By Aaron Krol
January 8, 2014 | While targeted gene silencing is a complex, cutting-edge feat in the lab, every female mammal on Earth has been pulling off an even greater stunt since time immemorial: in each of her cells, an entire X chromosome is silenced, preventing a damaging overexpression of the hundreds of X-linked genes that females, unlike males, carry two distinct copies of. The phenomenon, which is mediated by the little-understood XIST RNA-coding gene, made a splash in October at the annual meeting of the American Society of Human Genetics, when Dr. Jeanne Lawrence demonstrated that the same process could be used to silence a single copy of chromosome 21 in vitro, in cellular models of trisomy 21.
Now, a team at Johns Hopkins University Medical School led by Drs. Hao Wu and Jeremy Nathans has devised a way to visualize the patterns of X-inactivation in fully-developed animal models. X-inactivation is not uniform throughout an organism; instead, different cells may silence either the maternal or the paternal X chromosome. Early in embryonic growth, each cell in the developing animal "chooses" which X chromosome it will express, by some unknown and apparently random mechanism. These early founder cells, which differentiate into the various adult tissues, give rise to new generations of cells that all express the same X chromosome as their parents – meaning that early "decisions" about which X chromosome to silence permanently affect different areas of the body. The Johns Hopkins team attached two different fluorescent markers to the X chromosomes of mice to create visual maps of these patterns.
Microscope images of different mouse tissues with fluorescent labels showing which X chromosome is activated in each cell. Clockwise from top left: corneal endothelium, deep epidermis, cartilage, and cochlea. Image credit: Hao Wu, with permission of Neuron
To ensure that the fluorescent markers correctly sorted to the desired X chromosomes, the researchers bred their mice from genetically engineered parents. The mothers all had two X chromosomes carrying a gene for the green fluorescent protein GFP, while the fathers had X chromosomes carrying a gene for the red fluorescent protein tdTomato. Offspring of these parents would always have "green" maternal chromosomes and "red" paternal chromosomes. The researchers inserted these genes for fluorescence next to a variety of promoters that signal RNA polymerase to begin transcription in a particular cell type, so that in each strain of mice, the fluorescent proteins would only be expressed in a specific tissue. This allowed the team to take clear microscope images of X-inactivation in everything from the heart to the intestine to the brain, without image interference from other fluorescent cells.
The resulting pictures clearly show how random X-inactivation patterns early in development bias different tissues and regions toward the expression of one X chromosome or the other. The patterns are very similar to the best-known example of distributed X-inactivation: the coats of calico cats, which carry two different fur color alleles on their X chromosomes, resulting in random color patches where one or the other allele is expressed. However, this is the first time that the same patterns have been made visible in many different tissue types.
Many genetic diseases that occur on the X chromosome can be better understood with a clearer picture of X-inactivation patterns. For instance, in many mammals, females are less likely than males to suffer from color blindness, because the relevant gene resides on the X chromosome, so even a female carrying a mutated allele is likely to have a normal allele as well. However, the Johns Hopkins team's images of mouse retinal tissue shows that an eye may be overwhelmingly biased toward a single X chromosome, illuminating how heterozygous females may still be vulnerable to color blindness. To further explore these associations, the researchers also looked at X-inactivation in mice that are carriers for Norrie disease, an X-linked disorder that causes blindness in both humans and mice. They showed that when mice had one retina that disproportionately expressed the X chromosome carrying the Norrie disease mutation, the blood vessels around that retina would be deformed. Other X-linked, tissue-specific diseases in humans include hemophilia and the most common type of muscular dystrophy.
This research may also provide a new tool for exploring patterns of embryological development. The study appears today in Neuron.
Left and right retinas from the same mouse, showing differential X-inactivation. Image credit: Hao Wu, with permission of Neuron