CURRICULUM VITAE
Nicholas R. Friedman Address Department of Biological Sciences
Phone: (410) 455-1704
1000 Hilltop Circle
Email:
[email protected]
Baltimore, MD 21250
Education St. Mary’s College of Maryland, B.A., 2006 University of Maryland, Baltimore County, PhD Candidate, Biology, Fall 2007-present
Professional Experience University of Maryland, Baltimore County Laboratory Technician, 2007 Smithsonian Migratory Bird Center Field Assistant, 2007 Atlantic Bird Observatory Field Assistant, 2006 St. Mary’s College of Maryland Research Assistant, 2006 St. Mary’s College of Maryland Research Assistant, 2005
Publications Friedman. Nicholas R., L. M. Kiere, K. E. Omland. Convergent gains of red carotenoidbased coloration in the New World blackbirds. Auk 128(4):678-687. Jacobsen, Frode, Nicholas R. Friedman, and Kevin E. Omland. 2010. Congruence between nuclear and mitochondrial DNA: combination of multiple nuclear introns resolves a well-supported phylogeny of New World orioles (Icterus). Molecular Phylogenetics and Evolution 56(1):419-427. Friedman, Nicholas R., Christopher M. Hofmann, Beatrice Kondo, Kevin Omland. 2009. Correlated evolution of migration and dichromatism in the New World orioles (Icterus). Evolution 63(12):3269-3274. Leonard Duncan, Ichiro Nishii, Alexandra Harryman, Stephanie Buckley, Alicia Howard, Nicholas R. Friedman, Stephen M. Miller. 2007. The VARL gene family and the evolutionary origins of the master cell-type regulatory gene, regA, in Volvox carteri. Journal of Molecular Evolution 65:1-11. J. Jordan Price, Nicholas R. Friedman, Kevin E. Omland. 2007. Song and plumage evolution in the New World orioles show similar lability and convergence in patterns. Evolution 61(4):850-863. (Cover design by N. R. Friedman)
Conference Presentations Friedman, N. R., S. Manor, K. E. Omland. Examining rates of character change and diversification in blackbirds: why does red coloration evolve repeatedly? 1st Joint Conference on Evolutionary Biology, Ottowa, ON, July 2012. Oral Presentation.
Friedman, N. R., K. J. McGraw, K. E. Omland. The evolution of carotenoid pigmentation and coloration in orioles. Animal Behavior Society, Bloomington, IN, July 2011. Poster Presentation. Friedman, N. R., K. J. McGraw, K. E. Omland. Multiple gains of red coloration in blackbirds: convergence and parallelism in the evolution of carotenoid use. Animal Behavior Society, Williamsburg, VA, July 2010. Oral Presentation. Omland, K. E., J. J. Price, C. M. Hofmann, N. R. Friedman, F. Jacobsen. Temperate biases in studies of sexual dimorphism: plumage and song evolution in orioles. Animal Behavior Society, Williamsburg, VA, July 2010. Oral Presentation. Friedman, N. R., C. M. Hofmann, B. Kondo, and K. E. Omland. Correlated evolution of migration and sexual dichromatism in the New World orioles. American Ornithologists’ Union, Philadelphia, PA, August 2009. Oral Presentation. Omland, K. E., J. J. Price, S. M. Lanyon, C. M. Hofmann, N. R. Friedman, B. Kondo, and F. Jacobsen. Temperate zone biases in studies of sexual dimorphism: phylogenetic studies of plumage and song evolution in New World orioles. American Ornithologists’ Union, Philadelphia, PA, August 2009. Price, J. J., N. R. Friedman and K. E. Omland. Song evolution in the New World orioles. Animal Behavior Society, Burlington, VT, July 2007. Price, J. J., N. R. Friedman and K. E. Omland. Song and plumage evolution in the New World orioles show similar lability and convergence in patterns. American Ornithologists’ Union, Laramie, WY, August 2007.
Invited Seminars
Friedman, N. R., K. J. McGraw, K. E. Omland. Which of these birds is just like the other? Convergence and parallelism in blackbird color evolution. University of Maryland Baltimore County, Baltimore, MD. February 1, 2012. Friedman, N. R., K. J. McGraw, K. E. Omland. Color and pigment evolution in the New World blackbirds. Smithsonian Migratory Bird Center, Washington, DC, February 4, 2011.
Service Journal reviewer: The Auk, Journal of Avian Biology
Research Grants and Awards Society for the Study of Evolution Hamilton Award Finalist, 2012 Travel to Comparative Methods and Macroevolution in R, 2011 Sigma Xi Grant-in-aid of Research, 2010 Maryland Ornithological Society Research Grant, 2010 Sigma Xi Grant-in-aid of Research, 2005
Society Affiliations American Ornithologists’ Union Animal Behavior Society Society for the Study of Evolution Sigma Xi: The Scientific Research Society
ABSTRACT
Plumage color evolution in birds has been the focus of theoretical and empirical research on sexual selection since Darwin. Many of the yellow, orange, and red hues seen in bird plumage are the result of carotenoid pigmentation. While a great number of recent studies have examined the functions of carotenoid-based plumage coloration in a single species, few have examined the evolutionary history of this trait in a comparative phylogenetic context. Using the New World blackbirds as a model clade, I focus on two questions that a comparative phylogenetic approach can uniquely address. First, what is the history of evolutionary change in carotenoid color that led to the colors seen in extant blackbird taxa? Second, by what proximate mechanisms have carotenoid pigments evolved? In Chapter 1, I present an ancestral state reconstruction of carotenoid-based plumage coloration across the Icterid phylogeny, based on reflectance measurements of museum skins. My results show robust evidence that red coloration was gained repeatedly from a yellow common ancestor. In Chapter 2, I used pigment biochemistry of meadowlark (Sturnella) and Cacique (Cacicus) feathers to test whether independent gains of red coloration are the result of parallel or convergent metabolic mechanisms. Meadowlarks have evolved red coloration using a different set of carotenoids than caciques, but the caciques have evolved the same set of carotenoids twice. This suggests that red coloration evolved by convergent evolution among different blackbird clades, but evolved by parallel evolution within the caciques. Lastly, in Chapter 3 I examine the relationship between color and carotenoid pigmentation in orioles, a blackbird clade in which orange has been gained at least twice independently from a yellow common
ancestor. I found red-producing keto-carotenoids only in orange species and never in yellow species. This result is a striking contrast to our expectation for a continuous gradient of a carotenoid pigment concentration. These results suggest that repeated gains of C4-oxygenation ability best explain evolutionary changes in orange coloration in orioles. To summarize, I showed using phylogenetic comparative methods that blackbirds have repeatedly evolved towards redder carotenoid coloration. Using HPLC biochemistry, I showed that each of these gains of orange and red coloration is likely the result of a gain of C4-oxygenation ability. The prevalence of gains of orange and red coloration suggests that there may be a directional bias towards evolving longerwavelength carotenoid plumage. The research presented in these chapters provides the phylogenetic framework necessary for future studies to examine the functional causes underlying the repeated evolution of carotenoid-based coloration.
The evolution of carotenoid coloration and pigmentation in the New World blackbirds
By Nicholas R. Friedman
Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, Baltimore County in partial fulfillment of the requirements for the degree of Doctor of Philosophy
2013
UMI Number: 3563319
All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.
UMI 3563319 Published by ProQuest LLC (2013). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code
ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, MI 48106 - 1346
© Copyright by Nicholas R. Friedman 2013
Acknowledgements
I would first like to thank my advisor and mentor, Kevin Omland, for his support throughout my PhD. Working with Kevin over the last six years has been a real privilege. Kevin’s patience and encouragement helped guide me through my PhD, and his thoughtfully skeptical approach to ideas old and new helped shape my development as a scientist. I would like to thank my dissertation committee members for their constructive feedback and encouragement – their varying perspectives helped me both shape my questions and tell my story. I would like to thank Jordan Price for his continued mentorship and input, as well as my biology professors at St. Mary’s and UMBC. I would also like to thank my high school biology and English teachers, Tony and Sarah Everdell, without whose instruction I would have never dreamed of performing or writing about my research in evolutionary biology. My colleagues and friends in the Omland lab (Frode Jacobsen especially) helped me through troubleshooting and interesting discussions, as did Chris Hofmann and Lynna Kiere, whose efforts laid the foundation for this research. My collaborator Kevin McGraw and his students and postdocs helped me learn their analytical chemistry techniques and eventually use them as effective tools for comparative biology. I would like to thank Kurt Krosnowski, Kate Feller, Megan Porter and Jenny Gumm for their scientific advice and friendship. Lastly I would like to thank my friends and family, without whose support I would never have dared to begin, much less finish, a PhD.
TABLE OF CONTENTS
Chapter 1: Convergent gains of red carotenoid-based coloration in the New World blackbirds.............................................................................................................................1 Abstract....................................................................................................................2 Introduction..............................................................................................................3 Methods....................................................................................................................6 Color Measurement......................................................................................6 Character Scoring.........................................................................................7 Ancestral State Reconstruction..................................................................10 Results....................................................................................................................12 Color Variation..........................................................................................12 Ancestral States..........................................................................................15 Discussion..............................................................................................................15 Repeated Gains of Red Coloration............................................................17 Ancestral Yellow Coloration.....................................................................19 Conclusions................................................................................................20 Acknowledgements................................................................................................21 Literature Cited......................................................................................................21 Chapter 2: Convergence and parallelism in the evolution of red carotenoid pigmentation in caciques and meadowlarks.............................................................................................27 Abstract..................................................................................................................28 Introduction............................................................................................................29
Methods..................................................................................................................33 Sampling and Analysis..............................................................................33 Character Scoring......................................................................................34 Ancestral State Reconstruction..................................................................35 Results....................................................................................................................37 Discussion..............................................................................................................42 Convergence and Parallelism.....................................................................42 Causes of Repeated Gains of Red..............................................................44 Yellow Pigments in Yellow Feathers........................................................46 No Carotenoids in Black Feathers.............................................................47 Yellow Pigments in Yellow Feathers........................................................48 Conclusion.................................................................................................48 Acknowledgements................................................................................................49 Literature Cited......................................................................................................50 Appendix: Voucher specimen data........................................................................57 Chapter 3: History and mechanisms of carotenoid plumage evolution in the New World orioles.................................................................................................................................58 Abstract..................................................................................................................59 Introduction............................................................................................................60 Methods..................................................................................................................64 Sampling....................................................................................................64 HPLC Biochemistry...................................................................................65 Phylogenetic Comparative Methods..........................................................66
Results....................................................................................................................68 Carotenoid Compounds Observed.............................................................68 Color and Keto-carotenoid Concentration.................................................69 Ancestral State Reconstruction..................................................................70 Discussion..............................................................................................................72 Carotenoid Pigments in Oriole Feathers....................................................73 Why No Red Orioles?................................................................................75 History of Carotenoid Pigmentation in Orioles.........................................77 Conclusions................................................................................................79 Literature Cited..........................................................................................80 Appendix: Voucher specimen data............................................................86 Summary and Overall Conclusions...................................................................................88 Convergent gains of red carotenoid-based coloration in the New World blackbirds...........................................................................................................................89 Convergence and parallelism in the evolution of red carotenoid pigmentation in caciques and meadowlarks.................................................................................................90 History and mechanisms of carotenoid plumage coloration in the New World orioles (Icterus)..................................................................................................................91 Overall Summary and Future Questions................................................................92 References..............................................................................................................94
LIST OF FIGURES Figure 1-1. Distribution of mean spectral location values across the New World blackbirds excluding the orioles (a) and across the orioles (b)..........................................11 Figure 1-2. Cluster analysis of two measures of carotenoid coloration: spectral location and yellow chroma.............................................................................................................14 Figure 1-3. Ancestral state reconstruction of our composite scoring of carotenoid coloration across Icteridae.................................................................................................16 Figure 2-1. Gains and losses of carotenoid pigments reconstructed using parsimony on the Icterid molecular phylogeny........................................................................................41 Figure 3-1. Abbreviated diagrams of carotenoid pigments commonly observed in bird plumage, and the hypothesized reactions responsible for their metabolism (adapted from Andersson et al. 2007).......................................................................................................62 Figure 3-2. Comparison of breast plumage coloration and the total concentration of ketocarotenoids in feathers sampled from that same plumage across oriole species...............71 Figure 3-3. Parsimony reconstruction of ancestral color and pigment character states on the Jacobsen et al. (2010) Icterus phylogeny, with un-sampled taxa pruned, and using a composite of keto-carotenoid characters...........................................................................74
LIST OF TABLES Table 1-1. Character matrix used in ancestral state reconstructions....................................8 Table 1-2. Results of ancestral state reconstructions in terms of gains and losses, listed across scoring methods......................................................................................................18 Table 2-1. Pigments identified from feather extracts and their retention times, as well as their absorbance peaks.......................................................................................................38 Table 2-2. Average pigment concentrations form each species examined........................38 Table 3-1. Concentrations of carotenoid pigments extracted from oriole breast feathers, and color characters scored from reflectance measurements taken from specimens prior to feather removal..................................................................................................................67 Table 3-2. Results of comparative phylogenetic analyses using phylogenetic generalized least squares (PGLS) and Pagel’s discrete method (Pagel 1994)......................................67
&KDSWHU
1
CHAPTER 1: CONVERGENT GAINS OF RED CAROTENOID-BASED COLORATION IN THE NEW WORLD BLACKBIRDS (The Auk 128(4):1-10, 2011) Nicholas R. Friedman1*†, Lynna M. Kiere2, and Kevin E. Omland1
1
Department of Biological Sciences, University of Maryland – Baltimore County. 1000 Hilltop Circle, Baltimore, MD 21250 2
Laboratorio de Conducta Animal, Instituto de Ecología, Universidad Nacional Autónoma de México, Ciudad Universitaria, México
* To whom correspondence should be addressed † email:
[email protected]
&KDSWHU
2
Abstract Many birds have colorful plumage ornaments that utilize carotenoid pigments, and these are often displayed in signaling contexts. Researchers in behavioral ecology have focused on examining carotenoids in general, and red carotenoids in particular, as they may be an honest index of individual condition or quality. However, few studies have examined the evolutionary changes in carotenoid-based coloration across a phylogeny. In this study, we used reflectance spectrometry to examine carotenoid-based coloration across the New World blackbirds (Icteridae). We scored discrete character states based on these measurements and mapped them onto the icterid phylogeny. Our results indicate that red coloration has been gained six times in the blackbirds from a common ancestor that exhibited yellow ornamentation. This result was supported by both parsimony and likelihood methods of ancestral state reconstruction, and by each of three different scoring methods. Thus multiple lineages of icterids have convergently evolved red patches from a common ancestor that most likely used yellow. Several other studies have observed repeated gains of red coloration, suggesting that our observations may reflect a directional trend common among avian clades. Keywords: Ancestral state reconstruction, Carotenoids, Coloration, Convergent evolution, New World blackbirds
&KDSWHU
3
Introduction Different species of birds often vary greatly in plumage coloration, even when closely related. Understanding the historical origins of this variation can inform the study of both the mechanisms that produce plumage coloration and the functions that such coloration may perform. Carotenoid pigments are responsible for many of the red, orange, and yellow hues that are visible in feathers across a wide range of avian taxa (McGraw 2006a). Feather patches ornamented with carotenoids are used by many taxa in a signaling context (e.g., Murphy et al. 2009; Yasukawa et al. 2009). Studies of intraspecific variation in carotenoid-based coloration have shown that in many cases this coloration is dependent on the signaler’s condition (Dale 2006 and refs. therein). Indeed, saturation of carotenoid-based coloration may often be a sexually selected trait, as studies of house finches (Carpodacus mexicanus) and widowbirds (Euplectes) have demonstrated (Hill 1990; Pryke et al. 2001). However, there have been few studies of how carotenoid-based colors vary among species, or how this variation has evolved. The mechanisms underlying carotenoid-based coloration are complex (McGraw 2006a). While birds synthesize melanins endogenously (McGraw 2006b), they must obtain carotenoids from dietary sources (Fox 1962). Birds may deposit the carotenoid compounds they acquire from the diet (e.g., lutein, zeaxanthin) into their feathers directly (Brush 1990). For example, yellow warblers (Dendroica petechia) use only lutein and zeaxanthin to color their yellow feathers (McGraw 2006). Alternatively, some birds may chemically alter dietary carotenoids to form new modified carotenoid compounds that often have different absorbance spectra than the originals (e.g., astaxanthin, canthaxanthin; Fox et al. 1969; Inouye et al. 2001).
&KDSWHU
4
Many red-ornamented birds produce their plumage coloration by modifying their predominantly yellow dietary carotenoids to form red ones (Brush and Power 1976; Hill 1996 and refs. therein). Among these, house finches provide a well-studied example of this phenomenon. Despite the presence of only four carotenoid pigments in their diet and plasma, house finches deposit eleven unique carotenoid compounds into their feathers, six of which are red keto-carotenoids. Several researchers have suggested that the gain of such pigment modification mechanisms may be responsible for evolutionary transitions from yellow to red plumage (Hudon 1991; Kiere et al. 2009). Indeed, chemical analysis of feather pigmentation in bishops (Euplectes) has demonstrated that red coloration has evolved in that clade through a gain of red keto-carotenoids (Andersson et al. 2007; Prager and Andersson 2010). Many recent studies have focused on describing the proximate and ultimate mechanisms underlying intraspecific variation in carotenoid coloration (see McGraw 2006a; Dale 2006). However, studies are needed that address variation in carotenoid coloration from a phylogenetic perspective. By examining the historical changes in coloration that have led to the variation seen in extant species, phylogenetic studies can reveal interesting patterns of trait evolution (Pagel 1999). For example, a series of studies examining song and plumage evolution in two blackbird clades, the New World orioles and the caciques and oropendolas (Cacicus, Psarocolius, Gymnostinops and Ocyalus) found that degrees of homoplasy appeared to vary with mating system (Omland and Lanyon 2000; Price et al. 2007; Price and Whalen 2009). Studies of color and tail evolution in the bishops and widowbirds have found evidence of an apparent trend towards gains of red coloration (Prager and Andersson 2009, 2010). Patterns such as
&KDSWHU
5
these demand further investigation, and provide a unique opportunity to study the implications of sexual selection at a macroevolutionary scale. The New World Blackbirds (Icteridae) display a full range of carotenoid-based plumage colors and ornaments, making them an ideal system in which to study the evolution of these characteristics. Many previous studies have used Icteridae as a model clade to study the evolution of mating systems, migratory behavior, plumage coloration, and vocal behavior (Searcy and Yasukawa 1999; Johnson and Lanyon 2000; Omland and Lanyon 2000; Price and Lanyon 2002a; Hofmann et al. 2006; Kondo and Omland 2007; Price et al. 2007; Friedman et al. 2009; Price and Whalen 2009). Furthermore, many previous studies have examined the phylogenetic relationships of the New World blackbirds, which are supported by both mitochondrial and nuclear markers (Johnson and Lanyon 1999; Lanyon and Omland 1999; Omland et al. 1999; Price and Lanyon 2002b; Allen and Omland 2003; Lanyon and Barker 2007; Barker et al. 2008; Jacobsen et al. 2010). Model clades such as the Icteridae provide a unique opportunity for comparative studies to utilize existing data (e.g., Friedman et al. 2009). In this study, we build upon previous work on carotenoid-based coloration in the New World orioles (Hofmann et al. 2006) and the Cacicus group (Kiere et al. 2009), and expand this taxonomic coverage to include the entire blackbird family. Taxa with red carotenoid-colored plumage are rare among icterids, making up less than 20 of 101 recognized species (Jaramillo and Burke 1999; Clements 2007). However, such red-displaying taxa are found in three of four clades in the Icteridae (Lanyon and Omland 1999; Lanyon and Barker 2007). This pattern is most likely due to homoplasy, the presence of similar traits not derived from a common ancestor. However,
&KDSWHU
6
it is unclear whether it was convergent gains or losses that dominated the history of red plumage in blackbirds. Furthermore, it is as yet unclear which hue the most recent common ancestor of the blackbirds might have displayed. In this study, we used reflectance spectrometry and ancestral state reconstruction to examine patterns in the evolution of red coloration across the New World blackbirds. Ours goals in this study were threefold: 1) to infer ancestral states for carotenoid coloration across the Icteridae, 2) to distinguish whether carotenoid-based red coloration was repeatedly gained or lost in Icteridae, and 3) to determine the color of the ancestral blackbird.
Methods Color Measurement We measured carotenoid-based coloration across Icteridae using reflectance spectrometry. Reflectance spectrometry is a widely used method for quantifying color, as it describes color without respect to the spectral sensitivities of any particular visual system (Andersson and Prager 2006). Many previous studies performed by our research group have used reflectance spectrometry to measure feather coloration (Hofmann et al. 2006; Hofmann et al. 2007a; Hofmann et al. 2007b; Kiere et al. 2009). We built on these previous studies by using reflectance spectra from studies that measured color variation in the orioles (Hofmann et al. 2006) and the caciques (Kiere et al. 2009). Consequently, to produce new reflectance spectra for the remaining icterids we followed the techniques used in these previous studies.
&KDSWHU
7
Specifically, we used an Ocean Optics USB2000 reflectance spectrometer in conjunction with a PX-2 pulsed xenon light source to perform these measurements (Ocean Optics, Inc., Dunedin, FL), calibrated against a white Spectralon® standard (Labsphere, North Sutton, NH). We measured approximately five male museum skins for each taxon (mean 4.6 ± 1.21; Table 1). For taxa in which a breeding season has been described (Jaramillo and Burke 1999), we preferentially sampled specimens collected during this time of the year. Furthermore, in order to avoid measuring birds displaying immature plumage, we attempted to use individuals collected after their second year, as identified by molt or plumage (Pyle 1997; Jaramillo and Burke 1999). Finally, we attempted to sample individuals from multiple locations across each taxon’s range. We performed three replicate measurements of each body region (see Andersson and Prager 2006) that appeared red, orange or yellow. We also measured feathers that were blond, rusty, or chestnut to determine the presence or absence of carotenoid spectra (see below). We recorded and processed the resulting spectra using OOIBase software and a supplemental program provided by T.H. Chiou and T.W. Cronin. Character Scoring Scoring methods in color measurement are vital to studies of feather coloration, and many approaches have been described (Montgomerie 2006). In this study, we scored two characters describing feather color: 1) pigment use as either “carotenoid” or “melanin”, and 2) carotenoid coloration as “red”, “orange”, or “yellow”. The differences between colors produced by carotenoid and melanin pigments are qualitative, and can be diagnosed based on reflectance spectra (Hofmann et al. 2007a, b). We scored taxa that were described as entirely black as using melanic pigmentation (Jaramillo and Burke
&KDSWHU
8
Table 1: Character matrix used in ancestral state reconstructions. Only putatively carotenoid-colored feather patches were measured using reflectance spectrometry in this study (E = lesser wing coverts, B = breast, R = rump, N = nape, T = dorsal side of rectrices, Y = belly, P = pectoral tuft). Taxon
n
Agelaius assimilis
2
E
Red
Red
Red
Taxon
n
Icterus gularis yucatanensis
5
B
Orange
Yellow
Agelaius cyanopus
N/A
N/A
N/A
N/A
Orange
Absent
Icterus icterus ridgwayi
5
B
Orange
Yellow
Agelaius humeralis
5
E
N/A
Orange
N/A
Absent
Icterus jamacaii croconotus
5
B
Orange
Yellow
Agelaius icterocephalus
6
B
Orange
Yellow
Yellow
Yellow
Icterus jamacaii strictifrons
4
B
Orange
Yellow
Agelaius phoeniceus
7
E
Orange
Red
Red
Red
Icterus laudabilis
5
E
Orange
Yellow
Orange
Agelaius ruficapillus
5
B
Agelaius thilius
4
E
N/A
N/A
Absent
Icterus leucopteryx leucopteryx
5
B
Yellow
Yellow
Yellow
Yellow
Yellow
Yellow
Icterus maculialatus
5
B
Yellow
Yellow
Agelaius tricolor
5
E
Yellow
Red
Red
Red
Icterus mesomelas mesomelas
5
B
Yellow
Yellow
Agelaius xanthomus
5
Yellow
E
Yellow
Yellow
Red
Icterus mesomelas salvinii
5
B
Yellow
Yellow
Agelaius xanthophthalmus
Yellow
N/A
N/A
N/A
N/A
Absent
Icterus mesomelas taczanowskii
1
B
Yellow
Yellow
Amblycercus holosericeus
Yellow
N/A
N/A
N/A
N/A
Absent
Icterus nigrogularis nigrogularis
5
B
Yellow
Yellow
Yellow
3
B
Red
Red
Red
Icterus nigrogularis trinitatis
1
B
Yellow
Yellow
Yellow
10
R
Yellow
Yellow
Yellow
Icterus oberi
5
B
N/A
N/A
Absent
Cacicus chrysonotus
4
R
Yellow
Yellow
Yellow
Icterus parisorum
5
B
Yellow
Yellow
Yellow
Cacicus chrysopterus
3
R
Yellow
Yellow
Yellow
Icterus pectoralis
5
B
Yellow
Yellow
Yellow
Cacicus haemorrhous
5
R
Red
Red
Red
Icterus pustulatus formosus
4
B
Yellow
Yellow
Yellow
Cacicus melanicterus
5
R
Yellow
Yellow
Yellow
Icterus pustulatus sclateri
5
B
Yellow
Yellow
Yellow
Cacicus sclateri
N/A
N/A
N/A
N/A
Absent
Icterus spurius fuertesi
5
Y
N/A
N/A
Absent
Cacicus solitarius
N/A
N/A
N/A
N/A
Absent
Icterus spurius spurius
5
Y
N/A
N/A
Absent
14
R
Red
Red
Red
Icterus wagleri wagleri
5
B
Yellow
Yellow
Yellow
N/A
N/A
N/A
N/A
Absent
Lampropsar tanagrinus
N/A
N/A
N/A
N/A
Absent
Amblyramphus holosericeus Cacicus cela
Cacicus uropygialis Curaeus curaeus
Patch Discretized Cluster
Composite
Patch Discretized Cluster
Composite
Curaeus forbesi
N/A
N/A
N/A
N/A
Absent
Macroagelaius imthurni
4
P
Yellow
Yellow
Yellow
Dives dives
N/A
N/A
N/A
N/A
Absent
Macroagleaius subularis
3
P
N/A
N/A
Absent
Dives warszewiczi
N/A
N/A
N/A
N/A
Absent
Molothrus aeneus
N/A
N/A
N/A
N/A
Absent
20
N
N/A
N/A
Absent
Molothrus ater
N/A
N/A
N/A
N/A
Absent
Euphagus carolinus
N/A
N/A
N/A
N/A
Absent
Molothrus badius
N/A
N/A
N/A
N/A
Absent
Euphagus cyanocephalus
N/A
N/A
N/A
N/A
Absent
Molothrus bonariensis
N/A
N/A
N/A
N/A
Absent
Gnorimopsar chopi
N/A
N/A
N/A
N/A
Absent
Molothrus oryzivora
N/A
N/A
N/A
N/A
Absent
5
B
Yellow
Yellow
Yellow
Molothrus rufoaxillaris
N/A
N/A
N/A
N/A
Absent
N/A
N/A
N/A
N/A
Absent
4
T
Yellow
Yellow
Yellow
N/A
N/A
N/A
N/A
Absent
Dolichonyx oryzivorus
Gymnomystax mexicanus Gymnostinops guatimozinus
5
T
Yellow
Yellow
Yellow
Nesopsar nigerrimus
Gymnostinops montezuma
5
T
Yellow
Yellow
Yellow
Ocyalus latirostris
Hypopyrrhus pyrohypogaster
5
Y
Red
Red
Red
Icterus abeillei
5
B
Yellow
Yellow
Yellow
Psarocolius angustifrons
6
T
Yellow
Yellow
Yellow
Icterus auratus
3
B
Orange
Yellow
Orange
Psarocolius atrovirens
1
T
Yellow
Yellow
Yellow Yellow
Oreopsar bolivianus
Icterus auricapillus
5
B
Yellow
Yellow
Yellow
Psarocolius decumanus
10
T
Yellow
Yellow
Icterus bonana
5
B
N/A
N/A
Absent
Psarocolius oseryi
4
T
Yellow
Yellow
Yellow
Icterus bullocki bullocki
5
B
Orange
Yellow
Orange
Psarocolius viridis
4
T
Yellow
Yellow
Yellow
Icterus bullocki parvus
5
B
Orange
Yellow
Orange
Psarocolius wagleri
4
T
Yellow
Yellow
Yellow
Icterus cayanensis cayanensis
5
E
Yellow
Yellow
Yellow
Pseudoleistes guirahuro
3
E
Yellow
Yellow
Yellow
Icterus cayanensis periporphyrus
2
E
N/A
N/A
Absent
Pseudoleistes virescens
5
E
Yellow
Yellow
Yellow
Icterus cayanensis pyrrhopterus
6
E
N/A
N/A
Absent
Quiscalus lugubris
N/A
N/A
N/A
N/A
Absent
Icterus chrysater chrysater
5
B
Yellow
Yellow
Yellow
Quiscalus major
N/A
N/A
N/A
N/A
Absent
Icterus chrysater giraudii
3
B
Yellow
Yellow
Yellow
Quiscalus mexicanus
N/A
N/A
N/A
N/A
Absent
Icterus chrysocephalus
5
E
Yellow
Yellow
Yellow
Quiscalus nicaraguensis
N/A
N/A
N/A
N/A
Absent
Icterus cucullatus igneus
5
B
Orange
Yellow
Orange
Quiscalus niger
N/A
N/A
N/A
N/A
Absent
Icterus cucullatus nelsoni
5
B
Yellow
Yellow
Yellow
Quiscalus palustris
N/A
N/A
N/A
N/A
Absent
Icterus dominicensis dominicensis
6
E
Yellow
Yellow
Yellow
Quiscalus quiscula
N/A
N/A
N/A
N/A
Absent
Icterus dominicensis melanopsis
5
E
Yellow
Yellow
Yellow
Sturnella bellicosa
3
B
Red
Red
Red
Icterus dominicensis northropi
2
E
Yellow
Yellow
Yellow
Sturnella defillippi
4
B
Red
Red
Red
Icterus dominicensis portoricensis
5
E
Yellow
Yellow
Yellow
Sturnella lilianae
6
B
Yellow
Yellow
Yellow
Icterus dominicensis prosthemelas
5
E
Yellow
Yellow
Yellow
Sturnella loyca
3
B
Red
Red
Red
Icterus galbula galbula
5
B
Orange
Yellow
Orange
Sturnella magna
18
B
Yellow
Yellow
Yellow
Icterus graceannae
5
B
Yellow
Yellow
Yellow
Sturnella militaris
5
B
Red
Red
Red
Icterus graduacauda audubonii
4
B
Yellow
Yellow
Yellow
Sturnella neglecta
6
B
Yellow
Yellow
Yellow
Icterus graduacauda graduacauda
5
B
Yellow
Yellow
Yellow
Sturnella superciliaris
Icterus gularis gularis
5
B
Yellow
Yellow
Yellow
Xanthocephalus xanthocephalus
Icterus gularis tamaulipensis
5
B
Orange
Yellow
Orange
Xanthopsar flavus
6
B
Red
Red
Red
17
B
Yellow
Yellow
Yellow
3
B
Yellow
Yellow
Yellow
&KDSWHU
9
1999). To score the presence of carotenoids or melanins, we followed the methods of Hofmann et al. (2006). They defined the characteristics of carotenoid plumage as having a maximum reflectance greater than 10%, with a sigmoid spectral shape and a UV peak (further described in Hofmann et al. 2007a, b). In contrast, melanins produce no UV peak, but instead produce a spectrum wherein reflectance steadily increases with wavelength but does not plateau. We acknowledge that small quantities of phaeomelanin may be present in some plumage scored here as carotenoid-based and vice versa (McGraw et al. 2004). However, our intent in this study was not to comprehensively score melanin-based plumage, but rather to effectively score carotenoid-based coloration as absent when melanin-based coloration was observed instead. We measured reflectance spectra for all apparently carotenoid-pigmented feather patches on a total of 362 museum skins, sampled across 85 icterid species and subspecies (Clements 2007; specimen voucher numbers are reported in Table S1). Carotenoid coloration is found on different body regions across icterid species; for example, meadowlarks have yellow or red breasts, whereas caciques have yellow or red rumps (Jaramillo and Burke 1999). Consequently, we compared the color of whichever feather patch was most consistently carotenoid-colored for that clade (see Table 1). We examined spectral location (λR50) for each reflectance spectrum (Pryke et al. 2001; Andersson and Prager 2006; Hofmann et al. 2006; Montgomerie 2006). In addition, we calculated a measure of “yellow chroma” (representing color "purity"; see Montgomerie 2006). This measure was specific to an arbitrarily defined portion of the typical carotenoid reflectance curve that described percent reflectance in the green to yellow portion of the spectrum (500-600nm). “Red” chroma has been measured this way in
&KDSWHU
10
previous studies (Norris et al. 2004; Shawkey and Hill 2005), and our yellow chroma measurement is a simple extension of its concept. We used three methods to delineate discrete character states from continuous measures of color. First, we used areas of non-overlap among intervals of standard error for character values following the methods of Price and Lanyon (2002a). This simple technique has been used to delineate discrete character states from continuous measures of both song (Price et al. 2007) and plumage (Kiere et al. 2009). As all non-oriole blackbirds could only be scored as either red or yellow in color (see Fig. 1 below; also see Kiere et al. 2009), we used non-oriole taxa to describe three character states. We defined "red" and "yellow" states by the range of variation included within standard error intervals of non-oriole blackbirds, and defined "orange" as the intermediate range, which is only occupied by several orioles (this method is hereafter referred to as “discretized” scoring). Second, we scored a composite character that combines the discrete scoring with a fourth state to indicate when a taxon was found to use only melanin-based coloration (hereafter “composite” scoring). Third, we used a bivariate cluster analysis in R (cluster; Fig. 2) to assign taxa to different color groups (hereafter “cluster” scoring) (Maechler 2005). We estimated a dissimilarity matrix among taxa with respect to their spectral location and yellow chroma. We then used the "Partitioning Around Medioids" operation in cluster to describe fit into one, two, three, four, and five clusters. This procedure is analogous to the k-means method of cluster analysis (Maechler 2005), which has been used in previous studies of color (Friedman et al. 2009; Prager and Andersson 2010). Ancestral State Reconstruction
&KDSWHU
11
(a)
(b)
Figure 1: Distribution of mean spectral location values for taxa across the New World blackbirds excluding the orioles (a) and across the orioles (b). Note the absence of any icterid taxa with spectral locations from 560 to 570 nm. This indicates that there are no icterids that display color patches with intermediate values between orange and red.
&KDSWHU
12
We reconstructed discrete ancestral states for the three methods described above across the icterid molecular phylogeny (Lanyon and Omland 1999; Lanyon and Barker 2007). This tree was inferred from nucleotide sequences from two mitochondrial loci and four nuclear loci, and has been used in previous comparative studies of the blackbird family (Price et al. 2009). For each reconstruction, we used both simple parsimony and maximum likelihood in Mesquite (Maddison and Maddison 2010). We pruned all taxa lacking carotenoid-based ornamentation (38) from trees used to reconstruct the discretized and cluster characters, but included these taxa for the reconstruction of the composite character. As recommended by Wiens et al. (2007), we explored the use of both one-parameter (gains and losses equal) and two-parameter (gains or losses more likely) models of character change with estimated rates. However, due to computational constraints, two-parameter rate models could only be applied to the binary character (i.e., the cluster character). Results Color Variation We observed a bimodal distribution of taxon mean spectral location values across three of the icterid clades (Fig. 1a; n=37). However, this distribution was unimodal in the fourth clade, the New World orioles (Fig. 1b; n=39). Furthermore, only orioles exhibited plumage with spectral locations between 540nm and 560nm, a range that roughly describes hues that humans perceive as orange. Yellow chroma values were highly correlated with spectral location values (r2 = 0.95), ranged from 14% to 38%, and produced a bimodal distribution similar to that of spectral location.
&KDSWHU
13
By examining standard error intervals we observed that, outside of Icterus, the “yellow” icterids varied in their spectral location from 509nm to 537nm. We used this result to define our “yellow” character state by the range of 500-540nm. We also observed that the “red” icterids varied in spectral location from 580nm to 601nm. We used this result to define our “red” character state by the range of 580-605nm. We described birds exhibiting plumage with spectral location values in the intermediate range (540-580nm) as having “orange” plumage. Following the standard error interval methods described above, we scored 52 icterid taxa as “yellow”, 11 as “orange”, and 12 as “red” (38 others were scored as melanin; see Table 1). Cluster analysis of spectral location and yellow chroma revealed the presence of two highly significant groups (“red”: si = 0.86; “yellow”: si = 0.77; Fig. 2) that explained 100% of the observed color variation. All taxa were assigned to a group with a strong degree of confidence with the exception of several taxa described by the discretizing method (Hofmann et al. 2006) as orange in color (e.g., I. cucullatus igneus). Furthermore, cluster analyses with three or more clusters yielded non-significant assignments (data not shown). All the taxa scored as red using this method (Hofmann et al. 2006) were also scored as "red" using the above-described discretizing method (see Table 1). Lastly, we scored 38 taxa as having no carotenoid-based coloration. Of these, the 10 taxa that displayed chestnut or blonde colors all showed reflectance spectra with features suggesting that phaeomelanin was responsible for their hue. Thus, carotenoid coloration was scored as “absent” in these taxa.
&KDSWHU
14
Figure 2: Cluster analysis of two measures of carotenoid coloration: spectral location and yellow chroma. These two measurements are significantly correlated for carotenoid-colored patches (without phylogenetic correction; p < 0.0001, R2 = 0.94). Two groups were strongly supported; these are colored gray to indicate the “red” character state, and white to indicate the “yellow” character state.
&KDSWHU
15
Ancestral States Ancestral state reconstruction of the discretized character using simple parsimony showed five or six independent gains of red coloration from a yellow common ancestor (Table 2; Fig. S1). Analysis of the composite character using the same method showed a similar pattern of six independent gains of red coloration, but showed both yellow and red coloration to be repeatedly derived in the grackles and allies (Table 2). In our parsimony reconstructions, we inferred seven gains of orange coloration in orioles, but none in any other icterid clade. These all appeared to be recent gains, and orange coloration was only once a shared ancestral characteristic among any two oriole species. Ancestral state reconstruction of color using the cluster character also indicated a yellow ancestral icterid and five or six repeated gains of red carotenoid-based plumage in Icteridae (Table 2; Fig. S2). Likelihood estimates of ancestral states under a one-parameter model indicated six gains of red coloration from a common ancestor with yellow ornamentation for all scoring methods (Table 2; Fig. 3; Fig. S3). To support a scenario in which red coloration was ancestral, it was necessary to assume a model in which losses of red were at least 50 times more likely than gains. Such a model was not supported by likelihood estimates of rate parameters (p < 0.0001). Rather, the best fitting two-parameter rate model (losses only slightly more likely than gains; q12 = 3.90; q21 = 4.61) also showed strong support for a scenario involving six gains of red coloration, and also supported a yellow common ancestor (Table 2; Fig. S4). Discussion
&KDSWHU
Figure 3: Ancestral state reconstruction of our composite scoring of carotenoid coloration across Icteridae. This was calculated in Mesquite (Maddison and Maddison 2010) using the likelihood method under a one-parameter rate model. Pie diagrams indicate proportional likelihood scores, which describe degrees of support for each character state.
16
&KDSWHU
17
Repeated Gains of Red Coloration Our ancestral state reconstructions consistently indicated that there have been five or six independent gains of red carotenoid-based coloration in the New World blackbirds. The reconstructions show these gains occurring in the ancestor of the South American meadowlarks (genus Sturnella; five species), in the ancestor of the red-winged blackbirds (genus Agelaius; three species), and in four single species lineages (Hypopyrrhus pyrohypogaster, Amblyramphus holosericeus, Cacicus haemorrhous, and Cacicus uropygialis). Convergent gains of red coloration from a yellow ancestor were reported earlier in the caciques by Kiere et al. (2009). Our study finds a similar pattern in two other blackbird clades: the meadowlarks and allies, and in the grackles and allies. Our results strongly indicate that red carotenoid-based coloration evolved in Icteridae by repeated gains of red coloration from a yellow common ancestor. While we were able to skew the likelihood rate model to force a result of repeated losses of red, this skewed model was a significantly worse fit to the data when compared to the best-fitting model. The history of carotenoid-based coloration that we have inferred suggests the presence of a directional trend towards the repeated evolution of red plumage in this clade. We observed six changes in the direction of yellow to red, and none in the reverse direction (Table 2; but see Figs. S1 and S2). Evolutionary history is replete with putatively directional trends, which may be indicative of active or passive processes driving the repeated evolution of characters (Carroll 2001). Hill has argued that such a pattern should arise from an active process, particularly the selection across lineages for female preference for redder males (Hill 1996; Hill and McGraw 2004). This argument is based on two sometimes controversial assertions: 1) that selection should favor females
&KDSWHU
18
Table 2: Results of ancestral state reconstructions in terms of gains and losses, listed across scoring methods. For parsimony methods, we recorded both maximum and minimum numbers of gains and losses when equally parsimonious reconstructions were returned. For likelihood methods we used a likelihood decision threshold of 2.0. See text for likelihood model details. Reconstruction method
Scoring
Yellow Gains Losses
Red Gains Losses
Unordered parsimony
Discretized Composite Cluster Discretized Composite Cluster Cluster
0 4-6 0 0 6 0 0
5-6 6 5-6 6 6 6 6
Likelihood Mk1
Likelihood asymmetrical
13 - 14 17 - 20 5-6 14 19 6 6
0-1 0 0-1 0 0 0 0
Orange Gains Losses 7 7 N/A 7 7 N/A N/A
0 0-1 N/A 0 0 N/A N/A
Melanin Gains Losses N/A 9 - 11 N/A N/A 9 N/A N/A
N/A 6 - 10 N/A N/A 9 N/A N/A
&KDSWHU
19
that respond to more honest sexual signals (Krebs and Dawkins 1984); and 2) that red carotenoid-based coloration is more costly and thereby a more honest signal of male condition (Hill 1996). However, it is also important to consider the role of passive processes in producing apparent directional trends. Stochastic processes that follow bounded models of Brownian motion (Raup and Gould 1974; Gould 1988) also explain the repeated appearance of similar traits as lineages diversify, such as the trend towards larger body size in mammals (Cope’s Rule; Stanley 1973). Simulations and broader comparative studies are needed that examine the trend we observe in blackbirds to explore whether it may be driven by active or passive processes, and to determine whether similar trends are widespread across the passerine birds. Ancestral Yellow Coloration Our results suggest that the common ancestor of the New World Blackbirds most likely exhibited yellow plumage. However, it is worth mentioning that this conclusion depends on our assumption that the yellow plumage displayed on feather patches on different regions of the body was homologous. If this assumption is incorrect, our scoring of carotenoid coloration may have created a composite character made up of several independently evolving characters (McLennan and Brooks 1993). Scoring carotenoid coloration for each of these body regions independently leads to a reconstruction that they are each absent in the common ancestor. However, there are at present no data to support the assumption that carotenoid color patches on different body regions evolve independently. In contrast, recent studies in widowbirds and bishops that
&KDSWHU
20
included chemical analysis of pigments have shown yellow coloration to be homologous in all their taxa that express it (Andersson et al. 2007; Prager and Andersson 2009). In this study we have assumed that yellow coloration is homologous across feather patches. To assess whether this assumption is valid, our future studies of carotenoid coloration in the Icteridae will focus on the identity of the carotenoid pigments responsible for blackbird color variation. We also observed seven repeated gains of orange coloration in the New World orioles from a yellow common ancestor. Hofmann et al. (2006) addressed the evolution of orange coloration extensively in their study of the oriole clade, but they inferred a different history for carotenoid-based coloration. Using continuous ancestral state reconstruction methods, they inferred an ancestral oriole with intermediate yellow-orange coloration and subsequent changes towards both yellow and orange coloration. Our inference of a yellow common ancestor is influenced greatly by the predominance of yellow coloration among the other icterids. Furthermore, we noted a conspicuous lack of orange feather coloration in our extensive sampling of blackbirds outside Icterus (Fig. 1a). Thus, we conclude that orange feather coloration in Icteridae is a novelty that was gained either in the orioles’ common ancestor, or repeatedly within the oriole clade. Conclusions In conclusion, our results suggest that red coloration has been gained repeatedly in the New World Blackbirds from a common ancestor that exhibited yellow ornamentation. Thus, we conclude that the elaborate red ornaments displayed by many blackbirds are convergent in origin. Furthermore, we have described an apparent
&KDSWHU
21
directional trend towards red coloration in this clade that is similar to those described in previous studies of red and yellow coloration (Hill and McGraw 2004; Prager and Andersson 2010). These repeated inferences of directional trends towards red coloration suggest that this pattern could be common among passerine birds and may reflect either a general selection pressure or a stochastic directional process (Stanley 1973; Hill 1996). We will be directly exploring the ultimate causes of repeated gains of red coloration in blackbirds, and we encourage other studies that investigate these questions using behavioral (e.g., Hansen and Rohwer 1986), phylogenetic (e.g., Hofmann et al. 2006) and mechanistic approaches (e.g., Andersson et al. 2007). Our next step will be to reconstruct the evolution of the proximate mechanisms underlying coloration in the New World blackbirds. Acknowledgements J. J. Price, C. M. Hofmann and F. Jacobsen provided helpful advice on this manuscript. We would like to thank the USNM, ANSP, DMNH and FMNH for allowing the Omland lab to measure specimens in their collections. KEO was supported by a National Science Foundation CAREER grant DEB-0347083. The Omland Lab is a member of the Smithsonian Ornithology Group. NRF would like to thank P. and S. Everdell for prompting his interests in biology and nature. Literature Cited Allen E. S., Omland K. E. 2003. Novel Intron phylogeny supports plumage convergence in Orioles (Icterus). Auk 120:961-969.
&KDSWHU
22
Andersson, S., and M. Prager 2006. Quantifying Colors. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 41-89. Andersson, S., M. Prager, E. I. and Anette Johansson. 2007. Carotenoid content and reflectance of yellow and red nuptial plumages in widowbirds (Euplectes spp.). Functional Ecology 21:272-281. Barker, F. K., A. J. Vandergon, and S. M. Lanyon. 2008. Species status of the Redshouldered Blackbird (Agelaius assimilis): Implications for ecological, morphological, and behavioral evolution in Agelaius. Auk 125:87-94. Brush, A. H. 1990. Metabolism of carotenoid pigments in birds. FASEB J. 4:2969-2977. Brush, A. H., and D. M. Power. 1976. House finch pigmentation: carotenoid metabolism and the effect of diet. Auk 93:725-739. Carroll, S. B. 2001. Chance and necessity: the evolution of morphological complexity and diversity. Nature 409:1102-1109. Clements, J. F. 2007. The Clements checklist of birds of the world, 6th edition. Cornell University Press, Ithaca, NY. Dale, J. 2006. Intraspecific variation in coloration. In: Bird Coloration. II. Function and Evolution (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 36-86. Fox, D. L. 1962. Metabolic fractionation, storage and display of carotenoid pigments by flamingoes. Comparative Biochemistry and Physiology 6:1-40. Friedman, N. R., C. M. Hofmann, B. Kondo, and K. E. Omland. 2009. Correlated evolution of migration and sexual dichromatism in the New World Orioles (Icterus). Evolution 63:3269-3274. Gould, S. J. 1988. Trends as changes in variance: a new slant on progress and directionality in evolution. Journal of Paleontology 62:319-329. Hill, G. E. 1990. Female house finches prefer colourful males: sexual selection for a condition-dependent trait. Animal Behavior 40:563-572. Hill, G. E. 1996. Redness as a measure of the production cost of ornamental coloration. Ethology, Ecology, and Evolution 8:157-175.
&KDSWHU
23
Hill G. E., McGraw K. J. 2004. Correlated changes in male plumage coloration and female mate choice in cardueline finches. Animal Behavior 67:27-35. Hofmann, C. M., T. W. Cronin, and K. E. Omland. 2006. Using spectral data to reconstruct evolutionary changes in coloration: carotenoid color evolution in new world orioles. Evolution 60:1680-1691. Hofmann C. M., McGraw K. J., Cronin T. W., Omland K. E. 2007a. Melanin Coloration in New World Orioles I: Carotenoid Masking and Pigment Dichromatism in the Orchard Oriole Complex. Journal of Avian Biology 38:163-171. Hofmann, C. M., T. W. Cronin, and K. E. Omland. 2007b. Melanin Coloration in New World Orioles II: Ancestral State Reconstruction Reveals Lability in the Use of Carotenoid and Phaeomelanins. Journal of Avian Biololgy 38:172-181. Hudon, J. 1991. Unusual carotenoid use by the Western Tanager (Piranga ludoviciana) and its evolutionary implications. Can. J. Zool 69:2311-2320. Inouye, C. Y., G. E. Hill, R. D. Stradi, and R. Montgomerie. 2001. Carotenoid pigments in male house finch plumage in relation to age, subspecies, and ornamental coloration. Auk 118:900-915. Jacobsen F., Friedman N. R., Omland K. E. 2010. Congruence between nuclear and mitochondrial DNA: Combination of multiple nuclear introns resolves a wellsupported phylogeny of New World orioles (Icterus). Molecular Phylogenetics and Evolution 56:419-427. Jaramillo, A., and P. Burke. 1999. New World Blackbirds: The Icterids. Princeton University Press, Princeton, N. J. Johnson K. P., Lanyon S. M. 1999. Molecular systematics of the grackles and allies, and the effect of additional sequence (cyt B and ND2). Auk 116:759-768. Johnson, K. P., and S. M. Lanyon. 2000. Evolutionary changes in color patches of blackbirds are associated with marsh nesting. Behavioral Ecology 11:515-519. Kiere, L. M., C. M. Hofmann, J. J. Price, T. W. Cronin, and K. E. Omland. 2009. Discrete evolutionary color changes in caciques suggest different modes of carotenoid evolution between closely related taxa. Journal of Avian Biology 40:605-613.
&KDSWHU
24
Kondo, B., and K. E. Omland. 2007. Ancestral state reconstruction of migration: multistate analysis reveals rapid changes in New World orioles (Icterus spp.). Auk 124:410-419. Krebs, J. R. and R. Dawkins. 1984. Animal signals: mind-reading and manipulation. In Behavioural Ecology: an evolutionary approach, 2nd ed (Krebs, J. R. and N. B. Davies, eds), Sinauer: pp 380–402. Lanyon, S. M., and Barker F. K. 2007. Exploring patterns of morphological evolution in the New World blackbirds. In 125th Meeting of the American Ornithologists’ Union. Laramie, WY. Maddison, W. P. and D. R. Maddison. 2010. Mesquite: a modular system for evolutionary analysis. Version 2.72 http://mesquiteproject.org. Maechler, M. 2005. Cluster: Cluster analysis basics and extensions: cluster R package version 1.12.1. McGraw, K. J. 2006a. Mechanics of carotenoid-based coloration. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 177-242. McGraw, K. J. 2006b. Mechanics of melanin-based coloration. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 243-294. McGraw, K. J., and K. Wakamatsu, A. B. Clark, K. Yasukawa. 2004. Red-winged blackbirds Agelaius pheoniceus use carotenoid and melanin pigments to color their epaulets. Journal of Avian Biology 35:543-550. McLennan, D. A., and D. R. Brooks. 1993. The Phylogenetic Component of Cooperative Breeding in Perching Birds: A Commentary. The American Naturalist 141:790795. Montgomerie, R. 2006. Analyzing Colors. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 91-147. Murphy, T. G., D. Hernandez-Mucino, M. Osorio-Beristain, R. Montgomerie, and K. E. Omland. 2009. Carotenoid-based signaling by females in the tropical streakbacked oriole. Behavioral Ecology 20:1000-1006.
&KDSWHU
25
Norris, D. R., P. P. Marra, R. Montgomerie, T. K. Kyser, and L. M. Ratcliffe. 2004. Reproductive effort, molting latitude, and feather color in a migratory songbird. Science 306:2249-2250. Omland, K. E., and S. M. Lanyon. 2000. Recontructing Plumage Evolution in Orioles (Icterus): Repeated Convergence and Reversal in Patterns. Evolution 54:21192133. Omland, K. E., S. M. Lanyon, and S. J. Fritz. 1999. A Molecular Phylogeny of the New World Orioles (Icterus): The Importance of Dense Taxon Sampling. Molecular Phylogenetics and Evolution 12:224-239. Pagel, M. 1999. Inferring the Historical Patterns of Biological Evolution. Nature 401:877-885. Prager, M., and S. Andersson. 2009. Differential ability of carotenoid C4-oxygenation in yellow and red bishop species (Euplectes spp.). Comparative Biochemistry and Physiology B Molecular Biolology 154:373-380. Prager, M., and S. Andersson. 2010. Convergent evolution of red carotenoid coloration in Widowbirds and Bishops (Euplectes spp.). Evolution 64:3609-3619. Price, J. J., N. R. Friedman, and K.E. Omland. 2007. Song and Plumage Evolution in the New World Orioles (Icterus) Show Similar Lability and Convergence in Patterns. Evolution 61:850-863. Price, J. J., and S. M. Lanyon. 2002a. Reconsructing the evolution of complex bird song in the Oropendolas. Evolution 56:1514-1529. Price, J. J., and S. M. Lanyon. 2002b. A robust phylogeny of the oropendolas: polyphyly revealed by mitochondrial sequence data. Auk 119:335-348. Price, J. J., S. M. Lanyon and K. E. Omland. 2009. Losses of female song with changes from tropical to temperate breeding in the New World blackbirds. Proc. Biol. Sci 276:1971-1980. Price, J. J., and L. M. Whalen. 2009. Plumage evolution in the oropendolas and caciques: different divergence rates in polygynous and monogamous taxa. Evolution 63:2985-2998.
&KDSWHU
26
Pryke, S. R., M. J. Lawes, S. Andersson. 2001. Agonistic carotenoid signalling in male red-collared widowbirds: aggression related to the colour signal of both the territory owner and model intruder. Animal Behavior 62:695-704. Pyle, P. 1997. Identification Guide to North American Birds. Slate Creek Press, Bolinas, CA, USA. Raup, D. M., and S. J. Gould. 1974. Stochastic simulation and evolution of morphology towards a nomothetic paleontology. Systematic Biology 23:305-322. Searcy, W. A., K. Yasukawa, and S. M. Lanyon. 1999. Evolution of polygyny in the ancestors of red-winged blackbirds. Auk 116:5-19. Shawkey, M. D., and G. E. Hill. 2005. Carotenoids need structural colours to shine. Biology Letters 1:121-124. Stanley, S. M. 1973. An explanation for Cope's Rule. Evolution 27:1-26. Wiens J. J., Kuczynski C. A., Duellman W. E., Reeder T. W. 2007. Loss and re-evolution of complex life cycles in marsupial frogs: does ancestral trait reconstruction mislead? Evolution 61, 1886-1899. Yasukawa, K., L. K. Butler, and D. A. Enstrom. 2009. Intersexual and intrasexual consequences of epaulet colour in male red-winged blackbirds: an experimental approach. Animal Behavior 77:531-540.
&KDSWHU
27
CHAPTER 2: CONVERGENCE AND PARALLELISM IN THE EVOLUTION OF RED CAROTENOID PIGMENTATION IN CACIQUES AND MEADOWLARKS (Submitted to Evolution Fall 2012, Resubmission Planned Winter 2013)
NICHOLAS R. FRIEDMAN1, KEVIN J. MCGRAW2, KEVIN E. OMLAND1
1
Department of Biological Sciences, University of Maryland, Baltimore County, Baltimore, MD 21250 2
School of Life Sciences, Arizona State University, Tempe, AZ 85287
Email:
[email protected],
[email protected],
[email protected]
&KDSWHU
28
Abstract Species throughout the animal kingdom have evolved the use of carotenoid pigmentation to produce yellow, orange and red coloration. In birds, there are at least ten carotenoid compounds known to produce red plumage coloration; most of these are produced through metabolic modification of dietary precursor compounds. In this study, our goal was to determine whether different lineages of the New World blackbirds evolved red plumage coloration using similar compounds (parallelism) or different compounds (convergence). To study the mechanisms underlying these color changes, we examined carotenoid pigmentation across two blackbird clades: the meadowlarks and allies, and the caciques and oropendolas. We used high-performance liquid chromatography (HPLC) to identify the carotenoid compounds present in feathers from 14 blackbird taxa, and we mapped their presence or absence on a phylogeny. Our results show that the red plumage coloration found in meadowlarks includes a different set of carotenoids than the red coloration found in caciques, indicating that these gains of red are convergent. We also found that two closely related lineages of caciques gained red plumage coloration independently using the same set of carotenoids, providing an example of parallel evolution
offer
a glimpse of
variation among species evolves.
Key Words. – Homoplasy, ancestral state reconstruction, bird coloration, HPLC, Icteridae
&KDSWHU
29
Introduction Understanding the origins of morphological variation among species is one of the most central aims of evolutionary biology. Many species show similar morphological traits that are independently derived. Such cases of homoplasy can mislead taxonomy and identification (e.g., Omland and Lanyon 2000; Price et al. 2007), but are now easily identified using molecular markers. Identifying the presence of homoplasy invites a question: how did similar traits evolve in independent lineages? Most research has focused on the ultimate causes of homoplasy, describing the similar selection pressures that drive the evolution of traits towards a similar phenotype. However, less research has focused on the proximate causes of homoplasy, and in many cases the mechanisms by which independently derived traits are produced remain unknown. Molecular and phylogenetic tools are now available that permit identification of homoplastic traits and investigation of their underlying genetic, physiological, and biochemical mechanisms. For example, nearly a century and a half after Bates (1862) first proposed that natural selection might have lead to the repeated evolution of similar wing patterns in butterflies, researchers can now work towards understanding the molecular basis for these similar wing patterns (Reed and Serfas 2004; Reed et al. 2011). Researchers have long worked on the association between red flower coloration and avian pollination (Cronk and Ojeda 2008 and refs. therein), but have only recently identified the molecular basis of color variation among flowering plant species (Des Marais and Rausher 2010; Smith and Rausher 2011). Such research helps to compare the rates at which selection for a particular phenotype should lead to fixation between different molecular and biochemical mechanisms that produce that phenotype (Des
&KDSWHU
30
Marais and Rausher 2010). If a trait evolves repeatedly by different mechanisms, the available pathways to produce that trait might be diverse, and/or the phylogenetic distance great. Likewise if a trait evolves repeatedly using similar mechanisms, the available pathways to produce that trait might be few or constrained, and/or the phylogenetic distance small (Arthur 2001). Examining the proximate mechanisms underlying homoplasy provides a framework to test among different types of homoplasy that were previously difficult or impossible to distinguish: convergence, parallelism, and reversal. The definitions and usage of convergence and parallelism may be contentious, with some biologists using them synonymously, or nearly so (Arendt and Reznick 2007). To test between mutually exclusive hypotheses in this study, we have chosen to use the following definitions of convergence and parallelism. We use convergence (sensu strictu) to refer to the independent evolution of similar phenotypes by different mechanisms. We use parallelism (sensu Gould 2002; Hall 2003) to refer to the independent evolution of similar phenotypes by similar mechanisms (see also Futuyma 2005). This framework has been used in similar recent studies of the evolution of proximate mechanisms (Des Marais and Rausher 2010; Smith and Rausher 2011). Many birds have evolved the ability to deposit carotenoid pigments in their feathers to produce red, orange, or yellow colors. Carotenoid-based coloration has been the focus of a great deal of research over the last century (see Hudon 1994 and refs therein; McGraw 2006). Researchers have extensively examined both the pigments underlying this coloration and their relation to avian physiology and behavior (Inouye et al. 2001, McGraw et al. 2001; Dale 2006). This research has identified a strong role for
&KDSWHU
31
carotenoid-based coloration in sexual selection and visual signaling in many birds. In particular, females of many species may prefer to mate with males having more elaborate carotenoid-based traits (e.g., plumage, bare parts; Hill 2006). The carotenoids deposited in feathers cannot be synthesized endogenously by vertebrates, but are instead derived from those occurring in the diet (McGraw 2006 and refs. therein). Consequently, carotenoid-derived ornamental coloration may be dependent on diet and condition at the time of molt (Hill 1992; Hill and Montgomerie 1994). Given this condition dependence, several researchers have suggested that displays of carotenoid colors may be costly signals that provide information concerning individual health or genetic quality to potential mates or rivals (Kodric-Brown 1989; Hill 1991; but see Olson and Owens 1998). Whereas many studies have focused on the biochemical and ecological mechanisms that underlay intraspecific variation in carotenoid-based coloration, few have examined variation in carotenoid pigmentation across species in a phylogenetic context (see Dale 2006; McGraw 2006). The carotenoid compounds available in the diets of most wild birds (primarily lutein and zeaxanthin) produce yellow coloration when deposited unmodified into feathers. However, some species biochemically alter these dietary yellow carotenoids to produce modified red carotenoids such as canthaxanthin and astaxanthin (Brush 1967; Hudon 1991). There are many different carotenoid pigments found in birds feathers that can produce red coloration (see McGraw 2006), and these can differ even between closely related species or subspecies (Inouye et al. 2001). Furthermore, previous studies in widowbirds and bishops (Euplectes) have reported convergent gains of red plumage that are also convergent in pigmentation mechanism (Andersson et al. 2007;
&KDSWHU
32
Prager and Andersson 2009). Our goal in this study is to investigate the degree to which convergence versus parallelism explains repeated gains of red coloration across a closely related family of songbirds. To address this, we use ancestral state reconstruction to examine the history of gains and losses of carotenoid pigments underlying feather coloration. The New World blackbirds (Family Icteridae) are a speciose and behaviorally diverse clade that has served as a model clade for studying evolution. Consequently, much is known about the phylogenetic relationships, behavior, coloration and mating systems of icterids (e.g., Westneat et al. 1995; Lanyon and Omland 1999; Johnson and Lanyon 2000; Price and Lanyon 2002). Two icterid groups in particular include many taxa exhibiting red or yellow carotenoid-based plumage: the meadowlarks (Sturnella) and the caciques (Cacicus). Recent studies of these genera have inferred repeated gains of red coloration from an ancestor with yellow coloration (Kiere et al. 2009; Friedman et al. 2011). Such repeated gains of red coloration could be explained either by convergence (e.g., different enzymes producing different carotenoid compounds), or parallelism (e.g., the same enzymes and carotenoids). Reflectance spectra are statistically indistinguishable among distantly related red cacique taxa, suggesting that similar mechanisms may be responsible for independent origins of red coloration in this genus (Kiere et al. 2009). We use high-performance liquid chromatography (HPLC) to identify the types and amounts of carotenoid pigmentation in caciques and meadowlarks to distinguish between a scenario of convergent evolution versus one of parallel evolution. If we find that multiple distinct lineages exhibiting red plumage use the same red carotenoid(s) to produce red plumage, this would support the conclusion that repeated gains of red have
&KDSWHU
33
evolved by parallelism at the biochemical level. However, if we find that distinct lineages with red plumage use different carotenoids to color their feathers, this would support a conclusion of convergent evolution, demonstrating gains of red coloration by different biochemical mechanisms.
Methods Sampling and Analysis We sampled feathers from 49 vouchered museum skins representing 14 icterid species at the Academy of Natural Sciences in Philadelphia and the Delaware Museum of Natural History. We examined 3-5 individuals per taxon, representing all red taxa in both clades, as well as two species with black eumelanin-based coloration (C. solitarius and D. oryzivorous; Friedman et al. 2011). As several species in the meadowlarks and allies are sexually dichromatic with elaborate coloration only present in males (Jaramillo and Burke 1999), we compared male specimens across our sampled taxa. Furthermore, we used only specimens collected during the breeding season, and whose plumage appeared to be well preserved. Prior to sampling, we used reflectance spectrometry to measure the color of the patch to which the feathers belonged, using methods and equipment described in Friedman et al. (2011). From each specimen, we trimmed and measured approximately 2-5 mg of red or yellow-colored feather barbs for chemical analysis. Following procedures detailed by McGraw et al. (2005), we extracted carotenoids using acidified pyridine at 95ºC for 1-3hr. We then added 3 ml water and separated carotenoids by extraction into 2ml of hexane and tert-butyl methyl ether (1:1, v/v), and centrifugation at 3000rpm for 5 minutes. To prepare samples for HPLC analysis, we
&KDSWHU
34
evaporated the solvent to dryness under nitrogen and resuspended the pigment residue in the mobile phase (methanol : acetonitrile : dichloromethane, 42:42:16, v/v/v). This method separates carotenoids from any extracted polar non-carotenoid colorants (e.g., phaeomelanins, which we did not observe in solution). Our HPLC analyses of feather pigmentation followed those described by McGraw et al. (2006). We used a Waters 2695 instrument equipped with a Waters 2996 photodiode array and a Waters YMC Carotenoid column (5μm, 4.6mm x 250mm; Waters Corp., Milford, MA) that was heated to 30º C in a column heater. We used a three-step gradient solvent system that is capable of detecting both xanthophylls and carotenes; we pre-treated the column with 1% ortho-phosphoric acid to ensure elution of ketocarotenoids (Toomey and McGraw 2007). We collected absorbance data from 250 nm to 600 nm and identified pigments by comparison to retention times and absorbance spectra of authentic reference carotenoids and previous descriptions in the literature (Inouye et al. 2001; Britton et al. 2004; Andersson et al. 2007; Rowe and McGraw 2009). Pigment concentrations were estimated using the area under the HPLC elution curve at λmax and in comparison with external standards (following Rowe and McGraw 2009). Character Scoring Samples were labeled only by their museum catalog number during chemical analysis and scoring to ensure that the researcher performing these tasks was blind with respect to each sample’s taxonomic identity. We scored pigments as present for each taxon in which they were identified in at least half of the specimens we examined. We considered any identifiable compound with an chromatogram area less than 2 x 104 AU • s-1 to be present in an unquantifiable amount (indicated by a “+” in Table 2).
&KDSWHU
35
To describe the relationship between coloration and pigmentation, we used a measure of carotenoid hue (spectral location, or λR50; see Montgomerie 2006) from our recent study of blackbird coloration (Friedman et al. 2011). In that study, we used reflectance spectrometry to objectively measure carotenoid-colored feather patches on museum skins. Consequently, when we describe coloration in this paper as “yellow” or “red”, we are referring to the discrete character states quantified in our previous study. These character states define “yellow” as having a spectral location within 500-540nm, and “red” as having a spectral location within 580-605nm. While spectral location is an accepted measure of carotenoid hue, it does not include information about UV reflectance. To score discrete character states for color using information from the entire reflectance spectrum, we also used Stoddard and Prum’s (2008) tetracolorspace script to estimate photoreceptor stimulation values from each specimen’s reflectance spectrum. We used the “Partitioning Around Medioids” function in Maeckler et al.’s (2012) cluster package to delineate discrete character states from these photoreceptor stimulation values. However, we chose not to reconstruct ancestral states for receptor stimulation estimates in this study because doing so would create a compound character that could mislead ancestral state reconstruction (see McLennan and Brooks 1993).
Ancestral State Reconstruction To reconstruct ancestral states, we used the molecular phylogeny described by Price et al. (2009), which was inferred using RAxML analysis of two mitochondrial loci and four nuclear loci. This topology has been consistent across phylogenetic studies
&KDSWHU
36
within each clade (Price and Lanyon 2002; Barker et al. 2008), as well as recent a recent supertree (Hugall and Stuart-Fox 2012). As likelihood-based phylogenetic comparative methods cannot use trees that include taxa with missing data, we pruned this tree to include only taxa belonging to the caciques and oropendolas and the meadowlarks and allies (thus excluding the New World orioles (Icterus) and the grackles and allies). Consequently, this tree can appear to support a sister relationship between these two clades within Icteridae, where in fact current molecular evidence does not support such a relationship (see Price et al. 2009). Our sampling in this study encompassed 14 of the 30 recognized species within these two clades, and represents all of the carotenoid color variation observed by previous studies (Kiere et al. 2009; Friedman et al. 2011). Ancestral state reconstruction is a tremendously useful but parameter-intense method of analysis, and can thus be misleading without careful consideration of methods and their assumptions (Omland 1999). A recent study by Wiens et al. (2007) has demonstrated the need for researchers to use multiple methods and rate models before drawing conclusions from ancestral state reconstruction. In this study, we used both parsimony and likelihood-based methods as implemented in Mesquite (Maddison and Maddison 2011). We used rate models inferred in our previous study of carotenoid coloration that included a greater sample size (Friedman et al. 2011; hereafter “fixed rate model”), as well as models estimated from our current dataset (hereafter “estimated rate model”). Furthermore, we explored the sensitivities of our ancestral state reconstruction results to different rate model parameters within the likelihood framework. Specifically, this approach allowed us to compare both the support for and results of models assuming
&KDSWHU
37
different rates of character change, and different degrees of bias between gains and losses. Phylogenetic Comparative Methods To examine how pigment concentration and presence/absence contribute to color variation across species while correcting for phylogeny, we used the phylogenetic comparative methods detailed below. We scored the following characters: total plumage carotenoid concentration, total C4-ketocarotenoid concentration, discrete color category, and spectral location. We fit Brownian Motion (BM) and Ornstein-Uhlenbeck (OU) models of character evolution to our data using ouch (Butler and King 2004), and found that for each continuous character the BM model was the best fit. Consequently, we used the Phylogenetic Generalized Least Squares method (PGLS; Grafen 1989) under a BM model to test for correlations among continuous characters in geiger (Harmon et al. 2008; see online supplement). To test for a relationship between gains of keto-carotenoids and red coloration, we used Pagel’s (1994) discrete method as implemented in Mesquite (Maddison and Maddison 2012).
Results We identified the presence of nine distinct carotenoids in the yellow or red plumage of our study species: lutein, zeaxanthin, canthaxanthin, astaxanthin, doradexanthin, canary xanthophylls A and B, and 3-hydroxy-echinenone (Table 1). In addition, we detected five lutein isomers whenever lutein was present. These isomers are most likely artifacts produced during the pyridine extraction process (see McGraw et al. 2003). Mays et al. (2004) confirmed that such isomers are produced when the extraction
&KDSWHU
38
Table 1: Pigments identified from feather extracts and their retention times (RT), as well as their absorbance peaks (λmax). Pigment RT λmax Canary xanthophyll B 4.9 419, 443, 473 Canary xanthophyll A 5.2 418, 445, 473 Adonirubin 5.8 472 Alpha-doradexanthin 6.2 472 Isolutein (a)* 6 445, 472 Lutein 6.6 448, 476 Astaxanthin 7.5 480 Isolutein (b)* 8 448, 470 Zeaxanthin 7.8 452, 478 Isolutein (c)* 8.9 444, 471 Canthaxanthin 9.2 477 Isolutein (d)* 10.1 448, 475 3-hydroxy-echinenone 11.1 473 * = Isomers of lutein that result from the extraction process.
Table 2: Average pigment concentrations from each species examined (μg/g). Dietary Carotenoids
Lutein Isomers
Modified 'Yellows'
Modified 'Reds'
Taxon n ZXN LTN ILa* ILb* ILc* ILd* CXA CXB ADR CXN AXN ADX 3HE S. bellicosa 2 + 6.7 7.3 S. superciliaris 4 11.5 + 19.5 21.2 8.6 30.3 6.8 S. militaris 3 5.3 9.5 19.5 6.7 10.1 4.1 S. neglecta 4 3.7 1.1 0.4 + S. magna 6 4.3 0.7 0.5 1.1 + D. oryzivorous 3 X. xanthocephalus 3 11.9 3.8 C. cela 5 100.2 105.0 18.3 12.4 - 22.6 C. uropygialis 3 9.0 105.0 77.7 17.8 9.2 C. leucoramphus 2 78.9 209.2 28.4 22.4 - 13.6 C. haemorrhous 3 44.9 98.6 51.3 38.6 38.6 C. solitarius 1 P. wagleri 2 26.5 14.0 7.3 3.4 2.4 P. montezuma 1 14.0 13.7 3.3 C. melanicterus 3 65.2 111.6 18.2 14.2 - 15.8 Pigment Abbreviations: ZXN = Zeaxanthin, LTN = Lutein, Ila-d = Isolutein A-D, CXA and CXB = Canary Xanthophyll A and B, ADR = Adonirubin, CXN = Canthaxanthin, AXN = Astaxanthin, ADX = αDoradexanthin, 3HE = 3-hydroxy-echinenone, * = Asterisks denote isomers of lutein produced by the extraction process.
&KDSWHU
39
process is applied to a lutein standard. We found that all taxa with yellow plumage in both the meadowlarks and the caciques and oropendolas exhibited lutein as their major extraction product, and that most yellow taxa of the caciques and oropendolas had both lutein and zeaxanthin (Table 1). All taxa with red plumage included one or more C4-ketocarotenoids. However, this set of C4-ketocarotenoids differed between the red meadowlarks and the red caciques. We found -doradexanthin and 3-hydroxy-echinenone in red meadowlark feathers but not in red cacique feathers. In contrast, we found high concentrations of canary xanthophyll A and B (which are yellow pigments) in red cacique feathers but none in red meadowlark feathers (Table 1). We did not detect carotenoid pigments in the matte black-colored plumage of C. solitaries or D. oryzivorous. Total carotenoid concentrations were on average an order of magnitude lower in the meadowlarks (mean 23.53 μg/g, SD 32.39) than in the caciques and oropendolas (mean 205.16 μg/g, SD 156.92; p < 5 x 10-5 Welch two sample t-test). In contrast, total carotenoid concentrations did not differ significantly between red specimens and yellow specimens (p = 0.271 Welch two sample t-test). Controlling for phylogeny, we found no evidence of a correlation between total carotenoid concentration and spectral location (PGLS: df = 13, p = 0.23), or between total carotenoid concentration and discrete carotenoid hue (p = 0.28). Rather, there was a significant correlation between total C4-ketocarotenoid carotenoid concentration and spectral location (p < 1 x 10-4) and between total C4-ketocarotenoid concentration and discrete carotenoid hue (p < 1 x 10-4). Furthermore, we found a negative correlation
&KDSWHU
40
between lutein concentration and both spectral location (p < 0.05), and discrete carotenoid hue (p < 0.05). We found general agreement between parsimony and likelihood methods of ancestral state reconstruction. All parsimony and likelihood analyses reconstructed the ancestral blackbird node as having lutein present and all other carotenoids absent, albeit with varying degrees of uncertainty. Unordered parsimony reconstructed all nodes unambiguously for eight of the nine pigment characters examined in this study (Figure 1). All analyses also agreed in reconstructing losses of all carotenoid pigmentation in both lineages that lack apparent carotenoid-based coloration. There was considerable uncertainty among ancestral state reconstructions regarding the origin of zeaxanthin. Using parsimony, ancestral state reconstruction of zeaxanthin yielded 6 most parsimonious reconstructions. Using likelihood, reconstructions of zeaxanthin and lutein with an estimated rate model showed considerable uncertainty at several nodes (i.e., support below the decision threshold). However, likelihood reconstruction with the fixed rate model supported a single gain of zeaxanthin in the common ancestor of the caciques and oropendolas. Ancestral state reconstructions strongly suggested that two sets of C4ketocarotenoids were gained in succession in the meadowlarks (Figure 1). doradexanthin and adonirubin were gained in the ancestor of the red meadowlark clade, with a subsequent gain of 3-hydroxy-echinenone, canthaxanthin and astaxanthin in the ancestor of two of those species (S. militaris and S. superciliaris). However, ancestral state reconstructions suggested that a different set of C4-ketocarotenoids, canthaxanthin and adonirubin, were gained in the caciques and oropendolas, once in each red-colored
&KDSWHU
41
+ aDX, ADR
+ ZXN, 3HE, CXN, AXN, – LUT
– ALL
+ CXN, ADR, CXA, CXB, – ZXN + LUT
+ CXN, ADR, AXN CXA, CXB, – LUT
– ALL + ZXN
– ZXN
Melanin Yellow Red
Figure 1: Gains and losses of carotenoid pigments reconstructed using parsimony on the Icterid molecular phylogeny (Price et al. 2009). Hue data from reflectance spectrometry is also reconstructed using parsimony, and represented by branch colors (key at lower left). Gains are represented by + symbols, and losses by – symbols. Carotenoid compound names are abbreviated: 3HE = 3-hydroxy-echinenone, ADR = adonirubin, aDX = a-doradexanthin, AXN = astaxanthin, CXA = canary xanthophyll A, CXB = canary xanthophyll B, CXN = canthaxanthin, LUT = lutein, ZXN = zeaxanthin. Branch lengths indicative of molecular distances from Price et al. (2009).
&KDSWHU
42
taxon, and were accompanied by concurrent gains of canary xanthophylls A and B. However, the presence of astaxanthin was gained in the C. haemorrhous lineage but not in the C. uropygialis lineage. In summary, on all branches where red coloration was inferred to be gained, C4ketocarotenoids were also gained. However, red coloration gained in the meadowlarks differed from red coloration gained in the caciques by the presence of -doradexanthin and the absence of canary xanthophylls. Either lutein or zeaxanthin (but never both) was lost in each lineage showing red coloration.
Discussion Convergence and Parallelism We combined ancestral state reconstruction and pigment biochemistry techniques to infer evolutionary changes in carotenoid-based feather pigmentation. Our results show that meadowlarks and caciques evolved the same red plumage reflectance characteristics using different sets of carotenoid pigments (Table 2; Figure 1). Hence, this pattern of phenotypic expression is the result of convergent evolution. In contrast, we found that red coloration evolved twice in the caciques by what appear to be similar biochemical mechanisms. This finding suggests that the red coloration in two lineages of caciques represents an example of parallel evolution. The biochemical components of red coloration in C. haemmorhous and C. uropygialis are remarkably similar (Table 2). This similarity suggests that the two red cacique lineages may have evolved by changes in the same enzymatic pathway. Unfortunately the identities of the enzymes responsible for modifying carotenoids remain
&KDSWHU
43
unknown. However, researchers have proposed several biochemical pathways by which birds might alter dietary carotenoids to produce novel modified carotenoids (McGraw 2006; Prum et al. 2012). Lutein and zeaxanthin are two of the most common carotenoids in the terrestrial avian diet; they act as the precursors to most other plumage carotenoids. The model for avian carotenoid metabolism (McGraw 2006) predicts that the dehydrogenation and/or oxidation of sites on the precursor carotenoid’s end-rings are responsible for producing many of the various modified carotenoids found in avian plumage (but see Prum et al. 2012). Dehydrogenation of a hydroxyl group to form a C3ketocarotenoid abbreviates the molecule's conjugated system, decreasing the carotenoid's peak absorbance wavelength (Britton 1995). Such C3-ketocarotenoids (e.g., canary xanthophylls) are responsible for the yellow coloration found in canaries (Serinus canaria) and several other songbirds (McGraw 2006). In contrast, oxidation of the C4 site on the carotenoid ring to form a C4-ketocarotenoid lengthens the molecule's conjugated system, increasing the carotenoid's peak absorbance wavelength (Britton 1995). Such C4-ketocarotenoids are responsible for the red coloration found, for example, in male scarlet tanagers (Piranga olivacea; Brush 1967) and house finches (Carpodacus mexicanus; Brush and Power 1976; see McGraw 2006 for other such examples). Caciques have evolved red coloration twice, each time with gains of both the modified red pigments canthaxanthin and adonirubin and the modified yellow pigments canary xanthophyll A and B. This suggests that the evolution of red in caciques involved parallel gains in both the enzymes responsible for both dehydration and C4-oxidation. In contrast, the convergent gain of red coloration in the meadowlarks involved two sequential gains of C4-oxidation. Because of these sequential gains, we speculate
&KDSWHU
44
that the ancestor of S. militaris and S. superciliaris either expanded the substrate specificity of its original carotenoid-oxidizing enzyme, or gained the activity of an additional enzyme. The biochemical machinery necessary for transitions from yellow xanthophylls to red ketocarotenoids has long been present in the retinas of vertebrates, where ketolase/oxidase enzymes produce the astaxanthin found in oil droplets (Robinson 1994). Consequently, shifts from yellow to red coloration may have evoled by the neofunctionalization of existing carotenoid metabolic pathways, rather than as a novelty. Several recent studies of red coloration in the widowbirds and bishops (Euplectes) have also identified convergent origins of red coloration (Andersson et al. 2007; Prager and Andersson 2009, Prager and Andersson 2010). The authors found that modified red carotenoids such as canthaxanthin and -doradexanthin were responsible for producing red coloration in Euplectes ardens and E. orix, whereas large quantities of the modified yellow pigment 3-dehydrolutein were responsible for producing red coloration in E. axillaris (Andersson et al. 2007; Prager et al. 2009). Ancestral state reconstruction of coloration in this clade revealed at least two distinct gains of red coloration by different mechanisms (i.e., convergence sensu strictu; Prager and Andersson 2010). Convergent gains of red plumage coloration have also been observed in cardueline finches (Hill and McGraw 2004). Such repeated observations of changes in the direction of yellow to red suggest that selection for red plumage could be common among passerine birds. Causes of Repeated Gains of Red As with convergence in color pattern (Omland and Lanyon 2000), convergence and parallelism in color may be a consequence of either selection or constraint. Selection for red coloration might arise repeatedly under similar environmental variables such as
&KDSWHU
45
light environment, or might be constrained by dietary pigment availability. While these and other scenarios each have a sound theoretical basis as potential factors leading to red plumage coloration, they seem unlikely to explain the phylogenetic distribution of red in icterids. For example, while caciques do differ in their habitat preferences from montane cloud forests to dry lowland scrub, these differences do not explain color variation as there are yellow species that inhabit each extreme, and several are sympatric with red species (Jaramillo and Burke 1999). Migratory behavior, which correlates strongly with dichromatism in orioles (Friedman et al. 2009), could perhaps explain the maintenance of yellow coloration in the North American clade of migratory meadowlarks. However, this variable cannot explain color variation within the caciques and oropendolas, which are all sedentary. A persistent female preference for red coloration coupled with phyletic inertia might explain repeated convergence and parallelism in red coloration. Whether due to a degree of honesty inherent in red pigmentation (Hill 1996) or a pre-existing bias towards red in the female visual system (Hill and McGraw 2004), all female meadowlarks and caciques might prefer red. Under this scenario, phyletic inertia (i.e., slow evolutionary rate) would retard an inexorable march from yellow to red coloration. A final possible explanation for the repeated evolution of red coloration might be that the hue of carotenoid-based plumage coloration evolves in a character space that is intrinsically constrained. Birds must produce plumage coloration using only those carotenoid compounds that are found in the diet, either directly or as precursors. If these dietary carotenoids are metabolically modified, only a limited number of modifications are possible that change carotenoid absorptive properties (Britton 1995; but see Prum et
&KDSWHU
46
al. 2012). Thus carotenoid hue might be thought of as a bounded, discrete character that can be switched back and forth between yellow and red with the gain or loss of particular enzymes or their expression. Under these circumstances, red coloration might arise repeatedly from a yellow ancestor based on drift and starting conditions alone. Discrete Changes Between Yellow and Red Several evolutionary studies have discussed whether color evolves in a discrete or continuous manner (Roulin 2004; Hofmann et al. 2006; Kiere et al. 2009; Gumm and Mendelson 2011). Authors of a previous study of the caciques described interspecific color variation as discrete, with repeated transitions between yellow and red (Kiere et al. 2009). We found that the presence or absence of C4-ketocarotenoid explained color variation in the form of both spectral location and discrete hue. Indeed, C4ketocarotenoid pigments were present in every red-colored species, and absent in every yellow species (Table 2). Furthermore, continuous measurements of total carotenoid concentration did not explain variation in any measure of coloration. The distinction between treating color as a continuous character or a discrete character is important for other studies that reconstruct ancestral states for this character to test phylogenetic and comparative hypotheses. Researchers should choose appropriate character scoring approaches given the specifics of the system (Hofmann et al. 2006) Yellow Pigments in Yellow Feathers Prior to this study, several researchers had examined plumage pigmentation mechanisms in the New World blackbirds (McGraw et al. 2004; Hofmann et al. 2007; Newbrey and Reed 2011), but these previous studies were largely aimed at distinguishing between carotenoid from melanic pigmentation. Building on this work, we used
&KDSWHU
47
reflectance spectrometry to exclude feather patches whose coloration appeared to be due to melanin pigmentation (e.g, the buff nape of D. oryzivorus). In addition to the differences in the types of red plumage carotenoids detected, our survey of carotenoid pigmentation in the meadowlarks, caciques, and oropendolas also showed that: 1) the meadowlarks and allies use lutein to produce yellow coloration, whereas the caciques and oropendolas use both lutein and zeaxanthin; 2) the caciques and oropendolas deposit carotenoids in their feathers at a concentration roughly ten times greater than the meadowlarks and allies, with no apparent effect on hue. Species from many avian lineages produce yellow feathers and skin by using lutein (McGraw et al. 2003), or a combination of lutein and zeaxanthin (McGraw and Gregory 2004; McGraw and Shuetz 2004). However, those species that deposit only lutein in their feathers circulate zeaxanthin (McGraw et al. 2003), but appear unable to deposit the dietary compound into their feathers. If the New World blackbirds follow this pattern of circulation and deposition, this would suggest that the ancestor of the caciques and oropendolas evolved the ability to incorporate zeaxanthin into its feathers. No Carotenoids in Black Feathers When designing this study, we chose to sample black feathers from two species that were melanistic but nested within carotenoid-colored clades to test whether carotenoids might be present but masked in some species’ plumage. Such masked carotenoids in melanin-pigmented plumage have been observed in two blackbird species that express both types of pigmentation (Xanthocephalus xanthocephalus, Butcher 1991; Icterus spurius, Hofmann et al. 2007). However, we found no evidence of carotenoid pigmentation in the black breast feathers of male D. oryzivorus, or in the black rump
&KDSWHU
48
feathers of C. solitarius (Table 2). There are many explanations for the loss of elaborate traits (see Wiens 2001). In this case, the loss of carotenoid pigmentation may follow the gain of melanic pigmentation, perhaps due to selection against a costly signal that is never sent (Hill 1996). Yellow Pigments in Red Feathers The deposition of high concentrations of modified yellow pigments (Canary Xanthophyll A and B) into red cacique feathers is perplexing. We expect that these pigments should have no effect on the feather's carotenoid hue or redness because the wavelengths they absorb should be absorbed by red C4-ketocarotenoids as well. There are several possible functional and non-functional explanations for their presence. The presence of canary xanthophylls in red cacique feathers could function in reducing plumage reflectance at short wavelengths. Alternatively, the presence of high concentrations of modified yellow pigments in red cacique feathers might be due to a non-functional consequence of evolving red. Simultaneous origins of C4-ketocarotenoid and C3-ketocarotenoid pigments in red cacique lineages would be most parsimonious, however this seems to be an unlikely pair of chemical reactions to be performed by a single enzyme. Hence, the functional explanations for the repeated evolution of high concentrations of canary xanthophylls in red cacique feathers remain unclear and require further study. Conclusion Theory predicts that parallel evolution should be more common among closely related species, as they share similar developmental and genetic pathways (Gould 2002; Wake et al. 2011; but see Arendt and Reznick 2007). Indeed, we found that closely
&KDSWHU
related lineages evolved red coloration using similar mechanisms, whereas distantly related lineages evolved red coloration using different mechanisms. Some studies of color evolution have described traits that are produced by similar biochemical mechanisms, but are the result of mutations in different genes (e.g., Smith and Rausher 2011). Others have described traits produced by different regulators of a similar molecular mechanism (Prud’homme et al. 2006), or even different loci on the same gene (see Mundy 2005). We expect that independent cacique lineages were able to utilize similar metabolic pathways to convert dietary yellow pigments to red ketocarotenoids. In contrast, the distantly related meadowlark lineage has likely utilized a different metabolic pathway to arrive at a similar red phenotype. Red coloration in the caciques and other cases of parallel evolution blur the line between homology and homoplasy, as homologous pathways may be repeatedly utilized across evolutionary history.
Acknowledgements We thank the Academy of Natural Sciences in Philadelphia and the Delaware Museum of Natural History for allowing us to measure and sample feathers from specimens in their collections. We would also like to thank M. Toomey and M. Rowe for their assistance with HPLC analysis, and A. Brush, D. Futuyma, T. Mendelson, and J. Price for interesting and informative discussions regarding convergence and parallelism. N.R.F. was supported by grants from Sigma Xi and the Maryland Ornithological Society. K.E.O. was supported by a National Science Foundation CAREER grant DEB-0347083. The Omland lab is a member of the Smithsonian Ornithology Group.
49
&KDSWHU
50
Literature Cited Andersson, S., and M. Prager 2006. Quantifying Colors. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 41-89. Andersson, S., M. Prager, E.I. Anette Johansson. 2007. Carotenoid content and reflectance of yellow and red nuptial plumages in widowbirds (Euplectes spp.). Funct. Ecol. 21:272-281. Arendt, J., D. Reznick. 2007. Convergence and parallelism reconsidered: what have we learned about the genetics of adaptation? Trends Ecol. Evol. 23:26-32. Arthur, W. 2001. Developmental drive: an important determinant of the direction of phenotypic evolution. Evol. Dev. 3:271-278. Barker, F. K., A. J. Vandergon, S. M. Lanyon. 2008. Assessment of species limits among Yellow-Breasted Meadowlarks (Sturnella spp.) using mitochondrial and sexlinked markers. Auk 125:869-879. Bates, H. W. 1862. Contributions to an insect fauna of the Amazon Valley: Lepidoptera: Heliconidae. Trans. Linnean Soc. Lond. 23:495-566. Britton, G. 1995. UV/Visible Spectroscopy. In: Carotenoids. Vol1B. Spectroscopy (G. Britton, S. Liaaen-Jensen, H. Pfander, eds). Birkhäuser Verlag, Basel, Switzerland. Britton, G., S. Liaaen-Jensen, H. Pfander. 2004. Carotenoids. Handbook. Birkhäuser Verlag, Basel, Switzerland. Brush, A. H. 1967. Pigmentation in the Scarlet Tanager, Piranga olivacea. Condor 69:549-559. Brush, A. H., D. M. Power. 1976. House finch pigmentation: carotenoid metabolism and the effect of diet. Auk 93:725-739. Butcher, G. S. 1991. Mate choice in female Northern Orioles with a consideration of the role of the black male coloration in female choice. Condor 93:82-88. Butler, M. A., A. A. King. 2004. Phylogenetic comparative analysis: a modeling approach for adaptive evolution. Am. Nat. 164:683-695. Cronk, Q., I. Ojeda. 2008. Bird-pollinated flowers in an evolutionary and molecular context. J. Exp. Bot. 59:715-727.
&KDSWHU
51
Dale, J. 2006. Intraspecific variation in coloration. In: Bird Coloration. II. Function and Evolution (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 36-86. Des Marais, D. L., M. D. Rausher. 2010. Parallel evolution at multiple levels in the origin of hummingbird pollinated flowers in Ipomoea. Evolution 64:2044-2054. Eaton, M. D. 2005. Human vision fails to distinguish widespread sexual dichromatism among sexually "monochromatic" birds. PNAS 102:10942-10946. Friedman, N. R., C. M. Hofmann, B. Kondo, K. E. Omland. 2009. Correlated evolution of migration and sexual dichromatism in the New World orioles (Icterus). Evolution 63:3269-3274. Friedman, N. R., L. M. Kiere, K. E. Omland. 2011. Convergent gains of red carotenoidbased coloration in the New World blackbirds. Auk 128:678-687. Futuyma, D. J. 2005. Evolution. Sinauer Associates, Inc., Sunderland, MA. Gould, S. J. 2002. The structure of evolutionary theory. Harvard Univ. Press, Cambridge, MA. Grafen, A. 1989. The phylogenetic regression. Phil. Trans. R. Soc. Lond. B 326:119-157. Gumm J. M., Mendelson T. C., 2011. The evolution of multi-component visual signals in darters genus Etheostoma. Curr. Zool. 57: 125-139. Hall, B. K. 2003. Descent with modification: the unity underlying homology and homoplasy as seen through an analysis of development and evolution. Biol. Rev. 78:409-433. Harmon, L. J., J. T. Weir, C. D. Brock, R. E. Glor, W. Challenger. 2008. GEIGER: investigating evolutionary radiations. Bioinformatics 24:129-131. Hill, G. E. 1991. Plumage coloration is a sexually selected indicator of male quality. Nature 350:337-339. Hill, G. E. 1992. Proximate basis of variation in carotenoid pigmentation in male house finches. Auk 109:1-12. Hill, G. E. 1996. Redness as a measure of the production cost of ornamental coloration. Ethol. Ecol. Evol. 8:157-175.
&KDSWHU
52
Hill, G. E. 2006. Environmental regulation of ornamental coloration. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 507-560. Hill, G. E., K.J. McGraw. 2004. Correlated changes in male plumage coloration and female mate choice in cardueline finches. Anim. Behav. 67:27-35. Hofmann, C. M., T. W. Cronin, K. E. Omland. 2006. Using spectral data to reconstruct evolutionary changes in coloration: carotenoid color evolution in new world orioles. Evolution 60:1680-1691. Hofmann, C. M., K.J. McGraw, T.W. Cronin, K.E. Omland. 2007. Melanin coloration in New World orioles I: carotenoid masking and pigment dichromatism in the Orchard Oriole complex. J. Avian Biol. 38:163-171. Hudon, J. 1991. Unusual carotenoid use by the Western Tanager (Piranga ludoviciana) and its evolutionary implications. Can. J. Zool 69:2311-2320. Hudon, J. 1994. Showines, carotenoids, and captivity: a comment on Hill (1992). The Auk 111:218-221. Hugall, A. F., D. Stuart-Fox. 2012. Accelerated speciation in colour-polymorphic birds. Nature 485:631-635. Inouye, C. Y., G. E. Hill, R. D. Stradi, R. Montgomerie. 2001. Carotenoid pigments in male house finch plumage in relation to age, subspecies, and ornamental coloration. Auk 118:900-915. Jaramillo A., P. Burke. 1999. New World Blackbirds: The Icterids. Princeton: Princeton University Press. Johnson, K. P., S. M. Lanyon. 2000. Evolutionary changes in color patches of blackbirds are associated with marsh nesting. Behav. Ecol. 11:515-519. Kiere, L. M., C. M. Hofmann, J. J. Price, T. W. Cronin, K. E. Omland. 2009. Discrete evolutionary color changes in caciques suggest different modes of carotenoid evolution between closely related taxa. J Avian Biol. 40:605-613. Kodric-Brown, A. 1989. Dietary carotenoids and male mating success in the guppy: an environmental component to female choice. Behav. Ecol. Sociobiol. 25:393-401. Lanyon, S. M., K.E. Omland. 1999. A molecular phylogeny of the blackbirds (Icteridae): five lineages revelaed by cytochrome-b sequence data. Auk 116:629-639.
&KDSWHU
53
Lanyon S. M., F. K. Barker 2007. Exploring patterns of morphological evolution in the New World blackbirds. In: 125th Meeting of the American Ornithologists’ Union. Laramie: American Ornithologists’ Union. Maddison, W. P., D. R. Maddison. 2011. Mesquite: a modular system for evolutionary analysis. Version 2.75 http://mesquiteproject.org Maechler, M., Rousseeuw, P., Struyf, A., Hubert, M., Hornik, K. 2012. cluster: Cluster Analysis Basics and Extensions. R package version 1.14.2. Mays, H. L., K. J. McGraw, G. Ritchison, S. Cooper, V. Rush, R. S. Parker. 2004. Sexual dichromatism in the yellow-breasted chat Icteria virens: spectrophotometric analysis and biochemical basis. J Avian Biol. 35:125-134. McGraw, K. J. 2006. Mechanics of carotenoid-based coloration. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 177-242. McGraw, K. J., J. G. Shuetz. 2004. The evolution of carotenoid coloration in estrildid finches: a biochemical analysis. Comp. Biochem. Phys. B 139:45-51. McGraw, K. J., A. M. Stoehr, P. M. Nolan, G. E. Hill. 2001. Plumage redness predicts breeding onset and reproductive success in the House Finch: a validation of Darwin's theory. J Avian Biol. 32:90-94. McGraw, K. J., G. E. Hill, R. Stradi, R. S. Parker. 2002. The effect of dietary carotenoid access on sexual dichromatism and plumage pigment composition in the American Goldfinch. Comp. Biochem. Physiol. B 131:261-269. McGraw, K. J., M. D. Beebee, G. E. Hill, R. S. Parker. 2003. Lutein-based plumage coloration in songbirds is a consequence of selective pigment incorporation into feathers. Comp. Biochem. Physiol. B 135:689-696. McGraw, K. J., K. Wakamatsu, A. B. Clark, K. Yasukawa. 2004. Red-winged blackbirds Agelaius pheoniceus use carotenoid and melanin pigments to color their epaulets. J Avian Biol. 35:543-550. McGraw, K. J., J. Hudon, G. E. Hill. 2005. A simple and inexpensive chemical test for behavioral ecologists to determine the presence of carotenoid pigments in animal tissues. Behav. Ecol. Sociobiol. 57:391-397.
&KDSWHU
54
McGraw, K. J., P. M. Nolan, O. L. Crino. 2006. Carotenoid accumulation strategies for becoming a colorful House Finch: analyses of plasma and liver pigments in wild moulting birds. Funct. Ecol. 20:678-688. McLennan, D. A., D.R. Brooks. 1993. The Phylogenetic Component of Cooperative Breeding in Perching Birds: A Commentary. Am. Nat. 141:790-795. Mundy, N. I. 2005. A window on the genetics of evolution: MC1R and plumage coloration in birds. Proc. R. Soc. Lond. B 272:1633-1640. Newbrey, J. L., and W. L. Reed. 2011. Yolk and feather carotenoids in relation to female condition and reproduction in the yellow-headed blackbird. Auk 128(2): 382-392. Olson, V. A., I.P.F. Owens. 1998. Costly Sexual Signals: Are Carotenoid Rare, Risky, or required? Trends in Ecology and Evolution 13:510-515. Omland, K. E. 1999. The assumptions and challenges of ancestral state reconstruction. Syst. Biol. 48:604-611. Omland, K. E., S.M. Lanyon. 2000. Reconstructing plumage evolution in orioles (Icterus): repeated convergence and reversal in patterns. Evolution 54:2119-2133. Prager, M., S. Andersson. 2009. Differential ability of carotenoid C4-oxygenation in yellow and red bishop species (Euplectes spp.). Comp. Biochem. Physiol. B 154:373-380. Prager, M., S. Andersson. 2010. Convergent evolution of red carotenoid coloration in Widowbirds and Bishops (Euplectes spp.). Evolution 64:3609-3619. Price, J. J., S. M. Lanyon. 2002. A robust phylogeny of the oropendolas: polyphyly revealed by mitochondrial sequence data. Auk 119:335-348. Price, J. J., N. R. Friedman, K. E. Omland. 2007. Song and plumage evolution in the New World orioles (Icterus) show similar lability and convergence in patterns. Evolution 61:850-863. Price, J. J., S. M. Lanyon. K. E. Omland. 2009. Losses of female song with changes from tropical to temperate breeding in the New World blackbirds. Proc. Biol. Sci 276:1971-1980. Prud'homme, B., N. Gompel, A. Rokas, V.A. Kassner, T.M. Williams, S. Yeh, J.R. True, S.B. Carroll. 2006. Repeated morphological evolution through cis-regulatory changes in a pleiotropic gene. Nature 440:1050-1053.
&KDSWHU
55
Prum, R. O., A. M. LaFountain, J. Berro, M. C. Stoddard, H. A. Frank. 2012. Molecular diversity, metabolic transformation, and evolution of carotenoid feather pigments in cotingas (Aves: Cotingidae). J. Comp. Physiol. B. doi:10.1007/s00360-0120677-4. R Development Core Team. 2011. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/. Reed, R. D., M. S. Serfas. 2004. Butterfly wing pattern evolution is associated with changes in a notch/distal-less temporal pattern formation process. Curr. Biol. 14:1159-1166. Reed, R. D., R. Papa, A. Martin, H. M. Hines, B. A. Countermain, C. Pardo-Diaz, C. D. Jiggins, N. L. Chamberlain, M. R. Kronforst, R. Chen, G. Halder, H. F. Nijhout, W. O. McMillan. 2011. optix drives the repeated convergent evolution of butterfly wing pattern mimicry. Science 333:1137-1141. Robinson, S. R. 1994. Early vertebrate colour vision. Nature 367:121. Roulin, A. 2004. The evolution, maintenance and adaptive function of genetic colour polymorphism in birds. Biol. Rev. 79, 815-848. Rowe, M., K. J. McGraw. 2009. Carotenoids in the seminal fluid of wild birds: interspecific variation in fairy-wrens. Condor 110:694-700. Saks, L., K McGraw, P. Horak. 2003. How feather colour reflects its carotenoid content. Funct. Ecol. 17:555-561. Smith, S. D., M. D. Rausher. 2011. Gene loss and parallel evolution contribute to species difference in flower color. Mol. Biol. Evol. 28:2799-2810. Stoddard, M. C., R. O. Prum. 2008. Evolution of avian plumage color in a tetrahedral color space: a phylogenetic analysis of New World buntings. The American Naturalist 171:755-776. Toomey, M. B., K. J. McGraw. 2007. Modified saponification and HPLC methods for analyzing carotenoids from the retina of quail: implications for its use as a nonprimate model species. Invest. Ophth. Vis. Sci. 48:3976-3982. Wake, D. B., M. H. Wake, C. D. Specht. 2011. Homoplasy: from detecting pattern to determining process and mechanism of evolution. Science 331:1032-1035.
&KDSWHU
56
Westneat, D. F., A. B. Clark, K. C. Rambo. 1995. Within-brood patterns of paternity and paternal behavior in Red-winged Blackbirds. Behav. Ecol. Sociobiol. 37:349-356. Wiens, J. J. 2001. Widespread loss of sexually selected traits: how the peacock lost its spots. Trends Ecol. Evol. 16:517-523. Wiens, J. J., C. A. Kuczynski, W. E. Duellman, T. W. Reeder. 2007. Loss and reevolution of complex life cycles in marsupial frogs: Does ancestral trait reconstruction mislead? Evolution 61:1886-1899.
&KDSWHU
57
Appendix 1: Voucher specimen data Taxon Cacicus cela Cacicus cela Cacicus cela Cacicus cela Cacicus cela Cacicus haemorrhous Cacicus haemorrhous Cacicus haemorrhous Cacicus leucoramphus Cacicus leucoramphus Cacicus melanicterus Cacicus melanicterus Cacicus melanicterus Cacicus solitarius Dolichonyx oryzivorus Dolichonyx oryzivorus Dolichonyx oryzivorus Psarocolius montezuma Psarocolius wagleri Psarocolius wagleri Sturnella bellicosa Sturnella bellicosa Sturnella magna Sturnella magna Sturnella magna Sturnella magna Sturnella magna Sturnella magna Sturnella militaris Sturnella militaris Sturnella militaris Sturnella neglecta Sturnella neglecta Sturnella neglecta Sturnella superciliaris Sturnella superciliaris Sturnella superciliaris Xanthocephalus xanthocephalus Xanthocephalus xanthocephalus Xanthocephalus xanthocephalus
Museum ANSP ANSP ANSP DMNH DMNH ANSP ANSP ANSP ANSP ANSP DMNH DMNH DMNH DMNH DMNH DMNH DMNH DMNH DMNH DMNH ANSP ANSP ANSP ANSP ANSP DMNH DMNH DMNH ANSP ANSP ANSP ANSP ANSP ANSP ANSP ANSP DMNH DMNH DMNH DMNH
Specimen Number 133168 149129 182885 58614 65491 152332 152333 189079 144985 154733 17310 27438 45956 27829 407 2555 40804 17295 27441 45972 60297 109375 181439 182158 186457 45034 34813 45049 64527 105285 168284 183649 183651 183652 78990 78991 31027 4463 22229 23395
Locality Bolivia, Rio Chapare Colombia, Chela Ecuador, Santiago Ecuador, Napo Bolivia, Santa Cruz Colombia, Caquenta Colombia, Caquenta Guyana, Potaro-Siparuni Colombia, Cauca Colombia, Tolima Mexico, Sinaloa Mexico, Guerrero Mexico, Chiapas Paraguay, Boqueron USA, North Carolina, Wake USA, Florida, Dade Canada, Ontario, Algoma Honduras, La Ceiba Mexico, Veracruz Costa Rica, Cartago Ecuador, Chimbo Peru, Libertad USA, Missouri, Nodaway USA, Missouri, Nodaway USA, Pennsylvania, Bradford USA, New York, Tompkins USA, Oklahoma, Payne USA, Mississippi, Harrison Colombia, Bolivar Trinidad Colombia, Meta USA, Colorado, Weld USA, Colorado, Weld USA, Colorado, Weld Argentina, La Plata Argentina, La Plata Paraguay, Boqueron USA, California, Santa Clara USA, New Mexico, San Juan Mexico, Durango
&KDSWHU
58
CHAPTER 3: HISTORY AND MECHANISMS OF CAROTENOID PLUMAGE EVOLUTION IN THE NEW WORLD ORIOLES (ICTERUS)
NICHOLAS R. FRIEDMAN1, KEVIN J. MCGRAW2, KEVIN E. OMLAND1
1
Department of Biological Sciences, University of Maryland, Baltimore County, Baltimore, MD 21250 2
School of Life Sciences, Arizona State University, Tempe, AZ 85287
Email:
[email protected],
[email protected],
[email protected]
&KDSWHU
59
Abstract Carotenoid pigments are responsible for much of the brilliant yellow, orange, and red plumage coloration displayed by many songbird species. While many recent studies have focused on the functions of carotenoids in visual signaling, they seldom address the phylogenetic origins of coloration and its mechanisms. Here, we used the New World orioles (Icterus) as model clade to study the history of orange carotenoid-based coloration and pigmentation, sampling 47 museum specimens from across 12 species. We examined the identity and concentration of carotenoids in oriole feathers using HPLC, and compared these observations to reflectance measurements using phylogenetic comparative methods. Each of the seven yellow orioles we sampled, which were distributed across all three major oriole clades, used only lutein to color their feathers. Ancestral state reconstruction of this trait suggests that the oriole common ancestor had yellow feathers pigmented with lutein. We found keto-carotenoids in small concentrations in the plumage of each of the five orioles scored as orange. This correlation, which was significantly supported by multiple methods, suggests that discrete gains and losses of keto-carotenoids are behind independent gains of orange coloration in orioles. In constrast, total carotenoid concentration was not associated with hue, and total keto-carotenoid concentration poorly explained variation among species in which they were present. These findings suggest that oriole most likely evolved orange at least twice, each time by gaining the ability to metabolize dietary carotenoids by C4-oxygenation. The presence of this metabolic ability raises a new question: why aren't there any red orioles?
&KDSWHU
60
Introduction Color evolution is a convenient model for understanding evolutionary processes in general, and has been an inviting challenge to biologists since Darwin. In this study, we use an approach to addressing this challenge that synthesizes two of Tinbergen’s four questions, phylogeny and proximate mechanism, to examine the history of a trait’s evolution. Such an approach is possible now due to advances in modern biochemical and phylogenetic methods, and has been used successfully to study the evolution of coloration in mice, irises, and butterflies (Steiner et al. 2009; Smith and Rausher 2011; Reed et al. 2011). In this study, we use similar a similar approach to examine how evolution has shaped the proximate mechanisms of plumage coloration in songbirds to produce the color variation seen in extant species. The use of carotenoid pigments to color feathers is widespread among songbirds (McGraw 2006). Carotenoid pigments are not synthesized endogenously by animals, but must instead be obtained in their diet. Despite this limitation, carotenoids are responsible for much of the plumage coloration in songbirds that appears yellow, orange, or red to humans. For example, Yellow Warblers (Setophaga petechia) appear yellow due to the lutein that the species ingests in its diet (McGraw et al. 2003). In addition, many carotenoids that are not found in the avian diet are found in plumage, and these are most likely produced by biochemical modification of dietary carotenoids (Brush 1967; Fox et al. 1969). For example, the yellow in American Goldfinches (Spinus tristis) is the result of canary xanthophylls deposited into feathers (McGraw and Hill 2001). These yellow xanthophylls are modified yellow carotenoids most likely produced by the C3oxygenation of lutein or zeaxanthin (Prager and Andersson 2009). However, a similar
&KDSWHU
61
reaction produces red coloration when performed at the C4 site on a carotenoid ring. This produces a keto-carotenoid that appears red due to the extended length of its conjugated system (Britton 1995). Such modified red compounds are responsible for the red plumage coloration exhibited by House Finches (Carpodacus mexicanus; Hill 1992), and many other birds (see McGraw 2006). Many evolutionary transitions from yellow to red coloration in songbirds are expected to be the result of gains of this mechanism of C4oxygenation (Figure 1; Andersson et al. 2007; Prager and Andersson 2009; Friedman et al. Ch. 2). Orange carotenoid-based coloration is similarly produced by the deposition of keto-carotenoids into feathers (Hudon 1991), but is more rare among songbirds than yellow or red (N. Friedman and S. Lutrell unpublished data). Perhaps due to this rarity, few studies have focused on species with orange plumage, and fewer still have explored the proximate mechanisms by which they color their feathers (e.g., Hofmann et al. 2007; Reudink et al. 2009). However, the New World orioles (Icterus) offer a promising opportunity to study the evolution of orange plumage because they are a speciose genus with plumage that varies continuously from yellow to orange (Jaramillo and Burke 1999; Hofmann et al. 2006). Orioles have been studied extensively as a model clade, due to their diverse color patterns, breeding behaviors, and life histories (e.g., Omland and Lanyon 2000; Price et al. 2007; Friedman et al. 2009). The phylogenetic relationships of species in Icterus have been inferred using multiple independent loci (Omland et al. 1999; Allen and Omland 2003; Jacobsen et al. 2010), and their phylogeny continues to be resolved as both sequencing and phylogenetic inference methods advance (Jacobsen and Omland 2011). Studies by our research group in other blackbird clades have described
62 &KDSWHU
Legend C3-oxygenation C4-oxygenation
β-carotene
3-hydroxy-echinenone (3HE)
echinenone (ECH)
astaxanthin
adonirubin
canthaxanthin (CXN)
Modified Red Carotenoids (Keto-carotenoids)
β-cryptoxanthin
β-doradexanthin
Dietary Yellow Carotenoids
zeaxanthin (ZXN)
change observed in this study
3-dehydro-lutein
Modified Yellow Carotenoids
canary xanthophyll B (CXB)
Lutein (LUT)
α-doradexanthin
canary xanthophyll A (CXA)
Figure 1: Abbreviated diagrams of carotenoid pigments commonly observed in bird plumage, and the hypothesized reactions responsible for their metabolism (adapted from
Andersson et al. 2007). Compounds observed in this study have abbreviations below that are used throughout. Arrows pointing to the left indicate the addition of a ketone group at
the C3 position on the carotenoid ring (C3-oxygenation), while arrows pointing to the right indicate the addition of a ketone group at the C4 position on the carotenoid ring (C4oxygenation). Reactions hypothesized to be responsible in for gains of orange plumage coloration in orioles are highlighted in orange.
&KDSWHU
63
discrete variation in carotenoid coloration and pigmentation between yellow and red species (Kiere et al. 2009; Friedman et al. Ch. 2). The continuous variation of oriole coloration from yellow to orange (Hofmann et al. 2006) leads to a question that can be uniquely addressed in orioles: what biochemical changes explain variation in plumage coloration between yellow and orange plumage? The molecular and genetic mechanisms of carotenoid color variation among bird species are still unknown, however hypothetical models of these mechanisms have been described (see McGraw 2006; Prager and Andersson 2009; Prum et al. 2012). Ultimately, researchers must determine the identity of the C4-oxygenase that produces the ketocarotenoids responsible for red coloration to elucidate how variation among yellow and orange species is produced at the molecular level. However, an important first step to understanding the evolution of carotenoid-based coloration is to examine how pigments vary across species, and how they are related to color variation. This approach has been used by previous studies examining evolutionary transitions from yellow to red coloration (Prager and Andersson 2010; Friedman et al. Ch. 2). Many studies have examined carotenoid-based color variation within species (reviewed in Dale 2006). Most of these have suggested that variation in keto-carotenoid concentration explains variation in hue among individuals (Hill 1994; McGraw et al. 2006). Does oriole plumage coloration vary among species in a similar fashion? Specifically, does continuous variation in carotenoid concentration explain variation in color from yellow to orange? Alternatively, there might be discrete pigment differences involved in transitions from yellow to orange, as seen in transitions from yellow to red (Friedman et al. Ch. 2; Prager and Andersson 2009)? In addition, we ask whether orioles
&KDSWHU
64
have evolved orange independently using similar carotenoid pigments, or different ones (see Friedman et al. Ch. 2). To address these questions, we had three primary goals. First, to use HPLC biochemistry techniques to extract, identify, and quantify the carotenoids embedded in the breast feathers of yellow and orange orioles. Second, to reconstruct the history of gains and losses of carotenoid pigments on the oriole phylogeny. Lastly, we aimed to compare the types of carotenoids present in oriole plumage and their concentrations to quantitative measurements of plumage coloration using phylogenetic comparative methods. Methods Sampling Vouchered museum specimens are an ideal means to obtain rich data for comparative studies. Consequently, we sampled 3-5mg of feather material from the breast patches of 47 specimens at the Academy of Natural Sciences in Philadelphia and the Delaware Museum of Natural History from 12 Icterus species (Clements 2007). We sampled adult males collected during the breeding season, particularly those that appeared well preserved.
Color Scoring Prior to feather removal, we collected reflectance spectra from each museum specimen. This was accomplished using an Ocean Optics USB2000 reflectance spectrometer with a PX-2 pulsed xenon light source (Ocean Optics, Dunedin, FL), calibrated with a white Spectralon standard (Labsphere, North Sutton, NH). To score plumage coloration, we calculated reflectance midpoint values (λR50; Montgomerie 2006;
&KDSWHU
65
hereafter spectral location) for each reflectance spectrum. These values describe the hue of carotenoid-colored plumage, and have been used for this purpose in similar studies (e.g., Hofmann et al. 2006; Andersson et al. 2007). To score carotenoid hue as a discrete character, we followed the character states delineated in Friedman et al. (2011). In that study, our research group showed that there is a bimodal distribution of spectral location values across the New World blackbirds (Icteridae). Species with coloration that appears yellow or orange to humans had spectral location values between 500nm and 560nm, whereas those species that appear red to humans had spectral location values between 580nm and 605nm (Friedman et al. 2011; their Figure 1A). Only species in the oriole genus had spectral location values between 540nm and 560nm (Friedman et al. 2011; their Figure 1B). Consequently, we assigned arbitrary character states based on spectral location values: “Yellow” between 500nm and 540nm, “Orange” between 540nm and 560 nm, and “Red” between 580nm and 605nm (Friedman et al. 2011). In this study, we score discrete color based on those same criteria.
HPLC Biochemistry We used methods described in Friedman et al. (Ch. 2) to extract and identify carotenoid pigments in oriole feathers. Following McGraw et al. (2005), we extracted carotenoid pigments from 3-5mg of trimmed feather barbs using acidified pyridine. To separate the carotenoids from any other possible feather colorants, we extracted them into a solution of hexane and t-butyl methyl ether (1:1, v/v) under centrifugation. After this process, we visually inspected feathers for presence of phaeomelanin, but did not observe any. Prior to analysis with HPLC, we evaporated the organic layer under nitrogen for
&KDSWHU
66
overnight storage at -80º C. We then resuspended the carotenoids in the HPLC mobile phase (methanol: acetonitrile: dichloromethane, 42: 42: 16, v/v/v). Samples were analyzed using a Waters 2695 HPLC instrument and a Waters YMC Carotenoid column (5µm, 4.6mm x 250mm; Waters Corp., Milford, MA) at 30º C. Absorbance data was collected for each sample from 250nm to 600nm using a Waters 2996 photodiode array. As in Toomey and McGraw (2007), we used a three-part gradient HPLC protocol that is capable of separating and detecting both carotenes and xanthophylls. We identified and quantified carotenoids and their concentrations using external standards (following Rowe and McGraw 2009) while blind to each sample’s taxonomic identity. Phylogenetic Comparative Methods To examine the evolution of carotenoid coloration and pigmentation in a phylogenetic context, we used the phylogenetic relationships inferred by Jacobsen et al. (2010) based on six Z-linked nuclear introns. For our analyses, we used the maximum posterior probability tree from that study, which was produced in MrBayes 3.1 (Ronquist and Huelsenbeck 2003). This tree exhibits minor differences in topology compared to the mitochrondrial tree (Omland et al. 1999), particularly in clade C. As the topology of a more recent multi-locus species tree also differs in clade C (Jacobsen and Omland 2011), we compared ancestral state reconstructions in clade C across these three published phylogenies. For use in likelihood methods (which cannot accept taxa with missing data), we included only species for which we collected pigment data. We reconstructed ancestral states for color and pigment characters (Table 1) using both parsimony and likelihood-based inference methods in Mesquite 2.74 (Maddison and Maddison 2010). Ancestral state reconstruction requires the estimation of many
67 &KDSWHU
5 6 3 5 6 4 3 3 3 3 3 3
1.4 4.4 -
0.7 1.7 + -
5.1 8.8 0.4 1.0 -
-
-
-
Dietary Yellow
83.4 99.6 68.6 56.8 115.4 78.6 76.7 60.4 121.8 271.3 71.6 140.0
7.7 24.6 -
Modified Yellow
21.2 57.1 45.2 60.7 -
11.3 32.8 24.2 34.0 -
n [3HE] [ECH] [CXN] [AXN] [ADX] [ADR] [LUT] [ZXN] [CXA] [CXB]
Keto-carotenoids
5.8 10.5 1.9 + 5.4 -
121.7 99.6 68.6 157.2 115.4 157.5 76.7 60.4 121.8 271.3 196.3 140.0
"Orange" "Yellow" "Yellow" "Orange" "Yellow" "Orange" "Orange" "Yellow" "Yellow" "Yellow" "Orange" "Yellow"
Discrete Hue
Color Characters
551.8 533.5 513.7 551.2 514.2 550.5 549.3 519.3 526.7 526.0 557.3 516.7
Total [KetoTotal Spectral carotenoid] [Carotenoid] Location (nm)
Total Concentrations
Table 1: Concentrations of carotenoid pigments extracted from oriole breast feathers, and color characters scored from reflectance measurements taken from specimens prior to feather removal. All concentrations below (indicated by brackets) are in µg/g. Taxa follow Clements (2007). Discrete Hue scored as in Friedman et al. (2011).
Taxon I. bullockii I. cucullatus I. dominicensis I. galbula I. graduacauda I. gularis I. croconotus I. mesomelas I. nigrogularis I. pectoralis I. pustulatus I. prosthemelas
3HE = 3-hydroxy-echinenone, ECH = echinenone, AXN = astaxanthin, ADX = α-doradexanthin, ADR = adonirubin, LUT = lutein, ZXN = zeaxanthin, CXA = canary xanthophyll A, CXB = canary xanthophyll B
Variables
Test
< 0.050* < 0.001*** 0.040* 0.010** 0.382 0.234
P
+ + + + + +
Relationship
Table 2: Results of comparative phylogenetic analyses using phylogenetic generalized least squares (PGLS) and Pagel's discrete method (Pagel 1994). Explanatory variables are listed on the left and response variables on the right. [Total] = Total carotenoid concentration. [Keto] = Total keto-carotenoid concentration. Discrete Keto vs. Discrete Color Pagel 94 Discrete Keto vs. Spectral Location PGLS [Keto] vs. Spectral Location PGLS [Keto] vs. Discrete Color PGLS [Total] vs. Spectral Location PGLS [Total] vs. Discrete Color PGLS Significance level: ***p < 0.001, **p < 0.01, *p < 0.05
&KDSWHU
68
parameters, and may be vulnerable to violations of its assumptions (e.g., that the rate model used is accurate; Omland 1999; Wiens et al. 2007). To address this, we compared ancestral states across 1-parameter (Lewis 2001) and 2-parameter (Pagel 1994) rate models estimated from data in this study and estimated from a more thorough taxonomic sampling by Friedman et al. (2011). We used phylogenetic comparative methods to study the relationships between pigment and color characters. Using the ouch package in R (version 2.8; King and Butler 2009), we tested whether the Ornstein-Uhlenbeck model (Butler and King 2004) was a significantly better fit to our data and tree than the Brownian Motion (BM) model; it was not. Consequently, we used Phylogenetic Generalized Least Squares (PGLS; Grafen 1989) under a BM model in the ape package (version 3.0; Paradis et al. 2004) to test relationships among continuous and discrete characters. To examine relationships solely among discrete characters, we used Pagel’s Discrete test (Pagel 1994) as implemented in Mesquite 2.74. Together, these methods correct for the phylogenetic non-independence of comparative data. By combining multiple scoring and comparative methods, we test for correlated evolution of traits in a way that is robust to the limitations of a single method (Freckleton 2009).
Results Carotenoid Compounds Observed We detected the presence of seven distinct carotenoid compounds in oriole plumage, all of which were xanthophylls (Table 1). There are four carotenoid compounds found in the avian diet that are expected to act as precursors for the majority of
&KDSWHU
69
compounds found in avian plumage: -carotene, -cryptoxanthin, lutein, and zeaxanthin (Figure 1). As with previous studies of plumage pigmentation in blackbirds (Friedman et al. Ch. 2), we did not detect the presence of -carotene or -cryptoxanthin despite an HPLC protocol designed to detect these compounds if present (see McGraw 2006; Toomey and McGraw 2007).We detected lutein in high concentrations (mean = 103.67µg/g, SD = 58.78) in the plumage of each species examined in this study. We detected zeaxanthin in low to intermediate concentrations in the Altamira Oriole (I. gularis) and the Streak-backed Oriole (I. pustulatus; Table 1). However, we found no evidence of either -carotene or -cryptoxanthin in oriole plumage. We observed keto-carotenoids at low total concentrations (mean = 4.7µg/g, SD = 4.05) in the plumage of five species examined in this study. Among clade C orioles, we observed four of six species with plumage containing keto-carotenoid compounds. The Baltimore Oriole (I. galbula) and Bullock’s Oriole (I. bullockii) had plumage containing canthaxanthin and echinone, while I. gularis and I. pustulatus had plumage containing canthaxanthin and 3-hydroxy-echinenone (Table 1). In the plumage of the Orange-backed Troupial (I. croconotus), we at first found no evidence of any keto-carotenoids. However, subsequent examination of chromatography data from that species suggested the presence of echinenone in a low, unquantifiable concentration.
Color and Keto-carotenoid Concentration Using reflectance spectrometry methods described above and by Friedman et al. (2011), we scored five species as having “orange” coloration (Table 1). We found ketocarotenoids in the plumage of each of these species (albeit not in I. croconotus during our
&KDSWHU
70
initial scoring procedure). While these keto-carotenoids showed considerable variation in their concentration, keto-carotenoid concentration did not appear to explain color variation among species with “orange” coloration (Figure 2). Furthermore, the types of keto-carotenoid compounds present in each species’ plumage did not appear to explain color variation among species with “orange” coloration. We used the phylogenetic comparative methods described above and by Friedman et al. (Ch. 2) to examine the relationship between carotenoid pigmentation and hue across all species sampled in this study. Discrete carotenoid hue and the presence or absence of keto-carotenoids were perfectly correlated among species without phylogenetic correction. Keto-carotenoids were present in each “orange” species, and absent in every “yellow” species. Using Pagel’s (1994) discrete method, we found a significant correlation between these two characters (df = 8, p < 0.05). With PGLS, we also found significant correlations between total keto-carotenoid concentration and both spectral location and discrete hue (p < 0.05; p = 0.01; Table 2). Furthermore, we found that there was very strong evidence for a correlation between the presence or absence of ketocarotenoids and spectral location (p < 0.001). However, there was no evidence for any relationship between total carotenoid concentration and either discrete or continuous scoring of hue.
Ancestral State Reconstruction All species examined in this study were found to have lutein in their plumage; consequently the common ancestor of the New World orioles was reconstructed with lutein as present in the oriole common ancestor across all of the methods we used. We
&KDSWHU
CXN,ECH
CXN,3HE
530
540
550
ECH
520
Spectral Location (nm)
71
Lutein only Keto- present 0
2
4
6
8
10
[Keto-carotenoids] (µg/g) Figure 2: Comparison of breast plumage coloration and the total concentration of keto-carotenoids in feathe feathers sampled at same plumage across oriole species. Points filled orange denote species in which keto-carotenoids were from that observed as present, points filled yellow denote species in which keto-carotenoids were observed as absent. Labels indicate which keto-carotenoids were observed in each species (see Table 1 for key to abbreviations). Note that echinenone (ECH) was observed in an unquantifiable concentration in I. croconotus. The dotted line at 540nm refers to the arbitrary cutoff for discrete character states described in Friedman et al. (2011).
&KDSWHU
72
found two independent gains and one loss of “orange” coloration using unordered parsimony (Figure 3). However, several likelihood reconstruction methods conflicted with this result: ancestral state reconstruction using a 2-parameter likelihood model showed orange coloration as ancestral to clade C with two subsequent losses, and reconstruction using a 1-parameter likelihood model was uninformative due to a flat likelihood surface. The other yellow xanthophylls found in this study, zeaxanthin and the canary xanthophylls A and B, were reconstructed as gained in the clade C orioles, but not in clade B orioles. Canary xanthophylls A and B were most likely gained concurrently with orange coloration in clade C, as inferred by both parsimony and likelihood. Zeaxanthin was most likely either gained twice independently in I. gularis and I. pustulatus, or gained once in their common ancestor with a subsequent loss in I. nigrogularis. Of the keto-carotenoids found in oriole feathers, canthaxanthin was reconstructed as gained concurrently with “orange” coloration in the clade C orioles (Figure 3), but not in clade B orioles. Ancestral states for echinenone and 3-hydroxy-echinenone were reconstructed with considerable uncertainty individually. Unordered parsimony was uninformative for both characters within Clade C. Likelihood methods showed repeated terminal gains of both characters in Clade C orioles. However, a composite character of carotenoid pigments (see Table 1; reconstructed in Figure 3) showed the most resolved reconstruction of keto-carotenoids across the New World orioles, with a transition from producing echinenone to producing 3-hydroxy-echinenone occurring in Clade C.
Discussion
&KDSWHU
73
Carotenoid Pigments in Oriole Feathers We found that gains and losses of keto-carotenoids best explained variation in carotenoid-based breast coloration among orioles, a conclusion strongly supported by significant correlations between multiple methods of scoring and analysis (Table 2). Each gain of “orange” coloration saw a concurrent gain of keto-carotenoids (Figure 3). Continuous measures of keto-carotenoid concentration were correlated with coloration (PGLS, p < 0.05). However, variation in keto-carotenoid concentration did not appear to explain variation in hue among “orange” taxa (Figure 2). Furthermore, keto-carotenoids were absent or undetectable in all taxa with an observed spectral location value less than 540nm, including those described as “orange-yellow” in the literature (I. cucullatus, I. pectoralis; Jaramillo and Burke 1999). This finding is similar to results from similar studies in blackbirds and in widowbirds (Prager and Andersson 2009; Friedman et al., Ch. 2) in that gains of carotenoid C4-oxygenation are likely responsible for changes in carotenoid hue. Previous studies have described oriole color variation as continuous and unimodal (Hofmann et al. 2006; Friedman et al. 2011). Consequently, we had expected that continuous variation of keto-carotenoid concentration would explain color variation among orioles. Variation in pigment concentration has been reported to control intraspecific variation in hue in several well-studied species (McGraw and Gregory 2004; McGraw et al. 2006). Furthermore, Andersson et al. (2007) reported a shift in carotenoid hue towards longer wavelengths in widowbirds as the result of a substantial increase in carotenoid concentration.
&KDSWHU
74
However, our findings suggest that orioles instead gained orange plumage coloration by discrete changes in which pigments are used. Specifically, we found repeated gains of C4-oxygenation from an ancestor that lacked this trait. This result has important implications for the molecular mechanisms of color variation. The primary role of gains and losses in oriole color variation suggests that discrete changes in the presence or absence of a C4-oxygenase gene’s activity, as opposed to continuous changes in its expression, have been responsible for the evolution of carotenoid color variation in orioles. As keto-carotenoids are present within oil droplets in the avian retina (Toomey and McGraw 2009), the neo-localization of a C4-oxygenase already present in the retina to developing feather follicles is a compelling possible pathway to explain the evolution of orange or red coloration.
Why No Red Orioles? Orange plumage in orioles is most likely caused by the deposition of small concentrations of keto-carotenoid compounds into feathers, as in the Guianan Cock-ofthe-rock (Rupicola rupicola, Prum et al. 2012). These same compounds, when found in slightly larger concentrations in the feathers of other Icterid species, cause red coloration (Friedman et al. Ch. 2). This finding raises a perplexing question: why aren’t orange orioles red? We propose two hypotheses below that explore the evolutionary causes of this aspect of color evolution in orioles: one based on selection, and one based on biochemical constraints. First, selection may favor orange coloration over red coloration. For example, females may prefer orange coloration in males, or the costs of producing red coloration
&KDSWHU
75
(e.g., predation) may be so steep as to prevent its evolution (Lande 1980; Andersson 1994; Olson and Owens 1998). Studies in House Finches (C. mexicanus) have suggested that red is a more costly, and thereby more honest color than orange or yellow (Hill 1991; Hill 1996). However, a genetic correlation between trait and preference could maintain orange coloration in orioles as with yellow coloration in C. tristis (Hill and McGraw 2004). Second, it is possible that biochemical mechanisms may strongly constrain color evolution in this group. As in House Finches (McGraw et al. 2006), orioles deposit ketocarotenoids that are most likely produced from -carotene and -crytoxanthin precursors (McGraw 2006; Prager and Andersson 2009). However, orioles do not appear to deposit any keto-carotenoids produced from lutein or zeaxanthin precursors (e.g., astaxanthin, adoradexanthin). This result suggests that the putative enzyme that performs these C4oxygenation reactions may be substrate-specific. If this is the case, orioles may be unable to process lutein and zeaxanthin for use in keto-carotenoid production; orioles may indeed have a paucity of the carotenoids they need to become red despite a wealth of those they do not. Indeed, ornithologists have observed and even collected wild Baltimore Orioles with aberrantly red coloration (Flinn et al. 2007); this suggests that I. galbula has the capacity to produce red coloration but typically lacks the necessary carotenoids in its diet. Because -carotene and -cryptoxanthin are typically found in plasma but not in feathers (McGraw et al. 2006), our feather analyses do not enable us to determine what quantities of these compounds are available to orioles in the wild. If carotene and -cryptoxanthin are limited, orioles may be orange and not red due to a biochemical constraint against producing large quantities of keto-carotenoids.
&KDSWHU
76
History of Carotenoid Pigmentation in Orioles Our ancestral state reconstruction results generally show repeated gains of C4oxygenation from an oriole common ancestor that lacked this ability (Figure 3). Ketocarotenoid pigment types differed among species in clade C, and between clades B and C. These independent gains of keto-carotenoids in two different oriole clades may be the result of either parallel or convergent evolution (sensu strictu; Friedman et al. Ch. 2; Smith and Rausher 2011). If the same C4-oxygenase enzyme performs C4-oxygenation of -carotene, -cryptoxanthin, and echinenone to produce all the keto-carotenoids observed in this study, then our results would support a conclusion of parallel evolution of orange coloration. Our parsimony reconstruction of the composite carotenoid character suggests that there has been one gain of orange coloration in clade C, with a switch from echinenone to 3-hydroxy-echinenone in the common ancestor of I. gularis and I. pustulatus. Ancestral state reconstruction methods are varied, parameter-intensive, and potentially misleading when their assumptions are violated (Wiens et al. 2007). These assumptions include using the correct phylogeny, the correct model of evolution, and the correct scoring of characters as composite or independent (McLennan and Brooks 1993; Omland 1999). While our results appear to be sensitive to reasonable perturbations in several of these parameters, they are robust in their agreement on repeated gains of orange coloration and keto-carotenoid pigmentation: they show one gain of ketocarotenoids in an ancestor of clade C, and one gains of keto-carotenoids in clade B.
&KDSWHU
77
+ CXN, CXA, CXB, ECH
LUT Present in MRCA
+ 3HE, ZXN - ECH - ALL KETO, ZXN
+ ECH
Figure 3: Parsimony reconstruction of ancestral color and pigment character states on the Jacobsen et al. (2010) Icterus phylogeny, with un-sampled taxa pruned, and using a composite of keto-carotenoid characters. The shade of each branch and node represents the color of its inferred ancestral state. Gains of carotenoid pigments are indicated under each branch in black with a “+”, while losses are indicated in red with a “-“. The most recent common ancestor (MRCA) of the New World orioles was reconstructed with yellow plumage and the presence of lutein. Parsimony reconstructions of independent keto-carotenoid characters show equivocal reconstructions for ECH and 3HE, while likelihood reconstructions of these characters show them as repeatedly gained.
&KDSWHU
78
In the years since the publication of an Icterus phylogeny, research in our lab has revealed several topological differences in clade C among phylogenies produced from different loci and species tree methods (Omland et al. 1999; Jacobsen et al. 2010; Jacobsen and Omland 2011). This disagreement may have implications for the ancestral state reconstructions of echinenone and 3-hydroxy-echinenone performed in this study. However, much of this uncertainty is resolved by results from a clade C species tree (Jacobsen and Omland 2011), for which reconstructions using both parsimony and likelihood agree. Future work with a completed multi-locus oriole species tree (which is in development), combined with Bayesian ancestral state reconstruction methods (Ronquist 2004), should be able to account for this phylogenetic uncertainty in a single character mapping. As with any comparative phylogenetic study, there is a need for caution in interpreting our ancestral state reconstruction results. Previous studies have suggested that homoplasy is overwhelmingly common in the evolution of oriole plumage and song (Omland and Lanyon 2000; Price and Lanyon 2002; Hofmann et al. 2006; Friedman et al. Ch. 2). Consequently, erring on the side of homoplasy and uncertainty, as our likelihood reconstructions describe, may be prudent. However, work by Oakley and Cunningham (2000) suggests that phylogenetic comparative methods are generally robust to error and uncertainty in ancestral state reconstruction. To account for phylogenetic uncertainty in our comparative tests, we repeated these tests using a copy of the oriole Z phylogeny with clade C topology changed to reflect that of the most recent nuclear species tree (Jacobsen and Omland 2011). We found that all our correlations remained significant with this composite topology, with the exception of the correlation between keto-carotenoid
&KDSWHU
79
concentration and hue (data not shown). Thus, multiple topologies, multiple reconstruction methods, and multiple comparative methods all indicate that gains and losses of orange coloration in orioles is the result of discrete gains and losses of ketocarotenoids.
Conclusions Together with our research group’s results in a similar study of caciques and meadowlarks (Friedman et al. Ch. 2), we have identified the proximate mechanisms responsible for five independent shifts from yellow towards orange or red plumage coloration. Orioles appear to have evolved keto-carotenoid production through metabolism of -carotene and -cryptoxanthin, but not lutein and zeaxanthin as in meadowlarks and caciques (Andersson et al. 2007; Friedman et al. Ch. 2). Thus, we have described the evolution of several pigmentation mechanisms that are convergent in their use of different pigments, but may still be parallel in their use of C4-oxygenation to produce keto-carotenoids. Future studies using novel molecular techniques may be able to address this distinction between parallel and convergent gains at the genetic level (Pointer et al. 2011; Walsh et al. 2012). However, this study and those that precede it (Friedman et al. 2011; Friedman et al. Ch. 2) have used new methods of analysis and inference to describe a detailed history of carotenoid coloration in the Icteridae. By using the present to infer the past, phylogenetic comparative methods and ancestral state reconstruction are a more useful tool for understanding past visual signals and behaviors than fossils may ever be (Cunningham et al. 1998). Here, we have used such methods to produce the most detailed phylogenetic and mechanistic description of carotenoid color
&KDSWHU
80
evolution in any animal group to date. Thus, these studies together provide the phylogenetic framework necessary for future studies to examine the functional causes underlying the repeated evolution of carotenoid-based coloration.
Acknowledgements Literature Cited Allen, E. S., K. E. Omland. 2003. Novel intron phylogeny supports plumage convergence in orioles (Icterus). Auk 120:961-969. Andersson, M. 1994. Sexual Selection. Princeton University Press, Princeton, New Jersey, USA. Andersson, S., M. Prager, E. I. Anette Johansson. 2007. Carotenoid content and reflectance of yellow and red nuptial plumages in widowbirds (Euplectes spp.). Functional Ecology 21:272-281. Britton, G. 1995. UV/Visible Spectroscopy. In: Carotenoids. Vol. 1B. Spectroscopy (G. Britton, S. Liaaen-Jensen, H. Pfander, eds). Birkhäuser Verlag, Basel, Switzerland. Brush, A. H. 1967. Pigmentation in the Scarlet Tanager, Piranga olivacea. The Condor 69:549-559. Butler, M. A., A. A. King. 2004. Phylogenetic comparative analysis: a modeling approach for adaptive evolution. The American Naturalist 164:683-695. Clements, J. F., 2007. The Clements Checklist of Birds of the World. Comstock Publishing Associates, Ithaca, NY. Cunningham, C. W., K. E. Omland, T. H. Oakley. 1998. Reconstructing ancestral character states: a critical reappraisal. Trends in Ecology and Evolution 13:361367. Dale, J. 2006. Intraspecific variation in coloration. In: Bird Coloration. II. Function and Evolution (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 36-86.
&KDSWHU
81
Flinn, T., J. Hudon, D. Derbyshire. 2007. Tricks exotic shrubs do: when Baltimore Orioles stop being orange. Birding 5:62-68. Fox, D. L., A. A. Wolfson, and J. W. McBeth. 1969. Metabolism of -carotene in the American flamingo, Phoenicopterus ruber. Comparative Biochemistry and Physiology 29:1223-1229. Freckleton, R. P. 2009. The seven deadly sins of comparative analysis. Journal of Evolutionary Biology 22:1367-1375. Friedman, N. R., C. M. Hofmann, B. Kondo, K. E. Omland. 2009. Correlated evolution of migration and sexual dichromatism in the New World Orioles (Icterus). Evolution 63:3269-3274. Friedman, N. R., L. M. Kiere, K. E. Omland. 2011. Convergent gains of red carotenoidbased coloration in the New World blackbirds. Auk 126:678-687. Grafen, A. 1989. The phylogenetic regression. Philosophical Transactions of the Royal Society of London B. 326:119-157. Hill, G. E. 1991. Plumage coloration is a sexually selected indicator of male quality. Nature 350:337-339. Hill, G. E. 1996. Redness as a measure of the production cost of ornamental coloration. Ethology, Ecology, and Evolution 8:157-175. Hill, G. E., K. J. McGraw. 2004. Correlated changes in male plumage coloration and female mate choice in cardueline finches. Animal Behavior 67:27-35. Hill, G. E., R. Montgomerie, C. Y. Inouye, J. Dale. 1994. Influence of dietary carotenoids on plasma and plumage colour in the house finch: intra- and intersexual variation. Functional Ecology 8:343-350. Hofmann, C. M., T. W. Cronin, K. E. Omland. 2006. Using spectral data to reconstruct evolutionary changes in coloration: carotenoid color evolution in new world orioles. Evolution 60:1680-1691. Hofmann, C. M., K. J. McGraw, T. W. Cronin, K. E. Omland. 2007. Melanin coloration in New World orioles I: carotenoid masking and pigment dichromatism in the Orchard Oriole complex. Journal of Avian Biology 38:163-171. Hudon, J. 1991. Unusual carotenoid use by the Western Tanager (Piranga ludoviciana) and its evolutionary implications. Canadian Journal of Zoology 69:2311-2320.
&KDSWHU
82
Jacobsen, F., N. R. Friedman, K. E. Omland. 2010. Congruence between nuclear and mitochondrial DNA: Combination of multiple nuclear introns resolves a wellsupported phylogeny of New World orioles (Icterus). Molecular Phylogenetics and Evolution 56:419-427. Jacobsen, F., K. E. Omland. 2011. Species tree inference in a recent radiation of orioles (genus: Icterus): multiple markers and methods reveal cytonuclear discordance in the northern oriole group. Molecular Phylogenetics and Evolution 61:460-469. Jaramillo A., P. Burke. 1999. New World Blackbirds: The Icterids. Princeton University Press, Princeton, NJ. Kiere, L. M., C. M. Hofmann, J. J. Price, T. W. Cronin, K. E. Omland. 2009. Discrete evolutionary color changes in caciques suggest different modes of carotenoid evolution between closely related taxa. Journal of Avian Biology 40:605-613. King, A. A., M. A. Butler. 2009. ouch: Ornstein-Uhlenbeck models for phylogenetic comparative hypotheses (R package), http://ouch.r-forge.r-project.org Lande, R. 1980. Sexual dimorphism, sexual selection, and adaptation in polygenic characters. Evolution 34:292-305. Lewis, P. O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. Systematic Biology 50:913-925. Maddison, W. P., D. R. Maddison. 2010. Mesquite: A modular system for evolutionary analysis. Version 2.74. http://mesquiteproject.org McGraw, K. J. 2006. Mechanics of carotenoid-based coloration. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, eds.). Harvard University Press, Cambridge, MA, pp. 177-242. McGraw, K. J., A. J. Gregory. 2004. Carotenoid pigments in male American Goldfinches: what is the optimal biochemical strategy for becoming colourful. Biological Journal of the Linnean Society 83:273-280. McGraw, K. J., G. E. Hill. 2001. Carotenoid access and intraspecific variation in plumage pigmentation in male American Goldfinches (Carduelis tristis) and Northern Cardinals (Cardinalis cardinalis). Functional Ecology 15:732-739.
&KDSWHU
83
McGraw, K. J., M. D. Beebee, G. E. Hill, R. S. Parker. 2003. Lutein-based plumage coloration in songbirds is a consequence of selective pigment incorporation into feathers. Comparative Biochemistry and Physiology Part B 135:689-696. McGraw, K. J., J. Hudon, G. E. Hill. 2005. A simple and inexpensive chemical test for behavioral ecologists to determine the presence of carotenoid pigments in animal tissues. Behavioral Ecology and Sociobiology 57:391-397. McGraw, K. J., P. M. Nolan, O. L. Crino. 2006. Carotenoid accumulation strategies for becoming a colorful House Finch: analyses of plasma and liver pigments in wild moulting birds. Functional Ecology 20:678-688. McLennan, D. A., D. R. Brooks. 1993. The Phylogenetic component of cooperative breeding in perching birds: a commentary. The American Naturalist 141:790-795. Montgomerie, R. 2006. Analyzing Colors. In: Bird Coloration. I. Mechanisms and Measurements (G. E. Hill and K. J. McGraw, Eds.). Harvard University Press, Cambridge, MA, pp. 91–147. Olson, V. A., I. P. F. Owens. 1998. Costly sexual signals: are carotenoid rare, risky, or required? Trends in Ecology and Evolution 13:510-515. Omland, K. E. 1999. The assumptions and challenges of ancestral state reconstruction. Systematic Biology 48:604-611. Omland, K. E., S. M. Lanyon, S. J. Fritz. 1999. A molecular phylogeny of the New World orioles (Icterus): the importance of dense taxon sampling. Molecular Phylogenetics and Evolution 12:224-239. Omland, K. E., S. M. Lanyon. 2000. Recontructing plumage evolution in orioles (Icterus): repeated convergence and reversal in patterns. Evolution 54:2119-2133. Pagel, M. 1994. Detecting correlated evolution on phylogenies: a general method for the comparative analysis of discrete characters. Proceedings: Biological Sciences 255:37-45. Pointer, M. A., M. Prager, S. Andersson, N. I. Mundy. 2011. A novel method for screening a vertebrate transcriptome for gene involved in carotenoid binding and metabolism. Molecular Ecology Resources 12:149-159. Paradis, E., J. Claude, K. Strimmer. 2004. ape: analyses of phylogenetics and evolution in R language. Bioinformatics 20: 289-290.
&KDSWHU
84
Prager, M., S. Andersson. 2009. Differential ability of carotenoid C4-oxygenation in yellow and red bishop species (Euplectes spp.). Comparative Biochemistry and Physiology B Molecular Biology 154:373-380. Prager, M., S. Andersson. 2010. Convergent evolution of red carotenoid coloration in widowbirds and bishops (Euplectes spp.). Evolution 64:3609-3619. Price, J. J., S. M. Lanyon. 2002. Reconstructing the evolution of complex bird song in the oropendolas. Evolution 56:1514-1529. Price, J. J., N. R. Friedman, K. E. Omland. 2007. Song and plumage evolution in the New World orioles (Icterus) show similar lability and convergence in patterns. Evolution 61:850-863. Prum, R. O., A. M. LaFountain, J. Berro, M. C. Stoddard, H. A. Frank. 2012. Molecular diversity, metabolic transformation, and evolution of carotenoid feather pigments in cotingas (Aves: Cotingidae). Journal of Comparative Physiology B. 182:10951116. Reed, R. D., R. Papa, A. Martin, H. M. Hines, B. A. Countermain, C. Pardo-Diaz, C. D. Jiggins, N. L. Chamberlain, M. R. Kronforst, R. Chen, G. Halder, H. F. Nijhout, W. O. McMillan. 2011. optix drives the repeated convergent evolution of butterfly wing pattern mimicry. Science 333:1137-1141. Reudink, M. W., P. P. Marra, P. T. Boag, L. M. Ratcliffe. 2009. Plumage coloration predicts paternity and polygyny in the American redstart. Animal Behavior 77:495-501. Ronquist, F. and J. P. Huelsenbeck. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572-1574. Ronquist, F. 2004. Bayesian inference of character evoluton. Trends in Ecology and Evolution 19:475-481. Rowe, M., K. J. McGraw. 2009. Carotenoids in the seminal fluid of wild birds: interspecific variation in fairy-wrens. The Condor 110:694-700. Smith, S. D., M. D. Rausher. 2011. Gene loss and parallel evolution contribute to species difference in flower color. Molecular Biology and Evolution 28:2799-2810.
&KDSWHU
85
Steiner, C. C., H. Römpler, L. M. Boettger, T. Schönberg, H. E. Hoekstra. 2009. The genetic basis of phenotypic convergence in beach mice: similar pigment patterns but different genes. Molecular Biology and Evolution 26:35-45. Toomey, M. B., K. J. McGraw. 2007. Modified saponification and HPLC methods for analyzing carotenoids from the retina of quail: implications for its use as a nonprimate model species. Investigative Opthalmology & Visual Science 48:3976-3982. Toomey, M. B., K. J. McGraw. 2009. Seasonal, sexual, and quality related variation in retinal carotenoid accumulation in the house finch (Carpodacus mexicanus). Functional Ecology 23:321-329. Walsh, N., J. Dale, K. J. McGraw, M. A. Pointer, N. I. Mundy. 2012. Candidate genes for carotenoid coloration in vertebrates and their expression profiles in the carotenoid-containing plumage and bill of a wild bird. Proceedings of the Royal Society B. 279:58-66. Wiens, J. J., C. A. Kuczynski, W. E. Duellman, T. W. Reeder. 2007. Loss and reevolution of complex life cycles in marsupial frogs: Does ancestral trait reconstruction mislead? Evolution 61:1886-1899.
86
Appendix 1 : Voucher specimen data Taxon Icterus bullockii Icterus bullockii Icterus bullockii Icterus bullockii Icterus bullockii Icterus croconotus strictifrons Icterus croconotus strictifrons Icterus croconotus strictifrons Icterus cucullatus cucullatus Icterus cucullatus cucullatus Icterus cucullatus cucullatus Icterus cucullatus nelsoni Icterus cucullatus nelsoni Icterus cucullatus nelsoni Icterus dominicensis Icterus dominicensis Icterus dominicensis Icterus galbula Icterus galbula Icterus galbula Icterus galbula Icterus galbula Icterus graduacauda audobonii Icterus graduacauda audobonii Icterus graduacauda audobonii Icterus graduacauda audobonii Icterus graduacauda audobonii Icterus graduacauda graduacauda Icterus gularis gularis Icterus gularis tamaulipensis Icterus gularis tamaulipensis Icterus gularis tamaulipensis Icterus mesomelas taczanowski Icterus mesomelas taczanowski Icterus mesomelas taczanowski Icterus nigrogularis Icterus nigrogularis Icterus nigrogularis Icterus pectoralis Icterus pectoralis
Museum DMNH DMNH ANSP ANSP ANSP ANSP ANSP ANSP ANSP ANSP ANSP DMNH DMNH DMNH ANSP ANSP ANSP ANSP DMNH DMNH ANSP ANSP ANSP ANSP DMNH DMNH DMNH ANSP ANSP DMNH DMNH DMNH ANSP ANSP ANSP ANSP ANSP ANSP DMNH DMNH
Specimen # 44399 44405 44723 182969 182970 111815 111818 119411 40817 40818 40824 8306 44217 45130 3458 25105 35479 28095 42697 42706 65963 187398 40795 65695 34844 44410 44412 35486 129106 44609 44614 44615 108092 116521 116523 54428 58655 63340 30821 30831
Locality USA, Kansas, Morton Co. USA, Idaho, Latah Co., Moscow USA, California, San Diego USA, Texas, Tom Green Co. USA, Texas, Tom Green Co. Brazil, Descalvados Brazil, Descalvados Bolivia, Chatarona USA, Texas, Brownsville USA, Texas, Brownsville USA, Texas, Brownsville USA, Arizona, Fort Lowell USA, Arizona, Pima Co., Tuscon USA, Arizona, Pima Co., Tuscon Dominican Republic, San Domingo Dominican Republic, San Domingo Dominican Republic, San Domingo USA, Iowa, Winnebago Co. USA, Maine, Kennebec Co. USA, Minnesota, Fillmore Co. USA, Delaware, Rehoboth USA, New Jersey, Salem Co. USA, Texas, Bellville USA, Texas, Bellville Mexico, Guerrero USA, Texas, Webb Co., Laredo Mexico, Nuevo Leon, Monterrey Unknown Mexico, Oaxaca, Chivela Mexico, Tamaulipas Mexico, Tamaulipas Mexico, Veracruz Peru, Chagúal Peru, La Laja Peru, Palamba Unknown Venezuela, Buette Triste Colombia, Santa Marta Mexico, Oaxaca, San Gabriel Mexico, Guerrero, Tres Palos
87
Icterus pectoralis Icterus prosthemelas Icterus prosthemelas Icterus prosthemelas Icterus pustulatus pustulatus Icterus pustulatus pustulatus Icterus pustulatus pustulatus
DMNH ANSP ANSP ANSP DMNH DMNH DMNH
30839 76837 85971 91033 17246 27410 27411
Mexico, Guerrero, Joluchuca Nicaragua, Prinzapolke Panama, Changuinola Honduras, Lancetilla Mexico, Colima, Colima Mexico, Guerrero Mexico, Guerrero
&RQFOXVLRQV
Summary and Overall Conclusions
The many elaborate traits that animals display have long been an inviting challenge to biologists eager to understand the workings of the natural world. In the three studies described above, my goal was to examine how differences in elaborate traits arise among species. Many previous studies have focused on the function of sexually selected traits in a single species. In contrast, I used a phylogenetic comparative approach to explore how these traits vary among species. Hence, I utilized the New World blackbirds as a model clade to study the evolution of elaborate plumage coloration. The New World blackbirds are an ideal group in which to study the evolution of coloration for three reasons. First, they are speciose and diverse: there are 101 recognized Icterid species and they vary considerably with respect to their breeding biology (Jaramillo and Burke 1999; Clements 2007). Second, much is known about the blackbirds’ phylogenetic relationships both within and among their four major clades; this provides a framework for comparative studies (Johnson and Lanyon 1999; Lanyon and Omland 1999; Omland et al. 1999; Price and Lanyon 2002; Barker et al. 2008; Price et al. 2009). Lastly, many previous studies have reconstructed the history of sexually selected traits such as song and plumage in this group, providing a model clade advantage analogous to that found in a model organism (Johnson and Lanyon 2000; Omland and Lanyon 2000; Price and Lanyon 2004; Price et al. 2007). A casual glance at blackbird coloration allows a key observation: red coloration is rare in blackbirds, but is seen across three of the four major clades. Homoplasy is a simple explanation for this pattern, but belies the complexity involved in determining
88
&RQFOXVLRQV
what evolutionary changes were responsible for producing this variation among extant taxa. While there are several distinct means of asking questions about a visual signal (Tinbergen 1963), in my project I have approached this problem in two ways. First, I have reconstructed the evolutionary history of carotenoid-based coloration. Second, I have examined the proximate mechanisms underlying the differences in carotenoid-based coloration seen among extant blackbird taxa.
Convergent gains of red carotenoid-based coloration in the New World blackbirds In chapter one, I examined the history of carotenoid-based color evolution across the Icteridae by reconstructing ancestral states. I used reflectance spectrometry to measure coloration across 114 taxa and then delineated discrete character states based on their spectral location, a measure of carotenoid hue. I found that the common ancestor of the New World blackbirds most likely exhibited yellow carotenoid-based plumage. I also found six repeated gains of red coloration across the Icterids. This direction of color evolution was consistent: we observed no losses of red. Furthermore, this result was robust; to conclude repeated losses of red would require assuming a highly biased and poorly supported likelihood model. I also found no evidence that orange coloration has evolved in blackbirds other than the New World orioles, and no evidence that the New World orioles have ever evolved red coloration. This chapter describes a trend towards red coloration across the blackbirds. Combined with similar observations by other researchers (Hill and McGraw 2004; Prager and Andersson 2010), our findings support a general trend towards the evolution of red carotenoid-based coloration among songbirds. Such a trend might be the result of one of
89
&RQFOXVLRQV
three macroevolutionary process: 1) an intrinsic bias toward red coloration as proposed by Hill (1996); 2) slow evolutionary rate, combined with an ancestral state of yellow carotenoid-based coloration (i.e., phyletic inertia); 3) a character-associated diversification rate in which red species speciate slower or go extinct faster than yellow species. Research in the Omland lab is ongoing to test which of these hypotheses can be supported or rejected.
Convergence and parallelism in the evolution of red carotenoid pigmentation in caciques and meadowlarks In chapter two, I examined the proximate basis for repeated gains of red carotenoid-based coloration in the meadowlarks and allies and in the caciques and oropendolas. To accomplish this aim, I extracted pigments chemically and separated them using HPLC. Using the absorbance spectrum and retention times of external standards, I was able to identify carotenoid compounds present in each feather; using the area under the absorbance curve I was able to quantify each compound’s concentration. I then mapped their presence or absence onto the Icterid phylogeny and used phylogenetic comparative methods to test for relationships between pigments and the colors they produce. My HPLC results indicated that meadowlarks and caciques use different pigments to color their feathers red. Ancestral state reconstruction of these carotenoid pigments suggested that the common ancestor of the blackbirds most likely used lutein to color its feathers. Furthermore, red coloration has evolved twice in the caciques (Kiere et al. 2009), and my results indicate that it has evolved using the same set of carotenoid pigments.
90
&RQFOXVLRQV
These results suggest that meadowlarks and caciques have evolved red coloration using different mechanisms, an example of convergent evolution (sensu strictu; see Gould 2002). In contrast, the caciques have evolved red coloration twice using similar mechanisms, an example of parallel evolution at the biochemical level. I conclude from these results that red coloration evolved in meadowlarks and caciques using different carotenoid metabolic pathways, but evolved twice in caciques using the same metabolic pathway. Lastly, we found astonishingly high concentrations of modified yellow carotenoids (canary xanthophylls A and B) in red cacique rumps. These pigments require metabolic modification prior to deposition that has been hypothesized to be costly (Hill 1996), but should not contribute to these birds’ hue as their absorbance is masked by red keto-carotenoids. Why, then, are they deposited at all? If they are costly, then what is the use of a costly signal that is never sent?
History and mechanisms of carotenoid plumage evolution in the New World orioles (Icterus) In chapter three, I examined the proximate basis for repeated gains of orange coloration in the New World orioles. My aim was to test whether continuous changes in pigment concentration or discrete changes of pigment presence explained variation in carotenoid hue among species in this group. As in chapter two, I extracted carotenoid pigments chemically and separated them using HPLC. After identifying and quantifying carotenoid compounds present in the feathers, I mapped their presence or absence on the oriole phylogeny (Jacobsen et al. 2010). These ancestral state reconstructions suggested that orange coloration had been gained twice in orioles: once in the ancestor of clade C,
91
&RQFOXVLRQV
and once in the troupials. These independent gains likely occurred using different sets of keto-carotenoids. Each time orange coloration was gained, keto-carotenoids were gained concurrently. However, variation in keto-carotenoid concentrations did not appear to explain variation in carotenoid hue among orange species. To test for correlations between pigmentation and hue while correcting for phylogeny, I used the phylogenetic comparative methods PGLS and Pagel’s discrete test (Grafen 1989; Pagel 1994). I found that both the presence of keto-carotenoids and their concentration could explain variation in carotenoid hue among oriole species. However, much of the effect of concentration on hue appeared to be driven by the presence of keto-carotenoids rather than their particular concentration. While previous studies have demonstrated that carotenoid-based plumage coloration varies continuously in orioles (Hofmann et al. 2006), my results suggest that carotenoid pigments contribute to this variation primarily through discrete gains and losses of the ability to produce keto-carotenoids. These results are similar to those described in meadowlarks and caciques in chapter two, as well as in widowbirds and bishops (Prager and Andersson 2010). While changes in oriole pigmentation appear to be discrete (Kiere et al. 2009), this opens a new question: why then do orioles produce orange coloration if they are capable of performing the carotenoid metabolism necessary to produce red?
Overall Summary and Future Questions Comparative studies are essential in the study of evolution, as they allow researchers to examine the function of traits not only in the context of the present, but also in the context of their origin (Brooks and McLennan 1991). In reconstructing the
92
&RQFOXVLRQV
history of elaborate plumage coloration and its proximate bases, these chapters provide a foundation for future research into the evolution of sexually selected traits using the New World blackbirds as a model clade. The repeated gains of red and orange coloration observed throughout this study seem to have been caused by discrete changes in carotenoid metabolic ability leading to major effects in plumage coloration. Many of these gains appear to have occurred using similar metabolic pathways. This pattern may be the result of biochemical constraints on carotenoid pigmentation that make keto-carotenoids the only expedient avenue to producing carotenoid-based red coloration. Carotenoid C4-oxygenation produces red coloration by increasing the length of a carotenoid’s conjugated double bond, and thus lowering the energy of the photons it absorbs (Britton 1995). Other metabolic modifications of the carotenoid ring, such as methylation and hydroxylation are possible, but do not appear to have as strong an effect (Prum et al. 2012). Thus, carotenoid pigmentation mechanisms may only be able to coarsely respond to selection on coloration through metabolic modification. Through these chapters, I have documented the repeated evolutions of similar elaborate traits by similar or different biochemical mechanisms. However, the molecular and genetic basis of these pigmentation mechanisms remains unknown. This is likely the case for most other elaborate traits that result from sexual selection. Future studies of elaborate traits should focus not only on their function, but also how they vary among species. Furthermore, future studies should examine the mechanisms that produce elaborate traits, and consider how they may constrain both variation and rate during evolution.
93
&RQFOXVLRQV
References
Barker, F. K., A. J. Vandergon, S. M. Lanyon. 2008. Assessment of species limits among Yellow-Breasted Meadowlarks (Sturnella spp.) using mitochondrial and sex-linked markers. Auk 125:869-879. Brooks, D. R., D. A. McLennan. 1991. Phylogeny, ecology, and behavior: a research program in comparative biology. University of Chicago Press, Chicago, IL. Clements, J. F. 2007. The Clements checklist of birds of the world, 6th edition. Cornell University Press, Ithaca, NY. Gould, S. J. 2002. The structure of evolutionary theory. Harvard University Press, Cambridge, MA. Grafen, A. 1989. The phylogenetic regression. Phil. Trans. R. Soc. Lond. B 326:119-157. Hill, G. E. 1996. Redness as a measure of the production cost of ornamental coloration. Ethology, Ecology, and Evolution 8:157-175. Hill, G. E., K.J. McGraw. 2004. Correlated changes in male plumage coloration and female mate choice in cardueline finches. Animal Behavior 67:27-35. Jacobsen, F., N. R. Friedman, K. E. Omland. 2010. Congruence between nuclear and mitochondrial DNA: Combination of multiple nuclear introns resolves a wellsupported phylogeny of New World orioles (Icterus). Molecular Phylogenetics and Evolution 56:419-427. Jaramillo, A., and P. Burke. 1999. New World Blackbirds: The Icterids. Princeton University Press, Princeton, N. J. Johnson K. P., Lanyon S. M. 1999. Molecular systematics of the grackles and allies, and the effect of additional sequence (cyt B and ND2). Auk 116:759-768. Johnson, K. P., and S. M. Lanyon. 2000. Evolutionary changes in color patches of blackbirds are associated with marsh nesting. Behavioral Ecology 11:515-519. Kiere, L. M., C. M. Hofmann, J.J. Price, T.W. Cronin, K.E. Omland. 2009. Discrete evolutionary color changes in caciques suggest different modes of carotenoid evolution between closely related taxa. Journal of Avian Biology 40:605-613.
94
&RQFOXVLRQV
Lanyon, S. M., K.E. Omland. 1999. A molecular phylogeny of the blackbirds (Icteridae): five lineages revelaed by cytochrome-b sequence data. Auk 116:629-639. Omland, K. E., S. M. Lanyon, and S. J. Fritz. 1999. A molecular phylogeny of the New World orioles (Icterus): The importance of dense taxon sampling. Molecular Phylogenetics and Evolution 12:224-239. Omland, K. E., S.M. Lanyon. 2000. Recontructing plumage evolution in orioles (Icterus): repeated convergence and reversal in patterns. Evolution 54:2119-2133. Pagel, M. 1994. Detecting correlated evolution on phylogenies: a general method for the comparative analysis of discrete characters. Proceedings: Biological Sciences 255:37-45. Prager, M., S. Andersson. 2010. Convergent evolution of red carotenoid coloration in Widowbirds and Bishops (Euplectes spp.). Evolution 64:3609-3619. Price, J. J., and S. M. Lanyon. 2002. A robust phylogeny of the oropendolas: polyphyly revealed by mitochondrial sequence data. Auk 119:335-348. Price, J. J., S.M. Lanyon. 2004. Patterns of song evolution and sexual selection in the oropendolas and caciques. Behavioral Ecology 15:485-497. Price, J. J., N.R. Friedman, K.E. Omland. 2007. Song and plumage Evolution in the New World orioles (Icterus) show similar lability and convergence in patterns. Evolution 61:850-863. Price, J. J., S. M. Lanyon and K. E. Omland. 2009. Losses of female song with changes from tropical to temperate breeding in the New World blackbirds. Proc. R. Soc. Lond. B 276:1971-1980.
95
