Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Nov 9, 2004; 101(45): 15967–15969.
Published online Nov 1, 2004. doi:  10.1073/pnas.0405126101
PMCID: PMC528742

Chicken W: A genetically uniform chromosome in a highly variable genome


The Y chromosome of organisms with male heterogamety is expected to show reduced levels of genetic diversity, because the effective population size is one-fourth that of autosomes. However, studies in mammals, flies, and plants show that Y chromosome diversity is lower than expected even when differences in effective population size are taken into account. This may be explained by skewed reproductive success among males, leading to low male effective population size, or by a strong role of selection in shaping levels of nucleotide diversity in nonrecombining chromosomes. We tested these hypotheses in a system with female heterogamety by estimating nucleotide diversity in the female-specific W chromosome of the domestic chicken by resequencing of 7,643 base pairs in 47 birds from 10 highly divergent breeds. The screening revealed only one single segregating site, which is in sharp contrast to our previous observation, using a similar panel of birds of, on average, one segregating site every 39 base pairs in autosomal sequence. When taking sex-specific mutation rates and differences in effective population size into account, the observed degree of W chromosome polymorphism is 28-fold lower than expected for the frequency of segregating sites and 13-fold lower than expected for estimates of nucleotide diversity (autosomes, 6.5 × 10–3; W, 7.0 × 10–5). We note that selection is the only factor that can explain the reduced diversity in the sex-limited chromosome irrespective of mode of reproduction or whether there is male or female heterogamety. Reduced variability in female-specific W chromosomes is not easily explained by sexual selection.

Polymorphism levels vary among different regions of the genome. For but one reason, positive and negative selection will tend to decrease levels of genetic diversity at affected loci and at linked sites (1). By estimating local levels of nucleotide diversity (π ; ref. 2), inferences about selection and the evolutionary history of particular genomic regions can be made. On a larger scale, polymorphism levels may be expected to differ among autosomes and sex chromosomes. This derives from the fact that the amount of standing genetic variation is directly related to the effective population size (Ne; ref. 3), which varies between autosomes and sex chromosomes. Specifically, and by assuming random mating in an ideal population, the Ne of autosomes, the X chromosome, and the Y chromosome show a 4:3:1 relationship (4), and polymorphism levels are expected to scale accordingly (1, 3).

Nucleotide diversity in the nonrecombining region of the human Y chromosome has been estimated at 1.0–1.5 × 10–4 (5, 6), which is ≈20% of the autosomal average (6). However, this is lower than the 25% reduction predicted from a 4:1 relationship in Ne (4) and definitely lower if one also takes into account that the Y chromosome is exposed to more mutations than autosomes, a consequence of male-biased mutation (7, 8). Moreover, lower-than-expected levels of Y chromosome diversity have also been found in a number of other mammals (9, 10), as well as in Drosophila (11) and in the dioecious plant Silene latifolia (12). As a somewhat extreme example, a recent study of domestic horses showed complete monomorphism in 14 kb of noncoding Y chromosome sequence, which is in sharp contrast to diversity levels in other regions of the equine genome (13).

The overall trend of low levels of nucleotide diversity on the Y chromosome calls for a general explanation. In theory, reduced Y chromosome diversity may be due to several different factors (1). First, at skewed mating, where the operational male-to-female sex ratio is lower than unity, Ne of the Y chromosome will be <25% that of autosomes (4, 14). This is expected in domestic animals like the horse, where a limited number of stallions may have been used during domestication (13), but is also likely for natural populations of polygamous species. Second, as a nonrecombining chromosome, Y is sensitive to the effects of selection at any locus on the chromosome. For instance, positive selection in the form of selective sweeps has the potential to drive particular Y chromosome haplotypes to fixation (15). Also, purifying selection leading to background selection at linked loci can reduce levels of Y chromosome variability (16), although it is possible that when gene density is low, the deleterious mutation rate may not be high enough for background selection to reduce diversity to such an extent as that observed (11, 16). Third, sex-biased dispersal can affect the Ne among sex chromosomes and autosomes (17). For instance, when dispersal is male-biased, Ne for the Y chromosome will tend to increase (5).

A useful way of approaching these questions may be to study the female-specific W chromosome of organisms with female heterogamety (18), e.g., in birds (males, ZZ; females, ZW). The W chromosome has an expected 1:4 Ne relationship to autosomes, analogous to Y, so we can make similar default predictions for relative levels of W chromosome polymorphism (everything else being equal, a 25% reduction compared to autosomes is expected). As a nonrecombining chromosome, W is sensitive to the effects of selection. This means that if selection plays an important role in W chromosome diversity, polymorphism levels should be reduced. However, W is not directly affected by lowered male Ne arising from skewed reproductive success among males. In fact, when male Ne is low, the expected relationship between W and autosomes is less pronounced than 1:4, and slightly elevated levels of W polymorphism, compared to null expectations, may be expected.

We have recently characterized levels of nucleotide diversity in the genome of the domestic chicken using a panel of birds from divergent lines and breeds, selected to represent a wide range of the domestic chicken gene pool (19). This screening revealed extensive genetic variability with π for autosomal, noncoding sequence estimated at 6.5 ± 0.3 × 10–3 (Z-linked, 2.0 ± 0.1 × 10–3) and with on average one segregating site every 39 bp. Polymorphism in chicken is thus at higher levels, for instance, than in humans, and it is noteworthy that such extensive diversity implies that Ne of chicken is large (>100,000), giving no indication that domestic chickens descended from a small ancestral population. Here we use birds from the same panel of chicken breeds to estimate levels of nucleotide diversity in noncoding DNA of the female-specific W chromosome, with the aim of contrasting the degree of genetic variability on autosomes and the sex-limited chromosome in a system of female heterogamety.

Materials and Methods

Polymorphism screening was based on resequencing of 13 introns from three W-linked genes in unrelated female birds. Surveyed sequences (GenBank accession nos. are given in parentheses) included SPINW [intron 3 (AY628538)], CHD1W [intron 7, (AY628533), intron 10 (AY628478), intron 11 (AY628483), intron 12 (AY628488), intron 13 (AY628493), intron 15 (AY628498), intron 21 (AY628508), intron 24 (AY628518), and intron 25 (AY628523)], and UBAP2W [intron 1 (AY628448), intron 2 (AY628453), and intron 4 (AY628458)]. The three genes are all single-copy and thus are not sensitive to possible intrachromosomal gene conversion events. Only a handful of genes have so far been identified on the chicken W chromosome.

PCR reactions were performed in 20-μl volumes on a PerkinElmer 9600 Thermal Cycler by using 0.5 unit of AmpliTaq Gold (Applied Biosystems), 2.5–3 mM MgCl2 (Applied Biosystems), 0.08 mM dNTPs, 1× PCR Gold Buffer (Applied Biosystems), 5 pmol of each primer, and 50 ng of template DNA. Primer sequences and PCR conditions are described in Table 2, which is published as supporting information on the PNAS web site. PCR products were purified with ExoSAP-IT reagent (Amersham Biosciences) followed by direct sequencing in forward and reverse directions by using the DYEnamic ET Dye-Terminator Kit (Amersham Biosciences). Reactions were electrophoresed on a MegaBACE 1000 sequencing instrument (Amersham Biosciences).

Sequences were edited in autoassembler (Applied Biosystems), and overlapping forward and reverse sequences were compared to obtain a consensus sequence for each bird and intron. The screening panel consisted of 47 female birds of 10 breeds, including the red jungle fowl, the wild ancestor of the domestic chicken (Table 1). This subset of breeds has been identified to constitute a broad representation of the total gene pool of the domestic chicken (20). One male sample was always included to control for W-specific amplification. For calculation of nucleotide diversity, we used dnasp 4.0 (21). The Hudson–Kreitman–Aguadé (HKA) test (22) was performed by using the hka computer program written by Jody Hey and available at http://lifesci.rutgers.edu/~heylab/index.html.

Table 1.
Breeds of chicken used for polymorphism screening and distribution of the two identified haplotypes

Results and Discussion

A total of 7,643 bp of noncoding sequence from 13 introns of three W-linked genes (CHD1W, UBAP2W, and SPINW) were surveyed. The results of this screening were in sharp contrast to those previously obtained for autosomal sequence. Only one single segregating site (a C to A substitution in position 389 of CHD1W intron 11, rare allele frequency of 0.43; Table 1) was identified on the W chromosome, with nucleotide diversity = 7.0 ± 0.0 × 10–5. Because the same breeds, the same number of chromosomes, and the same methodology have been used for screening different parts of the chicken genome, differences in polymorphism levels can be quantified by comparing the present data with the results from ref. 19. The frequency of single-nucleotide polymorphisms on autosomes (1 of 39 bp) is ≈200 times higher than on W (1 of 7,643 bp), and estimates of nucleotide diversity for autosomal variability (6.5 × 10–3) are ≈90-fold higher than W chromosome variability.

Ne and the mutation rate directly govern the amount of nucleotide diversity in neutral DNA sequences (3). Divergence in noncoding and likely neutral orthologous sequence of chicken and turkey can be used as an estimator of the mutation rate in galliform birds. For autosomal sequence, chicken–turkey divergence is 10.08%, whereas for the W chromosome, it is 5.74% (23), i.e., a mutation rate 1.75 times higher in autosomes. With the combined effect of a 4-fold difference in Ne (assuming random mating), we thus expect polymorphism levels to be 7 times (4 × 1.75) higher on autosomes than on the W chromosome. Therefore, our data indicate that genetic variability is 13 [nucleotide diversity, (6.5 × 10–3/7.0 × 10–5)/7] to 28 [frequency of segragating sites, ((1/39)/(1/7,643))/7] times lower on the domestic chicken W chromosome than expected from neutral predictions. An HKA test, which takes differences in Ne and divergence into account, shows that the ratio of polymorphism levels on autosomes and the W chromosome deviates significantly from a neutral model (P = 0.003).

A smaller male than female Ne, one of several factors suggested to reduce Y chromosome variability in mammals (6), cannot explain these observations. In the wild ancestor of the domestic chicken, the red jungle fowl, there is high variance in male reproductive success with a lower Ne of males than of females (24, 25). Moreover, poultry breeding is associated with highly skewed reproductive success among cocks, and this is likely to have been the case throughout the chicken domestication process (26). The effect of these types of mating system on W chromosome variability would be elevated levels of polymorphism, which is the opposite of what we observe.

Our data seem best explained by a strong role of selection in shaping levels of W chromosome variability in the domestic chicken. In the absence of recombination outside the pseudo-autosomal region, the entire W chromosome forms a single haplotype and is thus affected by selection at any locus. If selection causes low levels of W chromosome variability in the domestic chicken, this is particularly interesting for at least two reasons. First, because the Y chromosome contains male-specific (in mammals, mainly testis-specific; ref. 27) genes, it has been suggested that sexual selection is a major force driving Y chromosome evolution, including polymorphism levels (28). However, low levels of genetic variability in the female-specific W chromosome of chicken cannot easily be explained by sexual selection arising from male–male competition, and hence it indicates that sexual selection is not required for genetic variability in the sex-limited chromosome to be reduced. To our knowledge, there is no evidence of reversed sex roles leading to sexual selection among females in the domestic chicken.

Second, what could be the targets of selection on W? This question is motivated by the fact that the avian W chromosome is gene-poor and the limited number of genes so far identified mainly represent gametologous copies of genes shared with the Z chromosome (29), reflecting a common ancestry of the Z and W sex chromosomes before their divergence. Although independently evolving, most of these genes on W are highly similar to their Z chromosome gametologs, with very low rates of nonsynonymous substitution between gametologous copies (unpublished work) and little evidence of evolution of female-specific function. This would suggest that these W-linked genes are rarely subject to positive selection. However, there is at least one exception (HINTW; ref. 30), and there may be, of course, yet unidentified genes on W that have evolved female-specific function and that may form targets for directional selection. One possibility is that domestication has implied selection for traits encoded by female-specific genes on the chicken W chromosome. However, with the exception of sex, no phenotypic trait has been mapped to the domestic chicken W chromosome despite the ease by which sex linkage can be revealed in traditional breeding programs (26). Moreover, our screening included five females of the red jungle fowl, the wild ancestor of the domestic chicken. The W chromosomes of these fowls were as genetically depauperate as those of the domestic hen that we analyzed, indicating that loss of W chromosome variability predates domestication. Furthermore, preliminary screenings for W chromosome polymorphism in several natural populations of wild bird species have also suggested low levels of genetic variability (31), indicating that reduced W chromosome variation is not a consequence of domestication.

One interesting aspect of reduced levels of W chromosome variability in avian genomes is that the concomitant reduction in mtDNA variability should be expected. mtDNA and W chromosome haplotypes segregate clonally in birds (32), so lowered Ne of W should lower Ne of mtDNA, and vice versa. Testing this prediction would require comparing the overall levels of mtDNA variability in organisms with female and male heterogamety, respectively, while at the same time controlling for species-specific differences in Ne and mutation rate, something that may prove difficult.


The observations of lower-than-expected levels of genetic variability in the Y chromosome of organisms with male heterogamety and in the W chromosome of the domestic chicken with female heterogamety indicate that sex-limited chromosomes generally possess limited genetic diversity. Nonrecombining chromosomes are susceptible to the influence of selective forces acting on any sequence on the chromosome, and given that selection is the only factor broadly applicable to explain reduced diversity in the sex-limited chromosome, irrespective of mode of reproduction, or whether there is male or female heterogamety, we argue that selection may be a common denominator of reduced variability in Y and W sex chromosomes.

Supplementary Material

Supporting Table:


We thank Dr. Michèle Tixier-Boichard (the European Commission-funded AVIANDIV Project) for chicken DNA samples; Malin Ellinder and Anneli Sjöberg for technical assistance; and two anonymous referees for useful comments. Financial support was obtained from the Swedish Research Council.


Author contributions: H.E. designed research; S.B. performed research; S.B. and H.E. analyzed data; and H.E. and S.B. wrote the paper.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviation: Ne, effective population size.

Data deposition: The sequences reported in this paper have been deposited in the GenBank database (accession nos. AY628538, AY628533, AY628478, AY628483, AY628488, AY628493, AY628498, AY628508, AY628518, AY628523, AY628448, AY628453, and AY628458).


1. Charlesworth, B. & Charlesworth, D. (2000) Philos. Trans. R. Soc. London B 355, 1563–1572. [PMC free article] [PubMed]
2. Nei, M. (1987) Molecular Evolutionary Genetics (Oxford Univ. Press, Oxford).
3. Kimura, M. (1983) The Neutral Theory of Molecular Evolution (Cambridge Univ. Press, Cambridge, U.K.).
4. Caballero, A. (1995) Genetics 139, 1007–1011. [PMC free article] [PubMed]
5. Shen, P., Wang, F., Underhill, P. A., Franco, C., Yang, W. H., Roxas, A., Sung, R., Lin, A. A., Hyman, R. W., Vollrath, D., et al. (2000) Proc. Natl. Acad. Sci. USA 97, 7354–7359. [PMC free article] [PubMed]
6. Sachidanandam, R., Weissman, D., Schmidt, S. C., Kakol, J. M., Stein, L. D., Marth, G., Sherry, S., Mullikin, J. C., Mortimore, B. J., Willey, D. L., et al. (2001) Nature 409, 928–933. [PubMed]
7. Hurst, L. D. & Ellegren, H. (1998) Trends Genet. 14, 446–452. [PubMed]
8. Li, W. H., Yi, S. & Makova, K. (2002) Curr. Opin. Genet. Dev. 12, 650–656. [PubMed]
9. Burrows, W. & Ryder, O. A. (1997) Nature 385, 125–126. [PubMed]
10. Hellborg, L. & Ellegren, H. (2004) Mol. Biol. Evol. 21, 158–163. [PubMed]
11. Zurovcova, M. & Eanes, W. F. (1999) Genetics 153, 1709–1715. [PMC free article] [PubMed]
12. Filatov, D. A., Moneger, F., Negrutiu, I. & Charlesworth, D. (2000) Nature 404, 388–390. [PubMed]
13. Lindgren, G., Backstrom, N., Swinburne, J., Hellborg, L., Einarsson, A., Sandberg, K., Cothran, G., Vila, C., Binns, M. & Ellegren, H. (2004) Nat. Genet. 36, 335–336. [PubMed]
14. Charlesworth, B. (1996) Genet. Res. 68, 131–150. [PubMed]
15. Smith, J. M. & Haigh, J. (1974) Genet. Res. 23, 23–35. [PubMed]
16. Charlesworth, B., Morgan, M. T. & Charlesworth, D. (1993) Genetics 134, 1289–1303. [PMC free article] [PubMed]
17. Wang, J. & Caballero, A. (1999) Heredity 82, 212–226.
18. Ellegren, H. (2003) BioEssays 25, 163–167. [PubMed]
19. Sundstrom, H., Webster, M. T. & Ellegren, H. (2004) Genetics 167, 377–385. [PMC free article] [PubMed]
20. Rosenberg, N. A., Burke, T., Elo, K., Feldman, M. W., Freidlin, P. J., Groenen, M. A., Hillel, J., Maki-Tanila, A., Tixier-Boichard, M., Vignal, A., et al. (2001) Genetics 159, 699–713. [PMC free article] [PubMed]
21. Rozas, J., Sanchez-DelBarrio, J. C., Messeguer, X. & Rozas, R. (2003) Bioinformatics 19, 2496–2497. [PubMed]
22. Hudson, R. R., Kreitman, M. & Aguade, M. (1987) Genetics 116, 153–159. [PMC free article] [PubMed]
23. Axelsson, E., Smith, N. G. C., Sundstrom, H., Berlin, S. & Ellegren, H. (2004) Mol. Biol. Evol. 21, 1534–1547. [PubMed]
24. Zuk, M., Johnson, K., Thornhill, R. & Ligon, J. D. (1990) Evolution (Lawrence, Kans.) 44, 477–485.
25. Collias, N. E. & Collias, E. (1996) Anim. Behav. 51, 1337–1354.
26. Muir, W. M. & Aggrey, S. E. (2003) Poultry Genetics, Breeding and Biotechnology (CAB International, New York).
27. Lahn, B. T., Pearson, N. M. & Jegalian, K. (2001) Nat. Rev. Genet. 2, 207–216. [PubMed]
28. Roldan, E. R. & Gomendio, M. (1999) Trends Ecol. Evol. 14, 58–62. [PubMed]
29. Ellegren, H. (2002) Trends Genet. 18, 25–28. [PubMed]
30. Ceplitis, H. & Ellegren, H. (2004) Mol. Biol. Evol. 21, 249–254. [PubMed]
31. Montell, H., Fridolfsson, A. K. & Ellegren, H. (2001) Mol. Biol. Evol. 18, 2010–2016. [PubMed]
32. Berlin, S. & Ellegren, H. (2001) Nature 413, 37–38. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...