Low level of extra-pair paternity in an urban population of blackbirds

Abstract We studied extra-pair paternity in a socially monogamous bird, the European Blackbird (Turdus merula). The research was conducted in an urban park in Poland. We obtained genetic samples from 36 broods (123 nestlings) of 21 unique pairs. The proportion of broods with some extra-pair nestlings was 17.1% (6 broods), and overall, 13.1% (16 nestlings) of all nestlings were sired by extra-pair males. In one brood (2.8% of all the investigated broods), we found one egg laid by a female which did not have her own territory in the study area (brood parasitism). These figures are relatively low compared to other closely related passerines. We suggest that mate guarding in the studied population may effectively prevent females from engaging in extra-pair activity, and in the case of females, divorce and serial polygamy are better strategies for enhancing the genetic quality and genetic diversity of their offspring than infidelity.


Introduction
Knowledge of the genetic mating system is essential for understanding sexual selection in birds. According to the latest summary of studies (Brouwer & Griffith 2019), Extra Pair Paternity (hereafter EPP) has been detected in 76% of the socially monogamous avian species with biparental care (386 populations of 255 species), with high variations across species and populations.
Despite several decades of studies of EPP in birds (Burke & Bruford 1987), the question why some females have extra-pair offspring while others do not has yet to be satisfactorily resolved (Lifjeld et al. 2019). For males, the adaptive benefits of engaging in extra-pair activity are readily comprehensible, but the benefits to females remain unclear (Forstmeier et al. 2014;Nakagawa et al. 2015). For females engaging in extra-pair mating, such behaviour could be costly owing to the energy that would need to be expended when searching for optimal extra-pair mates (Dunn & Whittingham 2006), reduced investment by the social partner (Davies 1992;Matysioková & Remeš 2013) and the risk of sexually transmitted disease (Poiani & Wilks 2000). On the other hand, the benefits to such females can be divided into two categories: 1) direct benefits, such as extra-pair males providing females with extra food (Tryjanowski & Hromada 2005), protecting them against predators (Gray 1997), contributing to parental care (Townsend et al. 2010) and providing insurance against infertility (Sheldon 1994;Santema et al. 2020;Vuarin et al. 2021); 2) indirect genetic benefits, i.e. good paternal genes, genetic heterozygosity and/or genetic compatibility of parental genomes (Jennions & Petrie 2000;Foerster et al. 2003;Mays & Hill 2004;Griffith 2010;Arct et al. 2015). Finally, the genetic constraints hypothesis has been formulated in response to increasing evidence that extra-pair mating could occasionally be maladaptive to females. Accordingly, female extra-pair copulations may be the dynamic result of sexually antagonistic coevolution, where positive selection on males to sire extra-pair young will lead to increased extra-pair mating by females as a correlated evolutionary response (Griffith et al. 2002;Arnqvist & Kirkpatrick 2005;Forstmeier et al. 2011).
We present the results of a study of extra-pair paternity in the European Blackbird Turdus merula conducted in Szczecin (NW Poland). The European Blackbird (henceforth: Blackbird) is a model species with a well-known biology (Snow 1958;Glutz von Blotzheim et al. 1971-1982Stephan 1985;Cramp & Simmons 1988;Desrochers & Magrath 1996;Faivre et al. 2001;Evans et al. 2009), but paternity has only been studied once, by Creighton (2000), who used a small sample of broods and females (14 nestlings from seven broods of four females).
The Blackbird is a socially monogamous species with biparental care. The social mating system in the population studied here ranges from monogamy to polygyny (Wysocki 2004a). Social monogamy is prevalent, but polygamous matings are regular occurrences (monogamy -99.0% of 1231 broods, bigamy -0.4%, sequential polyandry -0.3%, sequential polygyny -0.1%, simultaneous polyandry -0.2%). The occasional simultaneous use of a nest by two pairs (Wysocki & Walasz 2004), successful polyterritorial polygyny (Wysocki & Jankowiak 2018) and one male that simultaneously attracted three females (the nest was finished by the third female) were also recorded. The divorce rate is high (within-season -Wysocki 2004b, between-season -Wysocki 2006. Analysis of copulation behaviour (Wysocki & Halupka 2004) suggested that extra-pair copulations were rare (< 1% of all observations). But as such behaviour may be concealed (Birkhead & Möller 1992), field observations might underestimate their real frequency. In addition, paternity could also be affected by cases of rape, frequently observed in Blackbirds (Creighton 2000, Wysocki -unpubl. data). Here, we provide an estimate of the proportion of Blackbird broods containing extra-pair young (EPY) in an urban population in NW Poland.

Materials and methods
The data for this study were gathered during research into the ecology of an urban population of blackbirds from 2016 up to and including 2020 in the Stefan Żeromski Park, Szczecin, NW Poland (53°260′N, 14°330′E). Situated in the city centre, this park (area 21.9 ha) is surrounded by streets and buildings. The density of the population ranged from 1.9 to 2.2 pairs/ha. For a more detailed description of the study area, see Cholewa et al. 2021. Over 90% of the breeding population (captured by mist net) were marked with combinations of four colour rings enabling individual identification without the need for re-capture. During the breeding season (March-August), 1-3 persons surveyed the study area every day to locate the pairs' territories, to track their nests and confirm the social paternity (birds feeding nestlings).
We obtained genetic samples from 36 broods (123 nestlings and both parents) of 21 unique pairs (20 females and 17 males, one female had two different partners in consecutive breeding seasons): 14 pairs were sampled once, 2 pairs twice (one was sampled in the same breeding season, and the second in two different breeding seasons), another 3 pairs three times (the first pair once in 2017 and twice in 2019; the second pair twice in 2018 and once in 2019, and the third pair twice in 2018 and once in 2020), one pair four times (two broods in 2016, one in 2017 and one in 2018), and one pair six times (one brood in 2017, two in 2018, two in 2019, and one in 2020).
The chicks were ringed in the nests between 6 and 8 days of age, so some chicks in some broods may have died before we were able to ring them. But since less than 5% of the nests we observed lost chicks between the 1 st and 8 th days of life (Wysocki, unpubl. data), we believe that the impact of such losses on the results of extra-pair paternity in the Blackbird is negligible. The trapping, ringing and marking of the birds were supervised by Dariusz Wysocki (ringing licence No. 390/2018 from the Polish Academy of Sciences). Permits for the survey as well as for feather and blood sampling were obtained from the Local Ethics Committee in Poznań (No. 6/2014 dated 23.04.2014; Poland).
The males and females were 2-7 years old (the median was 4 in males and 3 in females).
DNA was extracted from feathers and blood clots using the GeneMATRIX Tissue DNA Purification Kit (EURx Poland). Blood clots were separated manually from wipes, placed in LT buffer (EURx Poland) and incubated with Proteinase K overnight at 56°C. In the case of the feathers, 0.5 cm of the calamus was cut and crushed into smaller pieces using a sterile scalpel in a sterile Petri dish and also incubated in LT buffer with Proteinase K. After incubation, the extraction followed the standard tissue protocol. DNA extracts were assessed for their purity and DNA concentration using Nanodrop (Thermo Fisher Scientific). DNA extracts were stored at −20°C.
The multiplex reaction mixture contained 2 or 3 μl of DNA extract depending on the concentration (from 10 to 150 ng/µl), 2.65 μl or 3.65 μl of water depending on the amount of DNA, 7.5 μl of PCR Master Mix (QIAGEN Multiplex PCR Kit), 1.25 μl of Q-Solution and 0.6 μl of primer mix (concentration 0.7-0.95 μM, forward and reverse primer). The amplification conditions were as follows: initial denaturation for 15 min at 95°C, 40 cycles of 30s at 94°C, 90s at 57°C, 90s at 72°C and final elongation for 10 min at 72°C.
The results of the microsatellite loci amplification were checked using an ABI 3500Xl Genetic Analyzer. The genotyping was performed using a mixture of size standard [GeneScan™ 600 LIZ (Applied Biosystems)] with formamide and PCR product, which was denatured for 5 min at 95°C. We identified genotypes using GENEMAPPER v.4.1.
Basic statistical indices for genetic markers were obtained for 37 (20 females and 17 males) presumably unrelated individuals. For each locus, we assessed unbiased expected heterozygosity (H E ) and observed heterozygosity (H O ) using GenAlEx version 6.5 (Peakall & Smouse 2012). The statistical significance of the deviation from Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) was estimated for each locus using Genepop on the Web version 4 (Raymond & Rousset 1995;Rousset 2008).
Parameters of usefulness of the selected microsatellites in paternity verification were also estimated on 37 unrelated individuals and 123 nestlings. Specifically, we calculated F(null) (null allele frequency), the polymorphic information content (PIC), the combined non-exclusion probability when the first parent (female/mother in this case) is known (NE-1P) or second one of the opposite sex is known (NE-2P) using Cervus 3.0 (Kalinowski et al. 2007). The probability of exclusion (both parents known) (P1), probability of exclusion (only one parent known) (P2) and probability of exclusion (both parents excluded) (P3) were also calculated using GenAlEx.
First, we manually compared the microsatellite genetic profiles between adult birds and chicks to detect extra-pair young, namely for every nestling presumably paternal alleles (not inherited from mother) were compared with the genotype of social father. Nestlings with 2-6 mismatches in their genotypes with social father were classified as EPP chicks. Next, we applied Cervus 3.0.7 (Kalinowski et al. 2007) to compare genotypes of all adult birds (males and females) with genotypes of every young to assess the occurrence of EPP and identify extra-pair males. We used the following simulation parameters -confidence of 95% (strict) and 80% (relaxed) and 4 minimum typed loci -calculating the confidence with LOD (obtained by taking the natural log (log to base e) of the overall likelihood ratio). We simulated 1000 offspring, 50 candidate fathers and 50 candidate mothers. Next, we made parentage analysis for parent pair. After making simulations, we analysed results from Cervus. We compared their genotypes with their mother and potential father to checked that their genotypes confirm compatibility with parents or not. Manual method and Cervus are very good options to make parentage analysis and their efficient is very similar, but in one case when we find potential father with full compliance with chick, Cervus compare different father with one mismatch in his genotype.
To estimate the proportion of EPY among all the nestlings in the population, and the proportion of broods containing EPY, we used the resampling Extra-pair paternity of blackbirds method implemented in a Python script. From the 36 broods we drew 1000 random resamples, each consisting of broods of 21 unique pairs. For each resample, we calculated both statistics, as specified above. The grand averages of the resampling distributions represented point estimates of the parameters in question, and their 95% confidence limits were computed as 2.5 and 97.5 percentiles.

Results
All microsatellite loci appeared to be highly polymorphic in a sample of 37 unrelated individuals. Both expected and observed heterozygosity were high, exceeding 0.8 in the majority of loci (Table I). A statistically significant linkage disequilibrium (Bonferroni corrected P-value = 0.017) was found for only one among 28 comparisons (3.57% of all comparisons). Genotypes and potential fathers of the studied blackbirds are included in the supplement.
We also confirmed the usefulness of the analysed microsatellites in paternity studies: the markers had a very low estimated frequency of null alleles (in the majority of cases <3%) The combined non-exclusion probability for the first parent (both parents unknown) was 0.0203, and the combined non-exclusion probability for the second parent (one parent known) was 0.0015, which indicates that our microsatellite set is sufficient for detecting EPP and EPMs.
Cervus compared potential fathers (males sampled in the park) to our 17 nestlings from EPP nests. Our results confirmed extra-pair fathers for 9 nestlings from 4 EPP nests (Table S1). Two of the Extra Pair Males (hereafter EPMs) were nearest neighbours of the females, but in the case of the two others, their territories were 150 m and 250 m away from the females' territories. All the EPMs were older or the same age as the social partner.
We also detected one case of egg dumping in a single EPP nest (2.8%, 95% CI: 0.01 to 14.5%, of the investigated nests). This dumping was raised by another mother, because their genotype was different and we know that his biological mother didn't make a nest in this time. Assuming this event to be an example of within-species nest parasitism, the proportion of broods with some extra-pair nestlings remains the same (17.1%), but proportion of the nestlings sired by extrapair male seems to have lower proportion and falls to 13.1%.

Discussion
The proportion of broods and nestlings with EPY (17.1% and 13.1%) in the blackbirds in this population was smaller than average compared with other bird species (33% and 19%; reviewed by Brouwer & Griffith 2019) but similar to other socially monogamous bird species -18.7% of broods and 11.1% of offspring (Griffith et al. 2002). The population of blackbirds that we studied had a smaller proportion of broods and nestlings with EPY compared to other members of the genus Turdus. Thus, the respective statistics were 36.4% and 18.2% in the Whitenecked Thrush (T. albicollis) (Biagolini-Jr et al. 2016), 52.6% and 37.8% in the Clay-coloured Robin (T. grayi) (Stutchbury et al. 1998), and 71.9% and 48.1% in the American Robin (T. migratorius) (Rowe & Weatherhead 2007). We assume that because the density of this population is high (2 pairs per ha; , territories are relatively small, which may facilitate mate guarding. Both males and females occasionally undertook extra-territorial forays during the fertile period and solicited extra-pair copulations (Wysocki & Walasz 2004, Wysocki unpubl. data) but without apparent success. We also recorded females escaping from their home ranges that were followed by their social mates which were attempting to stop them. In such cases, however, the pair usually split up after a few days. The only successful EPC observed (during the 26 years of this long-term study and among several hundred within-pair copulations recorded) was solicited by a female when her partner was feeding fledglings from the previous brood. Both the within-and between-season divorce rates are very high (14.2% and 43.1% of pairs, respectively; Wysocki 2004aWysocki , 2004bWysocki , 2006. For females, therefore, divorce might be the preferred and most accessible tactic for enhancing the genetic quality and genetic diversity of their offspring. Analysis in Cervus clearly indicated one case of egg dumping in this population. Based on observational data, we conclude that such an event was highly probable, as the candidate mother, selected on the basis of genetic data (a female in her second year of life), was observed several times in the park but was never seen during nest-building. Conspecific brood parasitism is rarely observed in Blackbirds (from 0 to 3.1% of populations; Samas et al. 2014;Ruiz-Raya et al. 2016), but during our field work there were several occasions when a female incubating the eggs chased away another female approaching her nest, and in two cases we found two eggs laid on one day.

Disclosure statement
No potential conflict of interest was reported by the author(s).