The Y chromosome is of great interest to students and can be used to teach about many important biological concepts in addition to sex determination. This paper discusses mutation, recombination, mammalian sex determination, sex determination in general, and the evolution of sex determination in mammals. It includes a student activity that illustrates how sex is determined in people.
It may seem strange to begin a paper about the Y chromosome with a discussion of recombination and mutation. But bear with me. Your patience will be amply rewarded.
For many years, scientists thought of recombination as a useful tool for constructing chromosome maps. Indeed, until DNA sequencing became widely available in the 1980s and 1990s, all chromosome maps were constructed using recombination frequencies; the more closely two genes are located on a chromosome, the less frequently they recombine. Yet recombination had been going on for several billion years before people started constructing chromosome maps. Why has this complicated and costly process survived billions of years of natural selection? The answer lies in the properties of mutation.
Mutation is a change in DNA. There are dramatic mutations, such as single base-pair changes that cause sickle cell anemia or severe combined immunodeficiency (‘‘Boy in the Bubble’’ disease). However, most mutations are not nearly so devastating. Most mutations make no difference at all. They are either located in non-coding areas of the genome or are ‘‘silent’’ mutations within coding areas of genes. The next-most-common kind of mutation causes a small but slightly detrimental change to a protein. An organism with this kind of mutation will live but will be at a very slight selective disadvantage. Rarer still are the mutations that confer an adaptive advantage and are selected for.
What happens to the many slightly detrimental mutations that occur? Because of recombination, these mutations are weeded out over evolutionary time, hundreds or thousands of generations, because recombination events yield chromosomes that do not contain the mutated genes (Figure 1).
A gene that does not recombine will accumulate random mutations and eventually not code for a functional protein. Decay in genes that either don't recombine or no longer serve a function has been observed frequently. Most mammals have about 1000 genes that code for odor receptors, G proteins in cell membranes of cells in the nose. Each odor receptor has a unique shape and binds to a specific odorant. However, primates have only about 300 functional odor receptors. Additional odor receptors exist as ‘‘pseudogenes’’ in primate genomes. These stretches of DNA have recognizable ‘‘odor receptor‘‘ sequence but are clearly nonfunctional; there may be a stop codon in the middle of the gene. This loss of functional odor receptors occurred when primates developed excellent color vision. Because they relied on vision rather than smell to find food, there was no selection against individuals with fewer functional odor receptors.
Similarly, there are 15 known species of icefish. Adapted to living in frigid Antarctic oceans, these fish have no red blood cells. They have large gills and scaleless, highly vascularized skin that enables them to obtain and transport adequate oxygen directly from the cold, oxygen-rich waters. Icefish have no gene for alpha hemoglobin and only a partial, nonfunctional gene for beta hemoglobin (one hemoglobin molecule consists of two alpha chains and two beta chains). The ancestors of icefish had functional hemoglobin genes. These genes no longer function in icefish, so mutations in them are not selected against (Carroll, 2006).
Evolutionary History of the Y Chromosome
About 310 million years ago, there was no Y chromosome as we know it. At that time, when mammals were evolving from their common ancestors with reptiles and birds, what are now the X and Y chromosomes were a pair of autosomes similar to today's X chromosome. Sex in these ancestors was probably determined by the temperature at which the egg is incubated, as it is in many reptiles today. This can occur only in ectotherms, animals that do not maintain a constant body temperature. Otherwise, only males or only females would be produced, not a situation conducive to long-term reproductive success.
Sometime between 165 and 310 million years ago, SRY (sex-determining region Y) evolved in the ancestral lineage of mammals. SRY is the gene on the mammalian Y chromosome that determines that the organism will be male. Once SRY evolved, endothermy could also evolve since both males and females could develop at the same temperature. The evolution of chromosomally based ZW sex determination in birds enabled them to independently evolve endothermy and its associated characteristics - insulated skin covering made of keratin (feathers or fur) and a four-chambered heart.
Further evidence that X and Y were once a pair of autosomes comes from DNA sequences of the human X and chicken autosomes (Ross et al., 2005). Figure 2 shows a dotter plot comparing the human X chromosome with parts of the chicken ##1 and ##4 chromosomes. The human Xp (the shorter arm) is nearly identical to part of the chicken ##1 chromosome, whereas the human Xq (the longer arm) has large chunks of similarity with parts of the chicken ##4 chromosome. The autosome that became X and Y in mammals lives on as autosomes in chickens.
After the evolution of SRY, X and Y continued to be very similar until a large inversion occurred on the Y chromosome. At this point, there could be no recombination between X and Y along the entire length of this inversion. Additional alterations to the Y chromosome further inhibited its ability to recombine with the X. Genes on this stretch of DNA could be maintained on the X chromosome, because X chromosomes recombine when they are in females. But because these genes on the Y could never recombine, over 1000 of them have been lost. These genes, and their associated ‘‘sex-linked traits,’’ such as red and green cone proteins (color-blindness) and Factors VIII and IX (hemophilia), are found on the X chromosome but not on the Y and were presumably on the original pair of autosomes. Small pseudoautosomal regions exist at both ends of the X and Y chromosomes in humans and at one end in other mammals. Here, X and Y remain homologous, recombine, and pair during meiosis. You might ask why a non-recombining Y chromosome was selected for. It is likely that not crossing over with the X is an advantage, so SRY does not regularly get transferred to the X chromosome. Since there are an average of two to three crossovers per chromosome during meiosis, this is a very real concern.
There are 76 protein-coding genes on the human Y; they code for 26 distinct proteins, because several genes code for the same protein. These genes are almost exclusively concerned with male fertility: men with deletions in them have no associated health problems except infertility. These genes have probably migrated to the Y chromosome from various autosomes in the last few hundred million years. Since these genes do not recombine, how are they preserved over evolutionary time?
Sequencing & Structure of the Y Chromosome
The Y is one of the smallest chromosomes and should have been one of the first to be sequenced. Yet it was one of the last to be sequenced, because its structure posed enormous technical difficulties (Skaletsky et al., 2003). The Y chromosome contains eight palindromes, regions that are an astonishing 99.94%% to 99.997%% identical at the base-pair level. These palindromes contain between 1.45 million and 9000 base pairs in each arm and a small ‘‘spacer’’ region between the two arms. Imagine sequencing something like this –– how do you know which arm of the palindrome your sequence is from? How do you know if one base-pair difference in 1000 between two arms of a palindrome is real, an error in sequencing, or a mutational difference between two people? To deal with this challenge, the Y chromosome from one man was sequenced.
The resulting sequence showed that about 25%% of the male-specific euchromatin on the Y consists of these eight palindromes and contains most of the Y's functional genes (Figure 3).
The genes in these palindromes occur in pairs, one copy on each arm of the palindrome, and can therefore be maintained over evolutionary time. Gene conversion occurs between the arms of each palindrome, so mutations in these genes can be corrected (Figure 4A).
Palindromes are not rare in genomes. A palindrome occurs if part of a chromosome is duplicated and then inverted. In the HCG//LH beta chain region of human chromosome 19, there is one copy of LH beta chain and 7 copies of HCG beta chain, some of which are ‘‘head to tail’’ to each other. This palindrome has occurred recently, for HCG is a relatively new gene. A palindrome on an autosome will be selected against because it will lead to unequal recombination and chromosomal instability. Conversely, a palindrome on the Y chromosome will be selected for because it provides a duplicate copy of the genes it contains, enabling them to be maintained.
We are just beginning to understand the evolutionary costs of these palindromes. These costs probably explain why there are very few palindromes on autosomes where they are not necessary to preserve genes. The same mechanism used for gene conversion on the Y can result in nonreciprocal recombination between two arms of the palindromes (Lange et al., 2009). Such events yield isodicentric Y chromosomes that contain two centromeres, two copies of some genes, and no copies of other genes (Figure 4B).
Many people with these isodicentric Y chromosomes are healthy but infertile; in some cases, they produce enough sperm that they can father children by single-sperm injection techniques. But this problem may be more widespread than anyone could have imagined. Chromosomes with two centromeres are mitotically unstable. They attach to two spindle fibers during mitosis and are often discarded by the cell. Therefore, an XY person with an isodicentric Y chromosome may have many cells that are XO. It is possible that many women with Turner's syndrome are actually XY, with an isodicentric Y that was discarded from their presumptive gonads early in development. Seventy-five percent of women with Turner's syndrome have their mother's X, which means that they have no sex chromosome from their father. Unlike Down syndrome, which is caused by true nondisjunction and shows an increasing frequency with maternal age, Turner's syndrome is not correlated with maternal age. Turner's syndrome occurs in approximately 1 in 2500 live births. The potential extent of this problem can be glimpsed when we realize that only 1%% of ‘‘XO’’ embryos are born; the rest are miscarried early in pregnancy. If any significant fraction of these embryos is XY with an isodicentric Y from their father, the evolutionary price to pay for having chromosomally based sex determination is very high indeed. These are new findings (Lange et al., 2009). Research is being done to see whether significant numbers of women with Turner's syndrome have isodicentric Y chromosomes in some of their tissues.
Also unknown is when the mammalian X and Y chromosomes evolved. This had to occur after the time of the common ancestor of mammals and birds/reptiles (310 million years ago), and before the time of the common ancestor of placental mammals and marsupials (eutherians; 145 million years ago), because both placental mammals and marsupials have XY sex determination with SRY determining maleness. Things get interesting, however, when you look at the platypus, whose genome was sequenced in 2008 (Warren et al., 2008). The platypus has XY sex determination but does not have SRY. The platypus X chromosome has no homology to the human X. It has some homology to the bird Z, although this homology does not seem to be involved in sex determination. This leaves open the question whether the eutherian XY evolved 310 million years ago, shortly after the common ancestor of mammals and birds/reptiles, or less than 166 million years ago, at the time of the common ancestor of monotremes and eutherians (Figure 5). See Appendix 2 for details.
Will the Y Chromosome Decay?
Many people delight in predicting that the Y chromosome, already small, will continue to decay until it no longer exists. This is unlikely. First, SRY is required for male sex determination. Any Y chromosome with a mutation in SRY is very strongly selected against ( Appendix 1). Second, because most of the functional genes on the Y occur in palindromes, they will be maintained by the gene conversion methods that maintain them now. Just calculating how much the Y chromosome has shrunk over the past 165––310 million years and projecting further decay at this rate ignores this underlying biology.
SRY & Sex Determination
In mammals, the presence of the Y chromosome determines maleness. People who are XXY, XXXY, XXXXY, and XXXXXY are known and are phenotypically male, although not always healthy or fertile. The clue to how the Y chromosome determines maleness occurred with the discovery of a very few people who are XY but phenotypically female.
Anatomically, the gonads of a 6-week embryo, called ‘‘indifferent gonads,’’ look the same whether they are ovaries or testes. They become differentiated around the 10th or 11th week of pregnancy. The SRY gene codes for a protein produced during those weeks that causes the indifferent gonads to become testes; the embryo develops then as a male. If the SRY protein is not present, the indifferent gonads become ovaries and the embryo develops as a female.
About 20 people are known to have a mutation in the SRY gene. They are XY but are phenotypically female. They do not go through a normal puberty and have nonfunctional gonads known as streak ovaries. Because these streak ovaries become cancerous about a third of the time, they are surgically removed. These women go through a social puberty with appropriate hormone treatments. They have a normal uterus and can carry a pregnancy if an egg donor is available and if they receive hormone treatments during the pregnancy. Lacking functional ovaries, they cannot produce their own eggs. One of these XY females has a point mutation in the SRY gene that changes a single amino acid in the SRY protein, making it inactive. It is remarkable that changing one base pair, out of 6 billion base pairs in a diploid human genome, can cause as dramatic a change as male-to-female sex reversal ( Appendix 1).
An XY female is phenotypically more normal that a female with Turner's syndrome (XO). Ribosomes are made up of RNA and protein. One ribosome contains approximately 70 different proteins, each coded for by a gene. These genes are all over the human chromosome map; one of them is on the Y chromosome. It has a homologue on the X that is not inactivated, so cells of normal males and females have two functioning copies of this gene, allowing them to make enough ribosomes. Because an XY female has two copies of this ribosomal protein gene in each cell and a female with Turner's syndrome (XO) has only one, the former makes enough ribosomes whereas the latter does not. The phenotypic abnormalities in females with Turner's syndrome may be attributable to this reduced number of ribosomes.
I thank the Whitehead Institute's Partnership for Science Education and the Seminar Series for High School Teachers; much of this material came from lectures in this program. Also, a special thank-you to Whitehead Director David Page for his time and input during many helpful conversations. The Whitehead Institute for Biomedical Research is a leading nonprofit research and educational institution located in Cambridge, MA. Finally, I thank Nadav Kupiek for expert preparation of the artwork.
The following is a sequence of nucleotides in a DNA molecule:
1. Find the sequence of nucleotides on the complementary DNA strand.
2. What mRNA will the original DNA code for?
3. What protein will this mRNA code for?
4. If you change the 10th nucleotide in the original DNA from G to T, what mRNA will the DNA now code for?
5. What protein will this mRNA code for?
6. Compare the proteins in questions 3 and 5.
3. leucine glutamic acid asparagine proline arginine methionine arginine
5. leucine glutamic acid asparagine threonine arginine methionine arginine
6. In the original protein (##3), the fourth amino acid is proline. In the second protein (##5), this proline has been changed to threonine. This is the result of one base change in the original DNA. Because proline has a circular R group, it cannot fit into an alpha helix, so a mutation that changes proline to another amino acid is likely to have a significant effect on the tertiary structure, and therefore the function, of the protein.
The original DNA is on the human Y chromosome and codes for amino acids ##80––86 of SRY, a protein that is 204 amino acids long. The point mutation in question ##4 results in an inactive SRY protein. This mutation was found in an XY female and is the mutation that identified SRY as the gene determining maleness.
Appendix 2: Sex Determination in the Platypus
The platypus is a monotreme and most recently had a common ancestor with eutherians (placental mammals and marsupials) approximately 166 million years ago (Binida-Emonds et al., 2007) (Figure 5).
Chromosomally based sex determination is considered either XY or ZW, depending on which sex has two large chromosomes. In XY, the female has two large chromosomes; in ZW, the male has two large chromosomes. Both XY and ZW sex determination have evolved independently many times. Therefore, each kind of XY or ZW system uses a different mechanism for determining sex. Eutherian mammals have an XY system where SRY on the Y determines maleness, so XXY is male and XO is female. Fruit flies have an XY system that relies on the ratio of X chromosomes to autosomes to determine sex, so XXY is female and XO is male.
The platypus has five X chromosomes, labeled X1, X2, X3, X4, and X5. Each of these X chromosomes contains different genes and shows homology to different chicken (and human) chromosomes, which implies that they, too, evolved from autosomes. The platypus also has five Y chromosomes, labeled Y1, Y2, Y3, Y4, and Y5. Female platypuses have two copies of each of the five X chromosomes. Males have one copy of each of the five X chromosomes and one copy of each of the five Y chromosomes.
During meiosis in females, each of the five X chromosomes behave as an homologous pair, and each gamete gets one copy of each of the five X chromosomes. During meiosis in males, the X and Y chromosomes are arranged in a line, with X and Y chromosomes alternating. All five X chromosomes then go to one gamete, whereas all five Y chromosomes go to the other.
The platypus X and Y are very interesting. None of the platypus Y chromosomes contains an SRY gene. This means that the platypus XY evolved independently of the eutherian XY and determines sex in a different, still unknown, way. Even more curious, the X5 platypus chromosome shows considerable homology to the Z chromosome of the bird ZW system, and other platypus X chromosomes show lesser degrees of homology to the bird Z. But this Z chromosome material does not seem to be involved in sex determination in either the bird or the platypus. And it is unlikely that birds and platypuses would have the same sex-determination mechanism, given that birds have a ZW system and the platypus has an XY system.
This suggests important questions for future research. Did the platypus XY evolve independently of the bird ZW, and is it a coincidence that parts of them came from the same autosome? The platypus XY probably evolved independently of the eutherian XY. When did this happen? Did the common ancestor of all mammals have XY(SRY), with the platypus later developing a different sex determination system (Figure 5A)? Or did both the eutherian and platypus XY systems evolve after the two lineages split 166 million years ago (Figure 5B)? If the latter is true, it implies that for about half the time since the mammalian lineage split from the reptile/bird lineage, the early proto-mammals did not have chromosomally based sex determination and were probably ectotherms.