Is the Rate of Insertion and Deletion Mutation Male Biased?: Molecular Evolutionary Analysis of Avian and Primate Sex Chromosome Sequences

Corresponding author: Hans Ellegren, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18D, SE-752 36 Uppsala, Sweden., hans.ellegren{at}ebc.uu.se (E-mail)

Communicating editor: W. STEPHAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

The rate of mutation for nucleotide substitution is generally higher among males than among females, likely owing to the larger number of DNA replications in spermatogenesis than in oogenesis. For insertion and deletion (indel) mutations, data from a few human genetic disease loci indicate that the two sexes may mutate at similar rates, possibly because such mutations arise in connection with meiotic crossing over. To address origin- and sex-specific rates of indel mutation we have conducted the first large-scale molecular evolutionary analysis of indels in noncoding DNA sequences from sex chromosomes. The rates are similar on the X and Y chromosomes of primates but about twice as high on the avian Z chromosome as on the W chromosome. The fact that indels are not uncommon on the nonrecombining Y and W chromosomes excludes meiotic crossing over as the main cause of indel mutation. On the other hand, the similar rates on X and Y indicate that the number of DNA replications (higher for Y than for X) is also not the main factor. Our observations are therefore consistent with a role of both DNA replication and recombination in the generation of short insertion and deletion mutations. A significant excess of deletion compared to insertion events is observed on the avian W chromosome, consistent with gradual DNA loss on a nonrecombining chromosome.

THERE is compelling evidence from humans and other organisms that the mutation rate for nucleotide substitution is higher among males than among females (HURST and ELLEGREN 1998 ). However, the precise extent of this excess of male point mutation (_m) in humans has been an issue of debate (SHIMMIN et al. 1993 ; MCVEAN and HURST 1997 ; BOHOSSIAN et al. 2000 ; CROW 2000A ; EBERSBERGER et al. 2002 ; ELLEGREN 2002 ; MAKOVA and LI 2002 ). In addition, it is not unreasonable to think that different organisms may vary with respect to the relative excess of male mutation. Overall, however, a male-biased point mutation rate is consistent with the higher number of germline cell divisions in spermatogenesis than in oogenesis seen in many organisms (VOGEL and MOTULSKY 1996 ) and the assumption that DNA replication is important for germline mutation.

While the idea of male-biased mutation was first reached through indirect observations on X-linked human genetic disorders (by J. B. S. Haldane; HALDANE 1947 ) and later supported by molecular analysis of the parental origin of de novo mutations at disease loci, molecular evolutionary approaches have been crucial for addressing sex-specific mutation rates (MIYATA et al. 1987 ). The mammalian Y chromosome is transmitted through the male germline only and the evolutionary rate of neutral sequences on this chromosome should thus solely reflect the male mutation rate. The female mutation rate can be obtained by studying the evolutionary rate of X-linked sequences and by taking advantage of the fact that the X chromosome spends two-thirds of the time in the female germline. Also, comparison of the evolutionary rate of autosomal sequences with either Y- or X-linked sequences allows _m to be estimated. Such molecular evolutionary approaches have been used in a number of studies on sex-specific mutation rates in humans and other mammals, the most recent estimate of _m for point mutation in the human lineage being 3–5 (EBERSBERGER et al. 2002 ; MAKOVA and LI 2002 ). A similar approach has also been used in studies of organisms with female heterogamety. In birds, the W chromosome is transmitted through females only and the evolutionary rate of neutral, W-linked sequences should thus specifically reflect the rate of female mutation. Avian studies suggest _m for point mutation is 2–4 (ELLEGREN and FRIDOLFSSON 1997 ; KAHN and QUINN 1999 ; CARMICHAEL et al. 2000 ).

However, all data on the parental origin of spontaneous mutation causing human genetic disease are not supportive of a strong male-biased rate (HURST and ELLEGREN 1998 ). For two well-studied X-linked recessive disorders in particular, Duchenne muscular dystrophy (GRIMM et al. 1994 ) and hemophilia B (KETTERLING et al. 1994 ; SOMMER et al. 2001 ), maternal transmissions are at excess among new mutations. Importantly, it has been demonstrated that this female bias is associated with deletion mutations, not point mutations. Moreover, for some dominant autosomal disorders caused by short deletion mutations, like neurofibromatosis type 1 (LAZARO et al. 1996 ) and Williams syndrome (PEREZ-JURADO et al. 1996 ), mutation rates of males and females are about the same. These observations have led to the idea that deletion (and perhaps insertion) mutations in general are not replication dependent or at least do not correlate to the number of DNA replications in the germline (VOGEL and MOTULSKY 1996 ; CROW 1997 , CROW 2000B ). One possibility is that insertion and deletion mutations (hereafter referred to as indel mutations or just indels) are somehow related to meiotic recombination (BAUMER et al. 1998 ; LOPEZ-CORREA et al. 2000 ). If so, their rate should mainly reflect the rate of recombination rather than the time a particular sequence spends in the male and female germlines, respectively.

Observations of sex-specific mutation rates at human disease loci often present conflicting results. For several genetic disorders caused by point mutations the male-to-female mutation rate ratio differs considerably from that indicated by molecular evolutionary analysis, thought to represent the genome average. The pattern of parental origin of disease mutation caused by indels is also heterogeneous (HURST and ELLEGREN 1998 ) and it is likely that such mutations, which cause observable yet nonlethal phenotypic effects, are not representative for the overall genomic rate of spontaneous indel mutation. Alternative approaches are therefore needed.

In this study we present the first large-scale genetic analysis of sex-specific rates of indel mutation in noncoding DNA, based on evolutionary analysis of sex chromosome sequences. As this approach has proved fundamental for the understanding of rates of point mutation in relation to sex, we believe it has the potential to provide similar insight into the causes and mechanism of indel mutations in males and females. We use two different systems to test the underlying factors affecting the rate of indel mutation: the comparison of X and Y chromosome sequences in primates and the comparison of Z and W chromosome sequences in birds. If meiotic recombination involving crossing over is a main cause of indel mutation, we should expect to find relatively few indel mutations in the nonrecombining Y and W chromosomes (Table 1). In contrast, if the number of germline DNA replications is important for the generation of indels (KUNKEL and BEBENEK 2000 ), similar to the case for point mutations, we should expect to find an excess of indels on the male-specific mammalian Y chromosome but a deficit on the female-specific avian W chromosome.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

Avian sequence data:
Five different introns from the gametologous avian chromo-helicase-DNA-binding protein CHD1W and CHD1Z genes were sequenced (or available from our previous work) in a number of avian species. The full exon-intron structure of these genes has not been elucidated so we arbitrarily designate the introns A–E. However, their location relative to the full-length chicken CHD1Z cDNA sequence (GenBank no. AF004397) can be identified through the primer sequences in Table 2 or the references provided therein. Templates for DNA sequencing were obtained through PCR amplification of DNA prepared from blood. DNA sequencing was performed with Big Dye terminator cycle sequencing chemistry and analyzed on an ABI377 (Applied Biosystems, Foster City, CA) DNA sequencing instrument. Sequences were deposited in GenBank (AF525971, AF526084). The data set was augmented with sequence information from ELLEGREN and FRIDOLFSSON 1997 , KAHN and QUINN 1999 , and MONTELL et al. 2001 . We also included sequence data (GenBank nos. AF165968, AF165969, AF165970, AF165971 and AF165976, AF165879) from the third intron of the gametologous ATP synthase -subunit ATP5A1Z and ATP5A1W genes (CARMICHAEL et al. 2000 ). CHD1W and ATP5A1W are both within the nonrecombining part of the W chromosome.

The complete set of bird species included, from the order Passeriformes, jackdaw (Corvus monedula), raven (Corvus corax), siberian jay (Perisoerus infaustus), pied flycatcher (Ficedula hypoleuca), collared flycatcher (Ficedula albicollis), barn swallow (Hirundo rustica), willow warbler (Phylloscopus trochilus), wood warbler (Phylloscopus sibilatrix), blue tit (Parus caeruleus), bluethroat (Luscinia svecica), house sparrow (Passer domesticus), oleaginous-hemispingus (Hemispingus frontalis), and zebrafinch (Taenipygia guttata); from Galliformes, chicken (Gallus gallus), turkey (Meleagris gallopavo), quail (Coturnix coturnix), pheasant (Phasianus colchius), sage grouse (Centrocercus urophasianus), and black grouse (Tetrao tetrix); from Anseriformes, barnacle goose (Branta leucposis), snow goose (Chen caerulescens), tundra swan (Cygnus columbianus), eider (Somateria mollissima), goldeneye (Bucephala clangula), redhead (Aythya americana), and canvasback (Aythya valisineria); from Charadriiformes, black-headed gull (Larus ridibundus), glaucous gull (Larus hyperboreus), herring gull (Larus argentatus), brown skua (Catharacta antarctica), oystercatcher (Haematopus ostralegus), dunlin (Calidris alpina), dotterel (Charadrius morinellus), adelie penguion (Pygoscels adliae), Leach's storm petrel (Oceanodroma leucorrhoa), and northern fulmar (Fulmarus glacialis); from Falconiformes, sparrow hawk (Accipiter nisus), merlin (Falco columbarius), goshawk (Accipiter gentilis), Galapagos hawk (Buteo galapagoensis), golden eagle (Aquila chrysaetos), black vulture (Aegypsis monachus), kestrel (Falco tinninculus), and merlin (Falco columbarius); from Piciformes, usambiro barbet (Trachyphonus usambiro), acorn woodpecker (Melanerpes formicivorus), and striped woodpecker (Picoides borealis); from Strigiformes, Tengmalm's owl (Aegolius funereus) and long-eared owl (Asio otus); from Psittaciformes, blue-fronted amazon (Amazon aestiva), kookaburra (Dacelo spp), sharp-tailed conure (Aratinga acuticaudata), maroon-bellied conure (Pyrrhura frontalis), and barred parakeet (Bolburhynchus lineola).

Primate sequence data:
DNA was prepared from tissue samples from male marmoset Callithrix jacchus using a standard proteinase K and phenol-chloroform extraction protocol. Six different introns from three gametologous genes (DBX/DBY, SMCX/SMCY, and ZFX/ZFY) shared between the X and Y chromosomes were amplified with the primers described in Table 3 and sequenced as above. Sequences were deposited in GenBank (AF526085, AF526086, AF526087, AF526088, AF526089, AF526090, AF526091, AF526092, AF526093, AF526094, AF526095, AF526096, AF526097). In addition, we used published sequence data from the third intron of the amelogenin AMELX/AMELY genes in human, orangutan (Pongo pygmeaus), and Bolivian squirrel monkey (Saimiri boliviensis; X14439, X14440, U88979, and U88981–U88983). For the last intron of the ZFX/ZFY genes we obtained sequences from human, orangutan, baboon (Papio cynocephalus), and Bolivian squirrel monkey (X58930–X58932, X58935, X58936, X72698, U24118, and AF02232). DBY, SMCY, ZFY, and AMELY are located within the nonrecombining part of the Y chromosome.

Analysis of sequence data:
Alignment of intron sequences of bird and primate species was made with the ClustalW algorithm using default settings. Separate multiple alignments of both Z- and W-linked gametologous sequences from a number of different groups of related avian species (generally species within the same order) were constructed as specified in Table 2. For 18 of 20 alignments more than two bird species were used and in these cases phylogenetic trees were constructed in MEGA 2.1 using the neighbor-joining algorithm and Kimura two-parameter correction (KUMAR et al. 2001 ). From these trees indels were subsequently assigned to branches. Furthermore, separate alignments of both X- and Y-linked gametologous sequences from humans and a single or several primate species were constructed, using the species shown in Table 4. In all cases, the same species were used in alignments of gametologs on both sex chromosomes.

A gap in one or several of the sequences within an alignment was considered the result of one or more indel mutations. However, length differences in tandemly repetitive DNA were excluded, using the criterion of not considering gaps in regions with three or more repeat units present in any of the species. We also excluded gaps from regions with sequence homology to known interspersed repetitive elements, identified through BLAST searches against avian and primate sequences. In a few cases the alignment algorithm suggested the presence of two gaps separated by a single nucleotide. To be conservative in the estimation of rates of indel mutation we manually realigned such regions to minimize the number of indels. Measures of divergence (nucleotide substitution) were estimated in MEGA 2.1 using Kimura two-parameter correction (KUMAR et al. 2001 ). We obtained a mean estimate for each alignment by averaging all pairwise divergences between sequences included in the alignment.

For all data sets, the lengths and number of indels found in alignments were resampled using a bootstrap procedure to calculate confidence intervals for estimates of _m (for indels). For the avian alignments, in each replicate the lengths and number of indels derived from all of the alignments of each individual intron on both sex chromosomes were first randomly resampled with replacement. The resulting totals for each intron were then randomly resampled with replacement and a value of _m was calculated from each resultant data set. Confidence intervals (95%) were estimated from the distribution resulting from 10,000 replicates of the bootstrapping process. To calculate confidence intervals from the primate alignments of gametologs, the values for the length and number of indels found in each aligned intron on both sex chromosomes were resampled with replacement in 10,000 replicates. Ninety-five percent confidence intervals for _m (for indels) were generated from the resultant distribution.

We refer to the Y and W chromosomes as nonrecombining although both have at least one small pseudoautosomal region (PAR) in which recombination takes place during meiosis. However, all sequences analyzed in this study are from outside the PAR and thus from regions with no meiotic recombination.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

Rates of indel mutation on sex chromosomes:
We used intron sequences of gametologous genes shared between the Z and W chromosomes of birds (Table 2) and the X and Y chromosomes of primates (Table 4), to infer the incidence and character of indel mutation in males and females. In primates (human and at least one of baboon, orangutan, squirrel monkey, or marmoset), 6195 bp of X chromosome and 5232 bp of Y chromosome sequence were derived in eight alignments, always using the same species for the alignment of a particular gametologous intron. In birds, 11,065 bp of Z chromosome and 10,435 bp of W chromosome sequence were obtained in the same way, in 20 different alignments.

Numbers of indels on the respective chromosomes are summarized in Table 5. The incidences of indels on primate X and Y were similar, occurring at a rate of 1% per base pair in our set of alignments (X = 0.0102, Y = 0.0117; P = 0.47, Fisher's exact test). In contrast, indels were about twice as common in alignments derived from the avian Z chromosome as in those derived from the W chromosome (Z = 0.0211, W = 0.0108; P < 0.001). There was a higher incidence of indels on Z than on W in 18 out of 20 avian alignments of gametologous introns (P < 0.001), while there was no obvious bias for primate introns (5 with more indels on Y, 3 with more on X; P = 0.438). The primate data set seems homogeneous with respect to which species are used for comparison; for instance, using only the most divergent pair of species—human vs. marmoset—results remain similar (data not shown). Converting the observed frequencies of indel mutation in different chromosomes to male-to-female mutation rate ratios, estimates of _m for indels of 2.43 (95% confidence interval 1.51–3.85) for birds and 1.24 (0.67–4.26) for primates are obtained.

Sex-specific rates for point mutation were also estimated from the aligned sequences of gametologous introns. In primates, observed mean pairwise divergence of 0.079 for X chromosome and 0.174 for Y chromosome sequences translates into an estimate of _m for point mutation of 5.61. The corresponding estimate in birds was _m = 2.31, derived from mean pairwise divergence in multiple species alignments of 0.123 on Z and 0.0663 on W. These estimates agree reasonably well with those obtained in earlier studies (ELLEGREN and FRIDOLFSSON 1997 ; EBERSBERGER et al. 2002 ; MAKOVA and LI 2002 ). We thus conclude that in primates there seems to be no clear difference in the rate of indel mutation on X and Y although the rate for point mutation is male biased. In birds, however, the rates of indel as well as point mutation are higher on Z than on W.

For the primate data set indel mutations comprised 6.3% (the Y chromosome) and 11.4% (X) of the total number of mutations, consistent with previous observations suggesting that less than one-tenth of all mutations in the human genome are indels (NACHMAN and CROWELL 2000 ). Indels were slightly more common in avian chromosomes, making up 14.4% (Z) and 13.9% (W) of all mutations. Birds therefore seem intermediate to human and Drosophila, where indels constitute 20% of all mutations (PRITCHARD and SCHAEFFER 1997 ; PETROV and HARTL 1998 , PETROV and HARTL 1999 ).

Character of indel mutation:
Fig 1 depicts the size distribution of DNA sequences being inserted or deleted on sex chromosomes of primates and birds. A strong dominance of events involves very short sequences, in particular 1-bp indels. There is no significant difference (Kolmogorov-Smirnov test) in the size distribution of indels between mammals and birds. Moreover, the overall size distribution of indels does not differ between X and Y or between Z and W. However, when 1-bp and >1-bp indels are treated separately, there is a more pronounced excess of mutations on Z compared to W for 1-bp indels (Z/W = 2.72, _m = 3.59) than for >1-bp indels (Z/W = 1.49, _m = 1.82; P = 0.025, Fisher's exact test). There is also a difference in the relative incidence of indel mutation on X and Y when analyzing the data in this way. For 1-bp indels, Y/X is 1.54 while it is 0.92 for indels >1 bp (P = 0.204). One-base-pair indels thus seem particularly common on avian Z and mammalian Y.

View larger version (14K): In this window In a new window Download PPT slide   Figure 1. Relative size distributions of indel mutations on (A) avian Z, (B) avian W, (C) primate Y, and (D) primate X chromosomes.

Generally more than two species were available in avian alignments and from established phylogenies the ancestral state of indel sequences could be obtained by parsimony principles. On the whole, deletions outnumbered insertion events (deletion/insertion ratio = 2.57; ² = 19.78, P < 0.001). However, the Z and W chromosomes differed considerably in this respect with only a moderate excess on Z (1.85, P = 0.0146) and a much more distinct bias on W (7.25, P > 0.0001), the two ratios being significantly different (P = 0.0189, Fisher's exact test). As most of the primate data were obtained from pairwise alignments (i.e., without an outgroup) we were unable to perform a similar analysis for the X and Y chromosomes.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

The main observations from this study can be summarized as follows: (i) Indel mutations are frequent on the Y chromosome of primates and the W chromosome of birds; (ii) the rate of indel mutation is similar on the X and Y chromosomes of primates, indicating no bias with respect to sex; (iii) in birds, the rate of indel mutation is about twice as high on the Z as on the W chromosome, indicating a moderate male bias; (iv) 1-bp indels seem particularly common on primate Y and avian Z; and (v) the W chromosome has relatively more deletions than Z. As a consequence of ii and iii, the sex-specific rates of indel and point mutation may be different in primates, while they appear similar in birds. Below we discuss these observations with respect to the possible mechanistic basis for indel mutation.

Analyses of flanking markers in cases of de novo deletion mutations at human disease loci have provided evidence for processes involving meiotic recombination in the generation of indel mutations (LAZARO et al. 1996 ; BAUMER et al. 1998 ; LOPEZ-CORREA et al. 2000 ). However, our data indicate that indels in noncoding DNA also arise frequently in regions of chromosomes that do not recombine at meiosis. These observations, consistent for both mammals and birds, demonstrate that meiotic crossing over cannot be the main, and certainly not the sole, cause of short insertions and deletions in noncoding DNA.

The contrasting ratios between sex chromosomes seen for rates of indel mutation in primates and birds might be informative for elucidating the origin of indel mutations. As indicated above, if meiotic recombination is the most important factor for rate of indel mutation we should expect more indels on X than on Y and more on Z than on W. In contrast, if the number of cell divisions has a large effect we should expect more indels on Y than on X and more on Z than on W (Table 1). Our data are not consistent with either of these predictions. As similar rates of indels were observed on X and Y a possible scenario is therefore that recombination and number of cell divisions both play a role for indel mutation. Specifically, we hypothesize that in systems with male heterogamety and where the number of cell divisions in spermatogenesis significantly exceeds that in oogenesis, meiotic recombination and replication may be important for the generation of indels on X while replication should be the main factor causing indels on Y. Our data suggest that meiotic recombination and the relatively low number of DNA replications of X introduce indel mutations at about the same rate as the larger number of replications of Y. Moreover, we hypothesize that in systems with female heterogamety meiotic recombination and replication should both contribute to a higher incidence of indels on Z than on W, as found in birds.

It should be noted that our data do not exclude the possibility that there are other sources of indel mutation apart from replication and recombination and that alternative mechanisms could affect the two sexes (or the two sex chromosomes) equally. For example, indels might be introduced from DNA damage. The contribution of another mechanism(s) could potentially be indicated from the fact that in birds the excess of indels on Z compared to W is similar to the excess of point mutations on Z vs. W. This may be unexpected according to the hypothesis that both replication and recombination cause indels while point mutations are often considered replication dependent. However, recent analyses of large-scale genome sequence data suggest that recombination might introduce point mutations too (LERCHER and HURST 2002 ). It therefore remains to be demonstrated that other mechanisms do play a role in the generation of indel mutation.

Replication errors are a likely mechanistic explanation for the effect of number of cell divisions on rate of indel mutation, although other factors could also be invoked (see below). DNA replication is known to introduce short insertion and deletion mutations through various forms of strand misalignment (KUNKEL and SONI 1988 ; OSHEROFF et al. 2000 ). Template-primer slippage during replication of iterated sequences, like microsatellites, is one obvious example but misalignment can also occur in unique sequences (KUNKEL and BEBENEK 2000 ; also note that repetitive sequences were excluded from our analyses). For a unique sequence, there is experimental evidence that length mutations involving single nucleotides dominate (KUNKEL 1990 ). We can therefore make the prediction that if meiotic recombination and replication are important for indel mutation in noncoding DNA, length mutations involving only 1 bp should be particularly common when replication is at high rates. Our data are consistent with this prediction: 1-bp indels were more common on Y than on X and on Z than on W. This adds further support to an overall role of DNA replication in introducing indel mutation in noncoding DNA.

It is important to note that a correlation between number of DNA replications and rate of indel mutation does not necessarily imply that mutations induced prior to meiosis are by replication errors. Recombination-like processes are involved also during mitosis, in particular for the repair of incorrectly introduced nucleotides or of lesions in DNA. Although the propensity for 1-bp indels in Y and Z is consistent with a role of replication errors, we cannot conclusively distinguish between replication errors and recombination-like processes for the generation of indels prior to meiosis.

Studies of a number of organisms, including human, mouse, and Drosophila, have indicated that spontaneous deletions generally outnumber insertions (GRAUR et al. 1989 ; SAITOU and UEDA 1994 ; PETROV and HARTL 1998 ; COMERON and KREITMAN 2000 ; VINOGRADOV 2002 ). As the relative excess of deletions seems to vary between species this may be an important parameter in the long-term evolution of genome size (PETROV et al. 2000 ; PETROV 2001 ). The avian data indicate that the insertion:deletion ratio may also vary within genomes, in this case with a higher proportion of deletions on W than on Z. It is conceivable that this relates to the different mechanisms behind indel mutation on the two chromosomes as suggested above. We find this observation consistent with the fact that W is a decaying chromosome where gradual DNA loss has characterized the evolution of W following cessation of recombination with Z.

	CONCLUSIONS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

We now return to the question of whether the two sexes have different rates of indel mutation. The equal rates observed for X and Y would suggest that there is no male bias in the human lineage. However, extrapolating mutation rate data from sex chromosomes to overall sex-specific rates requires a common mutation mechanism that correlates with the time spent in the male and female germlines. The conclusion of recombination and replication both playing a role for indel mutation is at odds with this requirement. As we quantitatively cannot assess the relative importance of recombination and replication it is difficult to obtain a detailed estimate of the male-to-female indel mutation rate ratio. Moreover, the situation is complicated by the fact that the recombination rate is higher in females than in males and that the X chromosome does not recombine in males. However, if there were a significant sex difference in the indel mutation rate, e.g., of similar extent as for the rate of point mutation, it seems reasonable that it should be manifested by contrasting rates of indel mutation on the X and Y chromosomes. We therefore conclude that our data suggest that there is no strong male bias for indels in humans.

The situation is different in birds. Representing the homogametic sex, males recombine at higher rates than females do. From the combined effect of high recombination and large number of DNA replications, we should expect more indels to arise in males than in females. The higher incidence of indels on Z than on W is consistent with this expectation. We therefore end with the general hypothesis that indels arise mainly in males in organisms with female heterogamety but may arise with more similar rates in males and females in organisms with male heterogamety.

	FOOTNOTES

Sequence data from this article have been deposited with the GenBank Data Library under accession nos. AF525971–AF526097.

	ACKNOWLEDGMENTS

We thank Sofia Berlin, Anna-Karin Fridolfsson, and Anna Härlid for sequence data. Financial support was obtained from the Swedish Research Council. H.E. is a Royal Swedish Academy of Sciences Research Fellow supported by a grant from the Knut and Alice Wallenberg Foundation.

Manuscript received July 24, 2002; Accepted for publication January 25, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

BAUMER, A., F. DUTLY, D. BALMER, M. RIEGEL, and T. TUKEL et al., 1998  High level of unequal meiotic crossovers at the origin of the 22q11.2 and 7q11.23 deletions. Hum. Mol. Genet. 7:887-894.[Abstract/Free Full Text]

BOHOSSIAN, H. B., H. SKALETSKY, and D. C. PAGE, 2000  Unexpectedly similar rates of nucleotide substitution found in male and female hominids. Nature 406:622-625.[Medline]

CARMICHAEL, A. N., A. K. FRIDOLFSSON, J. HALVERSON, and H. ELLEGREN, 2000  Male-biased mutation rates revealed from Z and W chromosome-linked ATP synthase alpha-subunit (ATP5A1) sequences in birds. J. Mol. Evol. 50:443-447.[Medline]

COMERON, J. M. and M. KREITMAN, 2000  The correlation between intron length and recombination in Drosophila: dynamic equilibrium between mutational and selective forces. Genetics 156:1175-1190.[Abstract/Free Full Text]

CROW, J. F., 1997  The high spontaneous mutation rate: Is it a health risk? Proc. Natl. Acad. Sci. USA 94:8380-8386.[Abstract/Free Full Text]

CROW, J. F., 2000a  A new study challenges the current belief of a high human male:female mutation ratio. Trends Genet. 16:525-526.[Medline]

CROW, J. F., 2000b  The origins, patterns and implications of human spontaneous mutation. Nat. Rev. Genet. 1:40-47.[Medline]

EBERSBERGER, I., D. METZLER, C. SCHWARZ, and S. PAABO, 2002  Genomewide comparison of DNA sequences between humans and chimpanzees. Am. J. Hum. Genet. 70:1490-1497.[Medline]

ELLEGREN, H., 2002  Human mutation—blame (mostly) men. Nat. Genet. 31:9-10.[Medline]

ELLEGREN, H. and A. K. FRIDOLFSSON, 1997  Male-driven evolution of DNA sequences in birds. Nat. Genet. 17:182-184.[Medline]

FRIDOLFSSON, A.-K. and H. ELLEGREN, 1999  A simple and universal method for sexing non-ratite birds. J. Avian Biol. 30:116-121.

GRAUR, D., Y. SHUALI, and W. H. LI, 1989  Deletions in processed pseudogenes accumulate faster in rodents than in humans. J. Mol. Evol. 28:279-285.[Medline]

GRIMM, T., G. MENG, S. LIECHTI-GALLATI, T. BETTECKEN, and C. R. MULLER et al., 1994  On the origin of deletions and point mutations in Duchenne muscular dystrophy: most deletions arise in oogenesis and most point mutations result from events in spermatogenesis. J. Med. Genet. 31:183-186.[Abstract]

HALDANE, J. B. S., 1947  The rate of mutation of the gene for hemophilia and its segregation in males and females. Ann. Eugen. 13:262-271.

HURST, L. D. and H. ELLEGREN, 1998  Sex differences in the mutation rate. Trends Genet. 14:446-452.[Medline]

KAHN, N. W. and T. W. QUINN, 1999  Male-driven evolution among Eoaves? A test of the replicative division hypothesis in a heterogametic female (ZW) system. J. Mol. Evol. 49:750-759.[Medline]

KETTERLING, R. P., E. L. VIELHABER, T. J. LIND, E. C. THORLAND, and S. S. SOMMER, 1994  The rates and patterns of deletions in the human factor IX gene. Am. J. Hum. Genet. 54:201-213.[Medline]

KUMAR, S., K. TAMURA, I. B. JAKOBSEN, and M. NEI, 2001  MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17:1244-1245.[Abstract/Free Full Text]

KUNKEL, T. A., 1990  Misalignment-mediated DNA synthesis errors. Biochemistry 29:8003-8011.[Medline]

KUNKEL, T. A. and K. BEBENEK, 2000  DNA replication fidelity. Annu. Rev. Biochem. 69:497-529.[Medline]

KUNKEL, T. A. and A. SONI, 1988  Mutagenesis by transient misalignment. J. Biol. Chem. 263:14784-14789.[Abstract/Free Full Text]

LÁZARO, C., A. GAONA, P. AINSWORTH, R. TENCONI, and D. VIDAUD et al., 1996  Sex differences in mutational rate and mutational mechanism in the NF1 gene in neurofibromatosis type 1 patients. Hum. Genet. 98:696-699.[Medline]

LERCHER, M. L. and L. D. HURST, 2002  Human SNP variability and mutation rates are higher in regions of high recombination. Trends Genet. 18:337-340.[Medline]

LÓPEZ-CORREA, C., H. BREMS, C. LÁZARO, P. MARYNEN, and E. LEGIUS, 2000  Unequal crossover: a frequent cause of NF1 microdeletions. Am. J. Hum. Genet. 66:1969-1974.[Medline]

MAKOVA, K. D. and W. H. LI, 2002  Strong male-driven evolution of DNA sequences in humans and apes. Nature 416:624-626.[Medline]

MCVEAN, G. T. and L. D. HURST, 1997  Evidence for a selectively favourable reduction in the mutation rate of the X chromosome. Nature 386:388-392.[Medline]

MIYATA, T., H. HAYASHIDA, K. KUMA, K. MITSUYASU, and T. YASUNAGA, 1987  Male-driven molecular evolution: a model and nucleotide sequence analysis. Cold Spring Harbor Symp. Quant. Biol. 52:863-867.[Medline]

MONTELL, H., A. K. FRIDOLFSSON, and H. ELLEGREN, 2001  Contrasting levels of nucleotide diversity on the avian Z and W sex chromosomes. Mol. Biol. Evol. 18:2010-2016.[Abstract/Free Full Text]

NACHMAN, M. W. and S. L. CROWELL, 2000  Estimate of the mutation rate per nucleotide in humans. Genetics 156:297-304.[Abstract/Free Full Text]

OSHEROFF, W. P., W. A. BEARD, S. YIN, S. H. WILSON, and T. A. KUNKEL, 2000  Minor groove interactions at the DNA polymerase beta active site modulate single-base deletion error rates. J. Biol. Chem. 275:28033-28038.[Abstract/Free Full Text]

PÉREZ-JURADO, L. A., R. PEOPLES, P. KAPLAN, B. C. HAMEL, and U. FRANCKE, 1996  Molecular definition of the chromosome 7 deletion in Williams syndrome and parent-of-origin effects on growth. Am. J. Hum. Genet. 59:781-792.[Medline]

PETROV, D. A., 2001  Evolution of genome size: new approaches to an old problem. Trends Genet. 17:23-28.[Medline]

PETROV, D. A. and D. L. HARTL, 1998  High rate of DNA loss in the D. melanogaster and D. virilis species groups. Mol. Biol. Evol. 15:293-302.[Abstract]

PETROV, D. A. and D. L. HARTL, 1999  Patterns of nucleotide substitution in Drosophila and mammalian genomes. Proc. Natl. Acad. Sci. USA 96:1475-1479.[Abstract/Free Full Text]

PETROV, D. A., T. A. SANGSTER, J. S. JOHNSTON, D. L. HARTL, and K. L. SHAW, 2000  Evidence for DNA loss as a determinant of genome size. Science 287:1060-1062.[Abstract/Free Full Text]

PRITCHARD, J. K. and S. W. SCHAEFFER, 1997  Polymorphism and divergence at a Drosophila pseudogene locus. Genetics 147:199-208.[Abstract]

SAITOU, N. and S. UEDA, 1994  Evolutionary rates of insertion and deletion in noncoding nucleotide sequences of primates. Mol. Biol. Evol. 11:504-512.[Abstract]

SHIMMIN, L. C., B. H. CHANG, and W. H. LI, 1993  Male-driven evolution of DNA sequences. Nature 362:745-747.[Medline]

SOMMER, S. S., W. A. SCARINGE, and K. A. HILL, 2001  Human germline mutation in the factor IX gene. Mutat. Res. 487:1-17.[Medline]

VINOGRADOV, A. E., 2002  Growth and decline of introns. Trends Genet. 18:232-236.[Medline]

VOGEL, F., and A. G. MOTULSKY, 1996 Human Genetics—Problems and Approaches. Springer-Verlag, Berlin/Heidelberg/New York.

This article has been cited by other articles:

				K. D. Makova, S. Yang, and F. Chiaromonte Insertions and Deletions Are Male Biased Too: A Whole-Genome Analysis in Rodents Genome Res., April 1, 2004; 14(4): 567 - 573. [Abstract] [Full Text] [PDF]