Maternal immunoglobulin concentration in Collared. Flycatcher (Ficedula albicollis) eggs in relation to parental quality and laying order

Functional Ecology 2006 Maternal immunoglobulin concentration in Collared Blackwell Publishing Ltd Flycatcher (Ficedula albicollis) eggs in relation to parental quality and laying order R. HARGITAI,* J. PRECHL and J. TÖRÖK* *Behavioural Ecology Group, Department of Systematic Zoology and Ecology, Eötvös Loránd University, Pázmány P. sétány 1/C, H-1117 Budapest, Hungary, and Immunology Research Group, Hungarian Academy of Sciences, Pázmány P. sétány 1/C, H-1117 Budapest, Hungary Summary 1. The immune system of newly hatched birds is relatively immature; therefore pathogens can be particularly virulent. Females transfer passive immunity in terms of immunoglobulins to the eggs to protect their young against infections in the crucial early life stages. 2. As transmission of antibodies is likely to be nutritionally costly, mothers are expected to allocate these components differentially to eggs according to their own condition, the quality of their mate or the laying order of eggs. 3. We found that in Collared Flycatchers (Ficedula albicollis), yolk antibody levels positively correlated with female body condition, but showed no relationship with maternal age or body size. Furthermore, females with higher plasma heterophil : lymphocyte ratio and heterophil count, which may indicate higher level of stress, deposited lower amount of immunoglobulins to their eggs. These results suggest that females of better physiological condition were able to invest more immunoglobulins to their eggs. 4. Neither the plumage characteristics, nor the age of the male parent was related to yolk antibody concentration, and thus no evidence for differential allocation of antibodies in relation to male quality or attractiveness could be detected. 5. Last-laid eggs contained higher yolk immunoglobulin concentration than earlier-laid eggs within a clutch. This pattern could be interpreted as a way to improve the survival probability of the disadvantaged last-hatching nestling. Key-words: Antibody transmission, brood survival strategy, egg quality, maternal effects, passive immunity Functional Ecology (2006) doi: 10.1111/j.1365-2435.2006.01171.x Ecological Society Introduction Neonatal birds initially have an inefficient immune system and limited ability to synthesize antibodies (Apanius 1998; Pastoret et al. 1998), hence young individuals are particularly vulnerable to infections. Females transmit immune factors, such as lysozyme, an important component of innate antibacterial immunity to the albumen (Trziszka 1994; Saino et al. 2002a), and antibodies to the developing oocyte (Kowalczyk et al. 1985), which confer passive immune protection against the various pathogens their offspring may face after hatching. The deposition of these components of the maternal immune system provides a way to adaptively influence the phenotype of the progeny through nongenetic effects (Mousseau & Fox 1998; Grindstaff, Brodie & Ketterson 2003). Author to whom correspondence should be addressed. E-mail: harrita@freemail.hu The transfer of maternally derived antibodies is assumed to be highly beneficial, since it reduces the susceptibility of nestlings to local diseases and pathogens by allowing them to apply a highly specific and efficient acquired immune defence even at the first encounter with a pathogen (Fadly & Nazerian 1989; Graczyk et al. 1994; Smith et al. 1994; Gasparini et al. 2001). Furthermore, maternal antibodies may continue to affect the fitness of the young even after disappearing from the circulation, by influencing the development and the efficiency of the immune system (Carlier & Truyens 1995; Lemke & Lange 1999; Buechler et al. 2002; Lozano & Ydenberg 2002). On the other hand, at very high levels, maternal antibodies may have blocking effects preventing the stimulation of the neonatal immune mechanisms, and thus suppressing the offspring s ability to develop active immunity against later infections by the same antigen (Carlier & Truyens 1995). Consequently, there should be a balance between providing protection and allowing the differentiation of specific immune cells by the young. 829

830 R. Hargitai et al. However, this maternal investment is likely to incur energetic and nutritional costs to the female, since females elevate their antibody production during egg formation (Klasing 1998; Saino, Martinelli & Møller 2001a), and mounting an immune response requires significant supplies of proteins and other nutrients (Sklan, Melamed & Friedman 1994; Sheldon & Verhulst 1996; Klasing 1998; Lochmiller & Deerenberg 2000). Therefore, females in poor condition may face limitations in the allocation of immunoglobulins because of nutritional constraints (Grindstaff et al. 2003). Furthermore, several other factors may affect the immunoglobulin transfer to the eggs, such as the age (Barua, Yoshimura, & Tamura 1998) and health of the mother (Smith et al. 1994; Gasparini et al. 2001; Blount et al. 2002; Buechler et al. 2002), environmental factors (Müller et al. 2004) and even the quality or attractiveness of the father ( Saino et al. 2002b). Concentration of antibodies may also vary among individual eggs within a clutch (Blount et al. 2002; Saino et al. 2002b; Müller et al. 2004). In most altricial bird species, incubation begins before clutch completion, resulting in asynchronous hatching (Stoleson & Beissinger 1995; Stenning 1996). Consequently, the youngest sibling is at a significant competitive disadvantage (Price & Ydenberg 1995; Viñuela 2000). Recent work has demonstrated that female birds can differentially deposit steroid hormones (Schwabl 1993; Lipar & Ketterson 2000; Eising et al. 2001) and carotenoids (Royle et al. 1999; Blount et al. 2002; Hõrak, Surai & Møller 2002; Saino et al. 2002c) into their eggs, favouring or handicapping young, depending on the laying sequence. In the case of adaptive brood reduction, parents would benefit from allocating more resources to the eggs of the highest reproductive value (first-laid eggs), thus exaggerating the effects of hatching asynchrony on the last nestling (Lack 1968; Schwabl, Mock & Gieg 1997; Viñuela 1997; Gil et al. 1999). Thus, when food availability is insufficient to rear the whole brood, the smallest nestling would die quickly, improving its siblings chances for survival. In contrast, if hatching asynchrony is the consequence of some physiological or environmental constraint (Stoleson & Beissinger 1995), females may attempt to attenuate the competitive disadvantage of later hatching nestlings by providing them with relatively more micro- or macronutrients (brood-survival strategy; Clark & Wilson 1981; Slagsvold et al. 1984; Schwabl 1993; Lipar, Ketterson & Nolan 1999). Intraclutch variation in immunoglobulin concentration may be interpreted as a way to modify the survival probabilities and immunocompetence of nestlings in relation to laying order. In this study, we investigated levels of maternal IgG (sometimes referred to as IgY) in the eggs of Collared Flycatchers (Ficedula albicollis) in relation to parental quality and the laying sequence of the eggs. We hypothesized that females in better condition would transfer higher amount of antibodies to their eggs as they have higher availability to nutrients. We also expected that females with higher plasma level of carotenoids would lay eggs with more antibodies, as carotenoids are thought to act as protectors of leucocytes from the damaging effects of free radicals, and furthermore, they play important roles in immunostimulation (Edge et al. 1997; Møller et al. 2000; Chew & Park 2004). Furthermore, in accordance with the differential allocation hypothesis (Burley 1986; Sheldon 2000), we expected that females would adjust their antibody investment to the quality of their mate, either by providing more antibodies to the offspring of a high-quality or attractive father, or by compensating the disadvantaged young sired by a poor-quality or unattractive male. Finally, we predicted that immunoglobulin concentration would increase with laying sequence in concert with our previous findings that Collared Flycatchers apply a brood-survival strategy improving the viability of the last-hatching young (Hargitai et al. 2005; Rosivall, Szöllosi & Török 2005; J. Török, R. Hargitai, G. Hegyi, Z. Matus, G. Michl, P. Péczely, B. Rosivall & Gy. Tóth, unpublished data). Materials and methods STUDY SPECIES AND FIELD PROCEDURES The study was carried out in a nest-box-breeding population of Collared Flycatchers in an oak-dominated woodland in the Pilis Mountains (47 43 N, 19 01 E), Hungary, in 2001 02 and 2004 05. The Collared Flycatcher is a long-distance migratory, hole-nesting, insectivorous passerine. This species is predominantly monogamous, and females usually lay five to seven eggs of one clutch per breeding season. The forehead patch size of males is an important determinant of the male s social mating status (Gustafsson, Qvarnström & Sheldon 1995; Török et al. 1999), as well as success in sperm competition (Sheldon & Ellegren 1999; Michl et al. 2002), but in contrast to a Swedish population (Qvarnström 1999), it does not indicate body condition of males in our population (Hegyi et al. 2002, 2006a). The size of the white wing patch of males may reliably indicate individual quality, as it is condition-dependent, and predicts the male s survival probability (Török, Hegyi & Garamszegi 2003). This trait also plays a role in territorial competition among males (Garamszegi et al. 2006). Eggs were marked according to laying order with a waterproof marker, and collected before the clutch was complete to ensure that embryonic development could not affect yolk composition. Collected eggs were replaced with dummy eggs to prevent females from abandoning their nests. Egg length and breadth were measured with a calliper (to the nearest 0 1 mm), egg size was estimated from Hoyt s formula (Hoyt 1979; size = length breadth 2 0 51). Our analysis included 75 clutches (12, 21, 19 and 23 clutches from 2001, 2002, 2004 and 2005, respectively). In 2004 and 2005 only two eggs per clutch were collected, but these clutches were also included in the interclutch analyses, as interclutch variance of yolk immunoglobulin concentration was

831 Immunoglobulins in Collared Flycatcher eggs significantly higher than intraclutch variance (repeatability: 86%, F 40,41 = 12 94, P < 0 001). Clutches from 2004 and 2005 were parts of different experimental studies. In 2004, females intraperitoneally received an injection of either an 5% sheep red blood cell in 100 µl phosphate buffered saline (PBS; experimental group) or 100 µl PBS (control group) during nest-building. In 2005, an empty cage (control group) or a cage with a female Collared Flycatcher (experimental group) was deposited near the female s nestbox for c. 1 h twice a day during the nest-building period. As experimental manipulation had no effect on yolk immunoglobulin level (2004: F 1,17 = 0 042, P = 0 84; 2005: F 1,21 = 0 003, P = 0 96), and entering it as factor in the analyses did not change our results qualitatively (results not shown), this variable was not included in the tests presented. Females were captured in the nest-box during incubation, whereas males were trapped during the courtship or nestling feeding period. Birds were weighed with a Pesola spring balance (to the nearest 0 1 g), and the length of the tarsus was measured with a calliper (to the nearest 0 1 mm). Female body condition was calculated as the residual of a linear regression of body mass on tarsus length. In another data set of repeated measurements of the same bird, female condition during incubation significantly correlated with condition before egg-laying (r = 0 62, n = 14, P = 0 017). Thus, although female condition was measured during the incubation period instead of the laying period so as to prevent nest desertion, we assume that the condition of incubating females is closely related to their condition at egg-laying. In addition, the sizes of the white forehead patch and wing patch of males were measured. The product of largest width and largest height (to the nearest 0 1 mm with a calliper) was used as an indicator of forehead patch size (Hegyi et al. 2006a). Wing patch size was expressed as the sum of the length of the white areas from the tip of the primary coverts on the outer web of primaries four to eight (Török et al. 2003). IMMUNOLOGICAL ASSAY Yolks were separated from albumen, thoroughly homogenized in an equal (1 ml per gram of yolk) amount of 0 9% (wt/v) NaCl solution, and stored at 20 C until analysis. Albumen was not sampled because of the very small quantities of antibodies occurring there (Rose, Orlans & Buttress 1974). Antibody (IgG) concentrations were determined using an indirect enzyme-linked immunosorbent assay (ELISA). ELISA plates (Greiner) were coated with yolk samples diluted 1:20 000 in carbonate buffer (ph 9 6). Some 50 µl of this dilution was added to plates in triplicates of each sample and incubated overnight at 4 C. To assess total antibody concentration, a serial dilution of a standard of pooled Collared Flycatcher yolks (n = 10) was added in duplicate to each plate. Antibody concentration of the standard was estimated in relation to chicken IgG (Sigma I-4881 (Sigma Chemical Company, PO Box 14508, St Louis, MO, 63178, USA)) with a similar indirect ELISA, resulting in a 0 38 mg ml 1 chicken IgG equivalent concentration. The linear range of the sigmoid curve of the dilutions was between 1:4000 and 1:64 000, which were thus assigned a concentration of 95 ng ml 1 and 5 9 ng ml 1, respectively. Eggs from the same clutches were estimated in one plate. The plates were washed four times with PBS buffer (200 µl) and blocked with 1% BSA-PBS (50 µl, Sigma A-4503) for 1 h at 37 C. After washing the wells with 200 µl of PBS-Tween 20 (0 05%) buffer (four times), 50-µl aliquots of rabbit antichicken IgG antiserum, produced by immunizing rabbits with purified chicken IgG (Sigma I-4881), were added to each well in a dilution of 1:500 in PBS-Tween 20. In order to control for the possible background of the technique, normal rabbit serum was added as a negative control to the 1:20 000 dilution of the standard yolk, and this dilution of the standard was also used without the antichicken antibody step. The absorbances of the negative controls were always lower than 0 15 and lower than the lowest absorbance value of the yolk samples on the plate. Initially, some wells were also coated with ovalbumin to test the binding of the antichicken antibody to this protein, and absorbances were never higher than 0 08. Plates were incubated for 3 h at 37 C, then washed four times with PBS-Tween 20. Subsequently, 50 µl of peroxideconjugated goat antirabbit IgG (affinity purified, Bio- Rad 172 1019 (Bio-Rad Laboratories Inc., 1000 Alfred Nobel Dr., Hercules, CA 94547)) diluted 1 : 10 000 in PBS-Tween 20 were added to each well and incubated overnight at 4 C. After washing the wells with PBS- Tween 20 (four times) a substrate (ABTS, 2,2 -azinobis (3-ethylbenzthiazoline-6-sulphonic acid), Sigma A- 1888) and hydrogen peroxide in citrate buffer were added (50 µl), then plates were covered with an aluminium foil (to keep them protected from light) and incubated at room temperature for 1 5 h. During this time, absorbances were measured every 15 min with a Dynatech MR700 Microplate reader (Dynatech Laboratories Inc., 14340 Sullyfield Cir., Chantilly, VA, 22021, USA) using a 405 nm wavelength filter. The measurement in which the standard points reached their highest value was used for concentration calculations. Data presented correspond to the yolk IgG concentration (µg ml 1 ) of the undiluted yolk sample calculated from the mean of the triplicate relative to chicken IgG concentration (see above). Intra- and interassay CV% were < 12% and < 20%, respectively. If intra-assay CV% reached a higher value than 12%, the mean of the two closest values of the triplicate was used. Plasma IgG levels were analysed with the same procedure with the same dilutions using pooled Collared Flycatcher plasma as standard, that gave a concentration of 0 86 mg ml 1 relative to chicken IgG. In that case, interassay coefficients of variation were < 10%. The specificity of the primary antichicken antibody used in the ELISA was previously tested by Western blot, showing a reactivity to F. albicollis IgG heavy chain.

832 R. Hargitai et al. SEROLOGICAL ANALYSES A blood sample (50 70 µl) was taken from the female s brachial vein in heparinized capillaries, and centrifuged at 10 000 r.p.m. for 10 min on the same day. The haematocrit value was calculated as the ratio between the length of the capillary occupied by red blood cells and the length of the capillary occupied by total blood. Haematocrit value has been regarded as an index of the health status of the bird (Harrison & Harrison 1986; Svensson & Merilä 1996; Potti et al. 1999). Lower level of haematocrit indicates anaemia, which could be related to diseases such as ecto-, endo- or haemoparasitic infections, but also nutritional deficiencies can cause lowered haematocrit values (Svensson & Merilä 1996; Merino & Potti 1998; Hoi-Leitner et al. 2001). Plasma was extracted and stored at 20 C until later analysis. Female haematocrit and plasma immunoglobulin levels were measured only in 2004 and 2005. A drop of blood was smeared on a microscope slide, air dried, fixed in absolute methanol and later stained with Giemsa s solution. The proportion of different types of leucocytes was assessed by examining a total of 50 leucocytes under 1000 magnification with oil immersion. An index of the relative abundance of heterophil granulocytes to lymphocytes (H/L ratio) was used as a measure of serological stress level (Gross & Siegel 1983; Maxwell 1993). Total white blood cell count (WBC) was estimated by counting the number of leucocytes in 10 fields at 40 magnification, and dividing by 10 in order to obtain the average number of leucocytes in one field. This number was multiplied by 2000 (Fudge 2000). The WBC was corrected by the haematocrit value according to the formula: corrected WBC = WBC (haematocrit/normal haematocrit), in which normal haematocrit is 45% (Campbell 1988). Heterophil and lympocyte numbers were estimated by multiplying their proportions with corrected WBC. Elevated lymphocyte count is generally associated with parasitic or viral infections, while an increase in heterophil number is commonly observed in response to bacterial infections, inflammation or stress (Harmon 1998; Fudge 2000). We had data for leucocyte profile only for females in 2005. Total plasma carotenoid concentration was determined by spectrophotometric analysis using a NanoDrop ND-1000 spectrophotometer (Nanodrop Technologies Inc., 3411 Silverside Rd., Bancroft Building, Wilmington, DE 19810, USA). A total of 15 µl plasma was diluted in 85 µl absolute ethanol. After 1 min of vortex, samples were centrifuged at 1500 g for 10 min to precipitate flocculent proteins. The absorbance of the supernatant was measured at the carotenoid peak at 450 nm in duplicate for each sample. Linearity was confirmed in a dilution series (1:2 1:16) of four plasma samples. Total carotenoid concentration was calculated from the absorption coefficient (250 ml/(mg cm)) of carotenoids (Britton 1995). We had data for plasma carotenoid level only for females in 2005. STATISTICAL ANALYSES Laying date was entered as a deviation from the median laying date of the particular year. Age was a binary variable for males (1 year old and older), as there is a distinct difference in the breeding plumage (colour of remiges and size of wing patch) between subadult and adult males (Svensson 1992). Male age was initially included in the analyses of wing patch size to test for possible interaction between male age and wing patch size. Furthermore, wing patch size was standardized by male age. Male forehead patch size was also standardized by year and male age. Females were grouped into two age classes (young: 1 or 2 years old; old: 3 years old or older) on the basis of ringing data to analyse the possible effect of senescence on egg investment. To account for variation related to years, female tarsus length and plasma IgG concentration were standardized across years. Standardization across years was carried out by subtracting the mean value in the particular year from the actual value and dividing the result with the standard deviation for that year. As haematocrit value correlated with the length of total blood in the capillary, we used the residuals of the regression of haematocrit on blood length as a corrected haematocrit value. Ln-transformed H/L ratios and heterophil numbers were used in order to obtain normal distribution of data. Interclutch variation in yolk IgG concentration was analysed by linear mixed models including nest identity as random factor. Year, clutch size and parental age were used as categorical predictor variables, all other variables were entered as continuous variables. Sample sizes in different clutch size groups were 10, 28 and 35 for five, six and seven egg clutches, respectively. Sample sizes in female age groups were 11 and 13, while in male age groups they were 20 and 47. Year effect and its interaction with the predictor variable were always included in the initial model; no other interactions were considered in this study. Furthermore, laying order was entered in the initial model as fixed factor to control for its possible effect. In all models, a stepwise backward selection procedure was employed, removing non-significant interactions and effects from the model one by one in decreasing order of P. Separate analyses were performed with each independent variable, as entering them into one model would have greatly decreased the degrees of freedom. The effect of egg-laying order on egg immunoglobulin concentration was analysed in separate tests based on data only from 2001 and 2002 by linear mixed models including year, laying order and year by laying order interaction as fixed effects and nest identity as random effect. As in 2004 and 2005 only two eggs per clutch were collected, and as intraclutch variation may be different among years, we decided not to pool data from these years with data from 2001 and 2002. Eggs within a clutch were ranked into one of the following categories: first, second, middle (third or the mean of

833 Immunoglobulins in Collared Flycatcher eggs Fig. 1. Year-standardized means of yolk antibody concentration of Collared Flycatcher clutches in relation to female body condition. Fig. 2. Clutch mean of yolk antibody concentration (relative to chicken IgG, see Methods) of Collared Flycatcher eggs in relation to heterophil : lymphocyte ratio of the female in 2005. Fig. 3. Clutch mean of yolk antibody concentration (relative to chicken IgG, see Methods) of Collared Flycatcher eggs in relation to absolute heterophil count of the female in 2005. 3 4 or 3 5), penultimate and last egg. All tests were two-tailed and differences were considered significant at P < 0 05. Data were analysed with STATISTICA, version 5 5 (StatSoft Inc., Tulsa, OK) and SPSS 13 0 (SPSS Inc., 2335 Wacker Dr., Chicago, Illinois, 60606, USA). Results Our results revealed that clutch mean concentration of yolk antibodies positively correlated with female body condition (F 1,41 28 = 9 91, P = 0 003, estimate = 5 20, Fig. 1). We also found that the plasma immunoglobulin level of females marginally significantly positively correlated with the bird s body condition (GLM, F 1,30 = 3 41, P = 0 075, R = 0 29), suggesting that high antibody level could be a characteristic of a high-quality female. In accord with previous results (Kowalczyk et al. 1985), the concentration of yolk IgG (mean ± SD, 0 63 ± 0 22 mg ml 1, chicken IgG equivalent) was comparable to that in the maternal plasma (mean ± SD, 1 02 ± 0 34 mg ml 1, chicken IgG equivalent). The relationship between the concentration of IgG in maternal circulation and mean concentration of IgG in the clutch differed between years (year plasma IgG level interaction: F 1,27 41 = 3 90, P = 0 058): in 2004, there was no significant correlation, while in 2005, a significant positive correlation could be detected (2004: F 1,10 96 = 0 76, P = 0 40; 2005: F 1,17 = 5 01, P = 0 039, estimate = 0 16). No significant correlation between plasma levels of immunoglobulins and total carotenoids could be found (GLM, F 1,15 = 0 074, P = 0 79). Females with higher H/L ratios and higher heterophil counts laid clutches with lower level of antibodies (H/L ratios: F 1,17 = 9 04, P = 0 008, estimate = 54 28, Fig. 2; heterophil count: F 1,17 = 4 64, P = 0 046, estimate = 34 75, Fig. 3). In contrast, no relationships between the haematocrit value, lymphocyte count and plasma carotenoid concentration of the female and mean yolk antibody concentration could be demonstrated (haematocrit: F 1,31 77 = 0 001, P = 0 97; lymphocyte count: F 1,16 64 = 0 03, P = 0 87; plasma carotenoid concentration: F 1,16 19 = 0 07, P = 0 80). Furthermore, there were no significant relationships between mean antibody concentration of the clutch and maternal age or body size (age: F 1,16 86 = 0 66, P = 0 43; body size: F 1,68 43 = 1 90, P = 0 17). The possible associations between yolk antibody investment and male characteristics including age, forehead patch size and wing patch size were also analysed, but no significant relationship could be detected (age: F 1,59 21 = 0 29, P = 0 59; forehead patch size: F 1,63 93 = 1 76, P = 0 19; wing patch size: F 1,65 73 = 2 18, P = 0 14). There was no significant seasonal variation in clutch mean of yolk immunoglobulin concentration (F 1,66 56 = 0 64, P = 0 43). Furthermore, neither clutch size (5 7 eggs, F 2,75 00 = 0 43, P = 0 65), nor egg volume (F 1,212 42 = 0 88, P = 0 35) affected the antibody level of Collared Flycatcher eggs. Analysing the within-clutch pattern of clutches from 2001 and 2002, we found that concentration of yolk antibodies increased with laying order in both years (F 4,137 67 = 3 39, P = 0 011, Fig. 4; year laying order interaction: F 4,136 86 = 0 59, P = 0 67), with the last-laid egg having the highest immunoglobulin level. A post-hoc test (LSD test) indicated significant differences between the last egg and the other eggs within a clutch (last and

834 R. Hargitai et al. Fig. 4. Mean (± SE) yolk antibody concentration (relative to chicken IgG, see Methods) of Collared Flycatcher eggs in relation to the position within the laying sequence (n = 33 clutches). first: P = 0 003; last and second: P = 0 023; last and middle: P = 0 001; last and penultimate: P = 0 016). To determine whether the increase in yolk antibody level with laying sequence was related to environmental and parental variables, separate tests were performed with each predictor variable, entering year, laying order, the predictor variable in question, and the interaction between the predictor variable and laying order as fixed effects and nest identity as random effect in the model. We found that the within-clutch variation was independent of laying date (laying order laying date interaction: F 4,128 85 = 0 95, P = 0 44), clutch size (F 8,129 65 = 0 57, P = 0 80), male characteristics (forehead patch size: F 4,127 94 = 0 92, P = 0 45; wing patch size: F 4,128 96 = 1 50, P = 0 21; age: F 4,128 82 = 0 57, P = 0 86), and female body size (F 4,112 75 = 1 28, P = 0 28). The interactions with maternal age and condition could not be analysed because of small sample sizes. Discussion In this study, we have shown that concentration of maternally derived yolk antibodies of Collared Flycather eggs was related to the condition of the female, irrespective of the female s body size or age. The production and transmission of antibodies to eggs may represent a significant immunological and resource drain for the laying bird, since each egg contains 10 20% of the female s steady-state immunoglobulin level (Kowalczyk et al. 1985). Females may compensate for this loss by increasing antibody production in order to sustain their own baseline level (Klasing 1998; Saino et al. 2001a). Up-regulation of the maternal immune system may be nutritionally costly and condition-dependent as it relies on the availability of specific nutrients and minerals (Jackson, Law & Nockels 1978; Klasing 1998; Møller et al. 1998; Alonso-Alvarez & Tella 2001). Thus, only females in good condition could afford to produce and allocate large amounts of antibodies to their eggs, possibly because they were able to divert more energy from self-maintenance to antibody production without compromising the prospects of a successful nestling rearing. In line with this, we found an almost significant positive association between plasma immunoglobulin concentration and body condition of females. The interpretation of high plasma antibody level is controversial as it can be a reflection of either prior exposure to infection (Gustafsson et al. 1994; Ots & Hõrak 1998) or good immune capacity (Saino et al. 2001b; Christe et al. 2001; Morales et al. 2004). Our results, in accordance with those of Saino et al. (2001a), suggest that high antibody level could be a characteristic of a female of high phenotypic quality. Nevertheless, it is also possible that the higher antibody transmission of females in better body condition was the result of that they were suffering from an infection and thus produced immunoglobulins at a higher rate. Positive association between parasite infection and body mass has been reported in Great Tits (Parus major, Ots & Hõrak 1998); however, in that study, birds were captured during the nestling feeding period, when higher body mass may represent a disadvantage for the breeding bird (Norberg 1981; Hillström 1995), while we measured the condition of females during incubation. We are also inclined to consider that females producing eggs with higher antibody level were of better health state, as leucocyte indices also suggested such a relationship. In contrast to our results, Grindstaff, Demas & Ketterson (2005) reported no effect of food restriction or female body mass on either maternal plasma or egg yolk IgG concentration. The difference between our results and theirs could originate from the facts that they manipulated only protein level of food and their study was conducted in captivity with a precocial species. High values of heterophil : lymphocyte (H/L) ratios are traditionally regarded as a reliable indicator of stress both in chicken (Gross & Siegel 1983; Maxwell 1993) and free living bird studies (Hõrak, Ots & Murumägi 1998; Ilmonen et al. 2003; Suorsa et al. 2004). The H/ L ratios are known to be enhanced by several types of stress factors including nutritional, climatic and social stress (Maxwell 1993). We found that elevated H/L ratios of females were related to lower amounts of maternal antibodies in the eggs, suggesting that stressed females had a limited quantity of energy or nutrients relative to the demands of antibody transmission to the eggs. Furthermore, stress is known to enhance the plasma level of glucocorticoid steroids (Sapolsky 1992), which in turn induces immunosuppression (Besedovsky & del Rey 1996; Deerenberg et al. 1997; Råberg et al. 1998). Thus, high level of stress could have resulted in a lower rate of antibody production. We also observed that higher number of heterophils were associated with lower levels of yolk antibodies. As heterophilia may indicate infection or inflammation (Maxwell 1993; Harmon 1998; Saks, Ots & Hõrak 2003), our results suggest that the weak health status of the female may have restricted the amount of antibodies it could transfer to the eggs.

835 Immunoglobulins in Collared Flycatcher eggs In agreement with earlier results (Brown et al. 1989; Blount et al. 2002; Grindstaff et al. 2005), we found a positive correlation between yolk immunoglobulin level and the immunoglobulin concentration in the mother s plasma, though the relationship could not be demonstrated in both studied years. Plasma IgG turnover is very rapid in birds, being about 36 h in domestic hen (Patterson et al. 1962), so it is possible that by the time we sampled the incubating female, the relationship disappeared in 2004. The positive relationship in 2005 supports the findings of poultry studies showing that yolk antibody level reflect the plasma antibody level of the laying female. Carotenoids act as regulators and stimulators of immune responses, and furthermore they prevent immune cells from free radical attack (Møller et al. 2000; Chew & Park 2004); thus we may expect a positive relationship between levels of carotenoids and immunoglobulins in the plasma. However, plasma carotenoid level of Collared Flycatcher females showed no association with either circulating immunoglobulin concentration or clutch mean of yolk antibody level. This observation differs from the results of studies of Lesser Black-Backed Gulls (Larus fuscus; Blount et al. 2002) and Barn Swallows (Hirundo rustica; Saino et al. 1999), in which higher plasma carotenoid levels were related to lower plasma immunoglobulin concentrations, although these results are inconsistent with the hypothesis that higher plasma antibody level reflects better immunocompetence. The concentration of immunoglobulins was unrelated to clutch size, which implies that females laying more eggs also had a higher production of immunoglobulins to meet the demands of producing a larger clutch. In the Barn Swallow, Saino et al. (2002b) also could not find a statistically significant covariation between clutch size and yolk antibody concentration. The differential allocation hypothesis suggests that mothers should invest more in reproduction when mated to a high-quality or attractive male because their sons will be preferred as mates or because their offspring will inherit the good genes of their father (Burley 1986; Gil et al. 1999; Saino et al. 2002b; Rutstein et al. 2004). Up to the present, only one study has examined the association between the characteristics of the male and the immunological quality of the eggs, showing higher deposition of antibodies to the eggs of better-quality males (Saino et al. 2002b). In the Collared Flycatcher, the male s attractiveness or quality may be reflected by the bird s sexually selected ornamentation (Hegyi, Török & Tóth 2002; Michl et al. 2002; Török et al. 2003) and age (Hegyi, Rosivall & Török 2006b). However, we found no significant evidence that females adjusted their immunoglobulin investment according to the characteristics of their mate. Hence, no support for the differential allocation of antibodies was found in this species. In broods of Collared Flycatchers, nestlings hatch asynchronously (Rosivall et al. 2005), and as a consequence, the youngest nestling is at an initial disadvantage. The last-hatched nestling is expected to experience higher level of stress than its siblings owing to its lower competitive ability for limiting food resources (Price & Ydenberg 1995; Cotton, Wright, & Kacelnik 1999; Saino et al. 2001b), which could impair its health state. Females allocated elevated concentration of immunoglobulins to the last-laid egg, which could be an adaptive mechanism to increase the viability and resistance to pathogens of the last-hatching young. Accordingly, in an earlier study of Barn Swallows, higher concentration of immunoglobulins, which could be of maternal origin, has been reported in late-hatched nestlings (Saino et al. 2001b), and was related to a larger T-cell mediated immune response. A high level of maternal antibodies may also influence offspring growth rate by decreasing the physiological expenses of mounting an immune response, and thus leaving more nutrients for growth and development. For example, it has been demonstrated that Great Tit nestlings of ectoparasite-exposed mothers grew faster and had a higher chance to survive probably because of the increased maternal antibody transfer to the eggs (Heeb et al. 1998; Buechler et al. 2002). Thus, a high amount of maternally derived antibodies may allow the last-hatching nestling to resist infection without the physiological and growth-retarding costs of the stimulation of the immune system. Elevated early growth rates may help the smallest nestling to catch up with its siblings, increasing the probability that the whole brood survives after fledging (Lindén, Gustafsson & Pärt 1992). In some gull species, yolk antibody concentration was found to decrease through laying sequence, with first eggs having the highest values (Blount et al. 2002; Müller et al. 2004). This allocation pattern could represent a maternal strategy to facilitate brood reduction in gulls, consistent with the decline in carotenoid concentration and egg size with the laying order of eggs (Royle et al. 1999; Blount et al. 2002). In the Collared Flycatcher, however, previous studies have shown that the concentration of β-carotene (J. Török, R. Hargitai, G. Hegyi, Z. Matus, G. Michl, P. Péczely, B. Rosivall, & Gy. Tóth, unpublished data) and egg volume (Cichon 1997; Hargitai et al. 2005; Rosivall et al. 2005) increase along the laying sequence within a clutch, providing support for the idea that Collared Flycatchers pursue a brood-survival strategy. In conclusion, our findings support the prediction that antibody production and transfer to the egg yolk is nutritionally costly in birds, as only females of better physiological condition could afford to lay eggs with relatively higher concentration of immunoglobulins. In contrast, no evidence for differential allocation of yolk immunoglobulins in relation to the attractiveness or quality of the male mate could be detected. Antibody level was higher in the last-laid egg, which may be interpreted as a maternal mechanism to improve the immunocompetence and survival probability of

836 R. Hargitai et al. the last-hatching, and thus competitively disadvantaged, nestling. Future work should be focused upon the identification of the fitness consequences of higher maternal antibody allocation to offspring in wild birds. Acknowledgements We are grateful to L. Z. Garamszegi, G. Hegyi, M. Herényi, G. Michl, B. Rosivall, B. Szigeti, E. Szöllosi and numerous students for assistance during fieldwork. We are indebted to M. Pihlaja, J. Sendecka, H. Siitari and E. Virtanen for helpful suggestions for antibody analysis, and Zs. Greff for invaluable help in the laboratory. We thank H. Siitari and two anonymous reviewers for helpful comments on earlier version of the manuscript. This study was financially supported by the Hungarian Scientific Research Fund (OTKA, grants T49650 and T49678), the Hungarian Ministry of Education (grants no. FKFP 0304/2000 and 0021/2002), the Eötvös Loránd University and the Pilis Park Forestry. The Hungarian Ministry of Environment provided permissions for the collection of Collared Flycatcher eggs (permissions 1255/2/2001, 2256 3/2002, 2573/2/2004, 15951/2005). References Alonso-Alvarez, C. & Tella, J.L. (2001) Effects of experimental food restriction and body-mass changes on avian T-cell mediated immune response. Canadian Journal of Zoology 79, 101 105. Apanius, V. (1998) Ontogeny of immune function. Avian Growth and Development: Evolution Within the Altricital Precocial Spectrum (eds J.M. Starck & R.E. Ricklefs), pp. 203 222. Oxford University Press, New York. Barua, A., Yoshimura, Y. & Tamura, T. (1998) Effects of ageing and oestrogen on the localization of immunoglobulincontaining cells in the chicken ovary. Journal of Reproduction and Fertility 114, 11 16. Besedovsky, H.O. & del Rey, A.E. (1996) Immune neuroendocrine interactions: facts and hypotheses. Endocrine Reviews 17, 64 97. Blount, J.D., Surai, P.F., Nager, R.G., Houston, D.C., Møller, A.P., Trewby, M.L. & Kennedy, M.W. (2002) Carotenoids and egg quality in the lesser black-backed gull Larus fuscus: a supplemental feeding study of maternal effects. Proceedings of the Royal Society London Biology Sciences B 269, 29 36. Britton, G. (1995) UV/visible spectroscopy. Carotenoids: Spectroscopy (eds G. Britton, S. Liaaen-Jensen & H. Pfander), pp. 13 63. Birkhäuser-Verlag, Basel. Brown, J., Resurreccion, R.S., Dickson, T.G. & Horne, A. (1989) The relationship of egg yolk and serum antibody. I. Infectious bursal disease virus. Avian Diseases 33, 654 656. Buechler, K., Fitze, P.S., Gottstein, B., Jacot, A. & Richner, H. (2002) Parasite-induced maternal response in a natural bird population. Journal of Animal Ecology 71, 247 252. Burley, N. (1986) Sexual selection for aesthetic traits in species with biparenal care. American Naturalist 127, 415 445. Campbell, T.W. (1988) Avian Hematology and Cytology. Iowa State University Press, Ames. Carlier, Y. & Truyens, C. (1995) Influence of maternal infection on offspring resistance towards parasites. Parasitology Today 11, 94 99. Chew, B.P. & Park, J.S. (2004) Carotenoid action on the immune response. Journal of Nutrition 134, 257S 261S. Christe, P., de Lope, F., Gonzalez, G., Saino, N. & Møller, A.P. (2001) The influence of environmental conditions on immune responses, morphology and recapture probability of nestling house martins (Delichon urbica). Oecologia 126, 333 338. Cichon, M. (1997) Egg weight variation in Collared Flycatcher Ficedula albicollis. Ornis Fennica 74, 141 147. Clark, A.B. & Wilson, B.S. (1981) Avian breeding adaptations: hatching asynchrony, brood reduction and nest failure. Quarterly Review of Biology 56, 257 277. Cotton, P.A., Wright, J. & Kacelnik, A. (1999) Chick begging strategies in relation to brood hierarchies and hatching asynchrony. American Naturalist 153, 412 420. Deerenberg, C., Apanius, V., Daan, S. & Bos, N. (1997) Reproductive effort decreases antibody responsiveness. Proceedings of the Royal Society London Biology Sciences B 264, 1021 1029. Edge, R., McGarvey, D.J. & Truscott, T.G. (1997) The carotenoids as anti-oxidants a review. Journal of Photochemistry and Photobiology B 41, 189 200. Eising, C.M., Eikenaar, C., Schwabl, H. & Groothuis, T.G.G. (2001) Maternal androgens in black-headed gull (Larus ridibundus) eggs: consequences for chick development. Proceedings of the Royal Society London Biology Sciences B 268, 839 846. Fadly, A.M. & Nazerian, K. (1989) Hemorrhagic enteritis of turkeys: influence of maternal antibody and age at exposure. Avian Diseases 33, 779 786. Fudge, A.M. (2000) Laboratory Medicine Avian and Exotic Pets. W.B. Saunders Company, Philadelphia. Garamszegi, L.Z., Rosivall, B., Hegyi, G., Szöllosi, E., Török, J. & Eens, M. (2006) Determinants of male territorial behavior in a Hungarian collared flycatcher population: plumage traits of residents and challengers. Behavioral Ecology and Sociobiology in press. DOI: 10.1007/500265-006-0210-4 Gasparini, J., McCoy, K.D., Haussy, C., Tveraa, T. & Boulinier, T. (2001) Induced maternal response to the Lyme disease spirochaete Borrelia burgdorferi sensu lato in a colonial seabird, the kittiwake Rissa tridactyla. Proceedings of the Royal Society London Biology Sciences B. 268, 647 650. Gil, D., Graves, J., Hazon, N. & Wells, A. (1999) Male attractiveness and differential testosterone investment in zebra finch eggs. Science 286, 126 128. Graczyk, T.K., Cranfield, M.R., Shaw, M.L. & Craig, L.E. (1994) Maternal antibodies against Plasmodium spp. in African black-footed penguin (Sphenicus demersus). Journal of Wildlife Disease 30, 365 371. Grindstaff, J.L., Brodie, E.D. III. & Ketterson, E.D. (2003) Immune function across generations: integrating mechanism and evolutionary process in maternal antibody transmission. Proceedings of the Royal Society London B. 270, 2309 2319. Grindstaff, J.L., Demas, G.E. & Ketterson, E.D. (2005) Diet quality affects egg size and number but does not reduce maternal antibody transmission in Japanese quail Coturnix japonica. Journal of Animal Ecology 74, 1051 1058. Gross, W.B. & Siegel, H.S. (1983) Evaluation of the heterophil/ lymphocyte ratio as a measure of stress in chickens. Avian Diseases. 27, 972 979. Gustafsson, L., Nordling, D., Andersson, M.S., Sheldon, B.C. & Qvarnström, A. (1994) Infectious diseases, reproductive effort and the cost of reproduction in birds. Philosophical Transactions of the Royal Society of London B 346, 323 331. Gustafsson, L., Qvarnström, A. & Sheldon, B.C. (1995) Trade-offs between life history traits and a secondary sexual character in male collared flycatchers. Nature 375, 311 313.

837 Immunoglobulins in Collared Flycatcher eggs Hargitai, R., Török, J., Tóth, L., Hegyi, G., Rosivall, B., Szigeti, B. & Szöllosi, E. (2005) Effects of environmental conditions and parental quality on the inter- and intraclutch egg size variation in the collared flycatcher (Ficedula albicollis). Auk 122, 509 522. Harmon, B.G. (1998) Avian heterophils in inflammation and disease resistance. Poultry Science 77, 972 977. Harrison, G.J. & Harrison, L.R. (1986) Clinical Avian Medicine and Surgery. W.B. Saunders, London. Heeb, P., Werner, I., Kölliker, M. & Richner, H. (1998) Benefits of induced host responses against an ectoparasite. Proceedings of the Royal Society London B. 265, 51 56. Hegyi, G., Török, J. & Tóth, L. (2002) Qualitative population divergence in proximate determination of a sexually selected trait in the collared flycatcher. Journal of Evolutionary Biology 15, 710 719. Hegyi, G., Török, J., Tóth, L., Garamszegi, L.Z. & Rosivall, B. (2006a) Rapid temporal change in the expression and age-related information content of a sexually selected trait. Journal of Evolutionary Biology 19, 228 238. Hegyi, G., Rosivall, B. & Török, J. (2006b) Paternal age and offspring growth: environmental or genetic effects? Behavioral Ecology and Sociobiology in press. DOI: 10.1007/500265-006-0211-3 Hillström, L. (1995) Body mass reduction during reproduction in the pied flycatcher Ficedula hypoleuca: physiological stress or adaptation for lowered costs of locomotion. Functional Ecology 9, 807 817. Hoi-Leitner, M., Romero-Pujante, M., Hoi, H. & Pavlova, A. (2001) Food availability and immune capacity in serin (Serinus serinus) nestlings. Behavioral Ecology and Sociobiology 49, 333 339. Hõrak, P., Ots, I. & Murumägi, A. (1998) Haematological health state indices of reproducing Great Tits: a response to brood size manipulation. Functional Ecology 12, 750 756. Hõrak, P., Surai, P.F. & Møller, A.P. (2002) Fat-soluble antioxidants in the eggs of great tits Parus major in relation to breeding habitat and laying sequence. Avian Science 2, 123 130. Hoyt, D.F. (1979) Practical methods of estimating volume and fresh weight of bird eggs. Auk 96, 73 77. Ilmonen, P., Hasselquist, D., Langefors, Å. & Wiehn, J. (2003) Stress, immunocompetence and leukocyte profiles of pied flycatchers in relation to brood size manipulation. Oecologia 136, 148 154. Jackson, D.W., Law, G.R. & Nockels, C.F. (1978) Maternal vitamin E alters passively acquired immunity of chicks. Poultry Science 57, 70 73. Klasing, K.C. (1998) Nutritional modulation of resistance to infectious diseases. Poultry Science 77, 1119 1125. Kowalczyk, K., Daiss, J., Halpern, J. & Roth, T.F. (1985) Quantitation of maternal-fetal IgG transport in the chicken. Immunology 54, 755 762. Lack, D. (1968) Ecological Adaptations for Breeding in Birds. Methuen, London. Lemke, H. & Lange, H. (1999) Is there a maternally induced immunological imprinting phase à la Konrad Lorenz? Scandinavian Journal of Immunology 50, 348 354. Lindén, M., Gustafsson, L. & Pärt, T. (1992) Selection on fledging mass in the collared flycatcher and the great tit. Ecology 73, 336 343. Lipar, J.L. & Ketterson, E.D. (2000) Maternally derived yolk testosterone enhances the development of the hatching muscle in the red-winged blackbird Agelaius phoeniceus. Proceedings of the Royal Society London B 267, 2005 2010. Lipar, J.L., Ketterson, E.D. & Nolan, V. Jr (1999) Intra-clutch variation in testosterone content of red-winged blackbird eggs. Auk 116, 231 235. Lochmiller, R.L. & Deerenberg, C. (2000) Trade-offs in evolutionary ecology: just what is the cost of immunity? Oikos 88, 87 98. Lozano, G.A. & Ydenberg, R.C. (2002) Transgenerational effects of maternal immune challenge in tree swallows (Tachycineta bicolor). Canadian Journal of Zoology 80, 918 925. Maxwell, M.H. (1993) Avian blood leucocyte responses to stress. World s Poultry Science Journal 49, 34 43. Merino, S. & Potti, J. (1998) Growth, nutrition and blowfly parasitism in nestling pied flycatchers. Canadian Journal of Zoology 76, 936 941. Michl, G., Török, J., Griffith, S.C. & Sheldon, B.C. (2002) Experimental analysis of sperm competition mechanisms in a wild bird population. Proceedings of the National Academy of Sciences of the USA 99, 5466 5470. Møller, A.P., Christe, P., Erritzøe, J. & Mavarez, J. (1998) Condition, disease and immune defence. Oikos 83, 301 306. Møller, A.P., Biard, C., Blount, J.D., Houston, D.C., Ninni, P., Saino, N. & Surai, P.F. (2000) Carotenoid-dependent signals: indicators of foraging efficiency, immunocompetence or detoxification ability? Avian and Poultry Biology Reviews 11, 137 159. Morales, J., Moreno, J., Merino, S., Tomás, G., Martínez, J. & Garamszegi, L.Z. (2004) Association between immune parameters, parasitism, and stress in breeding pied flycatcher (Ficedula hypoleuca) females. Canadian Journal of Zoology 82, 1484 1492. Mousseau, T.A. & Fox, C.W. (1998) Maternal Effects as Adaptations. Oxford University Press, New York. Müller, W., Groothuis, T.G.G., Dijkstra, C., Siitarni, H. & Alatalo, R.V. (2004) Maternal antibody transmission and breeding densities in the Black-Headed Gull Larus ridibundus. Functional Ecology 18, 719 724. Norberg, R.Å. (1981) Temporary weight decrease in breeding birds may result in more fledged young. American Naturalist 118, 838 850. Ots, I. & Hõrak, P. (1998) Health impact of blood parasites in breeding Great Tits. Oecologia 116, 441 448. Pastoret, P., Gabriel, P., Bazin, H. & Govaerts, A. (1998) Handbook of Vertebrate Immunology. Academic Press, San Diego, CA. Patterson, R., Younger, J.S., Weigle, W.O. & Dixon, F.J. (1962) The metabolism of serum proteins in the hen and chick and secretion of serum proteins by the ovary of the hen. Journal of General Physiology 45, 501 513. Potti, J., Moreno, J., Merino, S., Frias, O. & Rodriguez, R. (1999) Environmental and genetic variation in the haematocrit of fledgling pied flycatchers Ficedula hypoleuca. Oecologia 120, 1 8. Price, K. & Ydenberg, R. (1995) Begging and provisioning in broods of asynchronously-hatched yellow-headed blackbird nestlings. Behavioral Ecology and Sociobiology 37, 201 208. Qvarnström, A. (1999) Genotype by environment interactions in the determination of the size of a secondary sexual character in the collared flycatcher (Ficedula albicollis). Evolution 53, 1564 1572. Råberg, L., Grahn, M., Hasselquist, D. & Svensson, E. (1998) On the adaptive significance of stress-induced immunosuppression. Proceedings of the Royal Society London B 265, 1637 1641. Rose, M.E., Orlans, E. & Buttress, A. (1974) Immunoglobulin classes in the hen s egg: their segregation in yolk and white. European Journal of Immunology 4, 521 523. Rosivall, B., Szöllosi, E. & Török, J. (2005) Maternal compensation for hatching asynchrony in the collared flycatcher. Journal of Avian Biology 36, 531 537. Royle, N.J., Surai, P.F., McCartney, R.J. & Speake, B.K. (1999) Parental investment and egg yolk lipid composition in gulls. Functional Ecology 13, 298 306.