An Example with Carotenoid Diversification in Birds

Exploring the Deterministic Landscape of Evolution: An Example with Carotenoid Diversification in Birds

Item Type text; Electronic Dissertation

Authors Morrison, Erin Seidler

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 02/10/2021 06:24:33

Link to Item http://hdl.handle.net/10150/624290

EXPLORING THE DETERMINISTIC LANDSCAPE OF EVOLUTION: AN EXAMPLE WITH CAROTENOID DIVERSIFICATION IN BIRDS

Erin Seidler Morrison

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2017 2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Erin Morrison, entitled EXPLORING THE DETERMINISTIC LANDSCAPE OF EVOLUTION: AN EXAMPLE WITH CAROTENOID DIVERSIFICATION IN BIRDS and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy.

______Date: February 21, 2017 Alexander Badyaev

______Date: February 21, 2017 Renée Duckworth

______Date: February 21, 2017 Michael Sanderson

______Date: February 21, 2017 Sergey Gavrilets

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: February 21, 2017 Dissertation Director: Alexander Badyaev 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of the requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that an accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Erin Seidler Morrison 4

ACKNOWLEDGEMENTS

I would first like to thank Alex Badyaev, without whom this dissertation would not have been possible. It is thanks to Alex that I have learned how to independently think and question the world around me. While I know it has been challenging, I sincerely appreciate that he did not give up on me at a time when I almost gave up on myself. With his support I took on a project that perhaps was more ambitious than I bargained for, but Alex patiently helped through needlessly complicated measures and half-baked hypotheses, and did not charge me too much for repeated grammatical errors. Alex has taught me how to be a scientist and a teacher, and how to both appreciate and criticize the science we do. I will always be grateful that he took a chance on me and for his continued mentorship and support. There are a number of people who have provided much needed support, discussion, and opinions on the work presented here, and throughout my graduate school career. I appreciate the expertise, unique perspectives, and encouragement provided by the members of my committee: Renée Duckworth, Michael Sanderson, and Sergey Gavrilets. Their insightful questions and advice along the way gave me opportunities to think about my work from different perspectives, and I have learned so much from having the opportunity to interact with them. I cannot express how appreciative I am for all of the help I have received from past and current members of my lab group: Stepfanie Aguillon, Samantha Anderson, Virginia Belloni, Katie Chenard, Christopher Gurguis, Kelly Hallinger, Dawn Higginson, Ellen Ouellette, Jared Padway, Ahvi Potticary, and Georgy Semenov. They have sat through countless lab meetings on my work and their suggestions, advice, creativity, humor and patience were always welcome. I particularly want to thank Dawn Higginson, she continues to listen to me and give advice no matter how busy she is, and I have learned a tremendous amount from her. Almost all of the data collected for these studies would not have been possible without the help of a small army of people. I would like to thank Emmet Andrews, Virginia Belloni, Matt Coope, Courtney Christie, Caitlin Davey, Sarah Davis, Rachael Delaney, Victoria Farrar, Lauren Harris, Kelly King, Xander Posner, Jordan Veal, and Adam Welu for their tireless work. It has been so exciting to watch those who came in as undergraduates develop as scientists in their own right, and their enthusiasm for research was always the best part of my day. I would also like to thank Renée Duckworth for patiently helping me with HPLC. I appreciate the support of the EEB administrative staff for their help throughout the years. Funding for the research presented here was provided in part by grants from the National Science Foundation and the Packard Foundation to Alex Badyaev. I was also supported by fellowships from Amherst College and the University of Arizona Galileo Circle. I want to thank those who adopted me into their families and made sure that I always had someone to eat, swim, run, ride, and watch movies and TV with, and it is entirely thanks to all of them that the desert has become home: Sam Anderson, Sarah Baillie, Erin Dombrady, Cole Eskridge, Tim Gendler, Judy Helfand, Dennis Helfand, Dawn Higginson, Bridget Keene, Charlie Keene, Evan Keleman, Gavin Leighton, Janet Levine, Josh Levine, Kristen Metzger, Ellen Ouellette, and Ahvi Potticary. I cannot express the gratitude I owe to my family, Wendy, Rob and Kasey Morrison. While they still may not understand what it is that I do, they have been supportive in every decision I have made and they always let Ted get in his two cents. I also appreciate Stephen Selzer for his support. Lastly, I want to thank my grandmother, Violet Selzer, who told me every Monday night for the first five and a half years of graduate school that she was sure I would do great and that her opinion was definitely impartial. 5

TABLE OF CONTENTS

ABSTRACT……………………………………………………………………………………….6

I. INTRODUCTION………………………………………………………………………………8

II. PRESENT STUDY…………………………………………………………………………...13

REFERENCES…………………………………………………………………………………..18

APPENDIX A. STRUCTURING EVOLUTION: BIOCHEMICAL NETWORKS AND METABOLIC DIVERSIFICATION IN BIRDS………………………………………...21

APPENDIX B. THE LANDSCAPE OF EVOLUTION: RECONCILING STRUCTURAL AND DYNAMIC PROPERTIES OF METABOLIC NETWORKS IN ADAPTIVE DIVERSIFICATIONS…………………………………………………………………...56

APPENDIX C. BEYOND NETWORK TOPOLOGY: COEVOLUTION OF STRUCTURE AND FLUX IN METABOLIC NETWORKS…………………………………………...70

APPENDIX D. RETENTION AND RECOMBINATION OF BIOCHEMICAL MODULES IN THE EVOLUTION OF AVIAN CAROTENOID METABOLISM………………..108

ABSTRACT

Establishing metrics of diversification can calibrate the observed scope of diversity within a lineage and the potential for further phenotypic diversification. There are two potential ways to calibrate differences between phenotypes. The first metric is based on the structure of the network of direct and indirect connections between elements, such as the genes, proteins, enzymes and metabolites that underlie a phenotype. The second metric characterizes the dynamic properties that determine the strength of the interactions among elements, and influence which elements are the most likely to interact. Determining how the connectivity and strength of interactions between elements lead to specific phenotypic variations provides insight into the tempo and mode of observed evolutionary changes. In this dissertation, I proposed and tested hypotheses for how the structure and metabolic flux of a biochemical network delineate patterns of phenotypic variation.

I first examined the role of structural properties in shaping observed patterns of carotenoid diversification in avian plumage. I found that the diversification of species-specific carotenoid networks was predictable from the connectivity of the underlying metabolic network.

The compounds with the most enzymatic reactions, that were part of the greatest number of distinct pathways, were more conserved across species’ networks than compounds associated with the fewest enzymatic reactions. These results established that compounds with the greatest connectivity act as hotspots for the diversification of pathways between species.

Next, I investigated how dynamic properties of biochemical networks influence patterns of phenotypic variation in the concentration and occurrence of compounds. Specifically, I examined if the rate of compound production, known as metabolic flux, is coordinated among compounds in relation to their structural properties. I developed predictions for how different 7 distributions of flux could cause distinct diversification patterns in the concentrations and presence of compounds in a biochemical network.

I then tested the effect of metabolic network structure on the concentrations of carotenoids in the plumage of male house finches (Haemorhous mexicanus) from the same population. I assessed whether the structure of a network corresponds to a specific distribution of flux among compounds, or if flux is independent of network structure. I found that flux coevolves with network structure; concentrations of metabolically derived compounds depended on the number of reactions per compound. There were strong correlations between compound concentrations within a network structure, and the strengths of these correlations varied among structures. These findings suggest that changes in network structure, and not independent changes in flux, influence local adaptations in the concentrations of compounds.

Lastly, the influence of carotenoid network structure in the evolutionary diversification of compounds across species of birds depends on how the structure of the network itself evolves.

To test whether the carotenoid metabolic network structure evolves in birds, I examined the patterns of carotenoid co-occurrence across ancestral and extant species. I found that the same groups of compounds are always gained or lost together even as lineages diverge further from each other. These findings establish that the diversification of carotenoids in birds is constrained by the structure of an ancestral network, and does not evolve independently within a lineage.

Taken together, the results of this dissertation establish that local adaptations and the evolutionary diversification of carotenoid metabolism are qualitatively predictable from the structure of an ancestral enzymatic network, and this suggests there is significant structural determinism in phenotypic evolution. 8

I. INTRODUCTION

The results of diversification are on display all around us, but how can this diversification be measured? Is it possible, for example, to determine if a lineage can potentially be more diverse than its current distribution of phenotypes? Furthermore, why are some phenotypic changes observed, but others are never realized? To resolve these questions there needs to be a way to calibrate phenotypic differences. Specifically, there needs to be a standardized way to explain the mechanistic differences that separate one phenotype and another and the sequence of these differences. First, the scope of potential opportunities for phenotypic change needs to be defined by outlining different ways by which modifications could occur, given the current organization of a phenotype (Maynard Smith 1970; Gavrilets 2004; Gerhart and Kirschner

2007). The structure of these opportunities determines the number and types of changes that have to occur to transition from one phenotype to another (Alberch 1982). Second, the likelihood of distinct phenotypic changes needs to be established. This can be determined by dynamic properties of the strength of interactions between the genes, proteins, or metabolites underlying a phenotype (Bruggeman and Westerhoff 2007; Phillips 2008), which can shape the fitness consequences of specific phenotypic changes (Poelwijk et al. 2007; Velenich and Gore 2013;

Jiménez et al. 2015). Developing the theory and tools to be able to assess both the specific changes that separate phenotypes and why we these particular changes take place is crucial in advancing our understanding of how adaptation and evolutionary diversification occur.

The application of a systems biology perspective to phenotypic diversification (Barabási and Oltvai 2004) allows for the development of specific hypotheses on the role of structural determinism in patterns of diversification (Webster and Goodwin 1982). Mapping observed phenotypes onto deterministic networks comprised of all of the potential interactions between 9 genes, proteins, enzymes, metabolites and regulatory mechanisms presents a way to directly test if the structural and dynamic properties of the interactions in the network constrain which phenotypes are generated out of all of the ones that are possible. Importantly, this approach is an opportunity to link microevolutionary processes that directly cause the gain or loss of new interactions underlying phenotypic changes to the macroevolutionary patterns of diversification among species and lineages.

Biochemical networks represent ideal systems to study the contributions of both the structural organization of the enzymatic reactions in pathways and the dynamic properties of reactions rates, known as metabolic flux, to the evolutionary diversification of compounds and their concentrations. A comprehensive catalogue of all of the possible enzymatic reactions known to produce compounds in nature (Kanehisa et al. 2014) provides an opportunity to determine the structural differences between different biochemical networks in terms of which compounds and reactions are present (Jeong et al. 2000; Bernhardsson et al. 2011). Differences in the dynamic properties of enzymatic reactions can be characterized by metabolic flux (Kacser and Burns 1973; Fell and Sauro 1985), which affects the concentrations of compounds produced by enzymatic reactions in a network (Almaas et al. 2004; Nidelet et al. 2016).

The focus of this dissertation is on how the existing topology of a biochemical network and dynamic properties of flux shape patterns of compound diversification. Determining the mechanisms that lead to variation in a compound’s concentration and occurrence among individuals and between species provides insight into the tempo and mode of evolutionary phenotypic changes in metabolism. As a result, predictions can be made about the scope of diversification in a biochemical network and the changes in metabolism that could factor into evolutionary adaptations and speciation. 10

In this dissertation, I developed a theoretical framework for the contributions of structural and dynamic properties of biochemical networks to patterns of evolutionary diversification, and applied these predictions to examine the mechanisms underlying the diversification patterns of carotenoid metabolic networks in birds. Birds are known for their spectacular diversity of plumage coloration. They derive their carotenoid color using carotenoids initially obtained from their diet that they can then modify into other compounds using enzymatic reactions (Brush

1990; McGraw 2006). Established reactions between carotenoids allowed me to construct a global metabolic network representing the diversity of reactions and pathways that have evolved in the ornaments of birds. The identification of carotenoid compounds in the plumage of hundreds of species of birds representing over 100 MYA of avian evolution (Badyaev et al.

2015) provided an opportunity to assess patterns of carotenoid network diversification over a large evolutionary time scale. Furthermore, the observed variation in carotenoid-based coloration of house finches (Haemorhous mexicanus) within the same population, provided a way to test the dynamic properties of metabolic flux in the carotenoid network using variations in compound concentrations among individuals. These studies allowed me to connect proximate changes in dynamic properties of metabolic flux within species to the broader macroevolutionary patterns of diversification in network structure and compound composition across species.

Explanation of Dissertation Format

The research presented in this dissertation integrates concepts in evolutionary biology with biochemistry using systems biology approaches and measures, phylogenetic comparative analyses, and a long-term study of a natural population. The document is presented in four appendices, each formatted as an independent manuscript. 11

In Appendix A, I examined how the structure of the global metabolic network comprised of all the possible enzymatic reactions and carotenoids used by birds to produce plumage-bound carotenoids is associated with patterns of diversification of compounds across species. I tested whether the differences in the occurrence of compounds across species were related to their structural properties in the global carotenoid network, and assessed where the greatest divergence in the production of compounds among species occurred. My results suggested that compounds associated with the most enzymatic reactions act as hotspots of diversification.

Species’ metabolic networks diverge when they express distinct enzymatic pathways starting from conserved, highly connected compounds. These findings support significant structural determinism in patterns of diversification.

In Appendix B, I reviewed the relationship between biochemical network structure and dynamic properties of metabolic flux to develop a theoretical framework explaining the mechanisms underlying different patterns of evolutionary diversification in metabolic networks.

I specifically examined mechanisms of flux control that regulate the rates of compound production and reviewed the locations in the network that act as targets for flux controls. I presented a series of predictions for the emergence of different functional and evolutionary relationships between reactions and compounds in a biochemical network as a result of distinct properties of metabolic flux.

In Appendix C, I empirically tested how dynamic properties of metabolic flux are integrated into the structural properties of metabolic networks. Using the plumage carotenoids of individuals from the same population of wild male house finches (Haemorhous mexicanus), I examined how the concentrations of compounds vary in response to changes in the structure of the metabolic networks that produce the carotenoids. I found that the flux of most of the 12 compounds depended on network structure and that the concentrations of compounds within the same network structure are strongly correlated with each other. These results provided support for the hypothesis that network structural and dynamic properties coevolve, and suggest that the evolutionary diversification of metabolic networks depends more on the gain or loss of enzymatic reactions, rather than independent changes in flux on static network structures.

In appendix D, I tested how the structure of a metabolic network evolves. In particular, I assessed whether the enzymatic reactions in a metabolic network are an ancestral trait that are selectively expressed among species, or if they are independently derived in distantly related lineages. I reconstructed the ancestral carotenoid metabolic networks of 250 species of birds and examined how the structure and composition of compounds in the networks changed over evolutionary time. I showed that the same compounds located the closest together in the network co-occur the most frequently in both ancestral and species’ networks and thus the scope of compounds produced by networks in descendant lineages does not change in relation to their ancestral networks. These results establish that the diversification of avian carotenoid metabolism is caused by differences in the selective expression of an ancestral carotenoid metabolic network structure. 13

II. PRESENT STUDY

The methods, results, and conclusions of this study are presented in the papers appended to this dissertation. The following is a summary of the most important findings in these documents.

In this research, I developed models and predictions to test how the evolutionary diversification of biochemical networks is determined by both structural and dynamic properties of enzymatic pathways. I tested this framework by examining the evolutionary diversification of the carotenoid metabolic network across species of birds and the proximate changes in carotenoid metabolism among individuals in a natural population of house finches. I used theory from systems biology and biochemistry, phylogenetic comparative methods, and biochemical techniques to examine the mechanisms underlying patterns of carotenoid diversification in the plumage of birds.

In the first study (Appendix A), I examined how the structure of a metabolic network can explain patterns of diversification of compounds expressed among species. I tested three potential mechanisms that could underlie the diversification of compounds across species. One hypothesis was that differences in the species’ carotenoid networks could be caused by the use of different enzymatic pathways that start at conserved compounds associated with the most reactions, and thus species would differ the most in compounds with the fewest enzymatic reactions. A second hypothesis posited that there could be sequential addition of reactions to the same pathway, which would cause differences in the lengths of pathways between species. For my third hypothesis, I tested if the difference between species’ networks was caused by differences in the expression of distinct subgroups of interconnected compounds, or modules, 14 and predicted that if this were the case then closely connected compounds would occur in the same number of species.

I compiled a database of all of the carotenoids identified in 339 species of birds and constructed global carotenoid metabolic network of all of the possible known enzymatic reactions that could produce these compounds. Using the structure of this global network and a set of specific rules I established using biochemical properties of carotenoid metabolism, I built individual metabolic for 250 species based on the carotenoids identified in their plumage. I found that a compound’s connectivity, measured by the number of reactions per compound, and not the number of reactions it was located from the start of a pathway, contributed the most to the diversification of compounds across species. Carotenoids with greater connectivity occurred more often in species than compounds with fewer reactions and differences in the connectivity of species’ networks, defined by the average reactions per compound, accounted for more of the differences in compounds among species than did differences in the pathway lengths of species’ networks. These patterns occurred because species used different reactions associated with evolutionarily conserved compounds. I did not find strong support for the gain or loss of entire modules of interconnected compounds contributing to the diversification of species’ networks.

These results showed that differences in the connectivity of compounds shape patterns of diversification in species’ carotenoid networks and suggested that compound diversification was constrained by the current structure of enzymatic reactions.

In the second study (Appendix B), I developed a set of predictions for how interactions between dynamic and structural properties of a biochemical network determine opportunities for the diversification of compound concentration and occurrence in a network. My results from the first study established that patterns of carotenoid diversification were determined by the 15 connectivity of the avian carotenoid metabolic network, but I was still missing the theory to explain why compounds in specific structural positions in the network contribute more to diversification than others. I needed a way to explain the likelihood for the occurrence and concentrations of compounds in a biochemical network to vary together or individually in their expression. To accomplish this, I reviewed the relationship between the structural properties of biochemical networks and the dynamic properties of metabolic flux that determine the rate of production of compounds, and thus establish the functional relationships between compounds and reactions. I documented the mechanisms that control flux in biochemical networks and examined the structural properties of the locations in networks that are the targets of flux control.

I derived a series of predictions for how the concentrations of compounds would vary in response to changes in flux at different positions in the network, and established the evolutionary patterns in the gain and loss of reactions and compounds that should occur when there is selection for different properties of metabolic flux.

In the third study (Appendix C) I empirically test the theoretical framework I developed in the second study by examining how the dynamic properties of metabolic flux are integrated into the structure of the carotenoid metabolic network. I tested the hypothesis that the structure of a metabolic network has coevolved with properties of metabolic flux, and thus the production of certain compound concentrations should depend on distinct network structures. Alternatively, I hypothesized that both the structure and flux in a metabolic network could evolve independently, such that the concentrations of compounds could evolve on a static network structure. To evaluate these hypotheses, I examined how the variations in the concentrations of plumage- bound carotenoids of male house finches from the same population were related to the structure of each individual’s metabolic network. This experimental design allowed me to isolate the 16 proximate effects of changes in network structure on the concentrations of carotenoids in the absence of evolutionary changes or the influence of different environmental factors, such as temperature or diet. I extracted and analyzed compounds from the plumage of 442 individuals, which represents the largest existing within-species data set of plumage carotenoid compounds. I constructed metabolic networks for each individual based on the compounds identified in their plumage using established reactions known to occur in the species. I found that there were 11 unique structural variants of carotenoid networks among individuals. The concentrations of derived compounds depended on the number of reactions they were directly associated with and the overall network structure. There were strong correlations among compounds within the same network structure. Changes in the total flux of compounds within a structure contributed more than differences in how flux was partitioned among compounds to variations in compound concentrations. These findings provided strong support for the influence of network structure in determining properties of metabolic flux and the concentration of compounds. They suggest that the evolutionary diversification of metabolic networks depends more on the gain or loss of enzymatic reactions than on changes in the rate of reactions on a static network.

The fourth study (Appendix D) examined the evolution of biochemical network structure.

I specifically tested if the structure of the carotenoid metabolic network was an ancient ancestral trait, derived before the diversification of the avian lineage, or if it was independently derived in separate lineages. This study established if the diversification of avian carotenoid metabolism is the result of the evolution of differences in the selective expression of an ancestral network, or if and different lineages are constrained by the evolution of distinct network structures. I reconstructed the ancestral networks of the 250 species of birds used in the first study to assess how the structure and compound composition of avian carotenoid metabolic networks has 17 changed over evolutionary time. I found that the same pairs of compounds co-occur together in both ancestral and extant species’ networks, and that they are located close to each other in the network, separated by only a few reactions. The length of conserved pathways did not change between ancestral and descendant networks, and this suggests that the scope of products in avian carotenoid networks does not significantly change among diverging lineages. These findings provided evidence for the ancestral origin of the avian carotenoid network, and they supported the influence of evolutionarily conserved structural properties in delineating the opportunities for diversification in avian carotenoid metabolism. 18

REFERENCES

Alberch P. (1982) Developmental constraints in evolutionary processes. Evolution and

Development (ed. J.T. Bonner), pp. 313-332. Springer-Verlag: Berlin.

Almaas E., Kovacs B., Vicsek T., Oltvai Z.N., Barabási A.-L. (2004) Global organization of

metabolic fluxes in the bacterium Escherichia coli. Nature 427:839-843.

Badyaev A.V., Morrison E.S., Belloni V., Sanderson M.J. (2015) Tradeoff between robustness

and elaboration in carotenoid networks produces cycles of avian color diversification.

Biology Direct 10:45.

Barabási A.-L., Oltvai Z.N. 2004. Network biology: Understanding the cell's functional

organization. Nature Reviews Genetics 5:101-113.

Bernhardsson S., Gerlee P., Lizana L. (2011) Structural correlations in bacterial metabolic

networks. BMC Evolutionary Biology 11:20.

Bruggeman F.J., Westerhoff H.V. (2007) The nature of systems biology. Trends in Microbiology

15:45-50.

Brush A.H. (1990) Metabolism of carotenoid-pigments in birds. The FASEB Journal 4:2969-

2977.

Fell D.A., Sauro H.M. (1985) Metabolic control and its analysis. European Journal of

Biochemistry 148:555-561.

Gavrilets S. (2004) Fitness Landscapes and the Origin of Species. Princeton University Press:

Princeton, N.J.

Gerhart J., Kirschner M. (2007) The theory of facilitated variation. Proceedings of the National

Academy of Sciences of the United States of America 104:8582-8589. 19

Jeong H., Tombor B., Albert R., Oltvai Z.N., Barabási A.-L. (2000) The large-scale organization

of metabolic networks. Nature 407:651-654.

Jiménez A., Cotterell J., Munteanu A., Sharpe J. (2015) Dynamics of gene circuits shapes

evolvability. Proceedings of the National Academy of Sciences 112:2103-2108.

Kacser H., Burns J.A. (1973) The control of flux. Symposia of the Society for Experimental

Biology 27:65-104.

Kanehisa M., Goto S., Sato Y., Kawashima M., Furumichi M., Tanabe M. (2014) Data,

information, knowledge and principle: Back to metabolism in KEGG. Nucleic Acids

Research 42:D199-D205.

Maynard Smith, J. (1970) Natural selection and the concept of a protein space. Nature 225:563-

564.

McGraw K.J. (2006) The mechanics of carotenoid coloration in birds. Bird coloration, volume 1:

Mechanisms and measurements (ed. G.E. Hill and K.J. McGraw), pp. 177-242. Harvard

University Press: Cambridge, MA.

Nidelet T., Brial P., Camarasa C., Dequin S. (2016) Diversity of flux distribution in central

carbon metabolism of S. cerevisiae strains from diverse environments. Microbial Cell

Factories 15:1-13.

Phillips P.C. (2008) Epistasis - the essential role of gene interactions in the structure and

evolution of genetic systems. Nature Reviews Genetics 9:855-867.

Poelwijk F.J., Kiviet D.J., Weinreich D.M., Tans S.J. (2007) Empirical fitness landscapes reveal

accessible evolutionary paths. Nature 445:383-386.

Velenich A., Gore J. (2013) The strength of genetic interactions scales weakly with mutational

effects. Genome Biology 14:R76. 20

Webster G, Goodwin BC. 1982. The origin of species: a structuralist approach. Journal of Social

and Biological Structures 5:15-47. 21

APPENDIX A. STRUCTURING EVOLUTION: BIOCHEMICAL NETWORKS AND METABOLIC DIVERSIFICATION IN BIRDS

Published citation: BMC Evolutionary Biology (2016) 16:168

Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 22 DOI 10.1186/s12862-016-0731-z

RESEARCHARTICLE Open Access Structuring evolution: biochemical networks and metabolic diversification in birds Erin S. Morrison* and Alexander V. Badyaev

Abstract Background: Recurrence and predictability of evolution are thought to reflect the correspondence between genomic and phenotypic dimensions of organisms, and the connectivity in deterministic networks within these dimensions. Direct examination of the correspondence between opportunities for diversification imbedded in such networks and realized diversity is illuminating, but is empirically challenging because both the deterministic networks and phenotypic diversity are modified in the course of evolution. Here we overcome this problem by directly comparing the structure of a “global” carotenoid network – comprising of all known enzymatic reactions among naturally occurring carotenoids – with the patterns of evolutionary diversification in carotenoid-producing metabolic networks utilized by birds. Results: We found that phenotypic diversification in carotenoid networks across 250 species was closely associated with enzymatic connectivity of the underlying biochemical network – compounds with greater connectivity occurred the most frequently across species and were the hotspots of metabolic pathway diversification. In contrast, we found no evidence for diversification along the metabolic pathways, corroborating findings that the utilization of the global carotenoid network was not strongly influenced by history in avian evolution. Conclusions: The finding that the diversification in species-specific carotenoid networks is qualitatively predictable from the connectivity of the underlying enzymatic network points to significant structural determinism in phenotypic evolution. Keywords: Network structure, Metabolic pathways, Phenotypic diversity

Background (defined here as a deterministic network) caused by gen- Only a small proportion of theoretically possible changes omic or developmental epistasis [1, 6–11], internal inte- seemed to be realized in phenotypic evolution and diver- gration during development [12–15], and physical stability sification, with some outcomes appearing recurrently or historical contingency of gene and protein associations whereas others are seemingly forbidden [1–5]. Such de- [16–22]. Direct examination of the correspondence be- terminism and predictability of phenotypic outcomes is tween opportunities for diversification imbedded in such surprising considering the dimensionality of the genome, networks and realized phenotypic diversity is needed to il- the proteome, and the developmental dynamics linking luminate the structural properties of networks that delin- them and point to the existence of constraints in pheno- eate phenotypic diversity. typic variation. Theoretical and empirical studies have Phenotypic diversification on a deterministic network is suggested that such constraints may be a reflection of the result of the gain or loss of elements and interactions the connectivity of the network of interactions among el- that convey different fitness [1, 3, 22]. Mechanistically, the ements such as genes, proteins, enzymes and metabolites evolutionary representation and variability of network elements tends to be associated with their topological positions [23–28]. In particular, two structural properties of * Correspondence: [email protected] – Department of Ecology and Evolutionary Biology, University of Arizona, networks the number of reactions per element, which Tucson, AZ, USA represents the connectivity of the network, and the

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 2 of 17 23

number of reactions that separate elements in a network, evolutionary change than elements with fewer direct in- which defines the length of pathways between elements in teractions in a network [23, 39, 40]. Thus, the divergence the network – provide distinct ways by which elements among species’ networks should be driven by the gain or and interactions in the network are gained or lost and re- loss of interactions among highly connected elements, sult in different patterns of phenotypic diversification whereas the connected elements themselves should be (Fig. 1) [29–33]. conserved across species. Differences in the number of Greater connectivity of an element – the number of interactions that start from these conserved elements direct interactions it has with other elements in a net- should be reflected in differences in the overall network work – enables an evolving lineage to include different connectivity (number of interactions per element) across elements that both directly interact with the same elem- species’ networks, because a greater number of oppor- ent [34–36]. In this mode of network diversification tunities exist for species to express different interactions (hereafter pathway diversification), the gain of different at densely connected compounds. If pathway diversifica- interactions associated with the same element represents tion causes divergence among species’ networks, then the start of divergent pathways comprised of unique ele- we expect differences in the elements and interactions ments and interactions (Fig. 1a). For example, in meta- present across species networks to increase with the dif- bolic networks, the use of different enzymatic reactions ferences in the connectivity of their networks, such that from the same substrate metabolite produces different interactions and elements associated with the most con- products resulting in distinct metabolic pathways. The- nected compounds in the network should vary the most ory and empirical data suggest that metabolic and pro- across species. tein networks commonly evolve by the preferential Thelengthofpathways– the number of interactions attachment of new enzymatic reactions or protein inter- (e.g., enzymatic reactions) that connect elements in a net- actions to the most connected elements in these network – enables an evolving lineage to express different el- works [24, 34, 37, 38]. Correspondingly, the genes ements and reactions along the same pathway. This mode underlying proteins and enzymes with greater connectiv- of network diversification (hereafter pathway elongation), ity tend to be represented in a greater number of taxa, results from differences in the number of sequential inter- have longer evolutionary persistence and lower rates of actions from the same starting element (Fig. 1b). Most

Fig. 1 The structure of a deterministic network and potential evolutionary trajectories. The possible interactions (arrows) between elements (small circles) represent potential opportunities for diversification on a deterministic network (shown in grey). The black, purple and orange shaded portions of the network show examples of different expressed networks, with each color denoting a different functional module made up of different elements and interactions. (a) Under the pathway diversification scenario, elements with the most interactions (higher connectivity) should be most conserved across networks, and the number and identity of the interactions associated with these connected elements should differ across networks. (b) Under the pathway elongation scenario, elements at the beginning of a sequential pathway of reactions should be the most conserved across networks, and the pathway length (the number of reactions that separate one element from another) and elements located further away from the start of the pathway should differ between networks. (c) Under the module diversification scenario, differences between networks are the result of the gain or loss of entire modules (unique groups of functionally coupled elements and interactions) and the gain or loss of elements would not be related to their connectivity or to their distance from a starting element in a pathway Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 3 of 17 24

genes, proteins, and metabolites are regulated by multi- among naturally occurring carotenoids (Additional file 1a) step interactions [35, 41] and thus in most cases, the acti- – is associated with patterns of avian diversification in vation or expression of an element is dependent on carotenoid-producing metabolic networks. The connectiv- several prior interactions. Changes in interactions at the ity and topology of enzymatic reactions of the global ca- beginning of a pathway may prevent the expression of in- rotenoid network have evolved largely in the context of teractions located further downstream in the pathway and bacterial evolution (e.g., [60, 61]) and subsets of this global result in shorter pathways and the loss of elements. Alter- network are utilized in the carotenoid metabolism of all natively, the addition of a new interaction to the end of a lineages studied to date, such as fungi, plants, insects and pathway can increase the length of the pathway and pro- animals (e.g., [62, 63]). Here we studied the patterns of duce a novel product. Models of network growth and em- utilization of this network associated with the production pirical results suggest that most of the change in networks of carotenoid pigmentation in the plumage and integu- occurs at their periphery, such that terminal elements are ment of 250 bird species. Specifically, we were interested most likely to be gained or lost, whereas the central or up- in the effect of the structure of the global metabolic net- stream elements are the most conserved [39, 42–44]. Lon- work on the frequency of occurrence of individual carot- ger pathways between elements in a network therefore enoid compounds and reactions across species. provide more opportunities for the use of different num- In birds, metabolism of carotenoids expressed in bers of sequential reactions from the same starting elem- feathers and integument necessarily starts with the con- ent, such that some species networks only express the sumption of dietary carotenoids (e.g., [64, 65]). This intermediate elements that lie along a pathway of interac- property of avian carotenoid biosynthesis allows for the tions from one element to another and the final product is identification of the starting points of metabolic path- never expressed. If network diversification is driven by dif- ways in species’ networks and provides an opportunity ferences in the elongation of a sequence of interactions to distinguish the effects of pathway diversification from among species, then we expect species’ networks to have the effects of pathway elongation and module diversifica- different pathway lengths from the same starting element. tion on network divergence across species. In birds, The difference in the length of the pathways among spe- pathway diversification from the same highly connected cies’ networks should be reflected in the diversification compounds, pathway elongation starting at the same among the elements and interactions present in each spe- dietary compounds, or the consumption of different cies. In this case, the elements located at the beginning of dietary compounds representing different functional pathways should be conserved across networks, and spe- modules in the network could produce evolutionary cies’ networks should diverge more from each other at ele- transitions across species’ networks. In the global carot- ments located closer to the ends of potential pathways. enoid network, opportunities for pathway diversification Networks are often organized in discrete functional and elongation vary across metabolic pathways that start modules in which a group of metabolites, enzymes, genes, at different dietary carotenoids (Figs. 2 and 3). Addition- or proteins interact more often with each other than with ally, the consumption of different dietary compounds re- other elements in the network [45, 46]. Functional mod- sults in access to different enzymatic reactions and ules play an important role in the evolvability of organisms metabolites that could comprise different functional [47–51]. Empirical studies have shown that genes in the modules (Fig. 2). Here, we first mapped species’ caroten- same regulatory modules tend to be co-expressed [52–55], oid networks onto the global avian carotenoid metabolic resulting in similar evolutionary rates of proteins in the network [66] and examined whether differences in en- same modules [56, 57]. Additionally, genes that underlie zyme connectivity or relative pathway position of individ- within-module enzymatic reactions have similar rates of ual carotenoid compounds were associated with their evolutionary gain and loss (e.g., [58, 59]), such that mul- evolutionary representation among species. We then re- tiple enzymatic reactions that comprise a pathway are peated these analyses for biochemical modules of inter- gained or lost together. Therefore, another mode of net- connected elements and examined their evolutionary work divergence among species could be the result of the representation in relation to their structural properties. gain or loss of complete functional modules (hereafter We examined the relative contribution of enzymatic con- module diversification) (Fig. 1c). If this is the case, then nectivity, metabolic pathway lengths, and module repre- species should differ in modules they express, and neither sentation on network divergence and identified the the connectivity of elements nor the length of a pathway structural properties of both individual compounds and between elements in a network should be related to the modules associated with diversification hotspots on differences in species’ networks. the global carotenoid network. We discuss the extent Here we examined the extent to which the structure of to which the structure of the carotenoid metabolic enzymatic reactions in the global carotenoid network – network can be used to understand and predict pat- that comprises all of the documented enzymatic reactions terns of realized phenotypic diversity. Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 4 of 17 25

Fig. 2 Schematics of the connected global enzymatic network of carotenoid compounds (66 compounds, 97 enzymatic reactions) found in species under this study (Additional files 1 and 2). Green nodes show dietary carotenoids. The distinct shaded areas represent the module assignments for the 53 compounds expressed at least once across species’ networks using simulated annealing [71, 72]. The numbers in the squares for each module denote the module number that corresponds to the module assignments for each compound in Additional file 1c

Methods compounds. We collected an exhaustive list of all the Data collection and metabolic network construction carotenoid compounds and reactions documented in The global carotenoid biosynthesis network includes birds (n = 339 species), using carotenoids that are all of the enzymatic reactions that occur among foundinplumage,integument(bill,tarsi,skin), naturally-occurring carotenoids in bacteria, plants, plasma, liver, fat, feces, retina, and seminal fluid, or fungi and animals (Additional file 1a, [66]). This net- are known to be consumed in the diet (Additional work delineates biochemical pathways of carotenoid file 1b; data current as of July 2015). The chromatog- biosynthesis based on the chemical properties of the raphy and mass spectrometry methods that are listed

Fig. 3 Structural diversity of carotenoid compounds in the avian space of the global carotenoid metabolic network (Fig. 2). Compounds differ in connectivity (reactions per compound), shown in the histogram on the left, and their distance (number of reactions) from the four main dietary (starting) compounds (lutein, zeaxanthin, β-carotene, β-cryptoxanthin), shown in the graph on the right Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 5 of 17 26

in Additional file 1b document the presence or absence of carotenoid metabolic network. The uncorrected P-dis- specific compounds against known standards. All of the tance is the fraction of the number of compounds and distinct compounds identified in the species of birds were reactions that differ between each pair of networks (d) then used to construct the “avian subset” of the global ca- out of the total number of compounds and reactions in rotenoid metabolic network, consisting of 66 carotenoids the global network (NG): and 97 enzymatic reactions (Fig. 2). The global metabolic d network was then used as a template to construct P ¼ 250 species-specific carotenoid metabolic networks be- N G tween known dietary carotenoid compounds (the up- The pairwise P-distances were computed in Mesquite stream elements of carotenoid metabolic networks in (version 3.03) [69] using the PDAP:PDTREE (version birds), metabolized compounds (e.g., circulating in plasma 1.16) package [70]. The metabolic distance (D) between or found in other organism tissues), and the expressed networks represents the fraction of compounds and re- compounds identified from species’ plumage and integu- actions in which two networks differ out of the total ment (Additional file 2). Briefly, after mapping compounds number of compounds and reactions that occur in each found in the diet, plasma, and plumage or integument of of the networks: species under this study on the “avian space” of the global carotenoid biosynthesis network (Fig. 2), we recorded bio- d D ¼ chemical pathways that link dietary, intermediate and N 1 þ N 2 plumage-expressed compounds for each species (Add- N N itional files 1b and 2; details and justification in Badyaev et where 1 and 2 are the total number of compounds S S al. [66], which also see for phylogenetic analyses of avian and reactions in networks 1 and 2, respectively. The 53 carotenoid networks). For species that had no known compounds expressed in the global carotenoid network dietary or intermediate compounds (but not both), at least once among the species’ networks were parti- missing compounds and reactions were assigned tioned into ten structurally defined modules based on based on the mapping of the species’ known com- the density of the compounds’ enzymatic interconnectivity pounds and reactions on the global network and re- using the simulated annealing program netcarto (https:// cording all biochemically possible connections (e.g., amaral.northwestern.edu/resources/software/netcarto) between a known dietary and a known expressed [71, 72]. This approach to module partitioning has previ- compound or between a known intermediate and a ously been used to reliably assign metabolites to the cor- known expressed compound and a possible dietary rect functional pathway based only on the structural compound). Networks were not built for species if properties of the metabolites [71]. In the avian carotenoid the carotenoids expressed in their plumage or in- metabolic network, the modules are partitioned by differ- tegument were unknown even when all other com- ent dietary compounds; seven of the ten modules include ponents of the network were documented. Thus, not at least one starting, upstream dietary compound. For all of the compounds and reactions in the avian sub- module assignments of the individual compounds in the set of the global carotenoid metabolic network (Additional global carotenoid metabolic network refer to Fig. 2 and file 1a, Fig. 2) were present in the species-specific net- Additional file 1c. works. In the 250 species-specific complete networks that were constructed, 53 compounds and 81 enzymatic Network structural measurements reactions occurred at least once. Species under this study For each compound in the avian carotenoid network represent eleven avian orders (Anseriformes, Charadrii- (Fig. 2) we calculated the number of directly linked en- formes, Ciconiiformes, Columbiformes, Galliformes, zymatic reactions [73] and the distance from a dietary Passeriformes, Pelecaniformes, Phaethontifromes, Phoeni- compound (minimum number of reactions between a copteriformes, Piciformes, Trogoniformes) and span over compound and any of the dietary compounds in the net- 110 MYA of avian carotenoid diversification (Fig. 4a, 4b, work) to represent the connectivity and the pathway 4c, 4d and 4e, Additional file 3) [66]. position of each compound, respectively. The connectivity (C) of each of the modules in the global network and Metabolic distance and modularity in networks each of the species’ networks was the average number of We used a modified metabolic distance based on the reactions per compound: Jaccard distance [67] and Rodrigues and Wagner [68] to r C ¼ calculate the fraction of reactions and compounds differ- n ing between any two metabolic networks. Species’ networks were coded based on the presence of compounds where r is the total number of reactions in the mod- and reactions in the avian subset of the global ule or network and n is the total number of Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 6 of 17 27

Fig. 4 (a) Consensus tree of the non-passerine species in this study showing, for each species’ metabolic network, the number of compounds (number of bars; green bars –distinct dietary carotenoids; yellow, orange and red bars – metabolically derived compounds), average degree (y-axis of the legend), number of modules (number of bar groups), pathway length (x –axis of the legend, number of enzymatic reactions from the closest dietary compound). The tree is a part of a majority rule consensus tree of 249 species based on 1,000 randomly sampled trees from the Hackett All Species pseudo posterior distribution from Jetz et al. [116] (Additional file 3). The other subsets of the tree, show in the inset in the lower left corner, are displayed in Figures 4b, 4c, 4d, and 4e

compounds in the module or network. The diameter Species representation and realized phenotypic of each of the species’ networks is the shortest dis- diversification tance (number of reactions) between the two most The species representation of a compound or reaction is distant dietary and expressed compounds in the net- the number of species that have this compound or reac- work. The diameter of each of the modules in the glo- tion (e.g., [39]). Whereas species representation character- bal network is the fewest number of reactions between izes the evolutionary representation of a compound, it does the two most distant compounds in the module. Both not include information on species’ phylogenetic relation- the connectivity of the species' networks and the mod- ships, and instead enables the examination of metabolic ules and the diameter of the modules were computed network evolution from a structural, rather than historical using Cytoscape 2.8.2 [74] with NetworkAnalyzer 2.7 perspective (e.g., [39]). In a companion study we found that [75, 76] and RandomNetworks 1.0 [77]. the phylogenetic relationships among the species in this Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 7 of 17 28

Fig. 4 (b) Consensus tree of the suboscine species under this study. Legend in Figure 4a

study were not reflected in the similarity of their bio- exploration across avian lineages. Several other studies chemical networks; the small biochemical space on have taken similar approaches to compare structural which birds diversify and the structure of the biochem- features of metabolic networks across species of bac- ical network instead leads to recurrent convergence of teria, eukaryotes, and archaea independently of their distantly related and ecologically distinct taxa in meta- phylogenetic relationships (e.g., [24, 35, 78]). bolicnetworks[66].Havingexaminedthehistorical The realized diversification (R) of an enzymatic reac- sequence of exploration of the global carotenoid tion was measured as the fraction of species that do not network by extant avian species in that study, here have a reaction among all of the species that have the we explore whether the structure of the global carot- substrate compound for the reaction (nc), where nr is the enoid network is reflected in the pattern of network number of species that have the reaction: Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 8 of 17 29

Fig. 4 (c) Consensus tree of a subset the oscine species under this study. Legend in Figure 4a

nc − nr network with little or no divergence between spe- R ¼ ’ nc cies networks along that part of a pathway; mean- ing that the enzyme is conserved across species An enzymatic reaction with a realized diversifica- that also have the enzyme’ssubstratecompound. tion score of zero represents a location in the The realized diversification of an enzymatic Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 9 of 17 30

Fig. 4 (d) Consensus tree of a subset of the oscine species under this study. Legend in Figure 4a

reaction with a score close to 1 represents a point Results of major divergence between species (i.e., the en- Global carotenoid network structural properties and zyme is only present in a small fraction of the diversity of species’ networks total species that have the enzyme’ssubstrate Connectivity and the distance from dietary caroten- compound). oids of compounds varied widely in the avian subset Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 10 of 17 31

Fig. 4 (e) Consensus tree of a subset the oscine species under this study. Legend in Figure 4a

of the global carotenoid network (Figs. 2 and 3). All number of modules (1-6), and number of dietary caroten- but one compound were associated with at least one oids (1-6). reaction to a maximum of 10 reactions. Non-dietary compounds were one to eight reactions away from Structural determinants of compound occurrence among starting dietary carotenoids (Fig. 3). The species’ net- species works (Fig. 4a, 4b, 4c, 4d and 4e; Additional file 1b) The connectivity of a compound contributed the most differed widely in the number of total compounds (1-21), to its species representation; carotenoids with higher number of reactions (0-46), connectivity (0-4.53 average connectivity had greater species representation (Fig. 5a; reactions per compound), diameter length (0-8 reactions), bST = 0.73, t = 7.63, P < 0.001, n = 55). Species Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 11 of 17 32

Fig. 5 A compound’s connectivity contributed more to the compound’s occurrence than did the compound’s relative distance from a dietary compound. Shown are partial regressions of a compound’s species representation on (a), the number of reactions per compound and (b), its distance from a dietary compound representation of a compound did not vary with its in more species (Fig. 6a; Spearman’s ρ = 0.80, P = 0.006, distance from a dietary carotenoid (Fig. 5b; bST = -0.07, n = 10), but the diameter of a module was not related t = -0.72, P=0.48, n =55). to the occurrence of the module across species (Fig. 6b; ρ = 0.49, P =0.15, n = 10). Differences in the The role of modules in compound occurrence among numbers of species with each of the compounds in a species module were correlated with the connectivity of the The representation of functional modules of the avian module (Fig. 6c; ρ = 0.74, P = 0.01, n = 10), but not with carotenoid network varied across species' networks the diameter of the module (Fig. 6d; ρ =0.59, P =0.07, (Fig.6aandb).Modulesofhigher connectivity occurred n = 10).

Fig. 6 Species representations of interconnected compounds within modules were related to the connectivity, but not the length of pathways of these modules. Compounds in modules characterized by (a), greater overall connectivity were overrepresented across species’ networks, whereas the occurrence of compounds in modules was not related to (b), the diameter of the module. Vertical bars represent the standard error. Differences in the species representation of compounds in the same module increased with (c), greater module enzymatic connectivity, but was not related to (d), the diameter of the module Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 12 of 17 33

bST = 0.38, t = 3.10, P = 0.003, n = 81). The realized diversification of reactions in the network was not predicted by the distance of their substrate compounds from dietary compounds (Fig. 8b; bST = -0.05, t = -0.39, P = 0.70, n = 81).

Discussion To what extent is the exploration of a deterministic network and its associated phenotypic diversification the result of the network’s structural properties? The divergence between species’ networks could be driven by either the exploration of pathways from conserved compounds, the elongation of conserved pathways, or the addition of different modules. Our findings suggest that pathway diversification is the main mechanism of divergence among species’ metabolic networks; differences in

Fig. 7 Differences in enzymatic connectivity contributed more to network divergence than differences in diameter. Shown are partial regression plots of the metabolic distance between pairs of species’ networks that share the same dietary (starting) compounds and the difference in (a), network connectivity and (b), diameter length between each pair of networks

Structural determinants of metabolic distance among species networks In pairs of species networks that shared dietary carotenoids, differences in network connectivity accounted for more of the metabolic distance between species’ networks (Fig. 7a; bST = 0.67, t = 75.24, P < 0.001, n = 4,839) than did differences in the diameters of the networks (Fig. 7b; bST = 0.28, t = 31.50, P < 0.001, n = 4,839). Pairs of networks with large differences in the average number of reactions per compound were more metabolically distinct than networks with large differences in their Fig. 8 Realized diversification of the reactions associated with a diameters. compound (the fraction of species that do not have a reaction among all of the species that have the substrate compound for the reaction) was predicted by the connectivity of the substrate compound Structural properties of realized diversification of (reactions per compound), but not by the substrate compound’s enzymatic reactions distance from a dietary compound. Shown are partial regressions of the The connectivity of a substrate compound contributed realized diversification of a reaction on (a), the enzymatic connectivity ’ to the realized diversification across species of the reac- and (b), the distance from a dietary compound of the reactions substrate compound tions associated with the substrate compound (Fig. 8a; Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 13 of 17 34

the enzymatic connectivity among species’ networks carotenoid enzyme convergence across bird species [66]. contributed more to their metabolic divergence than did Instead, species-specific expression of compounds and differences in the length of their diameters (Fig. 7). In reactions by the selective expression of different enzyme- the avian subset of the global carotenoid metabolic net- encoding genes from the global carotenoid network, work, the connectivity of a compound strongly contrib- appears to be the dominant mode of avian carotenoid net- uted to further network diversification: compound work evolution [88, 89], with de novo evolution of new ca- connectivity contributed the most to both the frequency rotenoid pathways (e.g., [90–92]) playing a secondary role of compound occurrence across species (Fig. 5a) and the (Additional file 1b). A potential mechanism that could realized diversification of the reactions associated with drive pathway diversification of enzymatic reactions at the compound among species’ networks (Fig. 8a). In these connected compounds is differences in the control contrast, pathway elongation did not play a major role in of metabolic flux among species across different pathways the diversification of avian carotenoid networks: the rela- [93]. Alternatively, different threshold concentrations of a tive distance from a dietary compound was not related substrate compound associated with several enzymatic re- to a compound’s representation across species (Fig. 5b) actions may be required to activate different enzymatic re- or to the realized diversification of reactions associated actions [94, 95], such that the diversification of these with the compound among species’ networks (Fig. 8b). pathways among species should be dependent on changes The presence of distinct structural modules and differ- in the concentrations of these connected compounds. ences in the species representation of compounds within We showed that the evolutionary representation of these modules contributed to the metabolic divergence compounds and enzymatic reactions reflected their across species: the most densely connected modules structural properties in the global carotenoid network were the most prevalent across species’ networks. Meta- (Fig. 5a). Why do compounds with the greatest connectiv- bolic divergence across species, however, was not due to ity tend to be overrepresented across species? The longer the concurrent gain or loss of all of the compounds in a evolutionary persistence of the most connected elements module (Fig. 6c and d). Thus, pathway diversification is a common property of protein and gene deterministic strongly contributes to metabolic divergence among spe- networks across many taxa [e.g., 23, 24, 39, 40] and could cies: modules characterized by greater connectivity pro- reflect their role in maintaining the overall structural co- vided more opportunities for the use of distinct pathways. hesiveness and function of the network. The removal or A central assumption of these tests and their interpret- modification of highly connected elements could have ation, is that species are co-opting elements (genes or greater pleiotropic effects that are more harmful to the enzymes) that comprise the global avian carotenoid function of the network than the removal of less con- metabolic network and are selectively expressing a par- nected compounds [96–98]. This property can result in ticular subset of these elements, rather than evolving stronger selection against the loss of these elements (e.g., them de novo. Several lines of evidence support this as- [99]) or, alternatively, in lesser effectiveness of purifying sumption. First, there was no correspondence between selection for the deletion of centrally located elements in the historical relationships across study species and their the network [100, 101]. Further, metabolic flux theory sug- utilization of carotenoid network space (i.e., use or dis- gests that enzymes with the highest flux control coeffi- use of particular reactions and compounds; [66, 79]). In- cients should be located at the branching points of stead the structure of networks, in particular the link pathways in metabolic networks [102–105]. Such enzymes between pathway elongation and pathway diversification, experience stronger stabilizing selection than those that accounted for recurrent convergence of phylogenetically contribute less to the flow of metabolites through meta- distant and ecologically distinct species in the utilization bolic pathways (e.g., [106]), accounting for the link of network space and expression of carotenoid com- between enzymatic connectivity and evolutionary persist- pounds (ibid.). Although such a pattern could be pro- ence found in this study (Fig. 5a). These conclusions are duced by the independent evolution of enzymes with corroborated by the models of network evolution and em- identical functions, it is highly unlikely (e.g., [80]). In pirical studies of network growth that find that new ele- other taxa, horizontal gene transfer [58, 81–84] and ments in a network preferentially attach to evolutionarily symbiotic events [85] accounted for enzymatic conver- stable elements that have greater connectivity rather than gence in carotenoid metabolism between unrelated spe- to sparsely connected, but more evolutionary labile down- cies, but neither of these processes play a significant role stream elements [24, 28, 34, 38]. in avian carotenoid biosynthesis. Gene duplications It is possible that dietary compounds – the upstream- could similarly account for the evolution of convergent most elements of avian carotenoid networks – are not enzymes [24, 83, 86, 87], but the rate of gene duplica- evolutionarily stable enough to contribute to incremen- tions in birds [88] seems orders of magnitude lower that tal pathway elongation over evolutionary time. The evo- would be required to explain the documented rates of lutionary rates of the gain and loss of dietary carotenoids Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 14 of 17 35

were orders of magnitude higher than the evolutionary entire modules of elements in a deterministic network lability of other compounds across avian metabolic net- may be ordered or unordered, depending on their relative works [66], and our results show that dietary com- positions, but either would result in recurrent bursts of di- pounds were no more likely to be present in a network versification across lineages’ phenotypes [113–115]. Be- than metabolized downstream compounds (Fig. 5b). cause we found no evidence of avian carotenoid network Theory predicts that rate-limiting enzymes should occur diversification due to pathway elongation, we would not at upstream positions in pathways (e.g., [44]), however expect a sequential order in patterns of realized diversifi- the evolutionary instability of dietary compounds can cation in carotenoid pathways during avian evolutionary decrease the effectiveness of selection on these com- history. Instead, our finding that differences among spe- pounds. Instead, due to the high enzymatic connectivity cies’ networks were due to pathway diversification from of some compounds in carotenoid networks, pathways highly connected compounds, suggests that related spe- from different dietary starting points can ultimately pro- cies should have similar carotenoid networks only when duce the same end products (Fig. 2). Thus, network ro- they utilize the same pathways from the same shared com- bustness to evolutionary labile dietary compounds – a pound. The results of this study thus explain why pheno- central feature of avian carotenoid networks [66, 107] – typic diversification in expressed carotenoids between may also contribute to the evolutionary stability of the related species was overwhelmingly due to unordered connected compounds and explain why the diversifica- periodic bursts of biochemical diversification of several tion of species’ networks was centered on connected compounds at once in the same pathway module across compounds instead of the continued lengthening of species, with ecological divergence in the use of dietary ca- pathways from specific dietary compounds. rotenoids – the process closely associated with ecological Variance in the species representations of compounds speciation, pathway elongation, and species relatedness – and enzymatic reactions within the same modules (Fig. 6c playing a significantly weaker role [66, 107]. and d) implies that the modules partitioned by their structural properties do not correspond to actual biological Conclusions processes (e.g. [108]), despite the fact that the structural The goal of this study was to explicitly consider how the modules used in this study were associated with different structural interactions among elements of a trait affect dietary compounds. Differences in the number of species its diversification. Our results show that the structure of with each compound in a module, however, could be the the enzymatic reactions in the avian space of the global result of the connectivity of each of the compounds to carotenoid network delineates opportunities for diversifi- other modules, which has been shown to explain the evo- cation of expressed carotenoids in birds. Within-species lutionary rate of genes in protein interaction networks studies can establish the proximate mechanisms under- [109]. Furthermore, it is possible that species utilize all of lying the observed association of network topology, en- the enzymatic reactions and produce all of the compounds zymatic connectivity and evolutionary diversification in in a module but selectively express only some of the com- carotenoid compounds. Explicit consideration of spatial pounds in their plumage [107, 110–112], and so the vari- and temporal organization of interactions between genes, ation of the species representations of compounds in proteins, enzymes and other elements of deterministic modules captures this selective compound deposition of networks brings us closer to an understanding of the the products of a module. relationship between potential and realized phenotypic By identifying the topological structural properties in a diversity. deterministic network that underlie phenotypic differences we can begin to establish specific mechanisms for the microevolutionary sequences behind observed macroevo- Additional files lutionary patterns. For example, if highly connected network elements determine phenotypic differences, then Additional file 1: (a) Appendix S1: Confirmed enzymatic reactions in the “avian space” of global carotenoid biosynthesis network in bacteria, phenotypic diversification in a lineage might not occur in plants, and animals. This appendix contains references supporting the sequential order (structural or temporal) because different presence of specific compounds and the enzymatic reactions that pathways can be explored from the same initial conserved comprise the avian carotenoid biosynthesis global network. (b) Appendix S2: Characteristics of carotenoid metabolic networks for species used in element, and so we would expect weak phylogenetic signal the study. This appendix contains the structural measurements and among phenotypes. If pathway elongation is the source of references for compound identification and the method of identification phenotypic differences, then the dependence between for each of the species’ metabolic networks. (c) Appendix S3: Module assignments in the avian subset of the global carotenoid metabolic downstream and upstream elements imposes a clear se- network. This appendix contains the module assignments for each of the quential order to phenotypic diversification along the compounds in the global avian carotenoid metabolic network. The pathway, resulting in stronger historical associations number of the module corresponds to the partitioned regions in Fig. 2. across species’ networks. The incorporation or loss of (PDF 1501 kb) Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 15 of 17 36

Additional file 2: Species’ binary metabolic networks. This appendix 11. Rice SH. The evolution of developmental interactions: Epistasis, contains binary metabolic networks for each of the species included in canalization, and integration. In: Wolf JB, Brodie III ED, Wade MJ, the study. (XLSX 123 kb) editors. Epistasis and the evolutionary process. New York: Oxford University Press; 2001. p. 82–98. Additional file 3: Majority rule consensus phylogeny of species 12. Alberch P. From genes to phenotype: Dynamical systems and evolvability. included in the study. This appendix contains the Newick tree format of Genetica. 1991;84:5–11. the majority rule consensus phylogeny visually presented in Fig. 4, 5, 6, 7 13. Arthur W. Developmental drive: An important determinant of the direction and 8. The tree is based on 1,000 randomly sampled trees from the of phenotypic evolution. Evol Dev. 2001;3:271–8. Hackett All Species pseudo-posterior distribution downloaded from bird- 14. Forgacs G, Newman SA. Biological physics of the developing embryo. tree.org that is based on Jetz et al. 2012. (TXT 20 kb) Cambridge: Cambridge University Press; 2005. 15. Whyte LL. Internal factors in evolution. New York: George Braziller; 1965. Abbreviation 16. Bloom JD, Labthavikul ST, Otey CR, Arnold FH. Protein stability promotes – MYA, million years ago evolvability. Proc Natl Acad Sci U S A. 2006;103:5869 74. 17. Bridgham JT, Ortlund EA, Thornton JW. An epistatic ratchet constrains the direction of glucocorticoid receptor evolution. Nature. 2009;461:515–9. Acknowledgements 18. Harms MJ, Thornton JW. Historical contingency and its biophysical basis in We thank V. Belloni, V. Farrar and J. Andrews for help with the data glucocorticoid receptor evolution. Nature. 2014;512:203–7. collection, and R. Duckworth, M. Sanderson, D. Higginson, A. Potticary, C. 19. Newman SA. Physico-genetic determinants in the evolution of development. Gurguis, G. Semenov and three anonymous reviewers for thorough Science. 2012;338:217–9. comments on previous versions and helpful suggestions. 20. Pagel M, Pomiankowski A. Evolutionary genomics and proteinomics. Sunderland: Sinauer Associates; 2008. Funding 21. Povolotskaya IS, Kondrashov FA. Sequence space and the ongoing This work was supported by the David and Lucille Packard Foundation, expansion of the protein universe. Nature. 2010;465:922–7. Amherst College graduate fellowships, and the University of Arizona Open 22. Wagner A. The molecular origins of evolutionary innovations. Trends Genet. Access Publishing Fund. 2011;27:397–410. 23. Fraser HB, Hirsh AE, Steinmetz LM, Scharfe C, Feldman MW. Evolutionary Availability of data and material rate in the protein interaction network. Science. 2002;296:750–2. The datasets supporting the results of this article are available as additional 24. Light S, Kraulis P, Elofsson A. Preferential attachment in the evolution of files (Additional files 1, 2 and 3). metabolic networks. BMC Genomics. 2005;6:159. 25. Liu WC, Lin WH, Davis AJ, Jordán F, Yang HT, Hwang MJ. A network Authors’ contributions perspective on the topological importance of enzymes and their ESM designed the study. ESM and AVB analyzed the data. ESM wrote the phylogenetic conservation. BMC Bioinformatics. 2007;8:121. manuscript with help from AVB. Both authors have read and approved the 26. Yamada T, Bork P. Evolution of biomolecular networks-lessons from final manuscript. metabolic and protein interactions. Nat Rev Mol Cell Biol. 2009;10:791–803. 27. Zhao J, Ding G-H, Tao L, Yu H, Yu Z-H, Luo J-H, et al. Modular co-evolution Competing interests of metabolic networks. BMC Bioinformatics. 2007;8:311. The authors declare that they have no competing interests. 28. Maslov S, Krishna S, Pang TY, Sneppen K. Toolbox model of evolution of prokaryotic metabolic networks and their regulation. Proc Natl Acad Sci U S A. Consent for publication 2009;106:9743–8. Not applicable. 29. Banerjee A. Structural distance and evolutionary relationship of networks. BioSyst. 2011;107:186–96. Ethics approval and consent to participate 30. Borenstein E, Kupiec M, Feldman MW, Ruppin E. Large-scale reconstruction Not applicable. and phylogenetic analysis of metabolic environments. Proc Natl Acad Sci U S A. 2008;105:14482–7. Received: 15 February 2016 Accepted: 1 August 2016 31. Ebenhöh O, Handorf T, Kahn D. Evolutionary changes of metabolic networks and their biosynthetic capacities. IEE P Syst Biol. 2006;153:354–8. 32. Mithani A, Hein J, Preston GM. Comparative analysis of metabolic networks References provides insight into the evolution of plant pathogenic and non 1. Gavrilets S. Fitness landscapes and the origin of species. Princeton: pathogenic lifestyles in Pseudomonas. Mol Biol Evol. 2011;28:483–99. Princeton University Press; 2004. 33. Navlakha S, Kingsford C. Network archaeology: uncovering ancient networks 2. Gerhart J, Kirschner M. The theory of facilitated variation. Proc Natl Acad Sci from present-day interactions. PLoS Comp Biol. 2011;7, e1001119. U S A. 2007;104:8582–9. 34. Barabási A-L, Albert R. Emergence of scaling in random networks. Science. 3. Maynard SJ. Natural selection and the concept of a protein space. Nature. 1999;286:509–12. 1970;225:563–4. 35. Jeong H, Tombor B, Albert R, Oltvai ZN, Barabási A-L. The large-scale 4. Wagner GP. Homology, genes, and evolutionary innovation. Princeton: organization of metabolic networks. Nature. 2000;407:651–4. Princeton University Press; 2014. 36. Thieffry D, Huerta AM, Pérez-Rueda E, Collado-Vides J. From specific gene 5. Newman SA. The developmental genetic toolkit and the molecular regulation to genomic networks: a global analysis of transcriptional homology—analogy paradox. Biol Theory. 2006;1:12–6. regulation in Escherichia coli. Bioessays. 1998;20:433–40. 6. Badyaev AV, Walsh JB. Epigenetic processes and genetic architecture in 37. Barabási A-L. Luck or reason. Nature. 2012;489:507–8. character origination and evolution. In: Charmantier A, Garant D, Kruuk LEB, 38. Eisenberg E, Levanon EY. Preferential attachment in the protein network editors. Quantitative genetics in the wild. Oxford: Oxford University Press; evolution. Phys Rev Lett. 2003;91:138701. 2014. p. 177–89. 39. Bernhardsson S, Gerlee P, Lizana L. Structural correlations in bacterial 7. Bershtein S, Segal M, Bekerman R, Tokuriki N, Tawfik DS. Robustness-epistasis metabolic networks. BMC Evol Biol. 2011;11:20. link shapes the fitness landscape of a randomly drifting protein. Nature. 40. Hahn MW, Kern AD. Comparative genomics of centrality and essentiality 2006;444:929–32. in three Eukaryotic protein-interaction networks. Mol Biol Evol. 2005;22: 8. Breen MS, Kemena C, Vlasov PK, Notredame C, Kondrashov FA. Epistasis as 803–6. the primary factor in molecular evolution. Nature. 2012;490:535–8. 41. Xu K, Bezakova I, Bunimovich L, Yi SV. Path lengths in protein–protein 9. Gravner J, Pitman D, Gavrilets S. Percolation on fitness landscapes: effects of interaction networks and biological complexity. Proteomics. 2011;11:1857–67. correlation, phenotype, and incompatibilities. J Theor Biol. 2007;248:627–45. 42. Ramsay H, Rieseberg LH, Ritland K. The correlation of evolutionary rate 10. Poelwijk FJ, Kiviet DJ, Weinreich DM, Tans SJ. Empirical fitness landscapes with pathway position in plant terpenoid biosynthesis. Mol Biol Evol. reveal accessible evolutionary paths. Nature. 2007;445:383–6. 2009;26:1045–53. Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 16 of 17 37

43. Rausher MD, Miller RE, Tiffin P. Patterns of evolutionary rate variation 75. Assenov Y, Ramírez F, Schelhorn S-E, Lengauer T, Albrecht M. Computing among genes of the anthocyanin biosynthetic pathway. Mol Biol Evol. topological parameters of biological networks. Bioinformatics. 2008;24: 1999;16:266–74. 282–4. 44. Wright KM, Rausher MD. The evolution of control and distribution of 76. Doncheva NT, Assenov Y, Domingues FS, Albrecht M. Topological analysis adaptive mutations in a metabolic pathway. Genetics. 2010;184:483–502. and interactive visualization of biological networks and protein structures. 45. Hartwell LH, Hopfield JJ, Leibler S, Murray AW. From molecular to modular Nat Protoc. 2012;7:670–85. cell biology. Nature. 1999;402:C47–52. 77. McSweeney PJ. Randomnetworks. Version 1.0. 2008. [http://apps.cytoscape. 46. Ravasz E, Somera AL, Mongru DA, Oltvai ZN, Barabasi A-L. Hierarchical org/apps/randomnetworks] organization of modularity in metabolic networks. Science. 2002;297: 78. Ebenhöh O, Handorf T, Heinrich R. A cross species comparison of metabolic 1551–5. network functions. Genome Inform. 2005;16:203–13. 47. Badyaev A. Evolvability and robustness in color displays: Bridging the gap 79. Thomas DB, McGraw KJ, Butler MW, Carrano MT, Madden O, James HF. between theory and data. Evol Biol. 2007;34:61–71. Ancient origins and multiple appearances of carotenoid-pigmented feathers 48. Nagy L. Changing patterns of gene regulation in the evolution of arthropod in birds. Proc R Soc B. 2014;281:20140806. morphology. Am Zool. 1998;38:818–28. 80. Furnham N, Sillitoe I, Holliday GL, Cuff AL, Laskowski RA, Orengo CA, et al. 49. Raff EC, Raff RA. Dissociability, modularity, evolvability. Evol Dev. 2000;2:235–7. Exploring the evolution of novel enzyme functions within structurally 50. von Dassow G, Munro E. Modularity in animal development and evolution: defined protein superfamilies. PLoS Comp Biol. 2012;8, e1002403. elements of a conceptual framework for EvoDevo. J Exp Zool. 1999;285: 81. Altincicek B, Kovacs JL, Gerardo NM. Horizontally transferred fungal 307–25. carotenoid genes in the two-spotted spider mite Tetranychus urticae. Biol 51. Wagner GP, Altenberg L. Perspective: Complex adaptations and the Lett. 2012;8:253–7. evolution of evolvability. Evolution. 1996;50:967–76. 82. Kreimer A, Borenstein E, Gophna U, Ruppin E. The evolution of modularity 52. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display in bacterial metabolic networks. Proc Natl Acad Sci U S A. 2008;105:6976–81. of genome-wide expression patterns. Proc Natl Acad Sci U S A. 1998;95: 83. Moran NA, Jarvik T. Lateral transfer of genes from fungi underlies carotenoid 14863–8. production in aphids. Science. 2010;328:624–7. 53. Halfon MS, Grad Y, Church GM, Michelson AM. Computation-based discovery 84. Nováková E, Moran NA. Diversification of genes for carotenoid biosynthesis of related transcriptional regulatory modules and motifs using an in aphids following an ancient transfer from a fungus. Mol Biol Evol. experimentally validated combinatorial model. Genome Res. 2002;12:1019–28. 2012;29:313–23. 54. IhmelsJ,FriedlanderG,BergmannS,SarigO,ZivY,BarkaiN.Revealing 85. Sloan DB, Moran NA. Endosymbiotic bacteria as a source of carotenoids in modular organization in the yeast transcriptional network. Nat Genet. whiteflies. Biol Lett. 2012;8:986–9. 2002;31:370–77. 86. Kelley BP, Sharan R, Karp RM, Sittler T, Root DE, Stockwell BR, et al. 55. Niehrs C, Pollet N. Synexpression groups in eukaryotes. Nature. 1999;402:483–7. Conserved pathways within bacteria and yeast as revealed by global 56. Campillos M, von Mering C, Jensen LJ, Bork P. Identification and analysis of protein network alignment. Proc Natl Acad Sci U S A. 2003;100: evolutionarily cohesive functional modules in protein networks. Genome 11394–399. Res. 2006;16:374–82. 87. Kondrashov FA. Gene duplication as a mechanism of genomic adaptation 57. Chen Y, Dokholyan NV. The coordinated evolution of yeast proteins is to a changing environment. Proc R Soc B. 2012;279:5048–57. constrained by functional modularity. Trends Genet. 2006;22:416–9. 88. Zhang G, Li C, Li Q, Li B, Larkin DM, Lee C, et al. Comparative genomics 58. Pál C, Papp B, Lercher MJ. Adaptive evolution of bacterial metabolic reveals insights into avian genome evolution and adaptation. Science. networks by horizontal gene transfer. Nat Genet. 2005;37:1372–5. 2014;346:1311–20. 59. Wagner A. Evolutionary constraints permeate large metabolic networks. 89. Walsh N, Dale J, McGraw KJ, Pointer MA, Mundy NI. Candidate genes for BMC Evol Biol. 2009;9:231. carotenoid coloration in vertebrates and their expression profiles in the 60. Klassen JL. Phylogenetic and evolutionary patterns in microbial carotenoid carotenoid-containing plumage and bill of a wild bird. Proc R Soc B. biosynthesis are revealed by comparative genomics. PLoS One. 2010;5, e11257. 2012;279:58–66. 61. Umeno D, Tobias AV, Arnold FH. Diversifying carotenoid biosynthetic 90. Hudon J, Anciães M, Bertacche V, Stradi R. Plumage carotenoids of the pin- pathways by directed evolution. Microbiol Mol Biol Rev. 2005;69:51–78. tailed manakin (Ilicura militaris): Evidence for the endogenous production of 62. Britton G, Liaaen-Jensen S, Pfander H, editors. Carotenoids. Boston: rhodoxanthin from a colour variant. Comp Biochem Physiol B Biochem Mol Birkhäuser Verlag; 2004. Biol. 2007;147:402–11. 63. Schmidt K, Connor A, Britton G. Analysis of pigments: carotenoids and related 91. Prum RO, LaFountain AM, Berro J, Stoddard MC, Frank HA. Molecular diversity, polyenes. In: Goodfellow M, O'Donnell AG, editors. Chemical methods in metabolic transformation, and evolution of carotenoid feather pigments in prokaryotic systematics. Chichester: John Wiley & Sons; 1994. p. 403–61. cotingas (Aves: Cotingidae). J Comp Physiol B. 2012;182:1095–116. 64. Brush AH. Metabolism of carotenoid-pigments in birds. FASEB J. 1990;4:2969–77. 92. Prum R, LaFountain A, Berg C, Tauber M, Frank H. Mechanism of carotenoid 65. McGraw KJ. The mechanics of carotenoid coloration in birds. In: Hill GE, coloration in the brightly colored plumages of broadbills (Eurylaimidae). J McGraw KJ, editors. Bird coloration volume 1: Mechanisms and Comp Physiol B. 2014;184:651–72. measurements. Cambridge: Harvard University Press; 2006. p. 177–242. 93. Morrison ES, Badyaev AV. The landscape of evolution: Reconciling structural 66. Badyaev A, Morrison E, Belloni V, Sanderson M. Tradeoff between and dynamic properties of metabolic networks in adaptive diversifications. robustness and elaboration in carotenoid networks produces cycles of avian Integr Comp Biol. 2016;56:235-46. color diversification. Biol Direct. 2015;10:45. 94. Bongaerts GP, Vliegenthart JS. Effect of aminoglycoside concentration on 67. Jaccard P. The distribution of the flora in the alpine zone. New Phytol. reaction rates of aminoglycoside-modifying enzymes. Antimicrob Agents 1912;11:37–50. Chemother. 1988;32:740–6. 68. Rodrigues JFM, Wagner A. Genotype networks, innovation, and robustness 95. Matsuno R, Nakanishi K, Ohnishi M, Hiromi K, Kamikubo T. Threshold in a in sulfur metabolism. BMC Syst Biol. 2011;5:39. single enzyme reaction system: reaction of maltose catalyzed by 69. Maddison WP, Maddison DR. Mesquite: a modular system for evolutionary saccharifying α-Amylase from B. Subtilis. J Biochem. 1978;83:859–62. analysis. Version 3.03. 2015. [http://mesquiteproject.wikispaces.com/] 96. Albert R, Jeong H, Barabási A-L. Error and attack tolerance of complex 70. Midford PE, Garland T, Jr., Maddison WP. PDAP package of Mesquite. networks. Nature. 2000;406:378–82. Version 1.16. 2011. [http://mesquiteproject.org/pdap_mesquite/]. 97. Jeong H, Mason SP, Barabási A-L, Oltvai ZN. Lethality and centrality in 71. Guimerà R, Amaral LAN. Functional cartography of complex metabolic protein networks. Nature. 2001;411:41–2. networks. Nature. 2005;433:895–900. 98. Schmidt S, Sunyaev S, Bork P, Dandekar T. Metabolites: a helping hand for 72. Guimerà R, Amaral LAN. Cartography of complex networks: Modules and pathway evolution? Trends Biochem Sci. 2003;28:336–41. universal roles. J Stat Mech Theor Exp. 2005;2005:P02001-1-13. 99. Aris-Brosou S. Determinants of adaptive evolution at the molecular 73. Harary F. Graph theory. Reading: Addison-Wesley; 1969. level: The extended complexity hypothesis. Mol Biol Evol. 2005;22: 74. Smoot ME, Ono K, Ruscheinski J, Wang P-L, Ideker T. Cytoscape 2.8: New 200–9. features for data integration and network visualization. Bioinformatics. 100. Badyaev AV. “Homeostatic hitchhiking”: A mechanism for the evolutionary 2011;27:431–2. retention of complex adaptations. Integr Comp Biol. 2013;53:913–22. Morrison and Badyaev BMC Evolutionary Biology (2016) 16:168 Page 17 of 17 38

101. Kauffman S, Levin S. Towards a general theory of adaptive walks on rugged landscapes. J Theor Biol. 1987;128:11–45. 102. Heijnen JJ, van Gulik WM, Shimizu H, Stephanopoulos G. Metabolic flux control analysis of branch points: an improved approach to obtain flux control coefficients from large perturbation data. Metab Eng. 2004;6: 391–400. 103. LaPorte DC, Walsh K, Koshland DE. The branch point effect. Ultrasensitivity and subsensitivity to metabolic control. J Biol Chem. 1984;259:14068–75. 104. Pritchard L, Kell DB. Schemes of flux control in a model of Saccharomyces cerevisiae glycolysis. Eur J Biochem. 2002;269:3894–904. 105. Rausher MD. The evolution of genes in branched metaoblic pathways. Evolution. 2013;67:34–48. 106. Flowers J, Sezgin E, Kumagai S, Duvernell D, Matzkin L, Schmidt P, et al. Adaptive evolution of metabolic pathways in Drosophila. Mol Biol Evol. 2007;24:1347–54. 107. Higginson DM, Belloni V, Davis SN, Morrison ES, Andrews JE, Badyaev AV. Evolution of long-term coloration trends with biochemically unstable ingredients. Proc R Soc B. 2016;283:20160403. 108. Wang Z, Zhang J. In search of the biological significance of modular structures in protein networks. PLoS Comp Biol. 2007;3, e107. 109. Fraser HB. Modularity and evolutionary constraint on proteins. Nat Genet. 2005;37:351–2. 110. Fox DL. Metabolic fractionation, storage and display of carotenoid pigments by flamingoes. Comp Biochem Physiol. 1962;6:1–24. 111. Fox DL, Smith VE, Wolfson AA. Carotenoid selectivity in blood and feathers of lesser (African), Chilean and greater (European) flamingos. Comp Biochem Physiol. 1967;23:225–32. 112. McGraw KJ, Beebee MD, Hill GE, Parker RS. Lutein-based plumage coloration in songbirds is a consequence of selective pigment incorporation into feathers. Comp Biochem Physiol B Biochem Mol Biol. 2003;135:689–96. 113. Gerhart J, Kirschner M. Evolution and evolvability. In: Cells, embryos, and evolution. Malden: Blackwell Science; 1997. p. 580–614. 114. Reid RGB. Biological emergence: Evolution by natural experiment. Cambridge: MIT Press; 2007. 115. Yang AS. Modularity, evolvability, and adaptive radiations: a comparison of the hemi- and holometabolous insects. Evol Dev. 2001;3:59–72. 116. Jetz W, Thomas GH, Joy JB, Hartmann K, Mooers AO. The global diversity of birds in space and time. Nature. 2012;491:444–8.

Submit your next manuscript to BioMed Central and we will help you at every step:

• We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript at www.biomedcentral.com/submit 39

Appendix S1. Confirmed enzymatic reactions in "avian space" of global carotenoid biosynthesis network in bacteria, plants, and animals. Only nodes documented in birds (Appendix S2) are included. Bacteria Algae Plants Animals

CAROTENOID Origin node Reaction NODE PATH ENZ ISO References NODE PATH ENZ ISO References NODE PATH ENZ ISO References NODE PATH ENZ ISO References lutein 1?...70 Y...63, 70, 74 Y...63, 70, 74 Y . . . 8, 9, 32, 33, 34, 37, 38, 39, 40, 41, 46, 86, 34, 91 1167 . N N N . .NNN. .NNN. ? ? Y 46, 34, 93 1170 . N N N . .NNN. .NNN. . Y Y Y 8, 9, 33, 34, 42, 34, 107, 118 1152 . N N N . .NNN. .?NY. . Y N Y 34, 91 1116 . N N N . .NNN. .NNN. . Y Y N 8, 10, 11, 33, 34, 39, 46, 34, 91, 93 115.NNN. .NNN. . NNN. .YYN37, 41 116.NNN. .YYN105. NNN. .YYN32, 38, 39, 40 118.NNN. .NNN. . NNN. .YYN8, 10, 30, 51, 104, 123 119 .N NN. .NNN. . NNN. .??N8, 30, 104 1113 . N N N . .NNN. .NNN. . Y Y N 9, 39, 42, 118 1151 . N N N . .NNN. .NNN. . Y Y n 9, 12, 39, 42, 118 1145 . N N N . .NNN. .NNN. .NNN. 1146 .Y YN67 .NNN. . YYN67 ...N. (3R, 3'R) zeaxanthin 2Y...2,14, 22, 62 Y . . . 26, 62, 63, 70 Y . . . 62, 44, 63, 74 Y . . . 9,10, 11, 16, 29, 32, 38, 39, 44, 46, 54, 86 2269 . N N N . .NNN. . ? ? N 70 . Y Y N 46 2270 . N N N . .NNN. .NNN. . ? Y Y 9, 33, 34, 42, 34 2268 . N N N . .NNN. .NNN. . Y Y N 46, 54, 86 2267 .N NN. .??N. . NNN. .?YY33, 44, 46, 54 2220 . Y Y N 97 .NNN. .NNN. . Y Y N 32, 35, 38, 39, 40, 71, 82 2216 . N N N . .NNN. .NNN. . Y Y N 9, 10, 32, 42, 46, 34 2221 . N N N . . Y Y N 26, 63 . Y Y N 63 . Y Y N 15 2231 .N NN. .NNN. . NNN. .YYN9, 12, 42, 119, 120 2265 . N N N . .NNN. .NNN. . Y Y N 29, 46, 71 2232 .Y YN2, 14 . Y Y N 63 .NNN. . Y Y Y 9, 42, 70, 114 224.NNN. .NNN. . NNN. .YYN54, 121 2218 ? ? ? N . .??N. .NNN. . ? ? N 8, 111, 119 β‐carotene 3Y...14, 22, 23, 59, 62, 75 Y . . . 60, 61 Y . . . 62, 63, 70 Y . . . 3, 9, 16, 18, 28, 56, 57, 65, 64, 69, 71, 77, 78, 93 334.YYN2, 14, 22, 23, 59, 62, 75 . Y Y N 20, 61, 63 . Y Y N 63, 96 Y Y N 57, 65, 69, 93 3335 . Y Y N 2, 14, 22, 23, 59, 62, 75 . Y Y N 20, 61, 63 . N N N . .YY N 1, 12, 9, 57, 65, 93 3341 YYN 81 YYN70, 98 . Y Y N 96 Y Y N 5, 16, 18, 28, 64, 77, 78, 83 β‐cryptoxanthin 4Y...2,14, 22,23, 59, 62, 75 Y . . . 22, 61, 62, 63, 70 Y . . . 62, 70 Y . . . 9, 10, 65, 69, 56, 57, 71 442YYN2,14, 22, 23, 59, 62 . Y Y N 22, 62 . Y Y N 62, 63 . Y Y N 69 4436 . Y Y N 2,14, 22, 23, 59, 62 . Y Y N 22, 62 .NNN. . Y Y N 57, 93 4430 . Y Y N 97 .??N. .NNN. . Y Y N 32, 39, 40 anhydrolutein 5N....N....N....Y...32, 37, 41 7,8‐ dihydrolutein 6N....Y...105N....Y...32, 39, 40, 82 667.NNN. ?NY105. NNY. .YNY32, 39, 40, 82 66109 . N N N . .NNN. .NNN. . ? ? N 201 9‐Z‐7,8‐dihydrolutein 7N....?....N....Y...32, 39, 40, 82 776.NNN. .?NY105. NNY. .YNY32, 39, 40, 82 canary xanthophyll A 8N....N....N....Y...8, 9, 10, 11, 17, 32, 33, 46, 71, 91, 34, 104 8810 . N N N . .NNN. .NNN. . Y Y N 9, 11, 32, 33, 39, 34 8816 . N N N . .NNN. .NNN. . Y Y N 34, 91 889.NNN. .NNN. . NNN. .YYN9, 10, 11, 32, 33, 39, 34, 104, 123 88110 . N N N . .NNN. .NNN. . ? ? N 202 canary xanthophyll B 9N....N....N...74Y...8, 9, 11, 10, 17, 32, 33, 46, 71, 104 9917 . N N N . .NNN. .NNN. . ? ? N 36 998.NNN. .NNN. . NNN. .YYN9, 11 (3S, 6S,3'S, 6'S) tunaxanthin A 10 N.... N.... ?.... Y . . . 8, 11, 32, 33, 34, 39, 71, 34 10 1011 . N N N . .NNN. .?... Y N Y 8, 11, 39, 71, 108, 111 (3R, 6R,3'R, 6'R) tunaxanthin F 11 N.... N.... Y . . . 47, 48 Y . . . 8, 91 11 1166 . N N N . .NNN. .NNN. . ? ? N 8 11 1165 . N N N . .NNN. .NNN. . ? ? N 8 α‐doradexanthin 12 N.... Y . . . 70 N.... Y . . . 39, 42, 45, 70 12 1214 .N NN. .NNN. . NNN. .NNN130 12 1215 .N NN. .NNN. . NNN. .YYN45, 111 12 1232 . N N N . .NNN. .NNN. . Y Y N 71, 117 12 12100 . N N N . .NNN. .NNN. . ? ? N 200 (3S,4R,3'R,6'R) 4‐hydroxylutein 13 N.... N.... N.... Y . . . 9, 39, 42 13 1351 . N N N . .NNN. .NNN. . Y N N 42, 111 13 1314 . N N N . .NNN. .NNN. . Y ? N 9, 12, 39, 42, 111, 118, 130 fritschiellaxanthin 14 N.... Y . . . 70 N.... Y . . . 9, 39, 42, 70 14 1412 .N NN. .?N?. . ?NY. .NNN. 14 1415 .N NN. .NNN. . NNN. .Y?N9, 111 papilioerythrinone 15 N.... Y.... N.... Y . . . 45, 71 15 15108 . N N N . .NNN. .NNN. . ? ? N 201 3'‐dehydrolutein 16 N.... ?.. . . ?.... Y . . . 8, 9, 10, 11, 33, 34, 35, 46, 71, 34, 91, 93 16 161. N NN. .?... . NNN. . YYN9, 11, 34, 91 16 168. N NN. .NNN. . NNN74 NYN10, 11, 34, 92 40

16 169. N NN. .NNN. . NNN. . YYN10, 11, 34 16 1670 . N N N . .NNN. .NNN. . Y Y N 33, 34, 91 16 16109 . N N N . .NNN. .NNN. . ? ? N 201 piprixanthin 17 N.... N.... N.... Y . . . 36 17 1771 NNN. .NNN. .NNN . ? ? N 36 17 1718 YNN. .?NN. .NNN. Y ? N 36 rhodoxanthin 18 Y . . . 70 Y.... Y . . . 70 Y . . . 36 7,8,7',8'‐tetrahydrozeaxanthin 19 N.... N.... N.... Y . . . 32, 35, 40, 82 7,8‐dihydrozeaxanthin 20 Y Y . . 59, 97 N.... N.... Y . . . 32, 38, 39, 40, 71 20 2019 . N N N . .NNN. .NNN. . Y Y N 32, 35, 40, 82 antheraxanthin 21 N.... Y . . . 26, 63, 70 Y . . . 62, 63, 70, 74 Y . . . 9, 71 21 2122 . N N N . . Y Y N 63 . Y Y N 63, 74 .NNN. 21 212. N NN. .YYN63 . YYN63, 74, 75 . N N N . 21 2176 . N N N . .??N. . Y Y N 71, 115 .NNN. violaxanthin 22 NN.N. Y . . . 26, 63, 70 Y . . . 62, 63, 70, 74 Y . . . 71 22 2223 . N N N . . Y Y N 26, 63, 70 . Y Y N 62, 63, 70, 74 NN.. 22 2221 . N N N . . Y Y N 26, 63, 70 . Y Y N 62, 63, 70, 74 NN.. neoxanthin 23 N.... Y . . . 26, 63 Y . . . 62, 63, 70, 74, 105 Y . . . 49, 71 23 2324 . N N N . .NNN. .??N. . Y Y N 49 neochrome 24 N.... Y . . . 105, 112 Y . . . 70 Y . . . 49 idoxanthin 25 N.... ? . . . 111 N.... Y . . . 9,11, 13, 15, 16, 24, 28, 71 25 2534 . N N N . .NNN. .NNN. . Y ? N 120, 126, 127 25 2532 . N N N . .NNN. ..NN. . Y Y N 11, 13, 15, 16, 35 25 2574 . N N N . .NNN. .NNN. . Y Y N 11, 71 fucoxanthin 26 N.... Y . . . 70 N.... Y . . . 5, 51, 88, 71 26 2627 . N . N . .NN.. .NNN. Y Y N 5, 51, 88, 122 fucoxanthinol 27 N.... Y . . . 70, 124 NNNN. Y . . . 5, 51, 88, 71 27 2728 .N NN. .NNN. . NNN. .YYN5, 51 27 2729 . N N N . .N . .NNN. . ? ? N 122 amarouciaxanthin 28 N.... N.... N.... Y . . . 51 28 2829 .N NN. .NNN. . NNN. YYN5, 51, 88 paracentrone 29 N.... ? . . . 112 N.... Y . . . 5, 88 7,8 dihydro β‐cryptoxanthin 30 Y . . . 97 ? . . . 97 ? . . . 97 Y. 32, 39, 40 4‐hydroxyzeaxanthin 31 N.... N.... NNNN. Y . . . 8, 9, 16, 11, 71 31 3132 . N N N . .NNN. .NNN. . Y ? N 9, 12, 42 31 312. N NN. .NNN. . NNN. . YYN8, 9, 11 31 3174 .N NN. .NNN. . NNN. .YYN42, 120 adonixanthin 32 Y . . . 2, 14, 22, 59 Y . . . 20, 61, 63 Y . . . 80, 96, 125 Y. ..9, 11, 15, 16, 28, 29, 32, 35, 42 32 3234 . Y Y N 2, 14, 22, 59 . Y Y N 20, 61, 63 . N N N . . Y Y N 9, 12, 35, 104, 118 32 322. N NN. .NNN. . NNN. . YYN9, 16, 35, 39 32 3225 . N N N . .NNN. .NNN. . Y Y N 9, 42, 70 32 3231 . N N N . .NNN. .NNN. . Y Y N 8, 9, 11, 16 32 32102 . N N N . .NNN. .NNN. . ? ? N 200 13 cis‐(3R, 3'R) astaxanthin 33 Y . . . 73 Y . . . 84 ?.... Y . . . 15, 71, 85 33 3334 . Y N Y 73 . Y N Y 84 .?NY. . Y N Y 15, 27, 71, 85 (3'R, 3R) astaxanthin 34 Y . . . 2, 14, 22, 59, 102 Y . . . 20, 63, 70 Y . . . 74, 96 Y . . . 8, 11, 13, 15, 16, 27, 29, 35, 45, 57, 71 34 3432 . N N N . .NNN. .NNN. . Y Y N 8, 9, 11, 15, 29 34 3425 . N N N . NNNN. .NNN. . Y Y N 8, 11, 13, 15, 27 34 3475 . Y N Y 73 . Y N Y 84 .?NY. . Y N Y 15, 71, 85 34 3438 . N N N . .NNN. .NNN. . Y Y N 121 34 3433 . Y Y Y 73 . Y N Y 84 .?NY. . Y N Y 15, 27, 71, 85 34 34103 . N N N . .NNN. .NNN. . ? ? N 200 echinenone 35 Y . . . 2, 14, 22, 23, 25, 59, 62, 63, 75, 100 Y . . . 61, 63, 70 Y . . . 80, 96 Y . . . 15, 16, 28, 64, 71, 77, 78, 83, 93 35 353. N NN. .NNN. . NNN. . YYN4, 15, 16 35 3539 . Y Y N 81, 100 .YYN. .NNN. . Y Y N 6, 28, 64, 77, 78 35 3537 . Y Y N 2, 14, 22, 59, 62, 102 . Y Y N 62, 63 .NNN. . Y Y N 2, 9, 57, 64, 93 35 3536 . Y Y N 2, 14, 22, 23, 59, 62, 63, 102 . Y Y N 62, 63 . Y Y N 80, 96, 125 . Y Y N 57, 71, 101, 93 35 3541 .N NN. .NNN. . NNN. .YYN16 3'‐hydroxyechinenone 36 Y . . . 2, 14, 22, 59, 63, 102 Y . . . 20, 63, 70 Y . . . 80, 96 Y . . . 15, 17, 57, 71, 93 36 3632 . Y Y N 2, 14, 22, 59, 102 . Y Y N 20, 63 . Y Y . 80, 96, 125 . Y ? N 57, 71, 93 36 3638 . Y Y N 2,14, 59, 102 . Y Y N 20, 63 .NNN. .Y YN57, 71, 93 36 364 .N NN. .NNN. . NNN. .YYN9, 15 canthaxanthin 37 Y . . . 2,14,22, 25, 59, 63, 81 Y . . . 61, 63, 70 Y . . . . Y . . . 3, 15, 16, 28, 57, 64, 77, 78, 93 37 3735 . N N N . .NNN. .NNN. . Y Y N 4, 8, 15, 121 37 3738 . Y Y N 2, 14, 22, 59, 63, 102 . Y Y N 62, 63 . Y Y N 80, 96 . Y Y N 9, 17, 57, 93, 101, 121 37 3740 . N N N . .NNN. .NNN. . Y Y N 8, 16 37 3739 . N N N . .NNN. .NNN. . Y Y N 15, 16, 31, 82 adonirubin 38 Y . . . 2,14, 22, 59, 102 Y . . . 20, 63, 70 Y . . . 96 Y . . . 1, 3, 6, 7, 17, 57, 71, 101 38 3834 . Y Y N 2, 14, 22, 59, 102 . Y Y N 63 . Y Y N 74, 80, 96, 125 . Y YN9, 17, 57, 93, 101, 71, 77, 103 38 3836 .N NN. .NNN. . NNN. .YYN15 38 3837 . Y Y N 89 . Y Y N 62, 63 . Y Y N 80, 96 . Y Y N 89, 75 38 38107 . N N N . .NNN. .NNN. . ? ? N 200 4‐hydroxy‐echinenone 39 Y . . . 81, 100 Y . . . 70 N.... Y . . . 6, 16, 30, 31, 71, 77, 78, 82, 83, 99 39 3940 .N NN. .NNN. . NNN. .YYN15, 16, 31 39 3935 .N NN. .NNN. . NNN. .YYN16 41

39 3937 . Y ? N 81, 100 .??N. . NNN. . YYN6, 28, 35, 64, 77, 78, 83 isozeaxanthin 40 Y . . . 81 Y . . . 70, 98 Y . . . 80 Y . . . 8, 16, 28, 31, 53, 71, 80, 81 40 4041 . N N N . .NN.. .NNN. . Y Y N 8, 15, 16 40 4060 . N N N . .NN.. . Y Y N 80 .NNN. 40 4037 .Y YN81 .YYN70, 98 .NNN. . Y Y N 28, 64 40 4039 .Y YN81 .??N70. NNN. . NNN. β‐isocryptoxanthin 41 Y . . . 81 Y . . . 79, 98 Y . . . 70 Y . . . 5, 9, 18, 16, 31, 71, 77, 78, 83 41 413 .N NN. .N.N. . NNN. NYN16 41 4140 . Y Y N 81 YYN70, 98 . Y Y N 96 . Y Y N 28, 71 41 4135 . ? Y N 81 .??N70, 98 .NNN. . Y Y N 18, 28, 64, 77, 78, 83 α‐carotene 42 Y N . . 62, 63, 70, 76 Y . . . 62, 63, 70, 74 Y . . . 58, 59, 63, 74, 76 Y . . . 52, 53, 71 42 4246 . N N N . .NNN. . Y Y N 131 .??N. 42 4245 . ? Y N 59 . Y Y N 63 . Y Y N 63 .NN.. 42 4243 . N N N . .NNN. . ? ? N 96, 125 . Y ? . 72 α‐isocryptoxanthin 43 ?.... ?.... ?.... Y . . . 52, 71, 72 43 4344 . N N N . .NNN. .NNN. . Y Y N 52, 72 phoenicopterone 44 N.... Y . . . 70, 113 N.... Y . . . 53, 71, 72 zeinoxanthin 45 N.... Y . . . 58, 59, 63 Y . . . 58, 59, 63 N . . . . 45 451. N NN. .YYN63 . YYN63 . NNN. 45 4542 .N NN. .NNN. . NNN. .NNN. α‐cryptoxanthin 46 N.... Y . . . 63, 70 Y . . . 63, 70 Y . . . 46, 71, 106 46 461. Y YN67 .NNN. . YYN67 . . .N. 46 4642 .N NN. .NNN. . YYN131.??N8 rubixanthin 47 Y . . . 70, 94 Y . . . 70, 95 Y . . . 70 Y . . . 55, 57, 93, 71 47 4748 . Y Y N . .?YN. .??N. . Y Y N 57 47 4749 .? ?Y. .?YY. . YNY55 .NNN55, 104, 128 47 472. Y ?N129 ...... 47 474. Y ?N129 ...... 4‐oxo‐rubixanthin 48 ?.... ?.... ?.... Y . . . 57, 93 48 4850 . ? ? ? . .???. .NNN. .NNN. gazaniaxanthin 49 ?.... ?.... Y . . . 70 Y . . . 71, 104 49 4950 . ? ? N . .??N. .??.. . Y Y N 93, 104 49 4947 . ? ? Y . .??Y. Y Y Y 55 . N N N 55, 104, 128 4‐oxo‐gazaniaxanthin 50 ?.... ?.... ?.... Y . . . 57, 93, 104 50 5048 . ? ? ? . ???. NNN. . ? ? Y 93 (3S,4R,3'S,6'R) 4‐hydroxylutein 51 N.... N.... N.... Y . . . 9, 39, 42 51 5113 . N N N . .NNN. .NNN. . Y N N 42, 70, 111 51 5112 . N N N . .NNN. .NNN. . Y Y N 9, 12, 39, 42, 118 cis lutein 52 N.... ?.... ?.... Y . . . 34, 91 52 521. N NN. .?NN. . ?NY. . YNY34, 91, 93 3,4, 3',4'‐tetradehydroisozeanthin 60 N.... N.... Y . . . 80 N.... 60 6037 . N N N . .NNN. . Y N N 80 .NNN. γ‐carotene 61 Y . . . 62, 73, 94 Y . . . 63 Y . . . 58, 62, 63, 70 Y . . . 71, 72 61 613. Y YN62, 73 .YY 63 Y Y N 62, 63, 74 NNN. 61 6147 Y Y N 73, 94 ??N. . ? Y N 58, 70 ??N. 61 6142 . N N N 62 . Y Y . 63 . Y Y N 62, 63, 70 . N N N . δ‐carotene 64 Y . . . 62 Y . . . 63, 70 Y . . . 63, 74 Y . . . 71 64 6465 . N N N 63, 76 . Y Y N 62, 63 . Y Y N 62, 63, 87 NNN. 64 6442 . Y Y N 62 . Y Y N 62, 63 . Y Y N 62, 63, 87 NNN. (6S, 6'S) ε,ε‐ carotene 65 N.... Y.. . 63, 70 Y . . . 48, 63, 74, 87 Y . . . 8, 39, 46, 71 65 6566 . N N N . .??N. . Y Y N 47 .??N. ε‐carotene 3 diol 66 N.... ?.... Y . . . 47 Y.... 66 6665 . N N N . .N.N. .NNN. . ? ? N 8 66 6611 . N N N . .NNN. . Y Y N 47 .??N. (3R, 3'S) meso‐zeaxanthin 67 N.... N.... N . . . 44 Y . . . 16, 33, 44, 46, 54, 86, 34, 93 67 672. N NN. .NNN. . NNN. . ??Y33, 44, 54 67 671 NN NN. . NNN. . NNN. . Y YY34, 93 galloxanthin 68 N.... N.... N.... Y . . . 29, 46, 54, 71 β‐apo‐2'‐carotenol 69 ? Y . . 111 ? . . . 111 N . . . 111 Y . . . 46 3'‐epilutein 70 N.... ? . . . 112 ?.... Y . . . 9, 11, 33, 34, 39, 71, 34 70 7016 . N N N . .NNN. .NNN. . ? ? N 33, 46, 34 70 708. N NN. .NNN. . NNN. . YYN33 70 701. N N.. .??N. . NNN. . ??Y33, 34, 71, 34 70 702 .N NN. .NNN. . NNN. .YYY33, 34, 34, 91 70 7012 . N N N . .NNN. .NNN. . ? ? N 93 70 7051 . N N N . .NNN. .NNN. . Y Y N 9, 12, 39, 42, 118 resonance stabilized form 71 ?.... ?.... N.... Y . . . 36 71 7118 . ? N N . .?NN. .NNN. . ? ? N 36 β‐carotene ‐3,4, 3', 4'‐ tetrol 74 N.... Y . . . 70 N.... Y . . . 11, 42, 71 74 7431 .N NN. .NNN. . NNN. .YYN11 74 7425 . N N N . .NNN. .NNN. . Y Y N 11, 42, 120 9‐cis ‐ (3S, 3'S) astaxanthin 75 ?.... Y . . . 84 ?.... Y . . . 15, 71, 85 75 7534 . Y N Y 73 . Y N Y 84 .?NY. . Y N Y 15, 27, 71, 85 42

eschscholtzxanthin 76 N.... N.... Y. 70 Y.... 76 7677 . N N N . .NNN. . Y Y N 111, 116 .NNN. eschscholtzxanthone 77 N.... N.... Y . . . 116 N.... 77 7718 . N N N . .NNN. . Y Y N 116 .NNN. xipholenin (Note 1) 100 N.... N.... N.... Y . . . 200, 203 100 100101 . N N N . .NNN. .NNN. . ? ? N 200, 203 2,3‐didehydro‐xipholenin (Note 2) 101 N.... N.... N.... Y . . . 200, 203 rupicolin (Note 3) 102 N.... N.... N.... Y . . . 200 3'‐hydroxy‐3‐methoxy‐canthaxanthin 103 N.... N.... N.... ? . . . 200 103 103104 . N N N . .NNN. .NNN. . ? ? N 200 pompadourin (Note 4) 104 N.... N.... N.... Y . . . 200, 203 104 104105 . N N N . .NNN. .NNN. . ? ? N 200, 203 2,3‐Didehydro‐pompadourin (Note 5) 105 N.... N.... N.... Y . . . 200, 203 105 105106 . N N N . .NNN. .NNN. . ? ? N 200, 203 cotingin (Note 6) 106 N.... N.... N.... Y . . . 200, 203 brittonxanthin (Note 7) 107 N.... N.... N.... Y . . . 200, 203 cymbirhynchin (Note 8) 108 N.... N.... N.... Y . . . 201 eurylaimin (Note 9) 109 N.... N.... N.... Y . . . 201 4‐hydroxy‐canary xanthophyll A 110 N.... N.... N.... Y . . . 202

Y: confirmed present, N: confirmed absent, ?: expected, but no experimental evidence NODE ‐‐ carotenoid compound (number refers to network) PATH ‐‐ reaction from origin node to derived node ENZ ‐‐ presence of enzyme ISO ‐‐ evidence of isomerization Notes: 1: 3‐Methoxy‐3'‐hydroxy‐β,ε‐carotene‐4‐one 2: 3'‐hydroxy‐3‐methoxy‐2,3‐didehydro‐β,β‐carotene‐4‐one 3: 3'‐hydroxy‐3‐methoxy‐β,β‐carotene‐4‐one 4: 3,3'‐Dimethoxy‐β,β‐carotene‐4,4'‐dione or 3,3'‐dimethoxy‐canthaxanthin 5: 3,3'‐Dimethoxy‐2,3‐didehydro‐β,β‐carotene‐4,4'dione 6: 3,3'‐Dimethoxy‐2,3,2',3'‐tetradehydro‐β,β‐carotene‐4,4'dione 7: 3‐methoxy‐β,β‐carotene‐4,4'dione or 3‐methoxy‐canthaxanthin 8: 2,3‐didehydro‐papilioerythrinone 9: 7,8‐dihydro‐3'‐dehydro‐lutein 43

Literature Sources for Appendix S1: carotenoids in rainbow trout, salmon and chicken. Pure Appl. Chem. 57: 685-692.

1. Davies, B. H., W. J. Hsu, and C. O. Chichester. 1970. The 16. Schiedt, K., M. Vecchi, E. Glinz, and T. Storebakken. 1988. mechanism of the conversion of beta-carotene into Metabolism of carotenoids in salmonids: metabolism of canthaxanthin by the brine shrimp, Artemia salina L. astaxanthin and canthaxanthin in the skin of atlantic salmon (Crustacea: Branchiopoda). Comp. Biochem Physiol. 33: 601- (Salmo salar, L.). Helv. Chim. Acta 71:887-896. 615. 17. Stradi, R., G. Celentano, E. Rossi, G. Rovati, and M. Pastore. 2. Fraser, P. D., S. Hiroshi, and M. Norihiko. 1998. Enzymic 1995. Carotenoids in bird plumage: I. The carotenoid pattern confirmation of reactions involved in routes to astaxanthin in a series of Palearctic Carduelinae. Comp. Biochem. Physiol. formation, elucidated using a direct substrate in vitro assay. 110:131 -143. Eur. J. Biochem. 252: 229-236. 18. Tsushima, M., T. Kawakami, and T. Matsuno. 1993. 3. Fox, D. L., A. A. Wolfson, and J. W. McBeth. 1969. Metabolism of carotenoids in sea-urchin Pseudocentrotus Metabolism of -carotene in the american flamingo, depressus. Comp. Biochem. Physiol. 106: 737 -741. Phoenicopterus rubber. Comp. Biochem. Physiol. 29:1223- 1229. 19. Wyss, A., G. Wirtz, W. Woggon, R. Brugger, M. Wyss, A. Friedlein, H. Bachmann, and W. Hunziker. 2000. Cloning and 4. Guillou, A., G. Choubert, T. Storebakken, J. De La Noe, and expression of beta,beta-carotene 15,15'-dioxygenase. Biochem S. Kaushik. 1989. Bioconversion pathway of astaxanthin into Biophys Res Commun. 271:334-336. retinol2 in mature rainbow trout (Salmo gairdneri Rich.). Comp. Biochem. Physiol. 94: 484-485. 20. Liu, B.-H. and Y.-K. Lee 1999. Composition and biosynthetic pathways of carotenoids in the astaxanthin-producing green 5. Hallenstvet, M., E. Pyberg, and S. Liaaen-Jensen. 1978. alga Chlorococcum sp. Biotechnology Letters 21(11): 1007- Animal carotenoids - XIV Carotenoids of Psammechinus 1010. miliaris (sea-urchin). Comp. Biochem. Physiol. 60: 173-175. 21. Fraser, P. D. and P. M. Bramley (2004). The biosynthesis and 6. Hata, M., and Hata, M. 1969. Carotenoid metabolism in nutritional uses of carotenoids. Progress in Lipid Research Artemia salina L. Comp. Biochem. Physiol. 29: 985-994. 43(3): 228-265.

7. Herring, P. J. 1968. The carotenoid pigments of Daphnia 22. Martin, J., E. Gudina, et al. 2008. Conversion of beta-carotene magna Straus. II. Aspects of pigmentary metabolism. Comp. into astaxanthin: two separate enzymes or a bifunctional Biochem. Physiol. 24: 205-221. hydroxylase-ketolase protein? Microbial Cell Factories 7(1): 3. 8. Katsuyama, M., and Matsuno,T. 1988. Carotenoid and vitamin A, and metabolism of carotenoids, -carotene, canthaxanthin, 23. Punginelli, C., A. Wilson, et al. 2009. Influence of zeaxanthin astaxanthin, zeaxanthin, lutein and tunaxanthin in tilapia and echinenone binding on the activity of the orange Tilapia nilotica. Comp. Biochem. Physiol. B 90: 131-139. carotenoid protein. Biochimica et Biophysica Acta - Bioenergetics 1787(4): 280-288. 9. Matsuno,T. 1991. Xanthophylls as precursors of retinoids. Pure Appl. Chem. 63: 81-88. 24. Aas, G. H., B. Bjerkeng, et al. 1997. Idoxanthin, a major carotenoid in the flesh of Arctic charr (Salvelinus alpinus) fed 10. Matsuno, T., T. Hirono, Y. Ikuno, T. Maoka, M. Shimizu, and diets containing astaxanthin. Aquaculture 150(1-2): 135-142. T. Komori. 1986. Isolation of three new carotenoids and proposed metabolic pathways of carotenoids in hen's egg yolk. 25. Schwartzel E.M. and J.J Cooney. 1970 Isolation and Comp. Biochem. Physiol. 84: 477-481. identification of echinenone from Micrococcus roseus. J Bacteriology. 104: 272-274. 11. Matsuno, T., M. Katsuyama, T. Maoka, T. Hirono, and T. Komori. 1985. Reductive metabolic pathways of carotenoids 26. Esteban, R., B. Martínez, et al. 2009. Carotenoid composition in fish (3S, 3'S)-astaxanthin to tunaxanthin A, B and C. Comp. in Rhodophyta: insights into xanthophyll regulation in Biochem. Physiol. 80:779-789. Corallina elongate. European Journal of Phycology 44(2): 221 - 230. 12. Matsuno,T., H. Matsutaka, and S. Nagata. 1981. Metabolism of lutein and zeaxanthin to ketocarotenoids in goldfish, 27. Bjerkeng B., Hatlen B. and M. Jobling 2000 Astaxanthin and Carassius auratus. Bull. Jap. Soc. Sci. Fish. 47: 605-611. its metabolites idoxanthin and crustaxanthinin flesh, skin, and gonads of sexually immature and maturingArctic charr 13. Miki,W., K. Yamaguchi, S. Konosu, and T. Watanabe. 1984. (Salvelinus alpinus (L.)). Comp Biochem Physiol 99(3): 395- Metabolism of dietary carotenoids in eggs of red sea bream. 404. Comp. Biochem. Physiol. 77: 665 -668. 28. Maoka, T. and T. Matsuno 1989. Metabolism of carotenoids in 14. Misawa, N., Y. Satomi, K. Kondo, A. Yokoyama, S. terrestrial snail Euhadra Callizona amaliae. Comparative Kajiwara, T. Saito, T. Ohtani, and W. Miki. 1995. Structure Biochemistry and Physiology Part B: Comparative and functional analysis of a marine bacterial carotenoid Biochemistry 92(1): 41-43. biosynthesis gene cluster and astaxanthin biosynthetic pathway proposed at the gene level. J Bacteriol. 177: 6575- 29. Schiedt, K., Bischof, S. and Glinz, E. 1991. Recent progress 6584. on carotenoid metabolism in animals. Pure Appl. Chem. 63: 89-100 15. Schiedt, K., F. J. Leuenberger, M. Vecchi, and E. Glinz. 1985. Absorption, retention and metabolic transformations of 44

30. Matsuno, T., K. Katagiri, et al. 1985. Novel reductive 44. Maoka, T., A. Arai, et al. 1986. The first isolation of metabolic pathways of 4-oxo-beta-end group in carotenoids enantiomeric and meso-zeaxanthin in nature. Comparative of the spindle shell Fusinus perplexus. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 83(1): 121-124. Biochemistry 81(4): 905-908. 45. Stradi, R., E. Pini, et al. 2001. Carotenoids in bird plumage: 31. Henmi, H., M. Hata, et al. 1991. Studies on the carotenoids in the complement of red pigments in the plumage of wild and the muscle of salmon--V. Combination of astaxanthin and captive bullfinch (Pyrrhula pyrrhula). Comparative canthaxanthin with bovine serum albumin and egg albumin. Biochemistry and Physiology Part B: Biochemistry and Comparative Biochemistry and Physiology Part B: Molecular Biology 128(3): 529-535. Comparative Biochemistry 99(3): 609-612. 46. Bhosale, P., B. Serban, et al. 2007. Identification and 32. Stradi, R., J. Hudon, et al. 1998. Carotenoids in bird plumage: Metabolic Transformations of Carotenoids in Ocular Tissues the complement of yellow and red pigments in true of the Japanese Quail Coturnix japonica Biochemistry 46(31): woodpeckers (Picinae). Comparative Biochemistry and 9050-9057. Physiology Part B: Biochemistry and Molecular Biology 120(2): 223-230. 47. Siefermann-Harms, D., Hertzberg, S., Borch, G. and S. Liaaen-Jensen. 1981 Lactucaxanthin, an ɛ,ɛ-carotene-3,3-diol 33. Krinsky N.I., Landrum J.T. and R.A. Bone. 2003. Biologic from Lactuca sativa. Phytochemistry 20:85–88. mechanisms of the protective role of lutein and zeaxanthin in the eye. Annu. Rev. Nutr. 23:171–201 48. Bai, L., E. H. Kim, et al. 2009. Novel lycopene epsilon cyclase activities in maize revealed through perturbation of carotenoid 34. Khachik F. 2006. Distribution and metabolism of dietary biosynthesis. The Plant Journal 59(4): 588-599. carotenoids in humans as a criterion for development of nutritional supplements. Pure Appl. Chem. 78(8): 1551–1557. 49. Asai, A., M. Terasaki, et al. 2004. An epoxide-furanoid rearrangement of spinach neoxanthin occurs in the 35. Goodwin, T. W. 1986. Metabolism, Nutrition, and Function of gastrointestinal tract of mice and in vitro: formation and Carotenoids. Annual Review of Nutrition 6(1): 273-297. cytostatic activity of neochrome stereoisomers. J. Nutr. 134(9): 2237-2243. 36. Hudon J., Anciães M., Bertacche V. and R. Stradi 2007. Plumage carotenoids of the Pin-tailed Manakin (Ilicura 50. Hallenstvet, M., E. Ryberg, et al. 1978. Animal carotenoids-- militaris): evidence for the endogenous production of XIV carotenoids of Psammechinus miliaris (sea-urchin). rhodoxanthin from a colour variant. Comp Biochem and Comparative Biochemistry and Physiology Part B: Physiol Part B 147: 402–411. Comparative Biochemistry 60(2): 173-175.

37. McGraw, K. J., E. Adkins-Regan, Parker R.S. 2002. 51. Yonekura, L., M. Kobayashi, Terasaki M., and A. Nagao. Anhydrolutein in the zebra finch: a new, metabolically derived 2010 Keto-carotenoids are the major metabolites of dietary carotenoid in birds. Comparative Biochemistry and lutein and fucoxanthin in mouse tissues. J. Nutr. 140: 1824- Physiology Part B: Biochemistry and Molecular Biology 1831 132(4): 811-818. 52. Tsushima, M. and T. Matsuno 1990. Comparative biochemical 38. McGraw K.J., Hill G.E., Stradi R., Parker R.S. 2002. The studies of carotenoids in sea-urchins-I. Comparative effect of dietary carotenoid access on sexual dichromatism and Biochemistry and Physiology Part B: Comparative plumage pigment composition in the American goldfinch. Biochemistry 96(4): 801-810. Comparative Biochemistry and Physiology B 131: 261–269. 53. Fox, D. L. and T. S. Hopkins 1966. Comparative metabolic 39. Matsuno, T. 2001. Aquatic animal carotenoids. Fisheries fractionation of carotenoids in three flamingo species. Science 67(5): 771-783. Comparative Biochemistry and Physiology 17(3): 841-856.

40. Tsushima, M., Y. Ikuno, Nagata S., Kodama K. and T. 54. Toyoda, Y., L. R. Thomson, et al. 2002. Effect of Dietary Matsuno. 2002 Comparative biochemical studies of Zeaxanthin on Tissue Distribution of Zeaxanthin and Lutein in carotenoids in catfishes. Comparative Biochemistry and Quail. Invest. Ophthalmol. Vis. Sci. 43(4): 1210-1221. Physiology Part B: Biochemistry and Molecular Biology 133(3): 331-336. 55. Arpin, N. and S. Liaaen-Jensen 1969. Carotenoids of higher plants--II: Rubixanthin and gazaniaxanthin. Phytochemistry 41. McGraw, K. J. and M.C. Nogare 2004. Carotenoid pigments 8(1): 185-193. and the selectivity of psittacofulvin-based coloration systems in parrots. Comp Biochem Physiol Part B: Biochemistry and 56. Deviche, P., K. J. McGraw, et al. 2008. Season-, sex-, and Molecular Biology 138(3): 229-233. age-specific accumulation of plasma carotenoid pigments in free-ranging white-winged crossbills Loxia leucoptera. Journal 42. Ohkubo, M., M. Tsushima, et al. 1999. Carotenoids and their of Avian Biology 39(3): 283-292. metabolism in the goldfish Carassius auratus (Hibuna). Comparative Biochemistry and Physiology Part B: 57. Inouye C.Y., G. E. Hill, Stradi R.D. and R. Montgomerie Biochemistry and Molecular Biology 124(3): 333-340. 2001. Carotenoid pigments in male house finch plumage in relation to age, subspecies, and ornamental coloration. Auk 43. Andersson, Staffan, et al. 2007. Carotenoid content and 118(4): 900-915. reflectance of yellow and red nuptial plumages in widowbirds (Euplectes spp.). Functional Ecology 21:272-281. 58. Valadon, L. R. G. and R. S. Mummery 1969. Changes in Carotenoid Composition of Certain Roses with Age. Annals of Botany 33(4): 671-677. 45

59. Umeno, D., A. V. Tobias, et al. 2005. Diversifying Carotenoid 76. Ravanello, M. P., D. Ke, J. Alvarez, B. Huang, and C. K. Biosynthetic Pathways by Directed Evolution. Microbiol. Mol. Shewmaker. 2003. Coordinate expression of multiple bacterial Biol. Rev. 69(1): 51-78. carotenoid genes in canola leading to altered carotenoid production. Metabolic Engineering 5:255-263. 60. Niklitschek, M., J. Alcaino, et al. 2008. Genomic organization of the structural genes controlling the astaxanthin biosynthesis 77. Katayama, T., S. Makoto, S. Muneo, and C.O. Chichester. pathway of Xanthophyllomyces dendrorhous. Biological 1973. The biosynthesis of astaxanthin. XII. The conversion of 3 Research 41: 93-108. labelled β-carotene-15, 15 H2 into body astaxanthin in the lobster, Panulirus japonicus. International Journal of 61. Cunningham, F. X., Jr., H. Lee, et al. 2007. Carotenoid Biochemistry 4:223-226. Biosynthesis in the Primitive Red Alga Cyanidioschyzon merolae. Eukaryotic Cell 6(3): 533-545. 78. Katayama, T., Y. Kunisaki, M. Shimaya, K. L. Simpson, and C. O. Chichester. 1973a. The biosynthesis of astaxanthin-- 3 62. Sandmann, G. 1994. Carotenoid biosynthesis in XIV. The conversion of labelled beta-carotene-15,15'- H2 microorganisms and plants. European Journal of Biochemistry into astaxanthin in the crab, Portunus trituberculatus. 223(1): 7-24. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 46:269-272. 63. Ladygin, V. G. 2000. Biosynthesis of Carotenoids in the Chloroplasts of Algae and Higher Plants. Russian Journal of 79. Withers, N. W., R. S. Alberte, R. A. Lewin, J. P. Thornber, G. Plant Physiology 47(6): 796-814. Britton, and T. W. Goodwin. 1978. Photosynthetic unit size, carotenoids, and chlorophyll-protein composition of 64. Leuenberger, F., Thommen, H . 1970. Keto-carotenoids in the prochloron sp., a prokaryotic green alga. Proceedings of the Colorado beetle Leptinotarsa decemlineata . l. Insect Physiol. National Academy of Sciences of the United States of 16: 1 855-58. America 75:2301-2305.

65. Kayser 1977 Conversion of C14β-carotene to its 2-hydroxy 80. Cunningham, F. X., and E. Gantt. 2005. A study in scarlet: and 3-hydroxy metabolites by two moth species. Comp enzymes of ketocarotenoid biosynthesis in the flowers of Biochem Physiol. 59: 177-181. Adonis aestivalis. The Plant Journal 41:478-492.

66. Katayama T., Kamata T., Shimaya M., Deshimaru O. and 81. Hsieh, L. K., T.-C. Lee, C. O. Chichester, and K. L. Simpson. Chichester C. O. (1972) The biosynthesis of astaxanthin--VIII. 1974. Biosynthesis of Carotenoids in Brevibacterium sp. KY- 3 The conversion of labelled β-carotene-15,15'- H2 into 4313. Journal of Bacteriology 118:385-393. astaxanthin in prawn, Penaeus japonicus Bat3. Nippon Suisan Gakkaishi, 38, 1171-1175. 82. Matsuno, T., K. Katagiri, T. Maoka, and T. Komori. 1985. Novel reductive metabolic pathways of 4-oxo-beta-end 67. KEGG (reaction R)1851, enzyme 1. 14.13.12 9) group in carotenoids of the spindle shell Fusinus perplexus. Comparative Biochemistry and Physiology Part B: 68. Maoka T. 2011 Carotenoids in Marine Animals. 9:278- Comparative Biochemistry 81:905-908. 293293. Mar Drug. 83. Gilchrist, B. M., and W. L. Lee. 1976. The incorporation of 69. Heller K.G., Fleischmann P. and Lutz-Röder A. 2000 14C β-carotene into the marine isopod Idotea resecata Carotenoids in the spermatophores of bushcrickets (Stimpson, 1857) and the biosynthesis of canthaxanthin. (Orthoptera: Ephippigerinae). Proceed Royal Soc. 267: 1905- Comparative Biochemistry and Physiology Part B: 1908. Comparative Biochemistry 54:343-346.

70. Goodwin T.W. 1984 The biochemistry of the carotenoids. 84. Yuan, J.-P., and F. Chen. 1997. Identification of astaxanthin Vol. 1. Plants. Chapman and Hall Eds isomers in Haematococcus lacustris by HPLC-photodiode array detection. Biotechnology Techniques 11:455-459. 71. Goodwin T.W. 1984 The biochemistry of the carotenoids. Vol. 2. Animals. Chapman and Hall Eds. 85. Schiedt, K., F. J. Leuenberger, and M. Vecchi. 1981. Natural occurrence of enantiomeric and meso-astaxanthin. 5. Ex wild 72. Fox D.L., McBeth, J.W., and G. Mackinney. 1970 Some salmon (Salmo salar and Oncorhynchus). Helvetica Chimica dietary carotenoids and blood-carotenoid levels in flamingos- Acta 64:449-457. II. -carotene and -carotene consumed by the American flamingo. Comparative Biochemistry and Physiology 36:253- 86. Toomey, M. B., and K. J. McGraw. 2010. The effects of 262. dietary carotenoid intake on carotenoid accumulation in the retina of a wild bird, the house finch (Carpodacus mexicanus). 73. Choi, S.-K., H. Harada, S. Matsuda, and N. Misawa. 2007. Archives of Biochemistry and Biophysics 504:161–168. Characterization of two β-carotene ketolases, CrtO and CrtW, by complementation analysis in Escherichia coli. Applied 87. Cunningham, F. X., and E. Gantt. 2001. One ring or two? Microbiology and Biotechnology 75:1335-1341. Determination of ring number in carotenoids by lycopene epsilon-cyclases. Proceedings of the National Academy of 74. Cunningham, F. X., and E. Gantt. 1998. Genes and enzymes Sciences 98:2905-2910. of carotenoid biosynthesis in plants. Annu Rev Plant Physiol Mol Biol 49:557 - 583. 88. Hora, J., T. P. Toube, and B. C. L. Weedon. 1970. Carotenoids and related compounds. Part XXVII. Conversion of 75. Fraser, P. D., Y. Miura, and N. Misawa. 1997. In vitro fucoxanthin into paracentrone. Journal of the Chemical characterization of astaxanthin biosynthetic enzymes. Journal Society C: Organic 2:241-242. of Biological Chemistry 272:6128-6135. 46

89. KEGG (reaction R07568, enzyme CtlZ) and Physiology Part B: Biochemistry and Molecular Biology 113(2): 427-432. 90. Goodfellow D., Moss G. P. and B.C.L. Weedon. 1970. The Absolute Configuration of Lutein. J. Chem. Soc. D 13:1578- 105. Egeland, E. S., G. Johnsen, W. Eikrem, J. Throndsen, et 1578. al.1995. Pigments of Bathycoccus prasinos 91. Khachik, F., P. S. Bernstein, and D. L. Garland. 1997. (Prasinophyceae): methodological and chemosystematic Identification of lutein and zeaxanthin oxidation products in implications. J. PhycoL 31: 554-561. human and monkey retinas. Investigative Ophthalmology & Visual Science 38:1802-11. 106. Fernández, J. A. and J. Burgos.1981. Carotenoid pigments in the flesh and carapace of Aristaeomorpha foliacea and 92. Hata, M., and M. Hata. 1971. Carotenoid pigments in Heterocarpus dorsalis (crustacea: decapoda). Comparative goldfish (carassius auratus) II. colour change and carotenoid Biochemistry and Physiology Part B: Comparative pigment composition. International Journal of Biochemistry Biochemistry 69: 559-575. 2:182-184. 107. Miki, W., K. Yamaguchi, S. Konosu, T. Takane, et al.1985. 93. Stradi, R., G. Celentano, M. Boles, and F. Mercato. 1997. Origin of tunaxanthin in the integument of yellowtail (Seriola Carotenoids in bird plumage: The pattern in a series of red- quinqueradiata). Comparative Biochemistry 80(2): 195-201. pigmented carduelinae. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 108. Juola, F.A,. McGraw K. and Dearborn D.C, Carotenoids and 117:85-91. throat pouch coloration in the great frigatebird (Fregata minor). Biochemistry and Molecular Biology 149:370-377. 94. Takaichi, S., K. Shimada, and J. Ishidsu. 1990. Carotenoids from the aerobic photosynthetic bacterium, Erythrobacter 109. Hata and Hata 1971. Carotenoid pigments in goldfish longus: β-Carotene and its hydroxyl derivatives. Archives of (Carassius auratus L.)-III. Metabolism of ingested Microbiology 153:118-122. cynthiaxanthin. Tohoku J. Agri. Res., 21 (19716), pp. 183–188

95. Czeczuga, B. 1985. Carotenoids in representatives of the 110. Buchecker, R. 1982 A chemist’s view of animal carotenoids cladoniaceae. Biochemical Systematics and Ecology 13:83-88. in: Carotenoid Chemistry and Biochemistry. G. Britton, T.W. and Goodwin (Eds.), Pergamon Press, Oxford (1982), p. 175- 96. Cunningham, F. X. J., and G. E. 2011. Elucidation of the 193. Pathway to Astaxanthin in the Flowers of Adonis aestivalis. The Plant Cell 23:3055-3069. 111. Britton, G., Liaaen-Jensen, S and H. Pfander. 2004 Carotenoid. Handbook. Birkhäuser, Basel, Switzerland. 97. Takaichi, S., G. Sandmann, G. Schnurr, Y. Satomi, A. Suzuki, and N. Misawa. 1996. The carotenoid 7, 8-dihydro- end 112. Czeczuga, B. and R. D. Worthington.1997. Carotenoids in group can be cyclized by the lycopene cyclases from the lichens from the States of New Mexico and Texas in the bacterium Erwinia Uredovora and the higher Plant Capsicum United States of America. Feddes Repertorium 108: 387-399. Annuum. European Journal of Biochemistry 241:291-296. 113. Goodwin, T. W. 1974. Algal physiology and biochemistry, 98. Gribanovski-Sassu, O. 1972. Effect of diphenylamine on Blackwell Scientific Publications Ltd. carotenoid synthesis in Dictyococcus cinnabarinus. Phytochemistry 11:3195-3198. 114. Schiedt, K., S. Bischof and E. Glinz. 1993. Carotenoids. Part B. Metabolism, genetics, and biosynthesis, Harcourt Brace 99. Hudon, J., and A. H. Brush. 1990. Carotenoids produce flush Jovanovich Publishers. in the elegant tern plumage. The Condor 92:798-801. 115. Han, Q., K. Shinohara, Y. Kakubari and Y. Mukai.2003. 100. Schwartzel, E. M., and J. J. Cooney. 1972. Isolation of 4- Photoprotective role of rhodoxanthin during cold acclimation Hydroxyechinenone from Micrococcus roseus. Journal of in Cryptomeria japonica. Plant, Cell & Environment 26(5): Bacteriology 112: 1422-1424. 715-723.

101. McGraw, K. J., P. M. Nolan, and O. L. Crino. 2006. 116. Maoka, T., Y. Ito, Fujiwara and K. Hashimoto.1996. Carotenoid accumulation strategies for becoming a colourful Structures and antioxidative activity of retro Carotenoids from house finch: analyses of plasma and liver pigments in wild the Berries of the Japanese yew, Taxus cuspidata. J. Jap. Oil moulting birds. Functional Ecology 20:678-688. Chem. Soc. 45: 641-646.

102. Makino, T., H. Harada, H. Ikenaga, S. Matsuda, S. Takaichi, 117. Hsu, W.-J., D. B. Rodriguez and C. O. Chichester. 1972. The K. Shindoet al. (2008). Characterization of Cyanobacterial biosynthesis of astaxanthin. VI. the conversion of 14Clutein Carotenoid Ketolase CrtW and Hydroxylase CrtR by and 14C β-carotene in goldfish. International Journal of Complementation Analysis in Escherichia coli. Plant and Cell Biochemistry 3: 333-338. Physiology 49: 1867-1878. 118. Hata, M. and Hata, M. 1972 Carotenoid pigments in goldfish - 103. Teruhisa Katayama, Y. K., Makoto Shimaya, K.L. Simpson IV. Carotenoid metabolism, Bull. Jap. Soc. Sci. Fish., 38. 331- and C.O. Chichester 1973. The biosynthesis of astaxanthin— 338. XIV. The conversion of labelled β-carotene- 15,15-3H2 into astaxanthin in the crab, Portunus trituberculatus. Comp 119. Matsuno, T. and Katsuyama, M. 1982 Metabolism of Biochem 46: 269-272. zeaxanthin to rhodoxanthin in tilapia. Nippon Suisan Gakkaishi, 48 (1982), pp. 1491–1493. 104. Stradi, R., E. Rossi, G. Celentano and B. Bellardi. 1996. Carotenoids in bird plumage: The pattern in three Loxia 120. Matsuno, T., Nagata, S., Iwahashi, M, Koike, T., Okada M. species and in Pinicola enucleator. Comparative Biochemistry 1979. Intensification of color of fancy red carp with 47

zeaxanthin and myxoxanthophyll, major carotenoid 203. LaFountain, A. M., S. Kaligotla, S. Cawley, K. M. Riedl, S. J. constituents of spirulina. Bull. Jap. Soc. Scient. Fish., 45, pp. Schwartz, H. A. Frank, and R. O. Prum. 2010. Novel 627–632. methoxy-carotenoids from the burgundy-colored plumage of the Pompadour Cotinga Xipholena punicea. Archives of 121. Guillou, A., Choubert, G., de la Noe, J. 1992. Comparative Biochemistry and Biophysics 504:142-153. accumulations of labelled carotenoids (14C-astaxanthin, 3H- canthaxanthin and 3H-zeaxanthin) and their metabolic conversions in mature female rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. B. 102: 61-65.

122. Strand, A., O. Herstad and S. Liaaen-Jensen. 1998. Fucoxanthin metabolites in egg yolks of laying hens. Comparative Biochemistry and Physiology 119: 963-974.

123. McGraw, K. J., G. E. Hill, R. Stradi and R. S. Parker. 2002. The effect of dietary carotenoid access on sexual dichromatism and plumage pigment composition in the American goldfinch. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 131: 261-269.

124. Nitsche, H. (1974). Neoxanthin and fucoxanthinol in Fucus vesiculosus. Biochimica et Biophysica Acta (BBA) - General Subjects 338: 572-576.

125. Tian, L., Magallanes-Lundback, M., Musetti, V. and D. DellaPenna 2003 Functional Analysis of β- and -ring carotenoid hydroxylases in Arabidopsis The Plant Cell 15: 1320-1332.

126. Czeczuga, B. 1981. Carotenoids in fish. XXVIII. Carotenoids in Micropterus salmoides (Lalépéde) Centrarchidae. Hydrobiologia 78: 45-98.

127. Schiedt, K., Foss, P., Trond, S. and S. Liaaen-Jensen. 1989. Metabolism of carotenoids in salmonids-I. Idoxanthin, a metabolite of astaxanthin in the flesh of Atlantic salmon (Salmo salar, L.) under varying external conditions. Comparative Biochemistry and Physiology 92B: 277-281.

128. Hepperle, S.S., Li, Q. and A.L.L. East. 2005. Mechanism of cis/trans equilibration of alkenes via iodine catalysis. Journal of Physical Chemistry A 109: 10975-10981.

129. McDermott, J.C.B., Brown, D.J., Britton, G. and T.W. Goodwin. 1974. Alternative pathways of zeaxanthin biosynthesis in a Flavobacterium species. Experiments with nicotine as inhibitor. Biochemical Journal 144: 231-243.

130. Buchecker, R., Eugster, C.H. 1978. 183. Absolute konfiguration von -doradexanthin und von frtischiellaxanthin, einem neuen carotenoid aus Fritschiella tuberose IYENG. Helvetica Chimica Acta 61: 1962-1968.

131. KEGG (reaction R07850, enzyme 1. 14. 99. 45).

200. Prum, R. O., A. M. LaFountain, J. Berro, M. C. Stoddard, and H. A. Frank. 2012. Molecular diversity, metabolic transformation, and evolution of carotenoid feather pigments in cotingas (Aves: Cotingidae). J Comp Physiol B 182:1095- 1116. 201. Prum, R., A. LaFountain, C. Berg, M. Tauber, and H. Frank. 2014. Mechanism of carotenoid coloration in the brightly colored plumages of broadbills (Eurylaimidae). J Comp Physiol B 184:651-672. 202. LaFountain, A. M., H. A. Frank, and R. O. Prum. 2013. Carotenoids from the crimson and maroon plumages of Old World orioles (Oriolidae). Archives of Biochemistry and Biophysics 539:126-132. 48

Appendix S2. Characteristics of caroternoid metabolic networks for species used in the study. See Supplementary Material for methods and details.

Common name Scientific Name Diet nodes Nodes Edges Diam Path Degree Diet diam Cluster Assort Heter Centr Density Modularity Modules Sensit/edge Sens/node Method Main references Long Tailed Tit Aegithalos_caudatus 1 2 1 1 1.00 1.00 1.00 0.00 1.00 0.00 . 1.00 0.00 1 1.00 0.75 HPLC 62, 63 Redwinged Blackbird Agelaius_phoeniceus 4 17 37 6 2.47 4.35 4.00 0.41 3.33 0.37 0.15 0.18 0.53 4 0.00 0.06 HPLC 11, 38, 47 Redlegged Partridge Alectoris_rufa 3 17 31 6 2.62 3.65 4.00 0.36 2.66 0.32 0.12 0.15 0.61 3 HPLC 100, 101 Red Munia Amandava_amandava 3 5 4 3 1.57 1.60 1.00 0.00 1.75 0.62 0.33 0.30 0.17 3 0.13 0.24 HPLC 40 Zebra Waxbill Amandava_subflava 3 5 4 3 1.57 1.60 1.00 0.00 1.75 0.62 0.33 0.30 0.17 3 0.13 0.24 HPLC 40 Scaled fruiteater Ampelioides_tschudii 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Chestnutcrested cotinga Ampelion_rufaxilla 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Mallard Anas_platyrhynchos 6 8 6 2 1.25 1.50 2.00 0.00 1.67 0.71 0.19 0.14 0.50 4 0.08 0.13 HPLC2 9, 12, 53 Greylag Goose Anser_anser 3 3 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 3 . 0.33 HPLC2 12, 50 Helmeted manakin Antilophia_galeata 2 7 11 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Narina trogon Apaloderma_narina 4 16 32 8 3.06 3.76 7.00 0.35 3.04 0.46 0.18 0.15 0.53 4 HPLC 111 Cedar Waxwing Bombycilla_cedrorum 5 13 20 3 1.57 3.08 3.00 0.36 2.99 0.60 0.28 0.18 0.44 5 0.01 0.09 HPLC2 8, 11, 26, 62 Bohemian Waxwing Bombycilla_garrulus 2 9 18 3 1.39 4.00 3.00 0.53 2.90 0.25 0.21 0.33 0.41 2 0.00 0.19 HPLC 62, 63 Japanese Waxwing Bombycilla_japonica 2 9 18 3 1.39 4.00 3.00 0.53 2.90 0.25 0.21 0.33 0.41 2 0.00 0.19 HPLC 62 Trumpeter Finch Bucanetes_githagineus 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 63 Yellowrumped cacique Cacicus_cela 1 1 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 1 HPLC 108 Redrumped cacique Cacicus_haemorrhous 4 17 35 6 2.73 4.12 4.00 0.42 2.84 0.30 0.10 0.16 0.59 4 HPLC 108 Northern mountain cacique Cacicus_leucoramphus 2 2 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 2 HPLC 108 Mexican cacique Cacicus_melanicterus 2 2 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 2 HPLC 108 Scarletrumped cacique Cacicus_uropygialis 3 12 21 5 2.00 3.50 3.00 0.41 2.62 0.32 0.18 0.21 0.55 3 HPLC 108 Green broadbill Calyptomena_viridis 1 4 2 2 1.33 1.00 2.00 0.00 1.86 0.71 0.67 0.33 0.00 2 HPLC 106 Creambacked Woodpecker Campephilus_leucopogon 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 68 Pyrrhuloxia Cardinalis_sinuatis 4 16 30 6 2.76 3.75 5.00 0.38 2.84 0.44 0.20 0.16 0.54 4 HPLC 111 Northern Cardinal Cardinalis_cardinalis 5 20 40 9 3.01 4.00 6.00 0.33 3.49 0.48 0.19 0.14 0.51 5 0.01 0.06 HPLC2 11, 28, 43, 46 Black Siskin Carduelis_atrata 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 62, 63 Linnet Carduelis_cannabina 6 18 31 6 2.55 3.44 4.00 0.33 2.86 0.61 0.18 0.13 0.51 6 0.01 0.06 HPLC 3, 62, 69 European Goldfinch Carduelis_carduelis 3 10 18 3 1.39 3.60 3.00 0.48 2.90 0.43 0.22 0.27 0.41 3 0.00 0.16 HPLC2 28, 64, 65 European Greenfinch Carduelis_chloris 2 5 8 3 1.42 3.20 3.00 0.33 2.73 0.33 0.25 0.60 0.11 2 0.00 0.20 HPLC 54, 60, 64, 65 Red Siskin Carduelis_cucullata 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 62, 63 Common Redpoll Carduelis_flammea 5 17 31 6 2.55 3.65 4.00 0.35 2.86 0.54 0.19 0.15 0.51 5 0.01 0.07 HPLC 62, 67, 69 Hoary Redpoll Carduelis_hornemanni 5 18 33 8 2.90 3.67 6.00 0.33 3.12 0.55 0.24 0.14 0.51 5 0.01 0.07 HPLC 67 Oriental Greenfinch Carduelis_sinica 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 65 Yellowbreasted Greenfinch Carduelis_spinoides 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 65 Eurasian Siskin Carduelis_spinus 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 62, 64, 65 American Goldfinch Carduelis_tristis 4 7 8 3 1.42 2.29 3.00 0.24 2.73 0.75 0.30 0.29 0.11 4 0.00 0.14 HPLC 11, 48 House Finch Carpodacus_mexicanus 6 21 46 9 3.04 4.00 8.00 0.27 3.32 0.50 0.17 0.12 0.55 6 0.01 0.06 HPLC2 3, 6, 32, 42, 70, 71 Darkbreasted Rosefinch Carpodacus_nipalensis 3 12 24 4 1.80 4.00 4.00 0.33 3.08 0.35 0.26 0.24 0.44 3 0.00 0.08 HPLC 63 Beautiful Rosefinch Carpodacus_pulcherrimus 4 16 25 6 2.34 3.13 6.00 0.18 2.81 0.53 0.29 0.15 0.50 4 0.03 0.11 HPLC 63, 67 Pallas' Rosefinch Carpodacus_roseus 5 18 26 6 2.33 2.89 6.00 0.16 2.61 0.57 0.26 0.12 0.54 5 0.03 0.09 HPLC 62, 69 Streaked Rosefinch Carpodacus_rubicilloides 2 4 2 1 1.00 1.00 1.00 0.00 1.00 0.00 0.00 0.33 0.50 2 0.50 0.38 HPLC 63, 67, 69 Whitebrowed Rosefinch Carpodacus_thura 4 13 19 5 1.91 2.92 3.00 0.23 2.47 0.56 0.30 0.17 0.47 4 0.02 0.10 HPLC 63 Threebanded Rosefinch Carpodacus_trifasciatus 3 10 17 4 1.88 3.40 4.00 0.30 2.71 0.53 0.39 0.24 0.33 4 0.01 0.11 HPLC 63, 67 Hooded berryeater Carpornis_cucullatus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Swallowtailed manakin Chiroxiphia_caudata 2 7 12 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Bluebacked manakin Chiroxiphia_pareola 2 7 11 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.66 2 HPLC 105 Sootycapped Bush Tanager Chlorospingus_pileatus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 35 White Stork Ciconia_ciconia 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 TLC 49 Hooded Grosbeak Coccothraustes_abeillei 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Evening Grosbeak Coccothraustes_vespertina 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 HPLC 45 Bananaquit Coereba_flaveola 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 TLC 30 Yellowshafted Flicker Colaptes_auratus 4 18 34 8 2.88 3.78 6.00 0.33 3.23 0.49 0.23 0.15 0.52 4 0.01 0.07 HPLC 11, 62, 68 Redshafted Flicker Colaptes_auratus_cafer 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 62 Campo Flicker Colaptes_campestris 3 3 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 3 . 0.33 HPLC 68 Gilded Flicker Colaptes_chrysoides 3 11 19 6 2.29 3.45 5.00 0.27 3.15 0.55 0.44 0.24 0.35 3 0.02 0.12 HPLC 62 Greenbarred Woodpecker Colaptes_melanochloros 3 13 21 6 2.24 3.23 5.00 0.23 2.97 0.52 0.36 0.19 0.42 3 0.03 0.11 HPLC 62, 68 Lovely cotinga Cotinga_amabilis 1 5 10 6 2.52 3.11 6.00 0.34 2.49 0.35 0.29 0.28 0.36 3 HPLC 103 Purplebreasted cotinga Cotinga_cotinga 1 5 10 6 2.52 3.11 6.00 0.34 2.49 0.35 0.29 0.28 0.36 3 HPLC 103 Banded cotinga Cotinga_maculata 1 5 10 6 2.52 3.11 6.00 0.34 2.49 0.35 0.29 0.28 0.36 3 HPLC 103 Blue Tit Cyanistes_caeruleus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 2, 62 Blackandred broadbill Cymbirhynchus_macrorhynchos 1 4 3 4 2.00 1.60 4.00 0.00 1.80 0.31 0.17 0.40 0.22 2 HPLC 106 Great Spotted Woodpecker Dendrocopos_major 3 11 19 6 2.29 3.45 5.00 0.27 3.15 0.55 0.44 0.24 0.35 3 0.02 0.12 HPLC 62, 68 Yellowrumped Warbler Dendroica_coronata 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Palm Warbler Dendroica_palmarum 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Yellow Warbler Dendroica_petechia 4 4 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 4 . 0.25 HPLC 45 Pileated Woodpecker Dryocopus_pileatus 3 15 23 6 2.20 3.07 5.00 0.15 2.84 0.52 0.31 0.16 0.45 3 0.03 0.11 HPLC 68 Yellowhammer Emberiza_citrinella 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 62, 63 Blackheaded Bunting Emberiza_melanocephala 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 62, 63 49

Robin Erithacus_rubecula 1 2 1 1 1.00 1.00 1.00 0.00 1.00 0.00 . 1.00 0.00 1 1.00 0.75 HPLC 63 Gouldian Finch Erythrura_gouldiae 2 12 21 5 2.40 3.50 4.00 0.25 2.84 0.35 0.27 0.23 0.40 3 0.04 0.16 HPLC 63 Redheaded Parrot Finch Erythrura_psittacea 2 8 13 4 1.90 3.25 4.00 0.46 2.88 0.48 0.52 0.32 0.29 3 0.04 0.22 HPLC 63 Scarlet Ibis Eudocimus_ruber 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 TLC 21 Thickbilled Euphonia Euphonia_laniirostris 4 7 5 2 1.33 3.00 0.00 2.50 1.00 0.60 0.20 0.00 1 HPLC 11, 111 Orangecrowned euphonia Euphonia_saturata 2 5 5 2 1.33 3.00 0.00 1.67 0.50 0.58 0.40 0.22 2 HPLC 111 Yellowcrowned Bishop Euplectes_afer 2 3 3 2 1.25 2.00 1.00 0.00 1.67 0.35 1.00 0.67 0.00 1 0.00 0.33 HPLC 55, 63 Redcollared Widowbird Euplectes_ardens 4 19 40 9 3.01 4.21 6.00 0.35 3.49 0.42 0.20 0.16 0.51 3 0.01 0.06 HPLC 1 Fantailed Widowbird Euplectes_axillaris 2 6 9 3 1.53 3.00 3.00 0.24 3.07 0.47 0.50 0.47 0.12 2 0.03 0.19 HPLC 1 Yellow Bishop Euplectes_capensis 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Yellowmantled Widowbird Euplectes_macrourus 2 5 8 3 1.42 3.20 3.00 0.33 2.73 0.33 0.25 0.60 0.11 2 0.04 0.20 HPLC 1 Southern Red Bishop Euplectes_orix 4 17 35 9 3.08 4.12 6.00 0.31 3.36 0.45 0.23 0.17 0.47 4 0.01 0.07 HPLC 55, 63 Banded broadbill Eurylaimus_javanicus 1 6 5 4 1.73 2.00 4.00 0.00 2.13 0.25 0.19 0.29 0.37 2 HPLC 106 Blackandyellow broadbill Eurylaimus_ochromalus 1 6 5 4 1.73 2.00 4.00 0.00 2.13 0.25 0.19 0.29 0.37 2 HPLC 106 Wattled broadbill Eurylaimus_steerii 1 4 3 4 2.00 1.60 4.00 0.00 1.80 0.31 0.17 0.40 0.22 2 HPLC 106 Korean Flycatcher Ficedula_zanthopygia 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 1.00 HPLC 63 Red Fody Foudia_madagascariensis 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 63 Great Frigatebird Fregata_minor 4 5 4 2 1.20 1.60 2.00 0.00 1.40 0.33 0.33 0.30 0.44 2 0.50 0.20 HPLC 36 Chaffinch Fringilla_coelebs 4 12 20 4 1.83 3.33 4.00 0.19 2.61 0.53 0.31 0.20 0.39 4 0.01 0.09 HPLC 62 Brambling Fringilla_montifringilla 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Domestic Chicken Gallus_gallus_domesticus 4 15 34 6 2.61 4.53 5.00 0.37 3.53 0.47 0.18 0.20 0.44 4 0.00 0.07 HPLC2 12, 72 Common Yellowthroat Geothlypis_trichas 4 4 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 4 . 0.25 HPLC 11, 45 Crimson fruitcrow Haematoderus_militaris 3 11 11 5 2.05 5.00 0.10 1.93 0.43 0.11 0.11 0.67 4 HPLC 103 Scarlet Finch Haematospiza_sipahi 5 18 33 8 2.90 3.67 6.00 0.33 3.12 0.55 0.24 0.14 0.51 5 0.01 0.07 HPLC 63 Flamecrested Manakin Heterocercus_linteatus 3 11 19 2 1.25 1.50 6.00 0.00 1.50 0.35 0.14 0.18 0.35 3 HPLC 105 Yellowbreasted Chat Icteria_virens 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 HPLC 37 Bullock's oriole Icterus_bullockii 2 9 16 2 1.30 3.56 2.00 0.62 2.74 0.28 0.25 0.31 0.50 2 HPLC 107 Orangebacked troupial Icterus_croconotus 1 3 1 1 1.00 1.00 2.00 1.00 0.33 0.50 0.71 0.00 2 HPLC 107 Hooded oriole Icterus_cucullatus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Hispaniolan oriole Icterus_dominicensis 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Audobon's oriole Icterus_graduacauda 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Altamira oriole Icterus_gularis 4 13 24 5 1.85 3.69 3.00 0.36 2.98 0.41 0.25 0.21 0.51 3 HPLC 107 Northern Oriole Icterus_galbula 3 18 39 9 3.01 4.33 7.00 0.37 3.48 0.38 0.21 0.17 0.51 3 0.01 0.06 TLC 28 Venezuelan troupial Icterus_icterus 2 5 8 3 1.42 3.20 3.00 0.33 2.73 0.33 0.25 0.60 0.11 2 HPLC 111 Yellowtailed oriole Icterus_mesomelas 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Yellow oriole Icterus_nigrogularis 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Spotbreasted oriole Icterus_pectoralis 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Blackcowled oriole Icterus_prosthemelas 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 107 Streakedback oriole Icterus_pustulatus 4 13 24 5 1.85 3.69 3.00 0.36 2.98 0.41 0.25 0.21 0.51 3 HPLC 107 Pintailed Manakin Ilicura_militaris 3 9 12 4 2.00 2.67 4.00 0.29 2.82 0.51 0.29 0.28 0.36 3 0.06 0.17 HPLC 31 Whitebrowed purpletuft Iodopleura_isabellae 1 5 10 1 1.38 2.00 0.75 2.80 0.33 0.67 0.60 0.11 2 HPLC 102 Ringbilled Gull Larus_delawarensis 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 HPLC 39 Yellowlegged Gull Larus_michahellis 6 10 13 5 2.21 2.60 4.00 0.00 1.89 0.42 0.06 0.18 0.41 4 0.00 0.10 HPLC 52, 59 Silvereared Mesia Leiothrix_argentauris 2 9 16 5 2.21 3.56 4.00 0.33 2.91 0.39 0.41 0.31 0.29 3 0.03 0.19 HPLC 63 Redbilled Leiothrix Leiothrix_lutea 2 9 16 5 2.21 3.56 4.00 0.33 2.91 0.39 0.41 0.31 0.29 3 0.03 0.22 HPLC 62, 63 Bluecrowned manakin Lepidothrix_coronata 2 5 8 3 1.42 3.20 2.00 0.33 2.73 0.33 0.25 0.60 0.11 2 HPLC 105 Snowcapped manakin Lepidothrix_nattereri 2 5 8 3 1.42 3.20 2.00 0.33 2.73 0.33 0.25 0.60 0.11 2 HPLC 105 Whitefronted manakin Lepidothrix_serena 2 5 7 3 1.42 3.20 2.00 0.33 2.73 0.33 0.25 0.60 0.11 2 HPLC 105 Franklin's Gull Leucophaeus_pipixcan 1 1 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 1 . 1.00 HPLC 39 Rosecollared piha Lipaugus_streptophorus 1 5 10 6 2.52 3.11 6.00 0.34 2.49 0.35 0.29 0.28 0.36 3 HPLC 103 Red Crossbill Loxia_curvirostra 4 10 9 2 1.10 1.80 1.00 0.18 1.67 0.50 0.19 0.18 0.56 4 0.07 0.16 HPLC 13, 62, 64, 66, 69 Whitewinged Crossbill Loxia_leucoptera 6 21 39 6 2.45 3.71 4.00 0.33 2.88 0.48 0.14 0.12 0.58 6 0.00 0.05 HPLC2 14, 28, 62, 66 Siberian Rubythroat Luscinia_calliope 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 62, 63 Eastern striped manakin Machaeropterus_regulus 2 7 12 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Redbacked FairyWren Malurus_melanocephalus 4 19 40 9 3.01 4.21 6.00 0.35 3.49 0.42 0.20 0.16 0.51 4 0.01 0.06 HPLC 57 Goldenwinged Manakin Masius_chrysopterus 2 6 9 3 1.59 3.00 3.00 0.24 3.07 0.47 0.50 0.47 0.12 2 0.03 0.20 HPLC 31 Acorn woodpecker Melanerpes_formicivorus 2 6 5 2 1.29 1.67 2.00 0.00 1.67 0.35 0.20 0.27 0.50 2 HPLC 111 White Woodpecker Melanerpes_candidus 2 6 4 2 1.33 1.64 2.00 0.00 1.97 0.35 0.20 0.27 0.50 2 0.38 0.33 HPLC 68 Lewis's Woodpecker Melanerpes_lewis 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 68 Wild Turkey Meleagris_gallopavo 4 9 12 3 1.64 2.67 2.00 0.41 3.00 0.79 0.52 0.22 0.21 5 0.02 0.15 TLC 12 Yellow Wagtail Motacilla_flava 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Collared Grosbeak Mycerobas_affinis 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Whitewinged Grosbeak Mycerobas_carnipes 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Blackandyellow Grosbeak Mycerobas_icteroides 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Spotwinged Grosbeak Mycerobas_melanozanthos 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Star Finch Neochmia_ruficauda 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 0.50 HPLC 40 Palebellied tyrantmanakin Neopelma_pallescens 2 5 8 3 1.42 3.20 2.00 0.33 2.73 0.33 0.25 0.60 0.11 2 HPLC 105 Egyptian Vulture Neophron_percnopterus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 HPLC 51 Tristan Bunting Nesospiza_acunhae 2 6 9 3 1.53 3.00 3.00 0.24 3.07 0.47 0.50 0.47 0.12 2 0.03 0.19 TLC 58 Hihi (Stitchbird) Notiomystis_cincta 4 6 4 3 1.57 1.33 1.00 0.00 1.75 0.82 0.30 0.20 0.17 4 0.13 0.19 HPLC 17, 18 50

Blackandcrimson oriole Oriolus_cruentus 4 19 35 8 3.17 3.68 6.00 0.32 2.81 0.39 0.16 0.14 0.60 4 HPLC 112 Golden Oriole Oriolus_oriolus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Maroon oriole Oriolus_traillii 2 10 16 2 1.35 3.09 2.00 0.43 2.58 0.38 0.22 0.22 0.51 3 HPLC 112 Blackhooded Oriole Oriolus_xanthornus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Redcrested cardinal Paroaria_coronata 4 17 31 8 2.99 3.65 7.00 0.35 2.92 0.37 0.18 0.15 0.56 3 HPLC 111 Great Tit Parus_major 2 3 1 1 1.00 0.67 1.00 0.00 1.00 0.71 0.50 0.33 0.00 2 1.00 0.44 HPLC3 15, 25, 33, 34, 56, 61, 62 Yellowcheeked Tit Parus_spilonotus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Grey Partridge Perdix_perdix 2 8 13 4 1.97 3.25 3.00 0.00 2.81 0.56 0.38 0.29 0.15 4 0.04 0.31 TLC 12 Scarlet Minivet Pericrocotus_flammeus 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 62 Coal Tit Periparus_ater 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 62 Redtailed tropicbird Phaethon_rubricauda 3 14 27 8 3.09 3.86 6.00 0.43 2.90 0.40 0.23 0.19 0.50 3 HPLC 111 Ringnecked Pheasant Phasianus_colchicus 2 8 13 4 1.97 3.25 3.00 0.00 2.81 0.56 0.38 0.29 0.15 4 0.04 0.34 TLC 4, 12 Rosebreasted Grosbeak Pheucticus_ludovicianus 4 14 29 6 2.58 4.29 4.00 0.43 3.44 0.52 0.22 0.20 0.43 4 0.00 0.07 HPLC2 11, 28 Swallowtailed cotinga Phibalura_flavirostris 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Guianan red cotinga Phoenicircus_carnifex 2 2 0 3 1.78 3.14 0.00 0.42 2.83 0.28 0.33 0.43 0.00 2 HPLC 103 Andean Flamingo Phoenicopterus_andinus 3 13 23 4 1.89 3.54 4.00 0.37 3.02 0.49 0.27 0.19 0.50 4 0.01 0.17 TLC 20 Chilean Flamingo Phoenicopterus_chilensis 1 8 17 4 2.00 4.25 4.00 0.37 3.19 0.45 0.48 0.36 0.25 3 0.01 0.33 TLC 22 James's Flamingo Phoenicopterus_jamesi 3 13 23 4 1.89 3.54 4.00 0.37 3.02 0.49 0.27 0.19 0.50 4 0.01 0.17 TLC 20 Lesser Flamingo Phoenicopterus_minor 1 8 17 4 2.00 4.25 4.00 0.37 3.19 0.45 0.48 0.36 0.25 3 0.01 0.33 TLC 22 Greater Flamingo Phoenicopterus_roseus 1 8 17 4 2.00 4.25 4.00 0.37 3.19 0.45 0.48 0.36 0.25 3 0.01 0.33 TLC 22 American Flamingo Phoenicopterus_ruber 2 15 30 6 2.50 4.00 5.00 0.40 3.05 0.45 0.20 0.18 0.48 4 0.01 0.08 TLC 20, 24 Threetoed Woodpecker Picoides_tridactylus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 68 Hairy Woodpecker Picoides_villosus 4 20 36 8 2.85 3.60 6.00 0.26 3.11 0.50 0.21 0.13 0.52 4 0.02 0.07 HPLC 68 Goldenspangled piculet Picumnus_exilis 2 6 5 2 1.29 1.67 2.00 0.00 1.96 0.35 0.20 0.27 0.50 2 HPLC 111 Scalybellied Woodpecker Picus_squamatus 1 3 2 2 1.33 1.33 2.00 0.00 1.67 0.35 1.00 0.67 0.00 1 0.75 0.67 HPLC 68 Green Woodpecker Picus_viridis 4 19 35 8 2.87 3.68 6.00 0.31 3.10 0.49 0.22 0.14 0.53 4 0.01 0.07 HPLC 68 Pine Grosbeak Pinicola_enucleator 4 14 20 5 1.89 2.86 3.00 0.21 2.43 0.54 0.27 0.15 0.51 4 0.03 0.10 HPLC 62, 63, 64, 66, 69 Crimsonhooded manakin Pipra_aureola 2 7 12 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Roundtailed Manakin Pipra_chloromeros 5 17 32 8 2.91 3.76 7.00 0.35 3.39 0.57 0.25 0.15 0.47 5 0.01 0.07 TLC 29 Goldenheaded Manakin Pipra_erythrocephala 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 TLC 29 Bandtailed manakin Pipra_fasciicauda 2 7 12 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Wiretailed manakin Pipra_filicauda 2 7 12 4 2.00 3.00 4.00 0.32 2.82 0.35 0.29 0.36 0.36 2 HPLC 105 Redheaded Manakin Pipra_rubrocapilla 4 16 32 8 2.91 4.00 7.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 TLC 29 Goldenbreasted fruiteater Pipreola_aureopectus 2 2 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 2 HPLC 103 Fierythroated fruiteater Pipreola_chlorolepidota 2 2 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 2 HPLC 103 Handsome fruiteater Pipreola_formosa 2 2 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 2 HPLC 103 Redbanded fruiteater Pipreola_whitelyi 1 3 2 4 2.00 3.00 2.00 0.32 2.79 0.35 0.29 0.36 0.00 1 HPLC 103 Hepatic Tanager Piranga_flava 4 17 37 6 2.47 4.35 4.00 0.41 3.33 0.37 0.15 0.18 0.53 4 0.00 0.06 TLC 28 Western Tanager Piranga_ludoviciana 2 5 6 2 1.14 2.40 1.00 0.37 2.67 0.55 0.42 0.50 0.00 2 0.06 0.32 TLC 28 Scarlet Tanager Piranga_olivacea 4 19 40 9 3.01 4.21 6.00 0.35 3.49 0.42 0.20 0.16 0.51 4 0.01 0.06 TLC 11, 28 Summer Tanager Piranga_rubra 1 5 10 2 1.38 4.00 2.00 0.75 2.80 0.33 0.67 0.60 0.11 2 0.00 0.44 TLC 28 Roseate Spoonbill Platalea_ajaja 3 3 4 2 1.33 1.33 2.00 0.00 1.67 0.35 1.00 0.67 0.00 1 . 0.33 HPLC2 23, 63 Forest Weaver Ploceus_bicolor 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Cape Weaver Ploceus_capensis 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Village Weaver Ploceus_cucullatus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 4, 63 Nelicourvi Weaver Ploceus_nelicourvi 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Baya Weaver Ploceus_philippinus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Sakalava Weaver Ploceus_sakalava 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Africanmasked Weaver Ploceus_velatus 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 . 0.50 HPLC 63 Purplethroated cotinga Porphyrolaema_porphyrolaema 1 5 10 6 2.52 3.11 6.00 0.34 2.49 0.28 0.29 0.28 0.36 3 HPLC 103 Threewattled bellbird Procnias_tricarunculatus 1 3 4 1 1.00 1.00 0.67 2.00 1.00 0.00 1 HPLC 103 Longtailed broadbill Psarisomus_dalhousiae 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 HPLC 106 Montezuma oropendola Psarocolius_montezuma 2 2 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 2 HPLC 108 Chestnutheaded oropendola Psarocolius_wagleri 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 108 Blacknecked aracari Pteroglossus_aracari 3 11 17 5 2.09 3.09 3.00 0.28 2.31 0.43 0.24 0.20 0.48 3 HPLC 111 Jambu fruit dove Ptilinopus_jambu 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 113 Wompoo pigeon Ptilinopus_magnificus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 113 Beautiful fruit dove Ptilinopus_pulchellus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 113 Yellowbibbed fruit dove Ptilinopus_solomonensis 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 113 Redruffed fruitcrow Pyroderus_scutatus 2 10 16 2 1.30 3.20 2.00 0.51 2.50 0.34 0.25 0.24 0.50 2 HPLC 103 Goldnaped Finch Pyrrhoplectes_epauletta 1 2 1 1 1.00 1.00 1.00 0.00 1.00 0.00 . 1.00 0.00 1 1.00 0.75 OTH 38 Orange Bullfinch Pyrrhula_aurantiaca 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 OTH 38 Greyheaded Bullfinch Pyrrhula_erythaca 3 12 22 5 1.86 3.67 3.00 0.40 2.98 0.38 0.27 0.23 0.50 3 0.01 0.10 HPLC 38, 63 Redheaded Bullfinch Pyrrhula_erythrocephala 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 OTH 38 Eurasian Bullfinch Pyrrhula_pyrrhula 4 19 37 8 2.97 3.89 6.00 0.31 3.29 0.47 0.21 0.15 0.54 4 0.02 0.08 HPLC 62, 69 Cardinal Quelea Quelea_cardinalis 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 63 Redheaded Quelea Quelea_erythrops 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 63 Redbilled Quelea Quelea_quelea 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 63 Purplethroated fruitcrow Querula_purpurata 3 10 10 6 2.63 3.00 4.00 0.17 2.92 0.58 0.29 0.15 0.67 4 HPLC 103 Toco Toucan Ramphastos_toco 1 3 2 2 1.33 1.33 2.00 0.00 1.67 0.35 1.00 0.67 0.00 1 0.75 0.67 HPLC 63 51

Whitethroated toucan Ramphastos_tucanus 2 8 9 2 1.18 2.25 2.00 0.17 2.00 0.38 0.24 0.25 0.48 3 HPLC 111 Crimsonbacked Tanager Ramphocelus_dimidiatus 4 14 29 6 2.58 4.14 4.00 0.43 3.44 0.52 0.22 0.20 0.43 4 0.00 0.07 HPLC2 11, 28 Goldcrest Regulus_regulus 3 11 19 6 2.29 3.45 5.00 0.27 3.15 0.55 0.44 0.24 0.35 3 0.02 0.12 HPLC 62 Goldencrowned Kinglet Regulus_satrapa 3 18 39 9 3.01 4.33 7.00 0.37 3.48 0.38 0.21 0.17 0.51 3 0.01 0.06 HPLC 10 Desert Finch Rhodospiza_obsoleta 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 0.01 0.07 HPLC 62, 63 Goldenwinged Grosbeak Rhynchostruthus_socotranus 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 OTH 38 Andean cockoftherock Rupicola_peruvianus 1 3 4 1 1.00 2.00 0.67 2.00 0.00 0.00 1.00 0.00 2 HPLC 103 Guianan cockoftherock Rupicola_rupicola 2 10 16 3 1.39 3.09 3.00 0.46 2.45 0.33 0.22 0.22 0.54 3 HPLC 103 Guianan toucanet Selenidera_piperivora 3 11 17 5 2.09 3.09 3.00 0.28 2.31 0.43 0.24 0.20 0.48 3 HPLC 111 Common Canary Serinus_canaria 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 4, 63 Citril Finch Serinus_citrinella 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 62, 65 Yellowfronted Canary Serinus_mozambicus 1 4 6 2 1.14 3.00 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 0.06 0.44 HPLC 62, 63 Redfronted Serin Serinus_pusillus 2 9 16 2 1.30 3.56 2.00 0.62 2.74 0.28 0.25 0.31 0.50 2 0.01 0.22 HPLC 3, 62, 65 European Serin Serinus_serinus 2 9 16 2 1.30 3.56 2.00 0.62 2.74 0.28 0.25 0.31 0.50 2 0.01 0.22 HPLC 62, 64, 65 American Redstart Setophaga_ruticilla 2 9 16 2 1.30 3.56 2.00 0.62 2.74 0.28 0.25 0.31 0.50 2 0.01 0.22 OTH 38 Saffron Finch Sicalis_flaveola 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Greytailed piha Snowornis_subalaris 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Yellowbellied Sapsucker Sphyrapicus_varius 4 18 34 8 2.88 3.78 6.00 0.33 3.23 0.49 0.23 0.15 0.52 4 0.01 0.07 HPLC 68 Elegant Tern Sterna_elegans 4 8 16 5 2.33 4.00 5.00 0.37 2.55 0.37 0.33 0.32 0.36 2 0.00 0.13 TLC 27 Peruvian meadowlark Sturnella_bellicosa 2 6 5 2 1.29 1.67 2.00 0.00 1.67 0.35 0.20 0.27 0.50 2 HPLC 108 Eastern meadowlark Sturnella_magna 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 108 Redbreasted blackbird Sturnella_militaris 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 HPLC 108 Western meadowlark Sturnella_neglecta 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 108 Whitebrowed blackbird Sturnella_superciliaris 4 16 32 8 2.91 4.00 6.00 0.37 3.39 0.50 0.26 0.18 0.47 4 HPLC 108 Zebra Finch Taeniopygia_guttata 4 18 36 9 3.06 4.00 6.00 0.29 3.38 0.47 0.22 0.16 0.48 4 0.01 0.07 HPLC 41, 44 Golden Bushrobin Tarsiger_chrysaeus 2 3 3 2 1.25 2.00 1.00 0.00 1.67 0.35 1.00 0.67 0.00 1 0.00 0.33 HPLC 63 Sulfurbreasted Bushrike Telophorus_sulfureopectus 2 11 21 5 2.10 3.82 4.00 0.40 3.22 0.35 0.28 0.27 0.40 3 0.02 0.14 HPLC 63 Bokmakierie Telophorus_zeylonus 1 4 6 2 1.14 1.00 0.46 2.67 0.20 0.33 0.83 0.00 1 HPLC 111 Capercaillie Tetrao_urogallus 2 6 10 2 1.38 3.33 2.00 0.63 2.80 0.58 0.60 0.40 0.11 3 0.00 0.31 TLC 16 Wallcreeper Tichodroma_muraria 1 5 10 2 1.38 4.00 2.00 0.75 2.80 0.33 0.67 0.60 0.11 2 0.00 0.44 HPLC 62 Blackandgold Cotinga Tijuca_atra 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 103 Ecuadorian trogon Trogon_mesurus 4 16 32 8 2.91 4.00 7.00 0.37 3.39 0.50 0.26 0.18 0.47 4 HPLC 111 Eurasian Blackbird Turdus_merula 5 5 0 0 0.00 0.00 0.00 0.00 ...0.00 0.00 5 . 0.20 HPLC 19 Cassin's kingbird Tyrannus_vociferans 2 4 2 2 1.33 2.00 0.00 1.67 0.71 0.67 0.33 0.00 2 HPLC 111 Longtailed Rosefinch Uragus_sibiricus 3 13 20 6 2.27 3.08 6.00 0.23 2.82 0.60 0.38 0.18 0.42 4 0.02 0.10 HPLC 62, 63, 67, 69 Nashville Warbler Vermivora_ruficapilla 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 TLC 5 Virginia's Warbler Vermivora_virginiae 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 TLC 5 I'iwi Vestiaria_coccinea 4 16 31 8 3.09 3.88 7.00 0.37 3.08 0.45 0.19 0.17 0.51 4 HPLC 111 Yellowheaded blackbird Xanthocephalus_xanthocephalus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 HPLC 108, 109, 110 Whitewinged cotinga Xipholena_atropurpurea 4 15 23 8 3.19 3.45 5.00 0.26 3.16 0.55 0.19 0.12 0.68 4 HPLC 103 Whitetailed cotinga Xipholena_lamellipennis 4 16 28 8 3.21 3.52 5.00 0.27 3.23 0.55 0.20 0.12 0.68 5 HPLC 103 Pompadour cotinga Xipholena_punicea 4 16 29 8 3.38 3.39 8.00 0.25 3.09 0.54 0.18 0.11 0.66 5 HPLC 103,104 Japanese Whiteeye Zosterops_japonicus 1 1 0 0 0.00 0.00 0.00 0.00 . . . 0.00 0.00 1 . 1.00 OTH 38 Method: HPLC Highperformance liquid chromatography TLC Thin layer chromatography HPLC2 HPLC and TLC combined HPLC3 HPLC and Mass spectrometry OTH Mass spectrometry, others 52

Literature Sources for Appendix S2: 19 Faivre, B., Gregoire, A., Preault, M., Cezilly, F. & Sorci, G. Immune activation rapidly mirrored in a secondary sexual trait. Science 300, 103 (2003). 1 Andersson, S., Prager, M. & Johansson, E. I. A. Carotenoid 20 Fox, D. L. & Hopkins, T. S. Comparative metabolic content and reflectance of yellow and red nuptial plumages fractionation of carotenoids in three flamingo species. in widowbirds (Euplectes spp.). Functional Ecology 21, Comparative Biochemistry and Physiology 17, 841-856 272-281 (2007). (1966). 2 Arnold, K. E., Ramsay, S. L., Henderson, L. & Larcombe, 21 Fox, D. L. Carotenoids of the scarlet ibis. Comparative S. D. Seasonal variation in diet quality: antioxidants, Biochemistry and Physiology 5, 31-43 (1962). invertebrates and blue tits Cyanistes caeruleus. Biological 22 Fox, D. L., Smith, V. E. & Wolfson, A. A. Carotenoid Journal of the Linnean Society 99, 708-717 (2010). selectivity in blood and feathers of lesser (African), chilean 3 Badyaev, A. V., Belloni, V., Kennedy, L. & Delaney, R. and greater (European) flamingos. Comparative (unpubl. data). Biochemistry and Physiology 23, 225-232 (1967). 4 Brockmann, H. & Völker, O. Der gelbe Federfarbstoff des 23 Fox, D. L., Hopkins, T. S. & Zilversmit, D. B. Blood Kanarienvogels Serinus canaria canaria (L.) und das carotenoids of the roseate spoonbill. Comparative Vorkommen von Carotinoiden bei Vögeln. Hoppe-Seyler's Biochemistry and Physiology 14, 641-649 (1965). Zeitschrift fr physiologische Chemie 224, 193-215 (1934). 24 Fox, D. L., Wolfson, A. A. & McBeth, J. W. Metabolism of 5 Brush, A. H. & Johnson, N. K. The Evolution of Color b-carotene in the American flamingo, Phoenicopterus ruber. Differences between Nashville and Virginia's Warblers. Comparative Biochemistry and Physiology 29, 1223-1229 Condor 78, 412-414 (1976). (1969). 6 Brush, A. H. & Power, D. M. House finch pigmentation: 25 Hrak, P., Surai, P. F., Ots, I. & Mller, A. P. Fat soluble carotenoid metabolism and the effect of diet. Auk 93, 725- antioxidants in brood-rearing great tits Parus major: 739 (1976). relations to health and appearance. Journal of Avian Biology 7 Brush, A. H. Pigmentation in the scarlet tanager, Piranga 35, 63-70 (2004). olivacea. Condor 69, 549-559 (1967). 26 Hudon, J. & Brush, A. H. Probably dietary basis of a color 8 Brush, A. H. & Allen, K. Astaxanthin in the Cedar variant of the cedar waxwing. Journal of Field Ornithology Waxwing. Science 142, 47-48 (1963). 60, 361-368 (1989). 9 Butler, M. W. & McGraw, K. J. Relationships between 27 Hudon, J. & Brush, A. H. Carotenoids produce flush in the dietary carotenoids, body tissue carotenoids, parasite elegant tern plumage. Condor 92, 798-801 (1990). burden, and health state in wild mallard (Anas 28 Hudon, J. Unusual carotenoid use by western tanager platyrhynchos) ducklings. Arch Biochem Biophys 504, 154- (Piranga ludoviviana) and its evolutionary implications. 160, doi:10.1016/j.abb.2010.07.003 (2010). Canadian Journal of Zoology 69, 2311-2320 (1991). 10 Chui, C. K. S., McGraw, K. J. & Doucet, S. M. Carotenoid- 29 Hudon, J., Capparella, A. P. & Brush, A. H. Plumage based plumage coloration in golden-crowned kinglets pigment differences in manakins of the Pipra erythrocephala Regulus satrapa: pigment characterization and relationships superspecies Auk 106, 34-41 (1989). with migratory timing and condition. Journal of Avian 30 Hudon, J., Ouellet, H., Bénito-Espinal, . & Brush, A. H. Biology 42, 309-322 (2011). Characterization of an Orange Variant of the Bananaquit 11 Cohen, A. A., McGraw, K. J. & Robinson, W. D. Serum (Coereba flaveola) on La Désirade, Guadeloupe, French antioxidant levels in wild birds vary in relation to diet, West Indies. Auk 113, 715-718 (1996). season, life history strategy, and species. Oecologia 161, 31 Hudon, J., Anciaes, M., Bertacche, V. & Stradi, R. Plumage 673-683, doi:10.1007/s00442-009-1423-9 (2009). carotenoids of the Pin-tailed Manakin (Ilicura militaris): 12 Czeczuga, B. Carotenoids in the skin of certain species of evidence for the endogenous production of rhodoxanthin birds. Comparative Biochemistry and Physiology Part B: from a colour variant. Comparative biochemistry and Comparative Biochemistry 62, 107-109 (1979). physiology. Part B, Biochemistry & molecular biology 147, 13 del Val, E. et al. The liver but not the skin is the site for 402-411, doi:10.1016/j.cbpb.2007.02.004 (2007). conversion of a red carotenoid in a passerine bird. 32 Inouye, C. Y., Hill, G. E., Stradi, R. D., Montgomerie, R. & Naturwissenschaften 96, 797-801, doi:10.1007/s00114-009- Bosque, C. Carotenoid pigments in male house finch 0534-9 (2009). plumage in relation to age, subspecies, and ornamental 14 Deviche, P., McGraw, K. J. & Underwood, J. Season-, sex-, coloration. Auk 118, 900-915 (2001). and age-specific accumulation of plasma carotenoid 33 Isaksson, C., Ornborg, J., Prager, M. & Andersson, S. Sex pigments in free-ranging white-winged crossbills Loxia and age differences in reflectance and biochemistry of leucoptera. Journal of Avian Biology 39, 283-292 (2008). carotenoid-based colour variation in the great tit Parus 15 Eeva, T., Sillanpää, S. & Salminen, J. P. The effects of diet major. Biological Journal of the Linnean Society 95, 758- quality and quantity on plumage colour and growth of great 765 (2008). tit Parus major nestlings: a food manipulation experiment 34 Isaksson, C., Sturve, J., Almroth, B. C. & Andersson, S. The along a pollution gradient. Journal of Avian Biology 40, impact of urban environment on oxidative damage 491-499 (2009). (TBARS) and antioxidant systems in lungs and liver of 16 Egeland, E. S., Parker, H. & Liaaenjensen, S. Carotenoids in great tits, Parus major. Environ Res 109, 46-50 (2009). combs of Capercaillie (Tetrao urogallus) fed defined diets. 35 Johnson, N. K. & Brush, A. H. Analysis of Polymorphism Poultry Science 72, 747-751 (1993). in the Sooty-Capped Bush Tanager. Systematic Zoology 21, 17 Ewen, J. G. et al. Carotenoids, colour and conservation in an 245-262 (1972). endangered passerine, the hihi or stitchbird (Notiomystis 36 Juola, F. A., McGraw, K. J. & Dearborn, D. C. Carotenoids cincta). Anim Conserv 9, 229-235 (2006). and throat pouch coloration in the great frigatebird (Fregata 18 Ewen, J. G., Thorogood, R., Karadas, F., Pappas, A. C. & minor). Comparative Biochemistry and Physiology Part B: Surai, P. F. Influences of carotenoid supplementation on the Biochemistry and Molecular Biology 149, 370-377 (2008). integrated antioxidant system of a free living endangered 37 Mays Jr, H. L. et al. Sexual dichromatism in the yellow- passerine, the hihi (Notiomystis cincta). Comparative breasted chat Icteria virens: spectrophotometric analysis and Biochemistry and Physiology - Part A: Molecular & biochemical basis. Journal of Avian Biology 35, 125-134 Integrative Physiology 143, 149-154 (2006). (2004). 53

38 McGraw, K. J. in Bird Coloration. I. Mechanisms and carotenoid-based plumage traits: an experimental study. Measurements (eds G.E. Hill & K. J. McGraw) 177-242 Functional Ecology 22, 831-839 (2008). (Harvard University Press, 2006). 55 Prager, M., Johansson, E. I. & Andersson, S. Differential 39 McGraw, K. J. & Hardy, L. S. Astaxanthin is responsible ability of carotenoid C4-oxygenation in yellow and red for the pink plumage flush in Franklin's and Ring-billed bishop species (Euplectes spp.). Comparative biochemistry gulls. Journal of Field Ornithology 77, 29-33 (2006). and physiology. Part B, Biochemistry & molecular biology 40 McGraw, K. J. & Schuetz, J. G. The evolution of carotenoid 154, 373-380, doi:10.1016/j.cbpb.2009.06.015 (2009). coloration in estrildid finches: a biochemical analysis Comp. 56 Quesada, J. & Senar, J. C. Comparing plumage colour Biochem. Physiol. B 139, 45-51 (2004). measurements obtained directly from live birds and from 41 McGraw, Kevin J. & Toomey, Matthew B. Carotenoid collected feathers: the case of the great tit Parus major. Accumulation in the Tissues of Zebra Finches: Predictors of Journal of Avian Biology 37, 609-616 (2006). Integumentary Pigmentation and Implications for 57 Rowe, M. & McGraw, K. J. Carotenoids in the Seminal Carotenoid Allocation Strategies. Physiological and Fluid of Wild Birds: Interspecific Variation in Fairy-Wrens. Biochemical Zoology 83, 97-109, doi:doi:10.1086/648396 The Condor 110, 694-700, doi:10.1525/cond.2008.8604 (2010). (2008). 42 McGraw, K. J., Nolan, P. M. & Crino, O. L. Carotenoid 58 Ryan, P. G., Moloney, C. L. & Hudon, J. Color variation accumulation strategies for becoming a colourful House and hybridization among Nesospiza Buntings on Finch: analyses of plasma and liver pigments in wild inaccessible islands, Tristan da Cunha. Auk 111, 314-327 moulting birds. Functional Ecology 20, 678-688 (2006). (1994). 43 McGraw, K. J., Hill, G. E., Stradi, R. & Parker, R. S. The 59 Saino, N., Bertacche, V., Bonisoli-Alquati, A., Romano, M. Influence of Carotenoid Acquisition and Utilization on the & Rubolini, D. Phenotypic Correlates of Yolk and Plasma Maintenance of Species-Typical Plumage Pigmentation in Carotenoid Concentration in Yellow-Legged Gull Chicks. Male American Goldfinches (Carduelis tristis) and Northern Physiological and Biochemical Zoology 81, 211-225, Cardinals (Cardinalis cardinalis). Physiological and doi:doi:10.1086/527454 (2008). Biochemical Zoology 74, 843-852, doi:doi:10.1086/323797 60 Saks, L., McGraw, K. & Hrak, P. How feather colour (2001). reflects its carotenoid content. Functional Ecology 17, 555- 44 McGraw, K. J., Adkins-Regan, E. & Parker, R. S. 561 (2003). Anhydrolutein in the zebra finch: a new, metabolically 61 Sillanpää, S., Salminen, J.-P. & Eeva, T. Breeding success derived carotenoid in birds. Comparative Biochemistry and and lutein availability in great tit (Parus major). Acta Physiology Part B: Biochemistry and Molecular Biology Oecologica 35, 805-810 (2009). 132, 811-818 (2002). 62 Stradi, R. The Colour of Flight. (Solei Gruppos Editoriale 45 McGraw, K. J., Beebee, M. D., Hill, G. E. & Parker, R. S. Informatico, 1998). Lutein-based plumage coloration in songbirds is a 63 Stradi, R. in Colori in volo - il piumaggio degli uccelli (eds consequence of selective pigment incorporation into l. Brambilla, G. Canali, Mannucci E., & et al.) 117-146 feathers. Comparative Biochemistry and Physiology Part B: (Università degli Studi di Milano 1999). Biochemistry and Molecular Biology 135, 689-696 (2003). 64 Stradi, R., Celentano, G. & Nava, D. Separation and 46 McGraw, K. J., Hill, G. E. & Parker, R. S. Carotenoid identification of carotenoids in bird's plumage by high- pigments in a mutant cardinal: Implications for the genetic performance liquid chromatography-diode-array detection. and enzymatic control mechanisms of carotenoid Journal of Chromatography B: Biomedical Sciences and metabolism in birds. Condor 105, 587-592 (2003). Applications 670, 337-348 (1995). 47 McGraw, K. J., Wakamatsu, K., Clark, A. B. & Yasukawa, 65 Stradi, R., Celentano, G., Rossi, E., Rovati, G. & Pastore, K. Red-winged blackbirds Agelaius phoeniceus use M. Carotenoids in bird plumage: I. The carotenoid pattern in carotenoid and melanin pigments to color their epaulets. a series of Palearctic Carduelinae Comparative Journal of Avian Biology 35, 543-550 (2004). Biochemestry Physiology Part B: Comparative 48 McGraw, K. J., Hill, G. E. & Parker, R. S. The Biochemistry and Physiology 110, 131 -143 (1995). physiological costs of being colourful: nutritional control of 66 Stradi, R., Rossi, E., Celentano, G. & Bellardi, B. carotenoid utilization in the American goldfinch, Carduelis Carotenoids in bird plumage: the pattern in three Loxia tristis. Animal Behaviour 69, 653-660, species and in Pinicola enucleator. Comparative doi:10.1016/j.anbehav.2004.05.018 (2005). Biochemistry and Physiology Part B: Biochemistry and 49 Negro, J. J. & Garrido-Fernández, J. Astaxanthin is the Molecular Biology 113, 427-432 (1996). major carotenoid in tissues of white storks (Ciconia ciconia) 67 Stradi, R., Celentano, G., Boles, M. & Mercato, F. feeding on introduced crayfish (Procambarus clarkii). Carotenoids in Bird Plumage: The Pattern in a Series of Comparative Biochemistry and Physiology Part B: Red-Pigmented Carduelinae. Comparative Biochemistry and Biochemistry and Molecular Biology 126, 347-352 (2000). Physiology Part B: Biochemistry and Molecular Biology 50 Negro, J. J., Tella, J. L., Hiraldo, F., Bortolotti, G. R. & 117, 85-91 (1997). Prieto, P. Sex- and age-related variation in plasma 68 Stradi, R., Hudon, J., Celentano, G. & Pini, E. Carotenoids carotenoids despite a constant diet in the red-legged in bird plumage: the complement of yellow and red partridge (Alectoris rufa). Ardea 89, 275-279 (2001). pigments in true woodpeckers (Picinae). Comparative 51 Negro, J. J. et al. Coprophagy: An unusual source of Biochemistry and Physiology Part B: Biochemistry and essential carotenoids. Nature 416, 807-808 (2002). Molecular Biology 120, 223-230 (1998). 52 Pérez, C., Lores, M. & Velando, A. Availability of 69 Stradi, R., Pini, E. & Celentano, G. Carotenoids in bird nonpigmentary antioxidant affects red coloration in gulls. plumage: the complement of red pigments in the plumage Behavioral Ecology 19, 967-973, of wild and captive bullfinch (Pyrrhula pyrrhula). doi:10.1093/beheco/arn053 (2008). Comparative Biochemistry and Physiology Part B 128, 529- 53 Peters, A., Delhey, K., Denk, A. G. & Kempenaers, B. 535 (2001). Tarde-offs between immune investment and sexual 70 Toomey, M. B. & McGraw, K. J. Seasonal, sexual, and signaling in male mallards. American Naturalist 164, 51-59 quality related variation in retinal carotenoid accumulation (2004). in the house finch (Carpodacus mexicanus). Functional 54 Peters, A., Delhey, K., Andersson, S., Van Noordwijk, H. & Ecology 23, 321-329 (2009). Förschler, M. I. Condition-dependence of multiple 54

71 Toomey, M. B. & McGraw, K. J. The effects of dietary solid-state films, and solution. Archives of Biochemistry carotenoid intake on carotenoid accumulation in the retina and Biophysics 539:142-155. of a wild bird, the house finch (Carpodacus mexicanus). Arch Biochem Biophys 504, 161-168 (2010). 72 Tyczkowski, J. K., Yagen, B. & Hamilton, P. B. Metabolism of canthaxanthin, a red diketocarotenoid, by chickens. Poultry Science 67, 787-793 (1988). 100. García-de Blas, E., R. Mateo, J. Viuela, L. Pérez- Rodríguez, and C. Alonso-Alvarez. 2013. Free and esterified carotenoids in ornaments of an avian species: the relationship to color expression and sources of variability. Physiological and Biochemical Zoology 86:483-498. 101. García-de Blas, E., R. Mateo, F. Guzmán Bernardo, R. Rodríguez Martín-Doimeadios, and C. Alonso-Alvarez. 2014. Astaxanthin and papilioerythrinone in the skin of birds: a chromatic convergence of two metabolic routes with different precursors? Naturwissenschaften 101:407-416. 102. Mendes-Pinto, M. M., A. M. LaFountain, M. C. Stoddard, R. O. Prum, H. A. Frank, and B. Robert. 2012. Variation in carotenoid–protein interaction in bird feathers produces novel plumage coloration. Journal of The Royal Society Interface 9:3338-3350. 103. Prum, R. O., A. M. LaFountain, J. Berro, M. C. Stoddard, and H. A. Frank. 2012. Molecular diversity, metabolic transformation, and evolution of carotenoid feather pigments in cotingas (Aves: Cotingidae). J Comp Physiol B 182:1095-1116. 104. LaFountain, A. M., S. Kaligotla, S. Cawley, K. M. Riedl, S. J. Schwartz, H. A. Frank, and R. O. Prum. 2010. Novel methoxy-carotenoids from the burgundy-colored plumage of the Pompadour Cotinga Xipholena punicea. Archives of Biochemistry and Biophysics 504:142-153. 105. Hudon, J., A. Storni, E. Pini, M. Anciães, and R. Stradi. 2012. Rhodoxanthin as a Characteristic Keto-Carotenoid of Manakins (Pipridae). The Auk 129:491-499. 106. Prum, R., A. LaFountain, C. Berg, M. Tauber, and H. Frank. 2014. Mechanism of carotenoid coloration in the brightly colored plumages of broadbills (Eurylaimidae). J Comp Physiol B 184:651-672. 107. Friedman, N. R., K. J. McGraw, and K. E. Omland. 2014. History and mechanisms of carotenoid plumage evolution in the New World orioles (Icterus). Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 172–173:1-8. 108. Friedman, N. R., K. J. McGraw, and K. E. Omland. 2014. Evolution of carotenoid pigmentation in Caciques and Meadowlarks (Icteridae): repeated gains of red plumage coloration by carotenoid C4-oxygenation. Evolution 68:791- 801. 109. Newbrey, J. L., W. L. Reed, S. P. Foster, and G. L. Zander. 2008. Laying-Sequence Variation in Yolk Carotenoid Concentrations in Eggs of Yellow-Headed Blackbirds (Xanthocephalus xanthocephalus). The Auk 125:124-130. 110. Newbrey, J. L. and W. L. Reed. 2009. Growth of yellowheaded blackbird Xanthocephalus xanthocephalus nestlings in relation to maternal body condition, egg mass, and yolk carotenoids concentrations. Journal of Avian Biology 40:419-429. 111. Thomas, D. B., K. J. McGraw, H. F. James, and O. Madden. 2014. Non-destructive descriptions of carotenoids in feathers using Raman spectroscopy. Analytical Methods 6:1301-1308. 112. LaFountain, A. M., H. A. Frank, and R. O. Prum. 2013. Carotenoids from the crimson and maroon plumages of Old World orioles (Oriolidae). Archives of Biochemistry and Biophysics 539:126-132. 113. Berg, C. J., A. M. LaFountain, R. O. Prum, H. A. Frank, and M. J. Tauber. 2013. Vibrational and electronic spectroscopy of the retro-carotenoid rhodoxanthin in avian plumage, 55

Appendix S3. Module assignments in the avian subset of the global carotenoid metabolic network CAROTENOID Module Assignment lutein 1 (3R, 3'R) zeaxanthin 5 β-carotene 6 β-cryptoxanthin 6 anhydrolutein 1 7,8-dihydrolutein 1 9-Z-7,8-dihydrolutein 1 canary xanthophyll A 1 canary xanthophyll B 1 α-doradexanthin 3 (3S,4R,3'R,6'R) 4-hydroxylutein 3 fritschiellaxanthin 3 papilioerythrinone 3 3'-dehdyrolutein 1 piprixanthin 4 rhodoxanthin 4 7,8,7',8'-tetrahydrozeaxanthin 5 7,8-dihydrozeaxanthin 5 idoxanthin 5 fucoxanthin 8 7,8 dihydro β-cryptoxanthin 6 4-hydroxyzeaxanthin 5 adonixanthin 5 13 cis-(3S,3'S) astaxanthin 5 echineone 6 3'-hydroxyechinenone 6 canthaxanthin 6 adonirubin 6 4-hydroxy-echinenone 6 isozeaxanthin 6 β-isocryptoxanthin 6 α-carotene 2 α-isocryptoxanthin 2 phoenicopterone 2 α-cryptoxanthin 2 rubixanthin 9 4-oxo-rubixanthin 9 gazaniaxanthin 10 4-oxo-gazaniaxanthin 10 (3S,4R,3'S,6'R) 4-hydroxylutein 3 cis lutein 1 resonance stabilized form 4 xipholenin 3 2,3-didehydro-xipholenin 3 rupicolin 5 3'-hydroxy-3-methoxy-canthaxanthin 7 pompadourin 7 2,3-didehydro-pompadourin 7 cotingin 7 brittonxanthin 6 cymbirhynchin 3 7,8-dihydro-3'-dehydrolutein 1 4-hydroxy-canary xanthophyll A 1 Module numbers correspond to the paritioned regions in Fig. 2. Compounds that were not found in the plumage and/or integument of birds were not included. See Methods for details on how modules were assigned. 56

APPENDIX B. THE LANDSCAPE OF EVOLUTION: RECONCILING STRUCTURAL AND DYNAMIC PROPERTIES OF METABOLIC NETWORKS IN ADAPTIVE DIVERSIFICATIONS

Published citation: Integrative and Comparative Biology (2016) 56: 235-246

1/23/2017 RightsLink Printable License 1/23/2017 RightsLink Printable License 2. This permission covers the use of the material in the English language in the following OXFORD UNIVERSITY PRESS LICENSE TERMS AND CONDITIONS territory: world. If you have requested additional permission to translate this material, the terms and conditions of this reuse will be set out in clause 12. Jan 23, 2017 3. This permission is limited to the particular use authorized in (1) above and does not allow you to sanction its use elsewhere in any other format other than specified above, nor does it apply to quotations, images, artistic works etc that have been reproduced from other sources This Agreement between Erin S Morrison ("You") and Oxford University Press ("Oxford which may be part of the material to be used. University Press") consists of your license details and the terms and conditions provided by 4. No alteration, omission or addition is made to the material without our written consent. Oxford University Press and Copyright Clearance Center. Permission must be re-cleared with Oxford University Press if/when you decide to reprint. 5. The following credit line appears wherever the material is used: author, title, journal, year, License Number 4035030265141 volume, issue number, pagination, by permission of Oxford University Press or the License date Jan 23, 2017 sponsoring society if the journal is a society journal. Where a journal is being published on behalf of a learned society, the details of that society must be included in the credit line. Licensed content publisher Oxford University Press 6. For the reproduction of a full article from an Oxford University Press journal for whatever Licensed content publication Integrative and Comparative Biology purpose, the corresponding author of the material concerned should be informed of the Licensed content title The Landscape of Evolution: Reconciling Structural and Dynamic proposed use. Contact details for the corresponding authors of all Oxford University Press Properties of Metabolic Networks in Adaptive Diversifications journal contact can be found alongside either the abstract or full text of the article concerned, accessible from www.oxfordjournals.org Should there be a problem clearing these rights, Licensed content author Morrison, Erin S.; Badyaev, Alexander V. please contact [email protected] Licensed content date 2016-06-01 7. If the credit line or acknowledgement in our publication indicates that any of the figures, Type of Use Thesis/Dissertation images or photos was reproduced, drawn or modified from an earlier source it will be necessary for you to clear this permission with the original publisher as well. If this Institution name permission has not been obtained, please note that this material cannot be included in your Title of your work Evolution of the structural and dynamic properties of metabolic publication/photocopies. networks 8. While you may exercise the rights licensed immediately upon issuance of the license at Publisher of your work n/a the end of the licensing process for the transaction, provided that you have disclosed Expected publication date May 2017 complete and accurate details of your proposed use, no license is finally effective unless and until full payment is received from you (either by Oxford University Press or by Copyright Permissions cost 0.00 USD Clearance Center (CCC)) as provided in CCC's Billing and Payment terms and conditions. If Value added tax 0.00 USD full payment is not received on a timely basis, then any license preliminarily granted shall be Total 0.00 USD deemed automatically revoked and shall be void as if never granted. Further, in the event that you breach any of these terms and conditions or any of CCC's Billing and Payment Requestor Location Erin S Morrison terms and conditions, the license is automatically revoked and shall be void as if never University of Arizona Dept of Ecology & Evolutionary Biology granted. Use of materials as described in a revoked license, as well as any use of the 1041 E. Lowell St. BioSci West Rm 310 materials beyond the scope of an unrevoked license, may constitute copyright infringement TUCSON, AZ 85721 and Oxford University Press reserves the right to take any and all action to protect its United States copyright in the materials. Attn: Erin S Morrison 9. This license is personal to you and may not be sublicensed, assigned or transferred by you Publisher Tax ID GB125506730 to any other person without Oxford University Press’s written permission. Billing Type Invoice 10. Oxford University Press reserves all rights not specifically granted in the combination of (i) the license details provided by you and accepted in the course of this licensing Billing Address Erin S Morrison transaction, (ii) these terms and conditions and (iii) CCC’s Billing and Payment terms and University of Arizona Dept of Ecology & Evolutionary Biology conditions. 1041 E. Lowell St. BioSci West Rm 310 11. You hereby indemnify and agree to hold harmless Oxford University Press and CCC, and TUCSON, AZ 85721 their respective officers, directors, employs and agents, from and against any and all claims United States arising out of your use of the licensed material other than as specifically authorized pursuant Attn: Erin S Morrison to this license. Total 0.00 USD 12. Other Terms and Conditions: Terms and Conditions v1.4 Questions? [email protected] or +1-855-239-3415 (toll free in the US) or STANDARD TERMS AND CONDITIONS FOR REPRODUCTION OF MATERIAL +1-978-646-2777. FROM AN OXFORD UNIVERSITY PRESS JOURNAL 1. Use of the material is restricted to the type of use specified in your order details. 57 https://s100.copyright.com/AppDispatchServlet 1/3 https://s100.copyright.com/AppDispatchServlet 2/3 Integrative and Comparative Biology 58 Integrative and Comparative Biology, volume 56, number 2, pp. 235–246 doi:10.1093/icb/icw026 Society for Integrative and Comparative Biology

SYMPOSIUM

The Landscape of Evolution: Reconciling Structural and Dynamic Properties of Metabolic Networks in Adaptive Diversifications Erin S. Morrison1 and Alexander V. Badyaev Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ 85721-0001, USA From the symposium ‘‘Evolutionary Endocrinology: Hormones as Mediators of Evolutionary Phenomena’’ at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2016 at Portland, Oregon. 1E-mail: [email protected]

Synopsis The network of the interactions among genes, proteins, and metabolites delineates a range of potential phenotypic diversifications in a lineage, and realized phenotypic changes are the result of differences in the dynamics of the expression of the elements and interactions in this deterministic network. Regulatory mechanisms, such as hormones, mediate the relationship between the structural and dynamic properties of networks by determining how and when the elements are expressed and form a functional unit or state. Changes in regulatory mechanisms lead to variable expression of functional states of a network within and among generations. Functional properties of network elements, and the magnitude and direction of evolutionary change they determine, depend on their location within a network. Here, we examine the relationship between network structure and the dynamic mechanisms that regulate flux through a metabolic network. We review the mechanisms that control metabolic flux in enzymatic reactions and examine structural properties of the network locations that are targets of flux control. We aim to establish a predictive framework to test the contributions of structural and dynamic properties of deterministic networks to evolutionary diversifications.

Assessing the role of regulatory maintaining global structural properties, defined by mechanisms in phenotypic the topology and connectivity of the entire network diversification (Albert et al. 2000; Jeong et al. 2001; Schmidt et al. 2003; Vitkup et al. 2006), or to the distinct functional The link between the topology of genomic, proteo- roles of elements and interactions independent of mic, and metabolic network elements and the dy- their structural positions (Papp et al. 2004; Almaas namic properties of their interactions is crucial for et al. 2005; Mahadevan and Palsson 2005; Vitkup et the stability of a phenotype and opportunities for al. 2006). These contrasting explanations reflect the evolutionary diversification. The structure of all of debate as to whether selection acts on structure of the possible functional relationships between genes, deterministic networks or on distinct functional proteins, enzymes, and metabolites defines a deter- states within a network (Wagner 2007; Papp et al. ministic network, in which each distinct functional 2009). This distinction is the focus of our review. state corresponds to a potential phenotype (Box 1; Changes in structural properties of a network— Schuster et al. 2000; Covert and Palsson 2002; Alon such as in element connectivity and pathway length 2003; Baraba´si and Oltvai 2004; Covert et al. 2004). During diversifications within a lineage, some ele- (Box 1; Costa et al. 2007)—are determined by the ments and interactions of deterministic networks physical gain or loss of elements and interactions as a remain unchanged, whereas others vary widely result of gene duplications (Wagner 2001; Va´zquez across taxa (Fraser et al. 2002; Almaas et al. 2005; et al. 2003; Kondrashov 2012), mutations in existing Hahn and Kern 2005; Light et al. 2005; Bernhardsson genes (Wagner 2003; Berg et al. 2004), or horizontal et al. 2011; Badyaev et al. 2015). The difference in gene transfers (Light et al. 2005; Pa´l et al. 2005; evolutionary conservation of elements and interac- Klassen 2010). In contrast, functional states—such tions may be related either to their roles in as the rate of chemical reactions and the levels of

Advanced Access publication June 1, 2016 ß The Author 2016. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: [email protected]. 59 236 E. S. Morrison and A. V. Badyaev

fundamental constraints placed on the chemical and physical properties of elements and interactions in the network, the efficacy of functional states varies across environments: Some states are more locally optimal than others in dynamic properties such as energy consumption or reaction rates (Westerhoff et al. 1984; Ortega and Acerenza 1998; Ibarra et al. 2002; Price et al. 2004). The dynamic properties of a deterministic network determine how the elements and interactions in the network are utilized in a particular environment (Box 1). Thus, knowledge of the relationship between dynamic and structural properties of the network is needed to assess their roles in evolutionary diversification. Regulatory mechanisms control expression of functional states in a deterministic network (Almaas et al. 2004; Papp et al. 2004; Price et al. 2004; Reed and Palsson 2004; Almaas et al. 2005; Nam et al. 2012) and, therefore, the evolutionary potential of changes in functional Box. 1 Structural and dynamic properties of deterministic net- states and network expression across taxa works. (Westerhoff et al. 1984; Heinrich et al. 1991; Fell Deterministic network represents all possible interactions (gray 1997; Edwards et al. 2001; Ibarra et al. 2002; lines) among elements such as genes, proteins, enzymes, and Davidson and Erwin 2010). The efficacy of regulatory metabolites (gray circles) that could underlie a phenotype. Structural network properties describe the organization and changes in functional states in the network, however, location of the interactions among elements in a network. depends on topological locations of the regulatory Pathway length is the number of interactions between elements mechanisms within a deterministic network (Erwin (e.g., pathway length between A and I is four interactions), and and Davidson 2009). connectivity is the number of interactions per element (e.g. Here, we examine whether some topological posi- connectivity of B is three interactions). tions within a network are more likely to be regu- Dynamic network properties define interactions and elements that are more likely to be co-expressed and the strength of this lated than others to determine whether regulatory expression under different conditions. This can be due to abiotic changes related to structural properties produce dis- factors or internal regulation. Co-expressed elements and inter- tinct phenotypic changes. We focus on metabolic actions represent functional states. In the figure, the black pat- networks, because complete topologies of these net- terned outlines represent functional states of the network that works are now available for many species (Edwards are expressed in different species. and Palsson 1999, 2000a; Kanehisa et al. 2014). The Calibrating phenotypic differences on a deterministic network: study of metabolic flux—the rate of enzymatic reac- The relationship between structural and dynamic properties of a deterministic network can be used to calibrate differences be- tions across a network—provides an opportunity to tween its functional states. In the figure, the network of species 2 relate changes in enzyme activity at particular topo- differs from both of the networks of species 1 and 3 by two logical positions to phenotypic plasticity in the use of interactions and elements, whereas the networks of species 1 and a network within an individual and to phenotypic 3 differ by three interactions and elements. The current dynamic change across generations. We integrate knowledge properties of the network establish that the interaction between of topological properties of optimal flux control D and G is more likely than other interactions between the el- with studies that examine variation in flux caused ements expressed in these species, such as the interaction between D and E, which is not present in any of the species’ by diverse abiotic and biotic conditions (Kacser networks. and Burns 1981; Fell 1997; Ibarra et al. 2002; Segre` et al. 2002; Dekel and Alon 2005). We first present an overview of the mechanisms that can control flux in metabolic pathways. We then gene expression (Fell 1997; de la Fuente et al. 2002; review structural properties in metabolic networks Kuznetsov et al. 2002; Farkas et al. 2003)—are deter- that are associated with locations of optimal flux mined by the physical and chemical properties of control, and assess the impact of differences in the elements and interactions in a deterministic network topological locations of regulatory controls on ex- (Westerhoff et al. 1984; Ibarra et al. 2002). Due to pression of functional states. Evolution of deterministic networks 237 60

Mechanisms of metabolic flux control Chubukov et al. 2013; Schwender et al. 2014; Machado et al. 2015). Flux through a pathway is regulated by enzyme ac- In addition to changes to enzyme activity, meta- tivity and production (Fell 1992, 1997; Rossell et al. bolic flux can be regulated by transcriptional and 2006). Differences in the availability of the initial translational controls involved in enzyme produc- substrates of metabolic pathways and the affinity of tion. Transcription rates of specific enzymes in path- enzymes for these substrates contribute to flux vari- ways can vary widely in response to metabolite ation. Substrate concentrations over certain threshold concentrations (Goelzer et al. 2008; Bradley et al. levels activate enzymes, followed by an increase in 2009), signaling molecules (Cho et al. 2008), or abi- reaction rates until enzymes become saturated otic environmental perturbations (Gasch et al. 2000; (Matsuno et al. 1978; Bongaerts and Vliegenthart Causton et al. 2001; Enjalbert et al. 2006). The reg- 1988). Flux in a metabolic network can change rap- ulation of translation is also dependent on the rela- idly and reversibly, caused by changes in the avail- tive stability of mRNA transcripts (Smolke et al. ability of initial, often external, substrates (Nasution 2000, 2001; Bennett et al. 2008; Wang et al. 2015), et al. 2006; Wu et al. 2006; Taymaz-Nikerel et al. such that heritability of transcription rates can con- 2011; Taymaz-Nikerel et al. 2013), or due to short- tribute to the evolutionary stability of flux control term fluctuations in the enzyme affinity for the sub- (Emilsson et al. 2008; Gordon and Ruvinsky 2012; strate (e.g., caused by temperature or pH changes; Schaefke et al. 2013). Mechanisms of flux control can Dixon 1953; Szasz 1974; Bongaerts and Vliegenthart thus underlie both short-term, reversible changes and 1988; Saavedra et al. 2005; Sørensen et al. 2015). Flux more permanent, evolutionary changes in the expres- can be permanently changed, however, as the result sion of metabolic pathways. Below we review the of irreversible modifications to the enzyme that optimal placement of flux control in a metabolic changes its affinity for a substrate, often caused by network and assess how topological locations of reg- changes in the physical structure of the enzyme, such ulatory mechanisms affect short- and long-term evo- as due to mutation (Lamb et al. 1997). lutionary diversification. Metabolic flux is also affected by the allosteric regulation of enzymes, in which enzymes are acti- Integration of flux control mechanisms vated or deactivated by reversible covalent modifica- into the static structure of metabolic tions. Allosteric regulation adjusts enzyme activity to changes in abiotic and biotic environments of the networks metabolic network (Ralser et al. 2009; Link et al. Relationship between functional modularity and flux 2013). Feedback inhibition of enzymes by other me- control in metabolic networks tabolites is one of the ways allosteric regulation can Groups of enzymes and compounds that are inter- be accomplished (Umbarger 1956; Yates and Pardee linked by stronger regulatory mechanisms than other 1956). In these cases, metabolites produced at the elements in the network form a functional module end of pathways bind to the enzymes at the begin- (Hartwell et al. 1999). The coordinated regulation of ning of pathways and deactivate the enzymatic reac- enzymatic reactions within functional modules could tions to limit the further production of downstream be the result of optimizing flux control for a meta- compounds. Alternatively, covalent modifications bolic network functioning in a wide range of envi- might be caused by protein complexes binding to ronments. Indeed, species that are frequently exposed specific enzymes. For example, protein kinases and to many substrates and abiotic factors tend to have phosphoprotein phosphates activate or inhibit en- networks with greater structural and functional mod- zymes via phosphorylation and dephosphorization ularity (Borenstein et al. 2008; Kreimer et al. 2008). (Krebs and Beavo 1979). Protein complexes Alternatively, the coordinated regulation of enzymes themselves can be regulated by abiotic factors in a pathway could evolve to prevent their unneces- (Kaufmann et al. 1999; Jarmuszkiewicz et al. 2015), sary buildup in the limited volume of a cell, which hormones (Cohen 1988; Stra˚lfors and Honnor 1989), could occur if enzymes were regulated independently growth factors (Lee et al. 1991; Kholodenko et al. (Ellis 2001; Minton 2001; Wessely et al. 2011; de 1999), or neural impulses (Wang et al. 1988; Hijas-Liste et al. 2015). Bauerfeind et al. 1997). Variation in these factors Several models have been proposed to specifically can underlie adaptive responses of metabolic flux link flux control to network topology and functional without permanently altering pathway structure or modularity of metabolites and enzymes. One model the structure of the enzyme itself (ter Kuile and proposes that the coordination of flux through mul- Westerhoff 2001; Heinemann and Sauer 2010; tiple enzymes is controlled by a single rate-limiting 61 238 E. S. Morrison and A. V. Badyaev enzyme at the beginning of a pathway to prevent the Independent flux controls at the beginning buildup of intermediate metabolites (Blackman 1905; and end of pathways Krebs 1957). Indeed, enzymes located at the begin- The coordinated regulation and expression of enzy- ning of pathways tend to evolve greater flux control matic reactions may not extend over the entire path- than downstream enzymes (Eanes et al. 2006; Wright way. Uncoordinated regulation of flux along a and Rausher 2010; Olson-Manning et al. 2013; pathway modulates the synthesizing capacity of a Olson-Manning et al. 2015). Another model posits pathway, such that different parts of the network that rate-limiting enzymes are uncommon, such can respond to changing conditions independently that the regulation of flux is distributed across the and only the flux related to certain elements and enzymes along a pathway (Kacser and Burns 1973; interactions in a pathway is altered. Greater numbers Heinrich and Rapoport 1974; Fell 1992; Fell and of independent flux controls in a pathway are thus Thomas 1995; Rossell et al. 2006). In this case, expected in fluctuating environments (Soyer and changes in the activity of a single enzyme do not Pfeiffer 2010). Also, multiple regulatory mechanisms affect the flux in a pathway (Van Hoek et al. 1998; in the same pathway should be most optimal for Nilsson et al. 2001; Daran-Lapujade et al. 2004) and longer, linear pathways, because of a time delay multiple enzymes are all controlled by the same reg- before changes in the flux of an upstream enzyme ulatory mechanisms (Thomas and Fell 1998; Wessely reaches the end of a pathway (Seshasayee et al. 2009; et al. 2011; de Hijas-Liste et al. 2015). Although the Wessely et al. 2011). Independent flux controls at the targets of regulatory mechanisms differ between these first and last enzymes in a pathway mitigates these models, both ultimately result in the coordinated con- metabolic time delays (Klipp et al. 2002; McAdams trol of groups of multiple enzymes that are not asso- and Shapiro 2003; Zaslaver et al. 2004). For example, ciated with any structural property of a network differences in the transcriptional regulation of the (Pfeiffer et al. 1999; Ravasz et al. 2002; Schuster et first and last enzymes in a pathway resulted in al. 2002; Spirin et al. 2003; Ihmels et al. 2004; Kharchenko et al. 2005; C¸akir et al. 2006; Seshasayee lesser co-expression of enzymatic reactions in rela- et al. 2009; Zelezniak et al. 2014). tion to the distance between reactions (Spirin et al. In this case, functional modules, and not enzymes 2003; Ma et al. 2004; Kharchenko et al. 2005; Yu and in specific topological positions, are therefore the Gerstein 2006; Notebaart et al. 2008; Seshasayee et al. source of metabolic diversification on a biochemical 2009; Wessely et al. 2011). Similarly, the initial and network (Wagner and Altenberg 1996; Nagy 1998; terminating enzymes can be regulated by different von Dassow and Munro 1999; Raff and Raff 2000; mechanisms, such as when the last enzyme is regu- Badyaev and Foresman 2000; Badyaev 2007). Within lated by transcriptional or translational factors, an individual, changes in flux affect all metabolites in whilst the first enzyme—via feedback inhibition a functional module, but the relative proportions of based on the concentration of the last metabolite the enzymes remain constant due to their coordi- in the pathway (Moxley et al. 2009; de Hijas-Liste nated activity and expression (Fell and Thomas et al. 2015). 1995; Rossell et al. 2006). Thus, if the metabolic net- When distinct flux controls are located along a work is portioned into functional modules under pathway, changes in flux depend on the topological different regulatory mechanisms, then changes in positions of the enzymes in the pathways (Fig. 1B). the flux of enzymes will be unrelated to their struc- For example, the flux of enzymes located in up- tural positions, because the relative changes in flux of stream positions should change at a different rate enzymes in the same functional module will be con- than the enzymes located further downstream when stant (Fig. 1A). Targeted regulation of functional these enzymes are regulated by different control mech- modules leads to environment-specific expression of anisms. From an evolutionary perspective, the pres- these modules (Almaas et al. 2004; Papp et al. 2004; ence of several regulatory mechanisms may represent Reed and Palsson 2004). Indeed, the gain and loss of different ways in which the same pathway can be enzymes within a module co-occurred over evolutionary optimized to function in different environments. time, because proteins in the same functional module For example, in the aliphatic glucosinolate pathway tended to co-evolve at the same rate (Campillos et al. of Arabidopsis thaliana, flux was controlled by the 2006; Chen and Dokholyan 2006). When selection tar- first enzyme in the pathway in almost all environ- gets the coordinated regulation of a functional module, ments, but different regulatory factors governed en- structural positions of enzymes should not be related to zymes located further downstream (Olson-Manning evolutionary conservation of enzymes across species et al. 2015). Stabilizing selection was evident only in (Fig. 1A). the first enzyme in the pathway that had the largest Evolution of deterministic networks 239 62

Fig. 1 Structural locations of the control of metabolic flux affect phenotypic changes in the rate of enzymatic reactions in a biochemical pathway and the evolutionary diversification of metabolic networks across species. The circles and arrows on the left represent metabolites and enzymes, respectively, in a biochemical pathway. The solid and dashed arrows denote structural locations of distinct regulatory mechanisms on enzymes in a pathway. In the graphs on the right, proximate flux changes are measured by changes in the expression level of compounds in a pathway (flux change), and the evolutionary diversification of a compound in a pathway is determined by the number of species in a lineage that express the compound (species representation). To capture the structural positions of a compound in a network, connectivity measures the number of reactions per compound and pathway position is the number of reactions that separate a compound from the beginning of a pathway (Box 1). (A) When one regulatory mechanism coordinates the flux of all of the enzymes in a pathway, then the structural position of a compound in a pathway does not matter, because all of the compounds in the pathway will experience the same magnitude of flux change. Compounds that are part of the same functional module will be targeted by selection as a unit and will thus be gained or lost together across species with no relation to their structural positions. (B) When independent regulatory mechanisms control flux in different locations of the same pathway, changes in the flux of compounds will be related to their pathway position. In pathways with multiple regulatory controls, upstream compounds that are located fewer reactions away from the beginning of pathways have a large impact on changes in flux, and tend to be under stabilizing selection. Compounds located several reactions from the starting point of a pathway do not have a significant influence on the overall flux, and thus divergent selection should be stronger on compounds located at the end of pathways. (C) When there are different regulatory controls for pathways that either converge or diverge from the same compound (branch point), flux changes should be related to the connectivity of a compound. Due to their participation in multiple pathways, branch points have the greatest connectivity in a biochemical network, and thus tend to have a greater influence on metabolic flux than less connected compounds only associated with one pathway. Therefore, over evolutionary time, we would predict that the most connected compounds in the network will be conserved while divergent selection would occur among compounds with fewer enzymatic reactions that contribute less to flux control in pathways. influence on the overall flux in the pathway (Olson- Bernhardsson et al. 2011). It follows that when mul- Manning et al. 2013). Several studies have docu- tiple regulatory controls in a pathway are under se- mented distinct selection on enzymes in different lection, the upstream elements of the pathways locations in a pathway: Central or upstream enzymes should be conserved across species and downstream and metabolites that contributed more to the control elements should diverge (Fig. 1B). of flux in pathways tended to be under stabilizing selection, while downstream or terminal enzymes Flux control at branching points in that had less of an influence on the flux in pathways metabolic pathways were under divergent selection (Rausher et al. 1999; Locations within a metabolic network where separate Ramsay et al. 2009; Wright and Rausher 2010; pathways either converge to produce the same 63 240 E. S. Morrison and A. V. Badyaev metabolite or diverge from the same precursor can adaptive evolution of metabolism is supported by be targets of metabolic flux control. Metabolic con- the finding that branching point enzymes tend to trol theory predicts that the interactions between the occur in locations of optimal flux control (Eanes pathways at branching points should result in dis- 1999; Flowers et al. 2007; Rausher 2013). The metab- tinct patterns of flux control (Kacser 1983; Fell and olites that anchor these branch points tend to be Sauro 1985; Heijnen et al. 2004). At branching conserved over evolutionary time, whereas the less points, there is often a decoupling of regulation be- connected compounds within the pathways that tween incoming and outgoing reactions, and one either converge or diverge from the same highly con- pair of incoming and outgoing reactions from the nected branch point metabolite often experience di- shared metabolite tends to be more optimal than vergent selection (Fig. 1C; Fraser et al. 2002; Hahn another potential reaction pair (LaPorte et al. 1984; and Kern 2005; Bernhardsson et al. 2011; Badyaev et Heinrich et al. 1991; Stephanopoulos and Vallino al. 2015). 1991; Vallino and Stephanopoulos 1994; Spirin et al. 2003; Ihmels et al. 2004; Notebaart et al. 2008; Implications of the relationship between Seshasayee et al. 2009). When the flux is optimized structural and dynamic properties in in one of the branching pathways, it may shut down metabolic networks the expression of the other pathway (Kacser and Burns 1981) or, alternatively, the presence of two Examination of flux regulation in relation to the converging pathways can be advantageous when it structural properties of deterministic networks pro- allows for a greater flux into the following down- vides a way to understand proximate mechanisms of stream enzyme (Heinrich et al. 1991). Metabolites phenotypic change from a more global perspective. at branching points have high connectivity, such Instead of only being able to see where and how that these metabolites are associated with more in- changes occurred with respect to a current pheno- coming and outgoing enzymatic reactions compared type, we can begin to understand why certain phe- with the metabolites in less connected parts of the notypic changes are recurrent, whereas others are network. Given that many changes in flux are asso- rarely realized. As such, this approach links micro- ciated with branch points in pathways, changes in evolutionary and macroevolutionary changes. The the regulatory control of highly connected enzymes effect of network structure on the delineation of di- and metabolites should contribute more to changes versification opportunities depends on the integra- in flux than less connected metabolites and enzymes tion of regulatory mechanisms into the network (Zhang et al. 2007; Fig. 1C). structure. When entire modules of a metabolic net- Branching points in metabolic networks enable the work are under the same regulatory mechanism, the robustness of metabolism to the loss of pathways due network structure is an emergent property in the to environmental or genetic perturbations, or they evolutionary change of metabolism. In this case, can lead to specialization in the same environment the metabolism in the biochemical network is opti- (Edwards and Palsson 2000b; Zhang et al. 2007; Vogt mized to current abiotic and biotic factors and does 2010; Chen et al. 2011; Weng 2014). The redundancy not depend on the topology of enzymes. When there inherent in convergent branch points of pathways are multiple regulatory controls within pathways, the that produce the same metabolite from different sub- static structure becomes predictive of the potential strates buffers against the loss of one of the sub- for evolutionary change of certain enzymes and com- strates in the external environment (Badyaev et al. pounds, because these regulatory controls target dif- 2015; Higginson et al. In press). For example, when ferent structural locations to optimize metabolic flux. Escherichia coli was exposed to fluctuating levels of Linking the evolutionary stability of changes in reg- glucose and acetate, some strains evolved a generalist ulatory mechanisms to their corresponding pheno- phenotype that allowed them to use both substrates; typic changes within a deterministic network has whereas neither pathway was optimized compared implications for understanding the plasticity of a phe- with specialist strains, this strategy allowed the gen- notype and its role in diversification. We can predict eralist strain to adapt to changing conditions both how and when a phenotype should change, as (Herron and Doebeli 2011). Alternatively, the pres- well as the relative stability of the change. For exam- ence of divergent branching points allows the expres- ple, establishing topology of the enzymes that are tar- sion of different pathways from the same starting gets of hormonal signals gives insight into changes in metabolite in different environments, leading to di- functional properties of the network in response to versification and specialization (Lavington et al. hormonal signaling. Transient properties of hormonal 2014). The key role of branching points in the control, in which hormonal signaling changes in Evolution of deterministic networks 241 64 response to environmental perturbations (Schulte rate under different conditions (e.g., Ibarra et al. 2002; 2013), can lead to phenotypic plasticity in a meta- Almaas et al. 2004; Almaas et al. 2005; Herron and bolic network. Selection on functional properties of Doebeli 2011). Different dynamics of regulatory con- a metabolic network can produce adaptive radia- trol may characterize other functional states, such as tions in functional states (Kitano et al. 2010) as a in floral pigmentation (Rausher et al. 1999; Rausher et result of exposure to more stable environments or al. 2008) and insect flight performance (Eanes et al. genetic assimilation of hormone production in reg- 2006). ulatory mechanisms (Rissman et al. 1997; Flurkey et Establishing how the topology of regulatory mech- al. 2001; Ellis et al. 2003; Badyaev 2009). anisms in deterministic networks is linked to func- To test our predictions (Fig. 1), intraspecific tional and evolutionary changes gives us a changes in regulatory mechanisms need to be com- quantitative perspective on the underlying mecha- pared with interspecific patterns of diversification in nisms of phenotypic change and stability. Not only expressed metabolites and enzymes. Determining the can we pinpoint the specific differences between phe- topological locations of changes in flux would re- notypes, but we can also assess both the magnitude quire comparisons of the dynamic properties of the of these changes and possible sources of the variation same metabolic network in different environments based on differences across functional states in rela- (Almaas et al. 2004; Papp et al. 2004; Price et al. tion to the structure of their deterministic network. 2004; Reed and Palsson 2004; Lavington et al. In short, linking structural and dynamic properties 2014). The next step would be to assess the relation- of genetic, protein, and metabolic networks offers an ship between structural properties of regulatory opportunity to apply a predictive structure to ob- controls and evolutionary patterns of metabolic di- served evolutionary patterns. versification. Comparing how the metabolic network is used across species in a lineage establishes conser- Acknowledgments vation of compounds and enzymes over evolutionary time; these evolutionary differences should corre- We thank Frances Bonier, Robert Cox and Joel spond to the structural positions of changes in flux McGlothlin for organizing the symposium, NSF on the metabolic network. (IOS-1539936) and SICB (DEE, DCE, DEDB, DAB) More work is needed to assess how functional for supporting our participation, and R. Duckworth, states of metabolic networks change in multicellu- M. Higginson, A. Potticary, and G. Semenov for lar organisms. Many of the studies reviewed here thorough comments on previous versions and help- examinefluxinmicrobesandthemodeofevolu- ful suggestions. tion often differs between unicellular and multicellular organisms. For example, in unicellular Funding organisms pathways and elements can be gained This work was supported by the grants from the independently of their functional properties during National Science Foundation (DEB-1256375), the horizontal gene transfers from other organisms Packard Foundation Fellowship, Galileo Scholarship, (Lawrence and Roth 1996; Pa´l et al. 2005; Kreimer and Amherst College graduate fellowships. et al. 2008). This mode of evolution can result in indistinguishable evolutionary patterns to changes References in regulatory mechanisms, because connected compounds would be conserved due to the preferential Albert R, Jeong H, Baraba´si A-L. 2000. Error and attack tol- attachment of the horizontal transmission of acquired erance of complex networks. Nature 406:378–82. Almaas E, Kovacs B, Vicsek T, Oltvai ZN, Baraba´si A-L. 2004. enzymes to the same initial compounds (Eisenberg Global organization of metabolic fluxes in the bacterium and Levanon 2003; Light et al. 2005), or the di- Escherichia coli. Nature 427:839–43. vergence among downstream enzymes across spe- Almaas E, Oltvai ZN, Baraba´si A-L. 2005. The activity reac- cies is due to the horizontal gene acquisition of tion core and plasticity of metabolic networks. PLoS Comp enzymes at the end of pathways (Bernhardsson et al. Biol 1:e68. 2011). Alon U. 2003. Biological networks: the tinkerer as an engi- A greater focus on metabolic network divergence neer. Science 301:1866–7. in multicellular species is also an opportunity to Badyaev AV. 2007. Evolvability and robustness in color displays: bridging the gap between theory and data. Evol Biol evaluate the dynamics of metabolic networks for var- 34:61–71. iable functional properties. Almost all of the empir- Badyaev AV. 2009. Evolutionary significance of phenotypic ical studies discussed in this review examine the accommodation in novel environments: an empirical test targets of flux control in relation to optimal growth of the Baldwin effect. Phil Trans R Soc B 364:1125–41. 65 242 E. S. Morrison and A. V. Badyaev

Badyaev AV, Foresman KR. 2000. Extreme environmental regulation is insufficient to explain substrate-induced flux change and evolution: stress-induced morphological varia- changes in Bacillus subtilis. Mol Syst Biol 9:709. tion is strongly concordant with patterns of evolutionary Cohen P. 1988. Review lecture: protein phosphorylation and divergence in shrew mandibles. Proc R Soc Lond B hormone action. Proc R Soc B 234:115–44. 267:371–9. Costa LF, Rodrigues FA, Travieso G, Villas Boas PR. 2007. Badyaev AV, Morrison ES, Belloni V, Sanderson MJ. 2015. Characterization of complex networks: a survey of measure- Tradeoff between robustness and elaboration in carotenoid ments. Adv Phys 56:167–242. networks produces cycles of avian color diversification. Covert MW, Knight EM, Reed JL, Herrgard MJ, Palsson BØ. Biology Direct 10:45. 2004. Integrating high-throughput and computational data Baraba´si A-L, Oltvai ZN. 2004. Network biology: elucidates bacterial networks. Nature 429:92–6. Understanding the cell’s functional organization. Nat Rev Covert MW, Palsson BØ. 2002. Transcriptional regulation in Genet 5:101–13. constraints-based metabolic models of Escherichia coli. J Bauerfeind R, Takei K, De Camilli P. 1997. Amphiphysin I is Biol Chem 277:28058–64. associated with coated endocytic intermediates and under- Daran-Lapujade P, Jansen MLA, Daran J-M, van Gulik W, de goes stimulation-dependent dephosphorylation in nerve Winde JH, Pronk JT. 2004. Role of transcriptional terminals. J Biol Chem 272:30984–92. regulation in controlling fluxes in central carbon metabo- Bennett MR, Pang WL, Ostroff NA, Baumgartner BL, Nayak lism of Saccharomyces cerevisiae. J Biol Chem 279:9125–38. S, Tsimring LS, Hasty J. 2008. Metabolic gene regulation in Davidson EH, Erwin DH. 2010. Evolutionary innovation and a dynamically changing environment. Nature 454:1119–22. stability in animal gene networks. J Exp Zool B Mol Dev E Berg J, La¨ssig M, Wagner A. 2004. Structure and evolution of 314B:182–6. protein interaction networks: a statistical model for link de Hijas-Liste G, Balsa-Canto E, Ewald J, Bartl M, Li P, Banga dynamics and gene duplications. BMC Evol Biol 4:1–12. J, Kaleta C. 2015. Optimal programs of pathway control: Bernhardsson S, Gerlee P, Lizana L. 2011. Structural correla- dissecting the influence of pathway topology and feedback tions in bacterial metabolic networks. BMC Evol Biol 11:20. inhibition on pathway regulation. BMC Bioinformatics Blackman FF. 1905. Optima and limiting factors. Ann Bot os 16:163. 19:281–96. de la Fuente A, Brazhnik P, Mendes P. 2002. Linking the Bongaerts GP, Vliegenthart JS. 1988. Effect of aminoglycoside genes: inferring quantitative gene networks from microar- concentration on reaction rates of aminoglycoside-modify- ray data. Trends Genet 18:395–8. ing enzymes. Antimic Agen Chemother 32:740–6. Dekel E, Alon U. 2005. Optimality and evolutionary tuning of Borenstein E, Kupiec M, Feldman MW, Ruppin E. 2008. the expression level of a protein. Nature 436:588–92. Large-scale reconstruction and phylogenetic analysis of Dixon M. 1953. The effect of pH on the affinities of enzymes metabolic environments. Proc Natl Acad Sci USA for substrates and inhibitors. Biochem J 55:161–70. 105:14482–7. Eanes WF. 1999. Analysis of selection on enzyme polymor- Bradley PH, Brauer MJ, Rabinowitz JD, Troyanskaya OG. phisms. Annu Rev Ecol Syst 30:301–26. 2009. Coordinated concentration changes of transcripts Eanes WF, Merritt TJS, Flowers JM, Kumagai S, Sezgin E, and metabolites in Saccharomyces cerevisiae. PLoS Comp Zhu C-T. 2006. Flux control and excess capacity in the Biol 5:e1000270. enzymes of glycolysis and their relationship to flight me- C¸akir T, Patil KR, O¨ nsan ZI, U¨ lgen KÖ, Kirdar B, tabolism in Drosophila melanogaster. Proc Natl Acad Sci Nielsen J. 2006. Integration of metabolome data with meta- USA 103:19413–8. bolic networks reveals reporter reactions. Mol Syst Biol 2:50. Edwards JS, Ibarra RU, Palsson BØ. 2001. In silico predictions Campillos M, von Mering C, Jensen LJ, Bork P. 2006. of Escherichia coli metabolic capabilities are consistent with Identification and analysis of evolutionarily cohesive func- experimental data. Nat Biotechnol 19:125–30. tional modules in protein networks. Genome Res Edwards JS, Palsson BØ. 1999. Systems properties of the 16:374–82. Haemophilus influenzae Rd metabolic genotype. J Biol Causton HC, Ren B, Koh SS, Harbison CT, Kanin E, Jennings Chem 274:17410–6. EG, Lee TI, True HL, Lander ES, Young RA. 2001. Edwards JS, Palsson BØ. 2000a. The Escherichia coli MG1655 Remodeling of yeast genome expression in response to en- in silico metabolic genotype: Its definition, characteristics, vironmental changes. Mol Biol Cell 12:323–37. and capabilities. Proc Natl Acad Sci USA 97:5528–33. Chen F, Tholl D, Bohlmann J, Pichersky E. 2011. The family Edwards JS, Palsson BØ. 2000b. Robustness analysis of the of terpene synthases in plants: a mid-size family of genes Escherichia coli metabolic network. Biotechnol Prog for specialized metabolism that is highly diversified 16:927–39. throughout the kingdom. Plant J 66:212–29. Eisenberg E, Levanon EY. 2003. Preferential attachment in the Chen Y, Dokholyan NV. 2006. The coordinated evolution of protein network evolution. Phys Rev Lett 91:138701. yeast proteins is constrained by functional modularity. Ellis JM, Livesey JH, Evans MJ. 2003. Hormone stability in Trends Genet 22:416–9. human whole blood. Clin Biochem 36:109–12. Cho B-K, Barrett CL, Knight EM, Park YS, Palsson BØ. Ellis RJ. 2001. Macromolecular crowding: obvious but under- 2008. Genome-scale reconstruction of the Lrp regulatory appreciated. Trends Biochem Sci 26:597–604. network in Escherichia coli. Proc Natl Acad Sci USA Emilsson V, Thorleifsson G, Zhang B, Leonardson AS, Zink F, 105:19462–7. Zhu J, Carlson S, Helgason A, Walters GB, Gunnarsdottir Chubukov V, Uhr M, Le Chat L, Kleijn RJ, Jules M, Link H, S, et al. 2008. Genetics of gene expression and its effect on Aymerich S, Stelling J, Sauer U. 2013. Transcriptional disease. Nature 452:423–8. Evolution of deterministic networks 243 66

Enjalbert B, Smith DA, Cornell MJ, Alam I, Nicholls S, Brown Higginson DM, Belloni V, Davis SN, Morrison ES, Andrews AJP, Quinn J. 2006. Role of the Hog1 stress-activated pro- JE, Badyaev AV. In press. Evolution of long-term coloration tein kinase in the global transcriptional response to stress in trends with biochemically unstable ingredients. Proc Roy the fungal pathogen Candida albicans. Mol Biol Cell Soc B. 17:1018–32. Ibarra RU, Edwards JS, Palsson BØ. 2002. Escherichia coli K- Erwin DH, Davidson EH. 2009. The evolution of hierarchical 12 undergoes adaptive evolution to achieve in silico pre- gene regulatory networks. Nat Rev Gen 10:141–8. dicted optimal growth. Nature 420:186–9. Farkas IJ, Jeong H, Vicsek T, Baraba´si A-L, Oltvai ZN. 2003. Ihmels J, Levy R, Barkai N. 2004. Principles of transcriptional The topology of the transcription regulatory network in the control in the metabolic network of Saccharomyces cerevi- yeast, Saccharomyces cerevisiae. Physica A 318:601–12. siae. Nat Biotechnol 22:86–92. Fell DA. 1992. Metabolic control analysis: a survey of its Jarmuszkiewicz W, Woyda-Ploszczyca A, Koziel A, Majerczak theoretical and experimental development. Biochem J J, Zoladz JA. 2015. Temperature controls oxidative phos- 286:313–30. phorylation and reactive oxygen species production Fell DA. 1997. Understanding the control of metabolism. through uncoupling in rat skeletal muscle mitochondria. Miami (FL): Portland Press. Free Radical Biol Med 83:12–20. Fell DA, Sauro HM. 1985. Metabolic control and its analysis. Jeong H, Mason SP, Baraba´si A-L, Oltvai ZN. 2001. Lethality Eur J Biochem 148:555–61. and centrality in protein networks. Nature 411:41–2. Fell DA, Thomas S. 1995. Physiological control of metabolic Kacser H. 1983. The control of enzyme systems in vivo: elasticity flux: the requirement for multisite modulation. Biochem J analysis of the steady state. Biochem Soc Trans 11:35–40. 311:35–9. Kacser H, Burns JA. 1973. The control of flux. Symp Soc Exp Flowers JM, Sezgin E, Kumagai S, Duvernell DD, Matzkin Biol 27:65–104. LM, Schmidt PS, Eanes WF. 2007. Adaptive evolution of Kacser H, Burns JA. 1981. The molecular basis of dominance. metabolic pathways in Drosophila. Mol Biol E 24:1347–54. Genetics 97:639–66. Flurkey K, Papaconstantinou J, Miller RA, Harrison DE. 2001. Kanehisa M, Goto S, Sato Y, Kawashima M, Furumichi M, Lifespan extension and delayed immune and collagen aging Tanabe M. 2014. Data, information, knowledge and principle: in mutant mice with defects in growth hormone produc- back to metabolism in KEGG. Nucleic Acids Res 42:D199–205. tion. Proc Natl Acad Sci USA 98:6736–41. Kaufmann H, Mazur X, Fussenegger M, Bailey JE. 1999. Fraser HB, Hirsh AE, Steinmetz LM, Scharfe C, Feldman Influence of low temperature on productivity, proteome MW. 2002. Evolutionary rate in the protein interaction and protein phosphorylation of CHO cells. Biotechnol network. Science 296:750–2. Bioeng 63:573–82. Gasch AP, Spellman PT, Kao CM, Carmel-Harel O, Eisen Kharchenko P, Church GM, Vitkup D. 2005. Expression dy- MB, Storz G, Botstein D, Brown PO. 2000. Genomic ex- namics of a cellular metabolic network. Mol Syst Biol 1: pression programs in the response of yeast cells to environ- 2005.0016. mental changes. Mol Bio Cell 11:4241–57. Kholodenko BN, Demin OV, Moehren G, Hoek JB. 1999. Goelzer A, Bekkal Brikci F, Martin-Verstraete I, Noirot P, Quantification of short term signaling by the epidermal Bessie`res P, Aymerich S, Fromion V. 2008. Reconstruction growth factor receptor. J Biol Chem 274:30169–81. and analysis of the genetic and metabolic regulatory net- Kitano J, Lema SC, Luckenbach JA, Mori S, Kawagishi Y, works of the central metabolism of Bacillus subtilis. BMC Kusakabe M, Swanson P, Peichel CL. 2010. Adaptive diver- Syst Biol 2:20. gence in the thyroid hormone signaling pathway in the Gordon KL, Ruvinsky I. 2012. Tempo and mode in evolution stickleback radiation. Curr Biol 20:2124–30. of transcriptional regulation. PLoS Genet 8:e1002432. Klassen JL. 2010. Phylogenetic and evolutionary patterns in Hahn MW, Kern AD. 2005. Comparative genomics of cen- microbial carotenoid biosynthesis are revealed by compar- trality and essentiality in three Eukaryotic protein-interac- ative genomics. PLoS ONE 5:e11257. tion networks. Mol Biol E 22:803–6. Klipp E, Heinrich R, Holzhu¨tter H-G. 2002. Prediction of Hartwell LH, Hopfield JJ, Leibler S, Murray AW. 1999. From temporal gene expression. Eur J Biochem 269:5406–13. molecular to modular cell biology. Nature 402:C47–52. Kondrashov FA. 2012. Gene duplication as a mechanism of Heijnen JJ, van Gulik WM, Shimizu H, Stephanopoulos G. genomic adaptation to a changing environment. Proc R Soc 2004. Metabolic flux control analysis of branch points: an B 279:5048–57. improved approach to obtain flux control coefficients from Krebs EG, Beavo JA. 1979. Phosphorylation-dephosphoryla- large perturbation data. Metab Eng 6:391–400. tion of enzymes. Annu Rev Biochem 48:923–59. Heinemann M, Sauer U. 2010. Systems biology of microbial Krebs HA. 1957. Control of metabolic processes. Endeavour metabolism. Curr Opin Microbiol 13:337–43. 16:125–32. Heinrich R, Rapoport TA. 1974. A linear steady-state treat- Kreimer A, Borenstein E, Gophna U, Ruppin E. 2008. The ment of enzymatic chains. Eur J Biochem 42:89–95. evolution of modularity in bacterial metabolic networks. Heinrich R, Schuster S, Holzhu¨tter H-G. 1991. Mathematical Proc Natl Acad Sci USA 105:6976–81. analysis of enzymic reaction systems using optimization Kuznetsov VA, Knott GD, Bonner RF. 2002. General statistics principles. Eur J Biochem 201:1–21. of stochastic process of gene expression in eukaryotic cells. Herron MD, Doebeli M. 2011. Adaptive diversification of a Genetics 161:1321–32. plastic trait in a predictably fluctuating environment. Lamb DC, Kelly DE, Schunck W-H, Shyadehi AZ, Akhtar M, J Theor Biol 285:58–68. Lowe DJ, Baldwin BC, Kelly SL. 1997. The mutation T315A 67 244 E. S. Morrison and A. V. Badyaev

in Candida albicans sterol 14 -demethylase causes reduced of primary metabolism of Penicillium chrysogenum through enzyme activity and fluconazole resistance through reduced glucose perturbation in the bioscope mini reactor. Metab affinity. J Biol Chem 272:5682–8. Eng 8:395–405. LaPorte DC, Walsh K, Koshland DE. Jr. 1984. The branch Nilsson A, Pa˚hlman I-L, Jovall P-A˚ , Blomberg A, Larsson C, point effect. Ultrasensitivity and subsensitivity to metabolic Gustafsson L. 2001. The catabolic capacity of Saccharomyces control. J Biol Chem 259:14068–75. cerevisiae is preserved to a higher extent during carbon Lavington E, Cogni R, Kuczynski C, Koury S, Behrman EL, compared to nitrogen starvation. Yeast 18:1371–81. O’Brien KR, Schmidt PS, Eanes WF. 2014. A small Notebaart RA, Teusink B, Siezen RJ, Papp B. 2008. Co- system—high-resolution study of metabolic adaptation in Regulation of Metabolic Genes Is Better Explained by the central metabolic pathway to temperate climates in Flux Coupling Than by Network Distance. PLoS Comp Drosophila melanogaster. Mol Biol E 31:2032–41. Biol 4:e26. Lawrence JG, Roth JR. 1996. Selfish operons: horizontal trans- Olson-Manning CF, Lee C-R, Rausher MD, Mitchell-Olds fer may drive the evolution of gene clusters. Genetics T. 2013. Evolution of flux control in the glucosinolate 143:1843–60. pathway in Arabidopsis thaliana. MolBiolE30:14– Lee MT, Liebow C, Kamer AR, Schally AV. 1991. Effects of 23. epidermal growth factor and analogues of luteinizing hor- Olson-Manning CF, Strock CF, Mitchell-Olds T. 2015. Flux mone-releasing hormone and somatostatin on phosphory- control in a defense pathway in Arabidopsis thaliana lation and dephosphorylation of tyrosine residues of is robust to environmental perturbations and controls specific protein substrates in various tumors. Proc Natl variation in adaptive traits. G3 Genes Genom Genet Acad Sci USA 88:1656–60. 5:2421–7. Light S, Kraulis P, Elofsson A. 2005. Preferential attachment Ortega F, Acerenza L. 1998. Optimal metabolic control in the evolution of metabolic networks. BMC Genomics design. J Theor Biol 191:439–49. 6:159. Pa´l C, Papp B, Lercher MJ. 2005. Adaptive evolution of bac- Link H, Kochanowski K, Sauer U. 2013. Systematic identi- terial metabolic networks by horizontal gene transfer. Nat fication of allosteric protein-metabolite interactions that Genet 37:1372–5. control enzyme activity in vivo. Nat Biotechnol 31:357– Papp B, Pal C, Hurst LD. 2004. Metabolic network analysis of 61. the causes and evolution of enzyme dispensability in yeast. Ma H-W, Buer J, Zeng A-P. 2004. Hierarchical structure and Nature 429:661–4. modules in the Escherichia coli transcriptional regulatory Papp B, Teusink B, Notebaart RA. 2009. A critical view of network revealed by a new top-down approach. BMC metabolic network adaptations. HFSP Journal 3:24–35. Bioinform 5:199. Pfeiffer T, Sa´nchez-Valdenebro I, Nun˜o JC, Montero F, Machado D, Herrga˚rd MJ, Rocha I. 2015. Modeling the con- Schuster S. 1999. METATOOL: for studying metabolic net- tribution of allosteric regulation for flux control in the works. Bioinformatics 15:251–7. central carbon metabolism of E. coli. Front Bioeng Price ND, Reed JL, Palsson BØ. 2004. Genome-scale models Biotechnol 3:154. of microbial cells: evaluating the consequences of con- Mahadevan R, Palsson BØ. 2005. Properties of metabolic net- straints. Nat Rev Microbiol 2:886–97. works: structure versus function. Biophys J 88:L07–L9. Raff EC, Raff RA. 2000. Dissociability, modularity, evolvabil- Matsuno R, Nakanishi k, Ohnishi M, Hiromi K, Kamikubo T. ity. Evolution & Development 2:235–7. 1978. Threshold in a single enzyme reaction system: reac- Ralser M, Wamelink MMC, Latkolik S, Jansen EEW, Lehrach tion of maltose catalyzed by saccharifying -Amylase from H, Jakobs C. 2009. Metabolic reconfiguration precedes B. Subtilis. J Biol Chem 83:859–62. transcriptional regulation in the antioxidant response. Nat McAdams HH, Shapiro L. 2003. A bacterial cell-cycle regula- Biotechnol 27:604–5. tory network operating in time and space. Science Ramsay H, Rieseberg LH, Ritland K. 2009. The correlation of 301:1874–7. evolutionary rate with pathway position in plant terpenoid Minton AP. 2001. The influence of macromolecular crowding biosynthesis. Mol Biol E 26:1045–53. and macromolecular confinement on biochemical reactions Rausher MD. 2013. The evolution of genes in branched in physiological media. J Biol Chem 276:10577–80. metaoblic pathways. Evolution 67:34–48. Moxley JF, Jewett MC, Antoniewicz MR, Villas-Boas SG, RausherMD,LuY,MeyerK.2008.Variationinconstraint Alper H, Wheeler RT, Tong L, Hinnebusch AG, Ideker T, versus positive selection as an explanation for evolution- Nielsen J, et al. 2009. Linking high-resolution metabolic ary rate variation among anthocyanin genes. J Mol E 67: flux phenotypes and transcriptional regulation in yeast 137–44. modulated by the global regulator Gcn4p. Proc Natl Acad Rausher MD, Miller RE, Tiffin P. 1999. Patterns of evolution- Sci USA 106:6477–82. ary rate variation among genes of the anthocyanin biosyn- Nagy L. 1998. Changing Patterns of Gene Regulation in thetic pathway. Mol Biol E 16:266–74. the Evolution of Arthropod Morphology. Am Zool Ravasz E, Somera AL, Mongru DA, Oltvai ZN, Baraba´si A-L. 38:818–28. 2002. Hierarchical organization of modularity in metabolic Nam H, Lewis NE, Lerman JA, Lee D-H, Chang RL, Kim D, networks. Science 297:1551–5. Palsson BØ. 2012. Network context and selection in the Reed JL, Palsson BØ. 2004. Genome-scale in silico models of evolution to enzyme specificity. Science 337:1101–4. E. coli have multiple equivalent phenotypic states: assess- Nasution U, van Gulik WM, Proell A, van Winden WA, ment of correlated reaction subsets that comprise network Heijnen JJ. 2006. Generating short-term kinetic responses states. Gen Res 14:1797–805. Evolution of deterministic networks 245 68

Rissman EF, Wersinger SR, Taylor JA, Lubahn DB. 1997. Stephanopoulos G, Vallino JJ. 1991. Network rigidity and Estrogen receptor function as revealed by knockout studies: metabolic engineering in metabolite overproduction. neuroendocrine and behavioral aspects. Horm Behav Science 252:1675–81. 31:232–43. Stra˚lfors P, Honnor RC. 1989. Insulin-induced dephosphory- Rossell S, van der Weijden CC, Lindenbergh A, van Tuijl A, lation of hormone-sensitive lipase. Eur J Biochem Francke C, Bakker BM, Westerhoff HV. 2006. Unraveling 182:379–85. the complexity of flux regulation: A new method demon- Szasz G. 1974. The effect of temperature on enzyme activity strated for nutrient starvation in Saccharomyces cerevisiae. and on the affinity of enzymes to their substrates. Z Klin Proc Natl Acad Sci USA 103:2166–71. Chem Klin Bio 12:166–70. Saavedra E, Encalada R, Pineda E, Jasso-Cha´vez R, Moreno- Taymaz-Nikerel H, De Mey M, Baart G, Maertens J, Heijnen Sa´nchez R. 2005. Glycolysis in Entamoeba histolytica. FEBS J JJ, van Gulik W. 2013. Changes in substrate availability in 272:1767–83. Escherichia coli lead to rapid metabolite, flux and growth Schaefke B, Emerson JJ, Wang T-Y, Lu M-YJ, Hsieh L-C, Li rate responses. Metab Eng 16:115–29. W-H. 2013. Inheritance of gene expression level and selec- Taymaz-Nikerel H, van Gulik WM, Heijnen JJ. 2011. tive constraints on trans- and cis-regulatory changes in Escherichia coli responds with a rapid and large change in yeast. Mol Biol E 30:2121–33. growth rate upon a shift from glucose-limited to glucose- Schmidt S, Sunyaev S, Bork P, Dandekar T. 2003. Metabolites: excess conditions. Metab Eng 13:307–18. a helping hand for pathway evolution? Trends Biochem Sci ter Kuile BH, Westerhoff HV. 2001. Transcriptome meets 28:336–41. metabolome: hierarchical and metabolic regulation of the Schulte PM. 2013. What is environmental stress? Insights glycolytic pathway. FEBS Lett 500:169–71. from fish living in a variable environment. J Exp Biol Thomas S, Fell DA. 1998. The role of multiple enzyme acti- 217:23–34. vation in metabolic flux control. Adv Enzyme Regul Schuster S, Fell DA, Dandekar T. 2000. A general definition of 38:65–85. metabolic pathways useful for systematic organization and Umbarger HE. 1956. Evidence for a negative-feedback mech- analysis of complex metabolic networks. Nat Biotechnol anism in the biosynthesis of isoleucine. Science 123:848. 18:326–32. Vallino JJ, Stephanopoulos G. 1994. Carbon flux distributions Schuster S, Klamt S, Weckwerth W, Moldenhauer F, Pfeiffer at the pyruvate branch point in Corynebacterium glutami- T. 2002. Use of network analysis of metabolic systems in cum during lysine overproduction. Biotechnol Prog bioengineering. Biopro Biosys Eng 24:363–72. 10:320–6. Schwender J, Ko¨nig C, Klapperstu¨ck M, Heinzel N, Munz E, Van Hoek P, Van Dijken JP, Pronk JT. 1998. Effect of specific Hebbelmann I, Hay JO, Denolf P, De Bodt S, Redestig H, growth rate on fermentative capacity of baker’s yeast. Appl et al.. 2014. Transcript abundance on its own cannot be Environ Microbiol 64:4226–33. used to infer fluxes in central metabolism. Fro Plant Sci Va´zquez A, Flammini A, Maritan A, Vespignani A. 2003. 5:668. Modeling of protein interaction networks. Complexus Segre` D, Vitkup D, Church GM. 2002. Analysis of optimality 1:38–44. in natural and perturbed metabolic networks. Proc Natl Vitkup D, Kharchenko P, Wagner A. 2006. Influence of met- Acad Sci USA 99:15112–7. abolic network structure and function on enzyme evolu- Seshasayee ASN, Fraser GM, Babu MM, Luscombe NM. tion. Genome Biol 7:R39. 2009. Principles of transcriptional regulation and evolu- Vogt T. 2010. Phenylpropanoid biosynthesis. Molecular Plant tion of the metabolic system in E. coli. Genome Res 3:2–20. 19:79–91. von Dassow G, Munro E. 1999. Modularity in animal devel- Smolke CD, Carrier TA, Keasling JD. 2000. Coordinated, dif- opment and evolution: elements of a conceptual framework ferential expression of two genes through directed mRNA for EvoDevo. Journal of Experimental Zoology 285:307–25. cleavage and stabilization by secondary structures. Appl Wagner A. 2001. The yeast protein interaction network Environ Microbiol 66:5399–405. evolves rapidly and contains few redundant duplicate Smolke CD, Martin VJJ, Keasling JD. 2001. Controlling the genes. Mol Biol E 18:1283–92. metabolic flux through the carotenoid pathway using di- Wagner A. 2003. How the global structure of protein inter- rected mRNA processing and stabilization. Metab Eng action networks evolves. Proc R Soc B 270:457–66. 3:313–21. Wagner A. 2007. Gene networks and natural selection. In: Sørensen TH, Cruys-Bagger N, Windahl MS, Badino SF, Pagel M, Pomiankowski A, editors. Evolutionary genomics Borch K, Westh P. 2015. Temperature effects on kinetic and proteomics. Sunderland (MA): Sinauer Associates. parameters and substrate affinity of Cel7A cellobiohydro- p. 255–69. lases. J Biol Chem 290:22193–202. Wagner GP, Altenberg L. 1996. Perspective: Complex adapta- Soyer OS, Pfeiffer T. 2010. Evolution under fluctuating envi- tions and the evolution of evolvability. Evolution ronments explains observed robustness in metabolic net- 50:967–76. works. PLoS Comp Biol 6:e1000907. Wang JK, Walaas SI, Greengard P. 1988. Protein phosphory- Spirin V, Gelfand MS, Mirny LA. 2003. Computational anal- lation in nerve terminals: comparison of calcium/ ysis of metabolic modules and pathways in E. coli metabolic calmodulin-dependent and calcium/diacylglycerol- network. Int Soc Comp Biol 4:239–40. dependent systems. J Neurosci 8:281–8. 69 246 E. S. Morrison and A. V. Badyaev

Wang Z, Sun X, Zhao Y, Guo X, Jiang H, Li H, Gu Z. 2015. cerevisiae CEN.PK 113-7D following a glucose pulse. Appl Evolution of gene regulation during transcription and Environ Microbiol 72:3566–77. translation. Genome Biol E 7:1155–67. Yates RA, Pardee AB. 1956. Control of pyrimidine biosynthe- Weng J-K. 2014. The evolutionary paths towards complexity: sis in Escherichia coli by a feed-back mechanism. J Biol a metabolic perspective. New Phytol 201:1141–9. Chem 221:757–70. Wessely F, Bartl M, Guthke R, Li P, Schuster S, Kaleta C. Yu H, Gerstein M. 2006. Genomic analysis of the hierarchical 2011. Optimal regulatory strategies for metabolic pathways structure of regulatory networks. Proc Natl Acad Sci USA in Escherichia coli depending on protein costs. Mol Syst 103:14724–31. Biol 7:515. Zaslaver A, Mayo AE, Rosenberg R, Bashkin P, Sberro H, Westerhoff HV, Groen AK, Wanders RJA. 1984. Modern the- Tsalyuk M, Surette MG, Alon U. 2004. Just-in-time transcrip- ories of metabolic control and their applications. Biosci tion program in metabolic pathways. Nat Genet 36:486–91. Rep 4:1–22. Zelezniak A, Sheridan S, Patil KR. 2014. Contribution of net- Wright KM, Rausher MD. 2010. The evolution of control and work connectivity in determining the relationship between distribution of adaptive mutations in a metabolic pathway. gene expression and metabolite concentration changes. Genetics 184:483–502. PLoS Comput Biol 10:e1003572. Wu L, van Dam J, Schipper D, Kresnowati MTAP, Proell AM, Zhang Q, Teng H, Sun Y, Xiu Z, Zeng A. 2007. Metabolic Ras C, van Winden WA, van Gulik WM, Heijnen JJ. 2006. flux and robustness analysis of glycerol metabolism Short-term metabolome dynamics and carbon, electron, in Klebsiella pneumoniae. Bioprocess Biosystems Eng and ATP balances in chemostat-grown Saccharomyces 31:127–35. 70

APPENDIX C. BEYOND NETWORK TOPOLOGY: COEVOLUTION OF STRUCTURE AND FLUX IN METABOLIC NETWORKS

Submitted for publication to Journal of Evolutionary Biology

ABSTRACT

Accumulating evidence suggests that even full characterization of a biochemical network’s structure often does not adequately explain its function nor predict its evolutionary trajectory. Particularly unclear is whether the dynamics of flux are causally linked to a network’s static structure or emerge independently of it. Although the interaction between structural and dynamic properties of biochemical networks underlies their evolutionary diversification, the mechanism by which such interactions arise remains elusive and requires a direct empirical study of populations where both structural and dynamic properties vary among individuals’ biochemical networks. We reconstructed full carotenoid metabolic networks for

442 individual house finches (Haemorhous mexicanus) and uncovered 11 distinct structural variants of this network. We examined the consequences of structural diversity for the concentrations of plumage- bound carotenoids produced by these networks. We found that concentrations of metabolically-derived carotenoids, but not carotenoids at the beginning of metabolic pathways, depend on overall network structure and the number of reactions associated with a compound. Flux was partitioned similarly among individuals with the same network structure due to strong within-network correlations among compounds.

The greatest among-individual variation in flux occurred in network structures with the strongest correlations among compounds, suggesting that there is modular regulation of interconnected compounds in a structure. These findings indicate a significant influence of biochemical network structure on metabolic flux of derived carotenoids. Thus, the evolutionary diversification and local adaptations in carotenoid networks may depend more on the gain or loss of enzymatic reactions than changes in the activity of individual enzymes.

KEY-WORDS: adaptation, carotenoid metabolism, flux evolution, network structure

INTRODUCTION

Is the structure of a biochemical network a result of adaptation to specific dynamic properties of compound production, or can different dynamic properties be associated with the same network structure?

This determines whether changes in compound concentrations can occur under constant network topology, or only when reactions or substrates are gained or lost. There is evidence for both of these processes – evolutionary diversification of metabolic networks has been attributed to both structural changes in the occurrence of enzymatic reactions and dynamic changes in compound concentrations and enzyme activities (Jeong et al., 2000; Almaas et al., 2005; Badyaev et al., 2015; Nidelet et al., 2016), but the mechanistic links between structural and dynamic properties remain elusive (Stelling et al., 2002;

Lynch, 2007; Papp et al., 2009; Lee et al., 2012b).

Examining the correspondence between compound concentration and the number of pathways associated with the compound can untangle the origins of structural and dynamic network properties

(Torres-Sosa et al., 2012). Two key parameters are needed for testing this relationship. First is metabolic flux – a dynamic network property that represents the rate of compound production. The activity of an enzyme controls the magnitude of metabolic flux in a biochemical pathway and determines the amount of substrates and products (Kacser & Burns, 1973; Heinrich & Rapoport, 1974; Fell, 1997). The concentration of a substrate, the affinity of the enzyme for substrates, and the number of activated enzymes can all potentially regulate, and thus change, the activity of a reaction (reviewed in Morrison &

Badyaev, 2016a). Second is the connectivity of a compound – a structural network property characterized by the number of enzymatic reactions directly associated with the compound. The number of incoming reactions represents how many substrates are converted into the compound, and the number of outgoing reactions determines how many products it generates (Jeong et al., 2000; Wagner & Fell, 2001).

Connectivity of a compound evolves with the gain or loss of its associated enzymes, caused by gene duplications (Wagner, 2001; Berg et al., 2004; Kondrashov, 2012), mutations (Wagner, 2003b; Berg et al., 2004), horizontal gene transfer (Light et al., 2005; Pál et al., 2005; Klassen, 2010), and activation or deactivation of enzyme-encoding genes (Sadana, 1988; Piedrafita et al., 2015).

Coevolution of enzyme function and the structural positions of reactions in the network lead to the association between metabolic flux and connectivity of network compounds (Wright & Rausher,

2010; Rausher, 2013). Gain or loss of enzymatic reactions redistributes flux among the remaining enzymes associated with a compound. For example, the gain of an outgoing reaction requires the same amount of flux to be partitioned among more pathways, such that an increase in the production of one compound always leads to proportional changes in other compounds (Fig. 1a) (Fell & Thomas, 1995;

Rossell et al., 2006; Nilsson & Nielsen, 2016). Under this scenario, total flux can still change in a network structure, but its partitioning remains static, due to proportional changes in both the amount of substrates or enzyme activity of all reactions within a network. In such cases, the gain or loss of reactions changes network connectivity, and also causes changes in compound concentrations (Fig. 1a) (Emmerling et al., 2002; Lee et al., 2012a; Piedrafita et al., 2015). Under this scenario, a network structure is expected to be optimally designed for a certain state or property of metabolic flux (Wagner, 2003a; Eloundou-

Mbebi et al., 2016), leading to their coevolution.

Alternatively, the function of enzymes could be decoupled from the structure of biochemical pathways (Hartl et al., 1985; Lee et al., 2012b; Inoue & Kaneko, 2013), and independent evolution of enzyme activity occurs without gain or loss of enzymatic reactions (Kacser & Acerenza, 1993; Eanes,

1999; Almaas et al., 2004; Flowers et al., 2007; Olson-Manning et al., 2013). This can be caused by changes in how the activities of enzymes are regulated. The total flux of a network could be distributed differently among compounds depending on which reactions are co-regulated in a network (Morrison &

Badyaev, 2016a). All of the reactions in a pathway could be regulated as a module by the same mechanism, for example, such that the concentrations of compounds produced in this pathway change proportionally to each other. However, the regulation of some of the reactions in that same pathway could evolve independently (Soyer & Pfeiffer, 2010). This would cause differences in the distribution of flux among the same compounds depending on where regulatory changes occur in the pathway (Teusink &

Westerhoff, 2000). Under these scenarios, an increase in the concentration of a compound would not always be correlated with changes in the concentration of another compound within a network (Fig. 1b).

As a result, evolutionary changes in flux could occur on static network structure (Ebenhöh et al., 2005), leading to the decoupling of network structure and flux (Fig. 1b).

The relationship between flux and structure could vary locally among compounds depending on the number of reactions associated with each compound (Jeong et al., 2000; Barabási & Oltvai, 2004); the loss of the most interconnected compounds often disrupts the network’s functional properties (Albert et al., 2000; Schmidt et al., 2003). Thus, the flux of the most connected compounds can be more buffered from the effects of the gain or loss of reactions than less connected compounds. Similarly, if externally acquired substrates are present in excessive amounts, then concentrations of starting substrates in pathways may be independent of network structural properties. Contrasting the relationship between flux and structural changes among compounds that differ in their connectivity and metabolic derivation allows for direct insight into what causes the coevolution of structural and dynamic properties of a biochemical network.

Comparing the structures of biochemical networks and compound concentrations among individuals from the same population allows for the direct examination of the proximate effects of the gain or loss of enzymatic reactions on flux. This is because it isolates the effect of network structure on compound concentrations by removing confounding effects of substrate availability and temperature on compound production (Kim & Ryu, 1999; Sauer et al., 1999; Iyer et al., 2008). Here we use the metabolic network that produces plumage carotenoids in birds to test whether carotenoid concentration varies with the number and topology of reactions in the network. Network structure can be established based on the identity of the compounds found in an individual’s plumage in relation to a complete species-specific network (Badyaev et al., 2015; Morrison & Badyaev, 2016b), while network flux can be assessed directly by measuring relative concentrations of individual carotenoids moving through the network.

We examined the correlations between compound concentrations within a network structure with principal component analyses to determine the consistency of flux distribution among compounds when network structure is static. Strong correlations among compound concentrations indicate that flux is partitioned similarly among compounds, while weaker correlations indicate that fluxes of individual

compounds are regulated independently. To further establish whether distinct properties of flux partitioning occurred within each network structure, we assessed the strengths of the correlations among compound concentrations across network structures. Lastly, we examined how distinct flux distributions on the same network structure affected variation of compound concentrations. We tested the strength of correlations between compounds in relation to among-individual total carotenoid concentration within the same network. Greater variation in carotenoid concentrations in networks with weaker correlations among compounds would suggest that differences in flux partitioning contribute the most to the variation of compound concentrations within a network. Alternatively, if networks with the strongest correlations between compounds have the greatest variability in carotenoid concentrations, then there is modular regulation of flux in a network structure.

We studied the association between network topology and metabolic flux of plumage-bound carotenoids in a population of house finches (Haemorhous mexicanus). First, we built metabolic networks underlying the production of plumage carotenoids for each individual and quantified metabolic flux for 12 network compounds based on their observed concentrations. Second, we examined whether the concentration of each compound varied with network structure and compound connectivity. We tested these relationships separately for dietary and derived carotenoids, and among compounds with different numbers of reactions. We also assessed the strength of the correlations between compound concentrations and the impact that changes in structure have on these relationships. We discuss the contribution of structural and dynamic properties of carotenoid metabolism to intraspecific variation in compound concentrations, and consider the implications of these results for the evolutionary diversification of biochemical networks.

MATERIALS AND METHODS:

Study population

We analyzed 1,326 feather samples from 442 adult male house finches (Haemorhous mexicanus) in an individually color-marked study population in southeastern Arizona from 2003-2013 (protocol of feather

processing and fieldwork in Landeen & Badyaev, 2012). For each male, we sampled three to five feathers from each ornamental area (breast, rump, and crown) and processed the feathers from each ornament separately. Methods for feather carotenoid extraction, analysis, identification, and quantification are in

Higginson et al. (2016).

House finch carotenoid metabolic network

House finches plumage contain up to 19 carotenoids, including seven dietary carotenoids (β-carotene, β- cryptoxanthin, α-carotene, gazaniaxanthin, lutein, rubixanthin, zeaxanthin) and 12 carotenoids derived by the metabolism of dietary carotenoids (α-doradexanthin, β-isocryptoxanthin, 3’-dehydrolutein, 3’- hydroxy-echinenone, 4-oxo-rubixanthin, canary xanthophyll A, canary xanthophyll B, adonirubin, adonixanthin, astaxanthin, canthaxanthin, echinenone) (Inouye et al., 2001; McGraw et al., 2006;

Higginson et al., 2016). When all observed compounds and reactions known for the species are considered together, they form a metabolic network consisting of 24 compounds and 45 enzymatic reactions (Fig. 2a) (Badyaev et al., 2015; Morrison & Badyaev, 2016b). In this study we focused on a subset of 12 dietary and derived carotenoids that are linked to each other through enzymatic reactions: β- carotene, β-cryptoxanthin, lutein, zeaxanthin, β-isocryptoxanthin, 3’-dehydrolutein, 3’-hydroxy- echinenone, adonirubin, adonixanthin, astaxanthin, canthaxanthin, and echinenone.

Construction of individual metabolic networks

For each individual, we selected the plumage sample that contained the maximum number of carotenoid compounds. When all samples of an individual had the same number of compounds, a sample was selected at random. To construct an individual metabolic network, we mapped carotenoids identified in an individual sample on the full enzymatic network for the species (see Badyaev et al., 2015 and Morrison &

Badyaev, 2016 for justification). The mapping of groups of observed carotenoids resulted in 11 distinct network structures.

For each network structure, we calculated the numbers of incoming and outgoing reactions for each compound. The pathway position, maximum enzymatic connectivity and maximum betweenness centrality (Cb) were also calculated for each compound in the species’ network. The pathway position of a compound was the average of the minimum number of reactions the compound is from each of the dietary carotenoids (β-carotene, β-cryptoxanthin, lutein, zeaxanthin) in the species’ network. Dietary carotenoids were assigned a pathway position of zero reactions. The maximum enzymatic connectivity of a compound was the maximum number of total incoming and outgoing reactions directly associated with a compound in the species’ network, and is a local measure of compound connectivity. Betweenness centrality (CB) of a compound (n) -- a global structural measure of connectivity was defined as:

퐶퐵(푛) = ∑ (휎푠푡(푛)/휎푠푡) 푠≠푛≠푡

Where s and t are compounds different from n, σst represents the number of shortest pathways from s to t, and σst(n) is the number of shortest pathways from s to t that include n (Brandes, 2001). Betweenness centrality describes how often a compound (n) falls in the shortest path between pairs of compounds, and thus represents the influence of a compound on other network compounds (Yoon et al., 2006). We calculated the betweenness centrality using Cytoscape 2.8.2 (Smoot et al. 2011) with NetworkAnalyzer

2.7 (Assenov et al., 2008; Doncheva et al., 2012).

Statistical Analyses

We log-transformed concentrations to achieve normal distribution. General linear models were used to test for differences in the concentrations among individuals that vary in incoming or outgoing reactions and for differences in the concentrations of a compound among network structures. Pairwise differences between least-squares mean concentrations with different incoming or outgoing reactions, and between compound concentrations in unique network structures were tested using the Tukey-Kramer procedure

(Kramer 1956) due to unequal numbers of individuals per group. The standardized pairwise mean difference, the effect size, between the concentrations of a compound with different numbers of incoming

or outgoing reactions and between the concentrations of a compound in unique network structures was measured using Cohen’s d (Cohen, 1962), defined as the difference between the two group means (푥̅1 and

푥̅2) divided by the pooled standard deviation of the two groups:

|푥̅1 − 푥̅2| 푑12 = (푛 − 1)푠2 + (푛 − 1)푠2 √ 1 1 2 2 푛1 + 푛2 − 2

Where s2 is the variance and n is the number of samples in a group. We then ranked the effect sizes of structural changes for each compound and tested whether the magnitude of the change in the concentration of a compound in response to the addition or loss of reactions was correlated with a compound’s maximum pathway position, maximum connectivity, and maximum betweenness centrality.

For each network structure, we constructed linear principal components (PC) based on a correlational matrix of concentrations of individual carotenoids. We examined the eigenstructure of these matrices to compare the patterns of correlation among individual carotenoids within and across each network structure. All statistical analyses were done using SAS v. 9.4.

RESULTS:

Within-species variation in compound concentration and network structure

Individuals expressed from six to 19 carotenoids in their feathers (Fig. 2b). Eleven (91.67%) out of the 12 carotenoids varied in the number of incoming or outgoing reactions in individual metabolic networks

(Fig. 3). The coefficient of variation of a compound’s concentration ranged from 0.662 (zeaxanthin) to

2.842 (gazaniaxanthin) (Fig. 2a). Occurrence of compounds across 442 individuals varied from 14.03%

(β-cryptoxanthin) to 100% (lutein) (Fig. 2a).

Compound’s enzymatic connectivity and concentration

Concentrations of dietary carotenoids did not vary with the number of outgoing reactions (Fig. 3a-c; lutein: df = 2, F = 0.55, P = 0.577; zeaxanthin: df = 1, F = 1.32, P = 0.251; β-carotene: df = 1, F = 1.87, P

= 0.173), however concentrations of five (62.50%) out of the eight derived carotenoids did. Echinenone and adonirubin concentrations varied with the number of both incoming and outgoing reactions (Fig. 3d- e; echinenone: df = 1, F = 23.04, P < 0.001 and adonirubin: df = 1, F = 8.68, P = 0.003). The concentrations of 3’-hydroxy-echinenone and canthaxanthin depended on the number of outgoing reactions (Fig. 3f; 3’-hydroxy-echinenone: df = 1, F = 56.11, P < 0.001 and Fig. 3k; canthaxanthin: df = 1,

F = 31.66, P < 0.001), but the concentration of astaxanthin did not (Fig. 3j; df = 1, F = 0.94, P = 0.334).

The concentration of adonixanthin, but not of 3’-hydroxy-echinenone, 3’-dehydrolutein, and β- isocryptoxanthin, varied with the number of incoming reactions (Fig. 3f-i; adonixanthin: df = 1, F = 5.00,

P = 0.026; 3’-hydroxy-echinenone: incoming reactions df = 1, F = 1.34, P = 0.247, incoming reactions*outgoing reactions df = 1, F = 0, P = 0.967; 3’dehydrolutein: df = 1, F = 3.35, P = 0.068; and β- isocryptoxanthin: df = 1, F = 3.48, P = 0.063). The effect of the number of reactions on compound concentration was greater in derived compounds than dietary compounds (Fig. 4a; Spearman ρ = 0.602, P

= 0.050, n = 11), but was not related to either compound connectivity (Fig. 4b; ρ = 0.463, P = 0.15) or betweenness centrality (Fig. 4c; ρ = 0.200, P = 0.55).

Compound concentrations in distinct network structures

Concentrations of the 12 compounds that were expressed at least once in a distinct network, differed among network structures (Fig. 5). Two of these were dietary (β-carotene: df = 7, F = 2.51, P = 0.016 and lutein: df = 10, F = 3.60, P < 0.01) and four were derived (β-isocryptoxanthin: df = 8, F = 5.39, P <

0.001; canthaxanthin: df = 10, F = 3.62, P < 0.001; echinenone: df = 10, F = 3.56, P = 0.0002 and 3’- hydroxy-echinenone: df = 10, F = 3.02, P = 0.001). Of the six carotenoids whose concentrations did not vary among network structures, two were dietary (β-cryptoxanthin: df = 2, F = 0.06, P = 0.94 and zeaxanthin: df = 9, F = 0.92, P = 0.51) and four were derived (3’-dehydrolutein: df = 10, F = 1.36, P =

0.195; adonirubin: df = 10, F = 1.01, P = 0.432; adonixanthin: df = 5, F = 1.09, P = 0.367; and astaxanthin: df = 8, F = 1.51, P = 0.154). Network structures also differed in the total concentration of carotenoids (df = 10, F = 2.44, P = 0.008). The lowest concentrations of most of the compounds occurred

in the absence of dietary zeaxanthin (network D) (Fig. 5, Table S2). The highest concentrations of compounds occurred in network structures without derived carotenoids (networks E, G, J) (Fig. 5, Table

S2).

Structural integration of compound concentrations

In all network structures, PC1 accounted for proportional changes of all compounds, while PC2 reflected trade-offs among them (Fig. 6, Table S3). The strength of these correlations differed between network structures (Fig. 6, Table S3). The largest differences in total carotenoid concentrations occurred in networks with the strongest correlations among compounds (Fig. 7; t = 2.29, bST = 0.607, P = 0.048, n =

11).

DISCUSSION

To what extent do structural properties of a biochemical network reflect its dynamic properties? We found that the structure of a metabolic network influences carotenoid flux: derived carotenoid concentrations varied with reaction number (Fig. 3d-f, 3i, 3k; Table S1) and among distinct network structures (Fig. 5). This association between flux and network structure arose from coordinated flux partitioning among connected compounds within each structure (Fig. 6, Table S3). Differences in the strength of correlations between compound concentrations across network structures further demonstrate that the distribution of flux among the compounds follows network structure (Fig. 6, Table S3). The largest variation in compound concentrations occurred within network structures that had the strongest correlations of flux among compounds (Fig. 7), suggesting that the fluctuations in carotenoid concentrations on the same network are caused by coordinated changes in the total flux and not flux partitioning among the same compounds. Taken together, these findings support the prediction that the gain or loss of enzymatic reactions affects concentrations of expressed carotenoids more than changes in enzyme activity, leading to the coevolution of structural and dynamic properties of a network (Fig. 1a).

How does the gain or loss of reactions and compounds influence flux partitioning, and when should flux be decoupled from the structure of a network? We found that changes in network structure redistributed flux among compounds. These findings corroborate results of previous studies showing shifts in carbohydrate flux in glycolysis (Ralser et al., 2007) and carbon flux (Shi et al., 2010) in response to the activation or inactivation of reactions. In our study, flux dependency on the number of reactions differed among the compounds. The loss of derived compounds adonixanthin, β-cryptoxanthin, and astaxanthin led to an increase in the concentrations of other derived compounds (Fig. 5; Table S2), and concentrations of derived compounds depended on the number of incoming and outgoing reactions (Fig.

3). If the activity of an additional reaction is co-regulated by the same mechanisms as other reactions associated with a compound (Kacser & Burns, 1973; Mazat et al., 1996), then flux will be partitioned among more compounds, due to the conservation of mass through a pathway. Thus, when the amounts of substrates remain the same, the concentration of a compound should decrease with the addition of outgoing reactions (Brooks, 2004), as was observed in adonixanthin (Fig. 3i). On the contrary, the concentrations of the other derived compounds increased with the number of reactions (Fig. 3d-f; Fig.

3k). This could be caused by an increase in the total flux through the network, achieved by higher enzyme activity of incoming reactions (Kacser & Acerenza, 1993; Yang & Robb, 1994; Wehtje & Adlercreutz,

1997), or a greater amount of a consumed dietary compound (Wu et al., 2006; Taymaz-Nikerel et al.,

2011; Taymaz-Nikerel et al., 2013). The latter mechanism was supported by our finding that the total concentration of carotenoids (representing the mass available for carotenoid metabolism), differed between network structures (Fig. 5; Table S2). Alternatively, if the total flux remains constant when reactions are gained, as we found for some networks (Fig. 5; Table S2), then the addition of reactions could inhibit the activity of reactions directly associated with a compound (Kacser & Burns, 1981). This would result in a decrease of flux to the other products of these reactions, and therefore not affect the amount of the substrate partitioned into products.

The network structure affected flux of derived carotenoids to a greater extent than flux of dietary carotenoids (Fig. 4a), supporting our prediction that excess quantities of externally acquired dietary

compounds mask the effects of the gain or loss of reactions on their final concentrations. Indeed, birds consume more dietary carotenoids than they deposit into feathers or use to produce derived carotenoids

(Fox et al., 1969). The independence of dietary compounds from changes in network structure could account for the observed divergence of correlation structures between concentrations of dietary and derived carotenoids among network structures (Fig. 6; Table S3).

Although local changes in network structure did not influence the concentration of dietary carotenoids, the structural changes in the network caused by their absence affected concentrations of derived carotenoids (Fig. 5, Table S2). In particular, dietary zeaxanthin had the greatest effect on concentrations of derived compounds. In the house finch carotenoid network, derived carotenoids were located fewer reactions away from zeaxanthin than from other dietary compounds (Fig. 2a). If shorter pathways are energetically cheaper, then this structural property makes zeaxanthin a stronger contributor to the flux of derived compounds than other dietary compounds (Britton, 1976; Brush, 1981). The importance of zeaxanthin in maintaining avian carotenoid metabolism is further supported by the finding that some species preferentially accumulate higher proportions of zeaxanthin than other dietary compounds (McGraw et al., 2004). In this study, we found that zeaxanthin was the most prevalent and the least variable of the dietary carotenoids (Fig. 2a), suggesting a strong stabilizing selection for both the structural and dynamic contributions of dietary zeaxanthin to carotenoid metabolism in this population.

Strength of the association between network structure and metabolic flux can reflect the functional necessity of certain compounds in different environments. Environmentally-induced structural changes, such as fluctuations in the quantity of externally acquired compounds, can result in the redistribution of flux among products in the network (Handorf et al., 2005; Borenstein et al., 2008). Some compounds must always need to be produced despite environmental fluctuations (Barkai & Leibler, 1997;

Batchelor & Goulian, 2003; Kim et al., 2007; Shinar et al., 2009), whereas others are associated with specific environments (diCenzo et al., 2016). The avian carotenoid network structure includes both redundant and isolated biochemical pathways from dietary to derived compounds (Badyaev et al., 2015), and this facilitates the robustness of some compounds to changes in dietary inputs (Ma et al., 2009;

Shinar & Feinberg, 2010; Eloundou-Mbebi et al., 2016; Gao et al., 2016). We tested this idea by examining variation in concentration of compounds that differ in biochemical connectivity. We expected greater robustness in flux of the most connected compounds than compounds associated with fewer enzymatic reactions. However, we found no effect of connectivity or betweenness centrality of a compound on the strength of the relationship between its network structure and flux (Fig. 4b-c). This corroborates empirical findings that robustness of compound concentrations to environmental changes is conferred by the global interactions between all of the compounds in a network, and not by the structure of directly associated reactions (Ma et al., 2004; Inoue & Kaneko, 2013).

The finding that network structure accounted for variation in concentrations of plumage carotenoids at the population level suggests that the gain or loss of enzymatic reactions and dietary compounds is crucial for the evolution of local adaptations involving plumage-bound carotenoids. The fact that we observed such wide variation in the occurrence and concentration of carotenoids among individuals from the same population (Fig. 2a), suggests that activation and deactivation of enzymatic reactions can be accomplished rapidly and easily modulated (see also Badyaev & Duckworth, 2003).

Reversible regulatory changes could be driven by the environment during molt (Szasz, 1974; Ralser et al.,

2007; Link et al., 2013), or by hormonal inhibition or activation (Cohen, 1988; Strålfors & Honnor, 1989) that allow for short-term structural changes and rapid adaptation of carotenoid metabolism. When the environment remains consistent, these regulatory changes in enzyme activity may become permanent

(Emilsson et al., 2008; Gordon & Ruvinsky, 2012; Schaefke et al., 2013; Lopes et al., 2016; Mundy et al., 2016), leading to the evolution of distinct structures of metabolic networks.

The coevolution of network structure and flux dynamics delineates possible trajectories of evolutionary diversification of carotenoid metabolism. In all distinct networks under this study, individuals differed the most when compound concentrations changed proportionally (Fig. 6; Table S3).

Thus, the flux of groups of interconnected compounds that form functional modules are coordinated (Fell

& Thomas, 1995; Rossell et al., 2006). Indeed, the greatest among-individual variation in flux occurred in network structures that had the strongest positive correlations among compounds (Fig. 7). The modular

regulation of interconnected compounds suggests that selection acts on activity of all of the enzymes in a pathway. This is reflected in the evolutionary diversification of avian carotenoid metabolism, where the gain and loss of entire biochemical modules was more common than incremental gains or losses of individual compounds within the modules (Morrison & Badyaev, 2016b).

The mechanisms underlying changes in the structure of a biochemical network could also determine how and where diversification occurs in the network. Although only a small proportion of flux variation among individuals was caused by trade-offs among compound concentrations (Fig. 6; Table S3), these contribute the most to variation in network structure, and thus to flux. Novel enzymatic reactions are commonly added to the most connected compounds (Barabási & Albert, 1999; Jeong et al., 2000;

Barabási & Oltvai, 2004; Light et al., 2005), and this could constrain flux partitioning to only the most connected compounds in the network. As a result, variation in flux should be confined to compounds with fewer enzymatic reactions, leading to bursts of diversification in flux partitioning among them.

This study is one of the first comprehensive assessments of the correspondence between network structure and flux in a multicellular organism. Importantly, we found substantial variation in network structure and flux among individuals within a population, likely over shared genetic architecture underlying the network structure. The range of documented carotenoid network structures might allow for rapid adaptations of metabolic flux to changes in structural properties of the network, either as a result of environmental change or fitness consequences of the resulting products, ultimately leading to the coevolution of structural and dynamic network properties.

ACKNOWLEDGEMENTS

We thank D. Seaman, L. Kennedy, V. Belloni, K. King, and many research technicians for their help with collecting feathers and laboratory work. We thank K. Chenard, R. Duckworth, D. Higginson, K.

Hallinger, A. Potticary, and G. Semenov for helpful discussions. This research was supported by grants from the National Science Foundation, the David and Lucile Packard Foundation Fellowship, Amherst

College Graduate Fellowships and the University of Arizona Galileo Circle Scholarship.

REFERENCES

Albert, R., Jeong, H. & Barabási, A.-L. 2000. Error and attack tolerance of complex networks. Nature.

406: 378-382.

Almaas, E., Kovacs, B., Vicsek, T., Oltvai, Z.N. & Barabási, A.-L. 2004. Global organization of

metabolic fluxes in the bacterium Escherichia coli. Nature. 427: 839-843.

Almaas, E., Oltvai, Z.N. & Barabási, A.-L. 2005. The activity reaction core and plasticity of metabolic

networks. PLoS Comput. Biol. 1: e68.

Assenov, Y., Ramírez, F., Schelhorn, S.-E., Lengauer, T. & Albrecht, M. 2008. Computing topological

parameters of biological networks. Bioinformatics. 24: 282-284.

Badyaev, A.V., Morrison, E.S., Belloni, V. & Sanderson, M.J. 2015. Tradeoff between robustness and

elaboration in carotenoid networks produces cycles of avian color diversification. Biol. Direct.

10: 45.

Badyaev, A.V. & Duckworth, R.A. 2003. Context-dependent sexual advertisement: Plasticity in

development of sexual ornamentation throughout the lifetime of a passerine bird. J. Evol. Biol.

16: 1065-1076.

Badyaev, A.V. & Vleck, C.M. 2007. Context-dependent development of sexual ornamentation:

Implications for a trade-off between current and future breeding efforts. J. Evol. Biol. 20: 1277-

1287.

Barabási, A.-L. & Albert, R. 1999. Emergence of scaling in random networks. Science. 286: 509-512.

Barabási, A.-L. & Oltvai, Z.N. 2004. Network biology: Understanding the cell's functional organization.

Nat. Rev. Genet. 5: 101-113.

Barkai, N. & Leibler, S. 1997. Robustness in simple biochemical networks. Nature. 387: 913-917.

Batchelor, E. & Goulian, M. 2003. Robustness and the cycle of phosphorylation and dephosphorylation in

a two-component regulatory system. Proc. Natl. Acad. Sci. U.S.A. 100: 691-696.

Berg, J., Lässig, M. & Wagner, A. 2004. Structure and evolution of protein interaction networks: A

statistical model for link dynamics and gene duplications. BMC Evol. Biol. 4: 1-12.

Borenstein, E., Kupiec, M., Feldman, M.W. & Ruppin, E. 2008. Large-scale reconstruction and

phylogenetic analysis of metabolic environments. Proc. Natl. Acad. Sci. U.S.A. 105: 14482-

14487.

Brandes, U. 2001. A faster algorithm for betweenness centrality. J. Math Sociol. 25: 163-177.

Britton, G. 1976. Biosynthesis of carotenoids. In: Chemistry and Biochemistry of Plant Pigments, 2nd edn

(T.W. Goodwin, ed.), pp. 262-327. Academic Press, New York.

Brooks, S.P.J. 2004. Enzymes in the cell: What's really going on? In: Functional Metabolism: Regulation

and Adaptation (K.B. Storey, ed.), pp. 55-86. Wiley-Liss, Hoboken, NJ.

Brush, A.H. 1981. Carotenoids in wild and captive birds. In: Carotenoids and Colorants and Vitamin A

Precursors (J.C. Bauernfiend, ed.), pp. 539-562. Academic Press, New York.

Cohen, J. 1962. The statistical power of abnormal-social psychological research: A review. J. Abnorm.

Soc. Psychol. 65: 145-153.

Cohen, P. 1988. Review lecture: Protein phosphorylation and hormone action. Proc. R. Soc. Lond. B Biol.

Sci. 234: 115-144. diCenzo, G.C., Checcucci, A., Bazzicalupo, M., Mengoni, A., Viti, C., Dziewit, L. et al. 2016. Metabolic

modelling reveals the specialization of secondary replicons for niche adaptation in Sinorhizobium

meliloti. Nat. Commun. 7: 12219.

Doncheva, N.T., Assenov, Y., Domingues, F.S. & Albrecht, M. 2012. Topological analysis and

interactive visualization of biological networks and protein structures. Nat. Protoc. 7: 670-685.

Eanes, W.F. 1999. Analysis of selection on enzyme polymorphisms. Annu. Rev. Ecol. Syst. 30: 301-326.

Ebenhöh, O., Handorf, T. & Heinrich, R. 2005. A cross species comparison of metabolic network

functions. Genome Inform. 16: 203-213.

Eloundou-Mbebi, J.M.O., Küken, A., Omranian, N., Kleessen, S., Neigenfind, J., Basler, G. et al. 2016. A

network property necessary for concentration robustness. Nat. Commun. 7: 13255.

Emilsson, V., Thorleifsson, G., Zhang, B., Leonardson, A.S., Zink, F., Zhu, J. et al. 2008. Genetics of

gene expression and its effect on disease. Nature. 452: 423-428.

Emmerling, M., Dauner, M., Ponti, A., Fiaux, J., Hochuli, M., Szyperski, T. et al. 2002. Metabolic flux

responses to pyruvate kinase knockout in Escherichia coli. J. Bacteriol. 184: 152-164.

Fell, D.A. 1997. Understanding the Control of Metabolism. Portland Press, Miami, FL.

Fell, D.A. & Thomas, S. 1995. Physiological control of metabolic flux: the requirement for multisite

modulation. Biochem. J. 311: 35-39.

Flowers, J.M., Sezgin, E., Kumagai, S., Duvernell, D.D., Matzkin, L.M., Schmidt, P.S. et al. 2007.

Adaptive evolution of metabolic pathways in Drosophila. Mol. Biol. Evol. 24: 1347-1354.

Fox, D.L., Wolfson, A.A. & McBeth, J.W. 1969. Metabolism of β-carotene in the American flamingo,

Phoenicopterus ruber. Comp. Biochem. Phys. 29: 1223-1229.

Gao, J., Barzel, B. & Barabási, A.-L. 2016. Universal resilience patterns in complex networks. Nature.

530: 307-312.

Gordon, K.L. & Ruvinsky, I. 2012. Tempo and mode in evolution of transcriptional regulation. PLoS

Genet. 8: e1002432.

Handorf, T., Ebenhöh, O. & Heinrich, R. 2005. Expanding metabolic networks: Scopes of compounds,

robustness and evolution. J. Mol. Evol. 61: 498-512.

Hartl, D.L., Dykhuizen, D.E. & Dean, A.M. 1985. Limits of adaptation: The evolution of selective

neutrality. Genetics. 111: 655-674.

Heinrich, R. & Rapoport, T.A. 1974. A linear steady-state treatment of enzymatic chains. Eur. J.

Biochem. 42: 89-95.

Higginson, D.M., Belloni, V., Davis, S.N., Morrison, E.S., Andrews, J.E. & Badyaev, A.V. 2016.

Evolution of long-term coloration trends with biochemically unstable ingredients. Proc. R. Soc.

Lond. B Biol. Sci. 283: 20160403.

Inoue, M. & Kaneko, K. 2013. Cooperative adaptive responses in gene regulatory networks with many

degrees of freedom. PLoS Comput. Biol. 9: e1003001.

Inouye, C.Y., Hill, G.E., Stradi, R.D., Montgomerie, R. & Bosque, C. 2001. Carotenoid pigments in male

house finch plumage in relation to age, subspecies, and ornamental coloration. Auk. 118: 900-915.

Iyer, V.V., Sriram, G., Fulton, D.B., Zhou, R., Westgate, M.E. & Shanks, J.V. 2008. Metabolic flux maps

comparing the effect of temperature on protein and oil biosynthesis in developing soybean

cotyledons. Plant Cell Environ. 31: 506-517.

Jeong, H., Tombor, B., Albert, R., Oltvai, Z.N. & Barabási, A.-L. 2000. The large-scale organization of

metabolic networks. Nature. 407: 651-654.

Kacser, H. & Acerenza, L. 1993. A universal method for achieving increases in metabolite production.

Eur. J. Biochem. 216: 361-367.

Kacser, H. & Burns, J.A. 1973. The control of flux. Sym. Soc. Exp. Biol. 27: 65-104.

Kacser, H. & Burns, J.A. 1981. The molecular basis of dominance. Genetics. 97: 639-666.

Kim, J.-Y. & Ryu, D.D.Y. 1999. Physiological and environmental effects on metabolic flux change

caused by heterologous gene expression in Escherichia coli. Biotechnol. Bioprocess Eng. 4: 170.

Kim, P.-J., Lee, D.-Y., Kim, T.Y., Lee, K.H., Jeong, H., Lee, S.Y. et al. 2007. Metabolite essentiality

elucidates robustness of Escherichia coli metabolism. Proc. Natl. Acad. Sci. U.S.A. 104: 13638-

13642.

Klassen, J.L. 2010. Phylogenetic and evolutionary patterns in microbial carotenoid biosynthesis are

revealed by comparative genomics. PLoS ONE. 5: e11257.

Kondrashov, F.A. 2012. Gene duplication as a mechanism of genomic adaptation to a changing

environment. Proc. R. Soc. Lond. B Biol. Sci. 279: 5048-5057.

Kramer, C.Y. 1956. Extension of multiple range tests to group means with unequal numbers of

replications. Biometrics. 12: 307-310.

Landeen, E.A. & Badyaev, A.V. 2012. Developmental integration of feather growth and pigmentation and

its implications for the evolution of diet-derived coloration. J. Exp. Zool Part B. 318: 59-70.

Lee, J.W., Na, D., Park, J.M., Lee, J., Choi, S. & Lee, S.Y. 2012a. Systems metabolic engineering of

microorganisms for natural and non-natural chemicals. Nat. Chem. Biol. 8: 536-546.

Lee, S.H., Bernhardsson, S., Holme, P., Kim, B.J. & Minnhagen, P. 2012b. Neutral theory of chemical

reaction networks. New J. Phys. 14: 033032.

Light, S., Kraulis, P. & Elofsson, A. 2005. Preferential attachment in the evolution of metabolic networks.

BMC Genom. 6: 159.

Link, H., Kochanowski, K. & Sauer, U. 2013. Systematic identification of allosteric protein-metabolite

interactions that control enzyme activity in vivo. Nature Biotechnol. 31: 357-361.

Lopes, R.J., Johnson, J.D., Toomey, M.B., Ferreira, M.S., Araujo, P.M., Melo-Ferreira, J. et al. 2016.

Genetic basis for red coloration in birds. Curr. Biol. 26: 1427-1434.

Lynch, M. 2007. The frailty of adaptive hypotheses for the origins of organismal complexity. Proc. R.

Soc. Lond. B Biol. Sci. 104: 8597-8604.

Ma, H.-W., Kumar, B., Ditges, U., Gunzer, F., Buer, J. & Zeng, A.-P. 2004. An extended transcriptional

regulatory network of Escherichia coli and analysis of its hierarchical structure and network

motifs. Nucleic Acids Res. 32: 6643-6649.

Ma, W., Trusina, A., El-Samad, H., Lim, W.A. & Tang, C. 2009. Defining network topologies that can

achieve biochemical adaptation. Cell. 138: 760-773.

Mazat, J.-P., Reder, C. & Letellier, T. 1996. Why are most flux control coefficients so small? J. Theor.

Biol. 182: 253-258.

McGraw, K.J., Hill, G.E., Navara, K.J. & Parker, R.S. 2004. Differential accumulation and pigmenting

ability of dietary carotenoids in colorful finches. Physiol. Biochem. Zool. 77: 484-491.

McGraw, K.J., Nolan, P.M. & Crino, O.L. 2006. Carotenoid accumulation strategies for becoming a

colourful House Finch: Analyses of plasma and liver pigments in wild moulting birds. Funct.

Ecol. 20: 678-688.

Morrison, E.S. & Badyaev, A.V. 2016a. The landscape of evolution: Reconciling structural and dynamic

properties of metabolic networks in adaptive diversifications. Integr. Comp. Biol. 56: 235-246.

Morrison, E.S. & Badyaev, A.V. 2016b. Structuring evolution: Biochemical networks and metabolic

diversification in birds. BMC Evol. Biol. 16: 168.

Mundy, N.I., Stapley, J., Bennison, C., Tucker, R., Twyman, H., Kim, K.-W., Burke, T. et al. 2016. Red

carotenoid coloration in the Zebra Finch is controlled by a cytochrome P450 gene cluster. Curr.

Biol. 26: 1435-1440.

Nidelet, T., Brial, P., Camarasa, C. & Dequin, S. 2016. Diversity of flux distribution in central carbon

metabolism of S. cerevisiae strains from diverse environments. Microb. Cell Fact. 15: 1-13.

Nilsson, A. & Nielsen, J. (2016) Metabolic trade-offs in yeast are caused by F1F0-ATP synthase. Sci.

Rep. 6: 22264.

Olson-Manning, C.F., Lee, C.-R., Rausher, M.D. & Mitchell-Olds, T. 2013. Evolution of flux control in

the glucosinolate pathway in Arabidopsis thaliana. Mol. Biol. Evol. 30: 14-23.

Pál, C., Papp, B. & Lercher, M.J. 2005. Adaptive evolution of bacterial metabolic networks by horizontal

gene transfer. Nature Genet. 37: 1372-1375.

Papp, B., Teusink, B. & Notebaart, R.A. 2009. A critical view of metabolic network adaptations. HFSP J.

3: 24-35.

Piedrafita, G., Keller, M.A. & Ralser, M. 2015. The impact of non-enzymatic reactions and enzyme

promiscuity on cellular metabolism during (oxidative) stress conditions. Biomolecules. 5: 2101-

2122.

Ralser, M., Wamelink, M.M., Kowald, A., Gerisch, B., Heeren, G., Struys, E.A. et al. 2007. Dynamic

rerouting of the carbohydrate flux is key to counteracting oxidative stress. J. Biol. 6: 10.

Rausher, M.D. 2013. The evolution of genes in branched metabolic pathways. Evolution. 67: 34-48.

Rossell, S., van der Weijden, C.C., Lindenbergh, A., van Tuijl, A., Francke, C., Bakker, B.M. et al. 2006.

Unraveling the complexity of flux regulation: A new method demonstrated for nutrient starvation

in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U.S.A. 103: 2166-2171.

Sadana, A. 1988. Enzyme deactivation. Biotechnol. Adv. 6: 349-446.

Sauer, U., Lasko, D.R., Fiaux, J., Hochuli, M., Glaser, R., Szyperski, T. et al. 1999. Metabolic flux ratio

analysis of genetic and environmental modulations of Escherichia coli central carbon

metabolism. J. Bacteriol. 181: 6679-6688.

Schaefke, B., Emerson, J.J., Wang, T.-Y., Lu, M.-Y.J., Hsieh, L.-C. & Li, W.-H. 2013. Inheritance of

gene expression level and selective constraints on trans- and cis-regulatory changes in yeast. Mol.

Biol. Evol. 30: 2121-2133.

Schmidt, S., Sunyaev, S., Bork, P. & Dandekar, T. 2003. Metabolites: a helping hand for pathway

evolution? Trends Biochem. Sci. 28: 336-341.

Shi, L., Sohaskey, C.D., Pheiffer, C., Datta, P., Parks, M., McFadden, J. et al. 2010. Carbon flux rerouting

during Mycobacterium tuberculosis growth arrest. Mol. Microbiol. 78: 1199-1215.

Shinar, G. & Feinberg, M. 2010. Structural sources of robustness in biochemical reaction networks.

Science. 327: 1389-1391.

Shinar, G., Rabinowitz, J.D. & Alon, U. 2009. Robustness in glyoxylate bypass regulation. PLoS

Computational Biology, 5, e1000297.

Smoot, M.E., Ono, K., Ruscheinski, J., Wang, P.-L. & Ideker, T. (2011) Cytoscape 2.8: new features for

data integration and network visualization. Bioinformatics, 27, 431-432.

Soyer, O.S. & Pfeiffer, T. (2010) Evolution under fluctuating environments explains observed robustness

in metabolic networks. PLoS Comput. Biol. 6: e1000907.

Stelling, J., Klamt, S., Bettenbrock, K., Schuster, S. & Gilles, E.D. 2002. Metabolic network structure

determines key aspects of functionality and regulation. Nature. 420: 190-193.

Strålfors, P. & Honnor, R.C. 1989. Insulin-induced dephosphorylation of hormone-sensitive lipase. Eur.

J. Biochem. 182: 379-385.

Szasz, G. 1974. The effect of temperature on enzyme activity and on the affinity of enzymes to their

substrates. Z. Klin. Chem. Klin. Biochem. 12: 166-170.

Taymaz-Nikerel, H., De Mey, M., Baart, G., Maertens, J., Heijnen, J.J. & van Gulik, W. 2013. Changes in

substrate availability in Escherichia coli lead to rapid metabolite, flux and growth rate responses.

Metab. Eng. 16: 115-129.

Taymaz-Nikerel, H., van Gulik, W.M. & Heijnen, J.J. 2011. Escherichia coli responds with a rapid and

large change in growth rate upon a shift from glucose-limited to glucose-excess conditions.

Metab. Eng. 13: 307-318.

Teusink, B. & Westerhoff, H.V. 2000. ‘Slave’ metabolites and enzymes. Eur. J. Biochem. 267: 1889-

1893.

Torres-Sosa, C., Huang, S. & Aldana, M. 2012. Criticality is an emergent property of genetic networks

that exhibit evolvability. PLoS Comput. Biol. 8: e1002669.

Wagner, A. 2001. The yeast protein interaction network evolves rapidly and contains few redundant

duplicate genes. Mol. Biol. Evol. 18: 1283-1292.

Wagner, A. 2003a. Does selection mold molecular networks? Sci. STKE. 2003: pe41.

Wagner, A. 2003b. How the global structure of protein interaction networks evolves. Proc. R. Soc. Lond.

B Biol. Sci. 270: 457-466.

Wagner, A. & Fell, D.A. 2001. The small world inside large metabolic networks. Proc. R. Soc. Lond. B

Biol. Sci. 268: 1803.

Wehtje, E. & Adlercreutz, P. 1997. Water activity and substrate concentration effects on lipase activity.

Biotechnol. Bioeng. 55: 798-806.

Wright, K.M. & Rausher, M.D. 2010. The evolution of control and distribution of adaptive mutations in a

metabolic pathway. Genetics. 184: 483-502.

Wu, L., van Dam, J., Schipper, D., Kresnowati, M.T.A.P., Proell, A.M., Ras, C., et al. 2006. Short-term

metabolome dynamics and carbon, electron, and ATP balances in chemostat-grown

Saccharomyces cerevisiae CEN.PK 113-7D following a glucose pulse. Appl. Environ. Microbiol.

72: 3566-3577.

Yang, Z. & Robb, D.A. 1994. Partition coefficients of substrates and products and solvent selection for

biocatalysis under nearly anhydrous conditions. Biotechnol. Bioeng. 43: 365-370.

Yoon, J., Blumer, A. & Lee, K. 2006. An algorithm for modularity analysis of directed and weighted

biological networks based on edge-betweenness centrality. Bioinformatics. 22: 3106-3108.

Fig. 1 Correspondence of network structural and dynamic properties affects correlation between compound concentrations. Network structure is defined by the connectivity of its compounds (circles) and reactions (arrows). The concentrations of compounds represent dynamic properties of flux. (a) If the structure of a network corresponds to a specific distribution of flux, then there should be proportional changes between the concentrations of the compounds (top panel). Differences in compound connectivity (reactions per compound) would redistribute flux in the network and correspond to distinct compound concentrations (bottom panel). (b) If flux is partitioned differently on the same network structure, then changes in the concentrations of compounds would not be correlated (top panel). The concentration of a compound would vary independently of structural changes in the number of reactions (bottom panel). The asterisk denotes that the prediction was supported by this study.

a β-carotene 3'-dehydrolutein β-cryptoxanthin lutein β-isocryptoxanthin zeaxanthin

4'-hydroxy- canary zeaxanthin echinenone xanthopyll B canary isozeaxanthin xanthopyll A 3'-hydroxy- echinenone (3S,4R,3'S,6'R) 4-hydroxylutein

4-hydroxy- adonixanthin echinenone canthaxanthin idoxanthin α-doradexanthin

derived diet adonirubin α-carotene rubixanthin gaziaxanthin CV = 1 50% astaxanthin occurrence

b 25

4-oxo-rubixanthin 20

percent of individuals of percent 0 5 10 15 20 expressed compounds

Fig. 2 Occurrence and concentration of plumage carotenoids in male house finches. (a) The species’ carotenoid metabolic network includes 24 compounds (circles) and 45 reactions (arrows). Grey shading indicates the percentage of n = 442 individuals in which the carotenoid occurs, and the diameter of the circles indicates the among-individual coefficients of variation (CV) for the concentration (μg compound/g pigmented feather) of each compound. Dietary carotenoids (green circles) are the start of pathways that produce derived carotenoids (red circles) via enzymatic reactions (arrows). Small open circles show carotenoids not examined in this study. (b) The number of compounds identified in each individual.

Fig. 3 Carotenoid concentrations are associated with their structural properties. Shown are the least- squares means (± 1 standard error, SE) for the concentrations of dietary (a-c) and derived (d-k) carotenoids partitioned by number of incoming (in) or outgoing (out) reactions per compound. Horizontal

lines indicate no difference between groups at P < 0.05. The numbers above each bar are sample sizes. See Table S1 for a summary of the statistics.

Fig. 4 The relationship between compound concentration and network structure is dependent on compound pathway position, but not connectivity. (a) The number of reactions affected concentrations of derived compounds more than dietary compounds. Differences in the connectivity of compounds in the species network, as measured by (b) number of reactions and (c) betweenness centrality, were not related to the effect of changes in the number of reactions on compound concentrations. The number of reactions from a dietary compound separates starting dietary compounds (0 reactions) from derived compounds (>0 reactions from dietary compounds). The average concentration change is the absolute value of the mean difference (Cohen’s d) between concentrations of a compound with different numbers of reactions.

Fig. 5 Carotenoid concentration differs among network structures. Shown are the least-squares means for the concentrations of compounds across network structures (left). Solid lines show compounds whose concentration differed among network structures and dashed lines show compounds whose concentrations did not vary across network structures. Dietary carotenoids are shown in shades of green, the total concentration (sum of all compound concentrations in the network) in black, and the remaining colors represent derived compounds. Observed network structures (n = 11) are as follows: A = all compounds present; B = β-cryptoxanthin absent; C = β-cryptoxanthin and β-carotene absent; D = β- cryptoxanthin and zeaxanthin absent; E = adonixanthin absent; F = adonixanthin and β-cryptoxanthin absent; G = adonixanthin, β-cryptoxanthin and β-carotene absent; H = adonixanthin, β-isocryptoxanthin and β-cryptoxanthin absent; I = adonixanthin, β –isocryptoxanthin, β-cryptoxanthin and β-carotene absent; J = astaxanthin absent; K = astaxanthin and β-cryptoxanthin absent. Shown on the right is the maximum difference (Cohen’s d effect size) between the least-squares means of the concentrations of a compound in different networks (Table S3). The asterisk denotes significant pairwise differences in compound concentration between different structures at P < 0.05. See Table S2 for a summary of the statistics.

100

Fig. 6 Concentrations of compounds in a metabolic network are strongly correlated, but the strength of these correlations differs among network structures. Panels show correlations of each compound concentration with PC1 (left column) and PC 2 (right column) for each network structure A-K (see the legend of figure 4 for network descriptions). The eigenvalue for each PC and the percent of variation explained by it (in parentheses) are below each panel. Dietary compounds are shown in shades of green and the remaining colors represent derived compounds. Filled and unfilled bars denote positive and negative correlations, respectively. The number below each panel letter shows the number of individuals with this network structure. See Table S3 for a summary of the statistics.

101

Fig. 7 The magnitude of flux varies the most in networks with stronger correlations among compounds. In individuals with the same structure, flux varied with correlated changes in all network compounds. Concentration variation is the standard deviation of the total carotenoid concentrations (sum of all of the compound concentrations in a network) among individuals with the same network structure.

102

Table S1: Results of the pairwise Tukey-Kramer tests between the compound concentrations of individuals with different numbers of reactions per compound. Reaction group represents the number of reactions per compound (in bold), “in” denotes incoming reactions and “out” denotes outgoing reactions. The sample size (n) and the least-squares mean (± S.E.) of the compound concentration, log(μg compound/g feather), are below the reaction group name. Asterisks denote dietary compounds.

COMPOUND Reaction Group 1 Reaction Group 2 t P 1-2 out 3 out n = 11 n = 188 0.966 0.599 2.587 ± 0.083 2.505 ± 0.020 1-2 out 4 LUTEIN* n = 11 n = 243 0.792 0.708 2.587 ± 0.083 2.520 ± 0.018 3 out 4 n = 188 n = 243 0.573 0.835 2.505 ± 0.020 2.520 ± 0.018 1 out 2 out ZEAXANTHIN* n = 13 n = 401 1.15 0.251 2.094 ± 0.058 2.163 ± 0.010 0-1 out 2 out β-CAROTENE* n = 38 n = 346 1.37 0.173 1.800 ± 0.051 1.873 ± 0.017 1 in 2 in 3’-DEHYDROLUTEIN n = 18 n = 422 1.83 0.068 2.676 ± 0.058 2.568 ± 0.012 2-3 in 4 in ADONIXANTHIN n = 86 n = 153 2.24 0.026 2.089 ± 0.026 2.017 ± 0.019 1 out 2 out ASTAXANTHIN n = 170 n = 215 0.97 0.334 1.979 ± 0.032 1.938 ± 0.028 1-2 in, 0-1 out 3 in, 3 out ECHINENONE n = 40 n = 384 4.80 < 0.0001 2.377 ± 0.061 2.686 ± 0.020 1-2 in, 0-1 out 3 in, 0-1 out n = 20 n = 307 0.58 0.938 2.677 ± 0.096 2.781 ± 0.024 1-2 in, 0-1 out 1-2 in, 2 out n = 20 n = 8 7.00 < 0.0001 2.677 ± 0.096 3.370 ± 0.152 1-2 in, 0-1 out 3 in, 2 out n = 20 n = 101 7.67 < 0.0001 2.677 ± 0.096 3.482 ± 0.043 3’-HYDROXY-ECHINENONE 3 in, 0-1 out 1-2 in, 2 out n = 307 n = 8 3.84 0.0008 2.781 ± 0.024 3.370 ± 0.152 3 in, 0-1 out 3 in, 2 out n = 307 n = 101 4.45 < 0.0001 2.781 ± 0.024 3.482 ± 0.043 1-2 in, 2 out 3 in, 2 out n = 8 n = 101 2.27 0.107 3.370 ± 0.152 3.482 ± 0.043 1 out 2 out CANTHAXANTHIN n = 13 n = 410 5.63 < 0.0001 2.080 ± 0.098 2.639 ± 0.017 1-2 in, 0-2 out 3 in, 3 out ADONIRUBIN n = 20 n = 405 2.95 0.0034 1.987 ± 0.061 2.170 ± 0.013 1 in 2 in β-ISOCRYPTOXANTHIN n = 9 n = 382 1.87 0.063 1.787 ± 0.118 2.010 ± 0.018

103

Table S2: The maximum change (Cohen’s d effect size) in the concentration of a compound, log(μg compound/g feather), between distinct network structures. The network structure letter (A-K) is written above the corresponding minimum or maximum least-squares mean concentration (lsmean) of a compound. Asterisks denote dietary compounds. Results of the pairwise Tukey-Kramer tests between the compound concentrations of the minimum and maximum concentrations are shown. The maximum effect size among significant pairwise differences in compound concentrations at P < 0.05 were used for compounds that had significant pairwise differences in concentration between network structures. See Fig. 5 for the distribution of least-squares mean concentrations for all of the network structures. Network structures are as follows and correspond to the letters in Fig. 5 and Fig. 6: A=all compounds present; C= β-cryptoxanthin and β-carotene absent; D= β-cryptoxanthin and zeaxanthin absent; E=adonixanthin absent; G=adonixanthin, β-cryptoxanthin and β-carotene absent; J=astaxanthin absent.

Minimum Maximum Effect COMPOUND t P Concentration Concentration Size (lsmean ± S.E.) (lsmean ± S.E.) D J CANTHAXANTHIN 3.644 4.376 0.001 2.079 ± 0.139 2.855 ± 0.110 D G ECHINENONE 3.274 4.101 0.003 2.140 ± 0.148 2.807 ± 0.066 3’-HYDROXY- D G 3.096 3.878 0.006 ECHINENONE 2.873 ± 0.159 3.550 ± 0.071 TOTAL D G 2.352 3.361 0.034 CONCENTRATION 3.309 ± 0.125 3.767 ± 0.056 D G β-ISOCRYPTOXANTHIN 2.341 3.467 0.017 1.617 ± 0.150 2.186 ± 0.068 D J ADONIRUBIN 1.759 2.282 0.449 1.962 ± 0.113 2.290 ± 0.089 D E β-CAROTENE* 1.171 1.873 0.571 1.693 ± 0.132 1.975 ± 0.074 G J LUTEIN* 1.387 3.049 0.086 2.374 ± 0.050 2.689 ± 0.088 D G ASTAXANTHIN 1.361 2.488 0.242 1.660 ± 0.179 2.147 ± 0.080 C J 3’-DEHYDROLUTEIN 0.862 1.795 0.783 2.467 ± 0.072 2.660 ± 0.080 C E ZEAXANTHIN* 0.827 1.718 0.785 2.096 ± 0.066 2.241 ± 0.052 A J ADONIXANTHIN 0.561 0.493 0.996 1.953 ± 0.044 2.101 ± 0.067 J E β-CRYPTOXANTHIN* 0.214 0.353 0.934 1.940 ± 0.105 1.986 ± 0.074

Table S3: Results from the principal component analyses of compound concentration variation among all of the compounds present in a unique network structure. 104 The eigenvalue (λ) and percent variance are listed below each principal component.

NETWORK A: all compounds present (n = 30) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 Prin11 Prin12 λ = 8.517 λ = 1.444 λ = 0.645 λ = 0.480 λ = 0.354 λ = 0.256 λ = 0.119 λ = 0.089 λ = 0.034 λ = 0.029 λ = 0.022 λ = 0.010 70.98% 12.03% 5.37% 4.00% 2.95% 2.13% 0.99% 0.74% 0.29% 0.24% 0.19% 0.09% zeaxanthin 0.327 -0.013 -0.196 0.065 0.245 0.207 0.229 -0.337 -0.294 -0.592 -0.288 0.248 adonirubin 0.324 -0.195 -0.087 -0.154 -0.116 0.073 0.294 0.026 -0.190 0.029 0.814 0.143 β-isocryptoxanthin 0.312 0.079 0.009 -0.293 -0.121 0.485 -0.640 -0.264 0.261 0.034 0.063 0.090 3'-dehydrolutein 0.309 0.202 -0.361 0.264 0.027 -0.037 -0.134 0.141 0.031 -0.214 0.134 -0.748 3'-hydroxy-echinenone 0.307 -0.276 0.106 -0.007 -0.253 0.193 0.523 -0.249 0.385 0.278 -0.284 -0.276 astaxanthin 0.300 -0.263 0.188 -0.154 -0.392 0.065 -0.119 0.623 -0.333 -0.136 -0.303 0.000 adonixanthin 0.293 0.363 -0.149 -0.231 0.243 -0.040 0.052 -0.041 -0.466 0.622 -0.194 -0.029 echinenone 0.292 -0.195 0.326 0.046 -0.124 -0.671 -0.270 -0.448 -0.146 -0.038 0.015 -0.074 lutein 0.290 0.086 -0.385 0.541 -0.243 -0.182 -0.079 0.135 0.237 0.199 -0.068 0.501 β-cryptoxanthin 0.285 0.290 0.052 -0.479 0.243 -0.369 0.160 0.274 0.491 -0.218 -0.028 0.129 canthaxanthin 0.246 -0.359 0.327 0.328 0.688 0.137 -0.130 0.222 0.105 0.155 0.055 0.015 β-carotene 0.122 0.616 0.625 0.327 -0.168 0.182 0.151 0.007 -0.055 -0.084 0.117 0.033

NETWORK B: β-cryptoxanthin absent (n = 159) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 Prin11 λ = 7.189 λ = 1.821 λ = 0.628 λ = 0.383 λ = 0.314 λ = 0.282 λ = 0.156 λ = 0.087 λ = 0.068 λ = 0.048 λ = 0.023 65.36% 16.55% 5.71% 3.48% 2.86% 2.56% 1.42% 0.79% 0.62% 0.44% 0.21% adonirubin 0.349 0.011 -0.094 -0.272 -0.292 -0.084 -0.337 -0.491 -0.577 -0.028 -0.121 canthaxanthin 0.332 -0.243 -0.112 0.113 -0.093 -0.356 0.115 -0.456 0.629 -0.108 -0.204 echinenone 0.325 -0.302 -0.072 0.180 0.147 -0.141 0.257 0.247 -0.308 -0.677 0.210 3'-hydroxy-echinenone 0.318 -0.331 -0.099 0.098 0.115 -0.060 0.453 0.081 -0.257 0.689 -0.013 zeaxanthin 0.318 0.176 -0.423 -0.189 0.119 -0.364 -0.440 0.424 0.177 0.163 0.275 adonixanthin 0.302 0.200 -0.204 -0.509 0.402 0.532 0.256 -0.141 0.159 -0.113 -0.018 astaxanthin 0.298 -0.204 0.393 -0.343 -0.608 0.210 0.036 0.371 0.203 -0.001 0.051 β-isocryptoxanthin 0.291 -0.306 0.168 0.432 0.249 0.482 -0.544 -0.044 0.085 0.078 0.042 3'-dehydrolutein 0.281 0.440 -0.117 0.339 -0.132 0.087 0.058 0.305 -0.054 -0.061 -0.687 β-carotene 0.244 0.272 0.739 -0.114 0.415 -0.360 -0.016 -0.023 -0.035 0.040 -0.046 lutein 0.239 0.516 0.032 0.380 -0.273 0.116 0.204 -0.224 0.037 0.061 0.588

NETWORK C: β-cryptoxanthin and β-carotene absent (n = 10) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 λ = 7.635 λ = 1.401 λ = 0.505 λ = 0.245 λ = 0.098 λ = 0.080 λ = 0.030 λ = 0.006 λ = 0.000 λ = 0.000 76.35% 14.01% 5.05% 2.45% 0.98% 0.80% 0.30% 0.06% 0.00% 0.00% adonirubin 0.353 0.066 0.177 -0.135 -0.153 -0.003 -0.804 -0.285 -0.033 -0.263 zeaxanthin 0.341 0.216 0.014 0.165 -0.379 -0.524 0.244 -0.100 0.565 -0.044 3'-hydroxy-echinenone 0.338 -0.272 0.032 -0.076 0.129 -0.502 -0.038 0.630 -0.341 -0.146 echinenone 0.329 -0.349 0.020 0.042 -0.085 -0.054 -0.048 -0.240 -0.145 0.823 canthaxanthin 0.318 -0.360 -0.242 -0.010 0.368 -0.041 0.318 -0.551 -0.103 -0.398 astaxanthin 0.314 0.216 -0.114 -0.824 -0.169 0.245 0.254 0.086 -0.004 0.049 3'-dehydrolutein 0.307 0.349 -0.290 0.442 -0.378 0.249 0.134 0.035 -0.522 -0.087 lutein 0.304 0.339 -0.424 0.141 0.595 0.122 -0.218 0.200 0.314 0.196 β-isocryptoxanthin 0.285 -0.487 0.141 0.199 -0.177 0.564 0.036 0.316 0.386 -0.153 adonixanthin 0.263 0.322 0.782 0.112 0.342 0.113 0.241 -0.047 -0.110 0.030 105

NETWORK D: β-cryptoxanthin and zeaxanthin absent (n = 5) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 λ = 7.546 λ = 1.556 λ = 0.709 λ = 0.189 λ = 0.000 λ = 0.000 λ = 0.000 λ = 0.000 λ = 0.000 λ = 0.000 75.46% 15.56% 7.09% 1.89% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% adonirubin 0.353 -0.191 0.038 0.015 0.915 0.000 0.000 0.000 0.000 0.000 adonixanthin 0.353 0.150 -0.167 -0.156 -0.096 -0.890 0.000 0.000 0.000 0.000 canthaxanthin 0.346 -0.152 0.156 0.470 -0.179 0.020 -0.218 0.055 -0.727 0.000 3'-dehydrolutein 0.340 0.277 -0.066 -0.125 -0.069 0.223 -0.107 -0.849 0.000 0.000 3'-hydroxy-echinenone 0.340 -0.261 -0.168 0.044 -0.180 0.134 0.858 0.000 0.000 0.000 lutein 0.333 0.278 -0.217 -0.219 -0.058 0.264 -0.132 0.364 0.002 -0.702 echinenone 0.330 -0.269 0.191 0.468 -0.199 -0.011 -0.239 0.004 0.677 -0.094 β-isocryptoxanthin 0.321 0.298 -0.349 -0.013 -0.047 0.250 -0.153 0.343 0.098 0.686 astaxanthin 0.273 -0.242 0.618 -0.634 -0.171 0.081 -0.077 0.121 -0.020 0.163 β-carotene 0.041 0.691 0.572 0.262 0.100 -0.032 0.318 0.102 0.054 0.007

NETWORK E: adonixanthin absent (n = 16) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 Prin11 λ = 7.015 λ = 2.355 λ = 0.856 λ = 0.365 λ = 0.220 λ = 0.079 λ = 0.052 λ = 0.034 λ = 0.015 λ = 0.007 λ = 0.004 63.78% 21.40% 7.78% 3.32% 2.00% 0.72% 0.47% 0.31% 0.14% 0.06% 0.03% zeaxanthin 0.358 0.142 -0.085 0.317 -0.173 -0.077 -0.104 -0.089 -0.264 -0.777 0.127 adonirubin 0.342 0.050 -0.208 -0.172 0.725 0.138 0.198 -0.328 -0.289 0.024 -0.180 3'-hydroxy-echinenone 0.339 -0.263 -0.062 0.009 0.072 -0.109 -0.607 0.308 -0.382 0.333 0.274 echinenone 0.319 -0.329 -0.037 0.223 -0.198 0.015 -0.188 -0.095 0.197 0.066 -0.782 3'-dehydrolutein 0.318 0.309 -0.074 0.153 -0.364 -0.006 0.541 0.266 -0.371 0.364 -0.086 astaxanthin 0.314 -0.263 -0.260 -0.398 0.038 -0.497 0.283 0.329 0.360 -0.181 0.085 canthaxanthin 0.295 -0.378 -0.125 0.180 -0.209 0.293 0.193 -0.466 0.257 0.190 0.486 β-carotene 0.278 0.227 0.515 -0.475 -0.248 -0.295 -0.131 -0.452 -0.050 0.089 -0.004 lutein 0.267 0.398 -0.206 -0.374 -0.127 0.604 -0.247 0.224 0.303 -0.062 -0.008 β-isocryptoxanthin 0.235 -0.224 0.738 0.090 0.233 0.297 0.204 0.360 0.075 -0.146 0.042 β-cryptoxanthin 0.218 0.481 0.066 0.482 0.301 -0.301 -0.131 0.002 0.478 0.213 0.109

NETWORK F: adonixanthin and β-cryptoxanthin absent (n = 104) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 λ = 6.530 λ = 1.765 λ = 0.543 λ = 0.468 λ = 0.227 λ = 0.147 λ = 0.132 λ = 0.118 λ = 0.045 λ = 0.024 65.30% 17.65% 5.43% 4.68% 2.27% 1.47% 1.32% 1.18% 0.45% 0.24% adonirubin 0.362 -0.035 -0.134 0.243 0.000 -0.097 -0.826 -0.312 0.032 -0.010 echinenone 0.358 -0.216 -0.162 -0.185 -0.206 -0.236 0.129 0.018 -0.802 -0.081 3'-hydroxy-echinenone 0.350 -0.200 -0.235 -0.219 -0.238 -0.531 0.144 0.193 0.529 0.242 canthaxanthin 0.343 -0.232 -0.250 0.110 -0.138 0.485 0.389 -0.537 0.199 -0.141 zeaxanthin 0.320 0.331 -0.276 0.101 -0.218 0.503 -0.092 0.612 -0.020 0.133 β-isocryptoxanthin 0.304 -0.265 0.183 -0.587 0.587 0.280 -0.132 0.109 0.047 0.068 β-carotene 0.296 0.074 0.819 -0.061 -0.470 0.071 -0.017 -0.035 0.060 -0.027 astaxanthin 0.287 -0.304 0.226 0.684 0.392 -0.131 0.207 0.296 -0.009 -0.069 3'-dehydrolutein 0.268 0.529 -0.050 -0.121 0.199 -0.221 0.107 0.000 0.118 -0.720 lutein 0.253 0.549 0.054 0.068 0.276 -0.124 0.220 -0.318 -0.129 0.607 106

NETWORK G: adonixanthin, β-cryptoxanthin and β-carotene absent (n = 25) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 λ = 6.237 λ = 1.495 λ = 0.553 λ = 0.356 λ = 0.201 λ = 0.067 λ = 0.061 λ = 0.024 λ = 0.007 69.30% 16.61% 6.15% 3.96% 2.23% 0.75% 0.67% 0.26% 0.07% 3'-hydroxy-echinenone 0.369 -0.143 -0.020 -0.154 -0.670 -0.446 -0.273 0.299 -0.084 adonirubin 0.366 -0.026 0.442 0.246 0.250 -0.270 -0.434 -0.533 0.045 echinenone 0.357 -0.232 -0.358 -0.270 -0.171 0.087 0.407 -0.632 -0.122 canthaxanthin 0.340 -0.317 -0.257 -0.359 0.340 0.377 -0.444 0.225 0.289 zeaxanthin 0.335 0.363 -0.085 -0.292 0.513 -0.414 0.224 0.245 -0.345 lutein 0.310 0.485 -0.015 0.185 -0.204 0.574 -0.215 0.012 -0.465 β-isocryptoxanthin 0.308 -0.286 -0.355 0.766 0.135 -0.066 0.189 0.228 0.000 astaxanthin 0.305 -0.300 0.690 -0.073 -0.019 0.272 0.443 0.253 -0.015 3'-dehydrolutein 0.300 0.534 -0.023 0.041 -0.141 0.006 0.212 0.001 0.747

NETWORK H: adonixanthin, β-isocryptoxanthin and β-cryptoxanthin absent (n = 10) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 λ = 4.893 λ = 2.933 λ = 0.670 λ = 0.258 λ = 0.142 λ = 0.074 λ = 0.027 λ = 0.002 λ = 0.002 54.37% 32.59% 7.45% 2.86% 1.58% 0.82% 0.30% 0.03% 0.02% adonirubin 0.431 0.136 -0.035 0.295 -0.285 -0.006 -0.244 -0.610 -0.441 zeaxanthin 0.416 0.138 0.181 -0.386 0.130 -0.656 0.391 0.031 -0.159 canthaxanthin 0.404 -0.199 -0.035 0.374 -0.557 0.050 0.396 0.300 0.311 astaxanthin 0.394 0.072 -0.440 0.404 0.601 -0.123 -0.201 0.208 0.149 3'-hydroxy-echinenone 0.349 -0.304 0.400 -0.219 -0.089 0.016 -0.653 0.377 -0.035 echinenone 0.343 -0.339 0.206 -0.159 0.406 0.596 0.359 -0.216 -0.064 3'-dehydrolutein 0.253 0.469 -0.048 -0.370 -0.099 0.177 -0.158 -0.265 0.664 3'-dehydrolutein 0.125 0.547 -0.183 -0.172 -0.109 0.406 0.083 0.480 -0.456 β-carotene -0.066 0.437 0.730 0.470 0.191 0.028 0.056 0.060 0.089

NETWORK I: adonixanthin, β-isocryptoxanthin, β-cryptoxanthin and β-carotene absent (n = 5) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 λ = 7.618 λ = 0.329 λ = 0.051 λ = 0.002 λ = 0.000 λ = 0.000 λ = 0.000 λ = 0.000 95.23% 4.11% 0.63% 0.03% 0.00% 0.00% 0.00% 0.00% echinenone 0.359 -0.213 -0.231 0.235 -0.248 -0.810 0.000 0.000 adonirubin 0.358 0.238 -0.244 0.472 -0.067 0.323 0.038 -0.650 3'-hydroxy-echinenone 0.357 -0.217 -0.548 -0.158 -0.248 0.402 -0.325 0.415 lutein 0.357 -0.118 0.723 0.217 -0.182 0.101 -0.489 0.080 canthaxanthin 0.355 -0.324 0.244 -0.212 -0.269 0.194 0.744 0.000 3'-dehydrolutein 0.355 0.344 0.040 -0.755 0.017 -0.169 -0.198 -0.340 astaxanthin 0.355 -0.359 -0.023 0.021 0.863 0.000 0.000 0.000 zeaxanthin 0.332 0.696 0.046 0.184 0.150 -0.041 0.247 0.533 107

NETWORK J: astaxanthin absent (n = 8) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 Prin11 λ = 7.94 λ = 1.482 λ = 0.897 λ = 0.415 λ = 0.116 λ = 0.099 λ = 0.052 λ = 0.000 λ = 0.000 λ = 0.000 λ = 0.000 72.18% 13.47% 8.16% 3.77% 1.05% 0.90% 0.47% 0.00% 0.00% 0.00% 0.00% zeaxanthin 0.352 -0.040 -0.083 -0.038 -0.057 0.135 0.336 -0.856 0.000 0.000 0.000 3'-dehydrolutein 0.347 -0.045 -0.155 0.005 -0.078 -0.203 -0.571 -0.092 0.065 -0.584 0.355 β-cryptoxanthin 0.336 -0.204 0.133 0.156 -0.037 -0.050 -0.520 -0.081 -0.016 0.723 0.000 lutein 0.333 -0.142 -0.295 0.118 0.038 -0.208 0.237 0.224 -0.785 0.000 0.000 3'-hydroxy-echinenone 0.328 -0.054 0.340 0.041 0.105 -0.591 0.147 0.061 0.251 -0.157 -0.547 echinenone 0.321 0.178 0.139 -0.491 -0.306 -0.180 0.315 0.247 0.176 0.212 0.493 adonixanthin 0.299 -0.129 0.486 0.090 -0.403 0.582 0.000 0.197 -0.138 -0.245 -0.178 canthaxanthin 0.287 0.269 -0.280 -0.593 0.321 0.308 -0.214 0.102 -0.013 0.017 -0.401 adonirubin 0.265 -0.502 -0.147 0.170 0.498 0.260 0.244 0.242 0.382 -0.021 0.219 β-carotene 0.227 0.420 -0.492 0.484 -0.346 0.076 0.092 0.171 0.317 0.079 -0.158 β-isocryptoxanthin 0.170 0.618 0.388 0.307 0.501 0.076 0.014 -0.026 -0.139 -0.007 0.259

NETWORK K: astaxanthin and β-cryptoxanthin absent (n = 13) Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 Prin8 Prin9 Prin10 λ = 8.983 λ = 0.475 λ = 0.249 λ = 0.189 λ = 0.060 λ = 0.026 λ = 0.008 λ = 0.006 λ = 0.003 λ = 0.000 89.83% 4.75% 2.49% 1.89% 0.60% 0.26% 0.08% 0.06% 0.03% 0.00% adonirubin 0.330 0.074 -0.139 -0.144 0.210 -0.392 -0.343 -0.699 0.064 0.195 echinenone 0.327 0.196 0.160 -0.222 -0.111 0.313 -0.009 -0.193 -0.677 -0.416 canthaxanthin 0.324 0.180 0.242 -0.330 -0.158 0.367 -0.493 0.235 0.486 0.037 zeaxanthin 0.323 -0.164 -0.367 -0.118 -0.293 0.409 0.537 -0.251 0.188 0.284 3'-hydroxy-echinenone 0.320 0.250 0.236 -0.388 -0.034 -0.559 0.478 0.283 0.074 -0.008 3'-dehydrolutein 0.319 -0.338 -0.215 0.274 -0.279 -0.240 -0.074 0.024 0.260 -0.674 lutein 0.316 -0.393 0.185 0.219 -0.383 -0.182 -0.228 0.227 -0.374 0.490 adonixanthin 0.313 0.063 -0.635 -0.043 0.460 0.038 -0.134 0.470 -0.191 0.055 β-carotene 0.303 -0.441 0.461 0.139 0.628 0.193 0.191 -0.039 0.101 -0.038 β-isocryptoxanthin 0.284 0.606 0.099 0.716 -0.003 0.064 0.102 -0.023 0.082 0.088 108

APPENDIX D. RETENTION AND RECOMBINATION OF BIOCHEMICAL MODULES IN THE EVOLUTION OF AVIAN CAROTENOID METABOLISM

109

Abstract

Historical associations of genes and proteins are thought to delineate which pathways are available to their subsequent evolution, however the extent to which past functional involvements bias contemporary evolution are rarely quantified. Here were examine the extent to which the structure of an ancient biochemical network of carotenoid metabolism biases subsequent diversification of carotenoid networks in avian evolution. Specifically, we tested whether the evolution of avian carotenoid networks is most concordant with phylogenetically structured expansion from core biochemical reactions of common ancestors or with subsampling from the full ancestral network. We compared structural and phylogenetic associations in 467 carotenoid networks of extant and ancestral species to uncover the overwhelming effect of preexisting biochemical structure on avian carotenoid diversification over the last 50 million years. Birds repeatedly subsampled the full ancestral network during evolution, expressing and recombining at each evolutionary step entire network motifs (biochemical modules of compounds connected by short enzymatic pathways). Rapid and reversible switches between modules likely assured access to the full biochemical network during evolution. These findings corroborate the observation that ancestral biochemical modules have an important role in current adaptations and account for the lack of phylogenetic signal in avian carotenoid evolution and periodic convergence of evolutionary distant species in carotenoid metabolism.

110

Introduction

How far can the associations between genes and proteins diverge from each other to produce phenotypic divergence? Ancestral relationships between the elements underlying a phenotype delineate a limited number of evolutionary steps from one functional phenotype to another as a result of fitness consequences [1-3] and genomic or developmental epistasis [4-6]. The contingency of subsequent evolutionary trajectories on the structure of an ancestral phenotype is particularly evident in the evolution of metabolic networks, because divergence of homologous metabolic networks preserves the existing structure of the ancestral network. Structural properties of a metabolic network scale with its expansion- compounds associated with the most enzymatic reactions in the network are most likely to gain new reactions [7-9], and the influence of the original network structure is reflected in the occurrence of conserved pathway modules in subsequent diversifications [10, 11].

The diversification of carotenoid metabolism is an example of how the spectacular divergence in the production of different compounds builds on the existing structure of ancient enzymatic reactions.

Many core carotenoid metabolic pathways are highly conserved across prokaryotes and eukaryotes [12,

13], and homologous amino acid sequences and enzymes occur in distantly related bacterium and higher level plants [14, 15]. Carotenoid metabolism is hierarchical; species-specific branching pathways expand outwards from a few conserved basal reactions involved in the initial assembly of carotenoid molecules from precursors [16]. These findings establish the ancient origins of carotenoid metabolic pathways that are now integral in numerous biological functions, including photosynthesis, immunity, pigmentation, and vision [17, 18].

Almost all animals, however, are missing the basal enzyme that synthesizes carotenoids from a common precursor [19, 20], and yet they still have enzymatic pathways that can further modify carotenoids they acquire from their diet [21]. Given that animals do not all start carotenoid metabolism from the same substrates, to what extent do the historical associations between compounds and enzymatic reactions determine the contemporary diversification of carotenoid metabolism? Metabolic networks are comprised of pathways of interconnected enzymatic reactions that represent functionally independent 111 modules, and differences in the subsampling of pathway modules originating from the same ancestral network of carotenoid metabolism could drive divergence between species [22-24]. The network structure within each module would be conserved, but the combination of which modules were expressed would vary across species as a result of differences in the regulation of enzyme activities of individual modules

[25], or the acquisition of different dietary compounds [26]. Alternatively, diversification of carotenoid metabolism could be due to the evolutionary gain or loss of enzymatic reactions that build on the existing structure of an ancestral network [27, 28]. This could be caused by the addition of novel reactions to the end of pathways [29], or by the branching of pathways within pre-existing modules as the result of gene duplication and subfunctionalization [30, 31]. Each of these modes of diversification would have distinct effects on the tempo of evolutionary changes in metabolism. Subsampling modules from the same ancestral network would allow for widespread and rapid shifts in metabolism as opposed to slower, lineage-specific metabolic changes that would occur with the evolutionary expansion of functional modules.

Here we tested whether the evolutionary diversification of carotenoid compounds in the ornamental plumage of birds is determined by a conserved ancestral network, or if the structure of the network itself evolves with the diversification of birds. Using carotenoid networks present in 250 species from nine avian orders and 217 ancestral networks spanning 50 MYA of avian evolution, we tested if the frequency of compound co-occurrence in species changed across evolutionary time periods in relation to the pathway lengths between compounds. If the metabolic networks of all taxa are derived from subsampling modules on a conserved ancestral network structure, we would expect pairs of compounds connected by short pathways to co-occur more frequently than pairs of compounds separated by longer pathways in both ancestral networks and extant species (figure 1a). The expansion of enzymatic pathways within functional modules between ancestral networks and descendant species [32], however, would erase the influence of a common network structure. As new pathways are added to old ones, compounds separated by longer pathways would co-occur in descendant species (figure 1b). The findings of our study establish both the mode and limits of evolutionary diversification in avian carotenoid metabolism. 112

2. Material and methods

Individual carotenoid metabolic networks were constructed for 250 species of birds (appendix S1) based on established enzymatic reactions connecting the compounds identified in their plumage (see [26, 33] for a details). An ultrametric 50% majority-rule consensus tree of 243 species from 1,000 trees randomly sampled from the pseudo-posterior distribution of the Hackett All Species supertree [34] was constructed using SumTrees (version 4.1.0) [35] in DendroPy (version 4.1.0) [36] (figures 1c and S1a-d, appendix

S3). Following a similar approach to ancestral network reconstruction in [37], the ultrametric consensus tree was used in maximum likelihood estimations for ancestral state reconstructions of each of the 55 compounds and 91 reactions present at least once in species’ networks [26, 33]. We tested two binary models of evolution for each individual compound and reaction in the program r8s (version 1.8) [38, 39]: the Binary-1 model assumed equal rates and the Binary-2 model assumed separate rates for the gain and loss of a compound or reaction. Each of the compounds and reactions were considered discrete (present or absent) and unordered. The reconstructed ancestral states in the model with the lower Akaike Information

Criterion (AIC) value [40] were retained for each compound and reaction. Ancestral networks were comprised of the compounds and reactions present at each of the 217 internal nodes in the phylogeny

(appendix S1).

The phylogenetic profile for each compound [41] represents the subset of species and ancestral networks that contain the compound. This is a binary string that encodes the presence (1) or absence (0) of a compound in each of the 467 species and ancestral networks. To assess evolutionary changes in carotenoid network structure, we grouped networks into different age classes based on the age of their ancestral node. The phylogenetic similarity between compounds measured how frequently pairs compounds co-occurred in carotenoid networks in each age class by calculating the Jaccard coefficient

(Jij) [42] between their phylogenetic profiles using the package vegan (version 2.4-1) [43] in R (version

3.2.2) [44]: 113

푛푖푗 퐽푖푗 = 푛푖 + 푛푗 − 푛푖푗

Where nij is the number of networks with both compounds i and j, ni is the number of networks with compound i, and nj is the number of networks with compound j. Pairs of compounds with similar phylogenetic profiles (Jij = 1) are evolutionary conserved modules that coevolve in ancestral and species networks, while pairs of phylogenetically distinct compounds (Jij = 0) evolve independently of each other and occur in different networks [45, 46]. The minimum number of reactions separating a pair of compounds was assigned based on known carotenoid enzymatic reactions in birds [26, 33]. Pairs of compounds that were not connected by reactions were excluded from the study. All statistical analyses were implemented in SAS v. 9.4.

3. Results and Discussion

In this study we examined whether the diversification of plumage-bound carotenoids in birds was caused by differences in the expression of conserved functional modules derived from the structure of an ancestral carotenoid metabolic network structure or by the evolutionary expansion of the carotenoid metabolic network structure within divergent lineages. We found that pairs of compounds located closer together in the ancestral network co-occurred more often than compounds separated by several reactions in extant species (figure 1b; bST = -0.37, n = 250) and in ancestral networks in all seven time periods representing 50 MYA of avian evolution on the phylogeny (figure 1c). These results demonstrate that the evolution of avian carotenoid metabolism is directly predictable from a historical metabolic network structure. While we found that there were a lineage-specific modules that may have been added to the ancestral network more recently, the widespread subsampling of different combinations of conserved pathway modules contributes the most to the divergence in plumage carotenoids (figure S1a-d, appendix

S2).

Our results establish that enzymatic pathway modules in the avian carotenoid network are comprised of groups of closely connected compounds separated by only one or two reactions that always 114 occur together across the evolutionary history of carotenoid metabolism in birds. Furthermore, based on our finding that species express different combinations of modules, each of the pathway modules are functionally independent. Some of these modules are associated with specific dietary compounds, and are thus dependent on what species consume (figure S1e, appendix S2). Other modules, however, are comprised entirely of metabolically derived compounds at the end of pathways (figure S1e, appendix S2).

Due to the structural redundancy of pathways in the ancestral network [12], these derived modules can be accessed from different dietary compounds when there are shifts in diet. This may explain why the two most common dietary compounds, lutein and zeaxanthin, are not associated with a module. The occurrence of rate-limiting reactions that control the activity of multiple, sequential reactions [47, 48] or enzyme complexes that catalyze sequential reactions [49] would form groups of reactions that are always used together in metabolism. The modular structure and functional properties of the ancestral carotenoid metabolic network thus allows birds to exploit a wide range of ecological niches that vary in dietary compounds [50].

For subsampling of pathway modules to occur the structure of the entire ancestral network needs to remain intact across species, even when parts are not utilized for the production of a species’ plumage carotenoids. Rapid and reversible switches between pre-existing modules could be achieved if the enzymes in the ancestral network are non-specific, such that a single enzyme catalyzes multiple unique reactions with distinct substrates [51-53]. The same enzyme could therefore be affiliated with multiple pathway modules in the ancestral network, and so changes in which dietary compounds are acquired would lead to shifts in which of the modules are expressed without the loss of modules associated with absent dietary compounds. Ancestral modules associated with unique enzymes could still persist when not in use due to rapid and reversible changes in the activity of enzymes that control metabolic flux through the network. Differences in hormone levels and the epigenetic or genetic mechanisms that regulate metabolic flux could cause evolutionary changes in the activation of an enzyme without affecting its functional properties [54-58]. Alternatively, the structure of the ancestral network is retained across the evolutionary history of birds, because all of the pathway modules in the full network are utilized in the 115 production of carotenoids for biological functions unrelated to plumage coloration such as vision and the immune system [59, 60]. The carotenoid metabolism of species would vary based on which of these modules they co-opt for use in plumage coloration.

The finding that different combinations of pathway modules in the ancestral carotenoid network underlie the diversification of carotenoids across species demonstrates that opportunities for diversification can exceed the existing network structure. The selective expression of modules in the ancestral carotenoid metabolic network determines where opportunities for diversification can occur in avian carotenoid metabolism. Compounds that are connected to multiple pathway modules via enzymatic reactions would be expected to be more conserved over evolutionary history and across species networks than compounds that only have within-module connections [10]. Examination of patterns of carotenoid diversification across species’ networks confirms that compounds with the most enzymatic reactions act as hotspots of diversification in the ancestral network, because these compounds are conserved while the modules they participate in vary across species [26]. This allows closely related species to express distinct modules (figure S1a-d). The periodic convergence of carotenoid metabolism observed in distantly related taxa [12], however, establishes that the dependence on an ancestral network structure does place limits on avian carotenoid diversification. There are only so many ways to recombine the pathway modules, and distantly related species that consume the same dietary compounds have access to the same modules.

The results of this study explain the mechanisms behind the rapid diversification as well as the lack of a phylogenetic signal in avian carotenoid coloration. Our findings establish the important role that ancient pathways continue to play in contemporary adaptations.

116

References

1. Povolotskaya IS, Kondrashov FA. 2010 Sequence space and the ongoing expansion of the protein

universe. Nature 465, 922-927.

2. Harms MJ, Thornton JW. 2014 Historical contingency and its biophysical basis in glucocorticoid

receptor evolution. Nature 512, 203-207.

3. Maynard Smith J. 1970 Natural selection and the concept of a protein space. Nature 225, 563-564.

4. Gavrilets S. 2004 Fitness Landscapes and the Origin of Species. Princeton, NJ, USA: Princeton

University Press.

5. Poelwijk FJ, Kiviet DJ, Weinreich DM, Tans SJ. 2007 Empirical fitness landscapes reveal accessible

evolutionary paths. Nature 445, 383-386.

6. Bershtein S, Segal M, Bekerman R, Tokuriki N, Tawfik DS. 2006 Robustness-epistasis link shapes

the fitness landscape of a randomly drifting protein. Nature 444, 929-932.

7. Barabási A-L, Albert R. 1999 Emergence of scaling in random networks. Science 286, 509-512.

8. Light S, Kraulis P, Elofsson A. 2005 Preferential attachment in the evolution of metabolic networks.

BMC Genomics 6, 159.

9. Jeong H, Tombor B, Albert R, Oltvai ZN, Barabási A-L. 2000 The large-scale organization of

metabolic networks. Nature 407, 651-654.

10. Guimerà R, Amaral LAN. 2005 Functional cartography of complex metabolic networks. Nature 433,

895-900.

11. Ravasz E, Somera AL, Mongru DA, Oltvai ZN, Barabási A-L. 2002 Hierarchical organization of

modularity in metabolic networks. Science 297, 1551-1555.

12. Badyaev AV, Morrison ES, Belloni V, Sanderson MJ. 2015 Tradeoff between robustness and

elaboration in carotenoid networks produces cycles of avian color diversification. Biol. Direct 10, 45.

13. Sandmann G. 2002 Molecular evolution of carotenoid biosynthesis from bacteria to plants. Physiol.

Plant. 116, 431-440. 117

14. Armstrong GA, Alberti M, Hearst JE. 1990 Conserved enzymes mediate the early reactions of

carotenoid biosynthesis in nonphotosynthetic and photosynthetic prokaryotes. Proc. Natl. Acad. Sci

USA 87, 9975-9979.

15. Young PR, Lashbrooke JG, Alexandersson E, Jacobson D, Moser C, Velasco R, Vivier MA. 2012

The genes and enzymes of the carotenoid metabolic pathway in Vitis vinifera L. BMC Genomics 13,

243.

16. Umeno D, Tobias AV, Arnold FH. 2005 Diversifying carotenoid biosynthetic pathways by directed

evolution. Microbiol. Mol. Biol. Rev. 69, 51-78.

17. Goodwin TW. 1980 The Biochemistry of the Carotenoids, Vol. I Plants. London, UK: Chapman and

Hall.

18. Goodwin TW. 1984 The Biochemistry of the Carotenoids, Vol. II Animals. London, UK: Champan

and Hall.

19. Moran NA, Jarvik T. 2010 Lateral transfer of genes from fungi underlies carotenoid production in

aphids. Science 328, 624-627.

20. Goodwin TW. 1986 Metabolism, nutrition, and function of carotenoids. Annu. Rev. Nutr. 6, 273-297.

21. Davies BH. 1985 Carotenoid metabolism in animals: A biochemist's view. Pure Appl. Chem. 57, 679-

684.

22. Zhang Y, Li S, Skogerbo G, Zhang Z, Zhu X, Zhang Z, Sun S, Lu H, Shi B, Chen R. 2006

Phylophenetic properties of metabolic pathway topologies as revealed by global analysis. BMC

Bioinformatics 7, 252.

23. Peregrín-Alvarez JM, Tsoka S, Ouzounis CA. 2003 The phylogenetic extent of metabolic enzymes

and pathways. Genome Res. 13, 422-427.

24. Peregrín-Alvarez JM, Sanford C, Parkinson J. 2009 The conservation and evolutionary modularity of

metabolism. Genome Biol. 10, R63. 118

25. Lavington E, Cogni R, Kuczynski C, Koury S, Behrman EL, O’Brien KR, Schmidt PS, Eanes WF.

2014 A small system—high-resolution study of metabolic adaptation in the central metabolic

pathway to temperate climates in Drosophila melanogaster. Mol. Biol. Evol. 31, 2032-2041.

26. Morrison ES, Badyaev AV. 2016 Structuring evolution: Biochemical networks and metabolic

diversification in birds. BMC Evol. Biol. 16, 168.

27. Klassen JL. 2010 Phylogenetic and evolutionary patterns in microbial carotenoid biosynthesis are

revealed by comparative genomics. PLoS ONE 5, e11257.

28. Tanaka T, Ikeo K, Gojobori T. 2006 Evolution of metabolic networks by gain and loss of enzymatic

reaction in eukaryotes. Gene 365, 88-94.

29. Granick S. 1965 Evolution of heme and chlorophyll. In Evolving Genes and Proteins (eds V Bryson,

HJ Vogel). New York, NY, USA: Academic Press. pp. 67-88.

30. Jensen RA. 1976 Enzyme recruitment in evolution of new function. Annu. Rev. Microbiol. 30, 409-

425.

31. Horowitz NH. 1945 On the evolution of biochemical syntheses. Proc. Natl. Acad. Sci USA 31, 153-

157.

32. Ebenhöh O, Handorf T, Heinrich R. 2004 Structural analysis of expanding metabolic networks.

Genome Inform. 15, 35-45.

33. Badyaev AV, Morrison ES, Belloni V, Sanderson MJ. 2015 Tradeoff between robustness and

elaboration in carotenoid networks produces cycles of avian color diversification. Biol. Direct 10, 45.

34. Jetz W, Thomas GH, Joy JB., Redding DW, Hartmann K, Mooers AO. 2014 Global distribution and

conservation of evolutionary distinctness in birds. Curr. Biol. 24, 919-930.

35. Sukumaran J, Holder MT. 2016 SumTrees: Phylogenetic tree summarization 4.1.0. Available at

https://github.com/jeetsukumaran/DendroPy.

36. Sukumaran J, Holder MT 2010 DendroPy: A Python library for phylogenetic computing.

Bioinformatics 26, 1569-1571. 119

37. Ebenhöh O, Handorf T, Kahn D. 2006 Evolutionary changes of metabolic networks and their

biosynthetic capacities. IEE Proc.-Syst. Biol. 153, 354-358.

38. Sanderson MJ. 2003 r8s: inferring absolute rates of molecular evolution and divergence times in the

absence of a molecular clock. Bioinformatics 19, 301-302.

39. Marazzi B, Ané C, Simon MF, Delgado-Salinas A, Luckow M, Sanderson MJ. 2012 Locating

evolutionary precursors on a phylogenetic tree. Evolution 66, 3918-3930.

40. Akaike H. 1974 A new look at the statistical model identification. IEEE Trans. Automat. Contr. 19,

716-723.

41. Pellegrini M, Marcotte EM, Thompson MJ, Eisenberg D, Yeates TO. 1999 Assigning protein

functions by comparative genome analysis: Protein phylogenetic profiles. Proc. Natl. Acad. Sci USA

96, 4285-4288.

42. Jaccard P. 1912 The distribution of the flora in the alpine zone. New Phytol. 11, 37-50.

43. Oksanen J, Guillaume Blanchet F, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR,

O'Hara, RB, Simpson GL, Solymos P, et al. 2016 vegan: Community ecology package. Version 2.4-1.

Available at https://CRAN.R-project.org/package=vegan.

44. R Development Core Team. 2016 R: A language and environment for statstical computing. Version

3.2.2. Vienna, Austria: R Foundation for Statistical Computing. Available at http://www.R-

project.org.

45. Yamada T, Kanehisa M, Goto S. 2006 Extraction of phylogenetic network modules from the

metabolic network. BMC Bioinformatics 7, 130.

46. Zhao J, Ding G-H, Tao L, Yu H, Yu Z-H, Luo J-H, Cao Z-W, Li Y-X. 2007 Modular co-evolution of

metabolic networks. BMC Bioinformatics 8, 311.

47. Krebs HA. 1957 Control of metabolic processes. Endeavour 16, 125-132.

48. Fell DA. 1992 Metabolic control analysis: a survey of its theoretical and experimental development.

Biochem. J. 286, 313-330.

49. Srere PA. 1987 Complexes of sequential metabolic enzymes. Annu. Rev. Biochem. 56, 89-124. 120

50. Kreimer A, Borenstein E, Gophna U, Ruppin E. 2008 The evolution of modularity in bacterial

metabolic networks. Proc. Natl. Acad. Sci USA 105, 6976-6981.

51. Nobeli I, Favia AD, Thornton JM. 2009 Protein promiscuity and its implications for biotechnology.

Nat Biotech 27, 157-167.

52. James LC, Tawfik DS. 2003 Conformational diversity and protein evolution – a 60-year-old

hypothesis revisited. Trends Biochem. Sci. 28, 361-368.

53. Khersonsky O, Roodveldt C,Tawfik DS. 2006 Enzyme promiscuity: evolutionary and mechanistic

aspects. Curr. Opin. Chem. Biol. 10, 498-508.

54. Ketterson, ED, Nolan V Jr. 1999 Adaptation, exaptation, and constraint: a hormonal perspective. The

Am. Nat. 154, S4-S25.

55. Cohen, P. 1988 Review lecture: protein phosphorylation and hormone action. Proc. R. Soc. B 234,

115-144.

56. Hau M. 2007 Regulation of male traits by testosterone: implications for the evolution of vertebrate

life histories. Bioessays 29, 133-144.

57. Varum S, Rodrigues AS, Moura MB, Momcilovic O, Easley CAIV, Ramalho-Santos J, Van Houten

B, Schatten G. 2011 Energy metabolism in human pluripotent stem cells and their differentiated

counterparts. PLoS ONE 6, e20914.

58. Gordon KL, Ruvinsky I. 2012 Tempo and mode in evolution of transcriptional regulation. PLoS

Genet. 8, e1002432.

59. Toomey MB, Collins AM, Frederiksen R, Cornwall MC, Timlin JA, Corbo JC. 2015 A complex

carotenoid palette tunes avian colour vision. J. R. Soc. Interface 12, 20150563.

60. Krinsky NI, Yeum K-J. 2003 Carotenoid–radical interactions. Biochem. Biophys. Res. Commun. 305,

754-760.

121

hylogenetic P similarity

Figure 1. The same ancestral network structure consistently influences the diversification of carotenoid metabolism throughout the evolutionary history of birds. (a) If compounds separated by fewer reactions in an ancestral network (structural distance) co-occur more often in the same species

(phylogenetic profile similarity) than compounds separated by longer pathways, then structure matters for the diversification of compounds across species. If there is no relationship between the phylogenetic similarity and the structural distance in the ancestral network between compounds, then pathway modules are expanding and the ancestral network structure has been erased. The same pathway modules are conserved across (b) extant species and (c) ancestral networks, because in all cases pairs of compounds co-occur more often the more closely they are connected to each other in the same network structure. (c)

The majority-rule consensus tree of 243 species of birds (figure S1, appendix S3). Each graph below the phylogeny corresponds to the range of ages of the ancestral networks included in each analysis. The graph on the far left, for example, corresponds to ancestral networks from 50-30 MYA. 122

Appendix S1: The species and ancestral reconstructed networks. Each column is either a compound or a reaction (substrate-product) that occurs at least once in extant species networks. A compound or reaction present in a network is represented by a "1", and a "0" denotes absent compounds or reactions. The seven species in italics were not included in the majority-rule consensus tree. The numbers of the ancestral networks correspond to the internal node number labeled on the phylogeny in fig. S1a-d, and age corresponds to the age of the internal node on the majority-rule consensus tree. Listed below each compound and reaction is the best fitting model for the evolutionary rates of its gain and loss. Binary-1 assumes equal rates and Binary-2 assumes distinct rates of gain and loss (see methods for details).

Appendix S1 123 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Aegithalos_caudatus 0000100000000000000 0000 Agelaius_phoeniceus 0111100011000000100 0010 Alectoris_rufa 0111000000001111000 0010 Amandava_amandava 0110110000000000100 0000 Amandava_subflava 0110110000000000100 0000 Ampelioides_tschudii 0100000000000000000 0000 Ampelion_rufaxilla 0100000000000000000 0000 Anas_platyrhynchos 0111100000000000100 0000 Anser_anser 0001100000000000000 0000 Antilophia_galeata 0110000011000000111 0000 Apaloderma_narina 0111100000001000000 0010 Bombycilla_cedrorum 0111100011000000101 0010 Bombycilla_garrulus 0110000011000000100 0010 Bombycilla_japonica 0110000011000000100 0010 Bucanetes_githagineus 0111100000001000000 0010 Cacicus_cela 0100000000000000000 0000 Cacicus_haemorrhous 0111100011000000100 0010 Cacicus_leucoramphus 0110000000000000000 0000 Cacicus_melanicterus 0110000000000000000 0000 Cacicus_uropygialis 0101100011000000100 0000 Calyptomena_viridis 0110000000000000000 1100 Campephilus_leucopogon 0111100000001000000 0010 Cardinalis_cardinalis 0111100011001000101 0010 Cardinalis_sinuatus 0111100000001000000 0010 Carduelis_atrata 0100000011000000100 0000 Carduelis_cannabina 0111100000000000000 0010 Carduelis_carduelis 0110000011000000101 0010 Carduelis_chloris 0110000011000000100 0000 Carduelis_cucullata 0111100000001000000 0010 Carduelis_flammea 0011100000000000000 0010 Carduelis_hornemanni 0111100000001000000 0010 Carduelis_sinica 0100000011000000100 0000 Carduelis_spinoides 0100000011000000100 0000 Carduelis_spinus 0100000011000000100 0000 Carduelis_tristis 0111100011000000100 0000 Carpodacus_mexicanus 0111100011001000100 0010 Carpodacus_nipalensis 0110100011000000100 0010 Carpodacus_pulcherrimus 0110100000001111000 0010 Carpodacus_roseus 0110100000001111000 0010 Carpodacus_rubicilloides 0000100000000000000 0000 Carpodacus_thura 0101100000001000000 0000 Carpodacus_trifasciatus 0010100000000000000 0010 Carpornis_cucullata 0100000000000000000 0000 Chiroxiphia_caudata 0110000011000000111 0000 Chiroxiphia_pareola 0110000011000000111 0000 Chlorospingus_pileatus 0100000000000000000 0000 Ciconia_ciconia 0100000000000000000 0000 Coccothraustes_abeillei 0100000000000000000 0000 Coccothraustes_vespertinus0100000000000000000 0000 Coereba_flaveola 0100000011000000100 0000 Colaptes_auratus 0111101100001000000 0010 Colaptes_auratus_cafer 0111100000001000000 0010 Colaptes_chrysoides 0110100000001000000 0010 Colaptes_campestris 0110100000000000000 0000 Colaptes_melanochloros 0110101100001000000 0010 Cotinga_amabilis 0010000000000000000 0010 Cotinga_cotinga 0010000000000000000 0010 Cotinga_maculata 0010000000000000000 0010 Parus_caeruleus 0110000000000000000 0000 Cymbirhynchus_macrorhynchos0100000000001001000 0000 Dendrocopos_major 0110100000001000000 0010 Dendroica_coronata 0100000000000000000 0000 Dendroica_palmarum 0100000000000000000 0000 Appendix S1 124 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) Aegithalos_caudatus 0 0000010000000000000000 Agelaius_phoeniceus 0 0111111110100000000000 Alectoris_rufa 0 0111101110100000000100 Amandava_amandava 0 0000000000000000000000 Amandava_subflava 0 0000000000000000000000 Ampelioides_tschudii 0 0000000000000000000000 Ampelion_rufaxilla 0 0000000000000000000000 Anas_platyrhynchos 0 0000101000010000000000 Anser_anser 0 0001000000000000000000 Antilophia_galeata 0 0000000000000000000001 Apaloderma_narina 0 0111111110100000000100 Bombycilla_cedrorum 0 0111010000000000000000 Bombycilla_garrulus 0 0111000000000000000000 Bombycilla_japonica 0 0111000000000000000000 Bucanetes_githagineus 0 0111111110100000000100 Cacicus_cela 0 0000000000000000000000 Cacicus_haemorrhous 0 0111111110100000000000 Cacicus_leucoramphus 0 0000000000000000000000 Cacicus_melanicterus 0 0000000000000000000000 Cacicus_uropygialis 0 0000111110100000000000 Calyptomena_viridis 0 0000000000000000000000 Campephilus_leucopogon 0 0111111110100000000100 Cardinalis_cardinalis 0 0111111110100000000100 Cardinalis_sinuatus 0 0111111110100000000100 Carduelis_atrata 0 0000000000000000000000 Carduelis_cannabina 0 0111111110100001111000 Carduelis_carduelis 0 0111000000000000000000 Carduelis_chloris 0 0000000000000000000000 Carduelis_cucullata 0 0111111110100000000100 Carduelis_flammea 0 0111111110100001111000 Carduelis_hornemanni 0 0111111110100001100100 Carduelis_sinica 0 0000000000000000000000 Carduelis_spinoides 0 0000000000000000000000 Carduelis_spinus 0 0000000000000000000000 Carduelis_tristis 0 0000000000000000000000 Carpodacus_mexicanus 0 0111111111110001110100 Carpodacus_nipalensis 0 0111010100000000000000 Carpodacus_pulcherrimus 0 0111010100000001100100 Carpodacus_roseus 0 0111010100000001111100 Carpodacus_rubicilloides 0 0000010000000001100000 Carpodacus_thura 0 0000111110100001100100 Carpodacus_trifasciatus 0 0111010100000001100000 Carpornis_cucullata 0 0000000000000000000000 Chiroxiphia_caudata 0 0000000000000000000001 Chiroxiphia_pareola 0 0000000000000000000001 Chlorospingus_pileatus 0 0000000000000000000000 Ciconia_ciconia 0 0001000000000000000000 Coccothraustes_abeillei 0 0000000000000000000000 Coccothraustes_vespertinus 0 0000000000000000000000 Coereba_flaveola 0 0000000000000000000000 Colaptes_auratus 0 0111111110100000000100 Colaptes_auratus_cafer 0 0111111110100000000100 Colaptes_chrysoides 0 0111010100000000000100 Colaptes_campestris 0 0000000000000000000000 Colaptes_melanochloros 0 0111010100000000000100 Cotinga_amabilis 0 0111000000000000000000 Cotinga_cotinga 0 0111000000000000000000 Cotinga_maculata 0 0111000000000000000000 Parus_caeruleus 0 0000000000000000000000 Cymbirhynchus_macrorhynchos 0 0000000000000000000100 Dendrocopos_major 0 0111010100000000000100 Dendroica_coronata 0 0000000000000000000000 Dendroica_palmarum 0 0000000000000000000000 Appendix S1 125 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) Aegithalos_caudatus 0 000 00000 0 0000000000000 Agelaius_phoeniceus 0 000 00000 0 0001111111000 Alectoris_rufa 0 000 00000 0 0000000000000 Amandava_amandava 0 000 00000 0 0010000110000 Amandava_subflava 0 000 00000 0 0010000110000 Ampelioides_tschudii 0 000 00000 0 0000000000000 Ampelion_rufaxilla 0 000 00000 0 0000000000000 Anas_platyrhynchos 0 000 00000 0 0000000110000 Anser_anser 0 000 00000 0 0000000000000 Antilophia_galeata 0 000 00000 0 0001111101000 Apaloderma_narina 0 000 00000 0 0000000000000 Bombycilla_cedrorum 0 000 00000 0 0001111111000 Bombycilla_garrulus 0 000 00000 0 0001111111000 Bombycilla_japonica 0 000 00000 0 0001111111000 Bucanetes_githagineus 0 000 00000 0 0000000000000 Cacicus_cela 0 000 00000 0 0000000000000 Cacicus_haemorrhous 0 000 00000 0 0001111111000 Cacicus_leucoramphus 0 000 00000 0 0000000000000 Cacicus_melanicterus 0 000 00000 0 0000000000000 Cacicus_uropygialis 0 000 00000 0 0001111101000 Calyptomena_viridis 0 000 00000 0 0000000000000 Campephilus_leucopogon 0 000 00000 0 0000000000000 Cardinalis_cardinalis 0 000 00000 0 0001111111000 Cardinalis_sinuatus 0 000 00000 0 0000000000000 Carduelis_atrata 0 000 00000 0 0001111101000 Carduelis_cannabina 0 000 00000 0 0000000000000 Carduelis_carduelis 0 000 00000 0 0001111111000 Carduelis_chloris 0 000 00000 0 0001111111000 Carduelis_cucullata 0 000 00000 0 0000000000000 Carduelis_flammea 0 000 00000 0 0000000000000 Carduelis_hornemanni 0 000 00000 0 0000000000000 Carduelis_sinica 0 000 00000 0 0001111101000 Carduelis_spinoides 0 000 00000 0 0001111101000 Carduelis_spinus 0 000 00000 0 0001111101000 Carduelis_tristis 0 000 00000 0 0001111111000 Carpodacus_mexicanus 0 000 00000 0 0001111111000 Carpodacus_nipalensis 0 000 00000 0 0001111111000 Carpodacus_pulcherrimus 0 000 00000 0 0000000000000 Carpodacus_roseus 0 000 00000 0 0000000000000 Carpodacus_rubicilloides 0 000 00000 0 0000000000000 Carpodacus_thura 0 000 00000 0 0000000000000 Carpodacus_trifasciatus 0 000 00000 0 0000000000000 Carpornis_cucullata 0 000 00000 0 0000000000000 Chiroxiphia_caudata 0 000 00000 0 0001111111000 Chiroxiphia_pareola 0 000 00000 0 0001111101000 Chlorospingus_pileatus 0 000 00000 0 0000000000000 Ciconia_ciconia 0 000 00000 0 0000000000000 Coccothraustes_abeillei 0 000 00000 0 0000000000000 Coccothraustes_vespertinus 0 000 00000 0 0000000000000 Coereba_flaveola 0 000 00000 0 0001111101000 Colaptes_auratus 0 000 00000 0 0000000000110 Colaptes_auratus_cafer 0 000 00000 0 0000000000000 Colaptes_chrysoides 0 000 00000 0 0000000000000 Colaptes_campestris 0 000 00000 0 0000000000000 Colaptes_melanochloros 0 000 00000 0 0000000000110 Cotinga_amabilis 0 000 11110 0 0000000000000 Cotinga_cotinga 0 000 11110 0 0000000000000 Cotinga_maculata 0 000 11110 0 0000000000000 Parus_caeruleus 0 000 00000 0 0000000000000 Cymbirhynchus_macrorhynchos 0 000 00000 1 0000000000000 Dendrocopos_major 0 000 00000 0 0000000000000 Dendroica_coronata 0 000 00000 0 0000000000000 Dendroica_palmarum 0 000 00000 0 0000000000000 Appendix S1 126 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Aegithalos_caudatus 0 00000000000000000000000 Agelaius_phoeniceus 0 00000000000000001000011 Alectoris_rufa 0 01111111000000000000011 Amandava_amandava 0 00000000000000001000000 Amandava_subflava 0 00000000000000001000000 Ampelioides_tschudii 0 00000000000000000000000 Ampelion_rufaxilla 0 00000000000000000000000 Anas_platyrhynchos 0 00000000000000001000000 Anser_anser 0 00000000000000000000000 Antilophia_galeata 0 00000000000000001111100 Apaloderma_narina 0 01001001100000000000011 Bombycilla_cedrorum 0 00000000000000001000011 Bombycilla_garrulus 0 00000000000000001000011 Bombycilla_japonica 0 00000000000000001000011 Bucanetes_githagineus 0 01001000100000000000011 Cacicus_cela 0 00000000000000000000000 Cacicus_haemorrhous 0 00000000000000001000011 Cacicus_leucoramphus 0 00000000000000000000000 Cacicus_melanicterus 0 00000000000000000000000 Cacicus_uropygialis 0 00000000000000000000000 Calyptomena_viridis 0 00000000000000110000000 Campephilus_leucopogon 0 01001000100000000000011 Cardinalis_cardinalis 0 01001000100000001000011 Cardinalis_sinuatus 0 01001000100000000000011 Carduelis_atrata 0 00000000000000000000000 Carduelis_cannabina 0 00000000000000000000011 Carduelis_carduelis 0 00000000000000001000011 Carduelis_chloris 0 00000000000000001000000 Carduelis_cucullata 0 01001000100000000000011 Carduelis_flammea 0 00000000000000000000011 Carduelis_hornemanni 0 01001000100000000000011 Carduelis_sinica 0 00000000000000000000000 Carduelis_spinoides 0 00000000000000000000000 Carduelis_spinus 0 00000000000000000000000 Carduelis_tristis 0 00000000000000001000000 Carpodacus_mexicanus 0 01001000100000001000011 Carpodacus_nipalensis 0 00000000000000001000011 Carpodacus_pulcherrimus 0 01111111100000000000011 Carpodacus_roseus 0 01111111100000000000011 Carpodacus_rubicilloides 0 00000000000000000000000 Carpodacus_thura 0 01001000000000000000000 Carpodacus_trifasciatus 0 00000000000000000000011 Carpornis_cucullata 0 00000000000000000000000 Chiroxiphia_caudata 0 00000000000000001111100 Chiroxiphia_pareola 0 00000000000000001111100 Chlorospingus_pileatus 0 00000000000000000000000 Ciconia_ciconia 0 00000000000000000000000 Coccothraustes_abeillei 0 00000000000000000000000 Coccothraustes_vespertinus 0 00000000000000000000000 Coereba_flaveola 0 00000000000000000000000 Colaptes_auratus 0 01001000100000000000011 Colaptes_auratus_cafer 0 01001000100000000000011 Colaptes_chrysoides 0 01001000100000000000011 Colaptes_campestris 0 00000000000000000000000 Colaptes_melanochloros 0 01001000100000000000011 Cotinga_amabilis 0 00000000000000000000011 Cotinga_cotinga 0 00000000000000000000011 Cotinga_maculata 0 00000000000000000000011 Parus_caeruleus 0 00000000000000000000000 Cymbirhynchus_macrorhynchos 0 01001001000000000000000 Dendrocopos_major 0 01001000100000000000011 Dendroica_coronata 0 00000000000000000000000 Dendroica_palmarum 0 00000000000000000000000 Appendix S1 127 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Aegithalos_caudatus 0 0000000000000000000000 Agelaius_phoeniceus 0 1111111101001110001111 Alectoris_rufa 0 1111111101001110001111 Amandava_amandava 0 0000000000000000000000 Amandava_subflava 0 0000000000000000000000 Ampelioides_tschudii 0 0000000000000000000000 Ampelion_rufaxilla 0 0000000000000000000000 Anas_platyrhynchos 0 0000000000000010000011 Anser_anser 0 0000000000000000000000 Antilophia_galeata 0 0000000000000000000000 Apaloderma_narina 0 1111111101001110001111 Bombycilla_cedrorum 0 1111111100000000000000 Bombycilla_garrulus 0 1111111100000000000000 Bombycilla_japonica 0 1111111100000000000000 Bucanetes_githagineus 0 1111111101001110001111 Cacicus_cela 0 0000000000000000000000 Cacicus_haemorrhous 0 1111111101001110001111 Cacicus_leucoramphus 0 0000000000000000000000 Cacicus_melanicterus 0 0000000000000000000000 Cacicus_uropygialis 0 0000000001001110001111 Calyptomena_viridis 0 0000000000000000000000 Campephilus_leucopogon 0 1111111101001110001111 Cardinalis_cardinalis 0 1111111101001110001111 Cardinalis_sinuatus 0 1111111101001110001111 Carduelis_atrata 0 0000000000000000000000 Carduelis_cannabina 0 1111111101001110001111 Carduelis_carduelis 0 1111111100000000000000 Carduelis_chloris 0 0000000000000000000000 Carduelis_cucullata 0 1111111101001110001111 Carduelis_flammea 0 1111111101001110001111 Carduelis_hornemanni 0 1111111101001110001111 Carduelis_sinica 0 0000000000000000000000 Carduelis_spinoides 0 0000000000000000000000 Carduelis_spinus 0 0000000000000000000000 Carduelis_tristis 0 0000000000000000000000 Carpodacus_mexicanus 0 1111111101111111111111 Carpodacus_nipalensis 0 1111111100000000000000 Carpodacus_pulcherrimus 0 1111111100000000000000 Carpodacus_roseus 0 1111111100000000000000 Carpodacus_rubicilloides 0 0000000000000000000000 Carpodacus_thura 0 0000000001001110001111 Carpodacus_trifasciatus 0 1111111100000000000000 Carpornis_cucullata 0 0000000000000000000000 Chiroxiphia_caudata 0 0000000000000000000000 Chiroxiphia_pareola 0 0000000000000000000000 Chlorospingus_pileatus 0 0000000000000000000000 Ciconia_ciconia 0 0000000000000000000000 Coccothraustes_abeillei 0 0000000000000000000000 Coccothraustes_vespertinus 0 0000000000000000000000 Coereba_flaveola 0 0000000000000000000000 Colaptes_auratus 0 1111111101001110001111 Colaptes_auratus_cafer 0 1111111101001110001111 Colaptes_chrysoides 0 1111111100000000000000 Colaptes_campestris 0 0000000000000000000000 Colaptes_melanochloros 0 1111111100000000000000 Cotinga_amabilis 0 1111111100000000000000 Cotinga_cotinga 0 1111111100000000000000 Cotinga_maculata 0 1111111100000000000000 Parus_caeruleus 0 0000000000000000000000 Cymbirhynchus_macrorhynchos 0 0000000000000000000000 Dendrocopos_major 0 1111111100000000000000 Dendroica_coronata 0 0000000000000000000000 Dendroica_palmarum 0 0000000000000000000000 Appendix S1 128 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) Aegithalos_caudatus 0 00010000000000000000000 Agelaius_phoeniceus 0 11111111111000000000000 Alectoris_rufa 0 11000001111000000000000 Amandava_amandava 0 00000000000000001000000 Amandava_subflava 0 00000000000000001000000 Ampelioides_tschudii 0 00000000000001000000000 Ampelion_rufaxilla 0 00000000000001000000000 Anas_platyrhynchos 0 00000000000000001010000 Anser_anser 0 00000000000000011001000 Antilophia_galeata 0 00000000000000000000000 Apaloderma_narina 0 11110111111000000000000 Bombycilla_cedrorum 0 00011000000000010100000 Bombycilla_garrulus 0 00000000000000000000000 Bombycilla_japonica 0 00000000000000000000000 Bucanetes_githagineus 0 11111111111000000000000 Cacicus_cela 0 00000000000001000000000 Cacicus_haemorrhous 0 11010111111000000000000 Cacicus_leucoramphus 0 00000000000001100000000 Cacicus_melanicterus 0 00000000000001100000000 Cacicus_uropygialis 0 11010111100000000000000 Calyptomena_viridis 0 00000000000001000000000 Campephilus_leucopogon 0 11111111111000000000000 Cardinalis_cardinalis 0 11111111111000000100000 Cardinalis_sinuatus 0 11110111111000000000000 Carduelis_atrata 0 00000000000000000000000 Carduelis_cannabina 0 11111111111111000000000 Carduelis_carduelis 0 00000000000000000100000 Carduelis_chloris 0 00000000000000000000000 Carduelis_cucullata 0 11111111111000000000000 Carduelis_flammea 0 11111111111110000000000 Carduelis_hornemanni 0 11111111111100000000000 Carduelis_sinica 0 00000000000000000000000 Carduelis_spinoides 0 00000000000000000000000 Carduelis_spinus 0 00000000000000000000000 Carduelis_tristis 0 00000000000000011000000 Carpodacus_mexicanus 0 11111111111100000010000 Carpodacus_nipalensis 0 00011110011000000000000 Carpodacus_pulcherrimus 0 00011110011100000000000 Carpodacus_roseus 0 00011110011110000000000 Carpodacus_rubicilloides 0 00010000000100000000000 Carpodacus_thura 0 11110111100100000000000 Carpodacus_trifasciatus 0 00011110011100000000000 Carpornis_cucullata 0 00000000000001000000000 Chiroxiphia_caudata 0 00000000000000000000000 Chiroxiphia_pareola 0 00000000000000000000000 Chlorospingus_pileatus 0 00000000000001000000000 Ciconia_ciconia 0 00000000000001000001000 Coccothraustes_abeillei 0 00000000000001000000000 Coccothraustes_vespertinus 0 00000000000001000000000 Coereba_flaveola 0 00000000000000000000000 Colaptes_auratus 0 11111111111000000000000 Colaptes_auratus_cafer 0 11111111111000000000000 Colaptes_chrysoides 0 00011110011000000000000 Colaptes_campestris 0 00000000000001101000000 Colaptes_melanochloros 0 00011110011000000000000 Cotinga_amabilis 0 00000000000000000000000 Cotinga_cotinga 0 00000000000000000000000 Cotinga_maculata 0 00000000000000000000000 Parus_caeruleus 0 00000000000001100000000 Cymbirhynchus_macrorhynchos 0 00000000000000000000000 Dendrocopos_major 0 00011110011000000000000 Dendroica_coronata 0 00000000000001000000000 Dendroica_palmarum 0 00000000000001000000000 Appendix S1 129 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Aegithalos_caudatus 0 000000000000 Agelaius_phoeniceus 0 000000000000 Alectoris_rufa 0 000000000000 Amandava_amandava 0 000000000000 Amandava_subflava 0 000000000000 Ampelioides_tschudii 0 000000000000 Ampelion_rufaxilla 0 000000000000 Anas_platyrhynchos 0 000000000000 Anser_anser 0 000000000000 Antilophia_galeata 0 000000000000 Apaloderma_narina 0 000000000000 Bombycilla_cedrorum 0 000000000000 Bombycilla_garrulus 0 000000000000 Bombycilla_japonica 0 000000000000 Bucanetes_githagineus 0 000000000000 Cacicus_cela 0 000000000000 Cacicus_haemorrhous 0 000000000000 Cacicus_leucoramphus 0 000000000000 Cacicus_melanicterus 0 000000000000 Cacicus_uropygialis 0 000000000000 Calyptomena_viridis 0 000000000000 Campephilus_leucopogon 0 000000000000 Cardinalis_cardinalis 0 000000000000 Cardinalis_sinuatus 0 000000000000 Carduelis_atrata 0 000000000000 Carduelis_cannabina 0 000000000000 Carduelis_carduelis 0 000000000000 Carduelis_chloris 0 000000000000 Carduelis_cucullata 0 000000000000 Carduelis_flammea 0 000000000000 Carduelis_hornemanni 0 000000000000 Carduelis_sinica 0 000000000000 Carduelis_spinoides 0 000000000000 Carduelis_spinus 0 000000000000 Carduelis_tristis 0 000000000000 Carpodacus_mexicanus 0 000000000000 Carpodacus_nipalensis 0 000000000000 Carpodacus_pulcherrimus 0 000000000000 Carpodacus_roseus 0 000000000000 Carpodacus_rubicilloides 0 000000000000 Carpodacus_thura 0 000000000000 Carpodacus_trifasciatus 0 000000000000 Carpornis_cucullata 0 000000000000 Chiroxiphia_caudata 0 000000000000 Chiroxiphia_pareola 0 000000000000 Chlorospingus_pileatus 0 000000000000 Ciconia_ciconia 0 000000000000 Coccothraustes_abeillei 0 000000000000 Coccothraustes_vespertinus 0 000000000000 Coereba_flaveola 0 000000000000 Colaptes_auratus 0 000000000000 Colaptes_auratus_cafer 0 000000000000 Colaptes_chrysoides 0 000000000000 Colaptes_campestris 0 000000000000 Colaptes_melanochloros 0 000000000000 Cotinga_amabilis 0 000111100000 Cotinga_cotinga 0 000111100000 Cotinga_maculata 0 000111100000 Parus_caeruleus 0 000000001000 Cymbirhynchus_macrorhynchos 0 000000000000 Dendrocopos_major 0 000000000000 Dendroica_coronata 0 000000000000 Dendroica_palmarum 0 000000000000 Appendix S1 130 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Dendroica_petechia 0111100000000000000 0000 Dryocopus_pileatus 0110101100001000000 1110 Emberiza_citrinella 0110000000000000000 0000 Emberiza_melanocephala 0110000000000000000 0000 Erithacus_rubecula 0100000000000000100 0000 Erythrura_gouldiae 0110000000001111100 0010 Erythrura_psittacea 0110000000001000000 0010 Eudocimus_ruber 0000000000000000000 0000 Euphonia_laniirostris 0111110000000000100 0000 Euphonia_saturata 0110010000000000100 0000 Euplectes_afer 0110000000000000100 0000 Euplectes_ardens 0111100011001000100 0010 Euplectes_axillaris 0110010011000000100 0000 Euplectes_capensis 0110000000000000000 0000 Euplectes_macroura 0110000011000000100 0000 Euplectes_orix 0111100000001000100 0010 Eurylaimus_javanicus 0100001000001001100 0000 Eurylaimus_ochromalus 0100001000001001100 0000 Eurylaimus_steerii 0100000000001001000 0000 Ficedula_zanthopygia 0110000000000000000 0000 Foudia_madagascariensis 0111100000001000000 0010 Fregata_minor 0010000000110000000 0000 Fringilla_coelebs 0110100000000000100 0010 Fringilla_montifringilla 0100000000000000000 0000 Gallus_gallus_domesticus 0111100000000000000 0010 Geothlypis_trichas 0111100000000000000 0000 Haematoderus_militaris 0110100000001000001 0000 Haematospiza_sipahi 0111100000001000000 0010 Heterocercus_linteatus 0110100000001000000 0010 Icteria_virens 0100000000000000000 0000 Icterus_bullockii 0101000011000000100 0000 Icterus_croconotus 0100100000000000000 0000 Icterus_cucullatus 0100000000000000000 0000 Icterus_dominicensis 0100000000000000000 0000 Icterus_galbula 0111000011001000100 0010 Icterus_graduacauda 0100000000000000000 0000 Icterus_gularis 0111100011000000100 0000 Icterus_icterus 0110000011000000100 0000 Icterus_mesomelas 0100000000000000000 0000 Icterus_nigrogularis 0100000000000000000 0000 Icterus_pectoralis 0100000000000000000 0000 Icterus_prosthemelas 0100000000000000000 0000 Icterus_pustulatus 0111100011000000100 0000 Ilicura_militaris 0110100011000000111 0000 Iodopleura_isabellae 0001000000000000000 0000 Larus_delawarensis 0000000000000000000 0000 Larus_michahellis 0111100000000000100 0000 Leiothrix_argentauris 0110000000001000100 0010 Leiothrix_lutea 0110000000001000100 0010 Lepidothrix_coronata 0110000011000000100 0000 Lepidothrix_nattereri 0110000011000000100 0000 Lepidothrix_serena 0110000011000000100 0000 Larus_pipixcan 0000000000000000000 0000 Lipaugus_streptophorus 0010000000000000000 0010 Loxia_curvirostra 0100100011000000100 0000 Loxia_leucoptera 0111100011000000100 0010 Luscinia_calliope 0111100000001000000 0010 Machaeropterus_regulus 0110000011000000111 0000 Malurus_melanocephalus 0111100011001000100 0010 Masius_chrysopterus 0110000011000000110 0000 Melanerpes_candidus 0110001100000000000 1100 Melanerpes_formicivorus 0100100000001000000 0000 Melanerpes_lewis 0111100000001000000 0010 Appendix S1 131 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) Dendroica_petechia 0 0000000000000000000000 Dryocopus_pileatus 0 0111010100000000000100 Emberiza_citrinella 0 0000000000000000000000 Emberiza_melanocephala 0 0000000000000000000000 Erithacus_rubecula 0 0000000000000000000000 Erythrura_gouldiae 0 0111000000000000000100 Erythrura_psittacea 0 0111000000000000000100 Eudocimus_ruber 0 0000001000000000000000 Euphonia_laniirostris 0 0000000000000000000010 Euphonia_saturata 0 0000000000000000000010 Euplectes_afer 0 0000000000000000000000 Euplectes_ardens 0 0111111110100000000100 Euplectes_axillaris 0 0000000000000000000000 Euplectes_capensis 0 0000000000000000000000 Euplectes_macroura 0 0000000000000000000000 Euplectes_orix 0 0111111110100000000100 Eurylaimus_javanicus 0 0000000000000000000100 Eurylaimus_ochromalus 0 0000000000000000000100 Eurylaimus_steerii 0 0000000000000000000100 Ficedula_zanthopygia 0 0000000000000000000000 Foudia_madagascariensis 0 0111111110100000000100 Fregata_minor 0 0011000000000000000000 Fringilla_coelebs 0 0111010100000001100000 Fringilla_montifringilla 0 0000000000000000000000 Gallus_gallus_domesticus 0 0111111111100000000000 Geothlypis_trichas 0 0000000000000000000000 Haematoderus_militaris 0 0111010100000000000100 Haematospiza_sipahi 0 0111111110100001100100 Heterocercus_linteatus 0 0111010100000000000100 Icteria_virens 0 0000000000000000000000 Icterus_bullockii 0 0000101010100000000000 Icterus_croconotus 0 0000010000000000000000 Icterus_cucullatus 0 0000000000000000000000 Icterus_dominicensis 0 0000000000000000000000 Icterus_galbula 0 0111111110100000000100 Icterus_graduacauda 0 0000000000000000000000 Icterus_gularis 0 0000111110100000000000 Icterus_icterus 0 0000000000000000000000 Icterus_mesomelas 0 0000000000000000000000 Icterus_nigrogularis 0 0000000000000000000000 Icterus_pectoralis 0 0000000000000000000000 Icterus_prosthemelas 0 0000000000000000000000 Icterus_pustulatus 0 0000111110100000000000 Ilicura_militaris 0 0000000000000000000001 Iodopleura_isabellae 0 0000101010100000000000 Larus_delawarensis 0 0001000000000000000000 Larus_michahellis 0 0011101100000000000000 Leiothrix_argentauris 0 0111000000000000000100 Leiothrix_lutea 0 0111000000000000000100 Lepidothrix_coronata 0 0000000000000000000000 Lepidothrix_nattereri 0 0000000000000000000000 Lepidothrix_serena 0 0000000000000000000000 Larus_pipixcan 0 0001000000000000000000 Lipaugus_streptophorus 0 0111000000000000000000 Loxia_curvirostra 0 0000010000000001111000 Loxia_leucoptera 0 0111111110100001111000 Luscinia_calliope 0 0111111110100000000100 Machaeropterus_regulus 0 0000000000000000000001 Malurus_melanocephalus 0 0111111110100000000100 Masius_chrysopterus 0 0000000000000000000000 Melanerpes_candidus 0 0000000000000000000000 Melanerpes_formicivorus 0 0000010100000000000100 Melanerpes_lewis 0 0111111110100000000100 Appendix S1 132 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) Dendroica_petechia 0 000 00000 0 0000000000000 Dryocopus_pileatus 0 000 00000 0 0000000000110 Emberiza_citrinella 0 000 00000 0 0000000000000 Emberiza_melanocephala 0 000 00000 0 0000000000000 Erithacus_rubecula 0 000 00000 0 0000000100000 Erythrura_gouldiae 0 000 00000 0 0000000110000 Erythrura_psittacea 0 000 00000 0 0000000000000 Eudocimus_ruber 0 000 00000 0 0000000000000 Euphonia_laniirostris 0 000 00000 0 0010000110001 Euphonia_saturata 0 000 00000 0 0010000110001 Euplectes_afer 0 000 00000 0 0000000110000 Euplectes_ardens 0 000 00000 0 0001111111000 Euplectes_axillaris 0 000 00000 0 0011111111000 Euplectes_capensis 0 000 00000 0 0000000000000 Euplectes_macroura 0 000 00000 0 0001111111000 Euplectes_orix 0 000 00000 0 0000000110000 Eurylaimus_javanicus 0 000 00000 1 1000000100100 Eurylaimus_ochromalus 0 000 00000 1 1000000100100 Eurylaimus_steerii 0 000 00000 1 0000000000000 Ficedula_zanthopygia 0 000 00000 0 0000000000000 Foudia_madagascariensis 0 000 00000 0 0000000000000 Fregata_minor 0 000 00000 0 0000000000000 Fringilla_coelebs 0 000 00000 0 0000000110000 Fringilla_montifringilla 0 000 00000 0 0000000000000 Gallus_gallus_domesticus 0 000 00000 0 0000000000000 Geothlypis_trichas 0 000 00000 0 0000000000000 Haematoderus_militaris 0 100 11101 0 0000000000000 Haematospiza_sipahi 0 000 00000 0 0000000000000 Heterocercus_linteatus 0 000 00000 0 0000000000000 Icteria_virens 0 000 00000 0 0000000000000 Icterus_bullockii 0 000 00000 0 0001111101000 Icterus_croconotus 0 000 00000 0 0000000000000 Icterus_cucullatus 0 000 00000 0 0000000000000 Icterus_dominicensis 0 000 00000 0 0000000000000 Icterus_galbula 0 000 00000 0 0001111111000 Icterus_graduacauda 0 000 00000 0 0000000000000 Icterus_gularis 0 000 00000 0 0001111111000 Icterus_icterus 0 000 00000 0 0001111111000 Icterus_mesomelas 0 000 00000 0 0000000000000 Icterus_nigrogularis 0 000 00000 0 0000000000000 Icterus_pectoralis 0 000 00000 0 0000000000000 Icterus_prosthemelas 0 000 00000 0 0000000000000 Icterus_pustulatus 0 000 00000 0 0001111111000 Ilicura_militaris 0 000 00000 0 0001111111000 Iodopleura_isabellae 0 000 00000 0 0000000000000 Larus_delawarensis 0 000 00000 0 0000000000000 Larus_michahellis 0 000 00000 0 0000000110000 Leiothrix_argentauris 0 000 00000 0 0000000110000 Leiothrix_lutea 0 000 00000 0 0000000110000 Lepidothrix_coronata 0 000 00000 0 0001111111000 Lepidothrix_nattereri 0 000 00000 0 0001111111000 Lepidothrix_serena 0 000 00000 0 0001111101000 Larus_pipixcan 0 000 00000 0 0000000000000 Lipaugus_streptophorus 0 000 11110 0 0000000000000 Loxia_curvirostra 0 000 00000 0 0001111101000 Loxia_leucoptera 0 000 00000 0 0001111111000 Luscinia_calliope 0 000 00000 0 0000000000000 Machaeropterus_regulus 0 000 00000 0 0001111111000 Malurus_melanocephalus 0 000 00000 0 0001111111000 Masius_chrysopterus 0 000 00000 0 0001111111000 Melanerpes_candidus 0 000 00000 0 0000000000110 Melanerpes_formicivorus 0 000 00000 0 0000000000000 Melanerpes_lewis 0 000 00000 0 0000000000000 Appendix S1 133 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Dendroica_petechia 0 00000000000000000000000 Dryocopus_pileatus 0 01001000100000110000011 Emberiza_citrinella 0 00000000000000000000000 Emberiza_melanocephala 0 00000000000000000000000 Erithacus_rubecula 0 00000000000000000000000 Erythrura_gouldiae 0 01111111100000001000011 Erythrura_psittacea 0 01001000100000000000011 Eudocimus_ruber 0 00000000000000000000000 Euphonia_laniirostris 0 00000000000000001000000 Euphonia_saturata 0 00000000000000001000000 Euplectes_afer 0 00000000000000001000000 Euplectes_ardens 0 01001000100000001000011 Euplectes_axillaris 0 00000000000000001000000 Euplectes_capensis 0 00000000000000000000000 Euplectes_macroura 0 00000000000000001000000 Euplectes_orix 0 01001000100000001000011 Eurylaimus_javanicus 0 01001001000000000000000 Eurylaimus_ochromalus 0 01001001000000000000000 Eurylaimus_steerii 0 01001001000000000000000 Ficedula_zanthopygia 0 00000000000000000000000 Foudia_madagascariensis 0 01001000100000000000011 Fregata_minor 0 00000000001000000000000 Fringilla_coelebs 0 00000000000000001000011 Fringilla_montifringilla 0 00000000000000000000000 Gallus_gallus_domesticus 0 00000000000000000000011 Geothlypis_trichas 0 00000000000000000000000 Haematoderus_militaris 0 01001000000000000000011 Haematospiza_sipahi 0 01001000100000000000011 Heterocercus_linteatus 0 01001000100000000000011 Icteria_virens 0 00000000000000000000000 Icterus_bullockii 0 00000000000000000000000 Icterus_croconotus 0 00000000000000000000000 Icterus_cucullatus 0 00000000000000000000000 Icterus_dominicensis 0 00000000000000000000000 Icterus_galbula 0 01001000100000001000011 Icterus_graduacauda 0 00000000000000000000000 Icterus_gularis 0 00000000000000001000000 Icterus_icterus 0 00000000000000001000000 Icterus_mesomelas 0 00000000000000000000000 Icterus_nigrogularis 0 00000000000000000000000 Icterus_pectoralis 0 00000000000000000000000 Icterus_prosthemelas 0 00000000000000000000000 Icterus_pustulatus 0 00000000000000001000000 Ilicura_militaris 0 00000000000000001111100 Iodopleura_isabellae 0 00000000000000000000000 Larus_delawarensis 0 00000000000000000000000 Larus_michahellis 0 00000000000000001000000 Leiothrix_argentauris 0 01001000100000001000011 Leiothrix_lutea 0 01001000100000001000011 Lepidothrix_coronata 0 00000000000000001000000 Lepidothrix_nattereri 0 00000000000000001000000 Lepidothrix_serena 0 00000000000000001000000 Larus_pipixcan 0 00000000000000000000000 Lipaugus_streptophorus 0 00000000000000000000011 Loxia_curvirostra 0 00000000000000000000000 Loxia_leucoptera 0 00000000000000001000011 Luscinia_calliope 0 01001000100000000000011 Machaeropterus_regulus 0 00000000000000001111100 Malurus_melanocephalus 0 01001000100000001000011 Masius_chrysopterus 0 00000000000000001100000 Melanerpes_candidus 0 00000000000000110000000 Melanerpes_formicivorus 0 01001000000000000000000 Melanerpes_lewis 0 01001000100000000000011 Appendix S1 134 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Dendroica_petechia 0 0000000000000000000000 Dryocopus_pileatus 0 1111111100000000000000 Emberiza_citrinella 0 0000000000000000000000 Emberiza_melanocephala 0 0000000000000000000000 Erithacus_rubecula 0 0000000000000000000000 Erythrura_gouldiae 0 1111111100000000000000 Erythrura_psittacea 0 1111111100000000000000 Eudocimus_ruber 0 0000000000000000000000 Euphonia_laniirostris 0 0000000000000000000000 Euphonia_saturata 0 0000000000000000000000 Euplectes_afer 0 0000000000000000000000 Euplectes_ardens 0 1111111101001110001111 Euplectes_axillaris 0 0000000000000000000000 Euplectes_capensis 0 0000000000000000000000 Euplectes_macroura 0 0000000000000000000000 Euplectes_orix 0 1111111101001110001111 Eurylaimus_javanicus 0 0000000000000000000000 Eurylaimus_ochromalus 0 0000000000000000000000 Eurylaimus_steerii 0 0000000000000000000000 Ficedula_zanthopygia 0 0000000000000000000000 Foudia_madagascariensis 0 1111111101001110001111 Fregata_minor 0 0001100100000000000000 Fringilla_coelebs 0 1111111100000000000000 Fringilla_montifringilla 0 0000000000000000000000 Gallus_gallus_domesticus 0 1111111101111111111111 Geothlypis_trichas 0 0000000000000000000000 Haematoderus_militaris 0 1001100100000000000000 Haematospiza_sipahi 0 1111111101001110001111 Heterocercus_linteatus 0 1111111100000000000000 Icteria_virens 0 0000000000000000000000 Icterus_bullockii 0 0000000001001110001111 Icterus_croconotus 0 0000000000000000000000 Icterus_cucullatus 0 0000000000000000000000 Icterus_dominicensis 0 0000000000000000000000 Icterus_galbula 0 1111111101001110001111 Icterus_graduacauda 0 0000000000000000000000 Icterus_gularis 0 0000000001001110001111 Icterus_icterus 0 0000000000000000000000 Icterus_mesomelas 0 0000000000000000000000 Icterus_nigrogularis 0 0000000000000000000000 Icterus_pectoralis 0 0000000000000000000000 Icterus_prosthemelas 0 0000000000000000000000 Icterus_pustulatus 0 0000000001001110001111 Ilicura_militaris 0 0000000000000000000000 Iodopleura_isabellae 0 0000000001001110001111 Larus_delawarensis 0 0000000000000000000000 Larus_michahellis 0 0001100100000010000011 Leiothrix_argentauris 0 1111111100000000000000 Leiothrix_lutea 0 1111111100000000000000 Lepidothrix_coronata 0 0000000000000000000000 Lepidothrix_nattereri 0 0000000000000000000000 Lepidothrix_serena 0 0000000000000000000000 Larus_pipixcan 0 0000000000000000000000 Lipaugus_streptophorus 0 1111111100000000000000 Loxia_curvirostra 0 0000000000000000000000 Loxia_leucoptera 0 1111111101001110001111 Luscinia_calliope 0 1111111101001110001111 Machaeropterus_regulus 0 0000000000000000000000 Malurus_melanocephalus 0 1111111101001110001111 Masius_chrysopterus 0 0000000000000000000000 Melanerpes_candidus 0 0000000000000000000000 Melanerpes_formicivorus 0 0000000000000000000000 Melanerpes_lewis 0 1111111101001110001111 Appendix S1 135 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) Dendroica_petechia 0 00000000000001111000000 Dryocopus_pileatus 0 00011110011000000000000 Emberiza_citrinella 0 00000000000001100000000 Emberiza_melanocephala 0 00000000000001100000000 Erithacus_rubecula 0 00000000000000000000000 Erythrura_gouldiae 0 00000000000000000000000 Erythrura_psittacea 0 00000000000000000000000 Eudocimus_ruber 0 00000000000000000000100 Euphonia_laniirostris 0 00000000000000011000000 Euphonia_saturata 0 00000000000000000000000 Euplectes_afer 0 00000000000000000000000 Euplectes_ardens 0 11111111111000000000000 Euplectes_axillaris 0 00000000000000000000000 Euplectes_capensis 0 00000000000001100000000 Euplectes_macroura 0 00000000000000000000000 Euplectes_orix 0 11111111111000000000000 Eurylaimus_javanicus 0 00000000000000000000000 Eurylaimus_ochromalus 0 00000000000000000000000 Eurylaimus_steerii 0 00000000000000000000000 Ficedula_zanthopygia 0 00000000000001100000000 Foudia_madagascariensis 0 11111111111000000000000 Fregata_minor 0 00000000000000000000000 Fringilla_coelebs 0 00011110011100000000000 Fringilla_montifringilla 0 00000000000001000000000 Gallus_gallus_domesticus 0 11111111111001000000000 Geothlypis_trichas 0 00000000000001111000000 Haematoderus_militaris 0 00010110000000000100000 Haematospiza_sipahi 0 11111111111100000000000 Heterocercus_linteatus 0 00011110011000000000000 Icteria_virens 0 00000000000001000000000 Icterus_bullockii 0 11000000000000000000000 Icterus_croconotus 0 00010000000001000000000 Icterus_cucullatus 0 00000000000001000000000 Icterus_dominicensis 0 00000000000001000000000 Icterus_galbula 0 11101111111000000000000 Icterus_graduacauda 0 00000000000001000000000 Icterus_gularis 0 11110111100000000000000 Icterus_icterus 0 00000000000000000000000 Icterus_mesomelas 0 00000000000001000000000 Icterus_nigrogularis 0 00000000000001000000000 Icterus_pectoralis 0 00000000000001000000000 Icterus_prosthemelas 0 00000000000001000000000 Icterus_pustulatus 0 11110111100000000000000 Ilicura_militaris 0 00000000000000001000000 Iodopleura_isabellae 0 11000000000000000000000 Larus_delawarensis 0 00000000000000000001000 Larus_michahellis 0 00000001111000001000000 Leiothrix_argentauris 0 00000000000000000000000 Leiothrix_lutea 0 00000000000000000000000 Lepidothrix_coronata 0 00000000000000000000000 Lepidothrix_nattereri 0 00000000000000000000000 Lepidothrix_serena 0 00000000000000000000000 Larus_pipixcan 0 00000000000000000001000 Lipaugus_streptophorus 0 00000000000000000000000 Loxia_curvirostra 0 00010000000110000000000 Loxia_leucoptera 0 11111111111110000000000 Luscinia_calliope 0 11111111111000000000000 Machaeropterus_regulus 0 00000000000000000000000 Malurus_melanocephalus 0 11111111111000000000000 Masius_chrysopterus 0 00000000000000000000000 Melanerpes_candidus 0 00000000000000000000000 Melanerpes_formicivorus 0 00010110000000000000000 Melanerpes_lewis 0 11111111111000000000000 Appendix S1 136 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Dendroica_petechia 0 000000000000 Dryocopus_pileatus 0 000000000000 Emberiza_citrinella 0 000000000000 Emberiza_melanocephala 0 000000000000 Erithacus_rubecula 0 000000000000 Erythrura_gouldiae 0 000000000000 Erythrura_psittacea 0 000000000000 Eudocimus_ruber 0 000000000000 Euphonia_laniirostris 0 000000000000 Euphonia_saturata 0 000000000000 Euplectes_afer 0 000000000000 Euplectes_ardens 0 000000000000 Euplectes_axillaris 0 000000000000 Euplectes_capensis 0 000000000000 Euplectes_macroura 0 000000000000 Euplectes_orix 0 000000001110 Eurylaimus_javanicus 0 000000001110 Eurylaimus_ochromalus 0 000000001000 Eurylaimus_steerii 0 000000000000 Ficedula_zanthopygia 0 000000000000 Foudia_madagascariensis 0 000000000000 Fregata_minor 0 000000000000 Fringilla_coelebs 0 000000000000 Fringilla_montifringilla 0 000000000000 Gallus_gallus_domesticus 0 000000000000 Geothlypis_trichas 0 000000010000 Haematoderus_militaris 0 100111000000 Haematospiza_sipahi 0 000000000000 Heterocercus_linteatus 0 000000000000 Icteria_virens 0 000000000000 Icterus_bullockii 0 000000000000 Icterus_croconotus 0 000000000000 Icterus_cucullatus 0 000000000000 Icterus_dominicensis 0 000000000000 Icterus_galbula 0 000000000000 Icterus_graduacauda 0 000000000000 Icterus_gularis 0 000000000000 Icterus_icterus 0 000000000000 Icterus_mesomelas 0 000000000000 Icterus_nigrogularis 0 000000000000 Icterus_pectoralis 0 000000000000 Icterus_prosthemelas 0 000000000000 Icterus_pustulatus 0 000000000000 Ilicura_militaris 0 000000000000 Iodopleura_isabellae 0 000000000000 Larus_delawarensis 0 000000000000 Larus_michahellis 0 000000000000 Leiothrix_argentauris 0 000000000000 Leiothrix_lutea 0 000000000000 Lepidothrix_coronata 0 000000000000 Lepidothrix_nattereri 0 000000000000 Lepidothrix_serena 0 000000000000 Larus_pipixcan 0 000000000000 Lipaugus_streptophorus 0 000111100000 Loxia_curvirostra 0 000000000000 Loxia_leucoptera 0 000000000000 Luscinia_calliope 0 000000000000 Machaeropterus_regulus 0 000000000000 Malurus_melanocephalus 0 000000000000 Masius_chrysopterus 0 000000000000 Melanerpes_candidus 0 000000000000 Melanerpes_formicivorus 0 000000000000 Melanerpes_lewis 0 000000000000 Appendix S1 137 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Meleagris_gallopavo 0111100000000000000 0010 Motacilla_flava 0110000000000000000 0000 Mycerobas_affinis 0110000000000000000 0000 Mycerobas_carnipes 0100000000000000000 0000 Mycerobas_icterioides 0100000000000000000 0000 Mycerobas_melanozanthos 0100000000000000000 0000 Neochmia_ruficauda 0110000000000000000 0000 Neopelma_pallescens 0110000011000000100 0000 Neophron_percnopterus 0100000000000000000 0000 Nesospiza_acunhae 0110000011000000100 0000 Notiomystis_cincta 0111100000000000100 0000 Oriolus_cruentus 0111100000001111000 0010 Oriolus_oriolus 0110000000000000000 0000 Oriolus_traillii 0101000011001000000 0000 Oriolus_xanthornus 0110000000000000000 0000 Paroaria_coronata 0111100011001000000 0010 Parus_major 0110000000000000000 0000 Parus_spilonotus 0110000000000000000 0000 Perdix_perdix 0100100000000000000 0000 Pericrocotus_flammeus 0111100000001000000 0010 Parus_ater 0110000000000000000 0000 Phaethon_rubricauda 0111000000001000000 0010 Phasianus_colchicus 0100100000000000000 0000 Pheucticus_ludovicianus 0111100000000000000 0010 Phibalura_flavirostris 0100000000000000000 0000 Phoenicircus_carnifex 0100000000000000001 0000 Phoenicoparrus_andinus 0101000011000000100 0001 Phoenicopterus_chilensis 0001000000000000000 0000 Phoenicoparrus_jamesi 0101000011000000100 0001 Phoeniconaias_minor 0001000000000000000 0000 Phoenicopterus_roseus 0001000000000000000 0000 Phoenicopterus_ruber 0011000000000000000 0010 Picoides_tridactylus 0110000000000000000 0000 Picoides_villosus 0111101100001000000 1110 Picumnus_exilis 0100100000001000000 0000 Picus_squamatus 0100001100000000000 0000 Picus_viridis 0111101100001000000 0010 Pinicola_enucleator 0101100000001000100 0000 Pipra_aureola 0110000011000000111 0000 Pipra_chloromeros 0111100000001000001 0010 Pipra_erythrocephala 0111100000001000000 0010 Pipra_fasciicauda 0110000011000000111 0000 Pipra_filicauda 0110000011000000111 0000 Pipra_rubrocapilla 0111100000001000000 0010 Pipreola_aureopectus 0110000000000000000 0000 Pipreola_chlorolepidota 0110000000000000000 0000 Pipreola_formosa 0110000000000000000 0000 Pipreola_whitelyi 0100000000001000000 0000 Piranga_flava 0111100011000000100 0010 Piranga_ludoviciana 0100000011000000101 0000 Piranga_olivacea 0111100011001000100 0010 Piranga_rubra 0001000000000000000 0000 Platalea_ajaja 0000000000000000000 0000 Ploceus_bicolor 0110000000000000000 0000 Ploceus_capensis 0110000000000000000 0000 Ploceus_cucullatus 0110000000000000000 0000 Ploceus_nelicourvi 0110000000000000000 0000 Ploceus_philippinus 0110000000000000000 0000 Ploceus_sakalava 0110000000000000000 0000 Ploceus_velatus 0110000000000000000 0000 Porphyrolaema_porphyrolaema 0010000000000000000 0010 Procnias_tricarunculatus 0100000011000000000 0000 Psarisomus_dalhousiae 0110000000000000000 0000 Appendix S1 138 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) Meleagris_gallopavo 0 0111010000000000000000 Motacilla_flava 0 0000000000000000000000 Mycerobas_affinis 0 0000000000000000000000 Mycerobas_carnipes 0 0000000000000000000000 Mycerobas_icterioides 0 0000000000000000000000 Mycerobas_melanozanthos 0 0000000000000000000000 Neochmia_ruficauda 0 0000000000000000000000 Neopelma_pallescens 0 0000000000000000000000 Neophron_percnopterus 0 0000000000000000000000 Nesospiza_acunhae 0 0000000000000010000000 Notiomystis_cincta 0 0000000000000000000010 Oriolus_cruentus 0 0111111110100000000100 Oriolus_oriolus 0 0000000000000000000000 Oriolus_traillii 0 0000101010100000000100 Oriolus_xanthornus 0 0000000000000000000000 Paroaria_coronata 0 0111111100100000000100 Parus_major 0 0000000000000000000010 Parus_spilonotus 0 0000000000000000000000 Perdix_perdix 0 0011111100000000000000 Pericrocotus_flammeus 0 0111111110100000000100 Parus_ater 0 0000000000000000000000 Phaethon_rubricauda 0 0111101110100000000100 Phasianus_colchicus 0 0011111100000000000000 Pheucticus_ludovicianus 0 0111111110100000000000 Phibalura_flavirostris 0 0000000000000000000000 Phoenicircus_carnifex 0 0000000000000000000000 Phoenicoparrus_andinus 0 0001111110100000000000 Phoenicopterus_chilensis 0 0001111110100000000000 Phoenicoparrus_jamesi 0 0001111110100000000000 Phoeniconaias_minor 0 0001111110100000000000 Phoenicopterus_roseus 0 0001111110100000000000 Phoenicopterus_ruber 0 0111111110111100000000 Picoides_tridactylus 0 0000000000000000000000 Picoides_villosus 0 0111111110100000000100 Picumnus_exilis 0 0000010100000000000100 Picus_squamatus 0 0000000000000000000000 Picus_viridis 0 1111111110100000000100 Pinicola_enucleator 0 0000111110100001100100 Pipra_aureola 0 0000000000000000000001 Pipra_chloromeros 0 0111111110100000000100 Pipra_erythrocephala 0 0111111110100000000100 Pipra_fasciicauda 0 0000000000000000000001 Pipra_filicauda 0 0000000000000000000001 Pipra_rubrocapilla 0 0111111110100000000100 Pipreola_aureopectus 0 0000000000000000000000 Pipreola_chlorolepidota 0 0000000000000000000000 Pipreola_formosa 0 0000000000000000000000 Pipreola_whitelyi 0 0000000000000000000100 Piranga_flava 0 0111111110100000000000 Piranga_ludoviciana 0 0000000000000000000000 Piranga_olivacea 0 0111111110100000000100 Piranga_rubra 0 0000101010100000000000 Platalea_ajaja 0 0001001100000000000000 Ploceus_bicolor 0 0000000000000000000000 Ploceus_capensis 0 0000000000000000000000 Ploceus_cucullatus 0 0000000000000000000000 Ploceus_nelicourvi 0 0000000000000000000000 Ploceus_philippinus 0 0000000000000000000000 Ploceus_sakalava 0 0000000000000000000000 Ploceus_velatus 0 0000000000000000000000 Porphyrolaema_porphyrolaema 0 0111000000000000000000 Procnias_tricarunculatus 0 0000000000000000000000 Psarisomus_dalhousiae 0 0000000000000000000000 Appendix S1 139 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) Meleagris_gallopavo 0 000 00000 0 0000000000000 Motacilla_flava 0 000 00000 0 0000000000000 Mycerobas_affinis 0 000 00000 0 0000000000000 Mycerobas_carnipes 0 000 00000 0 0000000000000 Mycerobas_icterioides 0 000 00000 0 0000000000000 Mycerobas_melanozanthos 0 000 00000 0 0000000000000 Neochmia_ruficauda 0 000 00000 0 0000000000000 Neopelma_pallescens 0 000 00000 0 0001111111000 Neophron_percnopterus 0 000 00000 0 0000000000000 Nesospiza_acunhae 0 000 00000 0 0001111111000 Notiomystis_cincta 0 000 00000 0 0000000110001 Oriolus_cruentus 0 000 00000 0 0000000000000 Oriolus_oriolus 0 000 00000 0 0000000000000 Oriolus_traillii 0 000 00000 0 0101111000000 Oriolus_xanthornus 0 000 00000 0 0000000000000 Paroaria_coronata 0 000 00000 0 0001111000000 Parus_major 0 000 00000 0 0000000000001 Parus_spilonotus 0 000 00000 0 0000000000000 Perdix_perdix 0 000 00000 0 0000000000000 Pericrocotus_flammeus 0 000 00000 0 0000000000000 Parus_ater 0 000 00000 0 0000000000000 Phaethon_rubricauda 0 000 00000 0 0000000000000 Phasianus_colchicus 0 000 00000 0 0000000000000 Pheucticus_ludovicianus 0 000 00000 0 0000000000000 Phibalura_flavirostris 0 000 00000 0 0000000000000 Phoenicircus_carnifex 0 000 00000 0 0000000000000 Phoenicoparrus_andinus 0 000 00000 0 0001111101000 Phoenicopterus_chilensis 0 000 00000 0 0000000000000 Phoenicoparrus_jamesi 0 000 00000 0 0001111101000 Phoeniconaias_minor 0 000 00000 0 0000000000000 Phoenicopterus_roseus 0 000 00000 0 0000000000000 Phoenicopterus_ruber 0 000 00000 0 0000000000000 Picoides_tridactylus 0 000 00000 0 0000000000000 Picoides_villosus 0 000 00000 0 0000000000110 Picumnus_exilis 0 000 00000 0 0000000000000 Picus_squamatus 0 000 00000 0 0000000000110 Picus_viridis 0 000 00000 0 0000000000110 Pinicola_enucleator 0 000 00000 0 0000000100000 Pipra_aureola 0 000 00000 0 0001111111000 Pipra_chloromeros 0 000 00000 0 0000000000000 Pipra_erythrocephala 0 000 00000 0 0000000000000 Pipra_fasciicauda 0 000 00000 0 0001111111000 Pipra_filicauda 0 000 00000 0 0001111111000 Pipra_rubrocapilla 0 000 00000 0 0000000000000 Pipreola_aureopectus 0 000 00000 0 0000000000000 Pipreola_chlorolepidota 0 000 00000 0 0000000000000 Pipreola_formosa 0 000 00000 0 0000000000000 Pipreola_whitelyi 0 000 00000 0 0000000000000 Piranga_flava 0 000 00000 0 0001111111000 Piranga_ludoviciana 0 000 00000 0 0001111101000 Piranga_olivacea 0 000 00000 0 0001111111000 Piranga_rubra 0 000 00000 0 0000000000000 Platalea_ajaja 0 000 00000 0 0000000000000 Ploceus_bicolor 0 000 00000 0 0000000000000 Ploceus_capensis 0 000 00000 0 0000000000000 Ploceus_cucullatus 0 000 00000 0 0000000000000 Ploceus_nelicourvi 0 000 00000 0 0000000000000 Ploceus_philippinus 0 000 00000 0 0000000000000 Ploceus_sakalava 0 000 00000 0 0000000000000 Ploceus_velatus 0 000 00000 0 0000000000000 Porphyrolaema_porphyrolaema 0 000 11110 0 0000000000000 Procnias_tricarunculatus 0 000 00000 0 0001111000000 Psarisomus_dalhousiae 0 000 00000 0 0000000000000 Appendix S1 140 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Meleagris_gallopavo 0 00000000000000000000011 Motacilla_flava 0 00000000000000000000000 Mycerobas_affinis 0 00000000000000000000000 Mycerobas_carnipes 0 00000000000000000000000 Mycerobas_icterioides 0 00000000000000000000000 Mycerobas_melanozanthos 0 00000000000000000000000 Neochmia_ruficauda 0 00000000000000000000000 Neopelma_pallescens 0 00000000000000001000000 Neophron_percnopterus 0 00000000000000000000000 Nesospiza_acunhae 0 10000000000000001000000 Notiomystis_cincta 0 00000000000000001000000 Oriolus_cruentus 0 01111111100000000000011 Oriolus_oriolus 0 00000000000000000000000 Oriolus_traillii 0 01001000000000000000000 Oriolus_xanthornus 0 00000000000000000000000 Paroaria_coronata 0 01001000100000000000011 Parus_major 0 00000000000000000000000 Parus_spilonotus 0 00000000000000000000000 Perdix_perdix 0 00000000000000000000000 Pericrocotus_flammeus 0 01001000100000000000011 Parus_ater 0 00000000000000000000000 Phaethon_rubricauda 0 01001000100000000000011 Phasianus_colchicus 0 00000000000000000000000 Pheucticus_ludovicianus 0 00000000000000000000011 Phibalura_flavirostris 0 00000000000000000000000 Phoenicircus_carnifex 0 00000000000000000000000 Phoenicoparrus_andinus 0 00000000000000000000000 Phoenicopterus_chilensis 0 00000000000000000000000 Phoenicoparrus_jamesi 0 00000000000000000000000 Phoeniconaias_minor 0 00000000000000000000000 Phoenicopterus_roseus 0 00000000000000000000000 Phoenicopterus_ruber 0 00000000000011000000011 Picoides_tridactylus 0 00000000000000000000000 Picoides_villosus 0 01001000100000110000011 Picumnus_exilis 0 01001000000000000000000 Picus_squamatus 0 00000000000000000000000 Picus_viridis 0 01001000100000000000011 Pinicola_enucleator 0 01001000000000000000000 Pipra_aureola 0 00000000000000001111100 Pipra_chloromeros 0 01001000100000000000011 Pipra_erythrocephala 0 01001000100000000000011 Pipra_fasciicauda 0 00000000000000001111100 Pipra_filicauda 0 00000000000000001111100 Pipra_rubrocapilla 0 01001000100000000000011 Pipreola_aureopectus 0 00000000000000000000000 Pipreola_chlorolepidota 0 00000000000000000000000 Pipreola_formosa 0 00000000000000000000000 Pipreola_whitelyi 0 01001000000000000000000 Piranga_flava 0 00000000000000001000011 Piranga_ludoviciana 0 00000000000000000000000 Piranga_olivacea 0 01001000100000001000011 Piranga_rubra 0 00000000000000000000000 Platalea_ajaja 0 00000000000000000000000 Ploceus_bicolor 0 00000000000000000000000 Ploceus_capensis 0 00000000000000000000000 Ploceus_cucullatus 0 00000000000000000000000 Ploceus_nelicourvi 0 00000000000000000000000 Ploceus_philippinus 0 00000000000000000000000 Ploceus_sakalava 0 00000000000000000000000 Ploceus_velatus 0 00000000000000000000000 Porphyrolaema_porphyrolaema 0 00000000000000000000011 Procnias_tricarunculatus 0 00000000000000000000000 Psarisomus_dalhousiae 0 00000000000000000000000 Appendix S1 141 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Meleagris_gallopavo 0 1111111100000000000000 Motacilla_flava 0 0000000000000000000000 Mycerobas_affinis 0 0000000000000000000000 Mycerobas_carnipes 0 0000000000000000000000 Mycerobas_icterioides 0 0000000000000000000000 Mycerobas_melanozanthos 0 0000000000000000000000 Neochmia_ruficauda 0 0000000000000000000000 Neopelma_pallescens 0 0000000000000000000000 Neophron_percnopterus 0 0000000000000000000000 Nesospiza_acunhae 0 0000000000000000000000 Notiomystis_cincta 0 0000000000000000000000 Oriolus_cruentus 0 1111111101001110001111 Oriolus_oriolus 0 0000000000000000000000 Oriolus_traillii 0 0000000001001110001111 Oriolus_xanthornus 0 0000000000000000000000 Paroaria_coronata 0 1111111101001110000011 Parus_major 0 0000000000000000000000 Parus_spilonotus 0 0000000000000000000000 Perdix_perdix 0 0001100000000000000011 Pericrocotus_flammeus 0 1111111101001110001111 Parus_ater 0 0000000000000000000000 Phaethon_rubricauda 0 1111111101001110001111 Phasianus_colchicus 0 0001100000000000000011 Pheucticus_ludovicianus 0 1111111101001110001111 Phibalura_flavirostris 0 0000000000000000000000 Phoenicircus_carnifex 0 0000000000000000000000 Phoenicoparrus_andinus 0 0000000001001110001111 Phoenicopterus_chilensis 0 0000000001001110001111 Phoenicoparrus_jamesi 0 0000000001001110001111 Phoeniconaias_minor 0 0000000001001110001111 Phoenicopterus_roseus 0 0000000001001110001111 Phoenicopterus_ruber 0 1111111101001110001111 Picoides_tridactylus 0 0000000000000000000000 Picoides_villosus 0 1111111101001110001111 Picumnus_exilis 0 0000000000000000000000 Picus_squamatus 0 0000000000000000000000 Picus_viridis 0 1111111111001110001111 Pinicola_enucleator 0 0000000001001110001111 Pipra_aureola 0 0000000000000000000000 Pipra_chloromeros 0 1111111101001110001111 Pipra_erythrocephala 0 1111111101001110001111 Pipra_fasciicauda 0 0000000000000000000000 Pipra_filicauda 0 0000000000000000000000 Pipra_rubrocapilla 0 1111111101001110001111 Pipreola_aureopectus 0 0000000000000000000000 Pipreola_chlorolepidota 0 0000000000000000000000 Pipreola_formosa 0 0000000000000000000000 Pipreola_whitelyi 0 0000000000000000000000 Piranga_flava 0 1111111101001110001111 Piranga_ludoviciana 0 0000000000000000000000 Piranga_olivacea 0 1111111101001110001111 Piranga_rubra 0 0000000001001110001111 Platalea_ajaja 0 0000000000000000000000 Ploceus_bicolor 0 0000000000000000000000 Ploceus_capensis 0 0000000000000000000000 Ploceus_cucullatus 0 0000000000000000000000 Ploceus_nelicourvi 0 0000000000000000000000 Ploceus_philippinus 0 0000000000000000000000 Ploceus_sakalava 0 0000000000000000000000 Ploceus_velatus 0 0000000000000000000000 Porphyrolaema_porphyrolaema 0 1111111100000000000000 Procnias_tricarunculatus 0 0000000000000000000000 Psarisomus_dalhousiae 0 0000000000000000000000 Appendix S1 142 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) Meleagris_gallopavo 0 00011000000001010000000 Motacilla_flava 0 00000000000001100000000 Mycerobas_affinis 0 00000000000001100000000 Mycerobas_carnipes 0 00000000000001000000000 Mycerobas_icterioides 0 00000000000001000000000 Mycerobas_melanozanthos 0 00000000000001000000000 Neochmia_ruficauda 0 00000000000001100000000 Neopelma_pallescens 0 00000000000000000000000 Neophron_percnopterus 0 00000000000001100001000 Nesospiza_acunhae 0 00000000000000000000000 Notiomystis_cincta 0 00000000000000011000000 Oriolus_cruentus 0 11010111111000000000000 Oriolus_oriolus 0 00000000000001100000000 Oriolus_traillii 0 11000000000000000000000 Oriolus_xanthornus 0 00000000000001100000000 Paroaria_coronata 0 00110111111000000000000 Parus_major 0 00000000000000100000000 Parus_spilonotus 0 00000000000001100000000 Perdix_perdix 0 00111111111001000000000 Pericrocotus_flammeus 0 11111111111000000000000 Parus_ater 0 00000000000001100000000 Phaethon_rubricauda 0 11000001111000000000000 Phasianus_colchicus 0 00111111111001000000000 Pheucticus_ludovicianus 0 11111111111001000000000 Phibalura_flavirostris 0 00000000000001000000000 Phoenicircus_carnifex 0 00000000000001000100000 Phoenicoparrus_andinus 0 11100111111000000000001 Phoenicopterus_chilensis 0 11100111111000000000000 Phoenicoparrus_jamesi 0 11100111111000000000001 Phoeniconaias_minor 0 11100111111000000000000 Phoenicopterus_roseus 0 11100111111000000000000 Phoenicopterus_ruber 0 11101111111000000000000 Picoides_tridactylus 0 00000000000001100000000 Picoides_villosus 0 11111111111000000000000 Picumnus_exilis 0 00010110000000000000000 Picus_squamatus 0 00000000000000000000000 Picus_viridis 0 11111111111000000000000 Pinicola_enucleator 0 11110111100100000000000 Pipra_aureola 0 00000000000000000000000 Pipra_chloromeros 0 11111111111000000100000 Pipra_erythrocephala 0 11111111111000000000000 Pipra_fasciicauda 0 00000000000000000000000 Pipra_filicauda 0 00000000000000000000000 Pipra_rubrocapilla 0 11111111111000000000000 Pipreola_aureopectus 0 00000000000001100000000 Pipreola_chlorolepidota 0 00000000000001100000000 Pipreola_formosa 0 00000000000001100000000 Pipreola_whitelyi 0 00000000000000000000000 Piranga_flava 0 11111111111000000000000 Piranga_ludoviciana 0 00000000000000000100000 Piranga_olivacea 0 11111111111000000000000 Piranga_rubra 0 11000000000000000000000 Platalea_ajaja 0 00000001111000000000000 Ploceus_bicolor 0 00000000000001100000000 Ploceus_capensis 0 00000000000001100000000 Ploceus_cucullatus 0 00000000000001100000000 Ploceus_nelicourvi 0 00000000000001100000000 Ploceus_philippinus 0 00000000000001100000000 Ploceus_sakalava 0 00000000000001100000000 Ploceus_velatus 0 00000000000001100000000 Porphyrolaema_porphyrolaema 0 00000000000000000000000 Procnias_tricarunculatus 0 00000000000000000000000 Psarisomus_dalhousiae 0 00000000000001100000000 Appendix S1 143 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Meleagris_gallopavo 0 000000000000 Motacilla_flava 0 000000000000 Mycerobas_affinis 0 000000000000 Mycerobas_carnipes 0 000000000000 Mycerobas_icterioides 0 000000000000 Mycerobas_melanozanthos 0 000000000000 Neochmia_ruficauda 0 000000000000 Neopelma_pallescens 0 000000000000 Neophron_percnopterus 0 000000000000 Nesospiza_acunhae 0 000000000000 Notiomystis_cincta 0 000000000000 Oriolus_cruentus 0 000000000000 Oriolus_oriolus 0 000000000001 Oriolus_traillii 0 000000000000 Oriolus_xanthornus 0 000000000000 Paroaria_coronata 0 000000000000 Parus_major 0 000000000000 Parus_spilonotus 0 000000000000 Perdix_perdix 0 000000000000 Pericrocotus_flammeus 0 000000000000 Parus_ater 0 000000000000 Phaethon_rubricauda 0 000000000000 Phasianus_colchicus 0 000000000000 Pheucticus_ludovicianus 0 000000000000 Phibalura_flavirostris 0 000000000000 Phoenicircus_carnifex 0 000000000000 Phoenicoparrus_andinus 0 000000000000 Phoenicopterus_chilensis 0 000000000000 Phoenicoparrus_jamesi 0 000000000000 Phoeniconaias_minor 0 000000000000 Phoenicopterus_roseus 0 000000000000 Phoenicopterus_ruber 0 000000000000 Picoides_tridactylus 0 000000000000 Picoides_villosus 0 000000000000 Picumnus_exilis 0 000000000000 Picus_squamatus 0 000000000000 Picus_viridis 0 000000000000 Pinicola_enucleator 0 000000000000 Pipra_aureola 0 000000000000 Pipra_chloromeros 0 000000000000 Pipra_erythrocephala 0 000000000000 Pipra_fasciicauda 0 000000000000 Pipra_filicauda 0 000000000000 Pipra_rubrocapilla 0 000000000000 Pipreola_aureopectus 0 000000000000 Pipreola_chlorolepidota 0 000000000000 Pipreola_formosa 0 000000000000 Pipreola_whitelyi 0 000000000000 Piranga_flava 0 000000000000 Piranga_ludoviciana 0 000000000000 Piranga_olivacea 0 000000000000 Piranga_rubra 0 000000000000 Platalea_ajaja 0 000000000000 Ploceus_bicolor 0 000000000000 Ploceus_capensis 0 000000000000 Ploceus_cucullatus 0 000000000000 Ploceus_nelicourvi 0 000000000000 Ploceus_philippinus 0 000000000000 Ploceus_sakalava 0 000000000000 Ploceus_velatus 0 000000000000 Porphyrolaema_porphyrolaema 0 000111100000 Procnias_tricarunculatus 0 000000000000 Psarisomus_dalhousiae 0 000000000000 Appendix S1 144 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Psarocolius_montezuma 0110000000000000000 0000 Psarocolius_wagleri 0100000000000000000 0000 Pteroglossus_aracari 0101100000001000000 0000 Ptilinopus_jambu 0100000000000000000 0000 Ptilinopus_magnificus 0100000000000000000 0000 Ptilinopus_pulchellus 0100000000000000000 0000 Ptilinopus_solomonensis 0100000000000000000 0000 Pyroderus_scutatus 0101000011001000000 0000 Pyrrhoplectes_epauletta 0100000000000000100 0000 Pyrrhula_aurantiaca 0100000011000000100 0000 Pyrrhula_erythaca 0101100011000000100 0000 Pyrrhula_erythrocephala 0100000011000000100 0000 Pyrrhula_pyrrhula 0111100000001111000 0010 Quelea_cardinalis 0111100000001000000 0010 Quelea_erythrops 0111100000001000000 0010 Quelea_quelea 0111100000001000000 0010 Querula_purpurata 0110100000001000000 0000 Ramphastos_toco 0100000000001000000 0000 Ramphastos_tucanus 0100100011001000000 0000 Ramphocelus_dimidiatus 0111100000000000000 0010 Regulus_regulus 0110100000001000000 0010 Regulus_satrapa 0111000011001000100 0010 Rhodopechys_obsoletus 0111100000001000000 0010 Rhynchostruthus_socotranus 0100000011000000100 0000 Rupicola_peruvianus 0010000000000000000 0000 Rupicola_rupicola 0101000011001000000 0000 Selenidera_piperivora 0101100000001000000 0000 Serinus_canaria 0100000011000000100 0000 Carduelis_citrinella 0100000011000000100 0000 Serinus_mozambicus 0100000011000000100 0000 Serinus_pusillus 0101000011000000100 0000 Serinus_serinus 0101000011000000100 0000 Setophaga_ruticilla 0101000011000000100 0000 Sicalis_flaveola 0100000000000000000 0000 Snowornis_subalaris 0100000000000000000 0000 Sphyrapicus_varius 0111101100001000000 0010 Sterna_elegans 0010000000000000000 0000 Sturnella_bellicosa 0100100000001000000 0000 Sturnella_magna 0100000000000000000 0000 Sturnella_militaris 0111100000001000000 0010 Sturnella_neglecta 0100000000000000000 0000 Sturnella_superciliaris 0111100000001000000 0010 Taeniopygia_guttata 0111110000001000100 0010 Tarsiger_chrysaeus 0110000000000000100 0000 Telophorus_sulfureopectus 0110000011001000100 0010 Telophorus_zeylonus 0100000011000000100 0000 Tetrao_urogallus 0110000000000000000 0010 Tichodroma_muraria 0010000000000000000 0010 Tijuca_atra 0100000000000000000 0000 Trogon_mesurus 0111100000001000000 0010 Turdus_merula 0111100000000000000 0000 Tyrannus_vociferans 0110000000001000000 0000 Uragus_sibiricus 0110100000001000000 0010 Vermivora_ruficapilla 0100000000000000000 0000 Vermivora_virginiae 0100000000000000000 0000 Vestiaria_coccinea 0111100000001000000 0010 Xanthocephalus_xanthocephalus 0100000000000000000 0000 Xipholena_atropurpurea 0111100000001000000 0000 Xipholena_lamellipennis 0111100000001000000 0010 Xipholena_punicea 0111100000001000000 0010 Zosterops_japonicus 0100000000000000000 0000 node1 90.758000000000000000000 0000 node2 76.397010000000000000000 0000 Appendix S1 145 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) Psarocolius_montezuma 0 0000000000000000000000 Psarocolius_wagleri 0 0000000000000000000000 Pteroglossus_aracari 0 0000111110100000000100 Ptilinopus_jambu 0 0000000000000000000000 Ptilinopus_magnificus 0 0000000000000000000000 Ptilinopus_pulchellus 0 0000000000000000000000 Ptilinopus_solomonensis 0 0000000000000000000000 Pyroderus_scutatus 0 0000101010100000000100 Pyrrhoplectes_epauletta 0 0000000000000000000000 Pyrrhula_aurantiaca 0 0000000000000000000000 Pyrrhula_erythaca 0 0000111110100000000000 Pyrrhula_erythrocephala 0 0000000000000000000000 Pyrrhula_pyrrhula 0 0111111110100000000100 Quelea_cardinalis 0 0111111110100000000100 Quelea_erythrops 0 0111111110100000000100 Quelea_quelea 0 0111111110100000000100 Querula_purpurata 0 0111010100000000000100 Ramphastos_toco 0 0000000000000000000100 Ramphastos_tucanus 0 0000010100000000000100 Ramphocelus_dimidiatus 0 0111111110100000000000 Regulus_regulus 0 0111010100000000000100 Regulus_satrapa 0 0111111110100000000100 Rhodopechys_obsoletus 0 0111111110100000000100 Rhynchostruthus_socotranus 0 0000000000000000000000 Rupicola_peruvianus 0 0110000000000000000000 Rupicola_rupicola 0 0000101010100000000100 Selenidera_piperivora 0 0000111110100000000100 Serinus_canaria 0 0000000000000000000000 Carduelis_citrinella 0 0000000000000000000000 Serinus_mozambicus 0 0000000000000000000000 Serinus_pusillus00000101010100000000000 Serinus_serinus 0 0000101010100000000000 Setophaga_ruticilla 0 0000101010100000000000 Sicalis_flaveola 0 0000000000000000000000 Snowornis_subalaris 0 0000000000000000000000 Sphyrapicus_varius 0 0111111110100000000100 Sterna_elegans 0 0011101111000000000000 Sturnella_bellicosa 0 0000010100000000000100 Sturnella_magna 0 0000000000000000000000 Sturnella_militaris 0 0111111110100000000100 Sturnella_neglecta 0 0000000000000000000000 Sturnella_superciliaris 0 0111111110100000000100 Taeniopygia_guttata 0 0111111110100000000100 Tarsiger_chrysaeus 0 0000000000000000000000 Telophorus_sulfureopectus 0 0111000000000000000100 Telophorus_zeylonus 0 0000000000000000000000 Tetrao_urogallus 0 0111000000000000000000 Tichodroma_muraria 0 0111000000000000000000 Tijuca_atra 0 0000000000000000000000 Trogon_mesurus 0 0111111110100000000100 Turdus_merula 0 0000000000010000000000 Tyrannus_vociferans 0 0000000000000000000100 Uragus_sibiricus 0 0111010100000001100100 Vermivora_ruficapilla 0 0000000000000000000000 Vermivora_virginiae 0 0000000000000000000000 Vestiaria_coccinea 0 0111111110100000000100 Xanthocephalus_xanthocephalus 0 0000000000000000000000 Xipholena_atropurpurea 0 0111111110100000000100 Xipholena_lamellipennis 0 0111111110100000000100 Xipholena_punicea 0 0111111110100000000100 Zosterops_japonicus 0 0000000000000000000000 node1 90.758 0000000000000000000000 node2 76.397 0000000000000000000000 Appendix S1 146 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) Psarocolius_montezuma 0 000 00000 0 0000000000000 Psarocolius_wagleri 0 000 00000 0 0000000000000 Pteroglossus_aracari 0 000 00000 0 0000000000000 Ptilinopus_jambu 0 000 00000 0 0000000000000 Ptilinopus_magnificus 0 000 00000 0 0000000000000 Ptilinopus_pulchellus 0 000 00000 0 0000000000000 Ptilinopus_solomonensis 0 000 00000 0 0000000000000 Pyroderus_scutatus 0 000 00000 0 0001111000000 Pyrrhoplectes_epauletta 0 000 00000 0 0000000100000 Pyrrhula_aurantiaca 0 000 00000 0 0001111101000 Pyrrhula_erythaca 0 000 00000 0 0001111101000 Pyrrhula_erythrocephala 0 000 00000 0 0001111101000 Pyrrhula_pyrrhula 0 000 00000 0 0000000000000 Quelea_cardinalis 0 000 00000 0 0000000000000 Quelea_erythrops 0 000 00000 0 0000000000000 Quelea_quelea 0 000 00000 0 0000000000000 Querula_purpurata 0 110 11001 0 0000000000000 Ramphastos_toco 0 000 00000 0 0000000000000 Ramphastos_tucanus 0 000 00000 0 0001111000000 Ramphocelus_dimidiatus 0 000 00000 0 0000000000000 Regulus_regulus 0 000 00000 0 0000000000000 Regulus_satrapa 0 000 00000 0 0001111111000 Rhodopechys_obsoletus 0 000 00000 0 0000000000000 Rhynchostruthus_socotranus 0 000 00000 0 0001111101000 Rupicola_peruvianus 0 001 00000 0 0000000000000 Rupicola_rupicola 0 100 00000 0 0001111000000 Selenidera_piperivora 0 000 00000 0 0000000000000 Serinus_canaria 0 000 00000 0 0001111101000 Carduelis_citrinella 0 000 00000 0 0001111101000 Serinus_mozambicus 0 000 00000 0 0001111101000 Serinus_pusillus0000 00000 0 0001111101000 Serinus_serinus 0 000 00000 0 0001111101000 Setophaga_ruticilla 0 000 00000 0 0001111101000 Sicalis_flaveola 0 000 00000 0 0000000000000 Snowornis_subalaris 0 000 00000 0 0000000000000 Sphyrapicus_varius 0 000 00000 0 0000000000110 Sterna_elegans 0 000 00000 0 0000000000000 Sturnella_bellicosa 0 000 00000 0 0000000000000 Sturnella_magna 0 000 00000 0 0000000000000 Sturnella_militaris 0 000 00000 0 0000000000000 Sturnella_neglecta 0 000 00000 0 0000000000000 Sturnella_superciliaris 0 000 00000 0 0000000000000 Taeniopygia_guttata 0 000 00000 0 0010000110000 Tarsiger_chrysaeus 0 000 00000 0 0000000110000 Telophorus_sulfureopectus 0 000 00000 0 0001111111000 Telophorus_zeylonus 0 000 00000 0 0001111101000 Tetrao_urogallus 0 000 00000 0 0000000000000 Tichodroma_muraria 0 000 00000 0 0000000000000 Tijuca_atra 0 000 00000 0 0000000000000 Trogon_mesurus 0 000 00000 0 0000000000000 Turdus_merula 0 000 00000 0 0000000000000 Tyrannus_vociferans 0 000 00000 0 0000000000000 Uragus_sibiricus 0 000 00000 0 0000000000000 Vermivora_ruficapilla 0 000 00000 0 0000000000000 Vermivora_virginiae 0 000 00000 0 0000000000000 Vestiaria_coccinea 0 000 00000 0 0000000000000 Xanthocephalus_xanthocephalus 0 000 00000 0 0000000000000 Xipholena_atropurpurea 0 110 11101 0 0000000000000 Xipholena_lamellipennis 0 100 11101 0 0000000000000 Xipholena_punicea 0 110 11111 0 0000000000000 Zosterops_japonicus 0 000 00000 0 0000000000000 node1 90.758 000 00000 0 0000000000000 node2 76.397 000 00000 0 0000000000000 Appendix S1 147 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Psarocolius_montezuma 0 00000000000000000000000 Psarocolius_wagleri 0 00000000000000000000000 Pteroglossus_aracari 0 01001000000000000000000 Ptilinopus_jambu 0 00000000000000000000000 Ptilinopus_magnificus 0 00000000000000000000000 Ptilinopus_pulchellus 0 00000000000000000000000 Ptilinopus_solomonensis 0 00000000000000000000000 Pyroderus_scutatus 0 01001000000000000000000 Pyrrhoplectes_epauletta 0 00000000000000000000000 Pyrrhula_aurantiaca 0 00000000000000000000000 Pyrrhula_erythaca 0 00000000000000000000000 Pyrrhula_erythrocephala 0 00000000000000000000000 Pyrrhula_pyrrhula 0 01111111100000000000011 Quelea_cardinalis 0 01001000100000000000011 Quelea_erythrops 0 01001000100000000000011 Quelea_quelea 0 01001000100000000000011 Querula_purpurata 0 01001000000000000000011 Ramphastos_toco 0 01001000000000000000000 Ramphastos_tucanus 0 01001000000000000000000 Ramphocelus_dimidiatus 0 00000000000000000000011 Regulus_regulus 0 01001000100000000000011 Regulus_satrapa 0 01001000100000001000011 Rhodopechys_obsoletus 0 01001000100000000000011 Rhynchostruthus_socotranus 0 00000000000000000000000 Rupicola_peruvianus 0 00000000000000000000011 Rupicola_rupicola 0 01001000000000000000000 Selenidera_piperivora 0 01001000000000000000000 Serinus_canaria 0 00000000000000000000000 Carduelis_citrinella 0 00000000000000000000000 Serinus_mozambicus 0 00000000000000000000000 Serinus_pusillus000000000000000000000000 Serinus_serinus 0 00000000000000000000000 Setophaga_ruticilla 0 00000000000000000000000 Sicalis_flaveola 0 00000000000000000000000 Snowornis_subalaris 0 00000000000000000000000 Sphyrapicus_varius 0 01001000100000000000011 Sterna_elegans 0 00000000000000000000000 Sturnella_bellicosa 0 01001000000000000000000 Sturnella_magna 0 00000000000000000000000 Sturnella_militaris 0 01001000100000000000011 Sturnella_neglecta 0 00000000000000000000000 Sturnella_superciliaris 0 01001000100000000000011 Taeniopygia_guttata 0 01001000100000001000011 Tarsiger_chrysaeus 0 00000000000000001000000 Telophorus_sulfureopectus 0 01001000100000001000011 Telophorus_zeylonus 0 00000000000000000000000 Tetrao_urogallus 0 00000000000000000000011 Tichodroma_muraria 0 00000000000000000000011 Tijuca_atra 0 00000000000000000000000 Trogon_mesurus 0 01001000100000000000011 Turdus_merula 0 00000000000000000000000 Tyrannus_vociferans 0 01001000000000000000000 Uragus_sibiricus 0 01001000100000000000011 Vermivora_ruficapilla 0 00000000000000000000000 Vermivora_virginiae 0 00000000000000000000000 Vestiaria_coccinea 0 01001000100000000000011 Xanthocephalus_xanthocephalus 0 00000000000000000000000 Xipholena_atropurpurea 0 01001000000000000000011 Xipholena_lamellipennis 0 01001000000000000000011 Xipholena_punicea 0 01001000000000000000011 Zosterops_japonicus 0 00000000000000000000000 node1 90.758 00000000000000000000000 node2 76.397 00000000000000000000000 Appendix S1 148 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Psarocolius_montezuma 0 0000000000000000000000 Psarocolius_wagleri 0 0000000000000000000000 Pteroglossus_aracari 0 0000000001001110001111 Ptilinopus_jambu 0 0000000000000000000000 Ptilinopus_magnificus 0 0000000000000000000000 Ptilinopus_pulchellus 0 0000000000000000000000 Ptilinopus_solomonensis 0 0000000000000000000000 Pyroderus_scutatus 0 0000000001001110001111 Pyrrhoplectes_epauletta 0 0000000000000000000000 Pyrrhula_aurantiaca 0 0000000000000000000000 Pyrrhula_erythaca 0 0000000001001110001111 Pyrrhula_erythrocephala 0 0000000000000000000000 Pyrrhula_pyrrhula 0 1111111101001110001111 Quelea_cardinalis 0 1111111101001110001111 Quelea_erythrops 0 1111111101001110001111 Quelea_quelea 0 1111111101001110001111 Querula_purpurata 0 1001100100000000000000 Ramphastos_toco 0 0000000000000000000000 Ramphastos_tucanus 0 0000000000000000000000 Ramphocelus_dimidiatus 0 1111111101001110001111 Regulus_regulus 0 1111111100000000000000 Regulus_satrapa 0 1111111101001110001111 Rhodopechys_obsoletus 0 1111111101001110001111 Rhynchostruthus_socotranus 0 0000000000000000000000 Rupicola_peruvianus 0 1000000100000000000000 Rupicola_rupicola 0 0000000001001110001111 Selenidera_piperivora 0 0000000001001110001111 Serinus_canaria 0 0000000000000000000000 Carduelis_citrinella 0 0000000000000000000000 Serinus_mozambicus 0 0000000000000000000000 Serinus_pusillus00000000001001110001111 Serinus_serinus 0 0000000001001110001111 Setophaga_ruticilla 0 0000000001001110001111 Sicalis_flaveola 0 0000000000000000000000 Snowornis_subalaris 0 0000000000000000000000 Sphyrapicus_varius 0 1111111101001110001111 Sterna_elegans 0 0001100100000001111111 Sturnella_bellicosa 0 0000000000000000000000 Sturnella_magna 0 0000000000000000000000 Sturnella_militaris 0 1111111101001110001111 Sturnella_neglecta 0 0000000000000000000000 Sturnella_superciliaris 0 1111111101001110001111 Taeniopygia_guttata 0 1111111101001110001111 Tarsiger_chrysaeus 0 0000000000000000000000 Telophorus_sulfureopectus 0 1111111100000000000000 Telophorus_zeylonus 0 0000000000000000000000 Tetrao_urogallus 0 1111111100000000000000 Tichodroma_muraria 0 1111111100000000000000 Tijuca_atra 0 0000000000000000000000 Trogon_mesurus 0 1111111101001110001111 Turdus_merula 0 0000000000000000000000 Tyrannus_vociferans 0 0000000000000000000000 Uragus_sibiricus 0 1111111100000000000000 Vermivora_ruficapilla 0 0000000000000000000000 Vermivora_virginiae 0 0000000000000000000000 Vestiaria_coccinea 0 1111111101001110001111 Xanthocephalus_xanthocephalus 0 0000000000000000000000 Xipholena_atropurpurea 0 1001100101001110001111 Xipholena_lamellipennis 0 1111111101001110001111 Xipholena_punicea 0 1111111101001110001111 Zosterops_japonicus 0 0000000000000000000000 node1 90.758 0000000000000000000000 node2 76.397 0000000000000000000000 Appendix S1 149 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) Psarocolius_montezuma 0 00000000000001100000000 Psarocolius_wagleri 0 00000000000001000000000 Pteroglossus_aracari 0 11010111100000000000000 Ptilinopus_jambu 0 00000000000001000000000 Ptilinopus_magnificus 0 00000000000001000000000 Ptilinopus_pulchellus 0 00000000000001000000000 Ptilinopus_solomonensis 0 00000000000001000000000 Pyroderus_scutatus 0 11000000000000000000000 Pyrrhoplectes_epauletta 0 00000000000000000000000 Pyrrhula_aurantiaca 0 00000000000000000000000 Pyrrhula_erythaca 0 11110111100000000000000 Pyrrhula_erythrocephala 0 00000000000000000000000 Pyrrhula_pyrrhula 0 11111111111000000000000 Quelea_cardinalis 0 11111111111000000000000 Quelea_erythrops 0 11111111111000000000000 Quelea_quelea 0 11111111111000000000000 Querula_purpurata 0 00010110000000000000000 Ramphastos_toco 0 00000000000000000000000 Ramphastos_tucanus 0 00010110000000000000000 Ramphocelus_dimidiatus 0 11111111111001000000000 Regulus_regulus 0 00011110011000000000000 Regulus_satrapa 0 11101111111000000000000 Rhodopechys_obsoletus 0 11111111111000000000000 Rhynchostruthus_socotranus 0 00000000000000000000000 Rupicola_peruvianus 0 00000000000000000000000 Rupicola_rupicola 0 11000000000000000000000 Selenidera_piperivora 0 11010111100000000000000 Serinus_canaria 0 00000000000000000000000 Carduelis_citrinella 0 00000000000000000000000 Serinus_mozambicus 0 00000000000000000000000 Serinus_pusillus011000000000000000000000 Serinus_serinus 0 11000000000000000000000 Setophaga_ruticilla 0 11000000000000000000000 Sicalis_flaveola 0 00000000000001000000000 Snowornis_subalaris 0 00000000000001000000000 Sphyrapicus_varius 0 11111111111000000000000 Sterna_elegans 0 11000001111000000000000 Sturnella_bellicosa 0 00010110000000000000000 Sturnella_magna 0 00000000000001000000000 Sturnella_militaris 0 11111111111000000000000 Sturnella_neglecta 0 00000000000001000000000 Sturnella_superciliaris 0 11111111111000000000000 Taeniopygia_guttata 0 11111111111000000000000 Tarsiger_chrysaeus 0 00000000000000000000000 Telophorus_sulfureopectus 0 00000000000000000000000 Telophorus_zeylonus 0 00000000000000000000000 Tetrao_urogallus 0 00000000000001000000000 Tichodroma_muraria 0 00000000000000000000000 Tijuca_atra 0 00000000000001000000000 Trogon_mesurus 0 11111111111000000000000 Turdus_merula 0 00000000000001111010000 Tyrannus_vociferans 0 00000000000000100000000 Uragus_sibiricus 0 00011110011100000000000 Vermivora_ruficapilla 0 00000000000001000000000 Vermivora_virginiae 0 00000000000001000000000 Vestiaria_coccinea 0 11110111111000000000000 Xanthocephalus_xanthocephalus 0 00000000000001000000000 Xipholena_atropurpurea 0 11010111100000000000000 Xipholena_lamellipennis 0 11110111100000000000000 Xipholena_punicea 0 11010111111000000000000 Zosterops_japonicus 0 00000000000001000000000 node1 90.758 00000000000000000000000 node2 76.397 00000000000000000000000 Appendix S1 150 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) Psarocolius_montezuma 0 000000000000 Psarocolius_wagleri 0 000000000000 Pteroglossus_aracari 0 000000000000 Ptilinopus_jambu 0 000000000000 Ptilinopus_magnificus 0 000000000000 Ptilinopus_pulchellus 0 000000000000 Ptilinopus_solomonensis 0 000000000000 Pyroderus_scutatus 0 000000000000 Pyrrhoplectes_epauletta 0 000000000000 Pyrrhula_aurantiaca 0 000000000000 Pyrrhula_erythaca 0 000000000000 Pyrrhula_erythrocephala 0 000000000000 Pyrrhula_pyrrhula 0 000000000000 Quelea_cardinalis 0 000000000000 Quelea_erythrops 0 000000000000 Quelea_quelea 0 000000010000 Querula_purpurata 0 110110000000 Ramphastos_toco 0 000000000000 Ramphastos_tucanus 0 000000000000 Ramphocelus_dimidiatus 0 000000000000 Regulus_regulus 0 000000000000 Regulus_satrapa 0 000000000000 Rhodopechys_obsoletus 0 000000000000 Rhynchostruthus_socotranus 0 000000000000 Rupicola_peruvianus 0 001000000000 Rupicola_rupicola 0 100000000000 Selenidera_piperivora 0 000000000000 Serinus_canaria 0 000000000000 Carduelis_citrinella 0 000000000000 Serinus_mozambicus 0 000000000000 Serinus_pusillus0000000000000 Serinus_serinus 0 000000000000 Setophaga_ruticilla 0 000000000000 Sicalis_flaveola 0 000000000000 Snowornis_subalaris 0 000000000000 Sphyrapicus_varius 0 000000000000 Sterna_elegans 0 000000000000 Sturnella_bellicosa 0 000000000000 Sturnella_magna 0 000000000000 Sturnella_militaris 0 000000000000 Sturnella_neglecta 0 000000000000 Sturnella_superciliaris 0 000000000000 Taeniopygia_guttata 0 000000000000 Tarsiger_chrysaeus 0 000000000000 Telophorus_sulfureopectus 0 000000000000 Telophorus_zeylonus 0 000000000000 Tetrao_urogallus 0 000000000000 Tichodroma_muraria 0 000000000000 Tijuca_atra 0 000000000000 Trogon_mesurus 0 000000000000 Turdus_merula 0 000000000000 Tyrannus_vociferans 0 000000000000 Uragus_sibiricus 0 000000000000 Vermivora_ruficapilla 0 000000000000 Vermivora_virginiae 0 000000000000 Vestiaria_coccinea 0 000000000000 Xanthocephalus_xanthocephalus 0 000000010000 Xipholena_atropurpurea 0 110111010000 Xipholena_lamellipennis 0 100111010000 Xipholena_punicea 0 110111100000 Zosterops_japonicus 0 000000000000 node1 90.758 000000000000 node2 76.397 000000000000 Appendix S1 151 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node3 75.8000000000000000000 0000 node4 74.145100000000000000000 0000 node5 61.351110000000000000000 0000 node6 49.072110000000000000100 0000 node7 46.378100000000000000000 0000 node8 41.223100000000000000000 0000 node9 40.343100000000000000000 0000 node10 32.912100000000000000100 0000 node11 28.346100000000000000100 0000 node12 25.449100000000000000100 0000 node13 20.406100000000000000000 0000 node14 17.934100000000000000000 0000 node15 15.463100000000000000000 0000 node16 8.359100000000000000000 0000 node17 4.714100000000000000000 0000 node18 4.5100000000000000000 0000 node19 3.904100000000000000000 0000 node20 8.081000000000000000000 0000 node21 0.853100000000000000000 0000 node22 13.933100000000000000000 0000 node23 13.088100000000000000000 0000 node24 11.667100000000000000000 0000 node25 9.348100000000000000000 0000 node26 4.092100000000000000000 0000 node27 2.501100000000000000000 0000 node28 3.634100100000001000000 0000 node29 12.229100000000000000000 0000 node30 7.153100000000000000000 0000 node31 4.573100000000000000000 0000 node32 6.823100000000000000000 0000 node33 5.524100000000000000000 0000 node34 4.545100000000000000000 0000 node35 4.928100000000000000000 0000 node36 11.656000000000000000000 0000 node37 7.695100000000000000000 0000 node38 5.239100000000000000000 0000 node39 2.614111000011000000100 0000 node40 1.396100000000000000000 0000 node41 2.316101000011000000100 0000 node42 1.661101000011000000100 0000 node43 6.746100000000000000000 0000 node44 5.427100000000000000000 0000 node45 4.802100000000000000000 0000 node46 4.508100000000000000000 0000 node47 4.044100000000000000000 0000 node48 13.933000000000000000000 0000 node49 11.281110000000000000000 0000 node50 17.407111100011000000100 0010 node51 15.914100100000000000000 0000 node52 13.848000000000000000000 0000 node53 5.815111100000001000000 0010 node54 7.054101000011000000100 0000 node55 6.145100000011000000100 0000 node56 2.607100000011000000100 0000 node57 13.606100000011000000000 0000 node58 12.752100000011000000000 0000 node59 10.49100000000000000000 0000 node60 11.329100000000000000000 0000 node61 23.82100000000000000000 0000 node62 7.272100000000000000000 0000 node63 21.758100000000000000000 0000 node64 18.673100000000000000000 0000 node65 14.098100000000000000000 0000 Appendix S1 152 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) node3 75.8 0000000000000000000000 node4 74.145 0000000000000000000000 node5 61.351 0000000000000000000000 node6 49.072 0000000000000000000000 node7 46.378 0000000000000000000000 node8 41.223 0000000000000000000000 node9 40.343 0000000000000000000000 node10 32.912 0000000000000000000000 node11 28.346 0000000000000000000000 node12 25.449 0000000000000000000000 node13 20.406 0000000000000000000000 node14 17.934 0000000000000000000000 node15 15.463 0000000000000000000000 node16 8.359 0000000000000000000000 node17 4.714 0000000000000000000000 node18 4.5 0000000000000000000000 node19 3.904 0000000000000000000000 node20 8.081 0000000000000000000000 node21 0.853 0000000000000000000000 node22 13.933 0000000000000000000000 node23 13.088 0000000000000000000000 node24 11.667 0000000000000000000000 node25 9.348 0000000000000000000000 node26 4.092 0000000000000000000000 node27 2.501 0000000000000000000000 node28 3.634 0000010100000000000100 node29 12.229 0000000000000000000000 node30 7.153 0000000000000000000000 node31 4.573 0000000000000000000000 node32 6.823 0000000000000000000000 node33 5.524 0000000000000000000000 node34 4.545 0000000000000000000000 node35 4.928 0000000000000000000000 node36 11.656 0000000000000000000000 node37 7.695 0000000000000000000000 node38 5.239 0000000000000000000000 node39 2.614 0000111110100000000000 node40 1.396 0000000000000000000000 node41 2.316 0000101010100000000000 node42 1.661 0000101010100000000000 node43 6.746 0000000000000000000000 node44 5.427 0000000000000000000000 node45 4.802 0000000000000000000000 node46 4.508 0000000000000000000000 node47 4.044 0000000000000000000000 node48 13.933 0000000000000000000000 node49 11.281 0000000000000000000000 node50 17.407 0111110100000000000000 node51 15.914 0001010100000000000000 node52 13.848 0000000000000000000000 node53 5.815 0111111110100000000100 node54 7.054 0000101010100000000000 node55 6.145 0000000000000000000000 node56 2.607 0000000000000000000000 node57 13.606 0000000000000000000000 node58 12.752 0000000000000000000000 node59 10.49 0000000000000000000000 node60 11.329 0000000000000000000000 node61 23.82 0000000000000000000000 node62 7.272 0000000000000000000000 node63 21.758 0000000000000000000000 node64 18.673 0000000000000000000000 node65 14.098 0000000000000000000000 Appendix S1 153 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) node3 75.8 000 00000 0 0000000000000 node4 74.145 000 00000 0 0000000000000 node5 61.351 000 00000 0 0000000000000 node6 49.072 000 00000 0 0000000100000 node7 46.378 000 00000 0 0000000000000 node8 41.223 000 00000 0 0000000000000 node9 40.343 000 00000 0 0000000000000 node10 32.912 000 00000 0 0000000100000 node11 28.346 000 00000 0 0000000100000 node12 25.449 000 00000 0 0000000100000 node13 20.406 000 00000 0 0000000000000 node14 17.934 000 00000 0 0000000000000 node15 15.463 000 00000 0 0000000000000 node16 8.359 000 00000 0 0000000000000 node17 4.714 000 00000 0 0000000000000 node18 4.5 000 00000 0 0000000000000 node19 3.904 000 00000 0 0000000000000 node20 8.081 000 00000 0 0000000000000 node21 0.853 000 00000 0 0000000000000 node22 13.933 000 00000 0 0000000000000 node23 13.088 000 00000 0 0000000000000 node24 11.667 000 00000 0 0000000000000 node25 9.348 000 00000 0 0000000000000 node26 4.092 000 00000 0 0000000000000 node27 2.501 000 00000 0 0000000000000 node28 3.634 000 00000 0 0000000000000 node29 12.229 000 00000 0 0000000000000 node30 7.153 000 00000 0 0000000000000 node31 4.573 000 00000 0 0000000000000 node32 6.823 000 00000 0 0000000000000 node33 5.524 000 00000 0 0000000000000 node34 4.545 000 00000 0 0000000000000 node35 4.928 000 00000 0 0000000000000 node36 11.656 000 00000 0 0000000000000 node37 7.695 000 00000 0 0000000000000 node38 5.239 000 00000 0 0000000000000 node39 2.614 000 00000 0 0001111111000 node40 1.396 000 00000 0 0000000000000 node41 2.316 000 00000 0 0001111101000 node42 1.661 000 00000 0 0001111101000 node43 6.746 000 00000 0 0000000000000 node44 5.427 000 00000 0 0000000000000 node45 4.802 000 00000 0 0000000000000 node46 4.508 000 00000 0 0000000000000 node47 4.044 000 00000 0 0000000000000 node48 13.933 000 00000 0 0000000000000 node49 11.281 000 00000 0 0000000000000 node50 17.407 000 00000 0 0001111101000 node51 15.914 000 00000 0 0000000000000 node52 13.848 000 00000 0 0000000000000 node53 5.815 000 00000 0 0000000000000 node54 7.054 000 00000 0 0001111101000 node55 6.145 000 00000 0 0001111101000 node56 2.607 000 00000 0 0001111101000 node57 13.606 000 00000 0 0001111000000 node58 12.752 000 00000 0 0001111000000 node59 10.49 000 00000 0 0000000000000 node60 11.329 000 00000 0 0000000000000 node61 23.82 000 00000 0 0000000000000 node62 7.272 000 00000 0 0000000000000 node63 21.758 000 00000 0 0000000000000 node64 18.673 000 00000 0 0000000000000 node65 14.098 000 00000 0 0000000000000 Appendix S1 154 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node3 75.8 00000000000000000000000 node4 74.145 00000000000000000000000 node5 61.351 00000000000000000000000 node6 49.072 00000000000000001000000 node7 46.378 00000000000000000000000 node8 41.223 00000000000000000000000 node9 40.343 00000000000000000000000 node10 32.912 00000000000000000000000 node11 28.346 00000000000000000000000 node12 25.449 00000000000000000000000 node13 20.406 00000000000000000000000 node14 17.934 00000000000000000000000 node15 15.463 00000000000000000000000 node16 8.359 00000000000000000000000 node17 4.714 00000000000000000000000 node18 4.5 00000000000000000000000 node19 3.904 00000000000000000000000 node20 8.081 00000000000000000000000 node21 0.853 00000000000000000000000 node22 13.933 00000000000000000000000 node23 13.088 00000000000000000000000 node24 11.667 00000000000000000000000 node25 9.348 00000000000000000000000 node26 4.092 00000000000000000000000 node27 2.501 00000000000000000000000 node28 3.634 01001000000000000000000 node29 12.229 00000000000000000000000 node30 7.153 00000000000000000000000 node31 4.573 00000000000000000000000 node32 6.823 00000000000000000000000 node33 5.524 00000000000000000000000 node34 4.545 00000000000000000000000 node35 4.928 00000000000000000000000 node36 11.656 00000000000000000000000 node37 7.695 00000000000000000000000 node38 5.239 00000000000000000000000 node39 2.614 00000000000000001000000 node40 1.396 00000000000000000000000 node41 2.316 00000000000000000000000 node42 1.661 00000000000000000000000 node43 6.746 00000000000000000000000 node44 5.427 00000000000000000000000 node45 4.802 00000000000000000000000 node46 4.508 00000000000000000000000 node47 4.044 00000000000000000000000 node48 13.933 00000000000000000000000 node49 11.281 00000000000000000000000 node50 17.407 00000000000000001000011 node51 15.914 00000000000000000000000 node52 13.848 00000000000000000000000 node53 5.815 01001000100000000000011 node54 7.054 00000000000000000000000 node55 6.145 00000000000000000000000 node56 2.607 00000000000000000000000 node57 13.606 00000000000000000000000 node58 12.752 00000000000000000000000 node59 10.49 00000000000000000000000 node60 11.329 00000000000000000000000 node61 23.82 00000000000000000000000 node62 7.272 00000000000000000000000 node63 21.758 00000000000000000000000 node64 18.673 00000000000000000000000 node65 14.098 00000000000000000000000 Appendix S1 155 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node3 75.8 0000000000000000000000 node4 74.145 0000000000000000000000 node5 61.351 0000000000000000000000 node6 49.072 0000000000000000000000 node7 46.378 0000000000000000000000 node8 41.223 0000000000000000000000 node9 40.343 0000000000000000000000 node10 32.912 0000000000000000000000 node11 28.346 0000000000000000000000 node12 25.449 0000000000000000000000 node13 20.406 0000000000000000000000 node14 17.934 0000000000000000000000 node15 15.463 0000000000000000000000 node16 8.359 0000000000000000000000 node17 4.714 0000000000000000000000 node18 4.5 0000000000000000000000 node19 3.904 0000000000000000000000 node20 8.081 0000000000000000000000 node21 0.853 0000000000000000000000 node22 13.933 0000000000000000000000 node23 13.088 0000000000000000000000 node24 11.667 0000000000000000000000 node25 9.348 0000000000000000000000 node26 4.092 0000000000000000000000 node27 2.501 0000000000000000000000 node28 3.634 0000000000000000000000 node29 12.229 0000000000000000000000 node30 7.153 0000000000000000000000 node31 4.573 0000000000000000000000 node32 6.823 0000000000000000000000 node33 5.524 0000000000000000000000 node34 4.545 0000000000000000000000 node35 4.928 0000000000000000000000 node36 11.656 0000000000000000000000 node37 7.695 0000000000000000000000 node38 5.239 0000000000000000000000 node39 2.614 0000000001001110001111 node40 1.396 0000000000000000000000 node41 2.316 0000000001001110001111 node42 1.661 0000000001001110001111 node43 6.746 0000000000000000000000 node44 5.427 0000000000000000000000 node45 4.802 0000000000000000000000 node46 4.508 0000000000000000000000 node47 4.044 0000000000000000000000 node48 13.933 0000000000000000000000 node49 11.281 0000000000000000000000 node50 17.407 1111111100000010000000 node51 15.914 0000000000000000000000 node52 13.848 0000000000000000000000 node53 5.815 1111111101001110001111 node54 7.054 0000000001001110001111 node55 6.145 0000000000000000000000 node56 2.607 0000000000000000000000 node57 13.606 0000000000000000000000 node58 12.752 0000000000000000000000 node59 10.49 0000000000000000000000 node60 11.329 0000000000000000000000 node61 23.82 0000000000000000000000 node62 7.272 0000000000000000000000 node63 21.758 0000000000000000000000 node64 18.673 0000000000000000000000 node65 14.098 0000000000000000000000 Appendix S1 156 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) node3 75.8 00000000000000000000000 node4 74.145 00000000000000000000000 node5 61.351 00000000000000000000000 node6 49.072 00000000000000000000000 node7 46.378 00000000000000000000000 node8 41.223 00000000000000000000000 node9 40.343 00000000000000000000000 node10 32.912 00000000000000000000000 node11 28.346 00000000000000000000000 node12 25.449 00000000000000000000000 node13 20.406 00000000000000000000000 node14 17.934 00000000000001000000000 node15 15.463 00000000000001000000000 node16 8.359 00000000000001000000000 node17 4.714 00000000000001000000000 node18 4.5 00000000000000000000000 node19 3.904 00000000000001000000000 node20 8.081 00000000000000000000000 node21 0.853 00000000000001000000000 node22 13.933 00000000000001000000000 node23 13.088 00000000000001000000000 node24 11.667 00000000000000000000000 node25 9.348 00000000000000000000000 node26 4.092 00000000000000000000000 node27 2.501 00000000000001000000000 node28 3.634 00010110000000000000000 node29 12.229 00000000000001000000000 node30 7.153 00000000000001000000000 node31 4.573 00000000000001000000000 node32 6.823 00000000000000000000000 node33 5.524 00000000000000000000000 node34 4.545 00000000000000000000000 node35 4.928 00000000000000000000000 node36 11.656 00000000000000000000000 node37 7.695 00000000000001000000000 node38 5.239 00000000000000000000000 node39 2.614 11100111100000000000000 node40 1.396 00000000000000000000000 node41 2.316 11000000000000000000000 node42 1.661 11000000000000000000000 node43 6.746 00000000000001000000000 node44 5.427 00000000000000000000000 node45 4.802 00000000000000000000000 node46 4.508 00000000000001000000000 node47 4.044 00000000000001000000000 node48 13.933 00000000000000000000000 node49 11.281 00000000000001100000000 node50 17.407 00110110011000000000000 node51 15.914 00010110011001000000000 node52 13.848 00000000000000000000000 node53 5.815 11110111111000000000000 node54 7.054 11000000000000000000000 node55 6.145 00000000000000000000000 node56 2.607 00000000000000000000000 node57 13.606 00000000000000000000000 node58 12.752 00000000000000000000000 node59 10.49 00000000000000000000000 node60 11.329 00000000000001000000000 node61 23.82 00000000000000000000000 node62 7.272 00000000000000000000000 node63 21.758 00000000000000000000000 node64 18.673 00000000000000000000000 node65 14.098 00000000000001000000000 Appendix S1 157 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node3 75.8 000000000000 node4 74.145 000000000000 node5 61.351 000000000000 node6 49.072 000000000000 node7 46.378 000000000000 node8 41.223 000000000000 node9 40.343 000000000000 node10 32.912 000000000000 node11 28.346 000000000000 node12 25.449 000000000000 node13 20.406 000000000000 node14 17.934 000000000000 node15 15.463 000000000000 node16 8.359 000000000000 node17 4.714 000000000000 node18 4.5 000000000000 node19 3.904 000000000000 node20 8.081 000000000000 node21 0.853 000000000000 node22 13.933 000000000000 node23 13.088 000000000000 node24 11.667 000000000000 node25 9.348 000000000000 node26 4.092 000000000000 node27 2.501 000000000000 node28 3.634 000000000000 node29 12.229 000000000000 node30 7.153 000000000000 node31 4.573 000000000000 node32 6.823 000000000000 node33 5.524 000000000000 node34 4.545 000000000000 node35 4.928 000000000000 node36 11.656 000000000000 node37 7.695 000000000000 node38 5.239 000000000000 node39 2.614 000000000000 node40 1.396 000000000000 node41 2.316 000000000000 node42 1.661 000000000000 node43 6.746 000000000000 node44 5.427 000000000000 node45 4.802 000000000000 node46 4.508 000000000000 node47 4.044 000000000000 node48 13.933 000000000000 node49 11.281 000000000000 node50 17.407 000000000000 node51 15.914 000000000000 node52 13.848 000000000000 node53 5.815 000000000000 node54 7.054 000000000000 node55 6.145 000000000000 node56 2.607 000000000000 node57 13.606 000000000000 node58 12.752 000000000000 node59 10.49 000000000000 node60 11.329 000000000000 node61 23.82 000000000000 node62 7.272 000000000000 node63 21.758 000000000000 node64 18.673 000000000000 node65 14.098 000000000000 Appendix S1 158 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node66 10.462100000000000000000 0000 node67 7.225100000000000000000 0000 node68 16.35100100011000000100 0000 node69 11.247100000000000000100 0000 node70 7.068100000000000000100 0000 node71 4.6100000011000000100 0000 node72 10.375101000011000000100 0000 node73 7.392100000000000000000 0000 node74 2.208100000011000000100 0000 node75 1.399100000011000000100 0000 node76 8.404101000011000000100 0000 node77 7.881100000011000000100 0000 node78 7.801000000000000000000 0000 node79 5.904100000011000000100 0000 node80 4.872100000011000000100 0000 node81 6.54100000000000000000 0000 node82 5.202100000011000000100 0000 node83 4.506100000000000000000 0000 node84 1.437100000000000000000 0000 node85 7.532100000011000000100 0000 node86 5.165001100000000000000 0000 node87 0.604011100000000000000 0010 node88 1.636100100011000000100 0000 node89 5.761100000011000000100 0000 node90 13.627110100000001000000 0010 node91 11.831100000000000000000 0000 node92 13.285000000000000000000 0000 node93 9.076000100000000000000 0000 node94 6.797110100000001111000 0010 node95 12.598000000000000000000 0000 node96 9.895110010000000000100 0000 node97 19.957110000000000000000 0000 node98 12.715110010000000000100 0000 node99 6.875110110000000000101 0000 node100 11.292110000000000000000 0000 node101 6.34110000000001000000 0010 node102 7.296110000000000000000 0000 node103 10.799110000000000000000 0000 node104 10.012110000000000000000 0000 node105 0.156110000000000000000 0000 node106 9.111110000000000000000 0000 node107 6.407110000000000000100 0000 node108 4.427110000011000000100 0000 node109 3.789110000000000000000 0000 node110 3.198110000011000000100 0000 node111 2.674111100000001000100 0010 node112 6.977111100000001000000 0010 node113 4.707111100000001000000 0010 node114 39.459000000000000000000 0000 node115 30.889000000000000000000 0000 node116 37.725000000000000000000 0000 node117 10.151110000011000000100 0010 node118 6.624110000011000000100 0010 node119 35.472000000000000000000 0000 node120 23.811110000000000000000 0000 node121 15.772110000000000000000 0000 node122 14.56100000000000000000 0000 node123 13.761000000000000000000 0000 node124 38.725000000000000000000 0000 node125 30.027000000000000000000 0000 node126 16.568100000000000000000 0000 node127 2.713110000000001000100 0010 node128 17.312110000000000000000 0000 Appendix S1 159 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) node66 10.462 0000000000000000000000 node67 7.225 0000000000000000000000 node68 16.35 0000010100000001100000 node69 11.247 0000000000000000000000 node70 7.068 0000000000000000000000 node71 4.6 0000000000000000000000 node72 10.375 0000101010100000000000 node73 7.392 0000000000000000000000 node74 2.208 0000000000000000000000 node75 1.399 0000000000000000000000 node76 8.404 0000000000000000000000 node77 7.881 0000000000000000000000 node78 7.801 0000000000000000000000 node79 5.904 0000000000000000000000 node80 4.872 0000000000000000000000 node81 6.54 0000000000000000000000 node82 5.202 0000000000000000000000 node83 4.506 0000000000000000000000 node84 1.437 0000000000000000000000 node85 7.532 0000000000000000000000 node86 5.165 0010110000000001111000 node87 0.604 0111111110100001100000 node88 1.636 0000010000000001111000 node89 5.761 0000000000000000000000 node90 13.627 0111010100000001100100 node91 11.831 0000000000000001100000 node92 13.285 0000000000000001100000 node93 9.076 0000010000000001100000 node94 6.797 0111010100000001100100 node95 12.598 0000000000000000000000 node96 9.895 0000000000000000000010 node97 19.957 0000000000000000000000 node98 12.715 0000000000000000000000 node99 6.875 0000000000000000000000 node100 11.292 0000000000000000000000 node101 6.34 0111000000000000000100 node102 7.296 0000000000000000000000 node103 10.799 0000000000000000000000 node104 10.012 0000000000000000000000 node105 0.156 0000000000000000000000 node106 9.111 0000000000000000000000 node107 6.407 0000000000000000000000 node108 4.427 0000000000000000000000 node109 3.789 0000000000000000000000 node110 3.198 0000000000000000000000 node111 2.674 0111111110100000000100 node112 6.977 0111111110100000000100 node113 4.707 0111111110100000000100 node114 39.459 0000000000000000000000 node115 30.889 0000000000000000000000 node116 37.725 0000000000000000000000 node117 10.151 0111000000000000000000 node118 6.624 0111000000000000000000 node119 35.472 0000000000000000000000 node120 23.811 0000000000000000000000 node121 15.772 0000000000000000000000 node122 14.56 0000000000000000000000 node123 13.761 0000000000000000000000 node124 38.725 0000000000000000000000 node125 30.027 0000000000000000000000 node126 16.568 0000000000000000000000 node127 2.713 0111000000000000000100 node128 17.312 0000000000000000000000 Appendix S1 160 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) node66 10.462 000 00000 0 0000000000000 node67 7.225 000 00000 0 0000000000000 node68 16.35 000 00000 0 0001111101000 node69 11.247 000 00000 0 0000000100000 node70 7.068 000 00000 0 0000000100000 node71 4.6 000 00000 0 0001111101000 node72 10.375 000 00000 0 0001111101000 node73 7.392 000 00000 0 0000000000000 node74 2.208 000 00000 0 0001111101000 node75 1.399 000 00000 0 0001111101000 node76 8.404 000 00000 0 0001111101000 node77 7.881 000 00000 0 0001111101000 node78 7.801 000 00000 0 0000000000000 node79 5.904 000 00000 0 0001111101000 node80 4.872 000 00000 0 0001111101000 node81 6.54 000 00000 0 0000000000000 node82 5.202 000 00000 0 0001111101000 node83 4.506 000 00000 0 0000000000000 node84 1.437 000 00000 0 0000000000000 node85 7.532 000 00000 0 0001111101000 node86 5.165 000 00000 0 0000000000000 node87 0.604 000 00000 0 0000000000000 node88 1.636 000 00000 0 0001111101000 node89 5.761 000 00000 0 0001111101000 node90 13.627 000 00000 0 0000000000000 node91 11.831 000 00000 0 0000000000000 node92 13.285 000 00000 0 0000000000000 node93 9.076 000 00000 0 0000000000000 node94 6.797 000 00000 0 0000000000000 node95 12.598 000 00000 0 0000000000000 node96 9.895 000 00000 0 0010000110001 node97 19.957 000 00000 0 0000000000000 node98 12.715 000 00000 0 0010000110000 node99 6.875 000 00000 0 0010000110000 node100 11.292 000 00000 0 0000000000000 node101 6.34 000 00000 0 0000000000000 node102 7.296 000 00000 0 0000000000000 node103 10.799 000 00000 0 0000000000000 node104 10.012 000 00000 0 0000000000000 node105 0.156 000 00000 0 0000000000000 node106 9.111 000 00000 0 0000000000000 node107 6.407 000 00000 0 0000000110000 node108 4.427 000 00000 0 0001111111000 node109 3.789 000 00000 0 0000000000000 node110 3.198 000 00000 0 0001111111000 node111 2.674 000 00000 0 0000000110000 node112 6.977 000 00000 0 0000000000000 node113 4.707 000 00000 0 0000000000000 node114 39.459 000 00000 0 0000000000000 node115 30.889 000 00000 0 0000000000000 node116 37.725 000 00000 0 0000000000000 node117 10.151 000 00000 0 0001111111000 node118 6.624 000 00000 0 0001111111000 node119 35.472 000 00000 0 0000000000000 node120 23.811 000 00000 0 0000000000000 node121 15.772 000 00000 0 0000000000000 node122 14.56 000 00000 0 0000000000000 node123 13.761 000 00000 0 0000000000000 node124 38.725 000 00000 0 0000000000000 node125 30.027 000 00000 0 0000000000000 node126 16.568 000 00000 0 0000000000000 node127 2.713 000 00000 0 0000000110000 node128 17.312 000 00000 0 0000000000000 Appendix S1 161 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node66 10.462 00000000000000000000000 node67 7.225 00000000000000000000000 node68 16.35 00000000000000000000000 node69 11.247 00000000000000000000000 node70 7.068 00000000000000000000000 node71 4.6 00000000000000000000000 node72 10.375 00000000000000000000000 node73 7.392 00000000000000000000000 node74 2.208 00000000000000000000000 node75 1.399 00000000000000000000000 node76 8.404 00000000000000000000000 node77 7.881 00000000000000000000000 node78 7.801 00000000000000000000000 node79 5.904 00000000000000000000000 node80 4.872 00000000000000000000000 node81 6.54 00000000000000000000000 node82 5.202 00000000000000000000000 node83 4.506 00000000000000000000000 node84 1.437 00000000000000000000000 node85 7.532 00000000000000000000000 node86 5.165 00000000000000000000000 node87 0.604 00000000000000000000011 node88 1.636 00000000000000000000000 node89 5.761 00000000000000000000000 node90 13.627 01001000100000000000011 node91 11.831 00000000000000000000000 node92 13.285 00000000000000000000000 node93 9.076 00000000000000000000000 node94 6.797 01111111000000000000011 node95 12.598 00000000000000000000000 node96 9.895 00000000000000001000000 node97 19.957 00000000000000000000000 node98 12.715 00000000000000001000000 node99 6.875 00000000000000001000000 node100 11.292 00000000000000000000000 node101 6.34 01001000100000000000011 node102 7.296 00000000000000000000000 node103 10.799 00000000000000000000000 node104 10.012 00000000000000000000000 node105 0.156 00000000000000000000000 node106 9.111 00000000000000000000000 node107 6.407 00000000000000001000000 node108 4.427 00000000000000001000000 node109 3.789 00000000000000000000000 node110 3.198 00000000000000001000000 node111 2.674 01001000100000001000011 node112 6.977 01001000100000000000011 node113 4.707 01001000100000000000011 node114 39.459 00000000000000000000000 node115 30.889 00000000000000000000000 node116 37.725 00000000000000000000000 node117 10.151 00000000000000001000011 node118 6.624 00000000000000001000011 node119 35.472 00000000000000000000000 node120 23.811 00000000000000000000000 node121 15.772 00000000000000000000000 node122 14.56 00000000000000000000000 node123 13.761 00000000000000000000000 node124 38.725 00000000000000000000000 node125 30.027 00000000000000000000000 node126 16.568 00000000000000000000000 node127 2.713 01001000100000001000011 node128 17.312 00000000000000000000000 Appendix S1 162 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node66 10.462 0000000000000000000000 node67 7.225 0000000000000000000000 node68 16.35 0000000000000000000000 node69 11.247 0000000000000000000000 node70 7.068 0000000000000000000000 node71 4.6 0000000000000000000000 node72 10.375 0000000001001110001111 node73 7.392 0000000000000000000000 node74 2.208 0000000000000000000000 node75 1.399 0000000000000000000000 node76 8.404 0000000000000000000000 node77 7.881 0000000000000000000000 node78 7.801 0000000000000000000000 node79 5.904 0000000000000000000000 node80 4.872 0000000000000000000000 node81 6.54 0000000000000000000000 node82 5.202 0000000000000000000000 node83 4.506 0000000000000000000000 node84 1.437 0000000000000000000000 node85 7.532 0000000000000000000000 node86 5.165 0000000000000010000000 node87 0.604 1111111101001110001111 node88 1.636 0000000000000000000000 node89 5.761 0000000000000000000000 node90 13.627 1111111100000000000000 node91 11.831 0000000000000000000000 node92 13.285 0000000000000000000000 node93 9.076 0000000000000000000000 node94 6.797 1111111100000000000000 node95 12.598 0000000000000000000000 node96 9.895 0000000000000000000000 node97 19.957 0000000000000000000000 node98 12.715 0000000000000000000000 node99 6.875 0000000000000000000000 node100 11.292 0000000000000000000000 node101 6.34 1111111100000000000000 node102 7.296 0000000000000000000000 node103 10.799 0000000000000000000000 node104 10.012 0000000000000000000000 node105 0.156 0000000000000000000000 node106 9.111 0000000000000000000000 node107 6.407 0000000000000000000000 node108 4.427 0000000000000000000000 node109 3.789 0000000000000000000000 node110 3.198 0000000000000000000000 node111 2.674 1111111101001110001111 node112 6.977 1111111101001110001111 node113 4.707 1111111101001110001111 node114 39.459 0000000000000000000000 node115 30.889 0000000000000000000000 node116 37.725 0000000000000000000000 node117 10.151 1111111100000000000000 node118 6.624 1111111100000000000000 node119 35.472 0000000000000000000000 node120 23.811 0000000000000000000000 node121 15.772 0000000000000000000000 node122 14.56 0000000000000000000000 node123 13.761 0000000000000000000000 node124 38.725 0000000000000000000000 node125 30.027 0000000000000000000000 node126 16.568 0000000000000000000000 node127 2.713 1111111100000000000000 node128 17.312 0000000000000000000000 Appendix S1 163 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) node66 10.462 00000000000001000000000 node67 7.225 00000000000001000000000 node68 16.35 00010110000100000000000 node69 11.247 00000000000000000000000 node70 7.068 00000000000000000000000 node71 4.6 00000000000000000000000 node72 10.375 11000000000000000000000 node73 7.392 00000000000000000000000 node74 2.208 00000000000000000000000 node75 1.399 00000000000000000000000 node76 8.404 00000000000000010000000 node77 7.881 00000000000000000000000 node78 7.801 00000000000000000000000 node79 5.904 00000000000000000000000 node80 4.872 00000000000000000000000 node81 6.54 00000000000000000000000 node82 5.202 00000000000000000000000 node83 4.506 00000000000000000000000 node84 1.437 00000000000000000000000 node85 7.532 00000000000000000000000 node86 5.165 00111000000110000000000 node87 0.604 11111111111100000000000 node88 1.636 00010000000110000000000 node89 5.761 00000000000000000000000 node90 13.627 00011110011100000000000 node91 11.831 00000000000101000000000 node92 13.285 00000000000100000000000 node93 9.076 00010000000100000000000 node94 6.797 00011110011100000000000 node95 12.598 00000000000000000000000 node96 9.895 00000000000000000000000 node97 19.957 00000000000000000000000 node98 12.715 00000000000000000000000 node99 6.875 00000000000000001100000 node100 11.292 00000000000000000000000 node101 6.34 00000000000000000000000 node102 7.296 00000000000000000000000 node103 10.799 00000000000000000000000 node104 10.012 00000000000001100000000 node105 0.156 00000000000001100000000 node106 9.111 00000000000000000000000 node107 6.407 00000000000000000000000 node108 4.427 00000000000000000000000 node109 3.789 00000000000000000000000 node110 3.198 00000000000000000000000 node111 2.674 11111111111000000000000 node112 6.977 11111111111000000000000 node113 4.707 11111111111000000000000 node114 39.459 00000000000000000000000 node115 30.889 00000000000000000000000 node116 37.725 00000000000000000000000 node117 10.151 00000000000000000000000 node118 6.624 00000000000000000000000 node119 35.472 00000000000000000000000 node120 23.811 00000000000000000000000 node121 15.772 00000000000000000000000 node122 14.56 00000000000000000000000 node123 13.761 00000000000000000000000 node124 38.725 00000000000000000000000 node125 30.027 00000000000000000000000 node126 16.568 00000000000000000000000 node127 2.713 00000000000000000000000 node128 17.312 00000000000001100000000 Appendix S1 164 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node66 10.462 000000000000 node67 7.225 000000000000 node68 16.35 000000000000 node69 11.247 000000000000 node70 7.068 000000000000 node71 4.6 000000000000 node72 10.375 000000000000 node73 7.392 000000000000 node74 2.208 000000000000 node75 1.399 000000000000 node76 8.404 000000000000 node77 7.881 000000000000 node78 7.801 000000000000 node79 5.904 000000000000 node80 4.872 000000000000 node81 6.54 000000000000 node82 5.202 000000000000 node83 4.506 000000000000 node84 1.437 000000000000 node85 7.532 000000000000 node86 5.165 000000000000 node87 0.604 000000000000 node88 1.636 000000000000 node89 5.761 000000000000 node90 13.627 000000000000 node91 11.831 000000000000 node92 13.285 000000000000 node93 9.076 000000000000 node94 6.797 000000000000 node95 12.598 000000000000 node96 9.895 000000000000 node97 19.957 000000000000 node98 12.715 000000000000 node99 6.875 000000000000 node100 11.292 000000000000 node101 6.34 000000000000 node102 7.296 000000000000 node103 10.799 000000000000 node104 10.012 000000000000 node105 0.156 000000000000 node106 9.111 000000000000 node107 6.407 000000000000 node108 4.427 000000000000 node109 3.789 000000000000 node110 3.198 000000000000 node111 2.674 000000000000 node112 6.977 000000000000 node113 4.707 000000000000 node114 39.459 000000000000 node115 30.889 000000000000 node116 37.725 000000000000 node117 10.151 000000000000 node118 6.624 000000000000 node119 35.472 000000000000 node120 23.811 000000000000 node121 15.772 000000000000 node122 14.56 000000000000 node123 13.761 000000000000 node124 38.725 000000000000 node125 30.027 000000000000 node126 16.568 000000000000 node127 2.713 000000000000 node128 17.312 000000000000 Appendix S1 165 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node129 14.868110000000000000000 0000 node130 13.257000000000000000000 0000 node131 42.702100000000000000000 0000 node132 35.835100000000000000000 0000 node133 11.937110000000000000000 0000 node134 10.594000000000000000000 0000 node135 11.125000000000000000000 0000 node136 32.188000000000000000000 0000 node137 18.058100000001000000100 0000 node138 53.424100000000000000000 0000 node139 30.949110000000000000000 0000 node140 29.565000000000000000000 0000 node141 26.91100000000000000000 0000 node142 22.034100000000001000000 0000 node143 15.304100000000000000000 0000 node144 11.958110000000000000000 0000 node145 20.378100000000001000000 0000 node146 19.462100000000001000000 0000 node147 18.293000000000000000000 0000 node148 15.2110000000001000000 0000 node149 14.359010000000000000000 0000 node150 8.859010000000000000000 0010 node151 13.39000000000000000000 0000 node152 12.514010000000000000000 0000 node153 2.387111100000001000000 0010 node154 8.652000000000000000000 0000 node155 13.628110000000001000000 0000 node156 10.134100000000001000000 0000 node157 17.023000000000000000000 0000 node158 7.055000000000000000000 0000 node159 5.053000000000000000000 0000 node160 15.076110000011000000111 0000 node161 11.652110000011000000111 0000 node162 8.312110000011000000111 0000 node163 3.864110000011000000111 0000 node164 2.379110000011000000111 0000 node165 7.137110000011000000110 0000 node166 8.605110000011000000100 0000 node167 4.132110000011000000100 0000 node168 2.531110000011000000100 0000 node169 7.778110000000000000000 0000 node170 7.216110000001000000011 0000 node171 1.573110000011000000111 0000 node172 0.69110000011000000111 0000 node173 6.281110000000000000000 0000 node174 3.904111100000001000000 0010 node175 1.945111100000001000000 0010 node176 35.768100000000000000000 0000 node177 14.793100000000001001000 0000 node178 11.336100000000001001000 0000 node179 8.27100000000001001000 0000 node180 3.367100001000001001100 0000 node181 72.642000000000000000000 0000 node182 68.159000000000000000000 0000 node183 33.74110100000001000000 0000 node184 21.821110100000001000000 0010 node185 15.101110101100001000000 0010 node186 13.816100000000001000000 0000 node187 10.504110101100001000000 0010 node188 8.753110101100001000000 0010 node189 5.254110101100001000000 0010 node190 3.834110100000000000000 0000 node191 6.611100001100000000000 0000 Appendix S1 166 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) node129 14.868 0000000000000000000000 node130 13.257 0000000000000000000000 node131 42.702 0000000000000000000000 node132 35.835 0000000000000000000000 node133 11.937 0000000000000000000000 node134 10.594 0000000000000000000000 node135 11.125 0000000000000000000000 node136 32.188 0000000000000000000000 node137 18.058 0000000000000000000000 node138 53.424 0000000000000000000000 node139 30.949 0000000000000000000000 node140 29.565 0000000000000000000000 node141 26.91 0000000000000000000000 node142 22.034 0000000000000000000100 node143 15.304 0000000000000000000000 node144 11.958 0000000000000000000000 node145 20.378 0000000000000000000100 node146 19.462 0000000000000000000100 node147 18.293 0000000000000000000000 node148 15.2 0111000100000000000100 node149 14.359 0011000000000000000000 node150 8.859 0111000000000000000000 node151 13.39 0000000000000000000000 node152 12.514 0011000000000000000000 node153 2.387 0111111110100000000100 node154 8.652 0000000000000000000000 node155 13.628 0011000000000000000100 node156 10.134 0000000000000000000100 node157 17.023 0000000000000000000000 node158 7.055 0000000000000000000000 node159 5.053 0000000000000000000000 node160 15.076 0000000000000000000001 node161 11.652 0000000000000000000001 node162 8.312 0000000000000000000001 node163 3.864 0000000000000000000001 node164 2.379 0000000000000000000001 node165 7.137 0000000000000000000000 node166 8.605 0000000000000000000000 node167 4.132 0000000000000000000000 node168 2.531 0000000000000000000000 node169 7.778 0000000000000000000000 node170 7.216 0000000000000000000001 node171 1.573 0000000000000000000001 node172 0.69 0000000000000000000001 node173 6.281 0000000000000000000000 node174 3.904 0111111110100000000100 node175 1.945 0111111110100000000100 node176 35.768 0000000000000000000000 node177 14.793 0000000000000000000100 node178 11.336 0000000000000000000100 node179 8.27 0000000000000000000100 node180 3.367 0000000000000000000100 node181 72.642 0000000000000000000000 node182 68.159 0000000000000000000000 node183 33.74 0001010100000000000100 node184 21.821 0111010100000000000100 node185 15.101 0111010100000000000100 node186 13.816 0011000100000000000100 node187 10.504 0111010100000000000100 node188 8.753 0111010100000000000100 node189 5.254 0111010100000000000100 node190 3.834 0010010000000000000000 node191 6.611 0000000000000000000000 Appendix S1 167 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) node129 14.868 000 00000 0 0000000000000 node130 13.257 000 00000 0 0000000000000 node131 42.702 000 00000 0 0000000000000 node132 35.835 000 00000 0 0000000000000 node133 11.937 000 00000 0 0000000000000 node134 10.594 000 00000 0 0000000000000 node135 11.125 000 00000 0 0000000000000 node136 32.188 000 00000 0 0000000000000 node137 18.058 000 00000 0 0000000101000 node138 53.424 000 00000 0 0000000000000 node139 30.949 000 00000 0 0000000000000 node140 29.565 000 00000 0 0000000000000 node141 26.91 000 00000 0 0000000000000 node142 22.034 100 00000 0 0000000000000 node143 15.304 000 00000 0 0000000000000 node144 11.958 000 00000 0 0000000000000 node145 20.378 100 00000 0 0000000000000 node146 19.462 100 00000 0 0000000000000 node147 18.293 000 00000 0 0000000000000 node148 15.2 100 11101 0 0000000000000 node149 14.359 000 11110 0 0000000000000 node150 8.859 000 11110 0 0000000000000 node151 13.39 000 00000 0 0000000000000 node152 12.514 000 11100 0 0000000000000 node153 2.387 110 11101 0 0000000000000 node154 8.652 000 00000 0 0000000000000 node155 13.628 100 11001 0 0000000000000 node156 10.134 000 00000 0 0000000000000 node157 17.023 000 00000 0 0000000000000 node158 7.055 000 00000 0 0000000000000 node159 5.053 000 00000 0 0000000000000 node160 15.076 000 00000 0 0001111111000 node161 11.652 000 00000 0 0001111111000 node162 8.312 000 00000 0 0001111111000 node163 3.864 000 00000 0 0001111101000 node164 2.379 000 00000 0 0001111101000 node165 7.137 000 00000 0 0001111101000 node166 8.605 000 00000 0 0001111111000 node167 4.132 000 00000 0 0001111101000 node168 2.531 000 00000 0 0001111111000 node169 7.778 000 00000 0 0000000000000 node170 7.216 000 00000 0 0000100000000 node171 1.573 000 00000 0 0001111111000 node172 0.69 000 00000 0 0001111111000 node173 6.281 000 00000 0 0000000000000 node174 3.904 000 00000 0 0000000000000 node175 1.945 000 00000 0 0000000000000 node176 35.768 000 00000 0 0000000000000 node177 14.793 000 00000 1 0000000000000 node178 11.336 000 00000 1 0000000000000 node179 8.27 000 00000 1 0000000000000 node180 3.367 000 00000 1 1000000100100 node181 72.642 000 00000 0 0000000000000 node182 68.159 000 00000 0 0000000000000 node183 33.74 000 00000 0 0000000000000 node184 21.821 000 00000 0 0000000000000 node185 15.101 000 00000 0 0000000000110 node186 13.816 000 00000 0 0000000000000 node187 10.504 000 00000 0 0000000000110 node188 8.753 000 00000 0 0000000000110 node189 5.254 000 00000 0 0000000000110 node190 3.834 000 00000 0 0000000000000 node191 6.611 000 00000 0 0000000000110 Appendix S1 168 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node129 14.868 00000000000000000000000 node130 13.257 00000000000000000000000 node131 42.702 00000000000000000000000 node132 35.835 00000000000000000000000 node133 11.937 00000000000000000000000 node134 10.594 00000000000000000000000 node135 11.125 00000000000000000000000 node136 32.188 00000000000000000000000 node137 18.058 00000000000000000000000 node138 53.424 00000000000000000000000 node139 30.949 00000000000000000000000 node140 29.565 00000000000000000000000 node141 26.91 00000000000000000000000 node142 22.034 01001000000000000000000 node143 15.304 00000000000000000000000 node144 11.958 00000000000000000000000 node145 20.378 01001000000000000000000 node146 19.462 01001000000000000000000 node147 18.293 00000000000000000000000 node148 15.2 00000000000000000000011 node149 14.359 00000000000000000000000 node150 8.859 00000000000000000000011 node151 13.39 00000000000000000000000 node152 12.514 00000000000000000000000 node153 2.387 01001000000000000000011 node154 8.652 00000000000000000000000 node155 13.628 01001000000000000000000 node156 10.134 01001000000000000000000 node157 17.023 00000000000000000000000 node158 7.055 00000000000000000000000 node159 5.053 00000000000000000000000 node160 15.076 00000000000000001111100 node161 11.652 00000000000000001111100 node162 8.312 00000000000000001111100 node163 3.864 00000000000000001111100 node164 2.379 00000000000000001111100 node165 7.137 00000000000000001100000 node166 8.605 00000000000000001000000 node167 4.132 00000000000000001000000 node168 2.531 00000000000000001000000 node169 7.778 00000000000000000000000 node170 7.216 00000000000000000111100 node171 1.573 00000000000000001111100 node172 0.69 00000000000000001111100 node173 6.281 00000000000000000000000 node174 3.904 01001000100000000000011 node175 1.945 01001000100000000000011 node176 35.768 00000000000000000000000 node177 14.793 01001001000000000000000 node178 11.336 01001001000000000000000 node179 8.27 01001001000000000000000 node180 3.367 01001001000000000000000 node181 72.642 00000000000000000000000 node182 68.159 00000000000000000000000 node183 33.74 01001000000000000000000 node184 21.821 01001000000000000000011 node185 15.101 01001000100000000000011 node186 13.816 01001000100000000000000 node187 10.504 01001000100000000000011 node188 8.753 01001000100000000000011 node189 5.254 01001000100000000000011 node190 3.834 00000000000000000000000 node191 6.611 00000000000000000000000 Appendix S1 169 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node129 14.868 0000000000000000000000 node130 13.257 0000000000000000000000 node131 42.702 0000000000000000000000 node132 35.835 0000000000000000000000 node133 11.937 0000000000000000000000 node134 10.594 0000000000000000000000 node135 11.125 0000000000000000000000 node136 32.188 0000000000000000000000 node137 18.058 0000000000000000000000 node138 53.424 0000000000000000000000 node139 30.949 0000000000000000000000 node140 29.565 0000000000000000000000 node141 26.91 0000000000000000000000 node142 22.034 0000000000000000000000 node143 15.304 0000000000000000000000 node144 11.958 0000000000000000000000 node145 20.378 0000000000000000000000 node146 19.462 0000000000000000000000 node147 18.293 0000000000000000000000 node148 15.2 1001100000000000000000 node149 14.359 0001100100000000000000 node150 8.859 1111111100000000000000 node151 13.39 0000000000000000000000 node152 12.514 0001100100000000000000 node153 2.387 1111111101001110001111 node154 8.652 0000000000000000000000 node155 13.628 0001100100000000000000 node156 10.134 0000000000000000000000 node157 17.023 0000000000000000000000 node158 7.055 0000000000000000000000 node159 5.053 0000000000000000000000 node160 15.076 0000000000000000000000 node161 11.652 0000000000000000000000 node162 8.312 0000000000000000000000 node163 3.864 0000000000000000000000 node164 2.379 0000000000000000000000 node165 7.137 0000000000000000000000 node166 8.605 0000000000000000000000 node167 4.132 0000000000000000000000 node168 2.531 0000000000000000000000 node169 7.778 0000000000000000000000 node170 7.216 0000000000000000000000 node171 1.573 0000000000000000000000 node172 0.69 0000000000000000000000 node173 6.281 0000000000000000000000 node174 3.904 1111111101001110001111 node175 1.945 1111111101001110001111 node176 35.768 0000000000000000000000 node177 14.793 0000000000000000000000 node178 11.336 0000000000000000000000 node179 8.27 0000000000000000000000 node180 3.367 0000000000000000000000 node181 72.642 0000000000000000000000 node182 68.159 0000000000000000000000 node183 33.74 0000000000000000000000 node184 21.821 1111111100000000000000 node185 15.101 1111111100000000000000 node186 13.816 0001100000000000000000 node187 10.504 1111111100000000000000 node188 8.753 1111111100000000000000 node189 5.254 1111111100000000000000 node190 3.834 0000000000000000000000 node191 6.611 0000000000000000000000 Appendix S1 170 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) node129 14.868 00000000000000100000000 node130 13.257 00000000000000000000000 node131 42.702 00000000000000000000000 node132 35.835 00000000000000000000000 node133 11.937 00000000000000100000000 node134 10.594 00000000000000000000000 node135 11.125 00000000000000000000000 node136 32.188 00000000000000000000000 node137 18.058 00000000000000000000000 node138 53.424 00000000000000000000000 node139 30.949 00000000000000000000000 node140 29.565 00000000000000000000000 node141 26.91 00000000000000000000000 node142 22.034 00000000000000000000000 node143 15.304 00000000000001000000000 node144 11.958 00000000000001100000000 node145 20.378 00000000000000000000000 node146 19.462 00000000000000000000000 node147 18.293 00000000000000000000000 node148 15.2 00000000011000000000000 node149 14.359 00000000000000000000000 node150 8.859 00000000000000000000000 node151 13.39 00000000000000000000000 node152 12.514 00000000000000000000000 node153 2.387 11010111100000000000000 node154 8.652 00000000000000000000000 node155 13.628 00000000011000000000000 node156 10.134 00000000000000000000000 node157 17.023 00000000000000000000000 node158 7.055 00000000000000000000000 node159 5.053 00000000000000000000000 node160 15.076 00000000000000000000000 node161 11.652 00000000000000000000000 node162 8.312 00000000000000000000000 node163 3.864 00000000000000000000000 node164 2.379 00000000000000000000000 node165 7.137 00000000000000000000000 node166 8.605 00000000000000000000000 node167 4.132 00000000000000000000000 node168 2.531 00000000000000000000000 node169 7.778 00000000000000000000000 node170 7.216 00000000000000100000000 node171 1.573 00000000000000000000000 node172 0.69 00000000000000000000000 node173 6.281 00000000000000000000000 node174 3.904 11111111111000000000000 node175 1.945 11111111111000000000000 node176 35.768 00000000000000000000000 node177 14.793 00000000000000000000000 node178 11.336 00000000000000000000000 node179 8.27 00000000000000000000000 node180 3.367 00000000000000000000000 node181 72.642 00000000000000000000000 node182 68.159 00000000000000000000000 node183 33.74 00010110011000100000000 node184 21.821 00010110000000000000000 node185 15.101 00011110011000000000000 node186 13.816 00000000011000000000000 node187 10.504 00011110011000000000000 node188 8.753 00011110011000000000000 node189 5.254 00011110011000000000000 node190 3.834 00011000000001100000000 node191 6.611 00000000000000000000000 Appendix S1 171 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node129 14.868 000000000000 node130 13.257 000000000000 node131 42.702 000000000000 node132 35.835 000000000000 node133 11.937 000000000000 node134 10.594 000000000000 node135 11.125 000000000000 node136 32.188 000000000000 node137 18.058 000000000000 node138 53.424 000000000000 node139 30.949 000000000000 node140 29.565 000000000000 node141 26.91 000000000000 node142 22.034 100000000000 node143 15.304 000000000000 node144 11.958 000000000000 node145 20.378 100000000000 node146 19.462 100000000000 node147 18.293 000000000000 node148 15.2 100111010000 node149 14.359 000111100000 node150 8.859 000111100000 node151 13.39 000000000000 node152 12.514 000111000000 node153 2.387 110111010000 node154 8.652 000000000000 node155 13.628 100110010000 node156 10.134 000000000000 node157 17.023 000000000000 node158 7.055 000000000000 node159 5.053 000000000000 node160 15.076 000000000000 node161 11.652 000000000000 node162 8.312 000000000000 node163 3.864 000000000000 node164 2.379 000000000000 node165 7.137 000000000000 node166 8.605 000000000000 node167 4.132 000000000000 node168 2.531 000000000000 node169 7.778 000000000000 node170 7.216 000000000000 node171 1.573 000000000000 node172 0.69 000000000000 node173 6.281 000000000000 node174 3.904 000000000000 node175 1.945 000000000000 node176 35.768 000000000000 node177 14.793 000000001000 node178 11.336 000000001000 node179 8.27 000000001000 node180 3.367 000000001110 node181 72.642 000000000000 node182 68.159 000000000000 node183 33.74 000000000000 node184 21.821 000000000000 node185 15.101 000000000000 node186 13.816 000000000000 node187 10.504 000000000000 node188 8.753 000000000000 node189 5.254 000000000000 node190 3.834 000000000000 node191 6.611 000000000000 Appendix S1 172 - 14 = 15 = -carotene lutein -doradexanthin 1 = lutein -cryptoxanthin 8 = canary 9 = canary zeaxanthin zeaxanthin tetrahydro- 19 = 7,8,7',8'- hydroxylutein tunaxanthin F xanthophyll A xanthophyll B tunaxanthin A 3 = β 2 = zeaxanthin 25 = idoxanthin 26 = fucoxanthin 17 = piprixanthin 20 = 7,8-dihydro- 5 = anhydrolutein fritschiellaxanthin 18 = rhodoxanthin papilioerythrinone 10 = (3S,6S,3'S,6'S) 7 = 9-Z-7,8-dihydro- 4 = β 11 = (3R,6R,3'R,6'R) 16 = 3'-dehydrolutein 6 = 7,8-dihydro-lutein 12 = α 13 = 4 (3S,4R,3'R,6'R) Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node192 14.062110101100001000000 0000 node193 12.469100001100000000000 0000 node194 9.12100000000000000000 0000 node195 6.433100000000000000000 0000 node196 12.228100000000000000000 0000 node197 9.584100000000001000000 0000 node198 15.456100100000001000000 0000 node199 13.471000000000000000000 0000 node200 6.309100000000001000000 0000 node201 25.556110100000001000000 0000 node202 17.701010000000000000000 0000 node203 3.614000000000000000000 0000 node204 1.286000000000000000000 0000 node205 71.607000000000000000000 0000 node206 17.918100000000000000000 0000 node207 12.966100000000000000000 0000 node208 9.466100000000000000000 0000 node209 24.492001000000000000000 0000 node210 15.683001000000000000000 0000 node211 9.942001000000000000000 0000 node212 5.443101000011000000100 0001 node213 10.769000100000000000000 0000 node214 62.863000000000000000000 0000 node215 61.643000000000000000000 0000 node216 49.896000000000000000000 0000 node217 76.002000000000000000000 0000 Appendix S1 173 - - β - - -carotene 44 = 31 = 4- 43 = α 41 = β -cryptoxanthin structure 48 = 4-oxo- 50 = 4-oxo- echinenone echinenone ruboxanthin 32 = (3S, 3'R) adonixanthin hydroxylutein cryptoxanthin 52 = cis-lutein gazaniaxanthin 71 = resonance 39 = 4-hydroxy- 42 = α 38 = adonirubin 36 = 3'-hydroxy- 47 = rubixanthin 34 = astaxanthin 35 = echinenone isocryptoxanthin isocryptoxanthin phoenicopterone 30 = 7,8 dihydro 40 = isozeaxanthin hydroxyzeaxanthin 37 = canthaxanthin 49 = gazaniaxanthin 46 = α 51 = (3S,4R,3'S,6'R) 4 Species & ancestral Age binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 networks (MYA) node192 14.062 0000010100000000000100 node193 12.469 0000000000000000000000 node194 9.12 0000000000000000000000 node195 6.433 0000000000000000000000 node196 12.228 0000000000000000000000 node197 9.584 0011000000000000000100 node198 15.456 0000010100000000000100 node199 13.471 0000000000000000000000 node200 6.309 0000000000000000000100 node201 25.556 0001010100000000000100 node202 17.701 0001000000000000000000 node203 3.614 0001000000000000000000 node204 1.286 0001000000000000000000 node205 71.607 0000000000000000000000 node206 17.918 0000000000000000000000 node207 12.966 0000000000000000000000 node208 9.466 0000000000000000000000 node209 24.492 0001110100000000000000 node210 15.683 0001111110100000000000 node211 9.942 0001111110100000000000 node212 5.443 0001111110100000000000 node213 10.769 0001010100000000000000 node214 62.863 0000000000000000000000 node215 61.643 0000000000000000000000 node216 49.896 0000000000000000000000 node217 76.002 0000000000000000000000 Appendix S1 174 1-5 1-8 1-9 9-8 8-9 1-6 6-7 1-16 16-1 16-9 1-52 methoxy- xipholenin (eurylaimin) pompadourin canthaxanthin dehydro-lutein 102 = rupicolin 106 = contingin (cymbirhynchin) 100 = xipholenin 110 = 4-hydroxy- papilioerythrinone 103 = 3'-hydroxy-3- 104 = pompadourin 107 = brittonxanthin 109 = 7,8-dihydro-3'- 105 = 2,3-didehydro- 108 = 2,3-didehydro- 101 = 2,3-didehydro- canary xanthophyll A Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 networks (MYA) node192 14.062 000 00000 0 0000000000110 node193 12.469 000 00000 0 0000000000110 node194 9.12 000 00000 0 0000000000000 node195 6.433 000 00000 0 0000000000000 node196 12.228 000 00000 0 0000000000000 node197 9.584 000 00000 0 0000000000000 node198 15.456 000 00000 0 0000000000000 node199 13.471 000 00000 0 0000000000000 node200 6.309 000 00000 0 0000000000000 node201 25.556 000 00000 0 0000000000000 node202 17.701 000 00000 0 0000000000000 node203 3.614 000 00000 0 0000000000000 node204 1.286 000 00000 0 0000000000000 node205 71.607 000 00000 0 0000000000000 node206 17.918 000 00000 0 0000000000000 node207 12.966 000 00000 0 0000000000000 node208 9.466 000 00000 0 0000000000000 node209 24.492 000 00000 0 0000000000000 node210 15.683 000 00000 0 0000000000000 node211 9.942 000 00000 0 0000000000000 node212 5.443 000 00000 0 0001111101000 node213 10.769 000 00000 0 0000000000000 node214 62.863 000 00000 0 0000000000000 node215 61.643 000 00000 0 0000000000000 node216 49.896 000 00000 0 0000000000000 node217 76.002 000 00000 0 0000000000000 Appendix S1 175 1-46 1-51 8-10 2-20 2-16 9-17 2-31 51-13 13-51 51-12 13-14 14-15 12-15 12-32 10-11 42-46 42-43 43-44 20-19 17-18 17-71 71-18 31-32

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 . binary 1 . binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node192 14.062 01001000000000000000000 node193 12.469 00000000000000000000000 node194 9.12 00000000000000000000000 node195 6.433 00000000000000000000000 node196 12.228 00000000000000000000000 node197 9.584 01001000100000000000000 node198 15.456 01001000000000000000000 node199 13.471 00000000000000000000000 node200 6.309 01001000000000000000000 node201 25.556 01001000000000000000000 node202 17.701 00000000000000000000000 node203 3.614 00000000000000000000000 node204 1.286 00000000000000000000000 node205 71.607 00000000000000000000000 node206 17.918 00000000000000000000000 node207 12.966 00000000000000000000000 node208 9.466 00000000000000000000000 node209 24.492 00000000000000000000000 node210 15.683 00000000000000000000000 node211 9.942 00000000000000000000000 node212 5.443 00000000000000000000000 node213 10.769 00000000000000000000000 node214 62.863 00000000000000000000000 node215 61.643 00000000000000000000000 node216 49.896 00000000000000000000000 node217 76.002 00000000000000000000000 Appendix S1 176 2-32 4-30 3-41 3-35 32-31 32-25 25-32 32-34 34-32 34-25 25-34 41-40 40-41 41-35 35-41 37-40 40-37 39-40 35-39 39-35 35-37 37-35

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node192 14.062 0000000000000000000000 node193 12.469 0000000000000000000000 node194 9.12 0000000000000000000000 node195 6.433 0000000000000000000000 node196 12.228 0000000000000000000000 node197 9.584 0001100000000000000000 node198 15.456 0000000000000000000000 node199 13.471 0000000000000000000000 node200 6.309 0000000000000000000000 node201 25.556 0000000000000000000000 node202 17.701 0000000000000000000000 node203 3.614 0000000000000000000000 node204 1.286 0000000000000000000000 node205 71.607 0000000000000000000000 node206 17.918 0000000000000000000000 node207 12.966 0000000000000000000000 node208 9.466 0000000000000000000000 node209 24.492 0000000000000010000000 node210 15.683 0000000001001110001111 node211 9.942 0000000001001110001111 node212 5.443 0000000001001110001111 node213 10.769 0000000000000000000000 node214 62.863 0000000000000000000000 node215 61.643 0000000000000000000000 node216 49.896 0000000000000000000000 node217 76.002 0000000000000000000000 Appendix S1 177 1-1 2-2 3-3 4-4 4-36 39-37 37-39 35-36 36-32 36-38 38-36 37-38 38-37 38-34 34-38 47-48 49-50 18-18 42-42 34-34 37-37 38-38 26-26

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 1 binary 2 binary 1 . binary 2 networks (MYA) node192 14.062 00010110000000100000000 node193 12.469 00000000000000000000000 node194 9.12 00000000000000000000000 node195 6.433 00000000000000000000000 node196 12.228 00000000000000000000000 node197 9.584 00000000000000000000000 node198 15.456 00010110000000000000000 node199 13.471 00000000000000000000000 node200 6.309 00000000000000000000000 node201 25.556 00010110011000100000000 node202 17.701 00000000000000000000000 node203 3.614 00000000000000000001000 node204 1.286 00000000000000000000000 node205 71.607 00000000000000000000000 node206 17.918 00000000000001000000000 node207 12.966 00000000000001000000000 node208 9.466 00000000000001000000000 node209 24.492 00100110011000000000000 node210 15.683 11100110011000000000000 node211 9.942 11100111111000000000000 node212 5.443 11100111111000000000001 node213 10.769 00010110011000000000000 node214 62.863 00000000000000000000000 node215 61.643 00000000000000000000000 node216 49.896 00000000000000000000000 node217 76.002 00000000000000000000000 Appendix S1 178 6-109 8-110 12-100 32-102 34-103 38-107 15-108 16-109 100-101 103-104 104-105 105-106

Species & ancestral Age binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 binary 2 networks (MYA) node192 14.062 000000000000 node193 12.469 000000000000 node194 9.12 000000000000 node195 6.433 000000000000 node196 12.228 000000000000 node197 9.584 000000000000 node198 15.456 000000000000 node199 13.471 000000000000 node200 6.309 000000000000 node201 25.556 000000000000 node202 17.701 000000000000 node203 3.614 000000000000 node204 1.286 000000000000 node205 71.607 000000000000 node206 17.918 000000000000 node207 12.966 000000000000 node208 9.466 000000000000 node209 24.492 000000000000 node210 15.683 000000000000 node211 9.942 000000000000 node212 5.443 000000000000 node213 10.769 000000000000 node214 62.863 000000000000 node215 61.643 000000000000 node216 49.896 000000000000 node217 76.002 000000000000 a 179 Malurus melanocephalus Motacilla flava Icteria viens Dendroica petechial Setophaga ruticilla Dendroica palmarum Dendroica coronate Geothlypis trichas Vermivora ruficapilla Vermivora virginiae Xanthocephalus xanthocephalus Sturnella superciliaris Sturnella neglecta Sturnella magna Sturnella militaris Sturnella bellicose Psarocolius montezuma Psarocolius wagleri Cacicus melanicterus Cacicus cela Cacicus uropygilais Cacicus haemorrhous Cacicus leucoramphus Agelaius phoeniceus Icterus graduacauda Icterus gularis Icterus nigrogularis Icterus galbula Ictuers pustulatus Icterus bullockii Icterus mesomelas Icterus pectoralis Icterus Icterus croconotus Icterus prosthemelas Ictuers cucullatus Icterus dominicensis Chlorospingus pileatus Emberiza melanocephala Emberiza citronella Pheucticus ludovicianus Cardinalis sinuatus Cardinalis cardinalis Piranga rubra Piranga flava Piranga olivacea Nesospiza acunhae Paroaria coronate Coereba flaveola Ramphocelus dimidiatus Sicalis flaveola

a continued, fig. S1b b 50 40 30 20 10 0 MYA c

Figure S1. a) Majority-rule ultrametric consensus tree of a subset of the passerine species in this study showing the functional modules (A-N) present in each species’ carotenoid metabolic network (black shaded squares). Numbers at internal nodes identify the locations of reconstructed ancestral networks on the phylogeny (see Appendix S1). Modules are comprised of compounds that co-occur in species and ancestral networks (n = 467) at least 70% of the time (see figure S1e and Appendix S2 for descriptions of modules). The phylogeny is part of a majority-rule ultrametric consensus tree of 243 species derived from 1,000 randomly sampled trees from the Hackett All Species pseudo posterior distribution in Jetz et al. [1] (Appendix S3). The other subsets of the phylogeny, shown in the inset in the lower right corner, are displayed in Figures S1b, S1c, and S1d. b continued, fig. S1a 180

Fringilla coelebs Fringilla montifringilla Mycerobas affinis Mycerobas melanozanthos Mycerobas icterioides Myceropas carnipes Coccothraustes vespertinus Coccothraustes abeillei Rhynchostruthus socotranus Vestiaria coccinea Bucanetes githagineus Pinicola enucleator Pyrrhoplectes epaulette Pyrrhula erythrocephala Pyrrhula erythaca Pyrrhula pyrrhula Pyrrhula aurantiaca Rhodopechys obsoletus Carduelis chloris Carduelis sinica Carduelis spinoides Serinus mozambicus Serinus mozambicus Serinus canaria Serinus pusillus Serinus serinus Carduelis cannabina Carduelis spinus Carduelis tristis Carduelis atrata 87 Carduelis cucullata Carduelis hornemanni Carduelis flammea Loxia curvirostra Loxia leucoptera Carduelis citronella Carduelis carduelis Haematospiza sipahi Uragus sibiricus Carpodacus rubicilloides Carpodacus trifasciatus Carpodacus thura Carpodacus pulcherrimus Carpodacus roseus Carpodacus mexicanus Carpodacus nipalensis Euphonia laniirostris Amandava subflava Amandava amandava Erythrura psittacea Erythrura gouldiae Taeniopygia guttata Neochmia ruficauda Ploceus bicolor Ploceus cucullatus Ploceus nelicourvi Ploceus sakalva Ploceus philippinus 105 Ploceus capensis Ploceus velatus Euplectes afer Euplectes capensis Euplectes macroura Euplectes axillaris Euplectes ardens Euplectes orix Foudia madagascariensis Quelea quelea Quelea cardinalis Quelea erythrops

continued, fig. S1c b

70 60 50 40 30 20 10 0 MYA c

Figure S1. b) Majority-rule ultrametric consensus tree of a subset of the passerine species in this study. Legend in figure S1a. c continued, fig. S1b 181

A B C D E F G H I J K L M N Regulus strapa Regulus regulus Bombycilla cedrorum Bombycilla japonica Bombycilla garrulous Tichodroma muraria Turdus merula Erithacus rubecula Luscinia calliope Ficedula zanthopygia Tarsiger chrysaeus Aegithalos caudatus Zosterops japonicas Leiothrix lutea Leiothrix argentauris Parus caeruleus Parus ater Parus major Parus spilonotus Notiomystis cincta Oriolus cruentus Oriolus trailli Oriolus orioulus Orioulus xanthornus Pericrocotus flammeus Telophorus zeylonus Telophorus sulfureopectus Tyrannus vociferans Iodopleura isabellae Ampelioides tschudii Pipreola Formosa Pipreola chlorolepidota Pipreola aureopectus Pipreola whitelyi Ampelion rufaxilla Snowornis subalaris Phibalura flavirostris Cotinga maculate Cotinga cotinga Cotinga amabilis Procnias tricarunculatus Porphyrolaema porphyrolaema Xipholena atropurpurea Xipholena punicea Xipholena lamellipennis Lipaugus streptophorus Tijuca atra Haematoderus militaris Querula purpurata Pyroderus scutatus Carpornis cucullata Phoenicircus carnifex Rupicola peruvianus Rupicola rupicola Neopelma pallescens Antilophia galeata Chiroxiphia pareola Chiroxiphia caudate Masius chrysopterus Ilicura miltaris Lepidothrix serena Lepidothrix coronate Lepidothrix nattereri Heterocercus linteatus Pipra filicauda 171 Pipra Fasciicauda 172 Pipra aureola Machaeropterus regulus Pipra chloromeros Pipra erythrocephala Pipra rubrocapilla Calyptomena viridis Psarisomus dalhousiae Eurylaimus steerii Cymbirhynchus macrorhynchos Eurylaiums ochoromalus Eurylaimus javanicus

a continued, fig. S1d b

70 60 50 40 30 20 10 0 MYA c

Figure S1. c) Majority-rule ultrametric consensus tree of a subset of the passerine species in this study. Legend in figure S1a. d 182

continued, fig. S1c

A B C D E F G H I J K L M N Neophron percnopterus Picumnus exilis Campephilus leucopogon Dryocopus pileatus Colaptes auratus Colaptes melanochloros Colaptes campestris Colaptes chrysoides Picus viridis Picus squamatus Sphyrapicus varius Melanerpes lewis Melanerpes formicivorus Melanerpes candidus Picoides tridactylus Dendrocopos major Picoides villosus Pteroglossus aracari Selenidera piperivora Ramphastos toco Ramphastos tucanus Trogon mesurus Apaloderma narina Sterna elegans Larus pipixcan Larus delawarensis Larus michaehellis Phaethon rubricauda Ptilinopus magnificus Ptilinopus jambu Ptilinopus solomonensis Ptilinopus pulchellus Phoeniconaias minor Phoenicopterus ruber Phoenicoparrus andinus Phoenicoparrus jamesi Phoenicopterus roseus Phoenicopterus chilensis Ciconia ciconia Fregata minor Eudocimus ruber Platalea ajaja Anser anser Anas platyrhynchos

90 80 70 60 50 40 30 20 10 0 MYA a

Figure S1. d) Majority-rule ultrametric consensus tree of the non-passerine species in this study. Legend in figure S1a. e 183

K phoenicopterone α-carotene 7,8 dihydro 7,8,7',8'-tetrahydro-zeaxanthin α-isocryptoxanthin β-cryptoxanthin E cis-lutein 7,8-dihydro-zeaxanthin α-cryptoxanthin (3R,3'R)- β-isocryptoxanthin β-carotene β-cryptoxanthin zeaxanthin lutein A 3'-dehydrolutein B F echinenone anhydrolutein isozeaxanthin canary xanthophyll B 7,8-dihydrolutein 4-hydroxyzeaxanthin canary xanthophyll A H piprixanthin 9-Z-7,8- dihydro- G 7,8-dihydro-3'- 3'-hydroxy-echinenone lutein canthaxanthin resonance dehydro-lutein 4-hydroxy-echinenone (3S,4R,3'S,6'R) 4-hydroxy-canary 4-hydroxylutein rhodoxanthin xanthophyll brittonxanthin rupicolin (3S,4R,3'R,6'R) (3S,3'R)-adonixanthin 4-hydroxylutein adonirubin J 13 cis-astaxanthin fritschiellaxanthin I 2,3-didehydro- papilioerythrinone (3S,3'S)-astaxanthin C papilioerythrinone (cymbirhynchin)

α-doradexanthin 3'-hydroxy-3-methoxy-canthaxanthin idoxanthin D (3S,6S,3'S,6'S) fucoxanthin rubixanthin gazaniaxanthin pompadourin cotingin tunaxanthin A xipholenin 2,3-didehydro- pompadourin N L M 2,3-didehydro-xipholenin (3R,6R,3'R,6'R) 4-oxo- 4-oxo- tunaxanthin F rubixanthin gazaniaxanthin

Figure S1. e) The full avian carotenoid metabolic network. Each of the compounds (circles) and enzymatic reactions (arrows) shown here occurs at least once in the 250 extant taxa in the data set. See Badyaev et al. [2] and Morrison & Badyaev [3] for network construction methods. Green shaded circles are potential dietary compounds. Each of the grey shaded regions identifies a pathway module (A-N) based on the pairs of compounds that co-occur at least 70% of the time in species and ancestral networks (phylogenetic profile similarity is greater than or equal to 0.7). See appendix S2 for more details on each of the modules. 184

References

1. Jetz W, Thomas GH, Joy JB, Redding DW, Hartmann K. Mooers AO. 2014 Global distribution and conservation of evolutionary distinctness in birds. Curr. Biol. 24, 919-930.

2. Badyaev AV, Morrison ES, Belloni V, Sanderson M.J. 2015 Tradeoff between robustness and elaboration in carotenoid networks produces cycles of avian color diversification. Biol. Direct 10, 45.

3. Morrison, ES, Badyaev AV. 2016 Structuring evolution: Biochemical networks and metabolic diversification in birds. BMC Evol. Biol. 16, 168. 185

Appendix S2: Each module (A-N) includes a group of compounds that co-occur in species and ancestral networks (n = 467) at least 70% of the time (phylogenetic profile similarity is greater or equal to 0.7). There are 34 pairs of compounds that co-occur more than 70% of the time in species and ancestral networks and this represents 5% of the 731 pairs of compounds that are potentially connected to each other via enzymatic reactions. In the following table, we list the number of compounds in each module, the identity of the dietary and derived compounds that comprise the module, the identity of the dietary compounds that are indirectly connected to the module, and the number of extant taxa in the dataset with the module. The distribution of these modules on the avian carotenoid network and across the phylogeny are shown in fig. S1.

Module Number of Included diet Connected diet Number Derived compounds Identity compounds compounds compounds of taxa

β-carotene, lutein, β-carotene, β-isocryptoxanthin, A 5 zeaxanthin, 92 canthaxanthin echinenone, canthaxanthin, β-cryptoxanthin 4-hydroxy-echinenone

β-cryptoxanthin, lutein, eaxanthin, B 3 β-cryptoxanthin, 3’-hydroxy-echinenone, 108 β-carotene adonirubin adonirubin

lutein, 4-hydroxyzeaxanthin, zeaxanthin, C 4 adonixanthin, idoxanthin, 101 astaxanthin β-carotene, astaxanthin β-cryptoxanthin

lutein, 3’-hydroxy-3-methoxy- zeaxanthin, D 3 none canthaxanthin, pompadourin, 10 β-carotene, 2,3-didehydro-pompadourin β-cryptoxanthin

7,8-dihydro-zeaxanthin, E 2 none 7,8,7’,8’-tetrahydrozeaxanthin 4 zeaxanthin

186

Module Number of Included diet Connected diet Number Derived compounds Identity compounds compounds compounds of taxa 3’-dehydrolutein, canary F 3 none xanthophyll A, canary lutein, zeaxanthin 86 xanthophyll B piprixanthin, G 2 none lutein, zeaxanthin 9 resonance structure 7,8-dihydrolutein, H 2 none lutein, zeaxanthin 10 9-Z-7,8-dihydro-lutein (3S,4R,3’S,6’R) 4- I 2 none hydroxylutein, lutein, zeaxanthin 78 α-doradexanthin (3S,4R,3’R,6’R) 4- J 2 none hydroxylutein, lutein, zeaxanthin 6 fritschiellaxanthin α-isocryptoxanthin, K 2 none α-carotene 1 phoenicopterone rubixanthin, L 2 rubixanthin none 15 4-oxo-rubixanthin gazaniaxanthin, M 2 gazaniaxanthin none 6 4-oxo-gazaniaxanthin tunaxanthin A, N 2 tunaxanthin A, tunaxanthin F none 1 tunaxanthin F

187

#NEXUS

[Appendix S3: tree file of ultrametric consensus phylogeny based on 1,000 randomly sampled trees from "Hackett All Species set of 10000 trees with 9993 OTUs" published in: The global diversity of birds in space and time; W. Jetz, G. H. Thomas, J. B. Joy, K. Hartmann, A. O. Mooers doi:10.1038/nature11631] [Subsampled and pruned from birdtree.org on 05/27/2015 19:17:49 ] [See methods for ultrametric consensus tree, which was the tree used in this study for ancestral reconstructions]

BEGIN trees;

TREE 'ultrametric_consensustree' = ((((((Malurus_melanocephalus:48.677113,(((((Motacilla_flava:27.950802, ((((Icteria_virens:15.463236,((Dendroica_petechia:4.7142577,(Setophaga _ruticilla:4.5002923,(Dendroica_palmarum:3.9044554,Dendroica_coronata: 3.9044554)19:0.5958367)18:0.21396549)17:3.645214,(Geothlypis_trichas:8 .081144,(Vermivora_ruficapilla:0.85301393,Vermivora_virginiae:0.853013 93)21:7.22813)20:0.27832773)16:7.103764)15:2.075819,(((Xanthocephalus_ xanthocephalus:11.6686,((Sturnella_superciliaris:4.09151,(Sturnella_ne glecta:2.5012267,Sturnella_magna:2.5012267)27:1.5902828)26:5.2565312,( Sturnella_militaris:3.6335943,Sturnella_bellicosa:3.6335943)28:5.71444 65)25:2.3205595)24:1.4191732,(((Psarocolius_montezuma:4.573248,Psaroco lius_wagleri:4.573248)31:2.5796156,(Cacicus_melanicterus:6.823368,((Ca cicus_cela:4.544551,Cacicus_uropygialis:4.544551)34:0.9790053,(Cacicus _haemorrhous:4.9277506,Cacicus_leucoramphus:4.9277506)35:0.5958057)33: 1.299812)32:0.32949537)30:5.075801,(Agelaius_phoeniceus:11.655789,((Ic terus_graduacauda:5.2392273,((Icterus_gularis:1.395929,Icterus_nigrogu laris:1.395929)40:1.2184397,(Icterus_galbula:2.3163757,(Icterus_pustul atus:1.6611015,Icterus_bullockii:1.6611015)42:0.65527415)41:0.29799291 )39:2.6248586)38:2.4556375,((Icterus_mesomelas:5.42669,(Icterus_pector alis:4.802242,Icterus_icterus:4.802242)45:0.6244482)44:1.3194717,(Icte rus_croconotus:4.508374,(Icterus_prosthemelas:4.043882,Icterus_cuculla tus:4.043882,Icterus_dominicensis:4.043882)47:0.46449256)46:2.2377875) 43:0.94870275)37:3.9609249)36:0.57287526)29:0.8591091)23:0.45051646,(C hlorospingus_pileatus:13.933265,(Emberiza_melanocephala:11.280937,Embe riza_citrinella:11.280937)49:2.652328)48:0.0)22:4.000765)14:2.4715362, ((Pheucticus_ludovicianus:15.913931,((Cardinalis_sinuatus:5.8153744,Ca rdinalis_cardinalis:5.8153744)53:8.032927,(Piranga_rubra:7.053678,(Pir anga_flava:6.1451974,(Piranga_olivacea:2.6073956,Piranga_ludoviciana:2 .6073956)56:3.537802)55:0.9084803)54:6.7946224)52:2.065631)51:1.492982 4,((Nesospiza_acunhae:12.752187,(Paroaria_coronata:10.489793,Coereba_f laveola:10.489793)59:2.2623932)58:0.85370356,(Ramphocelus_dimidiatus:1 1.328818,Sicalis_flaveola:11.328818)60:2.277072)57:3.801024)50:2.60367 73)13:5.043211,((Fringilla_coelebs:7.271644,Fringilla_montifringilla:7 .271644)62:16.548254,((((Mycerobas_affinis:10.461879,Mycerobas_melanoz anthos:10.461879,Mycerobas_icterioides:10.461879,Mycerobas_carnipes:10 .461879)66:3.635935,(Coccothraustes_vespertinus:7.225395,Coccothrauste s_abeillei:7.225395)67:6.872419)65:4.574753,(Rhynchostruthus_socotranu s:16.35011,Vestiaria_coccinea:16.35011,Bucanetes_githagineus:16.35011, (Pinicola_enucleator:11.247011,(Pyrrhoplectes_epauletta:7.067632,(Pyrr 188

hula_erythrocephala:4.6002183,Pyrrhula_erythaca:4.6002183,Pyrrhula_pyr rhula:4.6002183,Pyrrhula_aurantiaca:4.6002183)71:2.467414)70:4.1793785 )69:5.1030974,((Rhodopechys_obsoletus:7.3916173,(Carduelis_chloris:2.2 084913,(Carduelis_sinica:1.3985488,Carduelis_spinoides:1.3985488)75:0. 8099425)74:5.183126)73:2.983084,(((Serinus_mozambicus:7.801493,(Serinu s_canaria:5.9044666,(Serinus_pusillus:4.8721323,Serinus_serinus:4.8721 323)80:1.0323343)79:1.8970263)78:0.079866506,(Carduelis_cannabina:6.53 9667,(Carduelis_spinus:5.2019477,(Carduelis_tristis:4.5059032,(Carduel is_atrata:1.43746,Carduelis_cucullata:1.43746)84:3.0684435)83:0.696044 15)82:1.3377196)81:1.3416926)77:0.52299184,(((Carduelis_hornemanni:0.6 039879,Carduelis_flammea:0.6039879)87:4.5606737,(Loxia_curvirostra:1.6 362138,Loxia_leucoptera:1.6362138)88:3.5284479)86:2.3670604,(Carduelis _citrinella:5.7606006,Carduelis_carduelis:5.7606006)89:1.7711219)85:0. 87262934)76:1.9703498)72:5.9754066,((Haematospiza_sipahi:11.830967,Ura gus_sibiricus:11.830967)91:1.7957532,((Carpodacus_rubicilloides:9.0762 37,(Carpodacus_trifasciatus:6.7968693,Carpodacus_thura:6.7968693,Carpo dacus_pulcherrimus:6.7968693,Carpodacus_roseus:6.7968693)94:2.2793677) 93:4.208863,(Carpodacus_mexicanus:12.598403,Carpodacus_nipalensis:12.5 98403)95:0.68669724)92:0.34162048)90:2.7233877)68:2.3224587)64:3.08519 84,(Euphonia_saturata:9.894907,Euphonia_laniirostris:9.894907)96:11.86 2859)63:2.0621333)61:1.2339032)12:2.8969996)11:4.565783,(((Amandava_su bflava:6.8754582,Amandava_amandava:6.8754582)99:5.8399525,((Erythrura_ psittacea:6.340413,Erythrura_gouldiae:6.340413)101:4.951838,(Taeniopyg ia_guttata:7.2955766,Neochmia_ruficauda:7.2955766)102:3.9966745)100:1. 4231598)98:7.2419944,((Ploceus_bicolor:10.011748,Ploceus_cucullatus:10 .011748,Ploceus_nelicourvi:10.011748,Ploceus_sakalava:10.011748,Ploceu s_philippinus:10.011748,(Ploceus_capensis:0.15637651,Ploceus_velatus:0 .15637651)105:9.855371)104:0.786907,((Euplectes_afer:6.4070783,((Euple ctes_capensis:3.7889807,(Euplectes_macroura:3.1984756,Euplectes_axilla ris:3.1984756)110:0.5905052)109:0.6379596,(Euplectes_ardens:2.6738596, Euplectes_orix:2.6738596)111:1.753081)108:1.9801377)107:2.7034297,(Fou dia_madagascariensis:6.97705,(Quelea_quelea:4.706626,Quelea_cardinalis :4.706626,Quelea_erythrops:4.706626)113:2.270424)112:2.1334581)106:1.6 881473)103:9.15875)97:12.559179)10:7.4317603,((Regulus_satrapa:30.8887 4,Regulus_regulus:30.88874)115:8.570384,((Bombycilla_cedrorum:10.15066 ,(Bombycilla_japonica:6.6241407,Bombycilla_garrulus:6.6241407)118:3.52 65193)117:27.574724,(Tichodroma_muraria:35.471947,(Turdus_merula:23.81 137,(Erithacus_rubecula:15.772334,(Luscinia_calliope:14.559886,(Ficedu la_zanthopygia:13.760958,Tarsiger_chrysaeus:13.760958)123:0.79892844)1 22:1.2124482)121:8.039037)120:11.660577)119:2.2534366)116:1.7337396)11 4:0.4892205)9:0.88001376,((Aegithalos_caudatus:30.026533,(Zosterops_ja ponicus:16.567932,(Leiothrix_lutea:2.7130141,Leiothrix_argentauris:2.7 130141)127:13.8549185)126:13.4586)125:8.698239,(Parus_caeruleus:17.311 525,(Parus_ater:14.867826,(Parus_major:13.256795,Parus_spilonotus:13.2 56795)130:1.6110315)129:2.4436991)128:21.413246)124:2.1035864)8:5.1550 975,(Notiomystis_cincta:42.701595,(((Oriolus_cruentus:10.594408,Oriolu s_traillii:10.594408)134:1.3424596,(Oriolus_oriolus:11.1245775,Oriolus _xanthornus:11.1245775)135:0.8122908)133:23.89786,(Pericrocotus_flamme us:32.187843,(Telophorus_zeylonus:18.05768,Telophorus_sulfureopectus:1 8.05768)137:14.130165)136:3.6468823)132:6.8668694)131:3.2818592)7:2.69 36557)6:12.282311,(((Tyrannus_vociferans:29.564753,Iodopleura_isabella e:29.564753)140:1.3842931,(((Ampelioides_tschudii:15.303812,(Pipreola_ 189

formosa:11.958488,Pipreola_chlorolepidota:11.958488,Pipreola_aureopect us:11.958488,Pipreola_whitelyi:11.958488)144:3.345323)143:6.730646,(Am pelion_rufaxilla:20.37828,((Snowornis_subalaris:18.293236,((Phibalura_ flavirostris:14.359037,(Cotinga_maculata:8.858968,Cotinga_cotinga:8.85 8968,Cotinga_amabilis:8.858968)150:5.50007,(Procnias_tricarunculatus:1 3.389689,(Porphyrolaema_porphyrolaema:12.513855,(Xipholena_atropurpure a:2.3871005,Xipholena_punicea:2.3871005,Xipholena_lamellipennis:2.3871 005)153:10.126755)152:0.87583375)151:0.9693487,(Lipaugus_streptophorus :8.65216,Tijuca_atra:8.65216)154:5.7068787)149:0.84130037,(Haematoderu s_militaris:13.62769,(Querula_purpurata:10.133677,Pyroderus_scutatus:1 0.133677)156:3.494014)155:1.5726479)148:3.0928984)147:1.1688815,(Carpo rnis_cucullata:17.022583,(Phoenicircus_carnifex:7.0549707,(Rupicola_pe ruvianus:5.0530977,Rupicola_rupicola:5.0530977)159:2.0018733)158:9.967 611)157:2.4395356)146:0.91616344)145:1.6561761)142:4.8753905,(Neopelma _pallescens:15.075984,(((Antilophia_galeata:3.8638563,(Chiroxiphia_par eola:2.3791919,Chiroxiphia_caudata:2.3791919)164:1.4846642)163:4.44855 26,(Masius_chrysopterus:7.137043,Ilicura_militaris:7.137043)165:1.1753 657)162:3.3396082,((Lepidothrix_serena:4.1321783,(Lepidothrix_coronata :2.531288,Lepidothrix_nattereri:2.531288)168:1.6008906)167:4.4725375,( Heterocercus_linteatus:7.778469,((Pipra_filicauda:1.5725914,(Pipra_fas ciicauda:0.6901019,Pipra_aureola:0.6901019)172:0.8824895)171:5.643419, (Machaeropterus_regulus:6.281017,(Pipra_chloromeros:3.904027,(Pipra_er ythrocephala:1.9450582,Pipra_rubrocapilla:1.9450582)175:1.9589686)174: 2.37699)173:0.9349931)170:0.56245905)169:0.82624704)166:3.0473006)161: 3.423967)160:11.833864)141:4.039198)139:22.474743,(Calyptomena_viridis :35.76839,(Psarisomus_dalhousiae:14.793243,(Eurylaimus_steerii:11.3355 41,(Cymbirhynchus_macrorhynchos:8.269681,(Eurylaimus_ochromalus:3.3672 233,Eurylaimus_javanicus:3.3672233)180:4.9024577)179:3.0658598)178:3.4 577024)177:20.97515)176:17.655396)138:7.5356345)5:12.790263,(Neophron_ percnopterus:72.64189,(((Picumnus_exilis:21.821146,((Campephilus_leuco pogon:13.8164625,((Dryocopus_pileatus:8.752832,(Colaptes_auratus:5.253 6206,(Colaptes_melanochloros:3.8341954,Colaptes_campestris:3.8341954,C olaptes_chrysoides:3.8341954)190:1.4194252)189:3.4992115)188:1.7515833 ,(Picus_viridis:6.611415,Picus_squamatus:6.611415)191:3.8930006)187:3. 3120468)186:1.2849336,((Sphyrapicus_varius:12.468637,(Melanerpes_lewis :9.120327,(Melanerpes_formicivorus:6.4333744,Melanerpes_candidus:6.433 3744)195:2.6869524)194:3.3483105)193:1.5934871,(Picoides_tridactylus:1 2.228185,(Dendrocopos_major:9.583935,Picoides_villosus:9.583935)197:2. 6442506)196:1.8339394)192:1.0392712)185:6.71975)184:11.919212,((Pterog lossus_aracari:13.471195,Selenidera_piperivora:13.471195)199:1.9844081 ,(Ramphastos_toco:6.3087997,Ramphastos_tucanus:6.3087997)200:9.146804) 198:18.284754)183:34.41903,(Trogon_mesurus:25.555859,Apaloderma_narina :25.555859)201:42.603527)182:4.4825053)181:1.1077957)4:1.6553385,(Ster na_elegans:17.701286,(Larus_pipixcan:3.6136222,(Larus_delawarensis:1.2 855505,Larus_michahellis:1.2855505)204:2.3280718)203:14.087665)202:57. 70374)3:0.5967932,(Phaethon_rubricauda:71.607086,(Ptilinopus_magnificu s:17.918142,(Ptilinopus_jambu:12.966486,(Ptilinopus_solomonensis:9.466 276,Ptilinopus_pulchellus:9.466276)208:3.50021)207:4.9516563)206:53.68 8946,(Phoeniconaias_minor:24.492142,((Phoenicopterus_ruber:9.94238,(Ph oenicoparrus_andinus:5.4425273,Phoenicoparrus_jamesi:5.4425273)212:4.4 998527)211:5.7403045,(Phoenicopterus_roseus:10.768596,Phoenicopterus_c hilensis:10.768596)213:4.9140887)210:8.809456)209:47.11495,(Ciconia_ci 190

conia:62.862747,(Fregata_minor:61.643303,(Eudocimus_ruber:49.89578,Pla talea_ajaja:49.89578)216:11.747524)215:1.219443)214:8.744343)205:4.394 7296)2:14.361656,(Anser_anser:76.001816,Anas_platyrhynchos:76.001816)2 17:14.361656)1;

END;