Improvement on molecular sex identification primers for Passeriform bird species

Introduction

With an in-depth understanding of sex-determination mechanisms, gender identification techniques for birds have rapidly developed (McQueen et al., 2001; Ellegren, 2002; Nakagawa, 2004). Since it is impossible to sex many sexually monomorphic species visually or with simple measurements, molecular sexing by polymerase chain reaction (PCR) amplification provides for reliable methods that have been used widely in many species (Quinn et al., 1990; Ellegren and Sheldon, 1997; Griffiths et al., 1998). One of them is to amplify the sex-linked CHD genes (CHD-Z and/or CHD-W) by PCR (Ellegren, 1996; Lessells and Mateman, 1996; Griffiths et al., 1998). Using this method, specific primers anneal to the conserved exonic regions of CHD and amplify across an intron with various lengths. The PCR products from the Z and W chromosomes can be discriminated after agarose electrophoresis, with males showing one band and females two bands.

As the most diverse order in Aves, the biology, ecology, ethology, evolution and adaptation of Passeriformes have been widely studied (Saino et al., 2008; Woxvold and Magrath, 2008). Saino et al. (2008) found that in enlarged broods of Hirundo rustica, which may result in harsh rearing conditions, the average phenotypic quality of nestlings was depressed for those with a male-biased sex ratio. In a study of the cooperatively breeding apostlebird, Struthidea cinerea, Woxvold and Magrath (2008) suggested that the sex ratio was biased according to the hatching order; mothers in small breeding groups produced significantly more males among the first-hatching brood, presumably because males were more helpful than females during the breeding period. Too many studies have been performed to describe them in detail, but a common feature is the use of molecular sexing methods to identify their sex based on the CHD genes (Ellegren and Sheldon 1997; Saino et al., 2008; Woxvold and Magrath, 2008).

Griffiths et al. (1998) designed the primer pair P2/P8 based on the CHD genes of the domestic chicken. Since then, many other primers have been developed for gender identification of birds, such as 2550F/2718R (Fridolfsson and Ellegren, 1999; Dubiec and Zagalska-Neubauer, 2006; Woxvold and Magrath, 2008), 1237L/1272H (Kahn et al., 1998) and sex1/sex2 (Wang and Zhang, 2009). The primer pair sex1/sex2 was designed from the CHD genes of the Brown-eared Pheasant (Crossoptilon mantchuricum) and can be used to sex many other pheasants accurately as well as some Passeriform species (Wang and Zhang, 2009). In this study, we mainly discuss the efficacy of sex1/sex2 in determining the sex of Passeriform species and describe how to improve them for gender identification.

Materials and methods

Feathers, egg shell membranes, muscles, or blood from 99 individuals belonging to 17 bird species in ten families were collected either in the field or from zoos (see details in Table 1). Genomic DNA was extracted by using DNA extraction kits (Trans Biotech, Beijing, China) and was diluted to 50–100 ng·L^–1 after determining the concentration spectrophotometrically (DU-640, Beckman Coulter GmbH, Germany). DNA extracted from feathers was used directly without dilution for PCR because of its low concentration.

Table  1.  Sex identification of 17 Passeriformes species using four pairs of primers

Family	Species	Source a (sampling location)	Traditional diagnosis	No. (M/F)
Passeridae	Passer montanus	M (museum of BNU)	Sequencing CHD-Z/W fragments	23 (12/11)
	Lonchura striata	B (Beijing, bird market)	Male song	10 (6/4)
Paridae	Parus major	M (DNR)	Sexual dimorphism	2 (1/1)
	Parus palustris	M (museum of BNU)	Record of the specieman tag	2 (1/1)
	Periparus venustulus	M (DNR)	Sexual dimorphism	3 (2/1)
Hirundinidae	Riparia riparia	M (museum of BNU)	Examination of the gonads	1 (0/1)
Turdidae	Luscinia svecicus	R (Tianjin Zoo)	Sexual dimorphism	2(1/1)
	Monticola gularis	R (Tianjin Zoo)	Sexual dimorphism	2 (2/0)
Aegithalidae	Aegithalos concinnus	B (DNR)	Observation of mating behavior	16 (9/7)
Laniidae	Lanius schach	M & E (Hainan)	Sequencing CHD-Z/W fragments	8 (4/4)
	Lanius fuscatus	M (Hainan)	Sequencing CHD-Z/W fragments	4 (2/2)
	Lanius cristatus	M (Hainan)	Sequencing CHD-Z/W fragments	5 (4/1)
Corvidae	Corvus corone	M (museum of BNU)	Sequencing CHD-Z/W fragments	4 (2/2)
	Cyanopica cyana	M & E (campus of BNU)	Observation of mating behavior	11 (5/6)
Fringillidae	Eophona migratoria	M (museum of BNU)	Sexual dimorphism	2 (1/1)
Emberizidae	Emberiza aureola	R (Tianjin, bird market)	Sexual dimorphism	2 (1/1)
Muscicapidae	Ficedula zanthopygia	R (Tianjin, bird market)	Sexual dimorphism	2 (1/1)
10 families	17 species			99
M, muscle; E, egg shell membrane; R, rectrices; B, blood; No., number of individuals; M/F, proportion of males to females. a Source means the sources of samples. BNU, Beijing Normal University; DNR, Dongzhai National Reserve (31°28′–32°09′N, 114°18′–114°30′E).

 | Show Table

DownLoad: CSV

The primer pairs P2/P8 (Griffiths et al., 1998) and sex1/sex2 (Wang and Zhang, 2009) were both used for sex identification of these Passeriformes of known gender (confirmed by traditional sex diagnoses, such as external morphology, behavioral observations, and post mortem examination of the gonads, Table 1). We used a 20-L PCR volume containing 100–200 ng DNA, 0.5 M of each primer, 10× PCR buffer, 2.0 mM MgCl₂, 0.2 mM of dNTP mix, and 0.75 U Taq DNA polymerase (all reagents were from Takara, Japan). The thermal cycling conditions for both primer pairs were initial denaturation at 94℃ for 3 min, then 35 cycles of denaturation for 30 s at 94℃, annealing for 45 s at 50℃ (we used constant temperature here for comparison, while better results could be obtained by adjusting the annealing temperature to fit different species) and extension for 50 s at 72℃, followed by a final extension for 10 min at 72℃. After electrophoresis on 3% agarose gels, we found different amplification efficiencies of sex1/sex2 in various samples, i.e., some samples showed preferential amplification of the longer CHD-W fragments. In order to explain this outcome, we sequenced the CHD fragments amplified by P2/P8 (the annealing sites of sex1/sex2 were within those of P2/P8) from some representative samples, such as Passer montanus (GU370350, GU370351), Corvus corone (GU370346, GU370347), and Lanius schach (GU370348, GU370349), at BGI Life Tech, Beijing, China.

Results

After aligning the annealing sites of the sex1/sex2 primers with the sequenced CHD fragments by Mega 3.1, we found that the last base pair at the 3′-end in sex1 and four base pairs in sex2 did not match with CHD-Z, but rather corresponded to the CHD-W fragments. To avoid unnecessary preferential amplification, we modified sex1 by deleting three base pairs at the 3′-end (referred to as sex1′) and substituted the four unsuitable base pairs in sex2 for adjacent base pairs (referred to as primer sex-mix: 5′-CCTTCRCTKCCATTRAAGCTRATCTGGAAT-3′, where R refers to G/A and K refers to G/T). Primer sets were then recombined as sex1′/sex2 and sex1′/sex-mix to re-determine the sex of sampled Passeriform species (Table 1).

Figure 1 shows the electrophoresis of the resulting PCR products amplified by the four primer pairs (P2/P8, sex1′/sex-mix, sex1′/sex2, and sex1/sex2) based on the same samples. Preferential amplification was observed either for CHD-Z or CHD-W when using P2/P8, sex1′/sex-mix, or sex1/sex2; while for sex1′/sex2, preferential amplification did not occur frequently. In total, 99 individuals belonging to 17 species in ten families of Passeriformes were sexed with different primer combinations (Table 1). The successful rates of sex determination tested with P2/P8, sex1/sex2, sex1′/sex-mix, and sex1′/sex2 were 85.6%, 89.6%, 92.7%, and 98.9%, respectively.

Figure  1.  Sex identification of Passeriform species using different primer pairs (the same number indicates the same individual). 1, Luscinia svecica; 2, Luscinia calliope; 3, 4, Emberiza aureola; 5, 6, Passer montanus; 7, 8, Corvus corone; M, 100-bp DNA ladder. Females: 1, 3, 5, 7; Males: 2, 4, 6, 8. The birds were sexed using (a) P2/P8, (b) sex1′/sex-mix, (c) sex1/sex2, and (d) sex1′/sex2.

DownLoad: Full-Size Img PowerPoint

Discussion

Sex identification techniques have undoubtedly benefited studies on evolutionary biology, ecology, and conservation genetics of birds (Saino et al., 2008; Woxvold and Magrath, 2008; Cockburn and Double, 2008). The exploration of sex ratio adjustment in birds has attracted much interest, but not all studies have shown adaptive patterns and some have even countered the predictions of existing theories (Frank, 1990; Bensch et al., 1999; Cockburn and Double, 2008). Studies with inconsistent results are difficult to interpret because they may not fully consider all possible constraints (West and Sheldon, 2002). Of particular relevance to our study on the use of primers in sex identification, many avian sex-allocation studies have identified the sex of the offspring late in their development, obtaining the so-called primary sex ratios, that may have experienced differential embryo or chick mortality due to chance environmental effects (Fiala, 1980). However, the problem of determining the sex of morphologically indistinguishable young birds and embryos has now been resolved by genetic techniques that allow the sexing of day-old nestlings and of embryos recovered from eggs, minimizing the confounding effect of later differential mortality among chicks (Griffiths et al., 1998; Fridolfsson and Ellegren, 1999). With the optimization of molecular sexing primers, the accuracy of various studies can expect to be greatly improved in the future.

In our study, the primer pairs P2/P8 and sex1/sex2 were both used for sex identification in Passeriformes. Regardless of their wide application (Ellegren and Sheldon, 1997; Whittingham and Dunn, 2000; Dreiss et al., 2006), certain deficiencies remain with each primer pair. We needed a high concentration of primers and high-quality DNA when using P2/P8, without which the preferential amplification of CHD-Z fragments occurred (Fig. 1a). However, the primer pair sex1/sex2 can generally cause the preferential amplification of CHD-W fragments (Fig. 1c, except in L. striata) because of their characteristics, i.e., they match CHD-W more closely than CHD-Z in the Passeriformes (at most a 2-bp difference in CHD-W versus at least a 4-bp difference in CHD-Z). We paired the modified sex1 (sex1′) with either sex2 or sex-mix, and found that sex1′/sex-mix caused evident preferential amplification of CHD-Z fragments in Aegithalos concinnus, C. corone, P. montanus, Emberiza aureola, Luscinia svecicus, Ficedula zanthopygia, and L. striata (see examples in Fig. 1b). We speculated that when using sex1′/sex-mix, the CHD-W fragments would hardly anneal with these primers, while this same combination can result in the preferential amplification of shorter CHD-Z fragments. Given that the preferential amplification of CHD-Z fragments may cause a more serious misidentification problem than that of CHD-W fragments that can at least imply the female sex, we ultimately chose sex1′/sex2 for sexing because sex1′ corresponds to the exons of both the CHD-W and CHD-Z fragments (i.e., no biased annealing to CHD-W occurs), and sex2 corresponds only with the exons of CHD-W fragments in Passeriformes (eliminating the potential preferential amplification of CHD-Z). Therefore, we postulated that limited preferential amplification of both CHD-Z and CHD-W fragments would occur. The results shown in Fig. 1d strongly support our hypothesis. Given that the differences in the successful sex rate were mainly caused by preferential amplification in our study, we recommend that the primer set sex1′/sex2 is a better choice as molecular sex identification primers. Nevertheless, the primer set sex1/sex2 is also useful in certain species with preferential amplification of CHD-W (whose lengths, much different from that of CHD-Z, would be preferred) because the appearance of CHD-W implies the female sex, and researchers can document the variable lengths of CHD-W from known female birds.

In total, we sexed 99 individuals from 17 Passeriform species using different primer combinations (Table 1). After comparing the products amplified by the different primer sets, we found that sex1′/sex2 showed clear, reliable results when sexing Passeriformes. Presently, they are being frequently used in population studies of Aegithalos concinnus, which shows cooperative breeding behavior (Hatchwell and Russell, 1996; McGowan et al., 2003) and we are interested in exploring the sex allocation in different breeding groups. Moreover, the sex identification of Lonchura striata (a model bird in studying the learning activity of songs in the Physiology Lab of Beijing Normal University) is also benefiting from the use of sex1′/sex2. In the near future, we can expect more applications of this primer pair to further studies in Passeriformes.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 30570234, 30330050). We thank Drs. Geoffrey Davison, Xiangjiang Zhan, and Yang Liu for their helpful comments on the manuscript.