Local Heterozygosity Effects on Nestling Growth an

Authors: Piotr Minias Katarzyna WojczulanisJakubas Robert Rutkowski Krzysztof Kaczmarek

Publish Date: 2015/07/22

Volume: 42, Issue: 4, Pages: 452-460

PDF Link

Abstract

Under inbreeding heterozygosity at neutral genetic markers is likely to reflect genomewide heterozygosity and thus is expected to correlate with fitness There is however growing evidence that some of heterozygosityfitness correlations HFCs can be explained by ‘local effects’ where noncoding loci are at linkage disequilibrium with functional genes The aim of this study was to investigate correlations between heterozygosity at seven microsatellite loci and two fitnessrelated traits nestling growth rate and nutritional condition in a recently bottlenecked population of great cormorant Phalacrocorax carbo sinensis We found that heterozygosity was positively associated with both nestling traits at the betweenbrood level but the individual withinbrood effects of heterozygosity were nonsignificant We also found that only one locus per trait was primarily responsible for the significant multilocus HFCs suggesting a linkage disequilibrium with nonidentified functional loci The results give support for ‘local effect’ hypothesis confirming that HFCs may not only be interpreted as evidence of inbreeding and that genetic associations between functional and selectively neutral markers could be much more common in natural populations than previously thoughtGenotypephenotype associations are often complex and difficult to disentangle but the impact of individual genetic variation on phenotypic quality and fitness has long been recognized as an important evolutionary mechanism Although heterozygosity has been reported to correlate with fitnessrelated traits such as reproductive output Slate et al 2000 Seddon et al 2004 Ortego et al 2009 survival Markert et al 2004 Jensen et al 2007 parasite resistance MacDougallShackleton et al 2005 AcevedoWhitehouse et al 2006 competitive ability Välimäki et al 2007 Minias et al 2015 developmental stability Vangestel et al 2011 and quality of ornamentation Aparicio et al 2001 Herdegen et al 2014 the strength of the heterozygosityfitness correlations HFCs is usually weak and it is still difficult to assess the generality of these associations in natural populations Chapman et al 2009 It is also acknowledged that HFCs may arise through several different mechanisms but there is no clear consensus on their relative importance Hansson and Westerberg 2002Heterozygosity at functional loci may affect fitness through overdominance ‘direct effect’ hypothesis when heterozygous individuals have an intrinsically higher fitness than homozygotes David 1998 This mechanism cannot however easily explain correlations between fitness and heterozygosity at noncoding markers such as microsatellite loci Heterozygosity at neutrally selected loci has been assumed to correlate with fitness only in inbred populations where it is expected to reflect genomewide heterozygosity which in turn should correlate with individual inbreeding coefficient Coulson et al 1998 Slate et al 2000 This mechanism is recognized as a ‘general effect’ of heterozygosity and it was initially suggested to explain a large majority of all HFCs reported for noncoding markers This interpretation was recently challenged by studies showing a weak association between the inbreeding coefficient and heterozygosity measured across a large number of neutrally selected loci Balloux et al 2004 Markert et al 2004 Slate et al 2004 An association between heterozygosity and inbreeding was suggested to occur only when the variation in individual inbreeding coefficients within populations is high Slate et al 2004 or when populations are strongly substructured Balloux et al 2004 However it seems that these specific conditions rarely occur in the wild and thus they are unlikely to explain most of the HFCs reported in empirical studies on vertebrates Pemberton 2004As an alternative to the ‘general effect’ hypothesis it was proposed that heterozygosity at neutral markers can have a ‘local effect’ on fitness assuming that it is correlated to heterozygosity at both linked and unlinked selected loci through genetic associations Ohta 1971 Hansson and Westerberg 2002 Szulkin et al 2010 This hypothesis however requires particular population structure and specific evolutionary or ecological circumstances such as small population size nonrandom mating population admixture or bottlenecks which can generate nonrandom associations of alleles at different loci known as linkage disequilibria Brouwer et al 2007 Szulkin et al 2010 Although linkage disequilibria were initially considered to be restricted to a narrow chromosomal segment around the target locus it is now realized that high levels of genomewide linkage disequilibria may occur in natural populations Slavov et al 2012 Hohenlohe et al 2012 suggesting that local heterozygosity effects could be much more widespread than previously thought Currently this view is gaining increasing empirical support with local heterozygosity effects reported in many vertebrate taxa AcevedoWhitehouse et al 2006 LieutenantGosselin and Bernatchez 2006 Tiira et al 2006 Charpentier et al 2008 including several species of birds Hansson et al 2004 Fossøy et al 2009 OlanoMarin et al 2011 Taking all these into account investigating the structure and consequences of local HFCs emerged as a new important goal of evolutionary biology reorienting basic questions of HFC studies LieutenantGosselin and Bernatchez 2006The prime objective of this study was to investigate correlations between heterozygosity at seven microsatellite loci and two fitnessrelated traits growth rate and nutritional condition in great cormorant Phalacrocorax carbo sinensis nestlings from a mediumsize inland colony in Poland We expected that genomewide inbreeding effects could be present in our population due to a severe bottleneck in the numbers of this treenesting subspecies P c sinensis that occurred throughout Europe in the middle of 20th century By the early 1960s the size of the entire northwestern European population was reduced to 800 pairs nesting in two Dutch colonies Goostrey et al 1998 whereas the Polish population was limited to only 150 pairs by the 1950s Głowaciński 1992 In the early 1970s the numbers of P c sinensis started to increase rapidly owing to a combination of relaxed human persecution availability of new colony sites and greatly improved food supply as the result of water eutrophication Hagemeijer and Blair 1997 reaching over 60000 breeding pairs in Northwest Europe by mid 1990s Van Eerden and Gregersen 1995 and ca 25000 breeding pairs in Poland at the beginning of 21th century Tomiałojć and Stawarczyk 2003 While this recent bottleneck episode could cause severe inbreeding depression across European populations we also acknowledge that it could generate considerable linkage disequilibria facilitating local effect of heterozygosity at neutral loci on fitnessrelated traits Thus the second aim of this study was to explore mechanisms underlying HFCs recorded within our great cormorant population by testing whether they reflect a genomewide general effect or a local effect of linkage disequilibriumThe study was conducted in the colony of great cormorants at Jeziorsko reservoir 51°73′N 18°63′E central Poland The colony was established in 1991 when 90 breeding pairs were recorded at the site Since then the size of the colony gradually increased reaching ca 500–600 pairs during the study period 2010–2011 and ca 800 pairs in 2013 Although the location and spatial organization of the colony changed over time the primary breeding habitat was the riparian willow woodland dominated by the white willow Salix alba and the grey willow Salix cinereaFor the purpose of this study we randomly selected 57 broods at the moment of hatching 29 broods in 2010 and 28 broods in 2011 As HFCs can be contextdependent and their magnitude may vary with environmental conditions and along the season Harrison et al 2011 our sampling period spanned over the whole main hatching period in the colony 24 and 22 days in 2010 and 2011 In the selected broods a minor fraction of all eggs was depredated during incubation 36  and 91  of eggs failed to hatch due to embryonic mortality or infertility In all these broods consisted of 252 eggs 220 of which produced hatchlings sampled for molecular analysis Blood samples for molecular analyses ca 10 μl were collected soon after hatching by puncturing the ulnar vein of nestlings No brood reduction occurred before sampling The samples were immediately suspended in 96  ethanol and stored until laboratory analyses When hatching of the whole brood was completed we collected the following measurements from all the hatchlings body mass ±1 g wing length ±1 mm culmen and tarsus length both ±01 mm Since great cormorants lay eggs in 1–3 day intervals clutch size of 3–6 eggs and start incubating from the first egg the hatching is asynchronous and broods typically contain chicks of different sizes To determine hatching order all measurements collected after hatching were reduced to the first principal component PC1 of the principal component analysis PCA PC1 accounted for 961  of the variability in all chosen variables and all body measurements had similar contributions to PC1 from 0251 to 0256 Hatching ranks were established based on size ranks assigned to each chick from PC1 values which followed methodology used in other species of waterbirds that exhibit marked hatching asynchrony eg Cash and Evans 1986 including several cormorant species Shaw 1985 Stockland and Amundsen 1988 All chicks were tagged on the tarsus with flexible Velcro™ strips of different colours These strips were enlarged according to the size of chicks during successive visits At the age of 13 days all chicks were marked with individuallynumbered metal rings and Velcro™ strips were removedBody mass of nestlings was repeatedly measured over the period of 22 days with 3 to 5day intervals resulting in at least five measurements per chick It was not possible to collect measurements near fledging as chicks older than 25 days may jump out of the nests if humans approach Platteeuw et al 1995 pers observ To calculate nestling growth rates we fitted logistic curves of the form y = A/1 + B × exp−KT to the measurements of body mass where y refers to the body measurement at age T A is an asymptotic value B is a constant of integration and K is the growth rate constant We used parameter K from the fitted curves as an indicator of chick growth rates Since we stopped collecting measurements when nestlings had not yet reached asymptotic values of their body mass the parameter A of the curve equation was constrained with the expected mean fledgling body mass of male 2379 g or female 1946 g great cormorant chicks Liordos and Goutner 2008

Keywords: