She Has Her Mother's Laugh: The Powers, Perversions, and Potential of Heredity (46 page)

Turning over facts like these in her mind, Lyon alighted on an idea that could explain them all. She sat down, typed it out in
seven paragraphs, and sent them off to the journal
Nature.

Lyon proposed that as a female embryo develops, its cells shut down one of their two X chromosomes. But each cell chooses at random between the two. After making this choice, the cell divides and causes its daughter cells to shut down the same X chromosome. They in turn pass down that same selection to all their own descendants. A female mouse's body is made up of lineages of cells, half of which have silenced one X chromosome, and half the other.

This kind of inner heredity could explain mottled mice. Lyon speculated that they had one X chromosome with a mutation that disrupted the development of the skin. A female mottled mouse had skin composed of clusters of cells. All the cells in each cluster shut down the same X chromosome. Some clusters produced normal fur as a result, and others produced altered colors.

Nature
published Lyon's short paper in 1961. Other biologists read it and wished they had thought of the idea themselves. Lyon meanwhile went on looking for more evidence. She investigated the fur of cats, finding that tortoiseshells and calicos had coat patterns that fit her model. A number of human diseases seemed to support her hypothesis as well.

As Lyon published her new evidence in further papers, other scientists generally found her idea irresistible. They dubbed it the Lyon hypothesis, or
just L.H. The random silencing of X chromosomes came to be known as “lyonization”—although Lyon herself disapproved of the name.

In 1963, when Lyon traveled to New York to speak at a scientific conference, newspapers and magazines lavished praise on her.
Time
marveled that the star of the meeting should turn out to be “a quiet Englishwoman who presented no paper and who is, of all things, editor of the semi-annual
Mouse News Letter.
”

But Lyon also incurred the wrath of a formidable opponent, a German-born geneticist named Hans Grüneberg. Grüneberg fled the Nazis in 1933, finding refuge in England and becoming a professor at University College London. In the mid-1900s, Grüneberg did more than anyone else to turn mice into a model for human heredity. He even wrote the definitive guide to the subject,
The Genetics of the Mouse.

Grüneberg had examined Lyon's thesis defense in 1950. A decade later, he read her paper in
Nature
and found it ridiculous. “
He may not have realized I wasn't a PhD student anymore—that I didn't have to ask him for permission,” Lyon later speculated.

While other scientists hailed her work, Grüneberg launched a crusade against her. In his own research, Grüneberg studied mice with a mutation on the X chromosome that produced defective teeth. According to the Lyon hypothesis, a female's teeth should be made of a patchwork of cells, some using the healthy version of the X chromosome and some using the defective one. But when Grüneberg peered into the mouths of the mice, he found that their teeth all looked the same.

When Grüneberg looked over studies of human diseases, he also failed to find compelling evidence for the Lyon hypothesis. “
It is concluded,” Grüneberg declared with the solemnity of a judge, “that the behaviour of sex-linked genes in man (like that in other mammals) gives no support to the Lyon hypothesis.”

Other scientists were appalled by Grüneberg's ruthlessness. Year after year, paper after paper, conference after conference, he kept up the attacks. His colleagues were embarrassed for him, too, because he refused to accept the evidence in favor of lyonization as it continued to pile up. One of the
most important studies in favor of L.H. came out in 1963.
Ronald Davidson, a geneticist at Johns Hopkins, and his colleagues studied a blood disease called G6PD deficiency. It's caused by a mutation on the X chromosome, causing a defect in proteins called G6PD that makes red blood cells fall apart. Men who inherit the G6PD mutation always suffer the disease. Women, on the other hand, can escape the symptoms if their other copy of the X chromosome has a normal version of the G6PD gene.

Davidson inspected individual skin cells from women who inherited the mutation. He showed that half of the cells silenced the X chromosome with the defective gene, and the other half silenced the working version. Overall, the women's cells produced enough G6PD to keep them healthy.

Grüneberg refused to accept Davidson's evidence. Instead, he started attacking Lyon's supporters, too. For a decade, Lyon later said, Grüneberg made her life difficult and depressing. Yet she maintained her unflappable tact. By the 1970s, scientists stopped asking if lyonization was real. They just wanted to know how it happened.

The answer turned out to lie in the many molecules that swarm around our DNA. These molecules—a combination of proteins and RNA molecules—control which genes become active and which remain silent. Some silence genes by winding stretches of DNA up tightly around spools. Others unwind it, allowing gene-reading molecules to reach the exposed DNA. Some proteins clamp down on a gene, shutting it down until they fall off. Since each cell may make many copies of a silencing protein, another will soon take its place. Cells can also shut down genes for the long-term by coating them with durable molecular shields. This shielding—called methylation—lasts beyond the life of a cell. When the cell divides, its two daughter cells build new shields to match the original pattern.

A number of scientists have dedicated their careers to finding the
molecules that shut down X chromosomes. Their search has led them to one stretch of DNA on the X chromosome, dubbed Xic, where several crucial genes reside. Early in the development of a female embryo, the two X
chromosomes in each cell are guided toward each other, their Xic regions lining up neatly. A flock of molecules descends on the pair of Xic regions, drifting between them in what is essentially a molecular version of eenie-meenie-minie-moe. Eventually they settle on one of the two Xic regions, where they switch on genes that will shut down the entire X chromosome.

One of the genes they switch on is called Xist. The cell uses Xist to manufacture long, snakelike RNA molecules. They slither along the X chromosome, finding a place where they can take hold. While one end of an Xist molecule grips the X chromosome, the other end snags proteins passing by to help it. Together, they twist and coil the X chromosome, until it has shrunk down to a compact nugget of DNA. The other X chromosome meanwhile remains active by keeping its own copy of the Xist gene silent.

Each cell in the early female embryo
rolls the genetic dice to pick which X chromosome to silence this way. Its pick is permanent. When the cell divides, it painstakingly unpacks the inactivated X chromosome to make a copy. In the two new daughter cells, the same X chromosome is folded back up again. The chromosome becomes like a box of old kitchen utensils you move from apartment to apartment, without ever making use of anything inside.

We can now see lyonization not only in molecular detail but also across the entire body. In 2014,
Jeremy Nathans and his colleagues at Johns Hopkins University figured out a way to make active X chromosomes light up. They inserted a gene into a mouse's X chromosome that could produce a red glowing protein if the scientists exposed the mice to a particular chemical. They engineered another line of mice to produce a green protein instead. With some careful breeding, they were able to produce litters of mice that inherited a green chromosome from one parent and a red one from the other. When they added both chemicals to different parts of the mice's bodies, their cells lit up like Christmas lights. Each cell was either red or green, depending on which chromosome fell quiet.

Neighboring cells typically glowed in different colors. But when Nathans pulled back to look over greater distances, new patterns took shape.
Purely by chance, large swaths of cells might mostly have the father's X turned on, while others had the mother's. This imbalance could affect entire organs. Some mice had brains with one hemisphere mostly red and the other mostly green. Some saw out of their left eye with retinal cells mostly using the father's X, while the mother's X gazed out of the right. The variations even included entire mice. In some animals, almost all the X chromosomes from one parent were shut through the whole body. In others, the opposite was true.

Much of the research scientists have carried out on X chromosomes has focused on their special capacity to make us sick. For men, carrying a single copy of the X chromosome means they can't hope to be rescued from a mutant gene by a working backup. As a result, most X-linked hereditary diseases almost exclusively strike men. Muscles require a protein called dystrophin to work properly, for example, and it just so happens the dystrophin gene sits on the X chromosome. Duchenne muscular dystrophy—a disease that causes muscles in many parts of the body to turn to jelly—almost always strikes boys. They inherit it from their unknowing mothers, who don't suffer the disease because some of their muscle cells make enough dystrophin to keep up their strength. Women, meanwhile, face their own troubles if silenced X chromosomes become active, throwing off their balances of proteins.

But Nathans and his colleagues suspect that lyonization might have an upside, too. It may expand the scope of heredity for women. In the brain, some neurons may inherit an active X chromosome that guides them to sprout branches in one pattern, while other neurons branch in another. The power of the human brain comes from its diversity—from different kinds of neurons, from different kinds of circuits, from different types of chemicals for communication. Lyonization may make women's brains inherently more diverse.

On Christmas Day 2014, Lyon enjoyed a holiday lunch and had a glass of sherry before taking a nap. She had been long retired at that point, with many laurels. In 1998, Cambridge had held a special ceremony to give her an official degree to replace her titular one. A Medical Research Council
building was named in her honor. The Genetics Society of America launched the annual Mary Lyon Award to recognize an outstanding geneticist. The biologist James Opitz complained that it was all “too small an honor for one, whom many I have known deemed ready for the Nobel Prize.” During her Christmas nap, Lyon died. Opitz only wished that in her final moments, she had her tortoiseshell cat, Cindy—a living demonstration of lyonization—on her lap to keep her company.

Mary Lyon did far more than reveal how women manage life with two X chromosomes. She opened up a new way of thinking about our inner heredity. Her hypothesis offered an example of how cells could commit themselves and their descendants to using some genes and not others. It turns out that similar commitments allow cells in the early embryo to turn into different tissues and organs. In the decades that have followed Lyon's pioneering work, other scientists have documented
more steps on the journey. It's a journey that starts at conception, continues through development, and lingers on for the rest of our lives.

At the moment of fertilization, when a sperm cell fuses to an egg and unloads its delivery of chromosomes and other molecules, a distinctive set of genes switches on. This special combination makes zygotes totipotent, meaning that they have total potency. A single zygotic cell has the potential to become any type of cell in the body, or even a cell in the placenta. When the zygote divides, it produces two new totipotent cells, and then four. If a doctor were to pluck any one of these totipotent cells and rear it in a petri dish, it could multiply into a complete embryo along with a placenta.

These cells, in other words, inherit not only DNA from their mother cells but their totipotency as well. This state endures from one generation of cells to the next, thanks to the molecules that hover around their DNA, determining which genes the cells can use and which stay silenced. A few master genes create powerful proteins, each of which can keep hundreds of other genes turned on and off. Those
master genes also sustain each other in feedback loops. One gene promotes another, which then turns on yet
another, which turns on the first. When a totipotent cell divides, its daughter cells inherit this same balanced network of proteins. The molecules go
right back to controlling the DNA in the two new cells, so that the new cell inherits the totipotency of its ancestor.

Totipotent cells can maintain this delicate balance for a few divisions. But then each new cell loses its totipotency, its future possibilities narrowed. The cells forming the outer shell of the embryo commit to becoming the placenta. The other cells, forming a clump of cells inside the shell, can only become part of the embryo itself. Instead of totipotent, these cells are now just “pluripotent,” meaning they still have several potential destinies.

The cells change their identities because their networks of genes and proteins rearrange themselves. When a totipotent cell makes proteins from its master genes, it doesn't produce them on a smooth assembly line. Sometimes its molecule machinery stalls and its supply of a protein runs low. Sometimes it races forward and produces a burst of molecules.

These fluctuations can throw the cell's feedback loops out of whack. One of the master genes in a totipotent cell, called Nanog, keeps many genes shut down. If a cell doesn't make enough Nanog proteins, the muzzled genes can spring into action—silencing Nanog itself. Once these gene networks flip, they can't flip back. The cell turns from totipotent to pluripotent.

Pluripotent cells get pushed farther across Waddington's landscape, falling into even deeper ravines and committing themselves to even narrower possibilities. Their random bursts of proteins continue to help drive them forward, along with signals that the cells get from their neighbors. The pluripotent cells end up in one of the three germ layers. Once a cell has become a mesoderm cell, it has surrendered its chance to become one of the other germ layers, to help build an eye or a lung. And with each new commitment, the methylation of genes—the long-term shielding of DNA—becomes more widespread. Cells begin to silence many of their genes so firmly that they can't be roused again. The networks of genes that maintain their identity as bone or muscle or gut become stronger, able to withstand errant bursts of proteins. When they divide, they reliably produce more of their kind, with the same methylation, the same coils to spool their DNA.