Re: What would happen if neither X-chromosome became Barr bodies?

Dear Chris,

The reason that you have not heard of cases of both X chromosomes being active in a 
female mammal is that it is incompatible with viability. We know this through a set of 
interesting observations and genetic experiments.

First, consider the general problem of alterations in gene dosage. The only full trisomy that 
is compatible with viability in humans is trisomy 21 (Down syndrome). The phenotypes 
caused by this trisomy are not the result of increased dosage of a single gene, but result 
from the additive effects of alterations in gene dosage of many genes. Chromosome 21 is 
quite a bit smaller than the X chromosome and has fewer genes.

We can look at the status of the human genome project to see the number of predicted 
genes on both chromosomes. Go to:
 http://
www.ncbi.nlm.nih.gov/genome/guide/human/

Click on chromosome 21 to see that there are 285 genes annotated to chromosome 21; 
click on the X to see that there are 1095 genes annotated to the X chromosome. So there 
are more than three times the number of genes on the X chromosome as on chromosome 
21.

The dosage difference in trisomy 21 is a 50% increase in the dosage of 285 genes. The 
dosage difference from males to females is a 100% increase in the dosage of 1095 genes. 
From another perspective, males suffer a monosomy of the X chromosome; there are no 
human monosomies that are compatible with viability.

All this is a way of showing you that some form of dosage compensation system is essential 
for any organism that has a chromosomal sex determination system. Mammals use X 
inactivation in females; all X chromosomes except one are inactivated, so normal XX females 
have one inactive X chromosome in each cell. Abnormal XXX females have two inactive X 
chromosomes in each cell.

X inactivation is initiated in an "X inactivation center," a specific region of the mammalian X 
chromosome. Once inactivation starts at this site, it spreads to the rest of the chromosome. 
Interestingly, the inactivation process does not "know" that X chromosome genes are being 
inactivated. We know this because there is a class of chromosome rearrangements called 
translocations, in which two different chromosomes are broken and rejoined to make a pair 
of new chromosomes. If one of these chromosomes is an X chromosome, one new 
chromosome will have the X inactivation center and part of another chromosome, while the 
other chromosome will have that part of the X chromosome lacking the X inactivation center 
and part of the other chromosome. The first new chromosome of this translocated pair will 
inactivate its X chromosome genes but also the autosomal genes on the same chromosome. 
The second new chromosome will fail to inactivate some X chromosome genes. Depending 
on the position of the chromosome breaks, the genic imbalance and the effects on viability 
will be more or less severe.

In Drosophila (fruit flies) a different system of dosage compensation is used. Both X 
chromosomes are active in females; the X chromosome in males is twice as active as either 
X chromosome in females. There are sex-specific lethal mutations in Drosophila that kill 
males only. These lethal mutations affect genes required for the hypertranscription of the X 
chromosome in males. The male X chromosome is therefore transcribed at the level of a 
single female X chromosome, a level incompatible with adult viability (although dying larvae 
can be studied).

We would expect that there could be sex-specific lethals in mammals. Mutations interfering 
with X inactivation should be lethal to females. I do not know of such mutations, but our 
ability to search for mutations of this kind in mammals is limited.

Finally, there are some cells in females that have both X chromosomes active. Pre-oogonial 
cells reactivate the X chromosome, so that both X chromosomes are active during 
oogenesis. I can speculate that this does not create a genic imbalance in these cells because 
1) not all the genes on the X chromosome are active in these cells (or any cell type, for that 
matter) , and 2) the X chromosome genes that are active have their levels of expression 
adjusted to produce the correct level of gene product by cell-type-specific transcription 
factors.

You also speculate that activating the X would cause feminization, since XXY males are 
feminized. I can see why you might think this, but the underlying biology of mammalian 
sexual differentiation suggests that this would not be the case.

Briefly, the "ground state" or "default" setting for mammalian sexual differentiation is 
female. Early mammalian embryos are identical in both sexes. There is a presumptive gonad 
and rudiments of both male- and female-specific structures. If an embryo contains a Y 
chromosome (or more accurately, a specific gene on the Y chromosome) in the cells of the 
presumptive gonad, androgen is made during development, the female structures do not 
develop and the male ones do. If there is no androgen (or if the cells cannot respond to 
androgen because of a mutation in the androgen receptor), the male structures do not 
develop and the female ones do.

Mutations that completely eliminate the function of the androgen receptor result in the most 
feminized mammals possible. For a description of the human syndrome, please see:
 http://
www.ncbi.nlm.nih.gov/htbin-post/Omim/dispmim?300068

The androgen receptor gene is X-linked, so these individuals are identified as chromosomal 
males at puberty, because they do not menstruate and are sterile (germ cell sex 
determination is not affected). Since they have only one X chromosome, there is no Barr 
body, yet these individuals are extremely feminized:

"The phenotype is often voluptuously feminine: Netter et al. (1958) reported this disorder in 
a famous photographic model, Marshall and Harder (1958) reported affected monozygotic 
twins who worked as airline stewardesses, and Polaillon (1891) described prostitution in an 
affected person."

This argument is meant to lead you to the conclusion that we should view the phenotype of 
XXY males as incompletely masculinized rather than as feminized. You may also find it 
useful to consider that a Barr body, being an inactive X chromosome, cannot have a genetic 
effect in principle because it is genetically inactive. Actually, a few genes on the X 
chromosome escape inactivation, but it is not clear to me that these have anything to do 
with the phenotype of XXY males.

There is a great deal of literature on your question. Try a PubMed search on "X inactivation," 
using "Limits" to restrict your results to "Reviews." I would recommend:

Trends Genet. 2003 May;19(5):243-7 

Annual Rev Genet. 2002;36:233-78. 

Curr Opin Genet Dev. 2002 Apr;12(2):219-24. 

Nat Rev Genet. 2001 Jan;2(1):59-67. 

Please also see the MGI Glossary:
 http://
www.informatics.jax.org/mgihome/other/glossary.shtml

Thank you for an interesting question.

Yours,

Paul Szauter
Mouse Genome Informatics