The gene is the monster program of all history. Millions of lines long and stored in DNA base-pair sequences, the programs inside genes are responsible for the size, shape, and structure of every living thing on the planet. They're the ultimate code, the mother of us all.
How unfortunate, then, that those programs are largely unread; how doubly unfortunate that their code is often full of bugs - system defects, flaws, errors - that cause them, when run, to produce anomalous results. In humans, other animals, and plants, these anomalies are known as genetic diseases.
You might think you could correct those defects and wipe out the diseases - if only you had a way of reading those programs, debugging them, and putting the upgraded versions back into the original gene. Then, the next time it ran inside that person's body, the gene would produce a healthy organ instead of some flawed and corrupt outgrowth. Goodbye Alzheimer's, multiple sclerosis, and cancer.
You could perform these miracles if you had a radically new type of read/write head, one that would read from, and write to, not magnetic media, not optical disk, but the genetic storage medium, DNA.
But those read/write heads are already in existence. Their development is largely the work of one man, a molecular biologist by the name of Leroy Hood.
Even in childhood, Lee Hood never did just one thing at a time. He grew up in Montana, where, among other things, he was a football star, actor in school plays, musician, debater, and editor of the school yearbook. He was the second student from his home state to win a Westinghouse Science Talent Search award, this for a geology project in a high-school science fair. He was so advanced in biology that, as a senior at high school, he helped teach the second-year/junior biology class, lecturing to his peers.
"It had quite a profound impact on me, because you really learn things by teaching," Hood says. "So, when I was finished, I knew a lot of biology."
He went to the California Institute of Technology in Pasadena for his bachelor's in the subject. Then, to learn something about the human animal (which was not overly emphasised at a science-and-technology nerd utopia like Caltech), but with no intention of ever practicing medicine, he got an MD from Johns Hopkins University in Maryland and went through the whole clinical routine - the shifts, rotations, and all the rest of it. He then returned to Caltech for a PhD in immunology. At that point, Hood decided he was equipped to do battle with the real world.
That was in the early 1970s, the dawning age of biotechnology, the era of genetic engineering, when untold marvels of medication and cure lay just over the horizon. One day very soon, theoretically, you'd be able to get cells to manufacture vaccines or hormones instead of just more cells. Or you'd be able to cure diseases by manipulating the faulty genetic sequences that caused them. All you'd have to do was rewrite human genes - a daunting but not impossible prospect.
A gene is a recipe - a blueprint, an instruction set - for a specific bodily component: a protein. Each separate gene encodes a different protein, and a given organism is the end product of all of its genes. One gene, one protein: that's the way a body is built. The human body, in fact, is the expression of some 100,000 different genes.
But to succeed in biotechnology two decades ago, you had to successfully work with proteins. That, in turn, meant being able to find the precise sequence of a given protein's amino acids, its basic constituent parts. This was known as "sequencing" the protein.
The trouble was, sequencing a protein was an extremely demanding and time-consuming task, one that involved the endless repetition of individually precise but nonetheless fabulously boring steps: Fractionate sample. Prepare lysis buffer. Add this, centrifuge that, pipette it out here. Soak, chill, heat, incubate. Suspend, wash, cover. Mix with inhibitors. Set up gel. Bless with rupees. Say prayers. The tedium of it all was absolutely unbearable, especially when you had to do it millions of times, again and again, forever. Hence the need for a protein-sequencing machine, a device that would do this stuff - or at least some of it - for you.
A protein-sequencing device had been invented in 1967, by Swedish chemist Pehr Edman, but it worked only on comparatively rich samples, ones that were brimming over with the specific protein molecules you were interested in. Many of the proteins that Hood was concerned with, however, existed in such extremely dilute concentrations that a completely new machine was required to sequence them. In the late 1970s, Hood and his Caltech colleagues developed it.
Their "gas-phase protein sequencer," as they called it, worked by breaking the specimen protein molecule apart and identifying each component amino acid in turn until the complete linear succession was known. With this device, Hood's group was able to sequence proteins using 100 times less material than ever before, discovering for the first time the chemical makeup of many important proteins, some of which would become major biotech products: interferons; colony stimulating factor, which helps recovering chemotherapy patients by boosting their white-blood-cell count; and erythropoietin (a hormone that treats anemia by stimulating the production of red blood cells), which turned into a billion-dollar drug.
But the protein "read" head was just the beginning; also possible was a protein "write" head, a device with which you could produce a protein - synthesise it out of bottled chemicals - if you knew the proper sequence of its amino acids. Such a device would let you run experiments on proteins to find out how they operated.
"A clever way to study protein function - in order to understand how they work, how they operate as molecular machines - is to synthesise them with mistakes at various places and see what that does," says Hood. "So, if you have a small protein of, say, 100 subunits, we could synthesise it and make mutations to see what it did."
But why stop with proteins? Why not go all the way and invent read/write heads that would perform the same stunts on the mother molecule, DNA? You'd be able to sequence DNA of unknown composition, and also do the reverse, creating DNA to order, piece by piece. "When we finished the first machine in 1977," Hood recalls, "we had a clear vision of the next three machines: the DNA synthesiser, the protein synthesiser, and the DNA sequencer."
Over the next five years, Hood and his crew would create all three. "The four instruments together allowed you to link the DNA and protein worlds in a way that had never been known before," Hood explains. "They were tools for moving back and forth, using information from one world to move into the other world, and vice versa."
The protein sequencer was too good a machine to keep under wraps, at least in the view of Hood's co-workers and friends, who wanted him to put the device on the market.
"Look, it's really unfair that you can do all of these sequences," they joked. "You have the only machines like this in the world. Aren't you morally obligated to commercialise these things?"
Well, it was a good point, but where was he going to find the time? By the early 1980s, Hood was a professor of biology at Caltech and the chair of the biology division. He was teaching his normal course load, and he was publishing papers like crazy, to the tune of more than 100 papers between 1965 and 1980, plus four textbooks. Plus, he had a wife and two kids, not to mention a passion for mountain climbing that he could afford to indulge only in the most hurried and abbreviated fashion. Pressed for time, he and some friends would be helicoptered up to a base camp, they'd make a mad dash to the summit and back, and then be helicoptered out again. He climbed several North American peaks this way (his wife, Valerie Logan, calls them his "macho climbs") but always stayed away from the oxygen-deprived heights of the Himalayas: "I need my brain cells," he explains.
Still, he could see the point of commercialising the protein reader. Besides, it might even make some money. So, Hood now embarked on a cross-country marketing tour, visiting all the main biological instrumentation firms - DuPont, Beckman Instruments, and so forth - explaining to middle-management types what a boon these machines would be to molecular biology, the drug industry, and the future course of research, if they would only consent to manufacture them. In all, he called upon 19 different corporations, and every last one declined the honour. "They're nice machines, but nobody really needs them," company officials explained patiently. "They just wouldn't sell that many copies. There's no money to be made."
That was in 1981. Hood responded with typical energy to the industry's judgments: he'd start his own company. So, he helped found Applied Biosystems Inc., in Foster City, just south of San Francisco. In 1983, Applied Biosystems merged with Perkin-Elmer. Today, the company manufactures, sells, and supports 25 different types of biological instrumentation systems and has offices in more than two dozen countries. Its machines are used for everything from DNA fingerprinting in criminal cases to the mass DNA-sequencing effort of the Human Genome Project. Since the flagship device, the Model 373 DNA Sequencer, was introduced in 1986, the company has sold nearly 3,000 DNA sequencers throughout the world. And at the price tag of £70,000, each has generated a healthy positive cash flow, both to the company and to Hood, who still gets patent royalties from sales of the machine. "The genome revolution would not have happened without them," says Craig Venter, head of The Institute for Genomic Research, whose group has sequenced parts of about 85 per cent of all human genes.
With these tools, quite obviously, some wondrous biological feats could be performed, and in short order, Hood and his group were performing them. There was, for instance, the miracle cure of the mouse-shivering mutation.
Mice suffering from the frightening disorder appear normal at birth, but at the age of two weeks, they start shivering uncontrollably and walk with a peculiar rolling gait. At two months, they undergo convulsions, and at three to five months, they die. Normal mice, by contrast, live two to three years.
The shivering is caused by a deficiency of myelin basic protein, an element in the sheathing that surrounds neural cells and permits the rapid transmission of nerve impulses. A deficiency of this protein implied a defect in the gene that encoded it, and this suggested that by recoding the gene, you could increase the supply of that protein. Here was the chance to cure a disease by rewriting the actual gene that caused it.
So they tried. Using the protein sequencer, Hood and company found the amino acid lineup of normal myelin basic protein. Then, by reference to the genetic code (a glossary of equivalencies between amino acids and DNA nucleotide triplets), they found the DNA sequence - the gene - for the normal protein. "It is a large gene stretching over 32,000 nucleotides of DNA, and it contains seven discrete coding regions," Hood explains.
Using the DNA sequencer, they read out the faulty gene of the shiverer mice and compared it with the intact gene from healthy mice. The faulty gene lacked five of the normal gene's seven coding regions: "Thus, the shiverer mice couldn't synthesise functional myelin basic proteins."
But if the shiverer mice were retrofitted with the correct gene, then they could synthesise them. Hood now took fertilised eggs from shiverer mice and, with a microneedle, injected into them intact genes from healthy mice. Theoretically, the new gene would be taken up by the chromosomes of the developing egg, which would mature into a normal adult.
Which is exactly what happened. Following the program contained in the newly incorporated gene, the once-diseased eggs grew into healthy mice, mice that made absolutely pure and perfect amounts of myelin.
As did their descendants. The new gene was passed on to the offspring of the cured mice, and to their offspring, and so on, for many generations; what would otherwise have been a bunch of extremely short-lived baby mice had been turned into a long and healthy family tree. It was one of the first cases in history of a genetic defect, a biological system bug, being reversed through the intentional manipulation of the faulty bits of code responsible.
But it's another machine, the Model 394 DNA Synthesiser, whose operation borders on surreal. It's a device, after all, that manufactures DNA to order - it makes the very stuff of life - right in front of your eyes.
To understand this device fully, one has to keep in mind that DNA is just another boring chemical. First discovered in 1869 by Swiss biochemist Friedrich Miescher, deoxyribonucleic acid can be synthesised like many other compounds, by mixing the right ingredients in the right amounts. Put them together in the right order, and you'll get a DNA molecule of the desired sequence. In a way, the whole spooky process should really be no big deal.
Still, it was a little weird asking, on my visit to Applied Biosystems, "Could I make some DNA?"
The answer was: "Why, of course. Sure. No problem."Ten minutes later, I was face to face with an Applied Biosystems Model 394 DNA
Synthesiser, the so-called "gene machine." About the size and shape of a microwave oven, it fits comfortably on a lab workbench. Along the front of it hang some 14 brown bottles of chemical ingredients, chief among which are the four marked "Bz dA," "Bu dG," "Bz dC," and "T." These are the supply source of DNA's four nucleotide bases: adenine, guanine, cytosine, and thymine.
There are just two output containers: a white gallon jug for waste, and a tiny transparent vial, about an inch long, where the end product, my custom-made DNA, will go.
"What sequence would you like to make?" the technician asked.
I was ready for this. I had written out, in my little yellow reporter's notebook, a sequence of five bases I'd dreamed up: ATGAC. Each nucleotide was represented once, plus an extra one for good measure."Go ahead, just type them in," she said.
Nowadays, the average computer user faces not only the standard QWERTY keys, but also the numeric keypad, cursor keys, function keys, programmable keys, status keys, power keys, plus one or more unrecognizable keys - a whole piano of keys, all just to type a letter to Mother. The informational input to the DNA Synthesiser, on the other hand, is entered by means of four lone buttons arranged in a column:
A
G
T
C
So I typed in my little sequence: A-T-G-A-C.
I paused. I pressed Enter.
And soon the machine was bubbling away, its valves opening and closing with quiet clicks. As with Hood's three other machines, this one is a masterpiece of plastic tubing and precision valves, most of which had been ferreted out by Hood's Caltech associate Mike Hunkapiller, now vice president of an Applied Biosystems division at Perkin-Elmer."When we were making the protein sequencer," Hood explains, "Mike found these very efficient valves that could operate with low volumes and not have leakage. He went around the world looking for them at places that designed and developed that type."
About 20 minutes later, the clicking ended, and in a series of 97 discrete steps, the instrument had made millions of copies of my custom-made DNA sequence, ATGAC. They're atop my desk at this moment, still in the tiny vial, right next to the stapler.
All of which can get a person to thinking. If the machine can produce that sequence, it could produce another one, a much longer one, any old string of DNA nucleotides, to order and on schedule.
Einstein DNA! Shakespeare! Elvis! All you'd need is a lock of hair, anything with the least vestige of a chromosome inside it, anything from which you could extract a tiny bit of genetic sequence. You could then produce that sequence. You could amplify it, purify it, and pour it into tiny vials.
In fact, you wouldn't even need the lock of hair. All you'd need is the information itself, just a smattering of the correct nucleotide ordering. O. J. Simpson's DNA sequences are undoubtedly on file somewhere. If you could get your hands on them, you could key a stretch of them into an Applied Biosystems Model 394 DNA Synthesiser, and an hour or so later out would come the molecules. Quite genuine! The real thing! A little bit of O. J. Simpson, right next to your stapler!
Okay, so synthesising an entire human genome is a long way off and much of the process is still only theoretical. Besides, you might not think it's a swell idea to have another person's DNA sitting on your desktop. There's something demeaning, cheapening, maybe even blasphemous, about the idea. And in fact, more than a few social critics have recoiled in horror at the thought of white-coated lab technicians reaching down into the private recesses of the DNA molecule and "engineering" it, making "improvements," getting the cells to manufacture needed enzymes or hormones instead of more cells. Do we have the inalienable right to hijack the molecules of life in this way? The prospect of tampering with human DNA, even for the purpose of curing disease, has always raised its own special brand of fear and loathing. The "playing God" objection turns up, regular as sunrise.
"Automatic sequencers provide a useful tool, no doubt, but they encourage simplistic assumptions about what life and humanity are to begin with," says medical anthropologist Barbara Koenig, who directs Stanford University's Centre for Biomedical Ethics. For Koenig, scientists like Hood have been "almost lionized" by the public, even when their "biologized" understanding of individuals is limiting and reductive.
Besides, it's bad enough that your own DNA should be tampered with; how much worse when the proposed alterations will be made not to your somatic cells (those exclusive of the gonads) but to your germ-line cells, any changes to which would be passed down to your descendants. Those prospects raise not only the nightmarish specter of eugenics, but fears about weakening the human organism altogether. "Monoculturing, or narrowing the gene pool," says Jeremy Rifkin, president of the Foundation on Economic Trends and critic of bioengineering, "leaves species less able to survive in changing environments and thus more vulnerable. Whenever you eliminate anything in an organism, you disrupt something else."
And while inventions like Hood's suggest that miraculous cures are around the corner, Rifkin isn't convinced: "Genes relate in so many ways to the larger environment in which they mutate, scientists have only the smallest understanding of the relationship between function and field."
But that's one side of the story. Successful germ-line alterations, you could easily argue, would be more valuable in the long run than somatic-line alterations. How much nicer it would be, if you had a genetic defect, to know you were saving your children from it while suffering the cost yourself, as opposed to your being cured of the disease but nevertheless passing it on to your kids.
Clearly, these arguments get emotional, probably because there's no way to resolve them, no objective decision-making procedure, as there is in the case of science proper, where experiments tend to settle a case. Which doesn't mean they shouldn't be discussed. Leroy Hood, for one, has always advocated discussion. In 1992, he and physicist Daniel Kevles edited a book on the subject, The Code of Codes (Harvard University Press), about the ethical and social implications of the human genome project. He wanted people to be more or less conversant with the facts involved. "I must say, I get really irritated," Hood says, "when I go into restaurants and see these posters: We serve no genetically altered foods.' That's nonsense. Every kind of food these days has been genetically engineered in the broadest sense. The hybrid corn or the hybrid wheats, those are 'engineered' in the genetic sense.
"And when you talk to these people, they have zero understanding of what they're talking about," he adds. "I make it a point to go up to the restaurant manager and say, 'Please explain to me that poster you have pinned up over there.'" In 1990, with the new tools of automated genetics and several start-up companies at his disposal, with the mouse-shivering molecular correction behind him, plus about 400 published works to his credit, Hood formulated his most ambitious plan of all. He imagined a new department at Caltech, one in which a group of molecular biologists, chemists, physicists, computer scientists, and others would join to remake the face of medicine. This dedicated band of biohackers would reduce to information every facet of an organism that could be reduced to information, which meant most of it, and with that knowledge - together with some new and improved read/write heads - they'd usher in a golden age of medical science. Hood went around telling his colleagues about this brave new scheme of his.
"What I see happening in medicine," he'd say, "is that over the next 25 years, we'll have identified perhaps 100 genes that predispose people to the most common diseases: cardiovascular, cancerous, metabolic, immunologic. We'll be able to do a DNA fingerprint on each individual: the computer will read out your potential future health history, and we'll have preventive measures that will let us intervene whenever there's a probability that you'll get one of these diseases - multiple sclerosis or rheumatoid arthritis or cardiovascular disease or whatever. The whole focus of medicine will be on keeping people well."
People would be cured of their illnesses, in other words, before they ever came down with them. Your entire genome would be read out, your faulty genes found, and then corrected by one or another type of gene therapy, much in the manner in which the mice had been cured of their shivering problem. To do it, all you'd need was some fancy new diagnostic machinery - some very powerful molecular-level read/write heads. It was Hood's aim to create them and send them out to all corners of the known universe.
But, much as he'd earlier been rebuffed by 19 different companies, Hood was now turned down by the officialdom of his home institution, Caltech. They didn't want him working on what they saw as mere instruments, mechanical appliances, toys. Such stuff, they told him, wasn't "real biology."
Bill Gates, for one.
In April 1991, Hood was invited up to the University of Washington, in Seattle, to deliver a set of guest lectures. The Microsoft chair and CEO attended all three. After the last one, Hood and Gates had dinner together at the Columbia Tower Club, a private gin mill and eatery on the 75th and 76th floors of the Columbia Tower building, the tallest in Seattle. There, with the Pacific Northwest stretching off in all directions, and with both the chair of the university's bioengineering department and the dean of the medical school in attendance, Hood and Gates plotted out the future of science, medicine, and the new field of molecular biotechnology.Six months later, in September, Gates presented the University of Washington with a no-strings-attached grant to the sum of £7.5 million. The university announced that Leroy Hood would come to the medical school as the William Gates III Professor of Molecular Biotechnology, that he'd be chair of that department, and that he'd be awarded various additional positions, honours, spoils, and perks. Bounteous as the offer was, Hood was of two minds about accepting it.
"It was a very hard decision to move up here," he says today. "Really traumatic."
After all, he'd been at Caltech continuously for 22 years. He'd be leaving behind one of his favourite colleagues, Eric Davidson, with whom he'd been investigating one of the prime mysteries of biology: the way in which the One Big Program that was the DNA molecule got chopped up, parcelled out, and expressed differently by different parts of the developing cell. It was a problem as old as Aristotle, who'd observed chicken eggs as they became chicks and watched the yolk develop into a beating heart. How did the different parts of a developing embryo know which specific bodily component to become -the heart, the brain, the liver, whatever?Aristotle, of course, knew nothing of DNA, programs, or gene expression, but his original question now arose in a new form. Each nucleated cell contained the DNA program for an entire organism, but no one cell ran the whole program. How did any given cell know which particular part of the program to run?
Solving that problem, though, was only one small piece of Hood's greater life plan, which included wiping out the genetic diseases of humans and envisioning "improving" the species in certain fateful ways. "It is certainly going to be possible to discover anti-aging and anti-cancer genes," he says, "and perhaps to permanently enhance qualities such as intelligence and memory." So, after a period of some soul-searching, Hood finally picked up his belongings and moved to Seattle.
"It's worked out as well as it possibly could," he says of the move. "Just all sorts of exciting things have happened."
One of them was Darwin Molecular, a new type of drug company that Hood co-founded in 1992 with money provided in part by Gates. Historically, drug development was a copycat business, with companies making slight alterations to old drugs, or experimenting with new compounds in a hit-or-miss fashion. It had been one of Hood's dreams to take a more intelligent approach, reading information out of the genes and using it to guide drug design.
Theoretically, you ought to be able to isolate some genetic disease, find and sequence the gene that caused it, and then invent a drug to combat the disease at the molecular level.
"For example, there are genes that predispose people to getting cancer," says David Galas, president and CEO of Darwin Molecular. "We want to make molecules that can address those genes, that can interact with the gene products."
Located in Bothell, across Lake Washington and a half-hour's drive from the university, Darwin's research lab is outfitted with machines whose designs are based on Applied Biosystems sequencers and synthesisers. Company researchers use the sequencers to read the gene ("We can read the sequence of a gene in a couple of days," says Galas), and they use the synthesisers to help create an array of potential drug molecules - ones that can beneficially affect the proteins encoded by that gene. Then, in a process known as directed molecular evolution, the scientists set these candidate molecules into competition with each other, letting them evolve until, out of the contest, only the fittest molecule will survive - the molecule that is best at treating the illness caused by the diseased gene.
"There are about 100,000 human genes," says Galas. "We now know about the functioning of much less than 1 per cent of those. We see our role as finding new genes, picking out carefully the ones we want to work on, and then finding the small molecules that can affect them."
Darwin is concentrating specifically on autoimmune diseases: multiple sclerosis, for example; rheumatoid arthritis; AIDS. The hope is to discover some new and aggressive molecular cures.
"But it's a complex problem," Galas admits. "It involves a level of control of the immune system that's well beyond what anybody else has ever attempted." Back at the University of Washington, Hood has created a radically new type of academic entity, the department of molecular biotechnology, whose focus is not so much to understand organisms as it is to build machines for understanding and manipulating them - a highly unconventional approach.
"In my experience, technology development is looked at somewhat askance by 'pure biologists,'" says Gerald Selzer of the National Science Foundation. "It's unusual to see, at least in biology, a group of people focused on technology development."
But it's never been unusual for Leroy Hood. "Developing new technologies gives you more leverage than any other thing you can do," he says. "What's unique about our lab is the way we're coupling leading-edge biology with leading-edge technology development."
The lab, on the top two floors of a concrete-and-glass structure overlooking Lake Washington, contains the normal assortment of biological solutions and glassware, plus something that you don't find in the average biology lab: an array of computer equipment, test platforms, and other gear.
Hood and his crew are working on a chip that will sequence unknown DNA in large batches instead of just a few nucleotides at a time. A chip the size of your thumbnail will have 65,000 DNA fragments on it, and each of them will react uniquely with different portions of unknown DNA. Place a drop of purified DNA on the chip, and read off whole stretches of sequence immediately.
Then, of course, you have to make sense of it all. Hood's group is working on that, too. Under development is an array of chips that will let you compare your newly found sequence against all others in a database.
Eventually, you'll have this great big database - a catalogue of human DNA - that you'll be able to leaf through like a book.
"You'll decide which part of the DNA you want to do experiments on, and you'll use a Hood synthesising machine to synthesise it," says Maynard Olson, who gave up a Howard Hughes medical research grant in St. Louis to be part of Hood's department. "You could go from sitting there browsing through the human genome to doing experiments on any selected part of the human genome in the laboratory, just a few hours later. Someday not too far away - 20 years from now - this will be molecular genetics."
And at that point you'll be able to read and write the mother code at will, just like Lee Hood always knew you'd be able to. You'll be able to hack that code, flawed and buggy, and experiment with it until you get it right - until you get it fully optimised and correct, straightened out, polished, fixed - the way it should have been written from the beginning.
Until then, Leroy Hood travels the country putting together his empire piece by piece. He has started a Prostate Cancer Consortium for the purpose of finding the gene that predisposes men to that ailment. And he has returned to work with Eric Davidson again on the embryo-development problem.
Biological dynamo Leroy Hood still doesn't do one thing at a time. Sometimes spending only two weeks a month at his home base, he traverses the country gathering together experts, gathering money - foundation money, grant money, private money - planning projects, and envisioning the next phase of the biotech revolution. He flies here, there, everywhere - king biohacker - gone in a blur. And in economy class, no less, just like a regular guy.
Ed Regis (edregis@aol.com) wrote about the Extropians in US Wired 2.10. His latest book is Nano: The Emerging Science of Nanotechnology (Little, Brown); he is currently writing a book about the hunt for new viruses.