Age-specific patterns of maternal investment in common gull egg yolk

Abstract

While the general patterns of age-specific changes in reproductive success are quite well established in long-lived animals, we still do not know if allocation patterns of maternally transmitted compounds are related to maternal age. We measured the levels of yolk testosterone, carotenoids and vitamins A and E in a population of known-aged common gulls (Larus canus) and found an age-specific pattern in yolk lutein and vitamin A concentrations. Middle-aged mothers allocated more of these substances to yolk compared to young and old mothers. These results can be explained through differences in age-specific foraging, absorption or deposition patterns of carotenoids and vitamins into yolk. If these molecules play a role in antioxidant defence and immune modulation, our results suggest a possible physiological pathway underlying the age-specific changes in reproductive success of long-lived birds in the wild.

1. Introduction

Increased interest in the pace-of-life syndrome hypothesis seen in recent years and the focus on physiological mediators behind the individual variation of life-history strategies [1] raise a need for studies of age-specific variation in physiological traits that are linked to life-history traits. Individual life-history strategy determines the optimal investment in reproduction versus maintenance, which shapes the patterns of age-associated variation in reproductive success [2]. Variation in age-specific phenotype of the mother can influence offspring phenotype beyond maternal contribution to the genetic constitution [3], through the maternal allocation patterns of resources transmitted to the offspring.

From the perspective of studying maternal allocation patterns to reproduction in relation to age, long-lived birds are good model organisms for several reasons. First, as long-lived species invest in self-maintenance over reproduction, variance in reproductive allocation has been shown to be larger than variance in adult survival [4], leading to reproductive senescence being more pronounced than somatic senescence. Second, because embryo development takes place within a sealed system, the egg, maternal investment at this stage can be easily measured [5]. In addition to nutrients, birds transfer many biologically active compounds to their eggs, including antioxidants, hormones, antibodies and antibacterial enzymes, with consequences on the morphology, physiology and behaviour of offspring [5]. While the general age-specific patterns of reproductive success (i.e. increase in younger age classes, reproductive senescence in older age classes, [4,6]) are quite well established in long-lived birds, we do not know if and how the active components of the yolk are affected by maternal age. None of these maternally transmitted compounds have ever been studied in the context of maternal age in wild birds with exactly known ages.

We studied maternal investment in egg yolk in a long-lived seabird species, the common gull (Larus canus), in a breeding colony of known-age birds (age range: 2–34). Reproductive success in common gulls increases to the 10th breeding year (13 years of age) and decreases thereafter [6]. We focused our study on three maternally transmitted compounds. First, we measured yolk testosterone levels, which are known to affect chick growth and development in birds [5,7]. Studies linking yolk testosterone to the precise age of the mother are, to our best knowledge, missing. Second, we assessed yolk carotenoid concentrations. Carotenoids are a diverse group of fat-soluble pigments that influence antioxidant status and immunity [8], but also the retinal health of birds [9]. These two yolk components are suggested to be linked through a compensatory mechanism, where an increase in yolk carotenoids prevents oxidative damage caused by the increase in yolk testosterone in developing bird embryos [7]. Third, we measured the levels of vitamins A and E, which are known to positively affect offspring growth across bird species [10].

We predicted that an age-specific pattern should be observable in yolk antioxidant and testosterone levels. A quadratic relationship with age is expected based on the understanding that the middle-aged birds may be in ‘prime' reproductive condition. A similar relationship between recruitment rate and age has been previously shown in common gulls [6]. The quadratic adjustment is especially appropriate in cross-sectional studies, because it reveals patterns otherwise masked by simpler linear adjustments [11].

2. Material and methods

We conducted the study in a breeding colony of known-aged common gulls in Estonia (58°46′ N, 23°26′ E). We collected the first egg from 30 nests after the second egg was laid to avoid nest abandonment. Common gull clutch consists invariably of three eggs. Maternal age varied from 4 to 23 years. We measured yolk testosterone concentrations with radioimmunoassay following previously published protocols [12], and the concentration of yolk carotenoids (eight types) and vitamins A and E by adapting a previously established high-performance liquid chromatography (HPLC) method [13]. The details and repeatability assessment of methods are available as the electronic supplementary material, table S1. We used quadratic regression analyses with laying date as a covariate for testing the associations between maternal age and dependent variables. The p-value < 0.05 was set as a criterion for statistical significance.

3. Results

Laying date only significantly affected egg weight, with eggs laid later in season being lighter (table 1). We identified eight carotenoids: lutein, astaxanthin, zeaxanthin, canthaxanthin, echinenone, β-cryptoxanthin, β-carotene, and one unknown carotenoid (electronic supplementary material, table S2, intercorrelations electronic supplementary material, table S3). Yolk lutein concentration was related to maternal age, displaying a significant quadratic term with age. The concentration of yolk lutein was lower in the eggs of young and old mothers compared to the eggs of middle-aged mothers (figure 1). The quadratic relationship of mother's age with vitamins A and E level showed a similar trend (although statistically non-significant, figure 1). Yolk vitamin E level showed a linear relationship with female age, being higher in younger females. Yolk testosterone levels showed no relationship with maternal age. When only reproductively senescent birds (age 13+) were included, lutein concentration in egg yolk showed a linear decrease with age (electronic supplementary material, table S5). When we analysed carotenoid proportions from total carotenoids, lutein, zeaxanthin and echinenone indicated a quadratic relationship with age, while canthaxanthin was proportionally highest in egg yolks of old females (electronic supplementary material, table S6). When carotenoid and vitamin concentrations where multiplied with egg mass, lutein, vitamins A and E and zeaxanthin all showed a significant quadratic relationship with age (electronic supplementary material, table S7).

Table 1.

Quadratic regression models of the associations between egg characteristics and mother's age, including laying date as a covariate. Sample size for each analysis was 30 (d.f. = 1,26). B, parameter estimate.

trait	effect	F	p-value	B ± s.e.	adjusted R2
egg mass	mother age	1.09	0.307	0.64 ± 0.62	0.20
	mother age2	0.58	0.454	−0.02 ± 0.02
	laying date	8.66	0.007	−2.61 ± 0.89
testosterone	mother age	0.29	0.594	0.74 ± 1.38	−0.01
	mother age2	0.16	0.694	−0.02 ± 0.05
	laying date	1.19	0.285	−1.13 ± 1.03
vitamin A	mother age	3.28	0.082	0.02 ± 0.01	0.03
	mother age2	3.76	0.063	<−0.001 ± 0.01
	laying date	0.43	0.518	0.01 ± 0.01
vitamin E	mother age	4.50	0.044	0.52 ± 0.25	0.08
	mother age2	4.07	0.054	−0.02 ± 0.01
	laying date	2.41	0.132	0.29 ± 0.19
total carotenoids	mother age	0.03	0.876	−0.40 ± 2.51	−0.07
	mother age2	0.11	0.746	0.03 ± 0.09
	laying date	0.21	0.650	0.86 ± 1.88
lutein	mother age	7.20	0.013	0.90 ± 0.33	0.13
	mother age2	6.91	0.014	−0.03 ± 0.01
	laying date	0.10	0.752	0.09 ± 0.25
astaxanthin	mother age	0.58	0.455	−0.27 ± 0.35	−0.03
	mother age2	0.95	0.339	0.01 ± 0.01
	laying date	0.02	0.899	0.03 ± 0.26
zeaxanthin	mother age	2.97	0.097	0.42 ± 0.24	0.001
	mother age2	2.74	0.110	−0.02 ± 0.01
	laying date	0.16	0.690	0.07 ± 0.18
canthaxanthin	mother age	1.17	0.290	−1.60 ± 1.48	0.02
	mother age2	1.81	0.190	0.08 ± 0.06
	laying date	0.04	0.852	0.21 ± 1.12
echinenone	mother age	2.48	0.127	0.22 ± 0.14	0.002
	mother age2	2.15	0.154	−0.01 ± 0.01
	laying date	1.22	0.280	0.11 ± 0.10
β-cryptoxanthin	mother age	0.22	0.644	0.07 ± 0.14	0.001
	mother age2	0.43	0.516	−0.004 ± 0.01
	laying date	1.81	0.190	0.14 ± 0.11
unknown carotenoid	mother age	0.02	0.879	0.03 ± 0.19	−0.03
	mother age2	0.03	0.858	−0.001 ± 0.01
	laying date	2.02	0.167	0.20 ± 0.14
β-carotene	mother age	0.53	0.471	0.07 ± 0.09	0.03
	mother age2	0.75	0.395	−0.003 ± 0.004
	laying date	3.30	0.081	0.12 ± 0.07

Figure 1.

Associations between female common gull age and egg characteristics. Continuous line indicates a significant association with age square of the female, dashed line indicates a trend.

4. Discussion

Here, we show that the yolk deposition of active components varies with maternal age. The most robust finding of our study was that yolk lutein content was the highest in eggs of middle-aged mothers, and lower in young and old breeders. Outside its role in integument coloration, the function of lutein in birds has been mostly considered in the framework of immune defence, antioxidant protection [14] and retinal health [9]. In yellow-legged gulls (Larus michahellis), yolk lutein concentration positively predicted chick tarsus length and body mass, while other carotenoids (zeaxanthin and dehydrolutein) were found to have no predictive power of chick phenotype [15]. To show an age-specific pattern in its deposition in yolk, lutein must (1) have an important self-maintenance function in adults and be therefore traded off against deposition to yolk, and/or (2) be differently available from diet for birds in different ages. While these two hypotheses are not mutually exclusive, both provide a possible link between maternal age and chick quality. Our previous longitudinal studies in this species have indicated that while physiological condition is not generally lower in older birds [16], reproductive senescence does occur, on both the level of signal traits [17] and offspring survival [6], indicating a possible trade-off between self-maintenance and reproductive integrity in older gulls.

Vitamin E showed a linear relationship with female age, being lower in younger female's egg yolk (but showed a decreasing trend when only studied in birds aged 13+ years, electronic supplementary material, table S4). When egg mass was taken into account, yolk vitamin A and E deposition also showed a quadratic relationship with age. Both of these vitamins have antioxidant properties, and have important functions in embryo development and in cell differentiation [10]. Deficiency in yolk vitamin E has been previously shown to limit growth in yellow-legged gull chicks [18]. When the proportion of carotenoid types from total carotenoid was analysed in relation to age, senescent pattern was again most significant in yolk lutein, but also in echinenone and zeaxanthin, while canthaxanthin concentrations were proportionally higher in eggs of older individuals. Canthaxanthin is easily absorbed from diet and effectively transferred to eggs [19], while other carotenoid types (especially lutein) might be more limiting.

The increase in yolk carotenoid and vitamin levels from young to middle-aged birds mirrors the increase in reproductive success in young to middle-aged gulls [6]. Lower reproductive success of younger gulls might indicate that it is dependent on age-specific condition of birds, or that birds in lower body condition selectively disappear from the older age classes. Thus, the increase in egg active component concentration from young to middle-aged birds provides support for the hypothesis that the ability to deposit these molecules in eggs is condition dependent, mediated either by greater efficiency in ensuring a high dietary intake, or by differences in processing and allocation of carotenoids and vitamins.

Yolk testosterone concentration in egg yolk was not related to maternal age. A recent study on Japanese quails (Coturnix japonica) suggested that while maternally derived testosterone is known to stimulate growth, these benefits may be counterbalanced by an increase in the production of reactive oxygen species, suggesting an interaction between testosterone and other egg components, including carotenoids with their antioxidant function [7]. In addition, a comparative study on birds indicated that androgens and antioxidants are co-adjusted within eggs and that maternally transmitted antioxidants might limit the potential effects of prenatal exposure to high testosterone levels on oxidative stress [20]. It would be, therefore, intriguing to hypothesize that the lower reproductive success of older gulls results from their inability to compensate the negative effects of testosterone-stimulated growth on oxidative stress of the embryo with sufficient levels of yolk carotenoids. Future experimental studies that supplement the eggs of the older gulls with carotenoids and follow-up on chick survival and recruitment rate could clarify this question.

In conclusion, our study shows, for the first time, an age-specific pattern in the deposition of yolk carotenoids and vitamins and suggests that this pattern could mediate the lower reproductive success observed in old birds in most long-lived species. While it remains unclear whether this age-specific pattern is caused by dietary differences between differently aged birds and/or physiological trade-offs, this finding brings us one step closer to understanding the mechanisms of reproductive senescence in the wild.

Ethics

This study was approved by Estonian Agricultural Ministry (licence 106).

Data accessibility

Raw data are available in the Dryad Digital Repository at http://dx.doi.org/10.5061/dryad.dc8dp2j [21].

Authors' contributions

T.S., M.G. and P.H. designed the study; J.U. and K.R. conducted fieldwork, K.R. provided demographic data; M.O., T.S. and J.U. conducted laboratory analyses; J.U., T.S. and P.H. drafted the manuscript and all authors revised it critically for important intellectual content. All authors agreed to be held accountable for the content therein and approved the final version of the manuscript.

Competing interests

We have no competing interests.

Funding

This work was supported by the Estonian Ministry of Education (ETF7190), Estonian Research Council (IUT21-1, IUT34-8), Scientific Grant Agency of the Slovak Republic VEGA 1/0686/15 to M.O. and the Marie Sklodowska-Curie grant agreement no. 701747 to T.S.