Seasonal Variation in the Regulation of Redox State and Some Biotransformation EnzymeActivities in the Barn Swallow (Hirundo rustica L.)Author(s): Sari Raja-aho, Mirella Kanerva, Tapio Eeva, Esa Lehikoinen, Petri Suorsa, Kai Gao,Dalene Vosloo, Mikko NikinmaaReviewed work(s):Source: Physiological and Biochemical Zoology, Vol. 85, No. 2 (March/April 2012), pp. 148-158Published by: The University of Chicago PressStable URL: http://www.jstor.org/stable/10.1086/664826 .Accessed: 21/03/2012 04:38

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access toPhysiological and Biochemical Zoology.

http://www.jstor.org

148

Seasonal Variation in the Regulation of Redox State and Some

Biotransformation Enzyme Activities in the Barn

Swallow (Hirundo rustica L.)

* S.R. and M.K. contributed equally to this article.

† Corresponding author; e-mail: sari.raja-aho@turku.fi.

Physiological and Biochemical Zoology 85(2):148–158. 2012. � 2012 by TheUniversity of Chicago. All rights reserved. 1522-2152/2012/8502-1130$15.00.DOI: 10.1086/664826

Sari Raja-aho1,*,†

Mirella Kanerva1,*Tapio Eeva1

Esa Lehikoinen1

Petri Suorsa1

Kai Gao1

Dalene Vosloo2

Mikko Nikinmaa1

1Department of Biology, University of Turku, 20014 Turku,Finland; 2School of Biological and Conservation Sciences,University of KwaZulu-Natal, Bag X54001, Durban 4000,South Africa

Accepted 1/21/2012; Electronically Published 3/7/2012

ABSTRACT

Little is known of the normal seasonal variation in redox stateand biotransformation activities in birds. In long-distance mi-gratory birds, in particular, seasonal changes could be expectedto occur because of the demands of migration and reproduc-tion. In this study, we measured several redox parameters inthe barn swallow (Hirundo rustica L.) during the annual cycle.We captured the wintering barn swallows before spring migra-tion in South Africa, and we captured the barn swallows thatarrived in spring, bred in summer, and migrated in autumn inFinland. The redox status and biotransformation activities ofbarn swallows varied seasonally. Wintering birds in South Africahad high biotransformation activities and appeared to expe-rience oxidative stress, whereas in spring and summer, theyshowed relatively low redox (superoxide dismutase [SOD], cat-alase [CAT], and glutathione reductase [GR]) and biotrans-formation enzyme activities. Autumn birds had very low bio-transformation enzyme activities and low indication ofoxidative stress but high activity of some redox enzymes (GRand glucose 6-phosphate dehydrogenase [G6PDH]). High ac-tivities of some redox enzymes (SOD, GR, and G6PDH) seemto be related to migration, whereas low activities of some redoxenzymes (SOD, CAT, and GR) may be associated with breeding.Barn swallows in South Africa may experience pollution-related

oxidative stress, which may hamper interpretation of normalseasonal variation in redox parameters.

Introduction

Environmental monitoring using birds often includes bio-markers related to redox state, which is determined by thebalance between oxidants and antioxidants (Beckman andAmes 1998; Finkel and Holbrook 2000; Halliwell and Gutte-ridge 2007). If an organism is experiencing oxidative stress (i.e.,if the redox state is shifted to the oxidative direction), theformation of various reactive oxygen species (ROS) exceeds theorganism’s antioxidant defense capacity, first causing distur-bances in cellular signaling dependent on ROS and, if moresevere, directly damaging biomolecules, such as DNA, proteins,and lipids (Beckman and Ames 1998; Finkel and Holbrook2000; Valavanidis et al. 2006; Halliwell and Gutteridge 2007;Monaghan et al. 2009). The net effect of oxidative stress maybe seen in breeding success, life span, activity, and many func-tional responses, such as the immune response (Costantini2008). Because oxidative stress is often associated with increasedenergy turnover, seasonal variation might occur, especially inlong-distance migrating birds (McWilliams et al. 2004; Cos-tantini et al. 2007). To assess the oxidative stress caused bychemical exposure, one needs to understand the normal sea-sonal changes in redox parameters, including relevant enzymeactivities (Stohs and Bagchi 1995).

The enzymes involved in regulating redox balance includeglutathione S-transferase (GST), glutathione peroxidase (GP),glutathione reductase (GR), superoxide dismutase (SOD), andcatalase (CAT; Halliwell and Gutteridge 2007; Costantini 2008;Monaghan et al. 2009; Koivula and Eeva 2010). Small anti-oxidant molecules involved in the regulation of redox balancecan be obtained via food, such as carotenoids, flavonoids, andvitamins E and C, or can be produced endogenously, such asglutathione (GSH), which is the major small redox state–reg-ulating molecule (Halliwell and Gutteridge 2007; Costantini2008; Monaghan et al. 2009). Additional important moleculesinvolved in the regulation of redox state are thioredoxins (Ni-shiyama et al. 2001), although their presence and induction inbirds as a result of redox disturbances have not been studied.The biomarkers used in studies of oxidative stress in birds arediscussed in detail in a recent review by Koivula and Eeva(2010). Oxidative stress is often associated with changes in

Seasonal Variation in Redox State and Biotransformation Activity in Birds 149

Table 1: Data on free-living barn swallows captured from roosts in South Africa and Finland during theannual cycle in 2007

Season No. barn swallows Male sex Female sex Capture dates Capture place

Winter 25 15 10 Feb. 21–Mar. 3 Potchefstroom, South AfricaSpring 24 22 2 May 21–23 Petteby, FinlandSummer 22 22 0 June 26, July 5–16 Petteby, FinlandAutumn 23 16 7 Aug. 21–Sept. 19 Petteby, Finland

Total 94 75 19

biotransformation activity (Halliwell and Gutteridge 2007).Compounds that activate the aryl hydrocarbon receptor (AhR)and consecutive xenobiotically induced gene expression, in-cluding biotransformation enzymes of the cytocrome P450family, cause the formation of oxygen free radicals (Whyte etal. 2000). Activation of AhR is often determined using ethox-yresorufin-O-deethylase (EROD) enzyme activity (Lorenzen etal. 1997).

One shortcoming in the use of redox state biomarkers andbiotransformation enzyme activities in environmental studiesis the general lack of the knowledge regarding their seasonalvariation. Knowledge regarding the function of different an-tioxidants, their levels in the blood or tissues of birds, and theseasonal variation in antioxidant activity is still scarce. Only afew studies have monitored the changes throughout the wholeyear (Norte et al. 2008a, 2008b, 2009; Cohen et al. 2009). Thesource of variation in the measured parameters is often difficultto ascertain for adult birds because of their unknown exposureto xenobiotics. The parameters used in biomonitoring may alsochange within the different stages of the life cycle of birds. Forexample, marked changes occur during nestling developmentin some hole-nesting insectivorous passerines (Tanhuanpaa etal. 1999). Seasonal physiological processes, such as molt (Fran-son et al. 2002), reproduction (Alonso-Alvarez et al. 2004;Wiersma et al. 2004; Bertrand et al. 2006), and heavy exerciseduring migration, cause variation in the redox state of wild andcaptive birds (Costantini et al. 2007, 2008; Larcombe et al.2008). Because information on these changes is scarce, we stud-ied the seasonal variation in parameters commonly used todescribe redox state and biotransformation activity using along-distance migratory bird, the barn swallow (Hirundo rusticaL.), as a model species.

Material and Methods

Study Area and Data

The free-living adult barn swallows ( ) used in this studyn p 94were captured from roosts during four different seasons in 2007(table 1). Barn swallow populations breeding in Sweden, Fin-land, the Baltic countries, or farther east most likely winter inthe eastern parts of South Africa, such as the Johannesburgarea (Szep et al. 2006, 2007; Ambrosini et al. 2009). Ringingrecoveries of the Finnish barn swallows in eastern South Africaalso verify this; altogether, 38 (49%) of 78 Finnish barn swallowsrecovered in Africa were recovered in eastern South Africa, and

11 of these were recovered from the area close to Johannesburg(data obtained from the Finnish Ringing Centre). We capturedthe wintering barn swallows before spring migration in Pot-chefstroom, South Africa (26�42′S, 27�06′E; ∼100 km from Jo-hannesburg). Barn swallows that were arriving in spring, breed-ing in summer, and migrating in autumn were captured fromthe common roosts in Petteby, Finland (60�17′N, 22�11′E). Wechose the dates of seasonal captures carefully on the basis ofthe well-documented phenology of Finnish barn swallows(Finnish Ringing Centre). Because we did not capture birdsfrom the nests, we could not determine the exact breeding statusof summer birds. On the basis of testis size (measured in mm),we can assume that the barn swallows were in a breeding phase.Most barn swallows arrive in Finland in the first or secondweek of May and lay their eggs (first brood) in the beginningof June (in 2007, laying started in 85 of 404 nests before May24; calculated according to Saino et al. 2004). The mean hatch-ing day of the first brood was June 19 (data not shown). Barnswallows breeding the second time usually start during the endof June (in 2007, the mean hatching day of the second broodwas July 29; data not shown). By the end of August, most adultbarn swallows are ready for migration, and no visual signs ofbreeding (e.g., incubation patch and cloacal protuberance) canbe found.

We measured the individuals for maximum wing length(Svensson 1992), longest and shortest tail feathers, fat score(ranging from 0 [no subcutaneous fat] to 8 [fat layer coversthe ventral side completely]; Kaiser 1993), and body mass (mea-sured to the nearest 0.1 g with a spring balance). In additionto determining the age of the barn swallows, we determinedtheir sex (75 males and 19 females) according to Jenni andWinkler (1994) and Svensson (1992). We then killed the barnswallows by cervical dislocation and pressed the trachea withthe thumb to make sure of a quick death. Barn swallow car-casses were dissected, and liver samples were collected and im-mediately frozen in liquid nitrogen. The remainder of eachcarcass was frozen (initially at �5�C; carcasses were subse-quently maintained at �20�C) for further analyses. Sex of thebarn swallows was confirmed by examining the gonads. Weused residual mass (deviations from linear regression of bodymass on wing length) and fat score as measures of energeticcondition in the following analyses. The residual mass also takesinto account nonlipid energy sources, such as muscles, and is

150 Raja-aho, Kanerva, Eeva, Lehikoinen, Suorsa, Gao, Vosloo, and Nikinmaa

therefore a more general measure of condition than is fat score(Long and Holberton 2004).

Barn swallows molt completely in the winter quarters. Afterthe molt of tail feathers in Africa, determining the age of barnswallows on the basis of plumage characteristics is impossible.Because their molt is slow (duration, 121–185 d), comparedwith that of most small passerines, barn swallows may start thespring migration while still molting (Jenni and Winkler 1994).We checked the molt stage in wing (primaries and secondaries)and tail feathers of 25 African barn swallows. We used theprimary score of wintering individuals (ranging from 0 for oldfeather to 5 for new feather, summed over nine developedprimaries and thus ranging from 0 to 45; Ginn and Melville1983). With the help of the primary score and the knownminimum duration of the molt (121 d; Ginn and Melville1983), we calculated the minimum number of days each in-dividual still needed to finish the molt (completion of molt p(45 � primary score)/(45/121)). Duration of primary molt canbe used as a proxy for total molt duration, because it usuallyextends over the entire molt period (Jenni and Winkler 1994).

Because the molting of primaries at the tip of the wing wasincomplete, we needed to estimate the final wing length ofwintering barn swallows to calculate the size-independent re-sidual mass used in the analyses. The final wing length wasestimated on the basis of fully grown wing feathers of springbarn swallows ( ). First, we measured the wing lengthn p 24up to the newly molted, already complete primaries (fifththrough eighth). Then, using the mean values of length dif-ferences between each of the primaries of spring barn swallows,we estimated the missing length of the molting wing and addedthe estimate to the measured length to obtain the final winglength.

The experiments that we performed comply with the currentlaws of Finland and South Africa. The licences for animal ex-periments were granted by the Southwest Finland RegionalEnvironment Centre (LOS-2006-L-416-254 and LOS-2007-L-22-254), the Central Animal Laboratory of the University ofTurku (1661/06), and the South African Department of Agri-culture, Conservation, Environment and Tourism (000377NW-06 and 000054NW-07).

Sample Processing

Frozen liver pieces were crushed in liquid nitrogen with a mor-tar and pestle. Thereafter, a piece of liver (approximately

cm) was homogenized in 600 mL of cold 0.1 M0.5 # 0.5K2HPO4 plus 0.15 M KCl buffer (pH 7.4) using TissueLyser(Qiagen), 30 s�1. A 100-mL portion of the homogenate waspipetted into an Eppendorf tube that contained 10 mL of 33mM 1-methyl-2-vinylpyridinium-trifluoromethanesulfonate(Sigma Chemicals) in 0.1 M HCl, for the determination ofoxidized glutathione (GSSG). A 50-mL proportion of homog-enate was pipetted in an Eppendorf tube for the determinationof reduced GSH level. Both tubes were frozen immediately inliquid nitrogen and stored at �80�C until measurements weredetermined. The rest of the homogenate was centrifuged for

15 min at 10,000 g (�4�C). The supernatant was divided intoaliquots, frozen in liquid nitrogen, and stored at �80�C.

Determination of Enzyme Activities and Small-Molecule Levels

The parameters were measured in triplicate (intra-assay coef-ficient of variability [CV] !10% in all cases) using 96-well and384-well microplates, which in most cases required reducingreagent volumes, compared with volumes recommended in themethod instructions. Because we used three control samples inevery plate, we were able to correct the interassay precision(CV, 4%–35%) with the ratio specific to the particular plate(range, 0.8–1.2). The levels of GR, GP, GST, and CAT activitywere measured with Sigma kits (Sigma Chemicals). CAT activityof summer birds was below the detection limit of the methodand therefore could not be measured. The GP activity wasmeasured using 2 mM H2O2 as the substrate. The inhibitionrate of SOD was measured using a Fluka kit (Fluka). The ratiobetween reduced and oxidized glutathione (GSH/GSSG ratio)and the total glutathione content (including both the reducedand oxidized forms of glutathione; totGSH) were measuredusing an OxisResearch kit (OxisResearch). Glucose-6-phos-phate dehydrogenase (G6PDH) activity was measured accord-ing to Noltmann et al. (1961), and the EROD activity wasmeasured according to Burke and Mayer (1974). Lipid hydro-peroxides (LHPs) were measured using the FOX2 method,modified from Nourooz-Zadeh et al. (1995) and Bou et al.(2008), using cumene hydroperoxide as a standard (SigmaChemicals). The protein content was determined with the Brad-ford method (Bradford 1976) using BioRad protein assay re-agent (BioRad) with bovine serum albumin (Sigma Chemicals)as a standard. All of the measurements were performed withan Envision plate reader (Perkin-Elmer) except for protein con-tent determinations, which were performed with a Victor 1plate reader (Perkin-Elmer).

Statistical Methods

We analyzed all data using generalized linear models (GLMs)with the GLIMMIX procedure of SAS, version 9.2 (SAS 2008)and principal component analysis (PCA) with Predictive An-alytics Software statistics 18.0. Response variables GST and GRdistributed normally (checked from the model residuals). Forthe nonnormally distributed data we used lognormal distri-bution (EROD, totGSH, GSH/GSSG ratio, LHP, SOD, CAT, GP,and G6PDH). We calculated 95% confidence intervals (CIs) toestablish bounds for the true effect (Steidl and Thomas 2001).

First, we used GLMs to analyze the seasonal variation inenzyme activity of EROD, GST, totGSH, GSH/GSSG ratio, LHP,SOD, CAT, GP, GR, G6PDH, and protein content. Pairwisecomparisons adjusted for the number of tests were made fromestimated marginal means (EMMs) with the Tukey-Kramer test.Second, we used PCA to discriminate the patterns of variationin the measured variables. PCA was performed using all in-dividual enzyme activities (except for CAT) and other moleculelevels, residual mass, and fat score as input variables (no ro-

Seasonal Variation in Redox State and Biotransformation Activity in Birds 151

Table 2: Generalized linear model statistics for the seasonalvariation in biotransformation activity and redox state ofbarn swallow during the annual cycle

Dependent variable F df P

ERODa 14.60 3, 90 !.0001GSTb 43.26 3, 90 !.0001totGSHa 25.69 3, 90 !.0001GSH/GSSG ratioa 93.03 3, 90 !.0001LHPa 4.57 3, 90 .0005SODa 3.19 3, 90 .0275CATa 4.55 2, 53c .0150GPa 3.31 3, 90 .0238GRb 9.97 3, 90 !.0001G6PDHa 9.19 3, 90 !.0001

Note. CAT p catalase; EROD p ethoxyresorufin-O-deethylase; GP pglutathione peroxidase; GR p glutathione reductase; GSH p reduced

glutathione; GSSG p oxidized glutathione; GST p glutathione S-transferase;

G6PDH p glucose-6-phosphate dehydrogenase; LHP p lipid hydroperoxides;

SOD p superoxide dismutase; totGSH p total glutathione content (including

both reduced and oxidized forms).aLognormal distribution.bNormal distribution.cCAT activity of summer birds was below the detection limit.

tation was used). The requirements for PCA were tested withthe Kaiser-Mayer-Olkin (KMO) test and Bartlett’s test for sphe-ricity ( ; , by Bartlett’s test). We also viewedKMO 1 0.5 P ! 0.05the correlation coefficients, significance levels, and commu-nalities to detect any nonsignificant parameter correlations( , ) or low communalities ( ). Finally, we2r ! 0.3 P 1 0.05 h ! 0.3tested the relationship between different body condition vari-ables and the biomarkers separately for each season. Sex, re-sidual mass, fat score, gonad size (mm), and completion ofmolt (for wintering birds only) were used as explanatory var-iables in the models. Because spring and summer birds weremainly males, and because sex was not associated with themeasured parameters in winter or autumn birds (only the GRactivity in the autumn group differed between females andmales; 5.72 mmol�1 min�1 mg vs. 4.93 mmol�1 min�1 mg), wedid not separate sexes in any of the subsequent analyses.

Results

All the measured biotransformation and redox state parametersshowed statistical differences among seasons (table 2; figs. 1,2). Biotransformation activity, as given by EROD and GSTactivities, is elevated among wintering barn swallows, comparedwith other groups (fig. 1A, 1B). An increase in the totGSH anda decrease in the GSH/GSSG ratio in wintering birds indicatethat they are experiencing oxidative stress (fig. 1C, 1D). Thesebirds also showed a tendency toward having the highest LHPlevels (fig. 1E), indicating the damage done by ROS, althoughLHP levels in the winter group were not statistically significantlydifferent from those in the spring and summer groups. TheSOD inhibition rates and CAT activities were highest in thewinter group and lower in the spring (for SOD inhibition rateand CAT activity) and summer (for SOD inhibition rate only)groups (fig. 2A, 2B). GP activity was lowest in the autumn birds(fig. 2C), which possibly explains the high GSH/GSSG ratioamong those birds (fig. 1D). The levels of GR and G6PDHactivity were elevated in winter and autumn birds (fig. 2D, 2E).Three components showing the eigenvalue 11 together ac-counted for 57% of the variation (table 3). Component 1 (PC1)showed positive loadings by the parameters totGSH, GST,EROD, LHP, and SOD and negative loadings with residualmass. Component 2 (PC2) was positively related to the param-eters G6PDH, GR, and GSH/GSSG ratio and was negativelyrelated to parameter GP. Fat score was positively related tocomponent 3 (PC3; table 3). The first two principal compo-nents (PC1 and PC2) together indicate increased biotransfor-mation activity and oxidative stress in winter quarters, reducedbiotransformation activity but intermediate oxidative stressduring the breeding period, and reduced biotransformationactivity and oxidative stress in autumn (fig. 3).

The residual mass of spring barn swallows was negativelyassociated with LHP and GR (table 4). Residual mass of win-tering birds was positively associated with G6PDH (table 4).The fat score of barn swallows in summer was positively as-sociated with GR and G6PDH (table 4). The size of male testis(mm) in summer was positively associated with totGSH and

GSH/GSSG ratio (table 4; fig. 4A, 4B). In wintering barn swal-lows, the completion of molt was positively associated withGST and LHP (table 4; fig. 5A, 5B).

Discussion

The parameters describing biotransformation activity and re-dox state differed significantly among seasons. Values were es-pecially high in winter barn swallows but were also elevated inautumn barn swallows (i.e., before spring and autumn migra-tions). Values for birds that had recently arrived to their breed-ing grounds resembled values for birds captured during thebreeding period. EROD and GST activities of barn swallowswere highest in winter. A possible explanation for high GSTlevels may be the energy-demanding, active molt in winteringbarn swallows. This is suggested by the fact that birds withnearly completed molt had lower GST activities than did thosein a more active molting phase. GST can also be used in me-tabolizing organic hydroperoxides, such as LHP (levels werealso high in actively molting birds), which is important inrenewing of feathers (Hayes et al. 2005; Halliwell and Gutteridge2007). An alternative explanation would be that toxic com-pounds increasing biotransformation activity, such as poly-chlorinated biphenyls (PCBs), polycyclic aromatic hydrocar-bons (PAHs), dioxin-like compounds or their metabolites, maybe present in the environment of wintering areas. Earlier studieshave found increased EROD activity in tree swallows and wrensafter exposure to PAHs (Bishop et al. 1999; Custer et al. 2001)as well as in nestlings of barn swallows and black guillemotsafter exposure to PCBs (Kuzyk et al. 2003; Custer et al. 2006).Wintering barn swallows in South Africa were captured alonga river that runs through rural areas with intensive agricultural

152

Figure 1. Seasonal differences (estimated marginal mean � 95% confidence interval) in enzyme activity of ethoxyresorufin-O-deethylase(EROD; A); glutathione S-transferase (GST; B); total glutathione content, including reduced and oxidized forms (total GSH; C); ratio betweenreduced and oxidized form of glutathione (GSH/GSSG ratio; D); and lipid hydroperoxide (LHP; E; logarithmic Y-axis). Mean values with thesame lowercase letter are not significantly different (Tukey’s test).

153

Figure 2. Seasonal differences (estimated marginal mean � 95% confidence interval) in enzyme activity of superoxide dismutase (SOD; A),catalase (CAT; B), glutathione peroxidase (GP; C), glutathione reductase (GR; D), and glucose-6-phosphate dehydrogenase (G6PDH; E). Meanswith the same lowercase letter are not significantly different (Tukey’s test).

154 Raja-aho, Kanerva, Eeva, Lehikoinen, Suorsa, Gao, Vosloo, and Nikinmaa

Table 3: Component matrix of the principal component(PC) analysis used to discriminate patterns of variation inthe measured parameters

Variable PC1 PC2 PC3

Eigenvalues:Total 2.795 2.314 1.120Percent of variance 25.405 21.036 10.186Cumulative % 25.405 46.442 56.628

Factor loadings:totGSH .732 .252 .249GST .707 �.371 .150Residual mass �.635 .464 .237EROD .631 .317 .041LHP .508 �.059 .416SOD .412 .339 �.295G6PDH .349 .702 .098GR .348 .687 �.380GSH/GSSG ratio �.463 .599 .006GP .065 �.376 .336Fat �.260 .462 .672

Note. Loadings highlighted in bold represent the highest absolute value of

the parameter when considering different PCs. EROD p ethoxyresorufin-O-

deethylase; GP, glutathione peroxidase; GR p glutathione reductase; GSH preduced glutathione; GSSG p oxidized glutathione; GST p glutathione S-

transferase; G6PDH p glucose-6-phosphate dehydrogenase; LHP p lipid

hydroperoxide; SOD p superoxide dismutase; totGSH p total glutathione

content (including both reduced and oxidized forms).

Figure 3. Two first principal components (PCs) of measured liver bi-otransformation and oxidative stress markers (total glutathione con-tent, including reduced and oxidized forms; glutathione S-transferase,ethoxyresorufin-O-deethylase, lipid hydroperoxide, superoxide dis-mutase, glucose-6-phosphate dehydrogenase, glutathione reductase, ra-tio between reduced and oxidized form of glutathione, and glutathioneperoxidase) and condition measures (residual mass and fat score) inbarn swallows, by season. Wintering barn swallows had high oxidativedamage and high biotransformation enzyme activities, spring and sum-mer birds had low biotransformation activity and intermediate oxi-dative damage, and autumn birds had low biotransformation activityand oxidative stress but high redox enzyme activities. The PC loadingsare shown in table 3.

activities as well as sources of municipal waste. In addition,upriver is an area with several gold mines. It is also reportedthat the PCB content of the air is higher in South Africa thanin Europe (Pozo et al. 2006). It is therefore possible that thebirds captured in South Africa were exposed to higher levelsof toxicants than were the birds captured in Finland. In futurestudies, it would be important to measure the levels of organicpollutants.

The intensive biotransformation metabolism of winteringbirds is one likely reason for them to have experienced oxidativestress. The low GSH/GSSG ratio and high GSH and LHP levelsof wintering barn swallows support this notion. Increased GRand G6PDH activities of winter and autumn birds appear tobe mainly associated with preparation for migration or withincreasing fat reserves in general. The latter enzyme is involvedin nicotinamide adenine dinucleotide/reduced form of nico-tinamide adenine dinucleotide cycling in the pentose phosphatepathway, providing reducing power for GR. Increased activitiesor changed levels of all these biomarkers are associated withoxidative stress (Halliwell and Gutteridge 2007). With regardto totGSH, an increase (Alonso-Alvarez et al. 2008), no change(Rattner et al. 2000; Hilscherova et al. 2003), and a decrease(Costantini and Bonadonna 2010) in oxidative stresses havebeen reported, indicating that it is not a suitable parameter fordemonstrating the occurrence of oxidative conditions.

On the basis of the present results, it is difficult to suggesta role for GP activity changes in life-history traits, such as

preparation for migration. However, its low activity, togetherwith the elevated GR activity, explains the high GSH/GSSG ratioin autumn birds. One possibility is that the increased GSH/GSSG ratio is related to the hyperphagy of premigratory birds.Before the migration flight, birds are fuelling mainly with fats(McWilliams et al. 2004), and high caloric intake is known tocause oxidative conditions (Halliwell and Gutteridge 2007),which can be effectively handled if a high GSH/GSSG ratio ismaintained. Another possibility is that the increased GSH/GSSG ratio is in anticipation of the need to tolerate oxidativestress in the wintering grounds. If this is the case, the findingsuggests that the oxidative conditions in the wintering groundsare not caused by human-induced pollution, because one wouldnot expect genetic adaptation to occur in so few generations.In studies involving great tits (Norte et al. 2009), the GP ac-tivities were higher during winter than during other seasons,which would fit our observations. In barn swallows, GP levelswere also high in spring and summer birds, compared withlevels in autumn birds. The pattern of GP activity may, however,be species specific and differ between migratory and sedentarybirds. In winter, great tits have to deal with cold weather and

Seasonal Variation in Redox State and Biotransformation Activity in Birds 155

Table 4: Generalized linear models for the relationship between the biotransformation activity, redoxstate, and body condition in the barn swallow during the annual cycle

Dependent variable, independent variable(s) Season Slope � SE F (df) P

GST:a

Completion of molt Winter .0004 � .0001 6.79 (1, 23) .0158LHP:b

Residual mass (g) Spring �.06 � .018 9.60 (1, 21) .0054Completion of molt Winter .02 � .004 14.86 (1, 23) .0008

totGSH:b

Size of testis (mm) Summer .3 � .09 9.24 (1, 20) .0065GSH/GSSG ratio:b

Size of testis (mm) Summer .6 � .19 9.24 (1, 20) .0065GR:a

Residual mass (g) Spring �.0004 � .0002 4.72 (1, 21) .0413Fat score Summer .001 � .0004 6.64 (1, 20) .0180

G6PDH:b

Residual mass (g) Winter .1 � .024 31.94 (1, 23) !.0001Fat score Summer .3 � .079 11.67 (1, 20) .0027

Note. GR p glutathione reductase; GSH p reduced glutathione; GSSG p oxidized glutathione; GST p glutathione S-

transferase; G6PDH p glucose-6-phosphate dehydrogenase; LHP p lipid hydroperoxides; num df p numerator degrees of

freedom; totGSH p total glutathione content (including both reduced and oxidized forms).aNormal distribution.bLognormal distribution.

short days. Norte et al. (2008a) hypothesized that the metab-olism of great tits is slower during cold weather. Regardless ofthe reason, increased biotransformation activity can cause ROSlevels to increase (Schlezinger et al. 2006), as was the case withwintering barn swallows. High SOD and CAT activities of win-ter birds may also be related to the biotransformation of toxiccompounds. Both enzymes are particularly important in ROSregulation, and the low activities of SOD in spring and summerbirds could suggest differences in redox signaling betweenbreeding and migratory birds.

Reproduction does not seem to change the redox state insummer birds, compared with that in birds that have not yetstarted reproducing (spring birds), although one should recallthat we could not determine the exact breeding status of sum-mer birds. One possible reason for this is that reproduction ofbarn swallows is not energetically so costly (relative to availableresources) that it would generate any imbalances. Another pos-sibility is that some (spring) birds may have recovered frommigration and prepared for reproduction in May, when theywere captured, because they had been in Finland for 2–3 wkalready. Notably, most studies of exercise and oxidative stresshave been performed with captive birds unaccustomed to long-distance flights (Costantini et al. 2008; Larcombe et al. 2010),which makes comparisons with our findings difficult. Markedsex differences have been observed with reproduction-inducedoxidative conditions (Alonso-Alvarez et al. 2004; Wiersma etal. 2004; Bertrand et al. 2006). This may be associated withmarked differences in energy gain and allocation between sexes.Because our spring and summer groups consisted mostly ofmales, we cannot compare the sex effect during the breeding

period. Testosterone elevated oxidative stress in zebra finch(Alonso-Alvarez et al. 2007) and red-legged partridge (Alonso-Alvarez et al. 2008). In our study, the males with larger testishad higher totGSH and GSH/GSSG ratio than did males withsmaller testis, indicating a greater reduction in redox status.Testis size was negatively related to the capture date; for birdscaptured late in the summer, the size of the testis was smaller.The size of the testis is not necessarily related to testosteroneproduction (Moore et al. 2002), but larger testis size suggeststhat sperm production is effective (Møller 1988). This couldindicate that males with more reducing power to cope againstROS might have better reproduction success than their con-specifics with lower GSH/GSSG ratio.

In a study by Franson et al. (2002), molt increased the activityof GR and GP, compared with the activity in nonmolting birds.In our study, we found no elevation in GR or GP activity inmolting birds during winter. However, overall, both GR andGP activities in wintering birds were high. GR activity of winterbirds, again, might be more related to preparation for migrationthan to molt, because the autumn birds had values that weresimilar to those for the winter birds. GST and LHP, which wereboth associated with the completion of molt, are metabolicallyconnected. GST can metabolize organic hydroperoxides by us-ing reduced glutathione (Hayes et al. 2005; Halliwell and Gut-teridge 2007). One hypothesis for the relationship betweenmolt, GST, and LHP is that, during feather growth, LHP isformed and GST metabolizes it. However, Jenni-Eiermann andJenni (1996) did not find an association between lipid metab-olites and molting in a comparison of postbreeding, migrating,and molting birds.

156 Raja-aho, Kanerva, Eeva, Lehikoinen, Suorsa, Gao, Vosloo, and Nikinmaa

Figure 4. Relationship between the size of testis and the redox state parameters (A) total glutathione content, including reduced and oxidizedforms, and (B) ratio between reduced and oxidized form of glutathione (GSH/GSSG ratio) in the barn swallow in summer.

Figure 5. Relationship between the completion of molt (days left for molt completion) and (A) biotransformation activity of glutathione S-transferase (GST) and (B) redox state parameter lipid hydroperoxide (LHP) in wintering barn swallows.

Redox parameters of birds differed between winter, springand summer, and autumn groups. Biotransformation activityand indications of oxidative stress were high in winter birds,whereas most of the redox enzymes were on an intermediatelevel. The spring and summer birds were profiled with lowlevels of redox enzyme and biotransformation activity and ex-hibited intermediate oxidative stress. In the autumn, barn swal-lows had very little biotransformation activity, and they seemedto have low oxidative stress even though some of the redoxenzymes (SOD, GR, and G6PDH) were highly active. It can beassumed that low enzyme activity and intermediate oxidativestress are associated with preparation for breeding or breedingitself, whereas high activity of some enzymes is associated withenergetic preparation for migration.

In environmental monitoring, it is important to be aware ofthe annual cycles and to take them into account in samplingfor physiological parameters. Also, studying the reasons behindthese changes helps us to understand the mechanisms of redoxregulation. To be able to generalize the results of the seasonalchanges in redox parameters across birds, more bird speciesshould be studied, because factors such as migratory behavior(resident-migratory) or the developmental stage of the organ-ism may affect the response. The best way to detect and studyoxidative stress is to use several different biomarkers, becausethe interaction between various antioxidants is very importantin the overall level of defense (Monaghan et al. 2009). In thecase of the barn swallow, the knowledge of exposure to con-taminants in the wintering grounds in South Africa would help

Seasonal Variation in Redox State and Biotransformation Activity in Birds 157

in interpreting which changes are normal seasonal fluctuationsand which are caused by xenobiotic biotransformation (Pozoet al. 2006).

Acknowledgments

We are grateful for valuable comments by two anonymousreviewers. Kirsi Reponen, Raimo Hyvonen, Maj Bjork, Ari Kar-hilahti, Jorma Nurmi, Jouni Saario, the Nylund family, AndreVosloo, Riaan Booysen, and Rina Booysen are thanked for theirhelp for assistance during capture and handling procedures.This study was financed by the Emil Aaltonen Foundation(S.R.), the Kone Foundation (M.K.), Academy of Finland (pro-ject 8119367 to T.E. and Centre of Excellence to M.N.), andthe University of Turku (Centre of Excellence to M.N.).

Literature Cited

Alonso-Alvarez C., S. Bertrand, G. Devevey, J. Prost, B. Faivre,and G. Sorci. 2004. Increased susceptibility to oxidative stressas a proximate cost of reproduction. Ecol Lett 7:363–368.

Alonso-Alvarez C., S. Bertrand, B. Faivre, O. Chastel, and G.Sorci. 2007. Testosterone and oxidative stress: the oxidationhandicap hypothesis. Proc R Soc B 274:819–825.

Alonso-Alvarez C., L. Perez-Rodriguez, R. Mateo, O. Chastel,and J. Vinuela. 2008. The oxidation handicap hypothesis andthe carotenoid allocation trade-off. J Evol Biol 21:1789–1797.

Ambrosini R., A.P. Møller, and N. Saino. 2009. A quantitativemeasure of migratory connectivity. J Theor Biol 257:203–211.

Beckman K.B. and B.N. Ames. 1998. The free radical theoryof aging matures. Physiol Rev 78:547–581.

Bertrand S., C. Alonso-Alvarez, G. Devevey, B. Faivre, J. Prost,and G. Sorci. 2006. Carotenoids modulate the trade-off be-tween egg production and resistance to oxidative stress inzebra finches. Oecologia 147:576–584.

Bishop C.A., N.A. Mahony, S. Trudeau, and K.E. Pettit. 1999.Reproductive success and biochemical effects in tree swallows(Tachycineta bicolor) exposed to chlorinated hydrocarboncontaminants in wetlands of the Great Lakes and St.Lawrence River basin, USA and Canada. Environ ToxicolChem 18:263–271.

Bou R., R. Codony, A. Tres, E.A. Decker, and F. Guardicila.2008. Determination of hydroperoxides in foods and bio-logical samples by the ferrous oxidation-xylenol orangemethod: a review of the factors that influence the method’sperformance. Anal Biochem 377:1–15.

Bradford M.M. 1976. Rapid and sensitive method for quanti-tation of microgram quantities of protein utilizing principleof protein-dye binding. Anal Biochem 72:248–254.

Burke M.D. and R.T. Mayer. 1974. Ethoxyresorufin: directfluorimetric assay of a microsomal o-dealkylation which ispreferentially inducible by 3-methylcholanthrene. Drug Me-tab Dispos 2:583–588.

Cohen A.A., K.J. McGraw, and W.D. Robinson. 2009. Serum

antioxidant levels in wild birds vary in relation to diet, season,life history strategy, and species. Oecologia 161:673–683.

Costantini D. 2008. Oxidative stress in ecology and evolution:lessons from avian studies. Ecol Lett 11:1238–1251.

Costantini D. and F. Bonadonna. 2010. Patterns of variationof serum oxidative stress markers in two seabird species.Polar Res 29:30–35.

Costantini D., M. Cardinale, and C. Carere. 2007. Oxidativedamage and anti-oxidant capacity in two migratory bird spe-cies at a stop-over site. Comp Biochem Phys C 144:363–371.

Costantini D., G. Dell’Ariccia, and H.P. Lipp. 2008. Long flightsand age affect oxidative status of homing pigeons (Columbalivia). J Exp Biol 211:377–381.

Custer T.W., C.M. Custer, K. Dickerson, K. Allen, M.J. Melan-con, and L.J. Schmidt. 2001. Polycyclic aromatic hydrocar-bons, aliphatic hydrocarbons, trace elements, and monoox-ygenase activity in birds nesting on the North Platte River,Casper, Wyoming, USA. Environ Toxicol Chem 20:624–631.

Custer T.W., C.M. Custer, B.L. Goatcher, M.J. Melancon, C.W.Matson, and J.W. Bickham. 2006. Contaminant exposure ofbarn swallows nesting on Bayou D’Inde, Calcasieu Estuary,Louisiana, USA. Environ Monit Assess 121:543–560.

Finkel T. and N.J. Holbrook. 2000. Oxidants, oxidative stressand the biology of ageing. Nature 408:239–247.

Franson J.C., D.J. Hoffman, and J.A. Schmutz. 2002. Bloodselenium concentrations and enzyme activities related to glu-tathione metabolism in wild emperor geese. Environ ToxicolChem 21:2179–2184.

Ginn H.B. and D.S. Melville. 1983. Moult in birds. BTO Guide19. BTO, Hertfordshire.

Halliwell B. and J.M.C. Gutteridge. 2007. Free radicals in bi-ology and medicine. Oxford University Press, Oxford.

Hayes J.D., J.U. Flanagan, and I.R. Jowsey. 2005. Glutathionetransferases. Annu Rev Pharmacol 45:51–88.

Hilscherova K., A. Blankenship, K. Kannan, M. Nie, L.L. Wil-liams, K. Coady, B.L. Upham, J.E. Trosko, S. Bursian, andJ.P. Giesy. 2003. Oxidative stress in laboratory-incubateddouble-crested cormorant eggs collected from the GreatLakes. Arch Environ Con Tox 45:533–546.

Jenni L. and R. Winkler. 1994. Moult and ageing of Europeanpasserines. Academic Press, London.

Jenni-Eiermann S. and L. Jenni. 1996. Metabolic differencesbetween the postbreeding, moulting and migratory periodsin feeding and fasting passerine birds. Funct Ecol 10:62–72.

Kaiser A. 1993. A new multi-category classification of subcu-taneous fat deposits of songbirds. J Field Ornithol 64:246–255.

Koivula M.J. and T. Eeva. 2010. Metal-related oxidative stressin birds. Environ Pollut 158:2359–2370.

Kuzyk Z.Z.A., N.M. Burgess, J.P. Stow, and G.A. Fox. 2003.Biological effects of marine PCB contamination on blackguillemot nestlings at Saglek, Labrador: liver biomarkers.Ecotoxicology 12:183–197.

Larcombe S.D., J.S. Coffey, D. Bann, L. Alexander, and K.E.Arnold. 2010. Impacts of dietary antioxidants and flight

158 Raja-aho, Kanerva, Eeva, Lehikoinen, Suorsa, Gao, Vosloo, and Nikinmaa

training on post-exercise oxidative damage in adult parrots.Comp Biochem Physiol B 155:49–53.

Larcombe S.D., C.A. Tregaskes, J.S. Coffey, A.E. Stevenson, L.Alexander, and K.E. Arnold. 2008. The effects of short-termantioxidant supplementation on oxidative stress and flightperformance in adult budgerigars Melopsittacus undulatus. JExp Biol 211:2859–2864.

Long J.A. and R.L. Holberton. 2004. Corticosterone secretion,energetic condition, and a test of the migration modulationhypothesis in the hermit thrush (Catharus guttatus), a short-distance migrant. Auk 121:1094–1102.

Lorenzen A., J.L. Shutt, and S.W. Kennedy. 1997. Sensitivity ofcommon tern (Sterna hirundo) embryo hepatocyte culturesto CYP1A induction and porphyrin accumulation by halo-genated aromatic hydrocarbons and common tern egg ex-tracts. Arch Environ Con Tox 32:126–134.

McWilliams S.R., C. Guglielmo, B. Pierce, and M. Klaassen.2004. Flying, fasting, and feeding in birds during migration:a nutritional and physiological ecology perspective. J AvianBiol 35:377–393.

Møller A.P. 1988. Testes size, ejaculate quality and sperm com-petition in birds. Biol J Linn Soc 33:273–283.

Monaghan P., N.B. Metcalfe, and R. Torres. 2009. Oxidativestress as a mediator of life history trade-offs: mechanisms,measurements and interpretation. Ecol Lett 12:75–92.

Moore I.T., N. Perfito, H. Wada, T.S. Sperry, and J.C. Wingfield.2002. Latitudinal variation in plasma testosterone levels inbirds of the genus Zonotrichia. Gen Comp Endocrinol 129:13–19.

Nishiyama A., H. Masutani, H. Nakamura, Y. Nishinaka, andJ. Yodoi. 2001. Redox regulation by thioredoxin and thio-redoxin-binding proteins. Iubmb Life 52:29–33.

Noltmann E., C.J. Gubler, and S.A. Kuby. 1961. Glucose 6-phosphate dehydrogenase (zwischenferment). 1. Isolation ofcrystalline enzyme from yeast. J Biol Chem 236:1225–1230.

Norte A.C., J.A. Ramos, P.M. Araujo, J.P. Sousa, and B.C. Shel-don. 2008a. Health-state variables and enzymatic biomarkersas survival predictors in nestling great tits (Parus major):effects of environmental conditions. Auk 125:943–952.

Norte A.C., B. Sheldon, J.P. Sousa, and J.A. Ramos. 2008b.Repeatability and method-dependent variation of blood pa-rameters in wild-caught great tits Parus major. Acta Ornithol43:65–75.

Norte A.C., J.A. Ramos, J.P. Sousa, and B.C. Sheldon. 2009.Variation of adult great tit Parus major body condition andblood parameters in relation to sex, age, year and season. JOrnithol 150:651–660.

Nouroozzadeh J., J. Tajaddinisarmadi, S. McCarthy, D.J. Bet-teridge, and S.P. Wolff. 1995. Elevated levels of authenticplasma hydroperoxides in niddm. Diabetes 44:1054–1058.

Pozo K., T. Harner, F. Wania, D.C.G. Muir, K.C. Jones, andL.A. Barrie. 2006. Toward a global network for persistentorganic pollutants in air: results from the GAPS study. En-viron Sci Technol 40:4867–4873.

Rattner B.A., D.J. Hoffman, M.J. Melancon, G.H. Olsen, S.R.Schmidt, and K.C. Parsons. 2000. Organochlorine and metalcontaminant exposure and effects in hatching black-crownednight herons (Nycticorax nycticorax) in Delaware Bay. ArchEnviron Con Tox 39:38–45.

Saino N., M. Romano, R. Ambrosini, R.P. Ferrari, and A.P.Møller. 2004. Timing of reproduction and egg quality covarywith temperature in the insectivorous Barn Swallow, Hirundorustica. Funct Ecol 18:50–57.

SAS Institute. 2008. SAS/STAT 9.2 User’s guide. SAS Institute,Cary, NC.

Schlezinger J.J., W.D.J. Struntz, J.V. Goldstone, and J.J. Stege-man. 2006. Uncoupling of cytochrome P450 1A and stim-ulation of reactive oxygen species production by co-planarpolychlorinated biphenyl congeners. Aquat Toxicol 77:422–432.

Steidl R.J. and L. Thomas. 2001. Power analysis and experi-mental design. Pp. 14–36 in S.M. Scheiner and J. Gurevitch,eds. Design and analysis of ecological experiments. OxfordUniversity Press, New York.

Stohs S.J. and D. Bagchi. 1995. Oxidative mechanisms in thetoxicity of metal-ions. Free Radical Bio Med 18:321–336.

Svensson L. 1992. Identification guide to european passerines.Fingraf, Sodertalje, Sweden.

Szep T., A.P. Møller, S. Piper, R. Nuttall, Z.D. Szabo, and P.L.Pap. 2006. Searching for potential wintering and migrationareas of a Danish Barn Swallow population in South Africaby correlating NDVI with survival estimates. J Ornithol 147:245–253.

———. 2007. Migratory connectivity in barn swallows andother hirundines. J Ornithol 148:257–260.

Tanhuanpaa S., T. Eeva, E. Lehikoinen, and M. Nikinmaa. 1999.Developmental changes in 7-ethoxyresorufin-O-deethylase(EROD) and delta-aminolevulinic acid dehydratase (ALA-D) activities in three passerines. Comp Biochem Physiol C124:197–202.

Valavanidis A., T. Vlahogianni, M. Dassenakis, and M. Scoullos.2006. Molecular biomarkers of oxidative stress in aquaticorganisms in relation to toxic environmental pollutants. Eco-tox Environ Safe 64:178–189.

Whyte J.J., R.E. Jung, C.J. Schmitt, and D.E. Tillitt. 2000. Ethox-yresorufin-O-deethylase (EROD) activity in fish as a bio-marker of chemical exposure. Crit Rev Toxicol 30:347–570.

Wiersma P., C. Selman, J.R. Speakman, and S. Verhulst. 2004.Birds sacrifice oxidative protection for reproduction. Proc RSoc B 271:S360–S363.