Behavioral Ecology Advance Access originally published online on July 28, 2004
Behavioral Ecology 2005 16(1):62-69; doi:10.1093/beheco/arh123
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sex-specific patterns of yolk androgen allocation depend on maternal diet in the zebra finch
School of Biology, University of St. Andrews, Fife, KY16 9TS, UK
Address correspondence to A. Rutstein. E-mail: ar27{at}st-andrews.ac.uk.
Received 13 November 2003; revised 28 March 2004; accepted 2 May 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Females are predicted to adjust their reproductive investment in relation to resource quality. In zebra finches (Taeniopygia guttata), diet quality has been found to influence egg mass both between and within clutches. We tested the prediction that diet quality also affects the quantity of maternally allocated yolk testosterone and 5
-dihydrotestosterone (DHT) between and within clutches. We also investigated whether this pattern differed between male and female eggs. Females laid eggs on a high-quality (HQ) or a low-quality (LQ) diet. Eggs were removed at laying and artificially incubated for 72 h, after which time embryos were sexed and yolk androgens assayed. Diet treatments were then swapped and the experiment repeated. Because there was evidence of a carry-over effect between breeding rounds, we based our conclusions mainly on the results from the first breeding round. On the HQ diet, but not on the LQ diet, infertile eggs contained more testosterone than did fertile eggs in round one. Although there were no overall differences in within-clutch patterns of androgen deposition between the diets, this changed when embryo sex was taken into account. On the HQ diet, testosterone decreased with laying sequence for male eggs but increased with laying sequence for female eggs. On the LQ diet, mothers' male eggs contained more testosterone and DHT than did female eggs regardless of position in the laying sequence. Our data suggest that there are complex, context-dependent mechanisms of sex-specific androgen allocation in this species. Key words: diet quality, sex allocation, Taeniopygia guttata, yolk androgens, zebra finch.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Diet and maternal condition are key predictors of avian reproductive success. When resources are abundant, females should increase their reproductive investment because per unit costs are relatively low and per unit benefits relatively high (Hochachka, 1992
; Pilz et al., 2003). Such increased investment is typically reflected in larger, heavier clutches (Houston et al., 1983; Smith et al., 1993) but can also take more subtle forms of investment. Females may lay eggs containing greater quantities of carotenoids (Blount et al., 2002; Royle et al., 2001), lipids (Nager et al., 2000; Royle et al., 1999), or immunoglobulins (Saino et al., 2002). In addition, females may preferentially direct resources toward the more costly sex (Appleby et al., 1997).
Zebra finches (Taeniopygia guttata) are protein-limited during egg production (Houston et al., 1995; Williams, 1996), such that females on a high-quality (HQ), protein-rich diet lay more and larger eggs (Monaghan et al., 1996; Rutstein et al., 2004; Selman and Houston, 1996; Williams, 1996). Furthermore, eggs laid on a HQ diet have greater hatching success than do those laid on a low-quality (LQ) diet (Gorman and Nager, 2003; Selman and Houston, 1996). Diet quality may also affect the pattern of egg quality within the clutch. Egg mass increases over the clutch on a HQ diet but decreases over the clutch on a LQ diet (Rutstein et al., 2004). This within-clutch difference in egg mass may be adaptive in asynchronously hatching species. Zebra finches hatch over a couple of days in the wild, and this asynchrony is more pronounced in the laboratory (Zann, 1996). When conditions are favorable, an increase in egg mass over the clutch might represent a brood survival strategy (Clark and Wilson, 1981; Slagsvold et al., 1984), whereby females invest more in later eggs to offset the disadvantage to these chicks of hatching last and having to compete with older, larger siblings. When conditions are harsh, however, a decline in egg mass with laying sequence might represent a brood reduction strategy by exaggerating the last chick disadvantage (Alisaukas, 1986; O'Connor, 1979).
A second mechanism by which females might adjust the within-brood competitive hierarchy is through the differential allocation of yolk androgens. Lipid-soluble steroids pass from the mother into the yolk during vitellogenesis, and females could potentially control allocation to each egg. The amount of testosterone and 5-dihydrotestosterone (DHT) an egg receives relative to the rest of the clutch may have important fitness consequences. One study to date has found that experimentally increased yolk testosterone decreases hatching success and chick survival in the American kestrel (Falco sparverius: Sockman and Schwabl 2000). In contrast, studies of canaries (Serinus canaria), red-winged blackbirds (Agelaius phoeniceus), and black-headed gulls (Larus ridibundus) suggest that maternally derived yolk androgens enhance chick growth rates (Eising et al., 2001; Schwabl, 1996), begging vigor (Eising and Groothuis, 2003; Schwabl, 1996), the size of the hatching muscle mass (Lipar and Ketterson, 2000), and social dominance (Schwabl, 1993). Thus, yolk androgens are generally thought to be beneficial to offspring.
In the majority of species studied, yolk androgens have been found to increase with laying sequence: canaries (Schwabl, 1993), red-winged blackbirds (Lipar et al., 1999), American kestrels (Sockman and Schwabl, 2000), lesser black-backed gulls (Larus fuscus: Royle et al., 2001), common terns (Sterna hirundo: French et al., 2001), black-headed gulls (Eising et al., 2001, Groothuis and Schwabl, 2002), European starlings (Sturnus vulgaris: Pilz et al., 2003), and house sparrows (Passer domesticus: Mazuc et al., 2003; Schwabl, 1997). This has generally been interpreted as a maternal compensatory mechanism to mitigate the disadvantage to later-hatched chicks (Schwabl, 1993, 1996). Decreasing levels of yolk testosterone with laying sequence have so far been reported in only two species: cattle egrets (Bubulcus ibis: Schwabl et al., 1997) and zebra finches (Gil et al., 1999). Cattle egrets are a facultative siblicidal species, and so the provisioning of first laid eggs with extra testosterone was thought to facilitate efficient brood reduction by the oldest chick (Schwabl et al., 1997). Zebra finches are not siblicidal but live in arid areas where HQ food may be in short supply (Zann, 1996), so a mechanism to promote brood reduction might be advantageous. However, breeding conditions are favorable in certain breeding seasons, especially after heavy rains, in which case, brood reduction may not always be adaptive. Therefore, some degree of flexibility might be expected in this species.
Differences in patterns of yolk androgen allocation within species have been found in relation to a number of social and environmental factors. For example, levels of testosterone and DHT over the whole clutch were found to increase with perceived mate attractiveness in zebra finches (Gil et al., 1999) and with female quality (measured in terms of clutch size and laying date) in European starlings (Pilz et al., 2003). Gil et al. (1999) proposed that, if androgens are costly to either the laying female or her chicks (see Royle et al., 2001; Sockman and Schwabl, 2000), only HQ offspring may be able to withstand or be worth this greater investment.
Yolk androgens may also be deposited differentially in male and female eggs when, for example, the impact of a given unit of parental investment enhances the reproductive value of one offspring sex more than the other (Leimar, 1996; Trivers and Willard, 1973). In Leghorn chickens (Gallus gallus domesticus) dominant females were found to allocate more testosterone to male eggs, whereas subordinate females allocated more testosterone to female eggs (Müller et al., 2002). Sex allocation theory predicts such a pattern of resource allocation because in this polygynous species, males have higher condition-dependent variance in reproductive success than do females.
Because zebra finches are socially monogamous and size-monomorphic, differential investment in the sexes might not seem likely. However, on a restricted diet, male-biased sex ratios have been found in this species at laying (Rutkowska and Cichon, 2002; Rutstein et al., 2004) and at hatching (Bradbury and Blakey, 1998; Kilner, 1998). This is thought to be adaptive because female fecundity (clutch size) is strongly related to female weight at fledging (Haywood and Perrins, 1992) and because females have lower rates of growth and survival on a poor quality diet (Bradbury and Blakey, 1998; de Kogel, 1997; Kilner, 1998; Martins, 2004). It is possible that such differences in male and female growth rates are mediated through sex differences in yolk hormone levels or through sex differences in sensitivity to yolk hormone levels (Henry and Burke, 1999).
In this study, we investigated differences in maternally deposited yolk testosterone and DHT in relation to diet quality. If androgens are costly to the female or her offspring, we predicted that females on a HQ diet would invest more than do those on a LQ diet. Second, because diet quality affects the number and quality of young that females can rear, we also predicted that females on a HQ diet would deposit relatively more androgens in later eggs (brood survival strategy), whereas we predicted that females on a LQ diet would deposit relatively less androgens in later eggs (brood reduction strategy). Third, we investigated sex differences in yolk androgen levels. Because female reproductive success is more dependent on nestling resources than that of males, we predicted that female eggs would contain greater concentrations of yolk androgens on a HQ diet, and that male eggs would contain greater concentrations of androgens on a LQ diet.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects
All birds were bred in our laboratory or came from other UK universities or private breeders. They all had previous breeding experience, but had not bred within 6 months before the experiment.
Experimental setup
Forty females were randomly assigned to one of two diet groups for 6 weeks before breeding. Those on the HQ regime were given mixed seed (foreign finch mix containing white, Japanese, yellow, panicum, and canary millet (14% protein) supplied by Haith's), supplemented daily with Haith's egg biscuit (a dried egg mixture to which water is added, containing 13.2% protein and 3.6% oil), hard-boiled hen's egg (13.5% protein, fresh eggs), Haith's PTX (a dried crumb mixture containing 27.2% protein and 17.9% oil), and fresh spinach. Those on the LQ regime were given foreign finch mix ad libitum and fresh spinach three times a week. Both groups had access to cuttlebone and fresh water ad libitum.
After 6 weeks, females were paired randomly with males with whom they had not previously bred. Pairs were placed in individual breeding cages measuring 75 x 40 x 40 cm, equipped with nest-boxes and nesting material. For the 6 weeks before pairing, all males had been caged in single sex groups in a separate room and received the standard laboratory diet of foreign finch mix, supplemented with fresh spinach and Haith's egg biscuit once a week. Once paired, males were kept on the same HQ or LQ diet as their partner, and these diets were maintained during egg laying and incubation. All cages were located within rooms in the animal house and maintained on a 16-h light/8-h dark lighting schedule under full-spectrum lights.
We checked nests daily for eggs. Newly laid eggs were removed (and replaced with dummy eggs), weighed to the nearest 0.01 g, and numbered. Eggs were artificially incubated at 37.5°C for 72 h to obtain sufficient embryonic tissue for molecular sexing. Eggs were then stored at –20°C for later dissection.
After 1 month, pairs were separated and transferred to single sex groups. After a 2-week rest period (during which all birds received the standard laboratory diet), experimental diets were reversed. Females on the HQ diet in round one received the LQ diet, and vice-versa. Males remained on the standard laboratory diet. After 6 weeks, females were paired up with the same male as in the first round, and the experiment was repeated.
Egg dissection
The embryo was removed from fertile eggs and stored at –20°C for DNA extraction and sexing. The yolk was separated from the white and stored at –20°C for androgen extractions and assays. Eggs in which there was no sign of embryonic development were presumed to be infertile.
DNA extraction and sexing
We extracted DNA from embryonic tissue by using the Puregene DNA extraction kit (Gentra Systems). Molecular sexing was carried out by using methodology and primers modified from the method of Griffiths et al. (1998) and described in Rutstein et al. (2004).
Yolk androgen extraction
After separating from the whites and the embryos, whole yolks were weighed and homogenized with 0.5 ml water. The androgens were extracted twice by using diethyl ether, and the ether factions were separated by snap-freezing in a dry ice-acetone bath. Neutral lipids were precipitated in 2 ml 90% alcohol, leaving overnight at –20°C, centrifuging, and drying the supernatant. The dried extract was dissolved in 2.5 ml assay buffer. Of this, 1.5 ml was kept for androgen (testosterone and DHT) analysis. The remaining 1 ml underwent an oxidation step (following methodology in the Amersham radioimmunoassay kit), which involved adding 75 µl oxidation reagent (5% sodium periodate, 0.5% potassium permanganate by weight) and incubating at room temperature for 20 min. This oxidizes testosterone such that DHT alone remains, and a DHT only assay can be conducted on the final extract. This was followed by a further two extractions, and the dried extract homogenized in 1 ml assay buffer for DHT analysis.
Yolk androgen assay
We conducted radioimmunoassays (RIAs) to analyze the total (testosterone + DHT) and DHT content of extracted yolks. The methodology followed that in a RIA kit (Amersham), using 200 µl tritiated DHT (5 µl in 80 ml assay buffer), 200 µl extracted yolk sample, and 200 µl anti-testosterone rabbit antisera at 1/6000 dilution. Standard curves for total (testosterone + DHT) and DHT assays were produced by using testosterone (12.5–400 pg/assay tube) and DHT (25–800 pg/assay tube) standards, respectively. The intra-assay coefficient of variation was 4.3 ± 0.3% (mean ± SE) for testosterone and 6.3% ± 0.6% for DHT; the interassay coefficient of variation was 17 ± 2.2% for both. The extraction recovery of total (testosterone + DHT) androgens was 75.6 ± 9.0%; the extraction recovery of DHT only was 59 ± 0.9%. The cross-reactivity of the antisera was 46% (see Nash et al., 2000), so the testosterone content was estimated as total –(0.46 x DHT).
Statistical analyses
Total yolk androgens were calculated by multiplying initial yolk mass by androgen concentration. Because the development of an embryo disrupts the distinction between yolk and albumen, the initial egg yolk mass was estimated by using regression equations obtained from data from fresh eggs laid on HQ and LQ diets. The regression equations were: for HQ diet eggs, y = 0.028 + (0.145 x); for LQ diet eggs, y = 0.045 + (0.129 x), where y is yolk mass and x is egg mass (Rutstein, 2003). Nonnormal data were log-transformed (log[androgen value + 1]). Because values for androgen concentration and total androgen levels gave very similar results, statistical values are given only for androgen concentration.
Egg mass and androgen data were analyzed by using the repeated-measures mixed general linear model (GLM) procedure in SAS (SAS Institute 1998). Female identity was entered as a random factor, diet quality as a fixed treatment, and eggs within the clutch as a repeated measure. GLMs used a Kenward Roger correction for degrees of freedom, which is appropriate for small sample sizes (Kenward and Roger, 1997).
Sex ratio (number of males/total number of sexed eggs) was analyzed at the brood level by fitting a generalized linear model (GLM), using the GLIMMIX macro in SAS. The error distribution was binomial with a logit link, weighted by brood size. Embryo sex was analyzed in relation to position in the laying sequence and treatment by using logistic regression. In all models, nonsignificant terms were sequentially removed, starting with the least significant interaction term.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clutch size and egg mass
Thirty females laid clutches on both the HQ and LQ diets (17 females laid on the HQ diet in round one and on the LQ diet in round two, and 13 females laid on the LQ diet in round one and on the HQ diet in round two). Seven females did not lay on either diet, two laid only on the HQ diet, and one laid only on the LQ diet.
In a paired analysis for females that laid on both diets, females laid significantly larger clutches on the HQ diet (HQ diet: median = 6 eggs [interquartile range = 5–7], LQ diet: 4 eggs [2–4], Wilcoxon signed-rank tests, W = 325.0, p < .0001). In a repeated-measures, paired GLM, there was a significant interaction between position in the laying sequence and diet with respect to egg mass, both for females that laid on the HQ diet first and the LQ diet second (F1,144 = 5.95, p = .02) (Figure 1a) and for females that laid on the LQ diet first and the HQ diet second (F1,112 = 4.11, p = .045) (Figure 1b). Egg mass increased significantly with laying sequence for females that laid on the HQ diet in round one (F1,83 = 4.07, p = .047) and for females that laid on the HQ diet in round two (F1,65 = 5.47, p = .02). In contrast, egg mass tended to decrease with laying sequence for females that laid on the LQ diet in round two (F1,46 = 3.94, p = .053) and did not change for females that laid on the LQ diet in round one (F1,36 = 0.73, p = .39).
|
There was also a significant interaction between diet and breeding round with respect to egg mass (F1,58 = 12.83, p = .0007) (Figure 1). Females that were on the HQ diet in the first round laid heavier eggs than did females that were on the HQ diet in round two. Unexpectedly, they continued to lay heavier eggs on the LQ diet (their second experimental diet) compared with the females that laid on the LQ diet in round one, suggesting a carry-over effect. There was no difference in the mass of male and female eggs in either diet group (F1,127 = 0.78, p = .38).
Sex ratio
The sex ratio of eggs produced on each diet were as follows: 38 male/33 female (HQ diet round 1), 14 male/11 female (LQ diet round 1), 26 male/30 female (HQ diet round 2), and 18 male/19 female (LQ diet round 2). None of these ratios was significantly biased, based on G tests (p > .2 in all cases). Because eggs within a clutch may not be independent, the mean clutch sex ratio was analyzed in a paired GLM for 21 females that laid fertile clutches on both diets (14 females that were on the HQ diet and seven that were on the LQ diet in round one). There was no difference in the clutch sex ratio between the two diets (mean ± SE, HQ diet: 0.48 ± 0.07, LQ diet: 0.51 ± 0.08, F1,20 = 0.10, p = .76). There was no effect of order of treatment (F1,19 = 0.01, p = .92), and there was no effect of laying sequence on embryo sex in either diet (F1,108 = 2.04, p = .16).
Androgen analyses
Fertile versus infertile eggs
Factors influencing testosterone and DHT concentrations are summarized in Table 1. Both testosterone and DHT decreased significantly with laying sequence on both diets (Table 1). The interaction between diet and laying sequence was not significant for either testosterone (F1,225 = 0.55, p = .46) or DHT (F1,211 = 0.17, p = .68). There was a significant three-way interaction between diet, egg fertility, and breeding round for both testosterone concentration (Figure 2a) and DHT concentration (Figure 2b). This is because for females that laid on the HQ diet in round one (and therefore the LQ diet in round two), infertile eggs contained more testosterone (F1,125 = 14.59, p = .0002) (Figure 2a) and more DHT (F1,134 = 5.55, p = .02) (Figure 2b) than did fertile eggs. There was no effect of diet on testosterone (F1,124 = 1.24, p = .27) or DHT concentration (F1,133 = 0.86, p = .36), and no interaction between diet and egg fertility for testosterone (F1,123 = 0.30, p = .58) or DHT (F1,119 = 2.09, p = .15). However, for females that laid on the LQ diet in round one (and therefore the HQ diet in round two), there was no difference between infertile and fertile eggs for either testosterone (F1,105 = 0.08, p = .78) (Figure 2a) or DHT concentration (F1,107 = 0.05, p = .82) (Figure 2b). Again, there was no effect of diet on testosterone (F1,106 = 0.45, p = .51) or DHT (F1,108 = 1.01, p = .32) and no interaction between diet and egg fertility for testosterone (F1,104 = 0.09, p = .77) or DHT (F1,106 = 0.54, p = .46). Because of the strong carry-over effect, we also compared the androgen data just for the first breeding round using an unpaired, repeated-measures GLM. There was a significant interaction between diet and fertility for testosterone (F1,111 = 3.98, p = .049) but not for DHT (F1,122 = 0.18, p = .67). The significant interaction for testosterone was because infertile eggs laid on the HQ diet contained significantly more testosterone compared with infertile eggs laid on the LQ diet (F1,23 = 9.05, p = .006), but there was no difference between fertile eggs on the two diets (F1,88 = 0.76, p = .38) (Figure 2a).
|
|
Analyses of fertile eggs only: effects of sex and laying sequence
In a paired analysis, conducted for females that laid at least one fertile egg in both clutches, there was a significant interaction between diet, embryo sex and breeding round both for testosterone (F1,131 = 5.88, p = .02) and DHT (F1,138 = 13.65, p = .0003). For testosterone, but not DHT, there was also an interaction between laying sequence, embryo sex, and breeding round (F1,131 = 15.10, p = .0002).
For females that laid on the HQ diet in round one, and therefore the LQ diet in round two (n = 14), there was a significant interaction between diet, embryo sex, and position in the laying sequence for testosterone concentration (F1,73 = 9.77, p = .003) (Figure 3). In round one, namely, when these females were on the HQ diet (Figure 3a), testosterone in male eggs decreased with position in the laying sequence (F1,21 = 7.41, p = .01), whereas testosterone in female eggs increased with position in the laying sequence (F1,25 = 10.68, p = .003), such that there was a significant interaction between laying sequence and embryo sex (F1,44 = 23.02, p < .0001). In round two, that is, on the LQ diet (Figure 3b), testosterone tended to be higher in female eggs than in male eggs (F1,32 = 4.01, p = .054) and decreased with position in the laying sequence for both sexes (F1,21 = 4.56, p = .04). On both diets, DHT tended to be greater for female eggs than for male eggs (F1,82 = 3.59, p = .06) and decreased with position in the laying sequence (F1,83 = 5.69, p = .02). There was no interaction between diet and laying sequence with respect to DHT concentration (F1,84 = 1.02, p = .32).
|
For females that laid on the LQ diet in round one and, therefore the HQ diet in round two (n = 9), there was a significant interaction between diet and embryo sex for both testosterone (F1,71 = 4.28, p = .04) (Figure 4a) and DHT (F1,72 = 10.57, p = .002) (Figure 4b). When these females were on the LQ diet, that is, round one, male eggs contained higher concentrations of androgens than did female eggs (T: F1,18 = 4.43, p = .049, DHT: F1,18 = 5.59, p = .03), but on the HQ diet, that is, round two, there was no difference between male and female eggs (testosterone: F1, 53 = 0.03, p = .85, DHT: F1,51 = 0.21, p = .65). On both diets, androgens decreased with position in the laying sequence for testosterone (F1,71 = 9.24, p = .003) and DHT concentration (F1,72 = 25.37, p < .0001).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Levels of investment
Females laid more and heavier eggs on the HQ diet compared with on the LQ diet. This is not surprising because zebra finches are thought to be protein-limited during egg formation (Houston et al., 1995
; Williams, 1996). However, this pattern could also result from maternal differential allocation. When resources are limited, reproduction is more costly in terms of the female's body condition and future reproductive potential (Hochachka, 1992). In addition, fewer offspring can be reared (Perrins, 1970), and these offspring will be of lower reproductive value (Hussell, 1972).
The egg mass data suggest that there was a carry-over effect between breeding rounds. Females on the HQ diet in round one laid heavier eggs on the LQ diet in round two than did females that were on the LQ diet in round one. This may be because females were able to store some limiting nutrient such as an essential amino acid that could then be used over a longer period (Williams, 1996). In addition, females were not allowed to rear young and so may have maintained a higher body condition than if they had reared chicks. This carry-over effect was also evident in certain aspects of the androgen data. In round one, infertile eggs laid on the HQ diet contained higher concentrations of testosterone than did fertile eggs, but there was no difference in testosterone concentration between fertile and infertile eggs laid on the LQ diet. In round two this pattern was reversed, showing that females maintained their level of investment between breeding attempts. Because of this carry-over effect, we will restrict our interpretation to the first round of the experiment.
There are two possible explanations for why infertile eggs on the HQ diet contained higher concentrations of testosterone than did fertile eggs. The first possibility is that high yolk testosterone concentrations are deleterious (Sockman and Schwabl, 2000) and therefore resulted in the early death of the embryo. The second possibility is that embryo development caused a rapid reduction in yolk testosterone. Support for the latter hypothesis comes from work on Leghorn chicken eggs by Elf and Fivizzani (2002), who found significant declines in androgen levels as early as day one of embryonic development. In the current study, yolk androgens were assayed when embryos were 3 days old, suggesting that declines in androgen levels could be substantial. However, a study by Eising et al. (2003), also on Leghorn chickens, found no changes in testosterone levels with incubation period.
In contrast, we found no difference in testosterone concentration between fertile and infertile eggs on the LQ diet in round one. This may have been because LQ diet eggs had initially lower concentrations of testosterone than did HQ eggs, and embryonic development (and hence testosterone metabolism) was slower in these eggs (Eising et al., 2001). If greater testosterone allocation is costly to mothers, the higher testosterone concentration in infertile HQ diet eggs provides support for the first of our predictions, that females should elevate their investment when conditions are favorable. However, further work is needed to understand how incubation affects changes in yolk androgen levels in this species.
Females on the LQ diet might deposit lower levels of testosterone if doing so is costly in some way to the female but also if high levels of testosterone are deleterious to the embryo (Gil et al., 1999, Sockman and Schwabl, 2000), an effect to which lower-quality embryos (from smaller, less well-provisioned eggs) may be particularly vulnerable (Royle et al., 2001). A study in the black-headed gull found no evidence of metabolic costs in chicks that hatched from androgen-treated eggs (Eising et al., 2003), although other costs cannot be ruled out, such as elevated immunosuppression (Ketterson and Nolan, 1999). Groothuis and Schwabl (2002) found that in a colony of black-headed gulls, lighter clutches contained more androgens, and suggested that females were compensating for the lower nutritional quality of these eggs. In an experimental field study on lesser black-backed gulls, supplementary-fed females laid eggs with lower levels of androgens (Verboven et al., 2003). However, we found the opposite to be true, and our data are in agreement with the field study carried out by Pilz et al. (2003) on European starlings, who found that higher-quality females (those that produced earlier and larger clutches) laid eggs with higher androgen concentrations. Such differences between studies are likely to reflect the complexity of the situation of which we currently have little understanding. Whether androgens are beneficial or deleterious to the female and/or embryo is likely to be dependent on a number of interacting factors such as the characteristics of the egg, including its size, position in the laying sequence, and sex (Groothuis and Schwabl, 2002), as well as the trade-offs operating on the laying females, which will be fundamentally different for long- and short-lived species (Linden and Møller, 1989).
Within-clutch patterns of egg mass and androgen deposition
We found a difference between diets in the pattern of within-clutch egg mass. Egg mass increased with laying sequence on the HQ diet, but not on the LQ diet, possibly reflecting an adaptive strategy by the female to influence the competitive hierarchy within the brood (Slagsvold et al., 1984). Heavier eggs toward the end of the laying sequence on the HQ diet would benefit these later hatching chicks and increase their chances of survival. However, we found no evidence for our second prediction that yolk androgen concentrations would also differ with laying sequence on the two diets. Overall, testosterone and DHT declined with position in the laying sequence on both diets, as described previously for this species (Gil et al., 1999), although when embryo sex was taken into account, a more complex pattern emerged.
In line with our third prediction, we found different patterns of yolk androgen concentrations in male and female depending on diet quality. On the HQ diet in round one, testosterone decreased with laying sequence for male eggs, but increased with laying sequence for female eggs. If females are the more costly sex (Kilner, 1998; Martins, 2004), this may be an adaptive mechanism to offset the disadvantage facing females hatching at the end of the clutch, by increasing their competitive ability (Schwabl, 1996). On the LQ diet in round one, male eggs contained higher concentrations of DHT and testosterone than did female eggs. Because daughters have relatively lower reproductive potential on a LQ diet (Haywood and Perrins, 1992), mothers may be suppressing their investment in female eggs. Higher androgens in male eggs may partly explain why males do better than do females on a restricted diet (Bradbury and Blakey, 1998; Kilner, 1998), although this cannot be the only factor, because females still fare worse than do males on a poor rearing diet when eggs are laid on the same diet (Martins, 2004). It is important to note that we cannot exclude the possibility that the levels that we measured reflected differential rates of metabolism or even production by the embryos rather than initial maternal allocation. However, because embryos were at a very early stage of development, we believe that assayed levels are likely to be representative of initial maternal allocation. This is in contrast to the study by Petrie et al. (2001), in which androgen levels in male and female peafowl eggs were assayed after 10-day incubation, at which time embryonic production of steroids cannot be ruled out (Woods and Erton, 1978). Further experimental work involving biopsies of fresh zebra finch eggs before incubation is necessary to unequivocally determine initial maternal allocation of androgens to male and female eggs. Nevertheless, the present study provides evidence that there are sex differences in androgen metabolism and/or allocation in zebra finches, consistent with other findings of differences between the sexes in this species (Bradbury and Blakey, 1998; Kilner, 1998; Martins, 2004; Rutstein et al., 2004).
Conclusions
Our data suggest that maternal deposition of yolk androgens varies according to diet quality, demonstrating a degree of flexibility on the part of the female in relation to her environment. There was evidence of a carry-over effect with respect to certain aspects of the egg mass and androgen data, but not others. This may be because females were reacting both to their own body condition and to the current resources available for rearing young, and may also reflect limitations in female plasticity. Certainly, in the first breeding round, females on the HQ diet were depositing more testosterone in infertile eggs compared with females on the LQ diet.
Females were also found to allocate androgens differentially to male and female eggs. The mechanism by which this is achieved and the effect that it has on the offspring is not known, but it may partly explain why female chicks have higher mortality than do male chicks when eggs have been laid and chicks reared on a restricted diet (Bradbury and Blakey, 1998; Kilner, 1998). In conclusion, the present study suggests that, together with hatching asynchrony and within-clutch differences in egg mass, female allocation of yolk androgens has the potential to affect within-clutch competitive hierarchies in a sex-specific manner. Research on yolk androgens and their effects in birds is still at a very early stage, and there are clearly large gaps in our knowledge concerning the extent of control the laying female has over the allocation of androgens to her eggs, including the detection or production of male and female eggs (for review, see Pike and Petrie, 2003). There is also comparatively little known about the effects of androgens on the offspring, as well as the chick's ability to modify its situation through its own production and metabolism of androgens. Answering these questions therefore represents the next major challenge in this field.
|
ACKNOWLEDGEMENTS |
|---|
We thank Biotechnology and Biological Sciences Research Council for funding this work, and Neil Hazon for the use of his hormone laboratory and assistance with the assays. We also thank Monique MacKenzie for statistical advice.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alisaukas RT, 1986. Variation in the composition of the eggs and chicks of American coots. Condor 88:84–90.
Appleby BM, Petty SJ, Blakey JK, Rainey P, and MacDonald DW, 1997. Does variation of sex ratio enhance reproductive success of offspring in tawny owls (Strix aluco)? Proc R Soc Lond B 264:1111–1116.[CrossRef]
Blount JD, Surai PF, Nager RG, Houston DC, Møller AP, Trewby ML, and Kennedy MW, 2002. Carotenoids and egg quality in the lesser black-backed gull Larus fuscus: a supplemental feeding study of maternal effects. Proc R Soc Lond B 269:29–36.[Medline]
Bradbury RR, and Blakey JK, 1998. Diet, maternal condition, and offspring sex ratio in the zebra finch, Poephila guttata. Proc R Soc Lond B 265:895–899.[CrossRef]
Clark AB, and Wilson DS, 1981. Avian breeding adaptations: hatching asynchrony, brood reduction, and nest failure. Q Rev Biol 56:253–277.[CrossRef]
de Kogel CH, 1997. Long-term effects of brood size manipulation on morphological development and sex-specific mortality of offspring. J Anim Ecol 66:167–178.[CrossRef]
Eising CM, Eikenaar C, Schwbl H, and Groothuis TGG, 2001. Maternal androgens in black-headed gull (Larus ridibundus) eggs: consequences for chick development. Proc R Soc Lond B 268:839–846.[Medline]
Eising CM, and Groothuis TGG, 2003. Yolk androgens and begging behaviour in black-headed gull chicks: an experimental field study. Anim Behav 66:1027–1034.[CrossRef]
Eising CM, Visser GH, Müller W, and Groothuis, TGG, 2003. Steroids for free? no metabolic costs of elevated maternal androgen levels in the black-headed gull. J Exp Biol 206:3211–3218.
Elf PK, and Fivizzani AJ, 2002. Changes in sex steroid levels in yolks of the leghorn chicken, Gallus domesticus, during embryonic development. J Exp Zool 293:594–600.[CrossRef][ISI][Medline]
French JB, Nisbet ICT, and Schwabl H, 2001. Maternal steroids and contaminants in common tern eggs: a mechanism of endocrine disruption? Comp Biochem Physiol C 128:91–98.[CrossRef]
Gil D, Graves J, Hazon N, and Wells A, 1999. Male attractiveness and differential testosterone investment in zebra finch eggs. Science 286:126–128.
Gorman HE, and Nager RG, 2003. State-dependent incubation behaviour in the zebra finch. Anim Behav 65:745–754.[CrossRef]
Griffiths R, Double MC, Orr K, and Dawson RJG, 1998. A DNA test to sex most birds. Mol Ecol 7:1071–1075.[CrossRef][Medline]
Groothuis TG, and Schwabl H, 2002. Determinants of within- and among-clutch variation in levels of maternal hormones in black-headed gull eggs. Funct Ecol 16:281–289.[CrossRef]
Haywood S, and Perrins CM, 1992. Is clutch size in birds affected by environmental conditions during growth. Proc R Soc Lond B 249:195–197.[Medline]
Henry MH, and Burke WH, 1999. The effects of in ovo administration of testosterone or an antiandrogen on growth of chick embryos and embryonic muscle characteristics. Poult Sci 78:1006–1013.
Hochachka W, 1992. How much should reproduction cost? Behav Ecol 3:42–52.
Houston DC, Donnan D, and Jones PJ, 1995. The source of nutrients required for egg production in zebra finches Poephila guttata. J Zool (Lond) 235:469–483.
Houston DC, Jones PJ, and Sibly RM, 1983. The effect of female body condition on egg laying in Lesser black-backed gulls Larus fuscus. J Zool (Lond) 200:509–520.
Hussell DJT, 1972. Factors affecting clutch size in Arctic passerines. Ecol Monogr 42:317–364.[CrossRef]
Kenward MG, and Roger JH, 1997. Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 53:983–987.[CrossRef][ISI][Medline]
Ketterson ED, and Nolan V Jr, 1999. Adaptation, exaptation, and constraint: a hormonal perspective. Am Nat 155:S4–S25.[CrossRef]
Kilner R, 1998. Primary and secondary sex ratio manipulation by zebra finches. Anim Behav 56:155–164.[CrossRef][ISI][Medline]
Leimar O, 1996. Life-history analysis of the Trivers and Willard sex-ratio problem. Behav Ecol 7:316–325.
Linden M, and Møller AP, 1989. Cost of reproduction and covariation of life history traits in birds. Trends Ecol Evol 4:367–371.[CrossRef]
Lipar JL, and Ketterson ED, 2000. Maternally-derived yolk testosterone enhances the development of the hatching muscle in the red-winged blackbird. Proc R Soc Lond B 267:2005–2010.[Medline]
Lipar JL, Ketterson ED, and Nolan V, 1999. Intraclutch variation in testosterone content of red-winged blackbird eggs. Auk 116:231–235.
Martins TLF, 2004. Sex-specific growth in zebra finch nestlings: a possible mechanism for sex ratio adjustment. Behav Ecol 15:174–180.
Mazuc J, Chastel O, and Sorci G, 2003. No evidence for differential maternal allocation to offspring in the house sparrow (Passer domesticus). Behav Ecol 14:340–346.
Monaghan P, Metcalfe NB, and Houston DC, 1996. Male finches selectively pair with fecund females. Proc R Soc Lond B 263:1183–1186.
Müller W, Eising CM, Dijkstra C, and Groothuis TGG, 2002. Sex differences in yolk hormones depend on maternal social status in leghorn chickens (Gallus gallus domesticus). Proc R Soc Lond B 269:2249–2255.[Medline]
Nager RG, Monaghan P, and Houston DC, 2000. Within-clutch trade-offs between the number and quality of eggs: Experimental manipulations in gulls. Ecology 81:1339–1350.[CrossRef]
Nash JP, Cuisset BD, Bhattacharyya S, Suter HC, le Menn F, and Kime DE, 2000. An enzyme linked immunosorbant assay (ELISA) for testosterone, estradiol, and 17,20ß-dihydroxy-4-pregenen-3-one using acetylcholinesterase as tracer: application to measurement of diet patterns in rainbow trout (Oncorhynchus mykiss). Fish Physiol Biochem 22:355–363.[CrossRef]
O'Connor RJ, 1979. Egg weights and brood reduction in the European swift (Apus apus). Condor 81:133–145.
Perrins CM, 1970. The timing of birds breeding seasons. Ibis 112:242–255.
Petrie M, Schwabl H, Brande-Lavridsen N, and Burke T, 2001. Maternal investment: sex differences in avian yolk hormone levels. Nature 412:498–498.[Medline]
Pike TW, and Petrie M, 2003. Potential mechanisms of avian sex manipulation. Biol Rev 78:553–574.[Medline]
Pilz KM, Smith HG, Sandell MI, and Schwabl H, 2003. Interfemale variation in egg yolk androgen allocation in the European starling: do high-quality females invest more? Anim Behav 65:841–850.[CrossRef]
Royle NJ, Surai PF, and Hartley IR, 2001. Maternally derived androgens and antioxidants in bird eggs: complementary but opposing effects? Behav Ecol 12:381–385.
Royle NJ, Surai PF, McCartney RJ, and Speake BK, 1999. Parental investment and egg yolk lipid composition in gulls. Funct Ecol 13:298–306.[CrossRef]
Rutkowska J, and Cichon M, 2002. Maternal investment during egg laying and offspring sex: an experimental study of zebra finches. Anim Behav 64:817–822.[CrossRef]
Rutstein AN, 2003 Reproductive investment in the zebra finch Taeniopygia guttata (PhD dissertation). St. Andrews: University of St. Andrews.
Rutstein AN, Slater PJB, and Graves JA, 2004 Diet quality and resource allocation in the zebra finch. Proc R Soc Lond B (Suppl.)271:286–289.
SAS Institute, 1998. SAS/STAT user's guide, version 8. Cary, North Carolina: SAS Institute, Inc.
Saino N, Ferrari R, Martinelli R, Romano M, Rubolini D, and Møller AP, 2002. Early maternal effects mediated by immunity depend on sexual ornamentation of the father in the barn swallow (Hirundo rustica). Proc R Soc Lond B 269:1005–1009.[Medline]
Schwabl H, 1993. Yolk is a source of maternal testosterone for developing birds. Proc Natl Acad Sci USA 90:11446–11450.
Schwabl H, 1996. Maternal testosterone in the avian egg enhances postnatal growth. Comp Biochem Physiol Part A Mol Integr Physiol 114:271–276.
Schwabl H, 1997. The contents of maternal testosterone in house sparrow Passer domesticus eggs vary with breeding conditions. Naturwissenschaften 84:406–408.[CrossRef][ISI][Medline]
Schwabl H, Mock DW, and Gieg JA, 1997. A hormonal mechanism for parental favouritism. Nature 386:231–231.[Medline]
Selman RG, and Houston DC, 1996. The effect of prebreeding diet on reproductive output in zebra finches. Proc R Soc Lond B 263:1585–1588.
Slagsvold T, Sandvik J, Rofstad G, Lorentsen O, and Husby M, 1984. On the adaptive value of intraclutch egg-size variation in birds. Auk 101:685–697.
Smith HG, Ottosson U, and Ohlsson T, 1993. Interclutch variation in egg mass among starlings Sturnus vulgaris reflects female condition. Ornis Scand 24:311–316.
Sockman KW, and Schwabl H, 2000. Yolk androgens reduce offspring survival. Proc R Soc Lond B 267:1451–1456.[Medline]
Trivers RL, and Willard DE, 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science 179:90–91.
Verboven N, Monaghan P, Evans DM, Schwabl H, Evans N, Whitelaw C, and Nager RG, 2003. Maternal condition, yolk androgens and offspring fitness: a supplemental feeding experiment in the lesser black-backed gull (Larus fuscus). Proc R Soc Lond B 270:2223–2232.[Medline]
Williams TD, 1996. Variation in reproductive effort in female zebra finches (Taeniopygia guttata) in relation to dietary supplements during egg laying. Physiol Zool 69:1255–1275.
Woods JE, and Erton LH, 1978. The synthesis of estrogens in the gonads of the chicken embryo. Gen Comp Endocrinol 36:360–370.[CrossRef][ISI][Medline]
Zann RA, 1996. The zebra finch. New York: Oxford University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. N. Rutstein, L. Gilbert, and J. L. Tomkins Experience counts: lessons from studies of differential allocation Behav. Ecol., September 1, 2005; 16(5): 957 - 960. [Full Text] [PDF]
|
|
|
|
|
|
A.N. Rutstein, H.E. Gorman, K.E. Arnold, L. Gilbert, K.J. Orr, A. Adam, R. Nager, and J.A. Graves Sex allocation in response to paternal attractiveness in the zebra finch Behav. Ecol., July 1, 2005; 16(4): 763 - 769. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||