© 2003 by The Society for Integrative and Comparative Biology
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Ancestral Gene Reconstruction and Synthesis of Ancient Rhodopsins in the Laboratory1
1 Department of Zoology, University of Toronto, Toronto, Ontario M5S 3G5, Canada
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONANCESTRAL RECONSTRUCTION AND...VERTEBRATE VISUAL PIGMENT...References |
|---|
Laboratory synthesis of ancestral proteins offers an intriguing opportunity to study the past directly. The development of Bayesian methods to infer ancestral sequences, combined with advances in models of molecular evolution, and synthetic gene technology make this an increasingly promising approach in evolutionary studies of molecular function. Visual pigments form the first step in the biochemical cascade of events in the retina in all animals known to possess visual capabilities. In vertebrates, the necessity of spanning a dynamic range of light intensities of many orders of magnitude has given rise to two different types of photoreceptors, rods specialized for dim-light conditions, and cones for daylight and color vision. These photoreceptors contain different types of visual pigment genes. Reviewed here are methods of inferring ancestral sequences, chemical synthesis of artificial ancestral genes in the laboratory, and applications to the evolution of vertebrate visual systems and the experimental recreation of an archosaur rod visual pigment. The ancestral archosaurs gave rise to several notable lineages of diapsid reptiles, including the birds and the dinosaurs, and would have existed over 200 MYA. What little is known of their physiology comes from fossil remains, and inference based on the biology of their living descendants. Despite its age, an ancestral archosaur pigment was successfully recreated in the lab, and showed interesting properties of its wavelength sensitivity that may have implications for the visual capabilities of the ancestral archosaurs in dim light.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONANCESTRAL RECONSTRUCTION AND...VERTEBRATE VISUAL PIGMENT...References |
|---|
Reconstructing ancestral proteins can be valuable in identifying important functional transitions in the evolutionary history of a protein, particularly when the entire ancestral protein sequence is reconstructed and then synthesized in vitro (Stewart, 1995
; Chang and Donoghue, 2000; Benner, 2002). This approach is only just beginning to be explored more fully, and has yielded insight into the history of proteins as diverse as chymase enzymes (Chandrasekharan et al., 1996), RNAses (Jermann et al., 1995; Zhang and Rosenberg, 2002), genes involved in developmental pathways (Sun et al., 2002), and mammalian retroposons (Adey et al., 1994). Particularly relevant to the success or failure of these approaches are the recent advances in ancestral reconstruction methods, and further developments in models of molecular evolution (Maddison, 1995; Yang et al., 1995; Lio and Goldman, 1998; Schultz and Churchill, 1999; Thorne, 2000; Huelsenbeck and Bollback, 2001; Whelan et al., 2001).
Visual pigments are particularly suitable for the study of molecular evolution, as they are easily expressed and purified in the laboratory, and many spectroscopic and biochemical assays are available to measure different aspects of their function. Furthermore, due to their importance as the first step activated in the sensory visual cascade, changes in their biochemical properties can have profound consequences on the visual capabilities of an organism (Goldsmith, 1990). By the same token, an organism's visual system may evolve unique characteristics in response to certain environmental conditions, which may then be traced to differences in visual pigment function (Hunt et al., 1996). The study of these visual pigments offers a rare opportunity to observe the clear effect of changes in single proteins on the organism as a whole (Watt and Dean, 2000).
Another, perhaps even more intriguing reason for reconstructing ancestral proteins lies in the hope of achieving a better understanding of the biology of ancient animals that may have possessed these proteins (Benner, 2002). Proteins that are involved in sensory systems would be particularly relevant to revealing the physiology or even the behavior of ancient animals that can no longer be studied directly. Experimental tests of laboratory-recreated ancestral proteins would provide information quite different from that obtained through studies of fossils. Though clearly interpretations based on recreations of single molecules are limited, under the best of circumstances one may hope to test some of the theories of ancient animal biology derived from other methods such as paleontological studies.
Fossils preserved well enough to shed light on physiology and behavior are rare (Ruben et al., 1999; Fisher et al., 2000). Moreover, ancient tissue samples with intact genetic material are even more rare. Although there have been many attempts to amplify ancient DNA from samples such as those preserved in amber, or from dinosaur bone extracted from Cretaceous period coal beds, their success has remained rather unclear. Generally, it appears that material older than several hundred thousand years may not be a reliable source of DNA except under unusual circumstances (Hoss et al., 1996; Austin et al., 1997; Fig. 1A), though studies using more recent material may have a higher rate of success (Orlando et al., 2002). Phylogenetic methods of ancestral state inference offer an interesting alternative to the difficulties of amplifying ancient DNA. When combined with laboratory synthesis of inferred ancestral genes, these approaches have much potential as a tool in exploring the past evolutionary history of proteins. We review here phylogenetic ancestral reconstruction methods, laboratory synthesis of ancestral genes, and applications of these methods to the synthesis of an ancestral archosaur rhodopsin gene in the laboratory.
|
|
ANCESTRAL RECONSTRUCTION AND GENE SYNTHESIS METHODS |
|---|
|
TOP
SYNOPSISINTRODUCTIONANCESTRAL RECONSTRUCTION AND...VERTEBRATE VISUAL PIGMENT...References |
|---|
Ancestral reconstruction methods
There are two different methods to infer ancestral states on a phylogeny, parsimony and likelihood/Bayesian analysis. Parsimony methods (implemented in programs such as PAUP*; Swofford, 2002
) minimize the amount of evolutionary change along the branches of the tree, with trees or ancestral states that require the fewest changes being preferred (Maddison, 1991; Swofford et al., 1996). Although weighted parsimony methods can accommodate different scenarios of character change, parsimony does not incorporate an explicit model of molecular evolution, nor does this method correct for multiple substitutions at a site, implicitly assuming a low rate of change. These properties of parsimony can sometimes be problematic from the standpoint of reconstructing ancestral states (Yang et al., 1995; Zhang and Nei, 1997; Nielsen, 2002).
Phylogenetic methods based on maximum likelihood analysis (implemented in programs such as PHYLIP, Felsenstein, 1991; MOLPHY, Adachi and Hasegawa, 1994; PAML, Yang, 1997; and NHML, Galtier and Gouy, 1998) use as an optimality criterion a likelihood score, calculated according to a specified model of evolution (Felsenstein, 1981). Optimization of the likelihood score can be used to specify topology and parameters such as branch lengths, character state frequencies, and ancestral states. Bayesian methods may also be used to calculate ancestral states with the highest posterior probability. This can be done using the maximum likelihood topology, branch lengths, and model parameters as priors (empirical Bayes method; Yang et al., 1995), or alternatively the posterior probabilities can be calculated by taking into account the uncertainty in the maximum likelihood topology and parameters using a Markov chain Monte-Carlo approach (hierarchical Bayes method; Huelsenbeck and Bollback, 2001). This can have some advantages over parsimony (Yang et al., 1995; Koshi and Goldstein, 1996; Lewis, 1998). In using an explicit model of molecular evolution, stochastic methods allow for the incorporation of knowledge of the mechanisms and constraints acting on coding sequences, as well as the possibility of comparing the performance of different models, ultimately resulting in the development of more realistic models (Goldman, 1993).
With stochastic methods such as maximum likelihood and Bayesian analysis, it is important to explore different models of molecular evolution to determine how robust the ancestral reconstruction results are (Huelsenbeck and Bollback, 2001; Huelsenbeck et al., 2002). Moreover, oversimplified or unrealistic models have been shown in certain cases to yield incorrect or otherwise misleading phylogenetic reconstructions (Cao et al., 1994; Huelsenbeck, 1997; Buckley, 2002), emphasizing the importance of model selection. Models can be generally divided into three different types: nucleotide, amino acid, and codon. Nucleotide models range from the simplest, Jukes-Cantor (Jukes and Cantor, 1969), which assumes equal base frequencies and rates of transitions and transversions, to much more complex models allowing unequal base frequencies (Felsenstein, 1981), transition/transversion bias (Kimura, 1980), among-site rate heterogeneity (Yang, 1994), and/or nonstationary base composition (Galtier and Gouy, 1998).
The simplest amino acid model is the Poisson, which assumes equal amino acid frequencies and rates of substitution among amino acids. This model is clearly unrealistic, and does not generally perform well. Models have also been developed that allow unequal amino acid frequencies (Hasegawa and Fujiwara, 1993), and among-site rate heterogeneity (Yang, 1994), in addition to a general time reversible (GTR) model for amino acids, which allows for unequal rates of substitutions in the rate matrix for all the different classes of amino acids (Yang, 1997). Rate matrices have been calculated for a number of data sets, including those of Dayhoff (Dayhoff, 1978; Kishino et al., 1990) and Jones (Jones et al., 1992; Cao et al., 1994) for globular proteins, and mitochondrial transmembrane proteins (Adachi and Hasegawa, 1996). This allows for the reduction in the number of parameters in the model of evolution. Recent developments include amino acid models that allow replacement rates to be proportional to the frequencies of both the replaced and resulting residues (+gwF model; Goldman and Whelan, 2002).
Codon-based models of molecular evolution show the most promise, as they have the potential to incorporate both nucleotide and amino acid information. The original codon-based models assumed equal nonsynonymous to synonymous rate ratios among sites and lineages (Goldman and Yang, 1994; Muse and Gaut, 1994). Subsequent models have allowed that ratio to vary across lineages or sites in the protein (Nielsen and Yang, 1998; Yang, 1998; Yang et al., 2000), or even incorporating unequal frequencies of different types of nonsynonymous substitutions based on the nature of the various amino acids (Yang et al., 1998).
Given the diversity of models now available, choice of model for use in phylogenetic analysis and ancestral state inference is critical. An inappropriate model of evolution can lead to inconsistency in the likelihood analysis, and convergence to an incorrect result (Huelsenbeck, 1998). Ancestral inference methods are particularly sensitive to model choice. The possibility of an incorrect result can be reduced by selecting a model of evolution that displays the best fit to the sequence data at hand. Toward this end, likelihood ratio tests can be used to compare two models of evolution that are nested with respect to one another, in order to determine whether the more complex model fits the sequence data significantly better than the simpler model (Felsenstein, 1981; Yang et al., 1994; Huelsenbeck and Rannala, 1997). If the models are not nested, they cannot be directly compared using a likelihood ratio test, and other methods, such as the generation of the distribution of the test statistic using Monte Carlo simulation, must be used (Goldman, 1993).
Ancestral gene synthesis methods
In encoding an inferred protein sequence as a synthetic gene, the degeneracy of the genetic code can be used to incorporate many useful properties that aid in ensuring rapid, problem-free synthesis and expression of the ancestral gene. Unique restriction sites can be incorporated at key positions in the gene, and potential primer sites optimized to reduce mispriming and hairpin structures in the primer. Codon usage bias can be optimized for a particular expression system in order to achieve desired expression levels (Sharp et al., 1988). In many expression systems, rare codons are known to cause translational problems due to limited tRNA availability, resulting in misincorporations, truncated proteins, and overall reduced translational efficiency (Kane, 1995). Though optimizing codon usage frequencies usually results in much higher expression levels, incorporation of unpreferred codons is occasionally useful in slowing translation of signal sequences so that cellular membrane translocation systems are not saturated (Karnik et al., 1987). Secondary structure in mRNA has also been implicated in lowered expression levels in E. coli (Griswold et al., 2003). GC content can affect levels of heterologously expressed proteins (Sinclair and Choy, 2002), and in any case should be adjusted to minimize potential difficulties in oligonucleotide synthesis, PCR, and DNA sequencing. There is some indication that silent sites can affect protein folding, particularly if the sites are between major protein domains (Cortazzo et al., 2002). If necessary, a mammalian translation initiation consensus sequence can be placed immediately preceding the initiation methionine codon (Kozak, 1984; Karnik et al., 1987). Tags or antibody epitopes to aid in protein purification can be introduced at either end of the gene.
Artificial genes can be assembled using restriction enzyme cleavage/ligation, PCR, or a combination of both methods. With either method, it is important to ensure some overlap between the fragments to be joined. PCR is often the preferred method of choice, due to its ease and speed of use, but care must be taken to avoid propagating errors introduced in the chemical oligonucleotide synthesis, or the subsequent cloning and amplification. Once the artificial ancestral gene has been fully assembled and sequenced, it is placed in an appropriate expression vector using restriction sites engineered at the ends of the gene. The choice of vector, expression system, and purification procedure will depend on properties of the particular protein being expressed.
|
VERTEBRATE VISUAL PIGMENT EVOLUTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONANCESTRAL RECONSTRUCTION AND...VERTEBRATE VISUAL PIGMENT...References |
|---|
The visual transduction cascade in animals is triggered via absorption of a photon by a visual pigment complex located in the photosensory cells of the eye. The activation of this visual pigment complex forms the first step in a biochemical cascade of events that eventually leads to a neural signal indicating that light has been perceived (Menon et al., 2001
). Though the specifics of the biochemical cascade can vary, with many notable differences between vertebrate and invertebrate phototransduction cascades (Pak, 1995; Pepe, 2001), the first step in vision is always activation of a visual pigment complex, whose overall structure and function has remained largely conserved. The visual pigment complex consists of a protein moiety that is covalently bound to a vitamin A-derived 11-cis retinal chromophore. This complex responds to light via isomerization of the chromophore, inducing a conformational change of the enveloping visual pigment protein. The change in conformation of the protein then opens a binding site for the second messenger in the transduction cascade, the G protein transducin. Activation of transducin upon binding to the visual pigment sets off the first steps of the visual transduction cascade in the photoreceptors of the eye.
The visual system is unusual in that it must respond to differences in light intensities across many orders of magnitude, from moonless nights, to bright sunny days. Vertebrates have solved this problem through the evolution of two different photoreceptor cell types in the retina in order to achieve such a dynamic range. One photoreceptor cell type, the rod, is active under crepuscular and nocturnal conditions, whereas a completely different one, the cone, is active during the day. Electrophysiological recordings of single photoreceptors have shown that in comparison with cones, rods tend to exhibit greater photosensitivity, and signal amplification. These are specializations designed to maximize photon capture and efficiently trigger the signal transduction cascade (Baylor, 1996). Cones, on the other hand, are less photosensitive, but tend to be much faster in responding to light.
These functional differences between rod and cone photoreceptors are thought to be mediated in part by properties of the different visual pigments expressed in each, with cone pigments tending to have faster regeneration times, faster decay of the meta II activated state, and lowered pigment stability which is probably related to changes in the pKa of the protonated Schiff base (Shichida et al., 1994; Imai et al., 1995; Rieke and Baylor, 2000; Ebrey and Koutalos, 2001). Phylogenetic studies show that vertebrate rod visual pigments, or rhodopsins, are in fact derived from cone opsins (Okano et al., 1992; Chang et al., 1995). This suggests that despite their origins from cone visual pigments, rhodopsins have subsequently evolved biochemical specializations to allow for better dim-light function.
One of the most striking differences between rod and cone visual pigments lies in the variability of their absorption sensitivities. In contrast to the dazzling array of wavelength sensitivities across the visible spectrum exhibited by the cone visual pigments (
340–570 nm), rod visual pigment sensitivities tend to be rather constrained about 500 nm (Lythgoe, 1972). Although the reasons for the clustering of the rod visual pigment sensitivities remain speculative, one of the most plausible explanations suggests it may be the result of a tradeoff between the spectrum of light available at night (which tends to be red-shifted), and the limitations imposed by thermally-driven isomerizations of the chromophore (which are less of a problem in blue-shifted pigments; Goldsmith, 1990). The idea of thermal noise imposing limits on visual sensitivities, and that this would be exacerbated at longer wavelengths, was first proposed by H.B. Barlow almost 50 years ago (Barlow, 1956).
Cone visual pigments, which are active mainly during the day, are not generally operating at the limits of detection and therefore are not subject to these limitations. They have evolved a wide variety of spectral sensitivities, forming the basis of color vision in many animals. It is known that much of the diversity in photoreceptor sensitivities appears to be the result of variation in the opsin protein itself as there are but a few known chromophores, mainly derivatives of 11-cis retinal (Nakanishi, 1991). Spectral tuning is thought to be influenced by the environment within the opsin binding pocket in which the chromophore resides, and this has been the subject of many mutagenesis studies in both rod and cone visual pigments to elucidate the mechanisms by which this may occur. In fact, both mutagenesis experiments and comparative evolutionary studies suggest that it can be quite easy to shift the wavelength sensitivity of a visual pigment with only a few amino acid substitutions (Kochendoerfer et al., 1999).
Laboratory synthesis of an ancestral archosaur rod visual pigment
Ancestors of a major lineage of diapsid reptiles, the ancestral archosaurs are thought to have existed about 240–250 million years ago (Fig. 1A). Their descendants include extant species such as birds, crocodiles and alligators. They are also the ancestors of several extinct lineages that included some of the most spectacular reptiles to ever exist, the late Cretaceous dinosaurs. What little is known of the physiology and behavior of the ancient archosaurs is inferred from the fossil record, and by analogy to their living descendants. Recreating an ancestral archosaur rod visual pigment, or rhodopsin, in the laboratory offers the opportunity to study biochemical aspects of their vision more directly.
The ancestral archosaur rhodopsin amino acid sequence was inferred on a phylogeny which reflects systematic relationships among the taxa for which rhodopsin sequences were available in the databases (Fig. 1B). Empirical Bayes methods were used to estimate amino acids with the highest posterior probabilities for the archosaur node at each site in the protein, with the maximum likelihood estimates of branch lengths and model parameters as priors (Yang, 1997). Different models were explored, and where possible, likelihood ratio tests (LRT) were performed in order to determine the relative fit of the models. For the ancestral archosaur node, the amino acid reconstructions of the three best-fitting models, as determined through LRT's, agreed at all but three sites where one reconstruction differed from the other two. A gene corresponding to the consensus amino acid sequence was synthesized, as well as others with mutations at each of the three sites representing the alternate reconstructions (Fig. 2; Chang et al., 2002b).
|
The artificial gene corresponding to the inferred ancestral archosaur rhodopsin protein sequence was chemically synthesized in long fragments of up to 230 bases, and placed into a mammalian expression vector (pMT). This ancestral archosaur gene was then transiently transfected into monkey kidney (COS-1) cells, harvested, regenerated with 11-cis retinal in the COS-cell membranes, solubilized, and immunoaffinity purified using the 1D4 monoclonal antibody (Fig. 3; Ferretti et al., 1986
; Chang et al., 2002a). In order to confirm that the correct protein had indeed been expressed in these cells, a Western blot was performed (Fig. 4).
|
|
As expected for a visual pigment, the purified ancestral archosaur rhodopsin bound to the 11-cis retinal chromophore to produce a stable pigment with an absorption peak in the visible range when measured on a double beam spectrophotometer. Interestingly, the absorption maximum of the archosaur rhodopsin was found to be at 508 nm (Fig. 5A), which is slightly red-shifted from that of many vertebrate rhodopsins which tend to cluster around 500 nm. Within reptiles, it is at the higher end of the range of values reported for reptiles and particularly birds, which tend to have longer wavelength-absorbing rhodopsins (Fig. 1B). Upon bleaching with light, the visible absorption peak shifted to 383 nm, which suggests that the rhodopsin has been converted to its active meta II form (Fig. 5A, inset). To confirm that the light-activated conformation of the ancestral archosaur rhodopsin was functionally active, a fluorescence assay was used to measure guanine-nucleotide uptake by the G protein transducin (Farrens and Khorana, 1995
). The photoactivated archosaur pigment was found to activate transducin at a rate similar to that of bovine rhodopsin (Fig. 5B).
|
These results indicate that the ancestral archosaur rhodopsin synthesized in the laboratory is able to activate the G protein transducin as bovine rhodopsin when assayed directly, and that its spectrum is slightly red-shifted from 500 nm, as are the rhodopsins of many birds. It is somewhat surprising that the archosaur rhodopsin activates transducin equally well as bovine rhodopsin, as we used bovine transducin for our assay, which presumably is quite different from an ancestral archosaur transducin. Even more interestingly, among reptiles the red-shifted absorption maxima of the ancestral archosaur rhodopsin most resembles that of the birds. Other reptiles, such as lizards and crocodiles, tend to have rhodopsins with absorption maxima at shorter wavelengths. This suggests that birds, more than other archosaur descendants, have tended to retain the ancestral visual state with respect to their rhodopsins, though clearly more studies are needed in order to address this issue. It may also be interesting in future studies to reconstruct ancestral G proteins, to better characterize the evolution of G protein–rhodopsin interactions.
|
FOOTNOTES |
|---|
1 From the Symposium Comparative and Integrative Vision Research presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2003, at Toronto, Canada.
2 E-mail: changb{at}zoo.utoronto.ca
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONANCESTRAL RECONSTRUCTION AND...VERTEBRATE VISUAL PIGMENT...References |
|---|
Adachi, J., and M. Hasegawa. 1994. MOLPHY. Institute of Statistical Mechanics, Tokyo, Japan.
Adachi, J., and M. Hasegawa. 1996. Model of amino acid substitution in proteins encoded by mitochrondrial DNA. J. Mol. Evol, 42:459-468.[Web of Science][Medline]
Adey, N. B., T. O. Tollefsbol, A. B. Sparks, M. H. Edgell, and C. A. Hutchison III. 1994. Molecular resurrection of an extinct ancestral promoter for mouse L1. Proc. Natl. Acad. Sci. U.S.A, 91:1569-1573.
Austin, J. J., A. B. Smith, and R. H. Thomas. 1997. Palaeontology in a molecular world—the search for authentic ancient DNA. Trends Ecol. Evol, 12:303-306.[CrossRef]
Barlow, H. B. 1956. Retinal noise and absolute threshold. J. Opt. Soc. of Am, 46:634-639.
Baylor, D. 1996. How photons start vision. Proc. Natl. Acad. Sci. U.S.A, 93:560-5.
Benner, S. A. 2002. The past as the key to the present: Resurrection of ancient proteins from eosinophils. Proc. Natl. Acad. Sci. U.S.A, 99:4760-4761.
Buckley, T. R. 2002. Model misspecification and probabilistic tests of topology: Evidence from empirical data sets. Syst. Biol, 51:509-523.[CrossRef][Web of Science][Medline]
Cao, Y., and J. Adachi. 1994. Phylogenetic place of guinea pigs: No support of the rodent-polyphyly hypothesis from maximum-likelihood analyses of multiple protein sequences. Mol. Biol. Evol, 11:593-604.[Abstract]
Chandrasekharan, U. M., S. Sanker, M. J. Glynias, S. S. Karnik, and A. Husain. 1996. Angiotensin II—forming activity in a reconstructed ancestral chymase. Science, 271:502-505.[Abstract]
Chang, B. S., M. A. Kazmi, and T. P. Sakmar. 2002a. Synthetic gene technology: Applications to ancestral gene reconstruction and structure-function studies of receptors. Methods Enzymol, 343:274-94.[Web of Science][Medline]
Chang, B. S. W., K. S. Crandall, J. P. Carulli, and D. L. Hartl. 1995. Opsin phylogeny and evolution: A model for blue shifts in wavelength regulation. Mol. Phylogenet. Evol, 4:31-43.[CrossRef][Medline]
Chang, B. S. W., and M. J. Donoghue. 2000. Recreating ancestral proteins. Trends Ecol. Evol, 15:109-114.[CrossRef][Medline]
Chang, B. S. W., K. Jonsson, M. Kazmi, M. J. Donoghue, and T. P. Sakmar. 2002b. Recreating a functional ancestral archosaur visual pigment. Mol. Biol. Evol, 19:1483-1489.
Cortazzo, P., C. Cervenansky, M. Marin, C. Reiss, R. Ehrlich, and A. Deana. 2002. Silent mutations affect in vivo protein folding in Escherichia coli. Biochem. and Biophys. Res. Comm, 293:537-541.
Dayhoff, M. O. 1978. A model of evolutionary change in proteins. Matrices for detecting distant relationships. In M. O. Dayhoff (ed.), Atlas of protein sequence and structure, pp. 345–358. National Biomedical Research Foundation, Washington, D. C.
Ebrey, T., and Y. Koutalos. 2001. Vertebrate photoreceptors. Prog. Ret. Eye Res, 20:49-94.[CrossRef][Web of Science][Medline]
Farrens, D. L., and H. G. Khorana. 1995. Structure and function in rhodopsin. II. Measurement of the rate of metarhodopsin-II decay by fluorescence spectroscopy. J. Biol. Chem, 270:5073-5076.
Felsenstein, J. 1981. Evolutionary trees from DNA sequences: A maximum likelihood approach. J. Mol. Evol, 17:368-376.[CrossRef][Web of Science][Medline]
Felsenstein, J. 1991. PHYLIP: Phylogeny inference package. University of Washington, Seattle, Washington.
Ferretti, L., S. S. Karnik, H. G. Khorana, M. Nassal, and D. D. Oprian. 1986. Total synthesis of a gene for bovine rhodopsin. Proc. Natl. Acad. Sci. U.S.A, 83:599-603.
Fisher, P. E., D. A. Russell, M. K. Stoskopf, R. E. Barrick, M. Hammer, and A. A. Kuzmitz. 2000. Cardiovascular evidence for an intermediate or higher metabolic rate in an ornithischian dinosaur. Science, 288:503-5.
Galtier, N., and M. Gouy. 1998. Inferring pattern and process: Maximum-likelihood implementation of a nonhomogeneous model of DNA sequence evolution for phylogenetic analysis. Mol. Biol. Evol, 15:871-879.[Abstract]
Goldman, N. 1993. Statistical tests of models of DNA substitution. J. Mol. Evol, 36:345-361.
Goldman, N., and S. Whelan. 2002. A novel use of equilibrium frequencies in models of sequence evolution. Mol. Biol. Evol, 19:1821-1831.
Goldman, N., and Z. Yang. 1994. A codon-based model of nucleotide substitution for protein-coding DNA sequences. Mol. Biol. Evol, 11:725-736.[Abstract]
Goldsmith, T. H. 1990. Optimization, constraint, and history in the evolution of eyes. Quart. Rev. Biol, 65:281-322.[CrossRef][Medline]
Griswold, K. E., N. A. Mahmood, B. L. Iverson, and G. Georgiou. 2003. Effects of codon usage versus putative 5'-mRNA structure on the expression of Fusarium solani cutinase in the Escherichia coli cytoplasm. Prot. Exp. Purif, 27:134-142.[CrossRef][Web of Science][Medline]
Hasegawa, M., and M. Fujiwara. 1993. Relative efficiencies of the maximum likelihood, maximum parsimony, and neighbor-joining methods for estimating protein phylogeny. Mol. Phylogenet. Evol, 2:1-5.[CrossRef][Medline]
Hoss, M., P. Jaruga, T. H. Zastawny, M. Dizdaroglu, and S. Paabo. 1996. DNA damage and DNA sequence retrieval from ancient tissues. Nucleic Acids Res, 24:1304-1307.
Huelsenbeck, J. P. 1997. Is the Felsenstein zone a fly trap? Syst. Biol, 46:69-74.[CrossRef][Web of Science][Medline]
Huelsenbeck, J. P. 1998. Systematic bias in phylogenetic analysis: Is the Strepsiptera problem solved? Syst. Biol, 47:519-537.[Web of Science][Medline]
Huelsenbeck, J. P., and J. P. Bollback. 2001. Empirical and hierarchical Bayesian estimation of ancestral states. Syst. Biol, 50:351-366.[Web of Science][Medline]
Huelsenbeck, J. P., B. Larget, R. E. Miller, and F. Ronquist. 2002. Potential applications and pitfalls of Bayesian inference of phylogeny. Syst. Biol, 51:673-688.[CrossRef][Web of Science][Medline]
Huelsenbeck, J. P., and B. Rannala. 1997. Phylogenetic methods come of age: Testing hypotheses in an evolutionary context. Science, 276:227-232.
Hunt, D. M., J. Fitzgibbon, S. J. Slobodyanyuk, and J. K. Bowmaker. 1996. Spectral tuning and molecular evolution of rod visual pigments in the species flock of cottoid fish in Lake Baikal. Vision Res, 36:1217-24.[CrossRef][Web of Science][Medline]
Imai, H., Y. Imamoto, T. Yoshizawa, and Y. Shichida. 1995. Difference in molecular properties between chicken green and rhodopsin as related to the functional difference between cone and rod photoreceptor cells. Biochemistry, 34:10525-31.[CrossRef][Medline]
Jermann, T. M., J. G. Opitz, J. Stackhouse, and S. A. Benner. 1995. Reconstructing the evolutionary history of the artiodactyl ribonuclease superfamily. Nature, 374:57-59.[CrossRef][Medline]
Jones, D. T., W. R. Taylor, and J. M. Thornton. 1992. The rapid generation of mutation data matrices from protein sequences. Comp. Appl. Biosci, 8:275-282.
Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules. In H. N. Munro (ed.), Mammalian protein metabolism, pp. 21–132. Academic Press, New York.
Kane, J. F. 1995. Effects of rare codon clusters on high-level expression of heterologous proteins in Escherichia coli. Curr. Opin. Biotech, 6:494-500.[CrossRef][Web of Science][Medline]
Karnik, S. S., M. Nassal, T. Doi, E. Jay, V. Sgaramella, and H. G. Khorana. 1987. Structure-function studies on bacteriorhodopsin. II. Improved expression of the bacterio-opsin gene in Escherichia coli. J. Biol. Chem, 262:9255-9263.
Kimura, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol, 16:111-120.[CrossRef][Web of Science][Medline]
Kishino, H., T. Miyata, and M. Hasegawa. 1990. Maximum likelihood inference of protein phylogeny and the origin of chloroplasts. J. Mol. Evol, 31:151-160.[CrossRef]
Kochendoerfer, G. G., S. W. Lin, T. P. Sakmar, and R. A. Mathies. 1999. How color visual pigments are tuned. Trends Biochem. Sci, 24:300-5.[CrossRef][Web of Science][Medline]
Koshi, J. M., and R. A. Goldstein. 1996. Probabilistic reconstruction of ancestral protein sequences. J. Mol. Evol, 42:313-20.[CrossRef][Web of Science][Medline]
Kozak, M. 1984. Compilation and analysis of sequences upstream from the translational start site in eukaryotic messenger-RNAs. Nucleic Acids Res, 12:857-872.
Lewis, P. O. 1998. Maximum likelihood as an alternative to parsimony for inferring phylogeny using nucleotide sequence data. In P. S. Soltis, D. E. Soltis, and J. J. Doyle (eds.), Molecular systematics of plants II: DNA sequencing, pp. 132–163. Kluwer, Boston.
Lio, P., and N. Goldman. 1998. Models of molecular evolution and phylogeny. Genome Res, 8:1233-44.
Lythgoe, J. N. 1972. List of vertebrate visual pigments. In H. J. A. Dartnall (ed.) Photochemistry of vision, pp. 604–623. Springer-Verlag, New York.
Maddison, W. P. 1991. Squared-change parsimony reconstructions of ancestral states for continuous-valued characters on a phylogenetic tree. Syst. Zool, 40:304-314.[CrossRef]
Maddison, W. P. 1995. Calculating the probability distributions of ancestral states reconstructed by parsimony on phylogenetic trees. Syst. Biol, 44:474-481.[CrossRef]
Menon, S. T., M. Han, and T. P. Sakmar. 2001. Rhodopsin: Structural basis of molecular physiology. Physiol. Rev, 81:1659-1688.
Muse, S. V., and B. S. Gaut. 1994. A likelihood approach for comparing synonymous and nonsynonymous nucleotide substitution rates, with application to the chloroplast genome. Mol. Biol. Evol, 11:715-724.[Abstract]
Nakanishi, K. 1991. Why 11-cis-retinal? Amer. Zool, 31:479-489.
Nielsen, R. 2002. Mapping mutations on phylogenies. Syst. Biol, 51:729-739.[CrossRef][Web of Science][Medline]
Nielsen, R., and Z. Yang. 1998. Likelihood models for detecting positively selected amino acid sites and applications to the HIV-1 envelope gene. Genetics, 148:929-936.
Okano, Y., D. Kojima, Y. Fukada, Y. Shichida, and T. Yoshizawa. 1992. Primary structures of chicken cone visual pigments: Vertebrate rhodopsins have evolved out of cone visual pigments. Proc. Natl. Acad. Sci. U.S.A, 89:5932-5936.
Orlando, L., D. Bonjean, H. Bocherens, A. Thenot, A. Argant, M. Otte, and C. Hanni. 2002. Ancient DNA and the population genetics of cave bears (Ursus spelaeus) through space and time. Mol. Biol. Evol, 19:1920-1933.
Pak, W. L. 1995. Drosophila in vision research. Invest. Ophthalmol. Vis. Sci, 36:2340-2357.
Pepe, I. M. 2001. Recent advances in our understanding of rhodopsin and phototransduction. Prog. Retinal Eye Res, 20:733-759.[CrossRef][Web of Science][Medline]
Rieke, F., and D. A. Baylor. 2000. Origin and functional impact of dark noise in retinal cones. Neuron, 26:181-6.[CrossRef][Web of Science][Medline]
Ruben, J. A., C. Dal Sasso, N. R. Geist, W. J. Hillenius, T. D. Jones, and M. Signore. 1999. Pulmonary function and metabolic physiology of theropod dinosaurs. Science, 283:514-516.
Schultz, T. R., and G. A. Churchill. 1999. The role of subjectivity in reconstructing ancestral character states: A Bayesian approach to unknown rates, states, and transformation asymmetries. Syst. Biol, 48:651-664.[CrossRef]
Sharp, P. M., E. Cowe, D. G. Higgins, D. C. Shields, K. H. Wolfe, and F. Wright. 1988. Codon usage patterns in Escherichia coli, Bacillus subtilis, Saccharomyces cerevisiae, Schizosaccharomyces pombe, Drosophila melanogaster and Homo sapiens—a review of the considerable within-species diversity. Nucleic Acids Res, 16:8207-8211.
Shichida, Y., H. Imai, Y. Imamoto, Y. Fukada, and T. Yoshizawa. 1994. Is chicken green-sensitive cone visual pigment a rhodopsin-like pigment? A comparative study of the molecular properties between chicken green and rhodopsin. Biochemistry, 33:9040-9044.[CrossRef][Medline]
Sinclair, G., and F. Y. M. Choy. 2002. Synonymous codon usage bias and the expression of human glucocerebrosidase in the methylotrophic yeast, Pichia pastoris. Prot. Exp. Purif, 26:96-105.[CrossRef][Web of Science][Medline]
Stewart, C.-B. 1995. Active ancestral molecules. Nature, 374:12-13.[CrossRef][Medline]
Sun, H. M., S. Merugu, X. Gu, Y. Y. Kang, D. P. Dickinson, P. Callaerts, and W. H. Li. 2002. Identification of essential amino acid changes in paired domain evolution using a novel combination of evolutionary analysis and in vitro and in vivo studies. Mol. Biol. Evol, 19:1490-1500.
Swofford, D. L. 2002. PAUP*, phylogenetic analysis using parsimony (*and other methods). Sinauer, Sunderland, Massachusetts.
Swofford, D. L., G. J. Olsen, P. J. Waddell, and D. M. Hillis. 1996. Phylogenetic inference. In D. M. Hillis, C. Moritz, and B. K. Mable (eds.), Molecular systematics, pp. 407–514. Sinauer, Sunderland, Massachusetts.
Thorne, J. L. 2000. Models of protein sequence evolution and their applications. Curr. Opin. Genet. Dev, 10:602-5.[CrossRef][Web of Science][Medline]
Watt, W. B., and A. M. Dean. 2000. Molecular-functional studies of adaptive genetic variation in prokaryotes and eukaryotes. Ann. Rev. Gen, 34:593-622.
Whelan, S., P. Lio, and N. Goldman. 2001. Molecular phylogenetics: State-of-the-art methods for looking into the past. Trends Gen, 17:262-272.[CrossRef][Web of Science][Medline]
Yang, Z. 1994. Maximum likelihood phylogenetic estimation from DNA sequences with variable rates over sites: Approximate methods. J. Mol. Evol, 39:306-314.[CrossRef][Web of Science][Medline]
Yang, Z. 1997. PAML: A program package for phylogenetic analysis by maximum likelihood. Comput. Appl. Biosci, 13:555-556.
Yang, Z. 1998. Likelihood ratio tests for detecting positive selection and application to primate lysozyme evolution. Mol. Biol. Evol, 15:568-573.[Abstract]
Yang, Z., N. Goldman, and A. Friday. 1994. Comparison of models for nucleotide substitution used in maximum-likelihood phylogenetic estimation. Mol. Biol. Evol, 11:316-324.[Abstract]
Yang, Z., S. Kumar, and M. Nei. 1995. A new method of inference of ancestral nucleotide and amino acid sequences. Genetics, 141:1641-1650.[Abstract]
Yang, Z., R. Nielsen, and M. Hasegawa. 1998. Models of amino acid substitution and applications to mitochondrial protein evolution. Mol. Biol. Evol, 15:1600-1611.[Abstract]
Yang, Z. H., R. Nielsen, N. Goldman, and A. M. K. Pedersen. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics, 155:431-449.
Zhang, J., and M. Nei. 1997. Accuracies of ancestral amino acid sequences inferred by the parsimony, likelihood, and distance methods. J. Mol. Evol, 44:S139-S146.
Zhang, J. Z., and H. F. Rosenberg. 2002. Complementary advantageous substitutions in the evolution of an antiviral RNase of higher primates. Proc. Nat. Acad. Sci. U.S.A, 99:5486-5491.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Anisimova and C. Kosiol Investigating Protein-Coding Sequence Evolution with Probabilistic Codon Substitution Models Mol. Biol. Evol., February 1, 2009; 26(2): 255 - 271. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||