by David Orrell
DNA, it seemed, was rather like a kind of software program for the expression of genes. If scientists could figure out the operating system, then perhaps they could hack into the program to read the future—and even change it.30
THE MASTER MOLECULE
You can easily inspect the DNA molecule in the comfort of your own home.* Take half a cup of split peas (with a nod to Mendel), add a pinch of salt and a cup of ice-cold water,and blend on high for fifteen seconds. This separates the pea cells. Strain the soup and add about two tablespoons of liquid detergent. Mix, then let sit for ten minutes. The detergent breaks open the cell nucleus, which holds the DNA. Pour the liquid into test tubes or similar glass containers until they are about a third full. Now add a pinch of meat tenderizer to each tube and stir very gently. The tenderizer includes enzymatic proteins that break up the proteins clustered around the DNA. (If you don’t have meat tenderizer, pineapple juice may work.) Finally, tilt the container and slowly pour rubbing alcohol down one side, as if you were carefully trying to fill a glass of beer, until the volume has doubled. If everything has gone as planned, some white stringy stuff will gently rise to the surface of the alcohol, which you can extract using a chopstick or whatever you have at hand. This is the DNA. It’s stringy because the long DNA molecule is partly unwound. A molecule of human DNA, if uncoiled, would be almost two metres long, though of course immensely thin.
If you like, carefully rinse the DNA and transfer it to a fresh glass. Add a shot of vodka, ice, and tonic. This Master Molecule drink will bring any party to “life”—at least a primitive, pea-like form of life.
*Based on a recipe from the Genetic Science Learning Center, University of Utah.
BIG SCIENCE MEETS THE MASTER MOLECULE
In the early 1980s, the U.S. Department of Energy was casting around for so-called big science projects that would complement its high-energy physics programs. In December 1984, at a summit to discuss techniques to detect genetic mutations in descendants of the survivors of Hiroshima and Nagasaki, someone proposed a project to determine the entire sequence of the human genome— the exact order of the As, Cs, Ts, and Gs (which differs slightly from person to person). The sheer length of the molecule meant that it would be impossible to decode in a reasonable time without the invention of new technologies. However, the promise seemed enormous: knowledge of DNA could help (or at least identify) victims of genetic diseases, lead to new kinds of drugs, and perhaps reveal the secrets of life.
The Human Genome Project, as it became known, has been compared to determining the exact address of each person on the planet. If the earth corresponds to the genome of a single cell, then a chromosome is a country, a gene is a town, and the base pairs— those billions of As, Cs, Ts, and Gs—are individual people. The enormous task was made possible by a range of new techniques for determining the sequence of DNA, which worked essentially by dicing it into small chunks, figuring out the sequence for each piece, and splicing it all together again on a computer. Attracted by the emphasis on numerical approaches, as well as the funding possibilities, many physicists, engineers, computer scientists, and mathematicians switched their attention from physical systems to living ones and joined the search for the so-called biological grail. (Of course, according to mechanistic science, there is no fundamental difference between physical and biological systems.) New disciplines were born, including bioinformatics, which uses mathematical and statistical tools to probe strings of DNA for hidden meaning, and systems biology, which analyzes the function of complex biological networks.
Rather than start straight in with the human genome, it was decided to warm up on some simpler organisms. The bacterium Haemophilus influenza had its DNA decoded in 1995. Baker’s yeast, Saccharomyces cerevisiae, gave up its secrets in 1996.31 The effort to determine the sequence of human DNA turned into a competition between a private company, Celera, headed by the former professional surfer Craig Venter (who sequenced DNA from his own body, perhaps to find the elusive surfing gene), and a public consortium of universities and research institutes. After a sometimes acrimonious battle—DNA, it seemed, was not just the molecule of identity, but the molecule of ego—the two sides eventually patched up their differences and, to great fanfare, jointly announced their triumph in 2001. The British prime minister, Tony Blair, said that it heralded “a revolution in medical science whose implications far surpass even the discovery of antibiotics.”32
One of the immediate surprises was that the human genome has only about 30,000 genes. The roundworm has about 19,000 genes, and even yeast has 6,000. Most estimates for humans, perhaps out of sheer pride, had put the number at at least 100,000. How could such a small number of genes produce so much complexity? Actually, it’s not the number of genes that count, but the different ways they can be combined and expressed, which is essentially limitless. The variability among humans involves even fewer genes: about 93 percent of those discovered are held in common, so individuals may differ in only a couple of thousand genes. But even this is far more than enough to ensure that no people apart from identical twins have the same genes—just as a phone system with only ten digits is enough to supply everyone on the planet with a unique number.
Another surprise was that about 98 percent of the DNA didn’t seem to code for any gene at all.33 The remainder was termed, somewhat prematurely, “junk DNA.” At least some of it has since been shown to play an important role in gene regulation, and perhaps even social attributes.34 The most embarrassing discovery, though, was that the small amount of DNA that does code is about 98 percent the same as a chimpanzee’s. We are more than just descended from apes, as Darwin showed; according to our DNA, we practically are apes.
Of course, we run into the problem of metric again. A measurement of the genetic text does not correspond to a measurement of particular outcomes. For example, the following two sentences differ by only a few percent in terms of their letters, but their meaning is entirely different:
Chimpanzees and human beings are very similar animals.
Chimpanzees and human beings are not very similar animals.
As in language, details count.35 There isn’t a straightforward, linear relationship between letters of code and the end result. The path between the two is crooked. In genetics, the word “not” corresponds to a gene that turns off transcription of another gene. Genes that activate or repress other genes in this way are ubiquitous.
PREDICTING HEALTH
One of the main goals of the genome project is to prevent diseases before they happen.36 The cost of sequencing DNA—estimated at $3 billion for the first human genome, or a dollar per base pair—continues to plummet. In the near future, it’s expected to reach a level affordable to many people—say, under a hundred dollars. Since DNA is present in all cells, it would only take a sample of blood or a swab from the inside of the mouth to determine someone’s genetic sequence. Small differences in the genome may correlate to susceptibility to certain diseases, anticipated longevity,and so on. Dozens of companies, including Celera, are searching for DNA tests that can foretell your future health—a genetic horoscope. As James Watson asserts: “We used to think our future was in the stars. Now we know it is in our genes.”37
Computer-based biological prediction is a new area that is evolving fast. Most genetic tests performed on a routine basis today are for so-called Mendelian diseases—those caused by a single gene. These can be devastating but are fortunately quite rare. Cystic fibrosis, for example, is associated with a recessive gene; if both parents carry it, the fetus may develop the disease. Detection is complicated by the fact that the relevant gene is extremely long, and the corresponding protein exists in hundreds of slightly mutated forms. Nevertheless, genetic testing of the parents can provide some warning.
Huntington’s disease is caused by a mutation that inserts too many repetitions of the sequence CAG (code for the amino acid glutamine) into a particular gene. The result is a deformed protein that slowly accumulates in the brain, leading to
progressive loss of movement control and eventually death. The age at which neurological symptoms first appear depends on the number of CAG repeats.
The predictive powers of the scientists seem almost magical: by analyzing a single gene, they can foretell a patient’s future symptoms. Unfortunately, the magic does not yet extend to a cure. If you have the defective gene, it appears in every cell in the body, so it cannot easily be corrected (though gene therapies hold potential). This raises a number of ethical issues around genetic testing. If your child had a high likelihood of contracting an incurable disease, would she really want to know? Would she want you to know? Would either of you want an insurance company or an employer to know? And when does genetic testing cross the line into eugenics?
The genes BRCA1 and BRCA2 were so named for their connection with a rare form of early-onset breast cancer. About 10 percent of breast cancers are believed to be hereditary, and about half of those may be associated with these two genes, which produce proteins responsible for DNA repair. Certain mutated forms are less efficient at this task, and can lead to breast or ovarian cancer. Genetic screening for the condition is complicated by two factors. First, each gene exists in hundreds of different forms, and it is hard to know which are dangerous, so the aim is usually to find out whether a family member with cancer has passed on a copy of the same gene version to their offspring, and the results must be carefully interpreted by a genetic counselor. Second, these genes are “owned” not by the person being tested, but by a Salt Lake City biotech firm called Myriad, who were the first to locate, sequence, and patent-protect them. Myriad therefore has a temporary monopoly on the genetic tests, which again raises a host of issues, especially for those who can’t afford the fees charged (typically thousands of dollars).38 Similar patents, covering the majority of the human genome, have been taken out by other companies including Celera, Human Genome Sciences, and Incyte in a kind of DNA land rush, with the aim of producing useful and marketable genetic tests.
Genes, of course, do not just cause diseases; in fact, they may even grant immunity. A person with blood type AB (about 4 percent of the population) is virtually immune to cholera, for example. Cholera is not a genetic disease, but it can have a genetic cure (or an economic one—it was eliminated from Europe and North America by improvements to the water supply). The gene that causes sickle-cell anemia, a blood condition that is especially common in people of African descent, also protects against malaria. Mutated forms of the gene CCR5 appear to foil the virus that causes AIDS.39
NATURE, NURTURE, OR NEITHER
If a particular gene is mutated, its protein products may cause Mendelian diseases; however, we cannot conclude from these special cases that every disorder has some unambiguous genetic root. Most diseases that have a hereditary component—including forms of cancer, heart disease, asthma, and diabetes—are not the result of a single mutant gene ticking away like a time bomb. Instead, they come from a combination of genetic and environmental factors. (It is often more appropriate to discuss gene versions rather than mutants. Genes vary, and the fact that your DNA differs from Craig Venter’s doesn’t mean that it isn’t normal, or that you are a mutant.) Predicting the appearance of these diseases is not a harder version of predicting the Mendelian diseases, but a completely different— call it Galtonian—class of problem. Just as global political problems are not usually caused by a handful of people and atmospheric storms are not stirred up by a single butterfly flapping its wings, so most health conditions cannot be traced back to a few rogue base pairs.
Heart disease, for example, is influenced to an extent by a gene known as APOE. This produces a protein that plays a role in the complex process of preventing cholesterol and fat from accumulating on the linings of blood vessels. The gene comes in slightly different versions, the most common of which are designated E2, E3, and E4. A person with two copies of E4 (i.e., who has inherited one from each parent) stands a relatively high risk of developing early heart disease.40 What counts is how the many different proteins involved interact as a group, however, so no single gene tells the full story. Current research therefore focuses on finding sets of genes that tend to vary together; but it is not clear that these are the sets responsible for disease.41 There are also other hard-to-quantify and inter-correlated factors that affect heart disease, such as fitness, obesity, noise pollution,42 tobacco use, stress, job status,43 the extent of one’s social network,44 the number of fast-food restaurants in the neighbourhood,45 mental attitude, and of course, sheer luck.
Another challenge in making predictions based on genetic data is that what counts is not the genes we have but the genes we express as proteins. This changes with time, environmental cues (such as fast-food restaurants), and location in the body. All cells in our body contain the same set of DNA instructions, but a liver cell is not the same as a heart cell, and a skin cell is different from a brain cell. They each interpret the text in a different way, using only what they need. In one particularly striking experiment, scientists transplanted mouse cells into chick embryos, and thus induced some of the chicks’ own cells to develop teeth. The strange thing is that while chicks normally don’t have teeth, they are thought to have descended from dinosaurs, which did. Evolution had repressed the growth of teeth (it had inserted the word “not”), but the genes for them had remained, like a memory from a past life.46
The expression of genes is not static and fixed. It’s promoted or repressed in a time-dependent, dynamic fashion by proteins that bind directly onto particular sites on the DNA. A gene may be switched on or off by the action of several proteins working in concert. Similarly, hormones such as those associated with stress exert their effect on the body by controlling gene transcription. After stress hormones are released from the adrenal glands, they enter into cells, combine with special receptor molecules, and bind to the DNA, where they control the expression of certain genes. Such hormones are, of course, particularly susceptible to environmental effects. Scud missiles were mostly ineffective against Israel during the first Gulf War, but the threat of them caused a huge spike in the number of heart attacks. Generally, stress is believed to weaken the immune system and make people more vulnerable to disease.47 Stress acts at a genetic level, but we wouldn’t say it was a genetic disease. Mental states such as depression also affect the expression of certain genes, but this does not mean their cause is rooted in those genes.
Even a developing fetus, safely ensconced inside the womb, is subject to complicated environmental effects, although the environment is now the mother’s body. One factor is the prenatal testosterone level, which affects the development of the nervous system and is influenced by the fetus’s sex. High testosterone levels loosely correlate with autism, immunosuppression, and aggression. On the bright side, those affected are often good at music and mathematics. Prenatal testosterone also influences expression of the genes that control digit length. In women, the ring and index fingers are usually around the same length, but in men the ring finger of the right hand is usually longer than the index finger.48 Therefore, finger length is a weak predictor of these traits, though the correlation is so weak that it is meaningless to read anything into it for a particular individual. (It has a certain Pythagorean logic, however: a long straight finger on a male right hand is linked with music, mathematics, and aggression.) Since the prenatal testosterone level is part of the dialogue between mother and developing child, it is impossible to say whether it is nature or nurture.
Practically all traits, in fact, are the result of interplay between nurture and nature. A child might be anxious because of an inherited nervous temperament (nature), or because of exposure to stress-causing experiences (nurture). But that nervous temperament might have been induced not by the parents’ genes, but by their anxious behaviour (nurture). Or those stress-causing experiences might be the result of an inherited tendency to seek risk (nature). Or the parents’ anxiety might have been caused by the child’s behaviour, making the child even more stressed, a
nd so on, in a positive feedback loop, until the men in the white coats show up. Nature and nurture are like two intertwined forces that sometimes work together and sometimes work in opposition. Even the nurturing experience of reading this book will change neuronal pathways in your brain, with perhaps strange and unpredictable effects on your nature. And to nature and nurture, we must also add blind chance—those subtle effects that are essentially random and cannot be controlled.
Chance plays a role in every stage of our lives, not least the moment when we are conceived. Aristotle believed that a child’s sex was determined by the temperature of the male’s sperm, which he in turn thought was a function of the weather. He therefore advised people who wanted a male heir to conceive during the summer.49 While the sex of some reptiles does depend on the temperature at which the eggs develop (the preponderance of males born in some species is an early indicator of global warming), a human child’s sex is determined by a random process—namely, whether the sperm that fertilizes the egg contains an X chromosome (female) or a Y (male).50 Similarly, a coin toss may determine whether a particular gene is expressed, a neural pathway laid down, or a cancer developed. And owing to small random differences in their development, even genetically identical twins are not really identical, as their fingerprints, those weather maps of identity, will show.