Antioxidant machinery differs between melanic and light nestlings of two polymorphic raptors.
ABSTRACT Colour polymorphism results from the expression of multiallelic genes generating phenotypes with very distinctive colourations. Most colour polymorphisms are due to differences in the type or amount of melanins present in each morph, which also differ in several behavioural, morphometric and physiological attributes. Melanin-based colour morphs could also differ in the levels of glutathione (GSH), a key intracellular antioxidant, because of the role of this molecule in melanogenesis. As GSH inhibits the synthesis of eumelanin (i.e. the darkest melanin form), individuals of darker morphs are expected to have lower GSH levels than those of lighter morphs. We tested this prediction in nestlings of two polymorphic raptors, the booted eagle Hieraaetus pennatus and the Eleonora's falcon Falco eleonorae, both of which occur in two morphs differing in the extent of eumelanic plumage. As expected, melanic booted eagle nestlings had lower blood GSH levels than light morph eagle nestlings. In the Eleonora's falcon, however, melanic nestlings only had lower GSH levels after controlling for the levels of other antioxidants. We also found that melanic female eagle nestlings had higher levels of antioxidants other than GSH and were in better body condition than light female eagle nestlings. These findings suggest an adaptive response of melanic nestlings to compensate for reduced GSH levels. Nevertheless, these associations were not found in falcons, indicating species-specific particularities in antioxidant machinery. Our results are consistent with previous work revealing the importance of GSH on the expression of melanic characters that show continuous variation, and suggest that this pathway also applies to discrete colour morphs. We suggest that the need to maintain low GSH levels for eumelanogenesis in dark morph individuals may represent a physiological constraint that helps regulate the evolution and maintenance of polymorphisms.
- SourceAvailable from: Suzanne M Gray[show abstract] [hide abstract]
ABSTRACT: Here, we review the recently burgeoning literature on color polymorphisms, seeking to integrate studies of the maintenance of genetic variation and the evolution of reproductive isolation. Our survey reveals that several mechanisms, some operating between populations and others within them, can contribute to both color polymorphism persistence and speciation. As expected, divergent selection clearly can couple with gene flow to maintain color polymorphism and mediate speciation. More surprisingly, recent evidence suggests that diverse forms of within-population sexual selection can generate negative frequency dependence and initiate reproductive isolation. These findings deserve additional study, particularly concerning the roles of heterogeneous visual environments and correlational selection. Finally, comparative studies and more comprehensive approaches are required to elucidate when color polymorphism evolves, persists, or leads to speciation.Trends in Ecology & Evolution 03/2007; 22(2):71-9. · 15.39 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The hypothesis that ornaments can honestly signal quality only if their expression is condition-dependent has dominated the study of the evolution and function of colour traits. Much less interest has been devoted to the adaptive function of colour traits for which the expression is not, or is to a low extent, sensitive to body condition and the environment in which individuals live. The aim of the present paper is to review the current theoretical and empirical knowledge of the evolution, maintenance and adaptive function of colour plumage traits for which the expression is mainly under genetic control. The finding that in many bird species the inheritance of colour morphs follows the laws of Mendel indicates that genetic colour polymorphism is frequent. Polymorphism may have evolved or be maintained because each colour morph facilitates the exploitation of alternative ecological niches as suggested by the observation that individuals are not randomly distributed among habitats with respect to coloration. Consistent with the hypothesis that different colour morphs are linked to alternative strategies is the finding that in a majority of species polymorphism is associated with reproductive parameters, and behavioural, life-history and physiological traits. Experimental studies showed that such covariations can have a genetic basis. These observations suggest that colour polymorphism has an adaptive function. Aviary and field experiments demonstrated that colour polymorphism is used as a criterion in mate-choice decisions and dominance interactions confirming the claim that conspecifics assess each other's colour morphs. The factors favouring the evolution and maintenance of genetic variation in coloration are reviewed, but empirical data are virtually lacking to assess their importance. Although current theory predicts that only condition-dependent traits can signal quality, the present review shows that genetically inherited morphs can reveal the same qualities. The study of genetic colour polymorphism will provide important and original insights on the adaptive function of conspicuous traits.Biological Reviews 12/2004; 79(4):815-48. · 10.26 Impact Factor
- Animal Behaviour 01/2008; 75:1351-1358. · 3.07 Impact Factor
Antioxidant Machinery Differs between Melanic and
Light Nestlings of Two Polymorphic Raptors
Ismael Galva ´n1*, Laura Gangoso2, Juan M. Grande3, Juan J. Negro1, Airam Rodrı ´guez1, Jordi Figuerola4,
1Department of Evolutionary Ecology, Estacio ´n Biolo ´gica de Don ˜ana (CSIC), Sevilla, Spain, 2Department of Ecology and Evolution, Biophore, University of Lausanne,
Lausanne, Switzerland, 3Department of Biology, University of Saskatchewan, Saskatoon, Canada, 4Department of Wetland Ecology, Estacio ´n Biolo ´gica de Don ˜ana (CSIC),
Sevilla, Spain, 5Ecology Unit, Instituto de Investigacio ´n en Recursos Cinege ´ticos, IREC (CSIC, UCLM, JCCM), Ciudad Real, Spain
Colour polymorphism results from the expression of multiallelic genes generating phenotypes with very distinctive
colourations. Most colour polymorphisms are due to differences in the type or amount of melanins present in each morph,
which also differ in several behavioural, morphometric and physiological attributes. Melanin-based colour morphs could
also differ in the levels of glutathione (GSH), a key intracellular antioxidant, because of the role of this molecule in
melanogenesis. As GSH inhibits the synthesis of eumelanin (i.e. the darkest melanin form), individuals of darker morphs are
expected to have lower GSH levels than those of lighter morphs. We tested this prediction in nestlings of two polymorphic
raptors, the booted eagle Hieraaetus pennatus and the Eleonora’s falcon Falco eleonorae, both of which occur in two morphs
differing in the extent of eumelanic plumage. As expected, melanic booted eagle nestlings had lower blood GSH levels than
light morph eagle nestlings. In the Eleonora’s falcon, however, melanic nestlings only had lower GSH levels after controlling
for the levels of other antioxidants. We also found that melanic female eagle nestlings had higher levels of antioxidants
other than GSH and were in better body condition than light female eagle nestlings. These findings suggest an adaptive
response of melanic nestlings to compensate for reduced GSH levels. Nevertheless, these associations were not found in
falcons, indicating species-specific particularities in antioxidant machinery. Our results are consistent with previous work
revealing the importance of GSH on the expression of melanic characters that show continuous variation, and suggest that
this pathway also applies to discrete colour morphs. We suggest that the need to maintain low GSH levels for
eumelanogenesis in dark morph individuals may represent a physiological constraint that helps regulate the evolution and
maintenance of polymorphisms.
Citation: Galva ´n I, Gangoso L, Grande JM, Negro JJ, Rodrı ´guez A, et al. (2010) Antioxidant Machinery Differs between Melanic and Light Nestlings of Two
Polymorphic Raptors. PLoS ONE 5(10): e13369. doi:10.1371/journal.pone.0013369
Editor: Kevin McGraw, Arizona State University, United States of America
Received July 2, 2010; Accepted September 16, 2010; Published October 14, 2010
Copyright: ? 2010 Galvan et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Financial support was obtained from the Spanish Ministry of Science and Innovation (project reference: CGL2009-10883-C02-02), from the Junta de
Castilla-La Mancha (project reference: PII1I09-0271-5037) and from Obra Social de la Caja de Canarias (Spain). J.J.N. was supported by Research Project CGL2006-
07481 of the Spanish Ministry of Science and Technology. I.G. benefited from a postdoctoral contract of the CSIC JAE-Doc program, A.R. from an I3P predoctoral
grant, and J.M.G. and L.G. from postdoctoral fellowships from the Spanish Ministry of Science and Innovation and from the Isabel Marı ´a Lo ´pez Memorial
Scholarship (J.M.G.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
Genetic polymorphism can lead to the production of strikingly
different phenotypes when the expression of a single gene depends
upon the alleles present. This phenomenon is known as colour
polymorphism when the main phenotypic difference is the colour
of the plumage, pelage or skin, and has traditionally been defined
as the existence of distinct, genetically determined colour morphs
within a single interbreeding population, in such a proportion that
the rarest form cannot be solely maintained by recurrent mutation
. The mechanisms responsible for the evolution and mainte-
nance of colour polymorphisms in wild populations of animals
have been the subject of intense research, especially in fish ,
reptiles  and birds .
Although the expression of colour polymorphisms depends
weakly on environmental factors such as temperature or food
[5; but see 6], their evolution and maintenance is determined by
the action of non-genetic agents through natural selection [7,8]. In
birds for example, the maintenance of colour polymorphisms may
be due to differential hunting success of colour morphs due to
variation in light conditions . However, selection does not
necessarily have to act directly on the colour trait to generate
differences in the adaptive value of each morph, as there are
several physiological, morphometric and behavioural traits
associated with each colour morph upon which natural selection
could act as well [4,10]. For example, if individuals of one colour
morph have higher competitive ability or immunocompetence
than those of the other morph, this would lead to viability selection
against the latter [11,12]. Consequently, pleiotropy in the genes
responsible for colour expression would generate covariation
between colour polymorphism and a number of individual
An additional and, as far as we know, unexplored association
between colour polymorphism and individual attributes can arise
when the characteristics of these attributes depend on the
biochemical basis of colour production, and not on pleiotropic
PLoS ONE | www.plosone.org1October 2010 | Volume 5 | Issue 10 | e13369
effects of the genes related to colouration [e.g. 10,14]. Most colour
polymorphisms are directly due to changes in the proportion of
areas covered by melanins in integumentary structures such as
feathers, or in the type of melanin these structures contain [4,13].
Therefore, the mechanisms that control the production of these
pigments may be responsible for the causal relationship between
colour polymorphism and some physiological attributes associated
with melanogenesis . Melanin pigments occur in two forms,
with eumelanin being responsible for darker colours such as blacks
and grey and pheomelanin responsible for lighter colours such as
brown, red and orange .
Glutathione (GSH) is a tripeptide thiol found in virtually all
animal cells that functions as the main physiological reservoir of
cysteine  and as the most important intracellular antioxidant
[18,19]. GSH is also intrinsically involved in the process of
melanin synthesis by melanocytes [20–22]. GSH influences
melanogenesis by directly inhibiting the action of tyrosinase (the
enzyme catalysing the first step of the process), by combating free
radicals that stimulate the action of tyrosinase, and by increasing
the ratio cysteine to dopaquinone . Consequently, GSH levels
determine the direction of melanogenesis: high tyrosinase activity
and a low ratio of cysteine to dopaquinone lead to the production
of eumelanin, while low tyrosinase activity and a high ratio of
cysteine to dopaquinone lead to the production of pheomelanin, or
even an absence of melanin synthesis [21,23]. Therefore, low
levels of GSH lead to eumelanogenesis while high levels lead to
pheomelanogenesis, such that eumelanin production proceeds by
default when antioxidant levels are low . Since the vast
majority of polymorphisms are expressed through variability in the
content of eumelanin and pheomelanin in the integument of the
individuals of the different morphs [e.g. 24–29], and eumelanin
production occurs when GSH levels are low [20,23], individuals
with large eumelanin-based traits or a greater overall proportion of
eumelanic integument may present a lower antioxidant capacity as
a consequence of this expression mechanism [20,21].
We performed the first correlative test of the hypothesis that
individuals of different melanin-based colour morphs differ in the
levels of intracellular GSH during melanogensis, and consequent-
ly, in their antioxidant capacity. The study involved two species of
birds of prey, the booted eagle Hieraaetus pennatus and the
Eleonora’s falcon Falco eleonorae. Birds of prey are ideal subjects
for such studies as they are the group of birds in which colour
polymorphisms are most common [30,31]. Both the booted eagle
and the Eleonora’s falcon occur in two distinct morphs: the typical
pale-coloured or ‘light’ morph and the less common dark or
‘melanic’ morph [sensu 13]. Adults and juveniles can be easily
assigned to either morph based on plumage colouration [32–35],
with melanic individuals having a greater amount of eumelanic
feathers (Fig. 1). The plumage morphs of the Eleonora’s falcon are
inherited independently of gender in a Mendelian fashion, with
melanic being the dominant allele . The genetic basis of the
booted eagle polymorphism is unknown. In both species, the
melanic morph is less common than the light morph, as is the case
in other polymorphic raptors [33,34, own unpublished data].
However, the species are phylogenetically distinct  and have
very different life histories. Booted eagles, like most diurnal
raptors, nest solitarily each spring and their breeding success
depends upon the productivity of their territories . In contrast,
Eleonora’s falcons breed colonially at the end of summer and time
their reproduction to coincide with the southward migration of
their avian prey, which can result in pronounced variation in food
availability for their young [38, authors pers. obs.]. We chose to
study both species to test for broad relationships between life-
history features and measures of antioxidant machinery.
We tested the hypothesis by obtaining blood samples from wild
nestlings of both species. We used nestlings because they have
already started to show morph differences at that age, and because
adults are difficult to catch. Furthermore, by using nestlings we
avoided several confounding variables such as differences in
experience between individuals, and reduced differences in age as
compared to adults. We predicted that, in both species, nestlings of
the melanic morph would have lower GSH levels than those of the
light morph. Given the antioxidant properties of GSH [18,19], we
also predicted that nestlings from the melanic morph should
maintain higher levels of alternative antioxidants than those of the
lighter morph, which would protect them from oxidative damage
incurred by their low GSH levels . Therefore, in addition to
comparing GSH levels between morphs, we also compared the
concentration of alternative antioxidants, as estimated by the levels
of uric acid and total antioxidant capacity of plasma, and also
oxidative damage, as estimated by the level of lipid peroxidation in
Materials and Methods
All fieldwork was conducted with the required authorizations
for capture, ringing and blood sampling of birds from the review
boards of the Regional Government of the Canary Islands
(Approval ID. 262/2009) and the Junta de Andalucı ´a (Approval
We sampled nestling booted eagles from territories monitored
by the Natural Processes Monitoring Team of Don ˜ana Biological
Station in Don ˇana National Park in south west Spain (37u109N,
6u239W). 65 eagle nestlings were bled from 52 nests when they
were 25–35 days old (34 nestlings from 28 nests in 2008 and 31
from 24 nests in 2009). Booted eagle nestlings at this age can easily
be assigned to morphs because the ventral plumage is dark brown
in the melanic morph but pale in the light morph (Fig. 1). We
sampled Eleonora’s falcon nestlings from a population of around
120 breeding pairs located in the Alegranza Islet on the north of
Lanzarote, in the Canary Islands (27u379–29u259N, 13u209–
18u199W) during September 2009. 132 falcon nestlings from 61
nests were bled when they were 25–30 days old (median =27.7
days old; their exact age in days was calculated by using the
formula provided by ), and their colour morph was
determined from the colour pattern of their undertail coverts, as
even at this young age, birds of the melanic morph have dark
brown undertail coverts while those of the light morph have pale
Blood samples (1 ml) were collected from the brachial vein and
stored in a cooler in the field, then later centrifuged in the
laboratory (within 6 hours of collection) and the plasma and red
blood cells stored separately at 280uC until analysis.
Determination of Total Glutathione Levels of
Briefly, the blood pellet was weighed to the nearest 0.0001 g
and thawed, and the red blood cells were drawn up with a pipette
avoiding the pellet surface (i.e. the buffy coat containing white
blood cells). Erythrocytes were immediately diluted (1:10 w/v) and
homogenized in a stock buffer [0.01 M phosphate buffered saline
(PBS) and 0.02 M ethylene diamine tetra acetic acid (EDTA)],
always working on ice to avoid oxidation. Three working solutions
were made up in the same stock buffer as follows: (I) 0.3 mM
nicotinamide adenine dinucleotide phosphate (NADPH), (II)
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org2 October 2010 | Volume 5 | Issue 10 | e13369
6 mM 5,59-Dithiobis(2-nitrobenzoic acid) (DTNB), and (III) 50
Units of glutathione reductase/mL. An aliquot (0.5 mL) of red
blood cell homogenate was vortexed with 0.5 mL of 10%
trichloroacetic acid for 5 seconds thrice within a 15 minute
period. Between each vortexing the samples were placed in a
darkened refrigerator to prevent oxidation. Finally, the mixture as
centrifuged at 1125 g for 15 minutes at 6uC and the supernatant
removed and placed on ice. Subsequent steps were carried out in
an automated spectrophotometer (A25-Autoanalyzer; Biosystems
SA, Barcelona). Solutions 1 and 2 were mixed at a ratio of 7:1
respectively and 160 uL of this mixture was added to 40 uL of the
sample supernatant in a cuvette. After 15 seconds, 20 uL of
solution 3 was added and then the absorbance at 405 nm was read
after 30 and 60 seconds. The change in absorbance between the
two readings was used to determine glutathione concentration in
red blood cells according to a standard curve generated by serial
dilution of glutathione from 1 mM to 0.031 mM. Repeatability of
this technique was previously determined on a sub-sample of
erythrocytes measured twice (r=0.85, n=20, P=0.002). Concen-
tration is presented as mmol glutathione/g of blood pellet.
Determination of Lipid Peroxidation in Erythrocytes
The principle of the test is based on the fact that most tissues
contain a mixture of thiobarbituric acid reactive substances
Figure 1. Booted eagle and Eleonora’s falcon nestlings of different colour morphs. Examples of (A) light-morph and (B) melanic-morph
booted eagle nestlings and of (C) light-morph and (D) melanic-morph Eleonara’s falcon nestlings from the study populations.
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org3 October 2010 | Volume 5 | Issue 10 | e13369
(TBARS), including lipid hydroperoxides and aldehydes, whose
concentrations increase due to oxidative stress. 1 mL of the
homogenate used for the tGSH analysis was mixed with 2 mL of a
solution of 15% trichloroacetic acid, 0.77% hydrochloric acid and
0.375% thiobarbituric acid and with 20 uL of 2% 2,6-di-tert-
butyl-4-methylphenol (BHT) in ethanol in a closed glass tube.
Tubes were then heated for 30 min at 90uC and then cooled in
ice-cold water for 10 min. The mixture was centrifuged at 2025 g
for 15 min and the absorbance of the supernatant was read at
535 nm. The concentration of peroxidised lipids was determined
in reference to a standard curve with 0, 1.25, 2.5 and 5 nmol/mL
of malondialdehyde (MDA) in H2O (i.e. end product of lipid
peroxidation) and presented as nM MDA/g of red blood cell
pellet. Repeatability was previously estimated on a subset of
samples measured three times (r=0.80, n=20, P=0.037).
Determination of Plasma Uric Acid Concentration and
Uric acid concentration of plasma was determined in a Cobas
Integra 400 Automated Analyser using the commercial enzymatic
colorimetric method (Roche Diagnostics S.L., San Cugat del
Valle ´s, Barcelona, Spain). Total Antioxidant Capacity (TAC) of
plasma was determined through a colorimetric assay. This method
is based upon the colour change, adapted from , caused by the
addition of hydrogen peroxide to colourless 2,2V-azinobis(3-
ethylbenzo-thiazoline-6-sulfonate) (ABTS), which oxidizes it into
a characteristic blue-green solution. 5 ml plasma was mixed with
200 ml 0.4 M acetate buffer (pH 5.8) and 30 ml of a solution of
30 mM acetate buffer, 2 mM H2O2and 10 mM ABTS. Absor-
bance was measured every 30 seconds for 10 minutes at 595 nm on
a 1420Multilabel CounterVictor3 (PerkinElmer),and antioxidant
capacity was measured as the rate of change of absorbance. TAC
was estimated as the average of two replicate measurements per
individual. Distilled water was used as a blank and TAC was
calibrated with a standard of Trolox (an hydrosoluble equivalent of
Vitamin E) and expressed in mM Trolox equivalent. The
repeatabiity of these techniques was determined on a subset of
samples measured two times (uric acid concentration: r=0.99,
n=10, P,0.0001; TAC: r=0.89, n=10, P,0.0001).
Nestling eagles and falcons were sexed using DNA extracted
from blood, which was amplified by PCR using the primers 2550F
and 2718R following Fridolfsson and Ellegren .
Generalized Linear Mixed Models (GLMMs) were used to
determine the predictors of total glutathione and lipid peroxida-
tion levels in erythrocytes and TAC and uric acid levels in plasma
of Eleonora’s falcon and booted eagle nestlings. All these variables
were dependent variables in single GLMMs, and morph (light vs.
melanic) and sex were added as fixed factors. Brood size was
added as a covariate, because this may be positively correlated
with oxidative stress levels [43,44]. We also included body mass as
a covariate, as this could affect oxidative stress, with nestlings that
develop more rapidly suffering greater stress . In the case of
Eleonora’s falcons, we included body mass and estimated age as
covariates. In every model, levels of other antioxidants were added
as covariates to understand the pattern of covariation between
them [e.g. 45]. We accounted for the common origin of nestlings
and year (in the case of booted eagles) by adding nest identity and
year as random factors, and the model was fitted using restricted
maximum likelihood (REML).
Unfortunately, some data on body mass and tarsus length of
booted eagles in 2008 was lost during a hardware accident. The
remaining data were thus analysed separately to avoid a decrease
in statistical power when analysing the effect of interest (i.e. colour
morph) with the complete dataset. Hence the results from the
booted eagle models with and without the morphometric data are
presented separately (2008: n=10 for body mass, n=0 for tarsus
length; 2009: n=27 for both body mass and tarsus length).
In all models, we included an interaction between morph and
sex to test whether there are sex-specific associations between
morph and antioxidant levels and oxidative stress. In addition to
the independent variables described above, the models also
included two parameters to control for potential variability during
the laboratory analyses of tGSH and TBARS. First, we included
the weight of the blood pellet before it was homogenized. Second,
the Eleonora’s falcon samples were analysed in three separate
assays because of the larger sample size. The assay in which each
sample was analysed was therefore added as a random factor.
Finally, we analysed variability in nestling body condition using
models where body mass was the dependent variable and tarsus
length was included as a covariate to control for body size .
The other predictor variables were those used in the models
examining levels of antioxidants and oxidative damage.
We removed non-significant terms from each saturated model
using a backward elimination procedure with a P-value of 0.1
being sufficient to eliminate the term. Random factors were always
maintained in the models, and in all cases, the distribution of the
residuals confirmed that the assumption of normality was fulfilled.
Satterthwaite correction was used to approximate the denomina-
tor degrees of freedom.
Total glutathione levels.
in glutathione levels between nestlings of the two colour morphs
(Table 1), with those of the melanic morph having lower levels
than those of the light morph (Fig. 2A). Considering the subsample
of individuals where morphometric data were available, nestling
mass was not significantly related to tGSH when added to the
previous model (F1,29.3=1.93, P=0.175).
In the same model, but including alternative antioxidants as
covariates, only TAC was negatively correlated with tGSH. How-
ever, the trend was non-significant (b=21.961024, F1,50=3.54,
P=0.066) and did not change the significant difference in tGSH
between morphs (F1,50=4.49, P=0.039).
In the model testing for variability in
TAC, the only factor that was significant was uric acid
concentration (b=57.86, F1,51=113.92, P,0.0001), although it
did not differ between morphs (see below). However, when uric acid
concentration was not considered, tGSH remained in the model, as
well as morph, sex (least square mean (LSM) 6 s.e.: males:
1224.10676.74 mmol/g, females: 1399.89680.95 mmol/g), nest
identity, year and the interaction between morph and sex (Table 1).
This interaction was due to melanic females having significantly
higher TAC levels than light females, whereas there was no
difference in TAC levels between males of the two morphs (Fig. 2B).
There was no effect of nestling body mass (F1,10.8=0.12, P=0.732).
Moreover, if tGSH and uric acid levels were not included in the
TAC model as covariates, the interaction between morph and sex
remained marginally significant (interaction: F1,30=3.75, P=0.062;
morph: F1,43=0.50, P=0.484; sex: F1,32.1=1.27, P=0.268;
Fig. 2B). Nest identity also had a significant effect (Z=2.25,
P=0.012), although year did not (Z=0.05, P=0.482).
There was a significant difference
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org 4 October 2010 | Volume 5 | Issue 10 | e13369
Table 1. Results of the models explaining variability in total glutathione (tGSH) levels in erythrocytes and plasma antioxidant levels
(TAC and uric acid) in booted eagle and Eleonora’s falcon nestlings.
Effectb F/ZdfPb F/Z dfPb F/Z dfP
Morph- 4.33 1,53 0.042- 0.01 1,21.10.908----
Sex----- 4.971,16.7 0.040----
Morph x Sex-----5.09 1,13.8 0.040----
tGSH----440.9311.05 1,15.60.0043.64 5.68 1,16.30.030
Morph- 0.00 1,1200.952--------
Morph x Sex-4.741,1020.032--------
Laboratory assay number-0.90-0.183--------
tGSH---- 291.2114.25 1,89.9
Figure 2. Physiological parameters of light and melanic morphs of nestling booted eagles of each sex. (A): Total glutathione (tGSH)
level in pelleted erythrocytes. (B): Plasma total antioxidant capacity (TAC). (C): Lipid peroxidation level of pelleted erythrocytes. (D): Body condition
(mass corrected for body size). Least squares means + standard error are shown.
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org5 October 2010 | Volume 5 | Issue 10 | e13369
The only significant terms in the model for uric acid levels were
TAC (b=0.01, F1,47.8=90.19, P,0.0001) and nest identity
(Z=1.78, P=0.037). When TAC was not included, only tGSH
and nest identity remained in the model (Table 1). Body mass did
not affect uric acid levels (F1,18.4=0.12, P=0.738).
The difference in the levels of oxidative
damage (TBARS) between melanic and light nestlings (LSM 6 s.e.:
26.8962.11 nmol MDA/g) was not significant (F1,48=2.80,
P=0.100). In this model, there was also an effect of sex, though
not of nest identity and year (Table 2). However, the difference in
(F1,8.01=7.23, P=0.027; Fig. 2C) after controlling for body mass
by including it as a covariate in the previous model (b=4.1661023,
F1,11.8=1.78, P=0.208; sex: F1,8.26=19.99, P=0.002; nest
identity: Z=2.33, P=0.010; year: Z=0.68, P=0.249).
Body condition was significantly related to
morph, sex, tarsus length and nest identity (Table 2). However,
there was also a significant morph by sex interaction (Table 2),
such that melanic females were in better condition than the other
three groups (Fig. 2D).
morphs became significant
morphs of the Eleonora’s falcon, but in this species the difference
depended upon the sex of the nestling. The model thus included
sex, morph and their interaction (Table 1). Melanic male nestlings
had higher tGSH levels than light males, though melanic female
tGSH levels also differed between colour
nestlings had lower tGSH levels than light females (Fig. 3A). There
was no difference between tGSH levels of light morph females and
melanic males (post-hoc test: P=0.690), but both had higher levels
than light morph males and melanic females (P=0.030), between
which there was no difference in tGSH level (P=0.590). Other
terms in the model were nestling mass, homogenate mass, nest
identity and laboratory assay number (Table 1).
When variability in all antioxidants was examined by adding
TAC and uric acid levels as covariates, tGSH levels differed
between morphs independently of their sex (F1,72.1=5.16,
P=0.026), with melanic falcons having lower tGSH levels than
light morph falcons (Fig. 3B). The sex and sex x morph interaction
were not significant (both P.0.3). In this species, however, TAC
(F1,71.1=5.61, P=0.020) and uric acid levels (F1,68.6=11.49,
P=0.001) covaried positively with tGSH (b=2.5661024and 0.05,
respectively). Other variables remaining in the model were body
mass (b=3.9561023, F1,72.8=31.23, P,0.0001), homogenate
(Z=2.24, P=0.013) and laboratory assay (Z=0.84, P=0.199).
TAC levels did not differ between
morphs, and the model only included the effects of nestling age
(b=224.10, F1,78.6=3.22, P=0.077) and uric acid (b=32.25,
F1,63.2=4.56, P=0.037; nest identity: Z=1.22, P=0.110). When
uric acid was not considered, the final model included nestling
body mass, tGSH and nest identity (Table 1).
The model for uric acid levels only included the effects of tGSH
(Table 1). The same result was obtained when TAC was not
considered in the model.
Table 2. Results of the models explaining variability in oxidative damage (TBARS) levels of erythrocytes and body condition in
booted eagle and Eleonora’s falcon nestlings.
Effectb F/Z dfPb F/Z dfP
Morph x Sex-----7.46 1,18.5 0.013
Tarsus length----- 15.85 1,21.2
Nest identity-0.23- 0.409- 1.44-0.075
Morph x Sex--------
2 20.025.871,70.8 0.018----
Age0.263.901,710.05211.84 112.37 1,93.1
Tarsus length----7.095.49 1,87.40.021
Laboratory assay number-0.99- 0.160----
tGSH2.218.04 1,83.10.00610.213.05 1,87.80.084
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org6 October 2010 | Volume 5 | Issue 10 | e13369
morphs. The model included nestling age, body mass, TAC levels,
tGSH, homogenate mass, nest identity and laboratory assay
number (Table 2).
The body condition of Eleonora’s falcon
nestlings did not differ between morphs (F1,42.1=0.01, P=0.935),
and only nestling age, sex (LSM 6 s.e.: males: 371.8567.32 g,
females: 401.2467.05 g), tarsus length, tGSH and nest identity
remained in the model (Table 2).
TBARS levels did not differ between
The present study adds to the relatively few reports of
physiological differences between colour morphs [14; see however
review in 4] and, as far as we know, is the first study reporting that
polymorphism covaries with levels of antioxidants and oxidative
Based upon previous findings in other bird species without
colour polymorphism [i.e. 20], we predicted that individuals of the
melanic morph of both species should have lower GSH levels than
individuals of the light morph, and that individuals of the melanic
morph should compensate for the low GSH levels required for
eumelanogenesis by increasing the level of alternative plasmatic
antioxidants. Although our results were consistent with both
hypotheses, there were several important differences between
species and sex, and some key interactions between the two. First,
whereas there were significant differences in GSH levels between
morphs of both sex in the booted eagle, the direction of the
difference depended on sex in the Eleonora’s falcon: melanic
males had higher levels than light males, while melanic females
had lower levels than light females. Second, the correlation
between GSH and alternative antioxidants was negative in eagles
but positive in falcons. Third, female booted eagle nestlings of the
melanic morph had higher total antioxidant capacity than females
of the light morph. Thus, female eagle nestlings of the melanic
morph had lower GSH levels but also higher levels of total
antioxidant capacity than those of the light morph. Moreover,
melanic eagles showed lower mean values of oxidative damage
than light eagles, although the difference was non-significant,
which suggests the existence of a compensatory mechanism
[e.g. 47], at least in this species.
The results showed both sex-specific and species-specific
differences between morphs in their levels of circulating antiox-
idants and body condition. Morph was not related to plasma
antioxidants or body condition in Eleonora’s falcons, despite a
larger sample size, which suggests there is no compensatory
antioxidant mechanism in this raptor species. However, melanic
falcons did not have higher levels of oxidative damage than light
falcons, which suggests that they may have more efficient
antioxidant machinery. Moreover, the lower levels of GSH in
melanic falcons of both sexes were only apparent after controlling
for variation in the levels of other antioxidants, suggesting that this
species possesses a modified or attenuated role for GSH in melanin
Species-specific associations between colour polymorphism and
antioxidant levels may result from pleiotropic effects of different
genes regulating colour production in each species. Probably the
most important gene in this context is the pro-opiomelanocortin
(POMC) gene, which codes for the production of several molecules
with pleiotropic effects [see 10]. These molecules are named
melanocortins, which are peptidic compounds that bind to
melanocortin receptors. Although several melanin-based colour
polymorphisms in wild birds depend on the expression of the
melanocortin-1 receptor gene [MC1R; 8,26,48–51], a number of
other genes coding for receptors of other different melanocortins
are also involved in melanin production . Thus, differences
between species in the number and/or identity of genes that
influence the expression of melanin-based polymorphism may
result in different patterns of covariation between colour and levels
of plasma antioxidants and oxidative damage.
Furthermore, the life history differences between the two species
likely expose them to different environmental factors, which may
in turn influence their levels of oxidative stress. For example,
Eleonora’s falcons breed colonially in late summer while booted
eagles breed as territorial pairs in early spring. Coloniality is
known to be strongly associated with the prevalence of
haematophagous ectoparasites and haematozoan infection ,
and the latter of these has been shown to generate oxidative stress
. Differences in the amount and regularity of food delivered to
the nestlings of each species may also explain the variation
between morphs in levels of plasma antioxidants and oxidative
stress. In booted eagles, the total amount of food nestlings receive
will depend upon the productivity of their territory, but in general
they deliver prey to the brood on a regular basis . In contrast,
falcons prey upon migratory birds. Since the number of migrating
birds fluctuates according to the prevailing wind conditions [38,
authors pers. obs.], the nestlings experience a very erratic food
supply, which likely causes sharp changes in physiological stress
and in uric acid levels [e.g. 54,55], that may unbalance their redox
Figure 3. Total glutathione (tGSH) levels in male and female Eleonora’s falcon nestlings of light and melanic morphs. Least squares
means + standard error from models that (A) did not control for the variability in other antioxidants (total antioxidant capacity and uric acid) and that
(B) included these variables as covariates are shown.
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org7 October 2010 | Volume 5 | Issue 10 | e13369
homeostasis. This may have resulted in different associations of
colour morph with the levels of plasma antioxidants and oxidative
damage in Eleonora’s falcons and booted eagles. Additional factors
such as geographic location, the different breeding periods
(autumn in Eleonora’s falcons vs. spring-summer in booted eagles)
or diet composition (exclusively birds in the falcons and including
other vertebrates in the eagles) may also be important.
Our results show that morphs differing in the eumelanin content
of their plumage also differ in their antioxidant machinery. We
suggest that these differences are a direct consequence of the
central role of GSH in eumelanin production [20–22], rather than
the result of pleiotropic genes affecting both melanization and
antioxidant levels . This would mean that a melanin-based
colour polymorphism may constrain the antioxidant levels of the
individuals that produce the melanic morph, which does not
exclude the possibility that pleiotropic effects on colour are also
Given the importance of oxidative stress in determining life-
history traits [reviewed in 56], there will likely be fitness
consequences for the individuals of each morph. It is worth
mentioning that the melanic morph of the Eleonora’s falcon seems
to be much less often observed in nature [32–34], and the same
tendency is observed in other polymorphic raptors . Indeed,
our results show that melanic booted eagle of any sex, and female
melanic Eleonora’s falcons, have lower GSH levels than
conspecifics of the light morh. It is possible that morph frequencies
are maintained by frequency-dependent selection, so that light
individuals may present fitness disadvantages of the same
magnitude than that represented by low GSH levels in melanic
individuals, resulting in a ratio costs:benefits that is similar in both
morphs [see 57 for an example of positive selection on darker
individuals].However, there is support for fitness disadvantages or
at least exceptionally low observation and capture frequencies in
the more eumelanized morph unrelated to direct natural selection
pressures on colour (e.g. predation or thermoregulation) in other
bird species [58; see however 12] and also in mammals [59,60]
and insects . Some reports indicate that birds of the more
eumelanized morph are more vulnerable to bacterial infections
 and ectoparasites [27; but see 62]. Therefore, perhaps
individuals belonging to the melanic morph pay a fitness cost,
despite the lower oxidative damage that we observed in melanic
nestling booted eagles as compared to light nestlings. The lower
oxidative damage found in melanic booted eagles as compared to
light eagles may occur because of the adaptive response to
compensate for the low GSH levels required for eumelanogenesis
may be physiologically costly, or because this adaptive response
does not always occur. Another possibility is that this adaptive
response can be only performed by individuals of high genotypic
quality, whereas low quality individuals are not able to make the
antioxidant compensation and suffer physiological consequences
after trying to maintain low GSH levels. Future studies should
explore these possibilities.
We thank Ester Ferrero, Francisco Miranda and Olaya Garcı ´a for help
during laboratory work, Jose ´ L. del Valle, Luis Garcı ´a, Rube ´n Rodrı ´guez
and Antonio Martı ´nez (Natural Processes Monitoring Team of Don ˜ana
Biological Station) for help during field work with booted eagles, and Juan
J. Moreno, Juan L. Barroso, Enrique Luque, Pilar Vicent, Gustavo Tejera
and Karim Manan for help during field work with Eleonora’s falcons.
Alexandre Roulin, Ian Stewart and Kevin McGraw made useful comments
on the manuscript.
Conceived and designed the experiments: IG JJN CAA. Performed the
experiments: IG LG JMG JJN AR JF CAA. Analyzed the data: IG JF.
Contributed reagents/materials/analysis tools: IG JF CAA. Wrote the
1. Huxley J (1955) Morphism in birds. Acta Int Congr Ornithol 11: 309–328.
2. Gray SM, McKinnon JS (2007) Linking color polymorphism maintenance and
speciation. Trends Ecol Evol 22: 71–79.
3. Sinervo B, Lively CM (1996) The rock-paper-scissors game and the evolution of
alternative male strategies. Nature 380: 240–243.
4. Roulin A (2004) The evolution, maintenance and adaptive function of genetic
colour polymorphism in birds. Biol Rev 79: 1–34.
5. Buckley PA (1987) Mendelian genes. In: Cooke F, Buckley PA, eds. Avian
Genetics, a Population and Ecological Approach. Orlando: Academic Press. pp
6. Roulin A, Almasi B, Rossi-Pedruzzi A, Ducrest A-L, Wakamatsu K, et al. (2008)
Corticosterone mediates the condition-dependent component of melanin-based
coloration. Anim Behav 75: 1351–1358.
7. Endler JA (1980) Natural selection on color patterns in Poecilia reticulata.
Evolution 34: 76–91.
8. Uy JAC, Moyle RG, Filardi CE, Cheviron ZA (2009) Difference in plumage
color used in species recognition between incipient species is linked to a single
amino acid substitution in the melanocortin-1 receptor. Am Nat 174: 244–254.
9. Galeotti P, Rubolini D, Dunn PO, Fasola M (2003) Colour polymorphism in
birds: causes and functions. J Evol Ecol 16: 635–646.
10. Ducrest A-L, Keller L, Roulin A (2008) Pleiotropy in the melanocortin system,
coloration and behavioural syndromes. Trends Ecol Evol 23: 502–510.
11. Kru ¨ger O (2002) Dissecting common buzzard lifespan and lifetime reproductive
success: the relative importance of food, competition, weather, habitat and
individual attributes. Oecologia 133: 474–482.
12. Brommer JE, Ahola K, Karstinen T (2005) The colour of fitness: plumage
coloration and lifetime reproductive success in the tawny owl. Proc R Soc B 272:
13. True JR (2003) Insect melanism: the molecules matter. Trends Ecol Evol 18:
14. Miles DB, Sinervo B, Hazard LC, Svensson EI, Costa D (2007) Relating
endocrinology, physiology and behaviour using species with alternative mating
strategies. Funct Ecol 21: 653–665.
15. Hooper RE, Tsubaki Y, Siva-Jothy MT (1999) Expression of a costly, plastic
secondary sexual trait is correlated with age and condition in a damselfly with
two male morphs. Physiol Entomol 24: 364–369.
16. Toral GM, Figuerola J, Negro JJ (2008) Multiple ways to become red: pigment
identification in red feathers using spectrometry. Comp Biochem Physiol B 150:
17. Benedetto JP, Ortonne JP, Voulot C, Khatchadourian C, Prota G, et al. (1981)
Role of thiol compounds in mammalian melanin pigmentation. Part I. Reduced
and oxidized Glutathione. J Inv Dermatol 77: 402–405.
18. Anderson ME (1998) Glutathione: an overview of biosynthesis and modulation.
Chem-Biol Interact 111-112: 1–14.
19. Wu G, Fang YZ, Yang S, Lupton JR, Turner ND (2004) Glutathione
metabolism and its implications for health. J Nutr 134: 489–492.
20. Galva ´n I, Alonso-Alvarez C (2008) An intracellular antioxidant determines the
expression of a melanin-based signal in a bird. PLoS ONE 3: e3335.
21. Galva ´n I, Alonso-Alvarez C (2009) The expression of melanin-based plumage is
separately moduated by exogenous oxidative stress and a melanocortin.
Proc R Soc B 276: 3089–3097.
22. Galva ´n I, Solano F (2009) The evolution of eu- and pheomelanic traits may
respond to an economy of pigments related to environmental oxidative stress.
Pigment Cell Melanoma Res 22: 339–342.
23. Ozeki H, Ito S, Wakamatsu, Ishiguro I (1997) Chemical characterization of
pheomelanogenesis starting from dihydroxyphenylalanine or tyrosine and
cysteine. Effects of tyrosinase and cysteine concentrations and reaction time.
Biochim Biophys Acta 1336: 539–548.
24. Preston CR (1980) Differential perch site selection by color morphs of the red-
tailed hawk (Buteo jamaicensis). Auk 97: 782–789.
25. Robbins LS, Nadeau JH, Johnson KR, Kelly MA, Roselli-Rehfuss L, et al.
(1993) Pigmentation phenotypes of variant extension locus alleles result from
point mutations that alter MSH receptor function. Cell 72: 827–834.
26. Baia ˜o PC, Schreiber EA, Parker PG (2007) The genetic basis of the plumage
polymorphism in red-footed boobies (Sula sula): a melanocortin-1 receptor (MC1R)
analysis. J Hered 98: 287–292.
27. Chakarov N, Boerner M, Kru ¨ger O (2008) Fitness in common buzzards at
the cross-point of opposite melanin-parasite interactions. Funct Ecol 22: 1062–
28. Roulin A, Gasparini J, Bize P, Ritschard M, Richner H (2008) Melanin-based
colorations signal strategies to cope with poor and rich environments. Behav
Ecol Sociobiol 62: 507–519.
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org8 October 2010 | Volume 5 | Issue 10 | e13369
29. Pannkuk EL, Siefferman LM, Butts JA (2010) Colour phases of the eastern
screech owl: a comparison of biomechanical variables of body contour feathers.
Funct Ecol 24: 347–353.
30. Roulin A, Wink M (2004) Predator-prey relationships and the evolution of
colour polymorphism: a comparative analysis in diurnal raptors. Biol J Linn Soc
31. Kru ¨ger O (2008) Engines of speciation: a comparative study in birds of prey.
J Evol Biol 21: 861–872.
32. Wink M, Wink C, Ristow D (1978) Biologie des Eleonorenfalken (Falco eleonorae):
2. Zur Vererbung der Gefiederphasen (hell-dunkel). J Ornithol 119: 421–428.
33. Ristow D, Wink C, Wink D, Scharlau W (1998) Colour polymorphism in
Eleonora’s falcon Falco eleonorae. Sandgrouse 20: 56–64.
34. Ristow D, Witte L, Wink M (2000) A characterisation of the homozygous dark
morph of Eleonora’s Falcon. Sandgrouse 22: 118–121.
35. Ferguson-Lees J, Christie DA (2001) Raptors of the World. London: Helm.
36. Lerner HRL, Mindell DP (2005) Phylogeny of eagles, Old World vultures, and
other Accipitridae based on nuclear and mitochondrial DNA. Mol Phylogenet
Evol 37: 327–346.
37. Martı ´nez JE, Pagan I, Calvo JF (2006) Interannual variations of reproductive
parameters in a booted eagle (Hieraaetus pennatus) population: the influence of
density and laying date. J Ornithol 147: 612–617.
38. Walter H (1979) Eleonora’s falcon: adaptations to prey and habitat in a social
raptor. Chicago: University of Chicago Press.
39. Ristow D, Wink M (2004) Seasonal variation in sex ratio of nestling Eleonora’s
falcons. J Raptor Res 38: 320–325.
40. Ristow D, Witte L, Wink M (2004) Sex determination of nestlings in Eleonora’s
Falcon Falco eleonorae: plumage characteristics and molecular sexing. In:
Chancellor RD, Meyburg BU, eds. Raptors Worldwide: proceedings of the VI
World Conference on Birds of Prey and Owls. Budapest: World Working Group
on Birds of Prey and Owls, MME/BirdLife Hungary. pp 459–466.
41. Erel O (2004) A novel automated direct measurement method for total
antioxidant capacity using a new generation, more stable ABTS radical cation.
Clin Biochem 37: 277–285.
42. Fridolfsson A-K, Ellegren H (1999) A simple and universal method for molecular
sexing of non-ratite birds. J Avian Biol 30: 116–121.
43. Costantini D, Casagrande S, de Filippis S, Brambilla G, Fanfani A, et al. (2006)
Correlates of oxidative stress in wild kestrel nestlings (Falco tinnunculus). J Comp
Physiol B 176: 329–337.
44. Alonso-Alvarez C, Bertrand S, Faivre B, Sorci G (2007) Increased susceptibility
to oxidative damage as a cost of accelerated somatic growth in zebra finches.
Funct Ecol 21: 873–879.
45. Cohen AA, McGraw KJ (2009) No simple measures for antioxidant status in
birds: complexity in inter- and intraspecific correlations among circulating
antioxidant types. Funct Ecol 23: 310–320.
46. Garcı ´a-Berthou E (2001) On the misuse of residuals in ecology: testing regression
residuals vs. the analysis of covariance. J Anim Ecol 70: 708–711.
47. Dimova EG, Bryant PE, Chankova SG (2008) ‘Adaptive response’—some
underlying mechanisms and open questions. Genet Mol Biol 31: 396–408.
48. Nachman MW, Hoekstra HA, D’Agostino SL (2003) The genetic basis of
adaptive melanism in pocket mice. Proc Natl Acad Sci USA 100: 5268–5273.
49. Doucet SM, Shawkey MD, Rathburn MK, Mays HL, Jr., Montgomerie R
(2004) Concordant evolution of plumage colour, feather microstructure and a
melanocortin receptor gene between mainland and island populations of a fairy-
wren. Proc R Soc B 271: 1663–1670.
50. Mundy NI (2006) Genetic basis of color variation in wild birds. In: Hill GE,
McGraw KJ, eds. Bird Coloration, Vol. I: Mechanisms and Measurements
Cambridge: Harvard University Press. pp 469–506.
51. Hubbard JK, Uy JAC, Hauber ME, Hoekstra HE, Safran RJ (2010) Vertebrate
pigmentation: from underlying genes to adaptive function. Trends Genet 26:
52. Tella JL (2002) The evolutionary transition to coloniality promotes higher blood
parasitism in birds. J Evol Biol 15: 32–41.
53. Mougeot F, Martı ´nez-Padilla J, Webster LMI, Blount JD, Pe ´rez-Rodrı ´guez L,
et al. (2009) Honest sexual signalling mediated by parasite and testosterone
effects on oxidative balance. Proc R Soc B 276: 1093–1100.
54. Alonso-Alvarez C, Ferrer M (2001) A biochemical study of fasting subfeeding,
and recovery processes in yellow-legged gulls. Physiol Biochem Zool 74:
55. Costantini D, Fanfani A, Dell’Omo G (2008) Effects of corticosteroids on
oxidative damage and circulating carotenoids in captive adult kestrels (Falco
tinnunculus). J Comp Physiol B 178: 829–835.
56. Monaghan P, Metcalfe NB, Torres R (2009) Oxidative stress as a mediator of life
history trade-offs: mechanisms, measurements and interpretation. Ecol Lett 12:
57. Roulin A, Altwegg R, Jensen H, Steinsland I, Schaub M (2010) Sex-dependent
selection on an autosomal melanic female ornament promotes the evolution of
sex ratio bias. Ecol Lett 13: 616–626.
58. Kru ¨ger O, Lindstro ¨m J, Amos W (2001) Maladaptive mate choice maintained
by heterozygote advantage. Evolution 55: 1207–1214.
59. Anderson TM, vonHoldt BM, Candille SI, Musiani M, Greco C, et al. (2009)
Molecular and evolutionary history of melanism in North American gray wolves.
Science 323: 1339–1343.
60. Gaubert P, Me ´zan-Muxart V (2010) Where have the ‘black genets’ gone? A
likely restriction of melanistic cases of the common genet (Genetta genetta) to its
introduced range. Mamm Biol 75: 353–357.
61. Saggese MD, Tizard I, Phalen DN (2008) Mycobacteriosis in naturally infected
ring-neck doves (Streptopelia risoria): investigation of the association between
feather colour and susceptibility to infection, disease and lesions type. Avian
Pathol 37: 443–450.
62. Roulin A (2004) Proximate basis of the covariation between a melanin-based
female ornament and offspring quality. Oecologia 140: 668–675.
Antioxidants and Polymorphism
PLoS ONE | www.plosone.org9 October 2010 | Volume 5 | Issue 10 | e13369