Julian K. Christians

Avian egg size: variation within species and inflexibility within individuals

Egg size is a widely‐studied trait and yet the causes and consequences of variation in this trait remain poorly understood. Egg size varies greatly within many avian species, with the largest egg in a population generally being at least 50% bigger, and sometimes twice as large, as the smallest. Generally, approximately 70% of the variation in egg mass is due to variation between rather than within clutches, although there are some cases of extreme intra‐clutch egg‐size variation. Despite the large amount of variation in egg size between females, this trait is highly consistent within individuals between breeding attempts; the repeatability of egg size is generally above 0.6 and tends to be higher than that of clutch size or laying date. Heritability estimates also tend to be much higher for egg size (> 0.5) than for clutch size or laying date (< 0.5). As expected, given the high repeatability and heritability of egg size, supplemental food had no statistically significant effect on this trait in 18 out of 28 (64%) studies. Where dietary supplements do increase egg size, the effect is never more than 13% of the control values and is generally much less. Similarly, ambient temperature during egg formation generally explains less than 15% of the variation in egg size. In short, egg size appears to be a characteristic of individual females, and yet the traits of a female that determine egg size are not clear. Although egg size often increases with female age (17 out of 37 studies), the change in egg size is generally less than 10%. Female mass and size rarely explain more than 20% of the variation in egg size within species. A females egg size is not consistently related to other aspects of reproductive performance such as clutch size, laying date, or the pairs ability to rear young. Physiological characteristics of the female (e.g. endogenous protein stores, oviduct mass, rate of protein uptake by ovarian follicles) show more promise as potential determinants of egg size. With regards to the consequences of egg‐size variation for offspring fitness, egg size is often correlated with offspring mass and size within the first week after hatching, but the evidence for more long‐lasting effects on chick growth and survival is equivocal. In other oviparous vertebrates, the magnitude of egg‐size variation within populations is often as great or greater than that observed within avian populations. Although there are much fewer estimates of the repeatability of egg size in other taxa, the available evidence suggests that egg size may be more flexible within individuals. Furthermore, in non‐avian species (particularly fish and turtles), it is more common for female mass or size to explain a substantial proportion of the variation in egg size. Further research into the physiological basis of egg‐size variation is needed to shed light on both the proximate and ultimate causes of intraspecific variation in this trait in birds.

Follicular Development and Plasma Yolk Precursor Dynamics through the Laying Cycle in the European Starling (Sturnus vulgaris)

We investigated the quantitative matching of plasma yolk precursor supply (the plasma pool) to follicle demand during yolk formation in European starlings (Sturnus vulgaris). Plasma concentrations of the two yolk precursors, vitellogenin (VTG) and very low density lipoprotein (VLDL), were only elevated coincident with rapid yolk development (RYD) and matched variation in total yolky follicle mass. VTG and VLDL were low (<0.4 μg/mL and <4.2 mg/mL, respectively) in nonbreeders and prebreeders with no yolky follicles, and at clutch completion. They increased to 4.02 μg/mL and 19.4 mg/mL in birds with a full follicle hierarchy (F1–F4), and concentrations then remained high and actually increased up to the point where only a single, yolky (F1) follicle remained. However, there was some evidence for mismatching of supply and demand: (a) precursor concentrations increased throughout the laying cycle even though the number of developing follicles decreased. We suggest that this is because of a requirement to maintain a large precursor pool to maintain high uptake rates; and (b) in birds with a full follicle hierarchy, precursor concentrations were negatively correlated with total follicle mass. This suggests that high uptake rates in large follicles can actually deplete circulating precursor concentrations. Plasma concentrations of both yolk precursors increased rapidly in the early morning with (predicted) time after ovulation, consistent with a lack of fine control of precursor concentrations. However, mean plasma VTG concentrations did not differ between morning or evening samples. In contrast, plasma VLDL concentrations were lower in the morning (16.8 mg/mL) than in the evening (22.9 mg/mL). Although there is marked individual variation in plasma VTG and VLDL (four‐ to eightfold), both precursors were repeatable in the short term (24 h), and plasma VTG was repeatable over a 14‐d interval between successive breeding attempts.

Controlling for Body Mass Effects: Is Part‐Whole Correlation Important?

Many physiological and life-history traits scale with body mass, and this mass dependence must often be taken into account when comparing group means or when analyzing correlations between traits. Residual analysis (Bennett 1987), multiple regression (Hayes and Shonkwiler 1996), and analysis of covariance (ANCOVA, Atchley et al. 1976; Packard and Boardman 1987) have been proposed as robust techniques to reduce or eliminate mass dependence, since the use of ratios (i.e., the trait value divided by body mass) is problematic (Packard and Boardman 1987). The first of these three methods all involve regression, so problems may arise when the physiological trait of interest is the mass of an organ or tissue and the covariate is body mass. In a regression between organ mass and body mass, both variables include the mass of the organ. This tends to create a positive relationship between the two variables, even when no correlation between organ mass and the rest of body mass (i.e., body mass minus organ mass) exists. Sokal and Rohlf (1995) referred to the correlation between a part (e.g., organ mass) and the whole (e.g., body mass) as “part-whole correlation.” A number of authors have addressed the issue of partwhole correlation within the context of scaling (e.g., Prothero 1986; LaBarbera 1989) and the analysis of ratios (e.g., Atchley et al. 1976; Jackson and Somers 1991 and references therein). Descriptions of the effects of part-whole correlation in ANCOVA, multiple regression, and residual analysis, however, have been only anecdotal. For instance, Tracy and Sugar (1989) and Burness et al. (1998) described examples where removal of partwhole correlation changed the results of ANCOVA and residual analysis, respectively. A more in-depth study of the effects of part-whole correlation on regression-type analyses (e.g., multiple regression, residual analysis, and ANCOVA) is lacking, and as a result, this potential problem is often ignored. For instance,

Enhanced immune function does not depress reproductive output

Costs of reproduction might be mediated by a physiological (resource allocation) trade–off between immune function and reproductive effort, and several recent studies have shown that an experimental increase in reproductive effort is associated with decreased immune function. Here we test the complementary prediction of this hypothesis: that increased immune function (specific antibody production) depresses reproductive output. Female European starlings (Sturnus vulgaris) were injected with a non–pathogenic antigen (sheep red blood cells) following completion of laying of their first clutch, to stimulate an in vivo humoral immune response (primary antibody production). We induced laying of a second clutch by removing the first clutch, and assessed changes in reproductive performance in individual females pre– and post–treatment. Injection of sheep red blood cells produced a significant antibody response in 96% (n=29) of treated females, with titres comparable to previous studies (range 1–7). However, increased antibody production did not decrease primary or secondary female reproductive effort (re–laying interval, egg size, clutch size, chick growth or fledging success), compared with control, saline–injected birds (n=22). These data do not support a simple resource allocation model for the cost of reproduction, based on a reciprocal, negative relationship between resources allocated to immune function and reproduction.

Regulatory variation at glypican-3 underlies a major growth QTL in mice

The genetic basis of variation in complex traits remains poorly understood, and few genes underlying variation have been identified. Previous work identified a quantitative trait locus (QTL) responsible for much of the response to selection on growth in mice, effecting a change in body mass of approximately 20%. By fine-mapping, we have resolved the location of this QTL to a 660-kb region containing only two genes of known function, Gpc3 and Gpc4, and two other putative genes of unknown function. There are no non-synonymous polymorphisms in any of these genes, indicating that the QTL affects gene regulation. Mice carrying the high-growth QTL allele have approximately 15% lower Gpc3 mRNA expression in kidney and liver, whereas expression differences at Gpc4 are non-significant. Expression profiles of the two other genes within the region are inconsistent with a factor responsible for a general effect on growth. Polymorphisms in the 3′ untranslated region of Gpc3 are strong candidates for the causal sequence variation. Gpc3 loss-of-function mutations in humans and mice cause overgrowth and developmental abnormalities. However, no deleterious side-effects were detected in our mice, indicating that genes involved in Mendelian diseases also contribute to complex trait variation. Furthermore, these findings show that small changes in gene expression can have substantial phenotypic effects.

PAPPA2, an Enzyme That Cleaves an Insulin-Like Growth-Factor-Binding Protein, Is a Candidate Gene for a Quantitative Trait Locus Affecting Body Size in Mice

Identifying genes responsible for quantitative variation remains a major challenge. We previously identified a quantitative trait locus (QTL) affecting body size that segregated between two inbred strains of mice. By fine mapping, we have refined the location of this QTL to a genomic region containing only four protein-coding genes. One of these genes, PAPPA2, is a strong candidate because it codes for an enzyme that cleaves insulin-like growth-factor-binding protein 5 (IGFBP-5), an important stimulator of bone formation. Among littermates that segregate only for the four-gene region, we show that the QTL has a significant effect on the circulating levels of IGFBP-5 and IGFBP-3 (the latter subject to limited degradation by PAPPA2), but not on levels of IGFBP-2 and IGFBP-4, which are not cleaved by PAPPA2. There are 14 nonsynonymous SNPs among QTL alleles, which may affect the activity of the translated protein. The refinement of the target region to four genes and the finding that the QTL affects IGFBP-5 levels suggest that PAPPA2 may be involved with normal postnatal growth. Our mapping results also illustrate the potentially fractal nature of QTL: as we mapped our QTL with increasing resolution, what appeared to be a single QTL resolved into three closely linked QTL (previous work), and then one of these was further dissected into two in this study.

Expression of pregnancy-associated plasma protein A2 during pregnancy in human and mouse

Pregnancy-associated plasma protein-A and -A2 (PAPPA and PAPPA2) are proteases that cleave IGF binding proteins (IGFBPs) and thereby increase the bioavailability of growth factors. PAPPA has long been recognized as a marker of fetal genetic disorders and adverse pregnancy outcomes. In contrast, although PAPPA2 is also highly expressed in human placenta, its physiological importance is not clear. To establish whether mice will be a useful model for the study of PAPPA2, we compared the patterns of expression of PAPPA2 in the placentae of mouse and human. We show, for the first time, that Pappa2 is highly expressed in mouse placenta, as is the case in humans. Specifically, it is expressed at the interface of the maternal and fetal layers of the mouse placenta at all gestational stages studied (10.5-16.5 days post coitum). Similarly, PAPPA2 is expressed in the syncytiotrophoblast layer of human placental villi and is also detected in some invasive extravillous trophoblasts in the first trimester. These results are consistent with a model whereby PAPPA2 cleaves IGFBPs produced in the maternal decidua to promote feto-placental growth, and indicate that this protein may play analogous roles in human and mouse placenta. PAPPA2 protein is detectable in the circulation of pregnant mice and humans during the first trimester and at term, raising the possibility that PAPPA2 may be a useful biomarker of placental dysfunction. Pappa2 expression also shows specific localization within the mouse embryo and therefore may play roles in fetal development, independent of its action in the placenta.

Genetic complexity of an obesity QTL (Fob3) revealed by detailed genetic mapping.

Obesity is proving to be a serious health concern in the developed world as well as an unwanted component of growth in livestock production. While recent advances in genetics have identified a number of monogenic causes of obesity, these are responsible for only a small proportion of human cases of obesity. By divergent selection for high and low fat content over 60 generations, we have created Fat (F) and Lean (L) lines of mice that represent a model of polygenic obesity similar to the situation in human populations. From previous crosses of these lines, four body fat quantitative trait loci (QTL) were identified. We have created congenic lines (Fchr15L), by recurrent marker-assisted backcrossing, to introgress the QTL region with the highest LOD score, Fob3 on Chr 15, from the L-Iine into the F-line background. We have further mapped this QTL by progeny testing of recombinants, produced from crosses between the F-line and congenic Fchrl5L mice, showing that the Fob3 QTL region is a composite of at least two smaller effect QTL—the proximal QTL Fob3a is a late-onset obesity QTL, whereas the distal Fob3b is an early-onset obesity QTL.

Fine mapping of a murine growth locus to a 1.4-cM region and resolution of linked QTL

Previous work identified a QTL affecting murine size (particularly tail length) in a cross between C57BL/6J and DBA/2J mice and refined its location to an 8-cM region between D1Mit30 and D1Mit57. The present study used recombinant progeny testing to fine map this QTL. Individuals from a partially congenic strain carrying chromosomes recombinant between D1Mit30 and D1Mit57 were mated to DBA/2J, generating 942 progeny. Two QTL affecting 10-week tail length were identified in this population: one at 9.7 cM distal to D1Mit30 (the position estimated in previous work), and another of smaller effect near D1Mit30. A second population (n = 787) was generated by mating siblings from the progeny test population that were heterozygous for the same segment of chromosome, including only recombinants between D1Mit265 and D1Mit57. In the latter population, two QTL were also identified: one at 10.2 cM distal to D1Mit30, and another of smaller effect at the distal end of the mapped region (at D1Mit150). When the two populations were analyzed together, the estimated location of the central QTL was 10.2 cM distal to D1Mit30 and there was marginally significant evidence of the distal QTL. The central QTL explained approximately 7% of the phenotypic variance, and the 95% confidence interval for its position (determined by bootstrapping) was a 1.4-cM region, approximately the region from D1Mit451 to D1Mit219. The central QTL also affected tail length and body mass at 3 and 6 weeks of age, but to a lesser degree than 10-week tail length.

Placental invasion, preeclampsia risk and adaptive molecular evolution at the origin of the great apes: evidence from genome-wide analyses.

INTRODUCTION Recent evidence from chimpanzees and gorillas has raised doubts that preeclampsia is a uniquely human disease. The deep extravillous trophoblast (EVT) invasion and spiral artery remodeling that characterizes our placenta (and is abnormal in preeclampsia) is shared within great apes, setting Homininae apart from Hylobatidae and Old World Monkeys, which show much shallower trophoblast invasion and limited spiral artery remodeling. We hypothesize that the evolution of a more invasive placenta in the lineage ancestral to the great apes involved positive selection on genes crucial to EVT invasion and spiral artery remodeling. Furthermore, identification of placentally-expressed genes under selection in this lineage may identify novel genes involved in placental development. METHODS We tested for positive selection in approximately 18,000 genes using the ratio of non-synonymous to synonymous amino acid substitution for protein-coding DNA. DAVID Bioinformatics Resources identified biological processes enriched in positively selected genes, including processes related to EVT invasion and spiral artery remodeling. RESULTS Analyses revealed 295 and 264 genes under significant positive selection on the branches ancestral to Hominidae (Human, Chimp, Gorilla, Orangutan) and Homininae (Human, Chimp, Gorilla), respectively. Gene ontology analysis of these gene sets demonstrated significant enrichments for several functional gene clusters relevant to preeclampsia risk, and sets of placentally-expressed genes that have been linked with preeclampsia and/or trophoblast invasion in other studies. CONCLUSION Our study represents a novel approach to the identification of candidate genes and amino acid residues involved in placental pathologies by implicating them in the evolution of highly-invasive placenta.