X Inactivation Seen As Contact Sport

One Chromosome Wraps Up Only After Opposing Xs Touch

At an early stage in a female embryo’s development, the two X chromosomes in each cell face a decision. The cell isn’t big enough for both of them; two copies of one chromosome could wreak havoc on the health of the nascent organism. In a process that is not fully understood, one of the chromosomes bows out. A noncoding RNA called Xist suddenly accumulates on this chromosome, sheathing it in an RNA coat and rendering it inactive. This decision happens simultaneously in every cell, but whether it is the father’s or mother’s chromosome that defers is entirely random in the body. After the process ends, each cell has a single active X, a mosaic of the Xs of both parents.

Photo by Graham Ramsay

Na Xu, Bryan Sun, and Jeannie Lee uncovered new steps that take place in female cells to ensure that only one X chromosome prevails.

In two recent papers, the lab of Jeannie Lee, HMS professor of genetics (pathology) at Massachusetts General Hospital, makes important breakthroughs in uncovering how the X chromosomes decide their fate. The studies address two steps in the inactivation process: the standoff between X chromosomes in which one prevails and the initial steps of inactivating the extra X.

Molecular Pillow Talk
The genes responsible for X inactivation work only on the chromosome they come from, but somehow the two chromosomes must coordinate. In a paper in the Jan. 20 Science, a team led by graduate student Na Xu showed that just before the onset of X inactivation, the two X chromosomes literally get together. The team used fluorescence in situ hybridization (FISH) to monitor the position of X chromosomes over time in mouse embryonic stem cells that were in the process of differentiating. They found that around days two through four, the X chromosomes moved closer together, but before and after this period they were more randomly spaced. This timing coincided with X inactivation, and the X inactivation center was the portion of the chromosome where the contact occurred.

Using chromosome conformation capture, a method for isolating DNA-bound proteins that are linked together, the team also verified that the DNA seems to be in close contact. “Not only are they close together under the microscope, but the molecules on the two chromosomes are really touching each other,” Lee said. When the X inactivation center was removed from chromosomes, they failed to pair up. But even if the region was inserted into a non-sex chromosome, it still acted as a point of contact with the X chromosome.

Dressed Up and No Place to Go
Somehow, this contact between the Xs results in one chromosome initiating its own inactivation. Xist RNA is almost absent before X inactivation, after which levels of the RNA soar on the inactive chromosome, but not on the active one. How this is carried out has been unclear.

Xist has a doppelganger; its antisense partner Tsix was identified by Lee’s lab several years ago. Tsix counteracts Xist and protects the future active X from being coated by the RNA. But the mechanism by which Tsix represses Xist has been debated. Several years ago, studies suggested that Xist RNA is regulated by a mechanism that changes its stability. The discovery of Tsix raised the question of whether the destabilizing factor is the paired antisense RNA. But in a paper in the March 3 Molecular Cell, Bryan Sun, a graduate student in Lee’s lab, provides evidence against this model.

Tsix, the team finds, regulates the transcription of its antisense partner. It does this primarily by changing the way that histones and DNA are packed into chromatin. The chromatin changes that they discovered, however, seem somewhat paradoxical. Areas that are less condensed, called euchromatin, usually have genes that are accessible for transcription; areas that are more tightly coiled, called heterochromatin, are usually silenced. But because Tsix opposes Xist, when the chromatin is opened up for Tsix, Xist becomes silent. Conversely, when the chromatin becomes more compacted when Tsix turns off, Xist suddenly becomes active. “The state of the chromatin reflects the state of Tsix,” not Xist, Lee said. “When Tsix is on, the transcription of Tsix converts the chromatin to euchromatin, to an activating form. But this activating form represses Xist.”

The X inactivation center on both chromosomes is in the open state before X inactivation, and the future active X remains that way. But somehow, Tsix levels plummet on the future inactive X, which causes the region to coil up into heterochromatin. In this case, the more tightly packed state enables Xist to switch on and form its coat. Lee notes that such a phenomenon is not unprecedented; a handful of genes in fruit flies are known to be expressed specifically
in heterochromatin.

Puzzling Answers
For reasons completely unknown, this paradoxical state of chromatin occurs only at the onset of X inactivation; once the process is over, the region returns to standard chromatin patterns. In addition to its effect on the chromatin, Tsix was also shown to interact with a molecule that attaches methyl groups to the Xist promoter, silencing it. Lee believes this is a secondary mechanism to make sure Xist stays off.

“This is the first time anybody’s shown a mechanism that actually precedes the expression of Xist and may therefore cause asymmetric Xist expression patterns,” said Lee. But this study also presents a new puzzle—how does Xist get turned on in what is normally a silencing state?

Laura Carrel, assistant professor of biochemistry and molecular biology at Penn State University, who wrote a review of the Science paper, said that both studies add key information, but open up entirely new questions about the process of X inactivation. In the Science paper, she said, the question is, “What is cross-talk?” If the chromosomes confer about their fate, what is the nature of the conversation? In the Molecular Cell paper, Lee’s team addresses some old questions, but with better tools, like the ability to parse out the role of Tsix versus Xist. It adds a new step to X inactivation, Carrel said, and also shows the process “is much more complicated than I would have guessed.”

—Courtney Humphries

top