Extra-pair paternity in a Neotropical rainforest songbird , the White-necked Thrush Turdus albicollis ( Aves : Turdidae )

Over the last two decades, several studies have shown that the mating systems of various birds are more complex than previously believed, and paternity tests performed with molecular techniques have proved, for instance, that the commonly observed social monogamy often presents important variations, such as extra-pair paternity. However, data are still largely biased towards temperate species. In our study, at an area of the Brazilian Atlantic Forest, we found broods containing at least one extra-pair young (EPY) in the socially monogamous White-necked Thrush Turdus albicollis (Vieillot, 1818). Paternity tests using six heterologous microsatellite loci revealed that four of 11 broods (36.4%) presented at least one extra-pair young (EPY). This rate of EPY is within the range found for other studies in the tropics. This is one of the few studies that present detailed paternity analyses of a Neotropical rainforest passerine. Our findings corroborate the early insights that breeding strategies involving cheating can also be widespread among Neotropical socially monogamous songbirds.

Over the last two decades, a number of studies have demonstrated that the mating systems of various birds are more complex than previously believed (reviewed in Griffith et al. 2002).For instance, paternity tests using molecular techniques have documented that different levels of extra-pair paternity (EPP) do occur.EPP can be defined as the proportion of offspring sired by only one social parent, usually the female, or the proportion of broods containing at least one extra-pair young (EPY) (Westneat et al. 1990, LiGon 1999, Griffith et al. 2002, neodorf 2004).The occurrence of EPP and EPY appear to be widespread among passerine birds (Westneat et al. 1990, Griffith et al. 2002, neodorf 2004), yet this information is largely biased towards populations from temperate zones (stutchbury & Morton 2008, Macedo et al. 2008), since more research on this subject has been conducted in that region.The Neotropical region holds almost half of the world's avifauna (Myers et al. 2000, haWkins et al. 2003), and Neotropical bird species differ from temperate species in many important aspects.For instance, they present reduced clutch sizes (Martin et al. 2000, Jetz et al. 2008) As the lack of parentage studies on Neotropical birds limits broad conclusions about passerine social systems, here we used microsatellite molecular markers to perform paternity tests in a population of the White-necked Thrush, Turdus albicollis (Vieillot, 1818), from a well-preserved Brazilian Atlantic Forest area.Our specific objective was to investigate the occurrence of EPY in this Neotropical songbird.
Field work was conducted at Carlos Botelho State Park (24°04'S 47°58'W), São Paulo State, southeastern Brazil.This Park holds 37,644 ha of Atlantic Rain Forest, and together with a set of adjacent conservation units, comprises one of the largest Atlantic Forest remnant in Brazil, with more than 1 million ha.

SHORT COMMUNICATION
Extra-pair paternity in a Neotropical rainforest songbird, the White-necked Thrush Turdus albicollis (Aves: Turdidae)  Whenever we found a nest we marked its location with a GPS (Garmin 62ST), and when nestlings were about five days old we sampled a drop of blood (10 µL) from each bird through a small incision at the tip of a claw.We also placed one mist net close to each nest to capture the parents when they flew into the nests to feed the nestlings, and blood samples were also obtained from them following the same procedure.Blood was stored in 1.5 ml tubes in 100% ethanol, and each animal received a unique combination of polyetilene colored bands for subsequent identification of adults by visual observations.After blood sampling and marking, nests were monitored by performing daily one hour focal observation sessions to confirm that we had sampled adults that were the social parents of the nestlings, and not individuals that happened to be close to the nests.
The DNA was extracted using a phenol-chloroform-isoamyl alcohol protocol (saMbrook et al. 1989), and animals were sexed by amplification of the homologous copies of the CHD gene (chromo-helicase-DNA-binding) using the primers P2/P8 (Griffiths et al. 1998).We also used the primer (P0) (han et al.  1).PCR reactions had a final volume of 10 µl, containing 0.2 mM of dNTPs, 1X amplification buffer, 3 mM MgCl2, 0.2 mM of each primer, 1 U Taq polymerase, and 100 ng of DNA.The reactions were performed in an Eppendorf thermocycler programmed for an initial denaturation at 94°C for 3 minutes, and 29 cycles of 94°C for 30 seconds, 30 seconds in the annealing temperature (see Table 1), and 72°C for 30 seconds, followed by a final extension at 72°C for 5 minutes.As allelic patterns could be clearly determined, alleles were scored for each locus in 7.5% polyacrylamide gels (schWarzová et al. 2008), run at 47 W during 1:40 hours, and results were visualized by silver staining (coMincini et al. 1995).We estimated the observed and expected heterozygosity, probability of heterozygosity deficit, and the probability of linkage disequilibrium using the software GENEPOP 4.0 (rayMond & rousset 1995).We also evaluated the possibility of null alleles, allelic dropout, and scoring errors due to stuttering using Micro-Checker (van oosterhout et al. 2004).The accuracy of the loci to perform the paternity tests was estimated by the probability that two random individuals in the population could present identical allelic composition (the probability of identity), and by calculating the probability that the set of loci could not exclude a pair of candidate unrelated parents from an random offspring, using Cervus 3.0 (kaLinoWski et al. 2007).All the above estimates were performed using only the adult individuals.The identifications of EPY were conducted through direct observation of allele inheritance (fLeischer 1996, Mitrus et al. 2014), and resulted when at least one young in a nest was sired by at least one extra-pair parent based on at least two loci.This is because mismatching at only one locus can be potentially caused by mutations or null alleles (Westneat & Mays 2005, Liu et al. 2015).We collected blood samples from 22 nestlings and 22 parents from 11 nests.The number of sampled nestlings per nest were 1 (n = 2 nests), 2 (n = 7), or 3 (n = 2) -in one nest with two eggs, only one egg hatched.The number of alleles per loci varied from 3 to 6 and none of the loci pairs were significantly linked.We also did not detect heterozygosity deficit (Table 2), null alleles, allelic dropout or scoring errors due to stuttering.The probability of identity of two random individuals in the population was 0.04%, and the probability that the set of loci could not exclude an unrelated pair of parents was 2.5%.
We detected cases of EPY in four of the 11 sampled nests (36.4%), which included one nest with one nestling, two cases in nests with two nestlings, and one case in a nest containing three nestlings.In three of these four nests we confirmed the social parents feeding the offspring.Raw genotypic data and allelic mismatches among nestlings and their social parents are presented in Appendix 1.
Cases of extra-pair mating have been reported for all species of Turdus studied so far.In a population of Clay-coloured Thrush, Turdus grayi (Bonaparte, 1838), from Panama (09°N), 52.6% of the broods contained EPY (stutchbury et al. 1998), and in a population of the American Robin, Turdus migratorius (Linnaeus, 1766), from Illinois, USA (40°N) this figure was 72% (roWe & Weatherhead 2007).Although we have analyzed a smaller number of broods, the frequency of EPY (36.4%) found for White-necked Thrush was smaller than these studies.However, this still can be considered high, as in Griffith et al. (2002) review 35% of 34 socially monogamous passerine species have exhibited EPY rates above 36.4%,indicating a potential phylogenetic effect within the Turdus.
A possible explanation for the exceptionally high levels of EPY in American Robin can involve breeding synchrony, a parameter commonly evoked to explain variations of EPY in passerine birds (Westneat et al. 1990, Griffith et al. 2002, neodorf 2004).When breeding is synchronic many females in a population are fertile at the same time and males may have limited opportunities to obtain extra-pair females, resulting in lower rates of extra-pair copulations (Griffith et al. 2002(Griffith et al. , neodorf 2004)).Although synchrony has not been estimated for this set of species, inferences can be made based on the duration of breeding seasons, i.e., populations or species with shorter breeding seasons should present more synchronic breeding (stutchbury & Morton 1995).We have observed active nests of White-necked Thrush during a period of four months and at least one marked pair nested twice in the season.Clay-coloured Thrush breeds during approximately three months in Panama (stutchbury et al. 1998), and they also can nest twice (dyrcz 1983).The breeding season of American Robin also lasts about four months (roWe & Weatherhead 2007), and they typically nest twice in the season (hoWeLL 1942(hoWeLL , younG 1955)).These data do not provide strong support for the idea that longer breeding seasons results in higher EPY rates in tropical species of Turdus.

2
Altitude varies from 30 to 1,003 m and annual rainfall varies from 777-2,264 mm (average 1,676 mm) (beisieGeL & Mantovani 2006).We searched for nests from September to February during two breeding seasons, 2013/2014 and 2014/2015, by monitoring approximately 7 km of trails, and 3 km of streams, in a sub-montane area varying in altitude from 680 to 850 m (oLiveira-fiLho & fontes 2000).To find the nests we monitored territories defended by males (Martin & GeupeL 1993) in areas of primary forest.The White-necked Thrush is a socially monogamous passerine (snoW & snoW 1963), endemic of the Neotropics, distributed in two disjunct regions: from Mexico to northern Argentina, and from northeastern Brazil to Uruguay (ridGeLy & tudor 1989, coLLar 2005).Surveys in the Atlantic Forest have shown that this is one of the most abundant species in this biome (siMpson et al. 2012, antunes et al. 2013), and the most frequently recorded using mist nets in this forest understory (aLves 2001).It builds bulky cup-shaped nests, mainly in large tree forks, invaginations of tree trunks, or inside bromeliads (snoW & snoW 1963, coLLar 2005).

Table 1 .
Microsatellite loci used for paternity tests in Turdus albicollis, primer sequences, annealing temperature (T), and source.