Evolution of sex chromosomes

Cytogenet Genome Res 99:125140 (2002)

DOI: 10.1159/000071584

The evolution of sex chromosomes

L.-J. Ayling and D.K. Griffin
Cell and Chromosome Biology Group, Department of Biological Sciences, Brunel University, Uxbridge (UK)

Abstract. Mammalian sex chromosomes appear, behave

and function differently than the autosomes, passing on their
genes in a unique sex-linked manner. The publishing of Ohnos
hypothesis provided a framework for discussion of sex chromosome evolution, allowing it to be developed and challenged
numerous times. In this report we discuss the pressures that

1. The advantages of sex

Without sex there would be almost no evolution, our fascination with
sex and the genetics may one day lead to the most important sexual problem
of all why do we bother in the first place. Steve Jones 1994.

Establishing the value of sex has proved difficult in theoretical population genetics and evolution studies (Hurst and Peck,
1996). Sex accelerates evolution, creating unique individuals,
with an ever-widening pool of ancestors. Sex is a worthwhile
(albeit expensive) way of reshuffling the genetic composition by
recombination, bringing together advantageous genes increasing the likelihood of survival, reproduction and the ability to
cope with the changing environment. Sexual organisms reaped
the benefits, rapidly evolving and growing more complex com-

drove the evolution of sex and the mechanisms by which it

occurred. We concentrate on how the sex chromosomes
evolved in mammals, discussing the various hypotheses proposed and the evidence supporting them.
Copyright 2002 S. Karger AG, Basel

pared to their asexual counterparts which led to sexual-dimorphism. This is often subtle, like the bright face of the baboon
but can be more overt and costly to the organism: For instance
male peacocks have an ornate display of feathers, attracting
predators and hindering escape and the elaborate 12-foot antlers of the Irish elk may have led to their extinction (Gerhart and
Kirschner, 1997; Crow, 1994). The advantages and disadvantages of sexual reproduction depend upon the immediate and
long-term gains. In the short-term, sex is costly and time consuming however, in the long run, new mixtures of genes constantly arise by genetic recombination. This may provide an
organism with the most advantageous combinations of genes,
making it more likely to survive, reproduce and pass on its
genes. These are considered in Table 1 and compared to parthenogenic reproduction.

2. The origin of sex dioecy from cosexuality?

Request reprints from Dr Darren K. Griffin, Cell and Chromosome Biology Group
Department of Biological Sciences, Brunel University
Uxbridge, Middlesex, UB8 3PH (UK)
telephone: 01895 274000 x2653; fax: 01895 274348
e-mail: Darren.Griffin@brunel.ac.uk

ABC

Fax + 41 61 306 12 34
E-mail karger@karger.ch
www.karger.com

2003 S. Karger AG, Basel

03010171/02/09940125$19.50/0

Accessible online at:

www.karger.com/cgr

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Received 19 November 2002; revision accepted 28 February 2003.

Although it is well established that sex encourages genetic

recombination and hence phenotypic variation, the question
still remains why are there separate sexes? Cosexuality
(where organisms are hermaphrodites with both male and
female sex organs) is common among marine invertebrates and
plants. However, among most animals dioecy (where there are
two morphologically distinct sexes) predominates. The female

Female Sterility
Mutation

Androdioecy

Male Sterility
Mutation

Cosexuality

Table 1. Parthenogenic reproduction versus

sexual reproduction

Male Sterility
Mutation

Female Sterility
Mutation

Sexual reproduction

Parthenogenic reproduction

Disadvantages

Advantages

Production of gametes is not efficient

Not all gametes give rise to new progeny.
Finding a mate and mating takes time, therefore less
offspring can be produced
Need to fend off competition from other males
Generally sexual organisms feed and protect their young

Do not need to produce gametes

A mate is not required, therefore more offspring can be
produced
No competition
Generally asexual organisms are not required to feed
and protect their young

Advantages

Disadvantages

Harmful mutations are removed by death or infertility,

purging the population
Spreads and maintains beneficial mutations through the
population
The number of ancestors double in each generation
Recombination in meiosis results in diversification

Accumulates the mutations until all individuals within a

population contain it
Beneficial mutations cannot spread through the
population
Only one new ancestor in each generation
No recombination therefore little diversification via
mutation alone

has an obvious role; she incubates the offspring, gives birth and
often raises it, the question still arises: why do males exist
except to impregnate the females? They have been likened to
parasites, having the pleasures of reproduction, passing on their
genes, with few of the pains involved (Jones, 1994). Charlesworth (1991) postulated that dioecy evolved from cosexuality.
Dioecy from cosexuality requires a minimum of two mutational steps. One step (gynodioecy) is a polymorphism (common mutation within the population) producing females from
an inability to produce male gametes via a male sterility mutation. The second phase: androdioecy, converts cosexuals into
males via a female sterility mutation. These steps occur simultaneously resulting in a mixed population of single sex (dioecy)
and cosexuals. Eventually, the remaining co-sexual individuals
would be selected against, leaving only organisms that display
dioecy (Charlesworth 1991; Charlesworth and Charlesworth,
2000). This process is demonstrated in Fig. 1. The process of
selecting in favour of dioecy is driven by two causal factors:
resource reallocation and avoidance of inbreeding (Charlesworth, 1991; Charlseworth and Charlesworth. 2000) i.e. in the
evolution of a female, mutations cause a decrease in male fertility and an increase in female fertility. Therefore, more time and
energy can be spent in producing one gamete instead of two,

126

Gynodioecy

Cytogenet Genome Res 99:125140 (2002)

thus reallocated essential resources such as time and energy.

Inbreeding results from self-fertilisation and increases homozygosity i.e. individuals are more likely to get deleterious recessive alleles after inbreeding than non-inbred offspring, exhibiting the disease phenotype. Dioecy reduces the ability for an
evolving female to produce sperm and to inbreed. Therefore
most animals evolved dioecy, but the mechanisms by which sex
is determined varies from species to species.

3. Sex determination mechanisms (SDM)

3.1. Temperature-dependent sex determination
Lizards, turtles and alligators have a temperature-dependent SDM. The egg incubation temperature stimulates transcription in the sex determination pathway (Bull, 1983) (see
section 6.1.4). In alligators, an incubation temperature of 27
31 C produces females but at 3135 C males result.
3.2. Genic mechanisms
The yeast Saccharomyces cerevisiae SDM is controlled by
the MAT (mating type) gene on chromosome 3. This has two
alleles: a and that determine mating type. The majority of its
Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Fig. 1. The evolutionary hypothesis that

males and females (dioecy) developed from cosexuals. Adapted from Charlesworth (1991).

Dioecy

No. of X
chromosomes

No. of sets of
autosomes

X:A ratio

Sex produced

3
3
2
2
1
1

2
3
2
3
2
3

1.5
1.0
1.0
0.67
0.5
0.33

meta-female
female
female
intersex
male
meta-male

Table 3. Summary of the various sex chromosome systems

Reptiles
Snakes
Lizards
Others
Insects
Birds
Mammals
Fish
Amphibia

Governing SDM

Governing
heterogametic sex

Occurrence in
population

ZZ/ZW
none
temperature dependant
none
ZZ/ZW
XX/XY
none
none

ZW female

almost universal
sporadic
practically zero
sporadic
universal
universal
sporadic
sporadic

ZW female
XY male

life it is haploid carrying either an a or allele. Cells mate with

opposite mating types by fusion and become diploid before
releasing haploid spores after meiosis (Winter et al., 1998).
Hymenoptera (ants, bees and wasps) drones (haploid males)
develop from unfertilised eggs, carrying one copy of the sexdetermining gene. Workers and queens (diploid females) develop from fertilised eggs, carrying two heterozygous copies of the
gene. Inbreeding produces weaker homozygous diploid males
with shorter life spans (Winter et al., 1998).
3.3. Chromosomal sex determination
3.3.1. The ZZ/ZW system determines sex in birds and
snakes: In most bird and snake species, females are heterogametic possessing a Z and W sex chromosome producing two
types of gametes containing either a Z or W chromosome.
Males are homogametic carrying two Z chromosomes and
produce one type of gamete (sperm). This is the ZZ/ZW system
of sex determination (Graves, 1998).
3.3.2. The XX/XY system determines sex in mammals: In
mammals males are heterogametic and females homogametic
and this is extensively conserved. Exceptions include wood
lemmings that have three sex chromosomes an X, a Y, and a
modified X (X*), which suppresses the activity of the Y. The
voles E. lutescen (that possess only one X chromosome regardless of sex) and E. tancrei (which has two X chromosomes) both
lack a Y (Fredga, 1988; Just et al., 1995).
3.3.3. Sex determination in Drosophila: Drosophila have an
XX/XY system, but sex is determined by an X:autosome ratio.
Table 2 shows the results of different ratios. Intermediate ratios
produce inter-sex flies, meta-males and meta-females. Inter-sex

flies have both male and female characteristics whereas metamales and females have poorly developed sexual characteristics
and often a shorter lifespan. This suggests that genes controlling
female sex development are on the X and male developmental
genes are on the autosomes. (Winter et al., 1998).
3.3.4. Sex determination in fish and amphibians: Only 2
3% of fish species have had their sex determining mechanism
examined. Cytogenetic studies are often impaired as fish usually have large numbers of small chromosomes. The sex determining mechanisms are varied and so far have been seen to
include XX/XY, ZW/ZZ, polygenic and XX/XO systems. The
SDMs also include more complex X-autosome and Y-autosome translocations such as: X1X1X2X2/X1X1X2, W1W2/ZZ,
X1X1X2X2/X1X2Y (Sola et al., 1981). Amphibians are difficult
to study, having a long generation time and the heterogeneity
appears independent of evolutionary divergence. The mechanisms seen include XX/XY and ZZ/ZW. The sex of some fish
and amphibian species are also strongly influenced by external
temperature and hormones (Schmid et al., 1989).
3.3.5. Chromosomal sex determination, a summary: The
classification of the chromosomes as X, Y, W and Z is used
solely to differentiate between the male and female heterogamety. The avian Z and W chromosome have a different complement of genes to that of the mammalian X and Y chromosome, suggesting that mechanism evolved independently (Baverstock et al., 1982, Dominguez-Steglich et al., 1990). These
systems are summarised in Table 3; in this review we consider,
primarily evolution of the mammalian sex chromosomes.

4. Evolution of the sex chromosomes

4.1. The morphology of the sex chromosomes
Sex chromosomes are generally microscopically distinguishable and have different shapes and/or banding patterns. The
human X chromosome constitutes approximately 5 % of the
total genome (Graves, 1998, 2000). The Y chromosome is generally much smaller, only 23 % of the human haploid genome
and only F40 genes have been identified on the Y chromosome. Many Y genes are non-functional pseudogenes with X
homologues. The marsupial sex chromosomes are significantly
smaller than that of the eutherians, with practically no Y chromosome in some species (Graves, 2000). Most mammalian sex
chromosomes are homologous over a very small portion and
rarely over an entire arm. This is termed the pseudoautosomal
region (PAR) (Burgoyne, 1982) and is the domain in which the
sex chromosomes synapse and undergo recombination (see section 6.5). The cytogenetic differences may result from structural rearrangements such as inversions, translocations, duplications and additions of heterochromatin. By contrast, in some
XY chromosome systems, the sex chromosomes share large
regions of homology and the X and Y differ minimally, these
cytogentically identical regions may have resulted from translocations of autosomes (Sola et al., 1981; Bull, 1983; Christidis,
1989; Schmid et al., 1989). The question therefore arises: by
what process did vertebrate gonosomes become heteromorphic
and why are some regions homomorphic?

Cytogenet Genome Res 99:125140 (2002)

127

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Table 2. Ratio of X chromosomes to sets of autosomes, and sex determination in Drosophila. Adapted from Winter et al. (1998)

Recombination Suppression
via Genic, Structural and/or Conformational changes

Decreased Function of loci in heteromorphic region

via a mechanism such as Mullers Rachet

Sex Chromosomes become Differentiated

Decreased meiotic activity of the Y

Fig. 2. The cyclic processes of Y chromosome degradation. Adapted from

Clark and Wall (1996).

4.2. Ohnos Law

Ohno (1967) discovered that some species of snakes have
morphologically different gonosomes that paired at a small
restricted region, others had a larger pairing region, while others still were so similar, they could only be identified as sexchromosomes by their size and banding patterns. He hypothesized that these differences represent intermediates between
undifferentiated and strongly differentiated Z and W chromosomes suggesting that the sex chromosomes evolved from a
homologous pair of autosomes, and during the degradation of
the Y, one X chromosome in females became subjected to inactivation (see section 7.1). Ohnos hypothesis has since been
applied to the mammalian XX/XY system (Graves, 1998) and
is now the accepted dogma for all sex chromosome evolution.
While snakes can be considered as intermediates in the ZZ/ZW
system of sex determination, the reindeer, Rangifer tarandus
provides intermediate examples in the XX/XY system. They
have large gonosomes, the q arms consist of substantial
amounts of heterochromatin (non-coding DNA), some parts
are homologous, but others are X specific. Unusually, the Y
chromosome has retained large amounts of euchromatin (coding DNA) homologous to the X (Lee et al., 1998).
4.3. The evolution of the heteromorphic sex chromosome
system
It is widely accepted that the sex chromosomes became differentiated first via suppression of crossing-over leading to the
eventual functional and structural degradation of the Y chromosome. Recombination generally is suppressed in a region
that contains heterochromatin, a structural rearrangement and/
or one containing a gene advantageous to one sex but detrimental to the other (see section 6.3). This recombination suppression allowed any mutations that occurred on the Y to become
fixed in the population. It also reduced the likelihood that the

128

Cytogenet Genome Res 99:125140 (2002)

mutation will be repaired. The degeneration of the Y, by Mullers Rachet, led to the sex chromosomes becoming heteromorphic (see next section). The increased differentiation of the
sex chromosomes resulted in a decreased meiotic activity of the
Y, causing further cycles of degeneration (Clark and Wall,
1996). This is simplified in Fig. 2.
4.4. Mullers Ratchet
Mullers Ratchet describes the mechanism by which the deleterious mutations accumulate on the Y chromosome (Muller,
1964; Felsenstein, 1974; Charlesworth, 1978). The mutation
rate in all chromosomes is approximately 1091011 per incorporated nucleotide (Allen, 1970; Thacker, 1985) however, the
mutations are more likely to become fixed in the Y chromosome by reducing the likelihood of recombination. Advantageous mutations are subjected to selective sweeps and fixation
becomes more rapid. Mullers Ratchet can be explained using
the Y chromosomes of three populations. The Y chromosomes
in population A have zero to three mutations, the chromosomes with zero mutations are lost from the population by random drift. The mutations prevent the Y chromosome from
recombining with the X, therefore the regeneration of mutant
free chromosomes are prevented and the ratchet is said to move
on one notch. The result of this is population B, the Y chromosomes here have one to four mutations. The chromosomes with
one mutation are lost by random drift and once again not
replaced. The resultant third generation, population C, have
between two and five mutations and so the ratchet continues to
be moved on a notch. This is illustrated in Fig. 3. The Y chromosomes were subjected to many inconsequential and random
alterations, causing the rapid acquisition and loss of many
sequences, making the Y chromosome the least conserved
genetic element in the genome (Dorit et al., 1995; Hammer,
1995; Whitfield et al., 1995).

5. Evolution of the mammalian X and Y chromosomes

Mammals are divided into three major groups, the monotremes, marsupials and placental mammals. It is assumed that
the sex chromosomes of all mammalian species arose from an
ancestor almost 200 million years ago (Hope et al., 1990).
Mammalian sex chromosome are characterised by a larger,
well-conserved X and a Y with few functional genes but more
pseudogenes, often with active X homologues. Each has an
identical PAR which pair at meiosis (see section 6.7). Ohnos
Law and Mullers Rachet only partially explain how this arose;
they do not explain why there is so little conservation between
mammalian PARs and why X genes in one mammalian taxa
may be found on various autosomes of others.
5.1. Addition attrition hypothesis
The addition-attrition hypothesis (Graves, 1995) (Fig. 4)
assumes the proto-gonosomes were small and enlarged by
cycles of autosomal additions followed by the gradual degradation (attrition) of the Y chromosome. This attrition resulted in
non-functional Y homologues of X genes, Y genes with new
novel functions and structural rearrangements. Y attrition
Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Proto Sex Chromosomes

Fig. 3. The mechanism of Mullers Rachet.

The diagram shows the Y chromosome and each
stripe represents a mutation. The chromosomes
in population A have 03 mutations, the chromosomes with 0 mutations are lost from the population by random drift. The lack of recombination
caused by the mutations prevents the chromosomal repair and Mullers Rachet is said to move
on one notch producing population B. Population
B has between 1, and 4 mutations, chromosomes
with one mutation are lost by random drift and
again not replaced. The resultant third generation, population C, has between 2, and 5 mutations and so the Rachet continues to move on.

Population A

Population B

Population C

Fig. 4. The addition-attrition hypothesis for

the evolution of mammalian sex chromosomes.
Autosomal segments (A1, A2, A3) are added to
the proto sex chromosomes creating the X chromosome Recently Added (XRA1, XRA2, XRA3
respectively). At meiosis the added region was
transferred onto the Y chromosome during recombination creating the Y chromosome Recently Added (YRA) region. The chromosomes were
subjected to selection pressures, reducing recombination to produce the accumulation of mutations and thus caused attrition. The sex chromosomes were subjected to cycles of addition, recombination and attrition. (Graves, 1995).

this point they could only be partially differentiated) along with

the autosome from which the partner lost the genetic material.
At the next meiosis these new genes were able to cross over to
the Y if they were located above the X-PAR to effectively
enlarge it. Once added to the Y the region was subjected to the
same selection pressures, drift and isolation from recombination as the Y chromosome. Without recombination the genes
either accumulated mutations resulting in their inactivation
and eventual loss, or gained a male-specific function whose
selection ensured their survival. The X chromosome was not
subjected to the same attrition as the Y chromosome as it could
be passed onto female offspring, with two copies able to recombine with one another.

Cytogenet Genome Res 99:125140 (2002)

129

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

eventually led to the poor conservation within mammals and

different events occurred in different lineages retaining different degrees of XY homology. The PAR of each species thus
appears to be a relic of the latest addition. Attrition of the Y
chromosome is demonstrated by studying the UBE1, AMEL
and RPS4 genes. UBE1 is present on both gonosomes in the
mouse and marsupial, but only on the X in primates suggesting
it has already been degraded and lost or the sequence has extensively mutated and cannot be detected by sequence analysis or
comparative painting. The human RPS4Y and AMELY (Y
genes) have a lower transcription rate than RPS4X and
AMELX (X genes) suggesting attrition also influences transcriptional activity (Graves, 1995). The X chromosome received the autosomal additions to produce a compound X. At
meiosis, the X chromosome paired with the Y chromosome (at

CpG
Islands

SRY Pseudogene
700bp

SRY
2.5kb
1.1kb transcript

Fig. 5. The basic molecular structure of the human SRY gene. Adapted
from Clark and Wall (1996).

Fig. 6. The basic interactions between SRY and DNA. The SRY protein
contains an HMG box which interacts directly with the AATAAG DNA consensus sequence to bend it at an angle essential for transcription.

6. Evolution of the Y chromosome

The Y chromosome appears to have preserved and acquired
a handful of genes essential for male survival and fertility (Jegalian and Lahn, 2001). Of the defined Y chromosome genes,
twelve are in the PAR (see section 6.7), some are pseudogenes,
degraded copies of X homologues and some are multi-copy,
male-specific genes unique to the Y and expressed only in the
testis (Graves, 1998).
6.1. The male sex determination factor (SXD)
Jost et al. (1973) demonstrated that mammalian sex differentiation begins in the foetal testis and suggested the existence
of a sex determination factor (SXD) previously called the TDF
(testis-determining factor). The key observation proving that
the mammalian Y chromosome contained the SXD was the
study of sex chromosome aneuploidy. Regardless of the number of X chromosomes, when the Y is present, the phenotype of
the individual is predominantly male (c.f. Drosophila section
3.3.3). Rare exceptions only occur (e.g. XY females) where the
SXD is deleted or inactivated. Through studying these rare
individuals the male determining gene SRY was isolated.
6.1.1. The discovery of SRY: The first candidate gene for
SXD was mapped to the distal region of Yp and named ZFY
(Page et al., 1987). This was later discounted due to the unusual
tissue expression in the mouse and its autosomal location in
marsupials and monotremes. More refined human mapping
resolved a more accurate minimal sex determination region
excluding ZFY. Studies of XX males with a Y translocation,
permitted the cloning of SRY (sex-determining region Y), map-

130

Cytogenet Genome Res 99:125140 (2002)

ping it to Yp11.3 (Sinclair et al., 1990). Studies of XY females

found that they carried mutations in this gene. Functional evidence was provided by inserting SRY into mouse XX embryos
resulting in transgenic males (Koopman et al., 1990, 1991).
With the exception of some species of vole (section 3.3.2), the
SRY probe always maps to the Y chromosome in eutherians.
6.1.2. The structure of the SRY gene: Sequence comparisons
between the SRY gene in mammals demonstrated poor conservation, a fairly surprising result considering its importance in
male development (Jegalian and Lahn, 2001). The basic molecular structure of SRY can be seen in Fig. 5. Downstream of
SRY is an SRY pseudogene, a second copy that has become
further degraded and has lost all function. Downstream of both
genes are CpG islands, sequences rich in cysteine and guanine
residues (i.e. have the consensus sequence CGCCCCGC).
Transcription factors such as WT-1 bind to these regions and
initiate their transcription and may be the first step in the male
developmental pathway.
SRY encodes a protein containing a 79-amino acid high
mobility group (HMG box) able to interact with the DNA consensus sequence AATAAG in the minor groove (Murphy et al.,
2001) as shown in Fig. 6. This interaction causes it to bend
through 65 in humans and 80 in mice (Graves, 1998) allowing close association between specific sequences or proteins
required for its transcriptional activity. The human and mouse
HMG-box is 71 % homologous and there is 63 % homology
between human and marsupials, outside the box there is no
homology suggesting its activity lies solely in the box (Graves,
1998). Mutations causing 46XY male-to-female sex reversal
are clustered within the HMG-box region. The HMG-box is
flanked by nuclear localisation sequences (NLS) to allow entry
into the nucleus via importin-beta, the binding and hydrolysis
of Ran-GTP dissociates the importin-beta:protein complex
(Forwood et al., 2001). Mutations in the carboxyl-NLS blocks
entry into the nucleus (Li et al., 2001), obstructing access to the
DNA and impeding its functioning. Furthermore the HMGbox directly interacts with the DNA binding domain of the
androgen receptor on the X chromosome to negatively regulate
transcription (Yuan, 2001).
6.1.3. Function of SRY: Sequence comparisons and functional analysis have provided evidence that SRY arose from
SOX3 (SRY-like HMG box containing). The Y chromosome
bears the SXD (SRY) and on the X the SXD is (SOX3) as
shown in Fig. 7b. The heterozygote SRY/SOX3 produces
males, and the homozygote SOX3/SOX3 produces females.
SOX3 is a transcription factor, activating female developmental pathways, but may also inhibit the transcription of
SOX9, key in the male determination pathway and sex differentiation (Graves, 1995). SOX9 mutations can lead to XY
females even when SRY is intact (Bowles et al., 2001). SOX3 is
very well conserved between human, mouse and marsupial and
is expressed in the central nervous system and genital ridge.
SRY however is far less conserved and only expressed in the
testis thus implying that SOX3 is the ancestral gene of SRY
(Ellis, 1996).
SRY may function by directly controlling the transcription
of SOX9, but more likely represses SOX3 to allow the transcription of SOX9 (Graves, 1998, 2000) and initiates a cascade
Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

CpG
Islands

f
PAR

PAR
SRY
SST

SST

SST

SST

SST

SSTM

SST

SSTM

XIST
SOX3

SOX3 SOX3

Proto Proto
Y
X

SRY SOX3

Proto Proto
Y
X

SRY SOX3

Proto Proto
Y
X

SST
XIST

SRY

Proto Proto
Y
X

SSTM
SRY

SSTM2

SOX3

Proto Proto
Y
X

X Y

Fig. 7. Flow diagram to depict a theoretical sequence of events that led to the evolution of the sex chromosomes. (a) Two
homologous chromosomes. (b) The mutation of SOX3 to SRY to form the proto sex chromosomes. (c) The mutation of a sexually
selected trait (SST) to a sexually selected trait in males (SSTM) on the proto Y chromosome. (d) Mutation on the proto X produced
the XIST which suppressed recombination. (e) Accumulation of mutations on the proto Y which may have resulted in rearrangements. (f) Additions of autosomes onto the sex chromosmes to produce the PAR and other SSTM genes.

SOX3

SOX3

SOX9

Male
Developmental
Pathway

SRY

SOX3*
(1 copy = Male)

SOX9

Male
Developmental
Pathway

SOX3* + SOX3
(2 copies = Female)

Fig. 8. (a) The interactions between SOX3, SRY and SOX9. The dotted
lines represent an inhibition, the solid arrows represent an activation and the
dashed arrows a mutation. (b) The suggested interaction of a theoretical
mutant SOX3 (SOX3)) Gene. The dotted lines represent inhibition, the solid
arrows activation and the dashed arrows a mutation.

with doublesex gene, the SXD in Drosophila (Ottolenghi et al.,

2002).
Although the mechanism of sex determination differs between mammals, birds and reptiles, the developmental pathway of the gonads appears to be conserved. In many reptiles
SDM is controlled by temperature. Within the correct range,
the transcription of SOX9 is stimulated, therefore reptilian
temperature and SOX9 have similar functions to eutherian
SRY and SOX9 respectively. The protein expressed by SOX9
has been detected in both sexes of reptilian embryos, in early
development SOX9 is down regulated in females but maintained or up-regulated in males (Torres-Maldonardo et al.,
2001). AMH is expressed earlier in the testis of the American
alligator suggesting SOX9 is more likely to be involved with
testis structural organisation rather than differentiation and
AMH regulates SOX9 (Western et al., 1991a), whereas mammalian SRY and/or SOX-9 are thought to regulate AMH

Cytogenet Genome Res 99:125140 (2002)

131

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

of activation and/or inhibition of other genes such as AMH (a

gene expressed in male determination). A mouse mutation
upstream of SOX9 results in the production of SRY-negative
males, thus the mutation may prevent the binding of the transcription repressor SOX3 (Chan and Rennert, 2002) (see
Fig. 8a).
Studies of XX hermaphrodites have detected a low expression of SRY, probably due to mosaicism with a Y-bearing cell
line (Ortenberg et al., 2002) suggesting that the complete genesis of testis requires a threshold of SRY concentration to be
reached. Studying XX-SRY-negative hermaphrodites permits
the discovery and confirmation of other genes involved in male
sex determination. This condition may partially be attributed
to an autosomal recessive mutation (Switonski et al., 2002) but
some cases have shown an elevated expression of SOX9 (Pailhoux et al., 2001). This was ascribed to a mutant SOX3 gene
that allowed the transcription of SOX9 (see below).
6.1.4. The origin of SRY: It is debatable whether the mutation of SOX3 to SRY was the initial feature that distinguished
the Y from the X chromosome. SRY has not yet been found in
monotremes and there is no direct evidence to suggest SRY is
the SXD in marsupials as it is broadly expressed. This suggests
SRY was not the original male SXD (Graves, 2000), perhaps
the X homologue of SOX3 mutated into SOX3* which may
lack the ability to inhibit SOX9 transcription, except when two
copies were present therefore functioning in a dose-sensitive
manner (Graves, 2000). This is simplified in Fig. 8b. A further
hypothesis has developed from the recently cloned marsupial
ATR gene. Marsupial ATRY is expressed at an appropriate
time, is testis specific and probably derived from a Y gene with
an X homologue (ATRX). ATRX is still present in the human
and mouse but ARTY is missing, maybe replaced by SOX3
before being mutated into SRY. Mutations in the human and
mouse ATRX have been connected with XY SRY positive
male-to-female sex reversal. A further causal candidate for this
condition is the human DMRT1 gene on 9p, involved in
human testis differentiation. DMRT1 has significant homology

SRY

SOX-9

AMH

Continued Testis
Determination
Cascade

Temperature
Dependant Sex
determination
Species

Specific Male
determining
Temperature

AMH

SOX-9

Continued Testis
Determination
Cascade

Fig. 9. Comparisons of SOX-9 and AMH in testis determination in the

mammalian XX/XY SDM and temperature dependent sex determination
species. The initiation of the cascade varies between systems of sex determination, however the equivalent genes are activated to produce the same
event.

(Western et al., 1999b). This is simplified in Fig. 9. Many more

genes are thought to be involved but are not shown; furthermore some of these interactions are yet to be confirmed.
6.2. The acquisition of a sexually selected trait in males
(SSTM)
After the acquisition of SRY (Fig. 7c) other mutations may
have occurred that conferred an advantage to males (but not
females) and accumulated on the Y chromosome. SSTM is a
theoretical gene thought to have a pre-mutated X homologue.
This selective advantage made the males stronger and more
likely to reproduce than their male counterparts lacking SSTM,
allowing the gene to spread throughout the population. Recombination between SRY or SSTM must be suppressed to prevent
their loss onto the X and retain the advantage (Ellis, 1996).
6.3. Recombination suppression
Suppression in recombination caused the Y chromosome to
become genetically isolated (see Fig. 7d.) Many factors can suppress recombination such as the creation of XIST (see section
7.1) from a mutation on the X chromosome, structural conformation such as an inversion loop and the gain of heterochromatin (Karpen et al. 1996; Jegalian and Lahn, 2001), these are
listed in Table 4. In this issue, Griffin et al. (2003), based on
studies in the genus Oreochromis, argue that recombination
suppression might have come about through an initial de novo
stochastic expansion of heterochromatin. It is proposed that
under conditions suitable for the evolution of sex chromosomes
(i.e. in the presence of a sex determining gene(s)) an accumulation could lead to a delay in synapsis and a subsequent reduction in recombination. Recombination suppression by whatever mechanism resulted in the ability to retain and spread the
advantageous genes throughout the male population. Conversely, this same mechanism probably caused adverse effects.
After 34 generations without recombination in Drosophila a
test cross was performed and Rice (1994) observed a significant
reduction in lifespan and ability to reproduce, probably due to
an accumulation of recessive mutations. Without recombination, deleterious mutant genes on the Y chromosome were able
to accumulate by Mullers Rachet (Muller 1964; Felsenstein
1974; Charlesworth 1978; section 4.3) and attrition (Graves,
1995; section 5.2); this is seen in Fig. 7e.

132

Cytogenet Genome Res 99:125140 (2002)

6.4. The acquisition of the pseudoautosomal region (PAR)

The pseudoautosomal region (PAR) ensures recombination
and correct segregation at meiosis. Failure of this results in
reduced fertility in mice (Hassold et al., 1991; Kipling et al.,
1996). PAR genes have autosomal or partial sex-linked inheritance (Burgoyne, 1982; Rappold, 1993); some may have been
incorporated into the Y specific region and may have gained
novel functions. The PAR is shown in Fig. 7f. Humans have
two PARs. (Cooke et al., 1985; Simmler et al., 1985; Freije et
al., 1992); PAR1 is 2,600 kb on the tips of the short arms
(Brown et al., 1988; Petit et al., 1988; Rappold and Lehrach,
1988), whereas the mouse has one PAR F2,000 kb (Singh and
Jones, 1982; Kipling et al., 1996). Deletions within this region
cause infertility due the failure of spermatogenesis, demonstrating the importance of PAR1 (Gabriel-Robez et al., 1990;
Mohandas et al., 1992). PAR2 is F500 kb (Kvalay et al., 1994;
Toder et al., 2000) and recombines in F4% of meiotic products
suggesting it is not essential to proper segregation but stabilises
the XY bivalents (Bickmore and Cooke, 1987).
Marsupials are thought not to contain a PAR (Sharp, 1982)
instead they form end-to-end attachments at mid-pachytene
enabling segregation (Graves, 1998). Recently, Page et al.
(2003) established, in Thylamys elegans that XY pairing is
maintained through a dense plate developed from their axial
elements. The monotremes have large sex chromosomes with a
microscopically visible PAR consisting of Xp and Yq (Wrigley
and Graves, 1988). The evolution of the PAR2 in humans is
recent and not present in other great apes and old world monkeys which diverged about 35 million years ago (Ellis et al.,
1990). Figure 10 shows how little conservation there is in the
PAR of the eutherian species and this can be explained by the
addition-attrition hypothesis. The PAR is a remnant of the last
autosomal addition that is likely to differ between species. The
steroid sulphatase gene (STS) with a non-functional Y homologue (Yen et al., 1987) is found in the PAR in mouse and pig
(Quilter et al., 2002), but outside in human and rat (Salido et
al., 1996). Furthermore in the prosimian species lemur and galago it is autosomal (Yen et al., 1988). KALX/Y and AMELX/Y
are localised close to STS on the sex chromosomes (Del Castillo
et al., 1992; Incerti et al., 1992). This is illustrated in Fig. 11.
Genes are thought to be transferred from the autosomes to
the PAR then incorporated into the Y chromosome by an
inversion (Graves, 1995). As the pig carries STS in the PAR
and the prosmians carry it on an autosome, it is unlikely that
this occurred. As the prosimians and the rodents are thought to
have diverged at the same time it is unlikely that the gene was
independently transferred to the pig, rodent and human gonosomes. A more plausible explanation is that STS was transferred onto the sex chromosomes over 100 million years ago
and transferred back onto the prosimian autosome after this
lineage diverged (Ellis, 1996). To confirm this, surrounding
genes should be identified and the location determined in the
other species.
6.5. The resultant Y
The Y chromosome in dasyurid marsupials is very small
(10 Mbp), only containing the SXD and spermatogenesis genes.
Other marsupials eliminate the Y chromosomes from the emDownloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Mammals
(XX/XY sex
determination
system)

Table 4. Factors that cause the suppression of

recombination

Factor

Mechanism of action

XIST gene

Causes the premature condensation in male meiosis to prevent recombination complexes

associating (See section 7.1) resulting in over 96% of the human Y-chromosome unable to
crossover (Jegalian and Lahn, 2001).

Inversion loops

Inversion loops cause a structural change, whether pericentric or paracentric it causes a

reduction in the likelihood of recombination. This is because one chromosome will have to
loop inside the other to pair in meiosis due to the structural rearrangemnent of the genes (see
Fig. 15). Furthermore, recombination may appear suppressed if two loci become positioned
near to an area that recombines infrequently such as the centromere.

Gain of
heterochromatin

Large regions of heterochromatin reduced the likelihood of recombination and are rarely seen
to form chiasmata in Drosophila (Karpen et al., 1996). The origin maybe from selfish or
parasitic DNA whose only function is to propagate itself.

Fig. 10. The PAR on the Y chromosomes of

eutherian species a breach of Ohnos Law.
Adapted from Graves et al. (1998).

(15)

Pig

STS = PAR

Prosimians

STS = Autosomal

Rat

STS = X-linked

Mouse

STS = PAR

Human

STS = X-linked

Fig. 11. Mammalian evolutionary tree denoting the location of the STS
Gene (approximate time of divergence in brackets, Myrs).

but a pseudogene in the mouse. The UBE1Y gene is active in

the mouse, pseudoautosomal in the platypus, a partial pseudogene in the monkey and has completely disappeared in the
human (Graves, 1998). Many other genes tell a similar story.
By examination of the nucleotide sequence of 19 genes in the
non-recombining portion of the sex chromosomes in various

Cytogenet Genome Res 99:125140 (2002)

133

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

bryo in somatic growth, no longer requiring it at all. Mole voles

have completely lost the Y chromosome suggesting that the
SXD has been transferred to a different chromosome (Toder et
al., 2000). The human genes SRY, RPS4Y, SMCY and RBMY
have been mapped to two small regions at either end of the
human Y chromosome. The marsupial gene homologues have
been mapped to the tiny Y chromosome suggesting they have
been present for the last 130 million years. These genes represent the Y conserved region (YCR), and appear to have separated by an inversion (Graves, 2000). The localisation and transcription of the genes vary greatly in different mammalian species. The remaining genes are part of the YRA (Y recently added) region. It has been established that genes from the human
YRA mapped to the XRA and was probably added by the addition-attrition hypothesis. The Y is largely made up of 3.4- and
2.5-kb repeat families DYZ1 and DYZ2, together making up
5070 % of DNA of the Y chromosome (Charlesworth, 1991;
Willard, 2001); it also contains a 650-kb alphoid satellite block
and highly repeated DNA Alu and LINE elements. All repeat
elements differ in length between species suggesting no significant function. This is shown in Fig. 12.
Y attrition has led to extensive variation in the gene content
and activity in mammals. STS is active in the mouse but is a
psueudogene in humans, whereas AMEL is active in the human

mammalian studies, Jegalian and Lahn (2001) hypothesized

that the degradation of the Y genes occurred in four discrete
groups starting about 300 million years ago. Each phase
involved a suppression of crossing over in meiosis, implying
that the smaller the number of differences the more recently
they diverged and that a larger number implies crossing over
halted earlier in evolution. This is demonstrated in Fig. 13.

al., 1991; McKay et al., 1992) and are termed X recent additions (XRA) (Glas et al., 1999). XRA genes are located in three
marsupial autosomes suggesting at least three different autosomal additions. XRA is conserved in eutherians suggesting the
addition occurred after the eutherian and metatherian lineages
diverged, but before the eutherian lineages diverged from each
other. The XRA and XCR regions are shown in Fig. 14 (Graves
and Watson, 1991). The conservation of the XRA differs
between eutherians as shown in Fig. 15 (Quilter et al., 2002).

7. Evolution of the mammalian X

The isolation and purification of human and mouse X gene
sequences has led to the production of probes that allow the
detection of homologous genes in other mammalian species by
cross-species hybridisation (zoo blots or ZOO-FISH). Genes on
human Xq and proximal Xp have been mapped to the X chromosomes of marsupials and monotremes (Watson et al., 1990;
Spencer et al., 1991b; McKay et al., 1992) suggesting they arose
from a common ancestor, F170 million years ago (X conserved
region, XCR). Human distal Xp genes map to marsupial and
monotreme autosomes (Spencer et al., 1991a; Watson et

Fig. 12. The human Y chromosome indicating the presence of the PAR, Y-chromosome conserved (YCR) region with known genes, Y chromosome recently added (YRA) region and area of
heterochromatin.

7.1. X inactivation
Genes are progressively mutated and lost on the Y, in order
to prevent a partial monosomy phenotype in the male, a system
of dosage compensation has evolved. Dosage compensation
mechanisms exist in Drosophila, Caenorhabditis elegans and
partially in marsupials (Ohno, 1967) (see Fig. 16). Eutherians
however have evolved a specialist mechanism of dosage compensation via X inactivation through the XIST gene to produce
similar phenotypes in both sexes as one X chromosome is inactivated in the female.
7.1.1. The Lyon hypothesis of X inactivation: Lyon and Russell (1961) independently postulated the phenomenon of X
inactivation (Lyonisation). They postulated that one X chromosome is randomly inactivated in all tissues at female embryogenesis, affecting the entire chromosome with all cell
descendants displaying the same inactivation pattern (Lifchytz
and Lindsley, 1972; Gartler et al., 1991). Breaching Lyons
hypothesis, marsupials (and maybe mice) preferentially inactivate the paternally derived X chromosome via an imprint (Takagi and Sasaki, 1975; Zuccotti and Monk, 1995). Mice are
now thought to lose their paternal imprint after the first dozen
rounds of embryonic cleavage, randomising inactivation during the differentiation of the inner cell mass (Hartshorn et al.,
2002).
7.1.2. Mechanisms of preferentially inactivating paternallyderived X chromosomes: Inactivation is controlled by the X
inactivation centre (Xic) at the Xce locus, XIST is accepted as
the gene controlling it. XIST is the only known gene to be
expressed exclusively on the inactive X and transcribed but not
translated. Originally, XIST may have evolved as a factor to
suppress recombination, inactivation may have been a coincidental effect (Ellis, 1996). The XIST RNA binds to the X chro-

134

Cytogenet Genome Res 99:125140 (2002)

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Fig. 13. The degeneration of the Y. Adapted

from Jegalian and Lahn (2001).

Fig. 14. The mammalian X. (a) The coloured

areas represent XCR. (b) The location of the human XRA genes in marsupials and monotremes.
Adapted from Graves and Watson (1991).

Human

Pig

PRKX
STS, KAL, AMELX

Mouse
Usp9x
Utx

KAL, STS, PRKX

AMELX
EIF2s3X
ZFX

Zfx

EIF2s3X, ZFX

Eif2s3x
USP9X
DBX
UTX

Smcx

USP9X
DBX, UTX

Amelx
SMCX
Fig. 15. Comparisons of the order of Xp genes
in the human, pig and mouse. Adapted from
Quilter et al. (2002).

SMCX

Xq

Xq

Sts

Xq

Cytogenet Genome Res 99:125140 (2002)

135

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Fig. 16. Dosage compensation. adapted from

Graves et al. (1998b). Y genes became mutated
and began to lose their function in cycles as the
additionattrition hypothesis suggests. The Y
gene produced fewer protein molecules (top),
hence the X gene was up-regulated or the Y allele
was up-regulated to amplify expression that may
have resulted in the gain of a new function (as
shown by (*) after a further mutation).

Female XX
(Active Xs)

Male XY
(Active X)

Oogenesis
(CpG sites remain methylated)

Spermatogenesis
(Demethylation of CpG sites)

Active Xs

Inactive X

Me

Promoter

CpG

CpG

15kb

XIST

XIST

Fertilisation

Female Offspring
Maternal
Active X

XX

mosome inactivating the X (Borsani et al., 1991; Brown et al.,

1991b; Rastan, 1994). Initially, XIST mRNA was detected at
the four-cell stage (Zuccotti and Monk, 1995), but has now been
found at the late two-cell cleavage stage (Hartshorn et al., 2002)
where, at this stage only, it is transcribed from both Xs. The
proto-active X switches off transcription between the two-cell
and four-cell stage. The XIST mRNA is then allowed to accumulate in the nucleus from the proto-inactive X. This pool
maintains the level of inactivation without requiring immediate further transcription. (Hartshorn et al., 2002). X-chromosome inactivation is thought to be progressive, switching off
genes portion-by-portion or gene-by-gene down the chromosome.
The XIST mRNA transcript directly associates with the X
chromosome to initiate and spread inactivation (Lifschytz and
Lindsley, 1972). The X chromosome becomes stable (Brown
and Willard, 1994), prematurely condenses (McCarrey and
Dilwork, 1992; Richeler et al., 1992; Salido et al., 1992) and
forms the inactive Barr body (Brockdroff et al., 1992; Brown et
al., 1992). Under a light microscope, the Barr body appears as a
condensed heterochromatic triangular structure juxtaposed to
the nuclear membrane in interphase and amongst the chromatin replicated last in S phase (Gartler et al., 1992). Mutation
studies in mouse Xist results in permanently active X chromosomes (Penny et al., 1996). A counting mechanism ensures that
all but one X chromosome is inactivated (Grumbach et al.,
1963).
Studies using sperm, eggs and pre-implantation embryos of
mice detected CpG islands in the promoter region of Xist.
These are methylated on the X chromosome of the egg and
maternally derived X in the embryo, but not on the X chromo-

136

Cytogenet Genome Res 99:125140 (2002)

Paternal
Inactive X

Male Offspring
Paternal

YX

Maternal
Active X

some in the sperm or the paternally derived X chromosome,

suggesting imprinting occurs in oogenesis but not spermatogenesis (Ariel et al., 1995, Zuccotti and Monk, 1995). This imprint
allows the maternally derived X chromosome to remain active
(especially in males as their X is always maternally derived).
The imprint however must be lost in spermatogenesis, via
demethylation in meiosis so that it may become preferentially
inactive in future offspring. It is thought likely that the binding
of the Xist mRNA to the X chromosome to prematurely condense it was the original mechanism, the imprints and methylations probably came later to stabilise the chromosome to
increase its inertness. The mechanism by which the paternal X
chromosome is thought to be preferentially inactivated is
shown in Fig. 17.
7.1.3. Random inactivation: Eutherians randomise X inactivation (Migeon, 1994). When deleterious recessive mutations
are present, this is advantageous as the full effects are prevented by producing functional protein by the X carrying the wildtype allele (see Fig. 18). The mechanism is similar to that in
marsupials, but the imprint is lost (Kay et al., 1993, 1994; Zuccotti and Monk, 1995). The Tsix gene, so called because as it is
an antisense transcript of Xist, has been found in the Xic region
in addition to Xist (Lee et al., 1999). The expression pattern
suggests Tsix regulates Xist. The expression of the Tsix mRNA
lasts longer on the active X than the Xist, enabling it to become
established as the active X, choosing which X to inactivate.
Tsix mRNA negatively regulates Xist, either by preventing Xist
transcription or by interacting with Xist mRNA directly. Mutated Tsix allows the constitutive production of Xist and the
chromosome carrying the mutation becomes the inactive X
(Willard et al., 2001).
Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Fig. 17. The mechanism of preferentially

inactivating the paternal X. The XIST gene, responsible for X inactivation, is demethylated in
spermatogenesis. Thus all female offspring will
have one active X and one inactive X, hence male
offspring always receive an active X.

Male
Y

Female

Male

Female

Parents

Offspring
Male has
disease
phenotype

Maternal

Fig. 18. How randomisation of inactivation

can benefit females carrying an X-linked recessive
mutation. (a). Mutation on a paternal X chromosome (b). Mutation on a maternal X chromosome. The recessive mutation is shown by P, the
normal gene is not shown.

Paternal

Preferentially inactivated
chromosome

Active Inactive

Active Inactive

Non-functional
protein produced

Functional protein
produced

Maternal

Paternal

Active Inactive

Active Inactive

Non-functional
protein produced

Randomisation

Functional protein
produced

Randomisation

Both functional and non-functional

proteins present producing a less
severe phenotype

Table 5. The functions of the Zfx/y genes in the human and mouse. The house keeping genes are
active genes with original function, the inactivated gene is underlined and the gene with a new function
(i.e. from a mutation) is highlighted by italics. Human ZFX and ZFY have the same function so have
two active housekeeping genes. Mouse Zfx and Zfy have different functions so have one active housekeeping gene, requiring the second gene to be inactive in females to compensate for the male specific
function of Zfy in males
Males
Zfx

Zfx

Zfx

housekeeping function
male-specific function

housekeeping function
housekeeping function

housekeeping function
housekeeping function

housekeeping function
inactivated

7.1.4. A further breach of Lyons hypothesis: Lyons hypothesis states that the entire inactive X is inactivated. The main
exceptions are the PAR genes, but others are also exempt. ZFX
(Xq21.3 q22.1) encodes a zinc finger protein that binds to
DNA, it escapes X inactivation in humans, but not in mice
(Adler et al., 1991). The explanation lies in the Y chromosome
homologue: ZFY. ZFX and ZFY are 97 % homologous suggesting that they bind the same or similar DNA sequences. In mice
Zfx is expressed in male and female adult tissues, whereas Zfy
is only expressed in adult and foetal male tissues, each have
different, non-overlapping functions (Schneider-Gadicke et al.,
1989; Palmer et al., 1990; Graves, 1998, 2000). This is summarised in Table 5.

8. Conclusions, commentary and the future of sex

Plato described Zeus creating sex as a punishment, splitting
apart hermaphrodites, making them forever seek their partner,
and conception could take place only if man chances upon

woman (Jones, 1994). Although a myth, it is possible that this

may have inspired Charlesworth (1991) in developing his theory that mammals evolved from cosexuals. The proto-gonosomes may have resulted from a male-sterility/female-fertility
and female-sterility/male-fertility mutation on different chromosomes producing the SXD. This argument however contains
a possible inconsistency. Each chromosome would be specific
to each sex, in vertebrates there is no chromosomal (or genic)
element specific to the homogametic sex. Therefore in mammals, the chromosome conferring female-fertility/male-infertility (i.e. the X) is carried in the males and likely to be detrimental to their reproduction. This can be overcome with a mutation
similar to the SOX3* postulated by Graves (2000) where it is
only detrimental to males when it is expressed in a two-fold
dose-dependent manner, producing infertile XXY males. Ellis
and Graves independently postulated that mammalian sex
originated from parthenogens, Ellis (1996) suggested that the
mutation of SOX3-to-SRY was the first. Graves (1995) suggested that sex was first determined by a different gene that later
mutated into, or was replaced by SOX3. In our opinion Graves

Cytogenet Genome Res 99:125140 (2002)

137

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Humans
Mice

Females

Zfy

is likely to be more accurate, as there were likely to be several

mutations before SRY was secured in mammals; maybe mammalian sex determination was controlled directly by SOX9 (as
in reptiles) or by another gene that controlled its transcription.
To the best of our knowledge however, there is no direct evidence to suggest either scheme is correct, and it still does not
explain how sexual organisms arose from asexual ones. We
believe that it is unlikely that the SOX3-to-SRY mutation was
the first mutation that differentiated the proto-sex chromosomes. This mutation was likely to have been corrected in DNA
synthesis. A more plausible explanation would be an initial
suppression in recombination via a rearrangement or acquisition of XIST. As the chromosomes were still homologous at this
stage, XIST would cause an inactivation of the proto-Y producing functional monosomy in both sexes. This may have resulted
in the doubling of expression of the sex chromosomes and as
the Y was degraded, dosage compensation was already established. Either way, recombination was suppressed allowing
additions to be made, mutations accumulated on the Y
(Graves, 1995; Muller, 1967) and any new advantageous mutations that occurred were fixed in the population (Muller,
1967).
The sex chromosomes are constantly evolving, what will
become of them? Maybe the whole genome will be transferred
to the gonosomes in several autosomal additions leaving only

two chromosomes, a very large X and relatively small Y due to

attrition. This will result in a heavy sperm (carrying the X) and
light sperm (carrying the Y). This may lead to significantly
more male offspring as the lighter Y sperm will be faster and
more likely to fertilise the egg. Furthermore, males would effectively be haploid (except for any active homologous Y genes)
whereas females would have two copies of the genome, resembling the haploid diploid system as seen in bees, but with X
inactivation (Graves et al., 1998a). Eutherians may follow marsupials and lose the PAR altogether, preventing further autosomal additions to be transferred to the Y as they do not pair via
the PAR in meiosis. Instead, addition maybe made to either the
X or Y producing the sex chromosomes XY1Y2 or X1X2Y
respectively, a system already seen among some marsupials,
monotremes and fish (Graves et al., 1998). A new SXD may
become mutated and succeed SRY. This may produce a completely different pair of sex chromosomes. Lemmings have a
modified X, along with the usual X and Y, which suppresses
the Y. Vole moles appear to lack a Y chromosome and the SRY
altogether. These may have a more advanced SDM that we
have yet to evolve. If this is true, then the Y chromosome so
long the subject of intense scientific interest (and, it might be
said, some macho posturing) may be doomed and ripe for
extinction.

References

138

Brown CJ, Ballabio A, Rupert JL, Lafreniere RG,

Grompe M, Tonlorenzi R, Willard HF: A gene
from the region of the human X inactivation centre
is expressed exclusively from the inactive X chromosome. Nature 349:3844 (1991).
Brown CJ, Hendrich BD, Rupert JL, Lafreniere RG,
Xing Y, Lawrence J, Willard HF: The human
XIST gene: analysis of a 17-kb inactive X-specific
RNA that contains conserved repeats and is highly
localized within the nucleus. Cell 71:527542
(1992).
Bull JJ: Evolution of Sex Determining Mechanisms
(Benjamin/Cummings, Don Mills 1983).
Burgoyne PS: Genetic homology and crossing over in
the X and Y chromosomes of mammals. Hum
Genet 61:8590 (1982).
del Castillo I, Cohen-Salmon M, Blanchard S, Lutfalla
G, Petit C: Structure of the X-linked Kallmann
syndrome gene and its homologous pseudogene on
the Y chromosome. Nature Genet 2:305310
(1992).
Chan WY, Rennert OM: Molecular aspects of sex differentiation. Current molec Med 2:2537 (2002).
Charlesworth B: Model for evolution of Y chromosomes and dosage compensation. Proc natl Acad
Sci, USA 75:56185622 (1978).
Charlesworth B: The evolution of sex chromosomes.
Science 251:10301033 (1991).
Charlesworth B: The evolution of chromosomal sex
determination and dosage compensation. Current
Biol 6:149162 (1996).
Charlesworth B, Charlesworth D: The degeneration of
Y chromosomes. Phil Trans R Soc London Series
B: Biol Sci 355:15631572 (2000).
Christidis L: Chromosomal repatterning and systemics
in the passeriformes (songbirds). Chrom Today
10:279294 (1989).

Cytogenet Genome Res 99:125140 (2002)

Clark MS, Wall WJ: Chromosomes, the Complex Code

(Chapman and Hall, London 1996).
Cooke HJ, Brown WR, Rappold GA: Hypervariable
telomeric sequences from the human sex chromosomes are pseudoautosomal. Nature 317:687692
(1985).
Crow JF: Advantages of sexual reproduction. Dev Genet 15:205213 (1994).
Dominguez-Steglich M, Meng G, Bettecken T, Muller
CR, Schmid M: The dystrophin gene is autosomally located on a microchromosome in chicken. Genomics 8:536540 (1990).
Dorit RL, Akashi H, Gilbert W: Absence of polymorphism at the ZFY locus on the human Y chromosome. Science 268:11831185 (1995).
Ellis NA: Human Genome Evolution, in Strachan T,
Read A (eds): Human Molecular Genetics, Chapter 10 (John Wiley and Sons, New York 1996).
Ellis NA, Goodfellow PJ, Pym B, Smith M, Palmer M,
Frischauf AM, Goodfellow PN: The pseudoautosomal boundary in man is defined by an Alu repeat
sequence inserted on the Y chromosome. Nature
337:8184 (1989).
Ellis N, Yen P, Neiswanger K, Shapiro LJ, Goodfellow
PN: Evolution of the pseudoautosomal boundary
in Old World monkeys and great apes. Cell
63:977986 (1990).
Felsenstein J: The evolutionary advantage of recombination. Genetics 78:737756 (1974).
Forwood JK, Harley V, Jans DA: The C-terminal
nuclear localization signal of the sex-determining
region Y (SRY) high mobility group domain mediates nuclear import through importin beta 1. J
biol Chem 276:4657546582 (2001).
Fredga K: Aberrant chromosomal sex-determining
mechanisms in mammals, with special reference to
species with XY females. Phil Trans R Soc London
Series B: Biol Sci 322:8395 (1988).

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Adler DA, Bressler SL, Chapman VM, Page DC, Disteche CM: Inactivation of the Zfx gene on the
mouse X chromosome. Proc natl Acad Sci, USA
88:45924595 (1991).
Allen EF, Albrecht I, Drake JW: Properties of bacteriophage T4 mutants defective in DNA polymerase.
Genetics 65:187200 (1970).
Ariel M, Robinson E, McCarrey JR, Cedar H: Gametespecific methylation correlates with imprinting of
the murine Xist gene. Nature Genet 9:312315
(1995).
Baverstock PR, Adams M, Polkinghorne RW, Gelder
M: A sex-linked enzyme in birds Z-chromosome
conservation but no dosage compensation. Nature
296:763766 (1982).
Bickmore WA, Cooke HJ: Evolution of homologous
sequences on the human X and Y chromosomes,
outside of the meiotic pairing segment. Nucl Acids
Res 15:62616271 (1987).
Borsani G, Tonlorenzi R, Simmler MC, Dandolo L,
Arnaud D, Capra V, Grompe M, Pizzuti A, Muzny
D, Lawrence C: Characterization of a murine gene
expressed from the inactive X chromosome. Nature 351:325329 (1991).
Bowles J, Koopman P: New clues to the puzzle of mammalian sex determination. Genome Biol 2:1025
(2001).
Brockdorff N, Ashworth A, Kay GF, McCabe VM,
Norris DP, Cooper PJ, Swift S, Rastan S: The
product of the mouse Xist gene is a 15-kb inactive
X-specific transcript containing no conserved ORF
and located in the nucleus. Cell 71:515526
(1992).
Brown WR: A physical map of the human pseudoautosomal region. EMBO J 7:23772385 (1988).
Brown CJ, Willard HF: The human X-inactivation
centre is not required for maintenance of Xchromosome inactivation. Nature 368:154156
(1994).

Jost A, Vigier B, Prepin J, Perchellet JP: Studies on sex

differentiation in mammals. Recent Prog Horm
Res 29:141 (1973).
Just W, Rau W, Vogel W, Akhverdian M, Fredga K,
Graves JA, Lyapunova E: Absence of Sry in species
of the vole Ellobius. Nature Genet 11:117118
(1995).
Karpen GH, Le MH, Le H: Centric heterochromatin
and the efficiency of achiasmate disjunction in
Drosophila female meiosis. Science 273:118122
(1996).
Kay GF, Penny GD, Patel D, Ashworth A, Brockdorff
N, Rastan S: Expression of Xist during mouse
development suggests a role in the initiation of X
chromosome inactivation. Cell 72:171182 (1993).
Kay GF, Barton SC, Surani MA, Rastan S: Imprinting
and X chromosome counting mechanisms determine Xist expression in early mouse development.
Cell 77:639650 (1994).
Kipling D, Salido EC, Shapiro LJ, Cooke HJ: High frequency de novo alterations in the long-range genomic structure of the mouse pseudoautosomal
region. Nature Genet 13:7880 (1996).
Koopman P, Munsterberg A, Capel B, Vivian N, Lovell-Badge R: Expression of a candidate sex-determining gene during mouse testis differentiation.
Nature 348:450452 (1990).
Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R: Male development of chromosomally
female mice transgenic for Sry. Nature 351:117
121 (1991).
Kvaloy K, Galvagni F, Brown WR: The sequence organization of the long arm pseudoautosomal region
of the human sex chromosomes. Hum molec Genet
3:771778 (1994).
Lee C, Griffin DK, OBrien PC, Yang F, Lin CC, Ferguson-Smith MA: Defining the anatomy of the
Rangifer tarandus sex chromosomes. Chromosoma
107:6169 (1998).
Lee JT, Davidow LS, Warshawsky D: Tsix, a gene antisense to Xist at the X-inactivation center. Nature
Genet 21:400404 (1999).
Li B, Zhang W, Chan G, Jancso-Radek A, Liu S, Weiss
MA: Human sex reversal due to impaired nuclear
localization of SRY: A clinical correlation. J biol
Chem 276:4648046484 (2001).
Lifschytz E, Lindsley DL: The role of X-chromosome
inactivation during spermatogenesis (Drosophilaallocycly-chromosome evolution-male sterilitydosage compensation). Proc natl Aca Sci, USA
69:182186 (1972).
McCarrey JR, Dilworth DD: Expression of Xist in
mouse germ cells correlates with X-chromosome
inactivation. Nature Genet 2:200203 (1992).
McKay LM, Watson JM, Graves JA: Mapping human
X-linked genes in the phalangerid marsupial Trichosurus vulpecula. Genomics 14:302308 (1992).
Migeon BR: X-chromosome inactivation: molecular
mechanisms and genetic consequences. Trends
Genet 10:230235 (1994).
Mohandas TK, Speed RM, Passage MB, Yen PH,
Chandley AC, Shapiro LJ: Role of the pseudoautosomal region in sex-chromosome pairing during
male meiosis: meiotic studies in a man with a deletion of distal Xp. Am J hum Genet 51:526533
(1992).
Muller HJ: The relation of recombination to mutational advance. Mutat Res 1:29 (1964).
Murphy EC, Zhurkin VB, Louis JM, Cornilescu G,
Clore GM: Structural basis for SRY-dependent 46X,Y sex reversal: modulation of DNA bending by a
naturally occurring point mutation. J molec Biol
312:481499 (2001).
Ohno S: Sex Chromosomes and Sex Linked Genes.
(Springer, New York 1967).

Ortenberg J, Oddoux C, Craver R, McElreavey K, Salas-Cortes L, Guillen-Navarro E, Ostrer H, Sarafoglou K: SRY gene expression in the ovotestes of
XX true hermaphrodites. J Urol 167:18281831
(2002).
Ottolenghi C, Fellous M, Barbieri M, McElreavey K:
Novel paralogy relations among human chromosomes support a link between the phylogeny of doublesex-related genes and the evolution of sex determination. Genomics 79:333343 (2002).
Page DC, Mosher R, Simpson EM, Fisher EM, Mardon
G, Pollack J, McGillivray B, de la Chapelle A,
Brown LG: The sex-determining region of the human Y chromosome encodes a finger protein. Cell
51:10911104 (1987).
Page J, Berrios S, Rufas JS, Parra MT, Suja JA, Heyting
C, Fernandez-Donoso R: The Pairing of X and Y
chromosomes during meiotic prophase in the marsupial species Thylamys elegans is maintained by a
dense plate developed from their axial elements. J
Cell Sci 116:551560 (2003).
Pailhoux E, Parma P, Sundstrom J, Vigier B, Servel N,
Kuopio T, Locatelli A, Pelliniemi LJ, Cotinot C:
Time course of female-to-male sex reversal in
38,XX fetal and postnatal pigs. Dev Dynam
222:328340 (2001).
Palmer MS, Berta P, Sinclair AH, Pym B, Goodfellow
PN: Comparison of human ZFY and ZFX transcripts. Proc natl Acad Sci, USA 87:16811685
(1990).
Petit C, Levilliers J, Weissenbach J: Physical mapping
of the human pseudo-autosomal region; comparison with genetic linkage map. EMBO J 7:2369
2376 (1988).
Penny GD, Kay GF, Sheardown SA, Rastan S, Brockdorff N: Requirement for Xist in X chromosome
inactivation. Nature 379:131137 (1996).
Quilter C, Sargent C, Archibald A, Affara A, Griffin
DK: Mapping of Y specific and X/Y homologous
genes in the pig: Implications for sex chromosome
evolution. Mammal Genome 13:588594 (2002).
Rappold GA: The pseudoautosomal regions of the human sex chromosomes. Hum Genet 92:315324
(1993).
Rappold GA, Lehrach H: A long range restriction map
of the pseudoautosomal region by partial digest
PFGE analysis from the telomere. Nucl Acids Res
16:53615377 (1988).
Rastan S: X chromosome inactivation and the Xist
gene. Curr Opin Genet Dev 4:292297 (1994).
Rice WR: Degeneration of a nonrecombining chromosome. Science 263:230232 (1994).
Richler C, Soreq H, Wahrman J: X inactivation in
mammalian testis is correlated with inactive Xspecific transcription. Nature Genet 2:192195
(1992).
Salido EC, Yen PH, Mohandas TK, Shapiro LJ: Expression of the X-inactivation-associated gene
XIST during spermatogenesis. Nature Genet
2:196199 (1992).
Salido EC, Li XM, Yen PH, Martin N, Mohandas TK,
Shapiro LJ: Cloning and expression of the mouse
pseudoautosomal steroid sulphatase gene (Sts). Nature Genet 13:8386 (1996).
Schneider-Gadicke A, Beer-Romero P, Brown LG,
Nussbaum R, Page DC: ZFX has a gene structure
similar to ZFY, the putative human sex determinant, and escapes X inactivation. Cell 57:1247
1258 (1989).
Schmid M, Nanda I, Steinlein C, Epplen JT: Evolution
of the sex chromosome in amphibia. Chrom Today
10:199222 (1989).
Sharp P: Sex chromosome pairing during male meiosis
in marsupials. Chromosoma 86:2747 (1982).
Simmler MC, Rouyer F, Vergnaud G, Nystrom-Lahti
M, Ngo KY, de la Chapelle A, Weissenbach J:
Pseudoautosomal DNA sequences in the pairing
region of the human sex chromosomes. Nature
317:692697 (1985).

Cytogenet Genome Res 99:125140 (2002)

139

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Freije D, Helms C, Watson MS, Donis-Keller H: Identification of a second pseudoautosomal region near
the Xq and Yq telomeres. Science 258:17841787
(1992).
Gabriel-Robez O, Rumpler Y, Ratomponirina C, Petit
C, Levilliers J, Croquette MF, Couturier J: Deletion of the pseudoautosomal region and lack of sexchromosome pairing at pachytene in two infertile
men carrying an X;Y translocation. Cytogenet Cell
Genet 54:3842 (1990).
Gartler SM, Dyer KA, Goldman MA: Mammalian X
chromosome inactivation. Mol Genet Med 2:121
160 (1992).
Gerhart J, Kirschner M: Cells, Embryos and Evolution:
Toward a Cellular and Developmental Understanding of Phenotypic Variation and Evolutionary Adaptability (Blackwell Science, Malden 1997).
Glas R, Marshall Graves JA, Toder R, Ferguson-Smith
M, OBrien PC: Cross-species chromosome painting between human and marsupial directly demonstrates the ancient region of the mammalian X.
Mammal Genome 10:11151116 (1999).
Graves JA: The origin and function of the mammalian
Y chromosome and Y-borne genes an evolving
understanding. Bioessays 17:311320 (1995).
Graves JA: Evolution of the mammalian Y chromosome and sex-determining genes. J Exp Zool
281:472481 (1998).
Graves JA: Human Y chromosome, sex determination,
and spermatogenesis a feminist view. Article.
Biol Reprod 63:667676 (2000).
Graves JA, Disteche CM, Toder R: Gene dosage in the
evolution and function of mammalian sex chromosomes. Cytogenet Cell Genet 80:94103 (1998a).
Graves JA, Wakefield MJ, Toder R: The origin and
evolution of the pseudoautosomal regions of human sex chromosomes. Hum molec Genet 7:1991
1996 (1998b).
Griffin DK, Harvey SC, Campos-Ramos R, Ayling L-J,
Bromage NR, Masabanda JS, Penman D: Early
origins of the X and Y chromosome: lessons from
tilapia. Cytogenet Genome Res 99:157163
(2003).
Grumbach MM, Morishima A, Taylor JH: Human sex
chromosome abnormalities in relation to DNA
replication and heterochromatinization. Proc natl
Acad Sci, USA 49:581589 (1963).
Hammer MF: A recent common ancestry for human Y
chromosomes. Nature 378:376378 (1995).
Hartshorn C. Rice JE. Wangh LJ: Developmentallyregulated changes of Xist RNA levels in simple
preimplantation mouse embryos, as revealed by
quantitative real-time PCR. Mol Reprod Dev
61:425436 (2002).
Hassold TJ, Sherman SL, Pettay D, Page DC, Jacobs
PA: XY chromosome nondisjunction in man is
associated with diminished recombination in the
pseudoautosomal region. Am J hum Genet
49:253260 (1991).
Hope RM, Cooper S, Wainwright B: Globin Macromolecular Sequences in Marsupials and Monotremes,
Mammals from Pouches and Eggs: Genetics,
Breeding and Evolution of Marsupials and Monotremes. (CSIRO Press, Melborne 1990).
Hurst LD, Peck JR: Recent advances in understanding
the evolution and maintenance of sex. Trends Ecol
Evol 11:4652 (1996).
Incerti B, Guioli S, Pragliola A, Zanaria E, Borsani G,
Tonlorenzi R, Bardoni B, Franco B, Wheeler D,
Ballabio A: Kallmann syndrome gene on the X and
Y chromosomes: implications for evolutionary divergence of human sex chromosomes. Nature Genet 2:311314 (1992).
Jegalian K, Lahn BT: Why the Y. Sci Am 284(2):4247
(2001).
Jones S: The Language of the Genes: Biology, History
and Evolutionary Future (HarperCollins, London
1994).

140

Toder R, Wakefield MJ, Graves JA: The minimal

mammalian Y chromosome the marsupial Y as a
model system Cytogenet Cell Genet 91:285292
(2000).
Torres-Maldonado L, Moreno-Mendoza N, Landa A,
Merchant-Larios H: Timing of SOX9 downregulation and female sex determination in gonads of the
sea turtle Lepidochelys olivacea. J Exp Zool 290:
498503 (2001).
Watson JM, Spencer JA, Riggs AD, Graves JA: The X
chromosome of monotremes shares a highly conserved region with the eutherian and marsupial X
chromosomes despite the absence of X chromosome inactivation. Proc natl Aca Sci, USA 87:
71257129 (1990).
Watson JM, Spencer JA, Riggs AD, Graves JA: Sex
chromosome evolution: platypus gene mapping
suggests that part of the human X chromosome was
originally autosomal. Proc natl Aca Sci, USA
88:1125611260 (1991).
Western PS, Harry JL, Graves JA, Sinclair AH: Temperature-dependent sex determination: upregulation of SOX9 expression after commitment to male
development. Dev Dynam 214:171177 (1999a).
Western PS, Harry JL, Graves JA, Sinclair AH: Temperature-dependent sex determination in the
American alligator: AMH precedes SOX9 expression. Dev Dynam 216:411419 (1999b).
Whitfield LS, Sulston JE, Goodfellow PN: Sequence
variation of the human Y chromosome. Nature
378:379380 (1995).

Cytogenet Genome Res 99:125140 (2002)

Willard HF, Carrel L: Making sense (and antisense) of

the X inactivation center. Proc natl Aca Sci, USA
98:1002510007 (2001).
Winter PC, Hickey GI, Fletcher HL: Instat Notes in
Genetics. (Bios Scientific Publishers, Oxford
1998).
Wrigley JM, Graves JA: Sex chromosome homology
and incomplete, tissue-specific X-inactivation suggest that monotremes represent an intermediate
stage of mammalian sex chromosome evolution. J
Hered 79:115118 (1988).
Yen PH, Allen E, Marsh B, Mohandas T, Wang N, Taggart RT, Shapiro LJ: Cloning and expression of steroid sulfatase cDNA and the frequent occurrence
of deletions in STS deficiency: implications for XY interchange. Cell 49:443454 (1987).
Yen PH, Marsh B, Allen E, Tsai SP, Ellison J, Connolly
L, Neiswanger K, Shapiro LJ: The human X-linked
steroid sulfatase gene and a Y-encoded pseudogene: evidence for an inversion of the Y chromosome during primate evolution. Cell 55:11231135
(1988).
Yuan X, Lu ML, Li T, Balk SP: SRY interacts with and
negatively regulates androgen receptor transcriptional activity. J biol Chem 276:4664746654
(2001).
Zuccotti M, Monk M: Methylation of the mouse Xist
gene in sperm and eggs correlates with imprinted
Xist expression and paternal X-inactivation. Nature Genet 9:316320 (1995).

Downloaded by:
Univ. of Michigan, Taubman Med.Lib.
141.213.236.110 - 8/1/2013 10:31:36 AM

Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ, Foster JW, Frischauf AM,
Lovell-Badge R, Goodfellow PN: A gene from the
human sex-determining region encodes a protein
with homology to a conserved DNA-binding motif.
Nature 346:240244 (1990).
Singh L, Jones KW: Sex reversal in the mouse (Mus
musculus) is caused by a recurrent nonreciprocal
crossover involving the X and an aberrant Y chromosome. Cell 28:205216 (1982).
Sola L, Cataudella S, Capanna E: New developments in
vertebrate cytotaxonomy. III. Karyology of bony
fishes: a review. Genetica 54:285328 (1981).
Spencer JA, Sinclair AH, Watson JM, Graves JA:
Genes on the short arm of the human X chromosome are not shared with the marsupial X. Genomics 11:339345 (1991a).
Spencer JA, Watson JM, Graves JA: The X chromosome of marsupials shares a highly conserved
region with eutherians. Genomics 9:598604
(1991b).
Switonski M, Jackowiak H, Godynicki S, Klukowska J,
Borsiak K, Urbaniak K: Familial occurrence of pig
intersexes (38,XX; SRY-negative) on a commercial fattening farm. Anim Reprod Sci 69:117124
(2002).
Takagi N, Sasaki M: Preferential inactivation of the
paternally derived X chromosome in the extraembryonic membranes of the mouse. Nature 256:
640642 (1975).