Visual system evolution and the nature of the ancestral snake.

doi: 10.1111/jeb.12663

Visual system evolution and the nature of the ancestral snake ~ S*, F. L. SAMPAIO*, C. JARED†, M. M. ANTONIAZZI†, E. R. LOEW‡, B . F . S I M OE J. K. BOWMAKER§, A. RODRIGUEZ¶, N. S. HART**, D. M. HUNT**††, J. C. PARTRIDGE**‡‡ & D. J. GOWER* *Department of Life Sciences, The Natural History Museum, London, UK †Laboratorio de Biologia Celular, Instituto Butantan, S~ao Paulo, Brazil ‡Department of Biomedical Sciences, Cornell University, Ithaca, NY, USA §Institute of Ophthalmology, University College London, London, UK ¶Unit of Evolutionary Biology, Zoological Institute, Technical University of Braunschweig, Braunschweig, Germany **School of Animal Biology and The Oceans Institute, The University of Western Australia, Perth, WA, Australia ††Lions Eye Institute, University of Western Australia, Perth, WA, Australia ‡‡School of Biological Sciences, University of Bristol, Bristol, UK

Keywords:

Abstract

fossoriality; lizards; opsins; Scolecophidia; Squamata; vision.

The dominant hypothesis for the evolutionary origin of snakes from ‘lizards’ (non-snake squamates) is that stem snakes acquired many snake features while passing through a profound burrowing (fossorial) phase. To investigate this, we examined the visual pigments and their encoding opsin genes in a range of squamate reptiles, focusing on fossorial lizards and snakes. We sequenced opsin transcripts isolated from retinal cDNA and used microspectrophotometry to measure directly the spectral absorbance of the photoreceptor visual pigments in a subset of samples. In snakes, but not lizards, dedicated fossoriality (as in Scolecophidia and the alethinophidian Anilius scytale) corresponds with loss of all visual opsins other than RH1 (kmax 490– 497 nm); all other snakes (including less dedicated burrowers) also have functional sws1 and lws opsin genes. In contrast, the retinas of all lizards sampled, even highly fossorial amphisbaenians with reduced eyes, express functional lws, sws1, sws2 and rh1 genes, and most also express rh2 (i.e. they express all five of the visual opsin genes present in the ancestral vertebrate). Our evidence of visual pigment complements suggests that the visual system of stem snakes was partly reduced, with two (RH2 and SWS2) of the ancestral vertebrate visual pigments being eliminated, but that this did not extend to the extreme additional loss of SWS1 and LWS that subsequently occurred (probably independently) in highly fossorial extant scolecophidians and A. scytale. We therefore consider it unlikely that the ancestral snake was as fossorial as extant scolecophidians, whether or not the latter are para- or monophyletic.

Introduction The origin of snakes from non-snake squamates (‘lizards’) has challenged evolutionary biologists for some 150 years (reviewed by Rieppel, 1988). Two main aspects have been debated: the identity of the closest relative(s) of snakes and the nature of the ancestral Correspondence: Bruno F. Sim~ oes and David J. Gower, Department of Life Sciences, The Natural History Museum, London SW7 5BD, UK. Tel.: (+44) 0207 942 5080; fax: (+44) 0207 942 5054; e-mails: [email protected] and [email protected]

snake: that is, phylogenetic and ecological origins of snakes (Bellairs, 1972; Rieppel, 1988; Caldwell, 2007). The dominant hypothesis for the general ecological origin of snakes is that of terrestrial burrowing (see Caprette et al., 2004; Lee, 2005 for an alternative perspective). Support for this lies in the anatomical similarity of snakes and multiple lineages of burrowing (fossorial) lizards (as well as proposed phylogenetic affinities among these taxa), in plausible adaptive explanations for phenotypic characteristics of snakes and in the phylogenetic distribution of fossoriality

ª 2015 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. 28 (2015) 1309–1320 JOURNAL OF EVOLUTIONARY BIOLOGY ª 2015 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY

1309

1310

~ S ET AL. B . F . S I M OE

among living snakes (e.g. Janensch, 1906; Mahendra, 1938; Bellairs & Underwood, 1951; Rieppel, 1988). The visual system has played a special role in snake origin debates and particularly in the burrowing ancestry hypothesis. Walls (1940, 1942; see also Mahendra, 1938) argued that many anatomical features of snake eyes pointed to a burrowing ancestry. Underwood (1967) modified Walls’ explanation by arguing for a role for nocturnality rather than only for dedicated burrowing, because snake eyes are simplified but those of some fossorial lizards are degenerate (see also RochonDuvigneaud, 1943; Underwood, 1977). The most dedicated of burrowers among extant snakes are the scolecophidians (at least some of which are ‘commonly’ termed ‘blindsnakes’), with their slender, cylindrical bodies, small narrow heads and substantially reduced eyes that lie under unspecialized head scales (Schmidt, 1950; Underwood, 1967). It has long been accepted that scolecophidians lie outside a clade (the Alethinophidia) comprising all other living snakes (e.g. Rieppel, 1988). However, there has been disagreement as to whether Scolecophidia is a mono- or paraphyletic taxon, and this has important implications for reconstructing features of the ancestral snake. This is because scolecophidian paraphyly would make it more likely (in terms of most parsimonious ancestral state reconstruction) that the last common ancestor of living snakes was scolecophidian like. In the modern era, morphological phylogenetics has consistently found support for scolecophidian monophyly (e.g. Lee et al., 2007). On the other hand, molecular evidence is somewhat equivocal or leans towards paraphyly. Thus, Vidal et al. (2010) recovered a monophyletic Scolecophidia, but Vidal et al. (2009) and Wiens et al. (2008, 2012) have presented evidence for paraphyly in which either the scolecophidian family Leptotyphlopidae or Anomalepididae is more closely related to other extant snakes (Alethinophidia) than to other (typhlopoid) scolecophidians. Although they made no reference to previous (anatomically or molecularly informed) challenges to scolecophidian monophyly (e.g. McDowell & Bogart, 1954; Underwood, 1967; Rieppel, 1988; Heise et al., 1995; Lee et al., 2007), Wiens et al. expressed surprise at their results and used them to consider the nature of the ancestral snake, concluding that scolecophidian paraphyly indicates that snakes were primitively or ancestrally burrowers that subsequently re-invaded surface habitats. There are very strong (adaptive) links between ecological niche and the complement, functionality and spectral tuning of photopigments expressed in retinal photoreceptor cells (rods and cones) (e.g. Carrol, 2007; Davies et al., 2012). Here, we report evolutionary patterns among the visual photopigment (opsin) genes of scolecophidians and of other squamates, and we assess their implications for reconstructions of the ancestral snake. We conclude that, whether or not they are

monophyletic, scolecophidians are an inaccurate model for the ancestral snake in terms of their vision biology and, perhaps, their degree of fossoriality. Instead, extant scolecophidians should be understood as highly specialized snakes whose visual systems show a particularly extreme degree of visual pigment loss that can be interpreted as a response to a fossorial lifestyle and that this degree of fossoriality is unlikely to have been shared by the last common ancestor of all extant snakes.

Materials and methods Taxon sampling and barcoding Samples were collected by fieldwork or obtained commercially. Taxon sampling for surveys of opsin genes comprised one individual each of 23 squamate reptile species, 17 of which were newly surveyed for this study (Table S1). Sampling includes four scolecophidians (representing all of the three major extant scolecophidian lineages), nine alethinophidian snakes (surface dwellers and burrowers) and ten lizards, seven of which are fossorial, including three limbless amphisbaenians (worm lizards). Samples used in the molecular analyses are reported in Table S1. Snakes and lizards were euthanized using approved (UK Home Office Schedule 1; Instituto Butantan ethics committee) procedures, and the eyes were extracted, coarsely macerated and stored in RNAlater (Ambion, Carlsbad, CA, USA) at 80 °C. For microspectrophotometry (MSP), we sampled Anilius scytale from the same locality as the material used in the molecular work, plus the scolecophidians Typhlops platycephalus and Typhlops hypomethes collected from Puerto Rico and an American Leptotyphlops sp. obtained through the commercial pet trade. RNA extraction and cDNA synthesis Total RNA was extracted using TRIzol followed by purification with PureLinkTM RNA Mini Kit (Life Technologies/Ambion, Carlsbad, CA, USA) following the manufacturer’s protocol. First-strand complementary DNA (cDNA) was synthesized by priming 500 ng of total RNA with 1 lL of oligo(dT)20 (50 lM; Invitrogen, Carlsbad, CA, USA) and by Superscript IIITM reverse transcriptase (Invitrogen) according to manufacturer’s instructions. RNA complementary to the cDNA was removed using 2 units of Escherichia coli RNase H (Ambion) followed by incubation at 37 °C for 20 min. PCR conditions for opsin gene amplification We attempted to amplify sws1, sws2, lws, rh1 and rh2 visual opsin genes using primers newly designed for conserved (among vertebrates) regions (Table S2). All


1318


Acknowledgments This work was supported by grants from the Leverhulme Trust (RPG-342 to DJG, NSH, DMH and JCP), the US Fish & Wildlife’s Wildlife Without Borders – Amphibians in Decline scheme (96200-1-G280 to DJG) – and the Department of Life Sciences of The Natural History Museum, London. Permits for research and export were granted by Direction de l’Environment de l’Amenagement et du Logement and the Direction des Services Veterinaires de la Guyane, Cayenne, French Guiana, and the Cameroon Ministry of Forests and Wildlife. BFS thanks Patricia Bianca for the use of her laboratory equipment and Pedro Fontana and Cris Caporrino for laboratory help in Instituto Butantan (S~ ao Paulo, Brazil). For assistance in the field, DJG thanks everyone involved in the 2014 Bioko Biodiversity Protection Program southern Caldera expedition (especially Gail Hearn and Vanessa Callahan), Gabriela Bittencourt-Silva, Antoine Fouquet, Philippe Gaucher, Jeannot and Odette (Camp Patawa), Marcel Kouete and Mark Wilkinson. Additional practical assistance was provided by Giovanna Gondim Montingelli, Simon Maddock, Gill Sparrow, Hussam Zaher and the NHM Sequencing Facility. Constructive criticism of earlier drafts was provided by Michael Caldwell, Andy Gardner, Michel Laurin and Jeff Streicher. AR was supported by a grant from the Alexander von Humboldt Foundation. Reptile photographs were provided by Ma€el Dewynter and Mark Wilkinson. DJG would like to acknowledge the late (and much missed) Garth Underwood for inspiring an interest in snake retinal evolution.

References Asenjo, A.B., Rim, J. & Oprian, D.D. 1994. Molecular determinants of human red/green color discrimination. Neuron 12: 1131–1138. Bellairs, A.D. 1972. Comments on the evolution and affinities of snakes. In: Studies in Vertebrate Evolution (K.A. Joysey & T.S. Kemp, eds), pp. 157–172. Oliver and Boyd, Edinburgh. Bellairs, A.D. & Underwood, G. 1951. The origin of snakes. Biol. Rev. 26: 193–237. Caldwell, M.W. 2007. The role, impact, and importance of fossils: snake phylogeny, origins, and evolution (1869–2006). In: Evolutionary Transitions and Origins of Major Groups of Vertebrates (J. Anderson & H.D. Sues, eds), pp. 253–302. Indiana University Press, Bloomington, IN. Caprette, C.L., Lee, M., Shine, R., Mokany, A. & Downhower, J.F. 2004. The origin of snakes (Serpentes) as seen through eye anatomy. Biol. J. Linn. Soc. 81: 469–482. Carrol, S.B. 2007. The Making of the Fittest: DNA and the Ultimate Forensic Record of Evolution. W. W. Norton & Company, New York, NY. Castoe, T.A., de Koning, A.P.J., Kim, H.M., Gu, W., Noonan, B.P., Naylor, G. et al. 2009. Evidence for an ancient adaptive episode of convergent molecular evolution. Proc. Natl. Acad. Sci. USA 106: 8986–8991.

Castoe, T.A., de Koning, A.P.J., Hall, K.T., Card, D.C., Schield, D.R., Fujita, M.K. et al. 2013. The Burmese python genome reveals the molecular basis for extreme adaptation in snakes. Proc. Natl. Acad. Sci. USA 110: 20645–20650. Chan, T., Lee, M. & Sakmar, T.P. 1992. Introduction of hydroxyl-bearing amino acids causes bathochromic spectral shifts in rhodopsin. Amino acid substitutions responsible for redgreen color pigment spectral tuning. J. Biol. Chem. 267: 9478–9480. Cowing, J.A., Poopalasundaram, S., Wilkie, S.E., Robinson, P.R., Bowmaker, J.K. & Hunt, D.M. 2002. The molecular mechanism for the spectral shifts between vertebrate ultraviolet- and violet-sensitive cone visual pigments. Biochem. J. 367: 129. Darriba, D., Taboada, G.L., Doallo, R. & Posada, D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9: 772. Davies, W.L., Cowing, J.A., Bowmaker, J.K., Carvalho, L.S., Gower, D.J. & Hunt, D.M. 2009. Shedding light on serpent sight: the visual pigments of henophidian snakes. J. Neurosci. 29: 7519–7525. Davies, W.I.L., Collin, S.P. & Hunt, D.M. 2012. Molecular ecology and adaptation of visual photopigments in craniates. Mol. Ecol. 21: 3121–3158. Douglas, D.A. & Arnason, U. 2009. Examining the utility of categorical models and alleviating artifacts in phylogenetic reconstruction of the Squamata (Reptilia). Mol. Phylogenet. Evol. 52: 784–796. Emerling, C.A. & Springer, M.S. 2014. Eyes underground: regression of visual protein networks in subterranean mammals. Mol. Phylogenet. Evol. 78: 260–270. Emerling, C.A. & Springer, M.S. 2015. Genomic evidence for rod monochromacy in sloths and armadillos suggests early subterranean history for Xenarthra. Proc. R. Soc. B 282: 20142192. Fasick, J., Cronin, T., Hunt, D. & Robinson, P. 1998. The visual pigments of the bottlenose dolphin (Tursiops truncatus). Vis. Neurosci. 15: 643–651. Fasick, J.I., Applebury, M.L. & Oprian, D.D. 2002. Spectral tuning in the mammalian short-wavelength sensitive cone pigments. Biochemistry 41: 6860–6865. Foureaux, G., Egami, M.I., Jared, C., Antoniazzi, M.M., Gutierre, R.C. & Smith, R.L. 2009. Rudimentary eyes of squamate fossorial reptiles (Amphisbaenia and Serpentes). Anat. Rec. 293: 351–357. Hart, N.S., Coimbra, J.P., Collin, S.P. & Westhoff, G. 2012. Photoreceptor types, visual pigments, and topographic specializations in the retinas of hydrophiid sea snakes. J. Comp. Neurol. 520: 1246–1261. Hedges, S.B. & Vidal, N. 2009. Lizards, snakes, and amphisbaenians (Squamata). In: The Timetree of Life (S.B. Hedges & S. Kumar, eds), pp. 383–389. Oxford University Press, New York, NY. Heesy, C.P. & Hall, M.I. 2010. The nocturnal bottleneck and the evolution of mammalian vision. Brain Behav. Evol. 75: 195–203. Heise, P.J., Maxson, L.R., Dowling, H.G. & Hedges, S.B. 1995. Higher-level snake phylogeny inferred from mitochondrial DNA sequences of 12S rRNA and 16S rRNA genes. Mol. Biol. Evol. 12: 259–265. Hoffstetter, R. 1968. Reviewed work: a contribution to the classification of snakes by Garth Underwood. Copeia 1968: 201–213.


1312


provided phylogenetic tree and were used to detect positive selection (x > 1) acting in particular lineages. The simplest branch model (one ratio) allows only one x ratio across the tree, whereas the more complex freeratio model assumes independent x ratios for each branch. Branch models were also used to estimate x for four branch categories (snakes vs. ‘lizards’; fossorial vs. nonfossorial squamates). All branch models (Yang et al., 2000) were compared with the simplest (one ratio) model using the likelihood ratio test (LRT), and the simpler model was rejected if P < 0.05. Site models (M1a nearly neutral and M2a positive selection; M7b and M8b & x) allow x to vary among sites (amino acids or codons). Site models M2a and M8 were compared (using LRT) with the simpler site models M1a and M7, respectively. Bayes empirical Bayes (Yang et al., 2005) implemented in models M2a and M8 b & x was used to identify sites under positive selection. In branch-site models, x can vary across both sites and lineages (Zhang et al., 2005), and this was used to detect positive selection at sites for snakes as a clade. Branch-site models were compared with the simplest model M1a by LRT. Ancestral state sequence reconstruction was estimated by marginal and joint reconstruction using CODEML. Potentially convergent molecular evolution within squamate rh1 (the only opsin gene that was amplified in all taxa) was assessed using an empirical Bayesian approach implemented in a version of PAML modified by Castoe et al. (2009) and using the method developed by Liu et al. (2011). Microspectrophotometry Microspectrophotometry was performed using methods described by Davies et al. (2009) for A. scytale and by Sillman et al. (2001) for T. platycephalus, T. hypomethes and Leptotyphlops sp. MSP sampling was restricted to single individuals of these three species, and the three scolecophidians were not the same species sampled in the molecular work. All procedures were performed under dim red light. Eyes were enucleated from darkadapted snakes killed using approved procedures. Retinas were mounted in saline containing 10% (w/v) dextran and compressed between two coverslips sealed with wax. For the A. scytale specimen, a modified Liebman dual-beam microspectrophotometer was used. For the three scolecophidians, a single-beam microspectrophotometer was used following the methods reported by Sillman et al. (2001). For the three scolecophidians, the amount of retinal material was small and the preparations were not entirely clean. The reported records are from small blobs at the ends of what appeared to be photoreceptors as well as from free-floating pieces. In each case, the noisiest records were ignored. There was no evidence for more than a single spectral class in each case. The

retina was separated from the pigment epithelium and macerated. A drop of the dispersed retina was placed between two cover slides and transferred to a MSP stage (for details, see Loew, 1994). The MSP data from the visual pigments were recorded every 1 nm from 350 to 750 nm. Selection criteria followed Loew (1994). The data were normalized by estimating the spectral maximum by eye and fitting a Gaussian function to the data points 20 nm either side of the wavelength. The peak absorbance (kmax) of each pigment was estimated by methods developed by Mansfield (1985) and MacNichol (1986) with the templates from Lipetz & Cronin (1988). Attempts to amplify visual opsin genes from genomic DNA For the four scolecophidian snake species surveyed for visual opsin genes and for A. scytale, we attempted to amplify partially the sws1 and lws opsin genes from genomic DNA (given that we failed to amplify them from eye cDNA, see below). We designed multiple primers in conserved areas of each exon (Table S4) – these primers have successfully amplified opsin genes from gDNA in some other squamates (B. F. Sim~ oes, F. L. Sampaio, N. S. Hart, D. M. Hunt, J. C. Partridge & D. J. Gower, unpublished data). Primers used by Davies et al. (2009) and the primers used to amplify the sws1 and lws from cDNA (Table S2) were also used. We tried to amplify fragments in 25 lL PCRs: 19 PCR buffer (Invitrogen), 1–1.75 mmol (mM) of MgCl2 (Invitrogen), 50 lmol L 1 of deoxynucleotides (Bioline), 0.4 lmol L 1 of each primer and 1 unit Platinum Taq polymerase (Invitrogen) and 100 ng of cDNA. We used the following PCR cycling parameters: initial denaturation at 95 °C for 10 min; 30 cycles of 15 s at 95 °C (denaturation), 30 s at 50–55 °C (annealing) and 1 min and 30 s at 72 °C (extension); and a final extension at 72 °C for 1.5 min. The touchdown profile from successful amplification of opsin genes from cDNA was also tried (see above). In no cases were we able to amplify sws1 or lws genes from these species’ gDNA.

Results Visual pigment complement We assembled a data set of opsin gene sequences for 23 squamates, comprising ten lizards and 13 snakes (Table S1). We successfully amplified four visual opsin genes (sws1, sws2, lws and rh1) from eye cDNA for all eight of the lizards we newly sampled, and a fifth (rh2) in all the lizards except Bachia cf. flavescens. In contrast, only three opsin genes (sws1, lws and rh1) were amplified in most of the snakes that we sampled, the only exceptions being A. scytale and the four sampled scolecophidians, for which only rh1 was amplified (Table 1; Figs 1


Vision evolution and snake ancestry

Table 1 Functional visual opsin genes successfully amplified and sequenced (●) in sampled squamates. Blank cells indicate lack of amplification. Fossorial taxa indicated in bold. All data newly generated except for those taxa marked with asterisk (Table S1). In general, we have found rh2 the most difficult to amplify in squamates using our methods, such that lack of amplification of this gene (especially in Bachia cf. flavescens) is not necessarily a compelling indication of its absence.

‘Lizards’ Uta stansburiana* Anolis carolinensis* Ophiodes striatus Bachia cf. flavescens Feylinia sp. Melanoseps occidentalis Takydromus sexlineatus Amphisbaena sp. Amphisbaena alba Amphisbaena infraorbitale Snakes Amerotyphlops brongersmianus Epictia collaris Liotyphlops beui Typhlophis squamosus Anilius scytale Tropidophis feicki Xenopeltis unicolor* Python regius* Python bivittatus* Polemon collaris Ophiophagus hannah* Pseustes poecilonotus Atractus flammigerus

rh1

rh2

sws2

sws1

lws

● ● ● ● ● ● ● ● ● ●

● ● ●

● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ●

● ● ● ● ● ● ● ● ● ● ● ● ●

● ● ● ● ● ●

and S8–S10). Despite extensive use of different primer pairs proven in their ability to amplify opsin genes in a broad range of reptiles (Tables S2 and S4), we were unable to amplify sws1 or lws opsin genes from eye cDNA or from genomic DNA (gDNA) in the four sampled scolecophidians or in A. scytale. Use of degenerate opsin primers (Davies et al., 2009) failed to amplify rh2 or sws2 from eye cDNA or gDNA in any snake. Spectral tuning Amino acids at known (from studies in other vertebrates) rh1, sws1 and lws spectral tuning sites (and ancestral state reconstructions) are reported in Tables S5–S7, respectively. The translated rh1 amino acid sequences for scolecophidians and A. scytale predict a small shortwave shift in spectral tuning (relative to the ancestral vertebrate) from a peak absorbance (kmax) of 500–494 nm, dictated by a substitution of aspartic acid (D) by asparagine (N) at site 83 (Nathans, 1990) (Table S7). MSP data are consistent with the lack of expression of sws1 and lws in the retinas of scolecophidians and A. scytale – in each case, only a single photopigment

1313

with kmax of approximately 490–500 nm was detected (Fig. 2). The small size of the eyes and photoreceptor cells of these snakes made MSP particularly challenging, such that readings from only a few cells for each species were made and very precise measures of kmax were generally not possible. However, our MSP data for A. scytale (497 2 nm: Fig. 2d) indicate a small (2– 5 nm) mismatch with the kmax predicted from our rh1 sequence data (493 nm: Table S5). The sws1 snake sequences present no evidence of likely major spectral tuning shifts based on what is known from studies in other vertebrates. All squamates sampled thus far are predicted (or known) (Table S6; Cowing et al., 2002; Hunt & Peichl, 2013) to have a UV-sensitive SWS1 visual pigment with kmax of c. 360 nm, with the probable exception of some sea snakes that have been found to have short-wavelength-sensitive (probably SWS1) pigments with kmax values displaced to longer wavelengths than the UV, in the range 428–430 nm (Hart et al., 2012). Most squamate LWS visual pigments are predicted to have a kmax of 560 nm (Table S7), but the lws sequences of Bachia cf. flavescens, Feylinia sp. and Tropidophis feicki have alanine (A) rather than serine (S) at site 180, which is predicted to shortwavelength shift the kmax of LWS visual pigments by 5 to c. 555 nm (Asenjo et al., 1994). The fossorial snake Atractus flammigerus shares this same lws S180A substitution and additionally has an A rather than threonine (T) at position 285. These two amino acid substitutions together are predicted (Yokoyama & Radlwimmer, 1998; Yokoyama et al., 2008) to cause a short-wavelength shift to a LWS kmax of c. 536 nm (Asenjo et al., 1994). Molecular evolution The sequence alignments for sws1, lws and rh1 each include more than 95% of the protein-coding region, and for sws2 and rh2, approximately 70% and 85%, of the coding region, respectively. The all-opsin phylogeny used to verify the identity of the amplified genes is shown in Fig. S8. The rh1, sws1 and lws phylogenies (Figs 1 and S8–S11) include some unorthodox higherlevel relationships, such as monophyly of ‘lizards’ (Fig. 1) and nonmonophyly of Archosauria (Figs S8 and S9), Batrachia (Figs S8 and S9) and Amniota (Fig. S8). However, several of these relationships are not strongly supported, and taxon sampling is sparse. Coding regions of opsin genes are non-neutral phylogenetic markers, and similarly, unorthodox relationships are a general occurrence in vertebrate opsin phylogenies (e.g. Mohun et al., 2010; Davies et al., 2012; Castoe et al., 2013). Most of the unorthodox relationships we recover are among major nonsquamate lineages, and we pay no further attention to these results here. The inferred rh1 phylogeny (Fig. 1) is broadly congruent with squamate phylogeny and strongly supported monophyla include Squamata, Serpentes,


1314


Alligator mississippiensis Pelodiscus sinensis Chrysemys picta

“Lizards”

Birds Mammals Amphibians (outgroup)

Squamata

0.05

Anolis carolinensis Uta stransburiana Ophiodes striatus Melanoseps occidentalis Feylinia sp. Bachia cf. flavescens Takydromus sexlineatus Amphisbaena sp. Amphisbaena alba Amphisbaena infrorbitale Xenopeltis unicolor Ophiophagus hannah Polemon collaris Pseustes poecilonotus Alethinophidia Colubroidea Atractus flammigerus Python bivittatus Python regius Anilius scytale Tropidophis feicki Serpentes Epictia collaris Typhlophis squamosus Amerotyphlops brongersmianus Scolecophidia Liotyphlops beui

Fig. 1 Rhodopsin 1 gene (rh1) maximum likelihood tree based on a GTR + G + I model of sequence evolution. Black circles represent internal branches with > 70% ML bootstrap and 1 Bayesian posterior probability, and grey circles represent less well-supported internal branches.

Scolecophidia, Alethinophidia and Colubroidea (Fig. S9). However, rather than forming a clade, the two anomalepidid scolecophidian (Typhlophis squamosus, Liotyphlops beui) rh1 sequences are robustly recovered as sister to the sampled leptotyphlopid (Epictia collaris) and typhlopid (Amerotyphlops brongersmianus) scolecophidians, respectively. Otherwise, no unorthodox squamate relationships (scolecophidian monophyly notwithstanding) are well supported (Figs 1 and S9). In the rh1 tree, scolecophidian monophyly and the nonmonophyly of anomalepidid scolecophidians are supported by convergent (as identified by empirical Bayesian analysis) changes at seven amino acid sites (sites 119, 159, 168, 209, 216, 229 and 241). Two of these sites (119 and 209) are associated with the stabilization of the retinal chromophore pocket (Park et al., 2008). Two further sites (159 and 168) are located in transmembrane domain 4, which has an impact in kmax tuning via spectral site 164 (Chan et al., 1992). Additionally, site 216 is in the same transmembrane domain 5 in which amino acids responsible for the stabilization of the retinal chromophore pocket are located (Park et al., 2008).

Analysis of rh1 and lws sequences for all sampled Squamata under branch models estimated the ratio of synonymous to nonsynonymous substitutions (dN/dS or x) as 0.127 and 0.159, respectively (Table 2), suggesting somewhat purifying selection in these genes. In contrast, sws1 (x = 0.039) appears to be under very strong purifying selection (Table 2). Analysis with branch models yielded higher x estimates for the visual opsin genes of snakes than of lizards (Table 2). For rh1, branch models estimated higher x estimates for fossorial than for nonfossorial squamates (Table 2). Under site models, only one spectral tuning site is identified as under positive selection in squamates – lws site 180 (Tables 3 and 4). The results of analysis with branch-site models suggest widespread positive selection at the amino acid level across sws1 in snakes (occurring in 32 of a total 359 amino acids; Table S8), including the three ‘known’ spectral tuning amino acid sites 93, 97 and 118 (Fasick et al., 1998, 2002; Wilkie et al., 2000). Analysis with site models identified ten rh1 amino acid sites under positive selection (Table 4), including two sites (119, 209) associated with stabiliza-



(a)

1.4

(b)

494 ± 3nm

2

490 ± 8nm

1.2 1.5

0.8 0.6 0.4 0.2 0 350 –0.2

400

450

500

550

600

650

700

750

1

0.5 0350

400

500

450

–0.5

Wavelength (nm)

–0.4

Relave absorbance

Relave absorbance

1

1315

550

600

700

650

750

Wavelength (nm)

–1

–0.6

(c) 1.4

(d) 500 ± 6nm

1.2

497 ± 2nm

0.025

10

0.6

0.4 0.2 0 350

750 400

–0.2

450

500

550

600

650

700

750

Wavelength (nm)

0.020

8

0.015

6

0.010

4

0.005

2

0.000

Number of cells

0.8

Absorbance

Relave absorbance

1

0

400

–0.4 –0.6

450

500

550

600

Wavelength (nm)

Fig. 2 Microspectrophotometry (MSP) of (a) Typhlops platycephalus: MSP prebleach absorbance spectrum showing mean relative absorbance for 16 rods; (b) Typhlops hypomethes: MSP prebleach absorbance spectrum showing mean relative absorbance for eight rods; (c) Leptotyphlops sp.: MSP prebleached absorbance spectrum showing mean relative absorbance for 18 rods; (d) Anilius scytale: mean absorbance spectra of 17 rods (bin size 2 nm), before and after bleaching with white light; the prebleach rod spectrum (open squares) is overlaid with a vitamin A1 (rhodopsin) visual pigment template (line).

tion of the retinal chromophore pocket (Park et al., 2008). Additionally, five of the rh1 amino acid sites under positive selection are located in transmembrane domains 5 and 6, which are known to influence kmax through spectral tuning amino acid sites 261, 265 and 269 (Chan et al., 1992; Yokoyama et al., 1995; Lin et al., 1998).

Discussion In snakes, but not lizards, dedicated fossoriality is coincident with the absence of all ancestral vertebrate visual opsins except RH1 (Table 1, Fig. 1). One explanation for this result is that functional copies of sws1 and lws genes are retained in the genome of highly fossorial snakes but that expression is very low. However, our failure to amplify sws1 and lws from either cDNA or gDNA or to find evidence for more than a single (RH1) photopigment using MSP (Fig. 2) suggests that functional sws1 and lws are lost from the scolecophidian and A. scytale genomes. This is supported by the failure to find morphological cones in histological studies of scolecophidians (Underwood, 1967, 1970, 1977 and studies cited therein). The lack of amplification from gDNA in scolecophidians and A. scytale suggests that possible sws1 and lws pseudogenes, if retained, are divergent in these taxa.

Table 2 Ratio of synonymous to nonsynonymous substitutions (dN/dS = x) for squamate visual opsin gene sequences under branch models. 2Dl = twice the difference logarithm of the likelihood value for the models; d.f. = degrees of freedom used to compare the models (corresponding with the number of free parameters). For classifications, see Table 1. Gene

Model

x

2Dl

sws1

One ratio Two ratio

0.039 xsnakes = 0.161, xlizards = 0.057 Variable by branch 0.127 xfossorial = 0.133, xnonfossorial = 0.122 xsnakes = 0.151, xlizards = 0.105 Variable by branch 0.159 xsnakes = 0.1915, xlizards = 0.148 Variable by branch

– 695.1

– 1

561.6 – 10.3

35 – 1

rh1

lws

Free ratio One ratio Two ratio

Free ratio One ratio Two ratio Free ratio

d.f.

5.94

P-value – 3.44

153

3.52 96 – 0.0013 0.0148

69.4 – 22.32

45 – 1

0.0112 – 2.31 06

107.46

35

2.73

09

All other (nonscolecophidian and non-Anilius scytale) snakes studied thus far have three functional visual opsin genes (expressed in the retina or located in gDNA) – rh1, sws1 and lws – that are homologous with those in other vertebrates (this study; Davies et al.,


1316


Table 3 Ratio of synonymous to nonsynonymous substitutions (dN/dS = x) for squamate visual opsin gene sequences under site models. For each gene, two pairs of models are compared to test for significant difference in goodness of fit to data. 2Dl = twice the difference logarithm of the likelihood value for the models; d.f. = degrees of freedom used to compare the models (corresponding with the number of free parameters). See Table 4 for the sites inferred to be under positive selection for each gene under each best-fit model. See Materials and Methods for further details. Gene

Model

2Dl

d.f.

P-value

Mean x

Parameter values

sws1

M1a M2a M7b M8 b & x M1a M2a M7 b M8 b & x M1a M2a M7 b M8 b & x

– 0 – 8.28 – 22.12 – 6.96 – 5.04 – 6.96

– 2 – 2 – 2 – 2 – 2 – 2

– 1 – 0.016 – 2.56 06 – 0.03 – 0.024 – 0.03

0.179 0.179 0.117 0.117 0.182 0.182 0.138 0.139 0.219 0.235 0.1748 0.1854

x0 x0 p p0 x0 x0 p p0 x0 x0 P p0

rh1

lws

Table 4 Amino acid sites identified (using Bayes empirical Bayes) as under positive selection, identified under site models in three visual opsin genes in squamates. See Table 3 and Materials and Methods for further details. Sites in bold are those associated with the stabilization of the chromophore retinal pocket in rh1 and spectral tuning in lws. Values in parentheses are Bayesian posterior probabilities for the amino acid sites under positive selection. Gene

Model

Sites under positive selection

sws1 rh1

M8 b & x M2a M8 b & x

lws

M2a M8 b & x

86 (0.88) – 281 (0.56) None 11 (0.59) – 33 (0.84) – 119 (0.73) – 123 (0.71) – 158 (0.87) – 173 (0.67) – 209 (0.91) – 213 (0.58) – 217 (0.68) – 255 (0.57) 180 (0.94) – 233 (0.96) – 286 55 (0.65) – 171 (0.58) – 174 (0.79) – 180 (0.95) – 181 (0.84) – 212 (0.64) – 229 (0.63) – 233 (0.91) – 275 (0.66) – 286 (0.98)

2009; Castoe et al., 2013; (Vonk et al., 2013). Snakes lack the rh2 and sws2 genes inferred to have been present in the ancestral vertebrate (e.g. Davies et al., 2012; Vonk et al., 2013) and which we found to occur widely among ‘lizards’. This is supported by published MSP studies of snakes (Sillman et al., 1997, 1999, 2001; Macedonia et al., 2009; Hart et al., 2012) and strongly suggests that the ancestral snake possessed functional copies of three (rh1, sws1 and lws) visual pigment genes (see also Davies et al., 2009, 2012). If loss and gain are equally weighted, it is quantitatively more or equally parsimonious (if Tropidophis is the sister taxon to nonAnilus alethinophidians or to Anilius, respectively), but it is much less biologically plausible to infer loss of functional sws1 and lws genes in the ancestral snake and their re-evolution within alethinophidians. Thus,

= = = = = = = = = = = =

0.050, x1 = 1, p0 = 0.861 (p1 = 0.136) 0.050, x1 = 1, x2 = 15.95, p0 = 0.864, p1 = 0.136 (p0 = 0) 0.187, q = 1.356 0.997 (p1 = 0.003), p = 0.192, q = 1.453, xS = 1.462 0.029, x1 = 1, p0 = 0.842 (p1 = 0.158) 0.029, x1 = 1, x2 = 1, p0 = 0.842, p1 = 0.102 (p0 = 0.056) 0.102, q = 0.634 0.989 (p1 = 0.011), p = 0.106, q = 0.712, xS = 1.136 0.03, x1 = 1, p0 = 0.802 (p1 = 0.198) 0.026, x1 = 1, x2 = 3.33, p0 = 0.800, p1 = 0.192 (p0 = 0.006) 0.096, q = 0.452 0.991 (p1 = 0.009), p = 0.103, q = 0.531, xS = 2.691

contrary to Walls (1942; see also e.g. Schwab, 2012), we doubt strongly that the evolution of vision in alethinophidian snakes involved ‘reconstituting their eyes almost “from scratch”’, at least in terms of visual opsin complement. Although an lws S180A substitution is known to impart a kmax shift of only 5 nm (Asenjo et al., 1994), this site is under positive selection in other vertebrates (Yokoyama, 2005) as well as squamates, as found here, suggesting that this substitution is adaptive. There is no evidence of variation in spectral tuning of RH1 within Scolecophidia, but two of the six (nontuning) amino acid sites (119 and 209) at which convergent evolution was detected lie in regions of the rh1 gene associated with the stabilization of the retinal chromophore pocket (Park et al., 2008), and we identified these same two sites as being under positive selection. Other convergent sites are located in transmembrane domains 4 and 5, which are known to be involved in RH1 pigment spectral sensitivity (Chan et al., 1992). Further studies, especially using site-directed mutagenesis and in vitro expression, are required to understand the consequences of these functional changes in RH1 evolution in these highly fossorial snakes. We note that current predictions of kmax from sequence data are based on direct measures of kmax in relatively few vertebrates, including very few reptiles (especially snakes). Until there are more species for which both MSP data and opsin gene sequences are available, we are unable to assess the possibility that unknown tuning sites and mechanisms occur in these taxa. Previously (e.g. Underwood, 1967), scolecophidians were believed to be the only snakes lacking cones, but our genetic and MSP data predict that the same is true also for the alethinophidian A. scytale. The dipsadid colubroid snake At. flammigerus is fossorial, although not to the extent seen in scolecophidians (or A. scytale).



Our taxon sampling for snake opsin gene sequences is currently sparse, but we have no reason to believe that the inferred LWS shortwave shift predicted for At. flammigerus is typical for fossorial snakes that express this gene. Indeed, (predicted or observed) typical 560 nm LWS photopigments occur in the somewhat fossorial Xenopeltis unicolor (Davies et al., 2009) and Polemon collaris (this study). The contrast between scolecophidians (and A. scytale) and burrowing lizards in their visual opsin gene complement is striking. Our lizard sampling includes highly fossorial Amphisbaena spp., a genus that has reduced eyes lying under head scales (Underwood, 1970, 1977; Foureaux et al., 2009). Amphisbaenians, however, unlike all snakes have retained rh2 and sws2 and, unlike the dedicated burrowing snakes, sws1 and lws genes as well. Why have fossorial lizards, including dedicated burrowers with reduced eyes, retained cone opsins that have been lost in burrowing snakes? One of the more obvious potential explanations is absolute time since adopting a burrowing lifestyle. However, this does not, as yet, provide a compelling answer because the oldest known scolecophidian snake (Standhardt, 1986) and amphisbaenian lizard (Sullivan & Lucas, 2000) fossils are both Puercan in age (c. 66–63.3 Ma). Nonetheless, we note that we have not sampled all major extant lineages of amphisbaenians and that there is evidence that Scolecophidia might not be monophyletic, which complicates estimates of their age. Amphisbaenians are diverse in their degree of eye reduction. The amphisbaenian species sampled here have eyes covered by head scales (as do scolecophidians), but other species have both more and less reduced eyes (e.g. Kearney, 2003), and it would be interesting to investigate the visual photopigments of this group more thoroughly and to understand better the absolute timing of evolutionary of eye reduction. There is some signal for a close phylogenetic affinity between snakes and amphisbaenians (e.g. Lee, 2005; Douglas & Arnason, 2009), but this hypothetical relationship is strongly disputed (e.g. Lee, 2005; Vidal & Hedges, 2005; Wiens et al., 2008; Hedges & Vidal, 2009; Pyron et al., 2013) and the strongly divergent visual opsin complements reported here suggest that (the possibly ancestral) fossoriality in each of the two groups is not homologous. The loss of functional copies of two visual opsin genes in the ancestral snake is convergent with that of mammals (Davies et al., 2012), with the same two opsin genes (sws2 and rh2) inferred to have been absent also in the ancestral therian. Aspects of mammalian vision (including some features shared with snakes, such as loss of photoreceptor coloured oil droplets) have been explained as the result of adopting a nocturnal (Walls, 1942; Heesy & Hall, 2010) or twilight/mesopic (Davies et al., 2012) ecotype in their early history, with their somewhat reduced visual system described as having passed through a functional ‘bottleneck’. The snake–

1317

mammal convergence bears additional similarities in that highly fossorial members of both clades have lost additional visual opsins. For example, the golden mole (like scolecophidians) retains only functional rh1 (Emerling & Springer, 2014). Rochon-Duvigneaud (1943) and Underwood (1970, 1977) argued from eye morphology that the ancestors of extant snakes likely passed through a nocturnal phase prior to becoming burrowers, in contrast to most burrowing lizards that, they believed, evolved from diurnal ancestors. However, at least two points caution against applying a simplistic extension of the similarities between mammals and snakes to decipher the ecology of snake ancestors: (i) that, where studied, (even highly) nocturnal lizards (geckos: Kojima et al., 1992) have retained visual opsin genes that are absent in all snakes and (ii) that rh1 and lws genes in snakes appear to be under more relaxed selection than sws1, whereas the opposite has been found in mammals (Zhao et al., 2009; Invergo et al., 2013). To the best of our knowledge, scolecophidians and A. scytale are the only rod monochromats thus far identified among tetrapod vertebrates other than cave salamanders (Kos et al., 2001), caecilian amphibians (Mohun et al., 2010), some deep-diving cetaceans (Meredith et al., 2013), xenarthran mammals (Emerling & Springer, 2015) and the golden mole (Emerling & Springer, 2014). The loss of cones and of all visual opsins except RH1 has happened at least twice in snakes (once each in the ancestor of Scolecophidia and of Anilius), more times if Scolecophidia are not monophyletic. That not only scolecophidians, but also A. scytale, have lost two visual opsins (SWS1, LWS) that were very likely present in the ancestral snake supports the idea that dedicated fossoriality in noncaenophidian snakes might be a derived rather than ancestral trait. This is consistent with G Underwood’s proposal (see Rieppel, 2012) that at least some of the snake groups that are both narrow-mouthed (i.e. nonmacrostomatan) and burrowing might be ‘regressed macrostomatans’ rather than their nonmacrostomaty (and fossoriality) being primitive for snakes (see also Vidal & David, 2004). Dedicated burrowing in noncolubroid snakes has led to loss of all visual opsins except RH1, which is unlikely to have been reversible. Whether or not Scolecophidia are paraphyletic, their highly limited photopigment complement thus suggests that the ancestral snake was not scolecophidian like in its visual biology, and thus perhaps in its degree of fossoriality. Several previous workers recognized that scolecophidians are highly specialized burrowers (Schmidt, 1950; McDowell & Bogart, 1954; Underwood, 1967; Hoffstetter, 1968; Kley, 2006), a view that our investigation of visual pigment complement supports. More broadly, it is mistaken to assume scolecophidian traits are necessarily plesiomorphic based only on their phylogenetic position outside Alethinophidia.


1318


Acknowledgments This work was supported by grants from the Leverhulme Trust (RPG-342 to DJG, NSH, DMH and JCP), the US Fish & Wildlife’s Wildlife Without Borders – Amphibians in Decline scheme (96200-1-G280 to DJG) – and the Department of Life Sciences of The Natural History Museum, London. Permits for research and export were granted by Direction de l’Environment de l’Amenagement et du Logement and the Direction des Services Veterinaires de la Guyane, Cayenne, French Guiana, and the Cameroon Ministry of Forests and Wildlife. BFS thanks Patricia Bianca for the use of her laboratory equipment and Pedro Fontana and Cris Caporrino for laboratory help in Instituto Butantan (S~ ao Paulo, Brazil). For assistance in the field, DJG thanks everyone involved in the 2014 Bioko Biodiversity Protection Program southern Caldera expedition (especially Gail Hearn and Vanessa Callahan), Gabriela Bittencourt-Silva, Antoine Fouquet, Philippe Gaucher, Jeannot and Odette (Camp Patawa), Marcel Kouete and Mark Wilkinson. Additional practical assistance was provided by Giovanna Gondim Montingelli, Simon Maddock, Gill Sparrow, Hussam Zaher and the NHM Sequencing Facility. Constructive criticism of earlier drafts was provided by Michael Caldwell, Andy Gardner, Michel Laurin and Jeff Streicher. AR was supported by a grant from the Alexander von Humboldt Foundation. Reptile photographs were provided by Ma€el Dewynter and Mark Wilkinson. DJG would like to acknowledge the late (and much missed) Garth Underwood for inspiring an interest in snake retinal evolution.

References Asenjo, A.B., Rim, J. & Oprian, D.D. 1994. Molecular determinants of human red/green color discrimination. Neuron 12: 1131–1138. Bellairs, A.D. 1972. Comments on the evolution and affinities of snakes. In: Studies in Vertebrate Evolution (K.A. Joysey & T.S. Kemp, eds), pp. 157–172. Oliver and Boyd, Edinburgh. Bellairs, A.D. & Underwood, G. 1951. The origin of snakes. Biol. Rev. 26: 193–237. Caldwell, M.W. 2007. The role, impact, and importance of fossils: snake phylogeny, origins, and evolution (1869–2006). In: Evolutionary Transitions and Origins of Major Groups of Vertebrates (J. Anderson & H.D. Sues, eds), pp. 253–302. Indiana University Press, Bloomington, IN. Caprette, C.L., Lee, M., Shine, R., Mokany, A. & Downhower, J.F. 2004. The origin of snakes (Serpentes) as seen through eye anatomy. Biol. J. Linn. Soc. 81: 469–482. Carrol, S.B. 2007. The Making of the Fittest: DNA and the Ultimate Forensic Record of Evolution. W. W. Norton & Company, New York, NY. Castoe, T.A., de Koning, A.P.J., Kim, H.M., Gu, W., Noonan, B.P., Naylor, G. et al. 2009. Evidence for an ancient adaptive episode of convergent molecular evolution. Proc. Natl. Acad. Sci. USA 106: 8986–8991.

Castoe, T.A., de Koning, A.P.J., Hall, K.T., Card, D.C., Schield, D.R., Fujita, M.K. et al. 2013. The Burmese python genome reveals the molecular basis for extreme adaptation in snakes. Proc. Natl. Acad. Sci. USA 110: 20645–20650. Chan, T., Lee, M. & Sakmar, T.P. 1992. Introduction of hydroxyl-bearing amino acids causes bathochromic spectral shifts in rhodopsin. Amino acid substitutions responsible for redgreen color pigment spectral tuning. J. Biol. Chem. 267: 9478–9480. Cowing, J.A., Poopalasundaram, S., Wilkie, S.E., Robinson, P.R., Bowmaker, J.K. & Hunt, D.M. 2002. The molecular mechanism for the spectral shifts between vertebrate ultraviolet- and violet-sensitive cone visual pigments. Biochem. J. 367: 129. Darriba, D., Taboada, G.L., Doallo, R. & Posada, D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9: 772. Davies, W.L., Cowing, J.A., Bowmaker, J.K., Carvalho, L.S., Gower, D.J. & Hunt, D.M. 2009. Shedding light on serpent sight: the visual pigments of henophidian snakes. J. Neurosci. 29: 7519–7525. Davies, W.I.L., Collin, S.P. & Hunt, D.M. 2012. Molecular ecology and adaptation of visual photopigments in craniates. Mol. Ecol. 21: 3121–3158. Douglas, D.A. & Arnason, U. 2009. Examining the utility of categorical models and alleviating artifacts in phylogenetic reconstruction of the Squamata (Reptilia). Mol. Phylogenet. Evol. 52: 784–796. Emerling, C.A. & Springer, M.S. 2014. Eyes underground: regression of visual protein networks in subterranean mammals. Mol. Phylogenet. Evol. 78: 260–270. Emerling, C.A. & Springer, M.S. 2015. Genomic evidence for rod monochromacy in sloths and armadillos suggests early subterranean history for Xenarthra. Proc. R. Soc. B 282: 20142192. Fasick, J., Cronin, T., Hunt, D. & Robinson, P. 1998. The visual pigments of the bottlenose dolphin (Tursiops truncatus). Vis. Neurosci. 15: 643–651. Fasick, J.I., Applebury, M.L. & Oprian, D.D. 2002. Spectral tuning in the mammalian short-wavelength sensitive cone pigments. Biochemistry 41: 6860–6865. Foureaux, G., Egami, M.I., Jared, C., Antoniazzi, M.M., Gutierre, R.C. & Smith, R.L. 2009. Rudimentary eyes of squamate fossorial reptiles (Amphisbaenia and Serpentes). Anat. Rec. 293: 351–357. Hart, N.S., Coimbra, J.P., Collin, S.P. & Westhoff, G. 2012. Photoreceptor types, visual pigments, and topographic specializations in the retinas of hydrophiid sea snakes. J. Comp. Neurol. 520: 1246–1261. Hedges, S.B. & Vidal, N. 2009. Lizards, snakes, and amphisbaenians (Squamata). In: The Timetree of Life (S.B. Hedges & S. Kumar, eds), pp. 383–389. Oxford University Press, New York, NY. Heesy, C.P. & Hall, M.I. 2010. The nocturnal bottleneck and the evolution of mammalian vision. Brain Behav. Evol. 75: 195–203. Heise, P.J., Maxson, L.R., Dowling, H.G. & Hedges, S.B. 1995. Higher-level snake phylogeny inferred from mitochondrial DNA sequences of 12S rRNA and 16S rRNA genes. Mol. Biol. Evol. 12: 259–265. Hoffstetter, R. 1968. Reviewed work: a contribution to the classification of snakes by Garth Underwood. Copeia 1968: 201–213.



Huelsenbeck, J.P. & Ronquist, F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. Hunt, D.M. & Peichl, L. 2013. S cones: evolution, retinal distribution, development, and spectral sensitivity. Vis. Neurosci. 31: 115–138. Invergo, B.M., Montanucci, L., Laayouni, H. & Bertranpetit, J. 2013. A system-level, molecular evolutionary analysis of mammalian phototransduction. BMC Evol. Biol. 13: 52. € Janensch, W. 1906. Uber Archaeophis proavus Mass., eine schlange aus dem Eoc€an des Monte Bolca. Beitrage zur Palaontologie und Geologie Oestreich-Ungarns und des Orients 1– 33. Katoh, K., Misawa, K., Kuma, K. & Miyata, T. 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30: 3059–3066. Kearney, M.M. 2003. Systematics of the Amphisbaenia (Lepidosauria: Squamata) based on morphological evidence from recent and fossil forms. Herpetological Monographs 17: 180. Kley, N.J. 2006. Morphology of the lower jaw and suspensorium in the Texas blindsnake, Leptotyphlops dulcis (Scolecophidia: Leptotyphlopidae). J. Morphol. 267: 494–515. Kojima, D., Okano, T., Fukada, Y., Shichida, Y., Yoshizawa, T. & Ebrey, T.G. 1992. Cone visual pigments are present in gecko rod cells. Proc. Natl. Acad. Sci. USA 89: 6841–6845. Kos, M., Bulog, B., Szel, A. & Rohlich, P. 2001. Immunocytochemical demonstration of visual pigments in the degenerate retinal and pineal photoreceptors of the blind cave salamander (Proteus anguinus). Cell Tissue Res. 303: 25. Lee, M.S.Y. 2005. Molecular evidence and marine snake origins. Biol. Lett. 1: 227–230. Lee, M.S.Y., Hugall, A.F., Lawson, R. & Scanlon, J.D. 2007. Phylogeny of snakes (Serpentes): combining morphological and molecular data in likelihood, Bayesian and parsimony analyses. Syst. Biodivers. 5: 371–389. Lin, S.W., Kochendoerfer, G.G., Carroll, K.S., Wang, D., Mathies, R.A. & Sakmar, T.P. 1998. Mechanisms of spectral tuning in blue cone visual pigments: visible and raman spectroscopy of blue-shifted rhodopsin mutants. J. Biol. Chem. 273: 24583. Lipetz, L.E. & Cronin, T.W. 1988. Application of an invariant spectral form to the visual pigments of crustaceans: implications regarding the binding of the chromophore. Vision. Res. 28: 1083. Liu, Z., Li, S., Wang, W., Xu, D., Murphy, R.W. & Shi, P. 2011. Parallel evolution of KCNQ4 in echolocating bats. PLoS ONE 6: e26618. Loew, E.R. 1994. A third, ultraviolet-sensitive, visual pigment in the Tokay gecko (Gekko gekko). Vision. Res. 34: 1427–1431. Macedonia, J.M., Lappin, A.K. & Loew, E.R. 2009. Conspicuousness of Dickerson’s collared lizard (Crotaphytus dickersonae) through the eyes of conspecifics and predators. Biol. J. Linn. Soc. 97: 749–765. MacNichol, E.F. 1986. A unifying presentation of photopigment spectra. Vision. Res. 26: 1543–1556. Mahendra, B.C. 1938. Some remarks on the phylogeny of the Ophidia. Anat. Anz. 86: 347–356. Mansfield, R.J.W. 1985. Primate photopigments and cone mechanisms. In: The Visual System (A. Fein & J.S. Levine, eds), pp. 89–106. Liss, New York, NY. McDowell, S. & Bogart, C. 1954. The systematic position of Lanthanotus and the affinities of the anguinomorphan lizards. Bull. Am. Mus. Nat. Hist. 105: 1–142.

1319

Meredith, R.W., Gatesy, J., Emerling, C.A., York, V.M. & Springer, M.S. 2013. Rod monochromacy and the coevolution of cetacean retinal opsins. PLoS Genet. 9: e1003432. Mohun, S.M., Davies, W.L., Bowmaker, J.K., Pisani, D., Himstedt, W., Gower, D.J. et al. 2010. Identification and characterization of visual pigments in caecilians (Amphibia: Gymnophiona), an order of limbless vertebrates with rudimentary eyes. J. Exp. Biol. 213: 3586–3592. Nathans, J. 1990. Determinants of visual pigment absorbance: role of charged amino acids in the putative transmembrane segments. Biochemistry 29: 937. Park, J.H., Scheerer, P., Hofmann, K.P., Choe, H.-W. & Ernst, O.P. 2008. Crystal structure of the ligand-free G-proteincoupled receptor opsin. Nature 454: 183–187. Pattengale, N.D., Alipour, M., Bininda-Emonds, O.R.P., Moret, B.M.E. & Stamatakis, A. 2010. How many bootstrap replicates are necessary? J. Comput. Biol. 17: 337–354. Pyron, R., Burbrink, F.T. & Wiens, J.J. 2013. A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Evol. Biol. 13: 93. Rieppel, O. 1988. A review of the origin of snakes. Evol. Biol. 22: 37. Rieppel, O. 2012. “Regressed” macrostomatan snakes. Fieldiana Life Earth Sci. 5: 99. Rochon-Duvigneaud, A. 1943. Les Yeux et la Vision des Vertebres. Masson et Cie, Paris. Schmidt, K.P. 1950. Modes of evolution discernible in the taxonomy of snakes. Evolution 4: 79. Schwab, R.I. 2012. Evolution’s Witness. Oxford University Press, New York, NY. Sillman, A.J., Govardovskii, V.I., Rohlich, P., Southard, J.A. & Loew, E.R. 1997. The photoreceptors and visual pigments of the garter snake (Thamnophis sirtalis): a microspectrophotometric, scanning electron microscopic and immunocytochemical study. J. Comp. Physiol. A 181: 89. Sillman, A.J., Carver, J.K. & Loew, E.R. 1999. The photoreceptors and visual pigments in the retina of a boid snake, the ball python (Python regius). J. Exp. Biol. 202: 1931–1938. Sillman, A.J., Johnson, J.L. & Loew, E.R. 2001. Retinal photoreceptors and visual pigments in Boa constrictor imperator. J. Exp. Zool. 290: 359–365. Stamatakis, A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313. Standhardt, B.R. 1986. Vertebrate Paleontology of the Cretaceous/ Tertiary Transition of Big Bend National Park, Texas. Louisiana State University and Agricultural and Mechanical College, Baton Rouge, LA. Sullivan, R.M. & Lucas, S.G. 2000. A Paleocene (Puercan) amphisbaenian in a regurgitalite from the Nacimiento Formation, San Juan basin, New Mexico. In: New Mexico’s Fossil Records 2 (S.G. Lucas, ed.), pp. 145–149. New Mexico Museum of Natural History and Science Bulletin, Albuquerque. Terakita, A. 2005. The opsins. Genome Biol. 6: 9. Underwood, G. 1967. A Contribution to the Classification of Snakes. British Museum Natural History Publications, London. Underwood, G. 1970. The eye. In: Biology of the Reptilia: Morphology B (C. Gans & T.S. Parsons, eds), pp. 1–97. Academic Press, New York, NY. Underwood, G. 1977. Simplification and degeneration in the course of evolution of squamate reptiles. Colloq. Intx. Cent. Natl. Rech. Sci. 266: 341–353.


1320


Vidal, N. & David, P. 2004. New insights into the early history of snakes inferred from two nuclear genes. Mol. Phylogenet. Evol. 31: 783. Vidal, N. & Hedges, S.B. 2005. The phylogeny of squamate reptiles (lizards, snakes, and amphisbaenians) inferred from nine nuclear protein-coding genes. C. R. Acad. Sci. Biol. 328: 1000–1008. Vidal, N., Rage, J.-C., Couloux, A. & Hedges, S.B. 2009. Snakes (Serpentes). In: The Timetree of Life (S.B. Hedges & S. Kumar, eds), pp. 390–397. Oxford University Press, New York, NY. Vidal, N., Marin, J., Morini, M., Donnellan, S., Branch, W.R., Thomas, R. et al. 2010. Blindsnake evolutionary tree reveals long history on Gondwana. Biol. Lett. 6: 558–561. Vonk, F.J., Casewell, N.R., Henkel, C.V., Heimberg, A.M., Jansen, H.J., McCleary, R.J.R. et al. 2013. The king cobra genome reveals dynamic gene evolution and adaptation in the snake venom system. Proc. Natl. Acad. Sci. USA 110: 20651– 20656. Walls, G.L. 1940. Ophthalmological implications for the early history of the snakes. Copeia 1940: 1–8. Walls, G.L. 1942. The Vertebrate Eye and its Adaptive Radiation. Fafner Publishing Company, New York, NY. Wiens, J., Kuczynski, C., Smith, S., Mulcahy, D., Sites, J. Jr, Townsend, T. et al. 2008. Branch lengths, support, and congruence: testing the phylogenomic approach with 20 nuclear loci in snakes. Syst. Biol. 57: 420–431. Wiens, J.J., Hutter, C.R., Mulcahy, D.G., Noonan, B.P., Townsend, T.M., Sites, J.W. et al. 2012. Resolving the phylogeny of lizards and snakes (Squamata) with extensive sampling of genes and species. Biol. Lett. 8: 1043–1046. Wilkie, S.E., Robinson, P.R., Cronin, T.W., Poopalasundaram, S., Bowmaker, J.K. & Hunt, D.M. 2000. Spectral tuning of avian violet and ultraviolet-sensitive visual pigments. Biochemistry 39: 7895–7901. Yang, Z. 2007. PAML 4: phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24: 1586–1591. Yang, Z.H., Nielsen, R., Goldman, N. & Pedersen, A. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics 155: 431–449. Yang, Z.H., Wong, W. & Nielsen, R. 2005. Bayes empirical bayes inference of amino acid sites under positive selection. Mol. Biol. Evol. 22: 1107–1118. Yokoyama, S. 2005. Statistical and molecular analyses of evolutionary significance of red-green color vision and color blindness in vertebrates. Mol. Biol. Evol. 22: 968–975. Yokoyama, S. & Radlwimmer, F.B. 1998. The “five-sites” rule and the evolution of red and green color vision in mammals. Mol. Biol. Evol. 15: 560–567. Yokoyama, R., Knox, B.E. & Yokoyama, S. 1995. Rhodopsin from the fish, Astyanax: role of tyrosine 261 in the red shift. Invest. Ophthalmol. Vis. Sci. 36: 939. Yokoyama, S., Yang, H. & Starmer, W.T. 2008. Molecular basis of spectral tuning in the red- and green-sensitive (M/LWS) pigments in vertebrates. Genetics 179: 2037–2043.

Zhang, J.Z., Nielsen, R. & Yang, Z.H. 2005. Evaluation of an improved branch-site likelihood method for detecting positive selection at the molecular level. Mol. Biol. Evol. 22: 2472–2479. Zhao, H., Rossiter, S.J., Teeling, E.C., Li, C., Cotton, J.A. & Zhang, S. 2009. The evolution of color vision in nocturnal mammals. Proc. Natl. Acad. Sci. USA 106: 8980–8985.

Supporting information Additional Supporting Information may be found in the online version of this article: Table S1 Identification and GenBank accession numbers of the samples used in this study. Table S2 New primers for opsin amplification from cDNA. Table S3 Primers for mitochondrial 16S rRNA amplification for barcoding. Table S4 New primers for attempted opsin amplication from genomic DNA. Table S5 Known amino acid spectral tuning sites for rh1 and predicted peak absorbance (kmax) for squamates. Table S6 Known amino acid spectral tuning sites for sws1 and predicted peak absorbance (kmax) for squamates. Table S7 Known amino-acid spectral tuning sites for lws and predicted peak absorbance (kmax) for squamates. Table S8 Visual opsin gene amino acid sites identified (under branch-site models using Bayes Empirical Bayes) as under positive selection for snakes (as the foreground branch). Figure S1 Maximum Likelihood visual opsin genes phylogenetic tree estimated by RAxML based on GTR + G + I model of sequence evolution. Figure S2 Maximum Likelihood rhodopsin 1 (rh1) gene phylogenetic tree estimated by RAxML based on GTR + G + I model of sequence evolution. Figure S3 Maximum Likelihood long-wavelength (lws) opsin gene phylogenetic tree estimated by RAxML based on GTR + G model of sequence evolution. Figure S4 Maximum Likelihood short-wavelength (sws1) opsin gene phylogenetic tree estimated by RAxML based on GTR + G model of sequence evolution.

Received 20 February 2015; revised 6 May 2015; accepted 18 May 2015


Mosaic and concerted evolution in the visual system of birds.

The king cobra genome reveals dynamic gene evolution and adaptation in the snake venom system.

The evolution of MICOS: Ancestral and derived functions and interactions.

Ancestral vertebrate complexity of the opioid system.

The nature of visual self-recognition revisited.

Understanding Brassicaceae evolution through ancestral genome reconstruction.

Type 2 blindsight and the nature of visual experience.

The Nature of Snake Poisons-Viewed by the Light of the Most Recent Researches.

Fossils and the mosaic nature of human evolution.

The graded and dichotomous nature of visual awareness.

The ancestral levels of transcription and the evolution of sexual phenotypes in filamentous fungi.

Molecular evolution of snake venom toxins.

The "balance of nature"-evolution of a Panchreston.

The evolution and expression of the moth visual opsin family.

Primal eukaryogenesis: on the communal nature of precellular States, ancestral to modern life.

Nature or the natural evolution of plaque: what matters?

The nature of chemical innovation: new enzymes by evolution.

The Origin and Evolution of Baeyer-Villiger Monooxygenases (BVMOs): An Ancestral Family of Flavin Monooxygenases.

Evolution of the complement system.

The structure of experience, the nature of the visual, and type 2 blindsight.

Feature issue: visual function and the ageing visual system.

Restriction and recruitment-gene duplication and the origin and evolution of snake venom toxins.

The nature of the boundary between cortical visual areas II and III in the cat.

Evolution of sexes from an ancestral mating-type specification pathway.