Kary Mullis: The Ultimate Gene Machine

Interviewed April 1992 by Anthony Liversidge

It was a shock for the audience, so soon after breakfast. The keynote speaker at "Nucleic Acids: New Frontiers," a San Diego conference of the American Association for Clinical Chemistry, was giving his slide talk. Then, right after a scientific chart, there flashed on the huge screen a sharply defined image of the speaker's ex-girlfriend clad only in a multicolored Mandelbrot fractal pattern, generated by computer and projected with considerable transparency onto her skin. As a nervous frisson rippled through the auditorium, the speaker Kary Mullis was unabashed. "This is my home town, and I can do what I like!" he joked. Indeed, no one seemed to mind this and other examples of his "creativity." As a young woman chemist said afterwards, Mullis might be "a sexist pig," but his ideas were "so refreshing, I could have listened to him all day."

Among the more scientific goodies Mullis served up was a technique to filter DNA from blood in 15 minutes (the conventional process takes a day), a suggestion that the historic Avagadro number system for counting molecules be replaced by a common-sense Mullis measure of "things per microliter," and a novel explanation for how AIDS defeats the immune system. But beyond the scientific pyrotechnics, charm, and provocations, Mullis also had the enduring respect of everyone in the room for the singlehanded invention of PCR--polymerase chain reaction--the most powerful lab advance in molecular biology in a decade.

PCR finds and multiplies tiny fragments of DNA millions of times in just hours. Picking out a few base pairs from normal DNA is like hunting for a needle in a haystack. PCR detects the bit of DNA aimed for and keeps doubling it. After 30 exponential steps, you have a haystack of a billion needles. The technique has replaced clumsy recombinant DNA cloning with a fast and slick way to make as much specimen DNA as desired and has inspired a burst of shortcuts in the still-vast project to decode the human genome.

Applications abound. In forensics, PCR helps solve murders from degraded bloodstains and fathom mysteries of disease from samples as old as 40 years. It helps diagnose genetic problems in human eggs and embryos, can detect the AIDS virus in blood before antibodies develop, and may yield an early warning system for cancer. Archaeologists and botanists can track evolution in musty specimens of dried leaves, skin, hair, feathers, and eggshells. They line up mice carcasses and plot the history of the species by means of DNA variations that PCR detects. PCR has proven that a 40,000-year-old mammoth is the ancestor of the elephant, and it was used to analyze DNA in an 18-million-year-old magnolia leaf preserved in a peat bog. There are plans to use PCR to trace ships responsible for oil slicks and to analyze President Lincoln's bloodstains and bone fragments for genetic weaknesses.

One problem with PCR is that it's so sensitive, the investigator's own hair and skin can contaminate the works. There is also a patent war. Mullis, now an independent consultant on PCR applications, invented the process while at Cetus, which owned the patent. The biotech company was sued by Du Pont (Du Pont lost) on the basis of a paper written 20 years earlier by Nobel-prize winner Har Gobind Khorana (then of MIT) that raised the possibility of a process such as PCR. Meanwhile, other companies are racing to develop rival systems. Mullis helped Cetus defend itself, although he resigned two years after he invented PCR, for which his reward was a meager $10,000 bonus. His brand of freewheeling creativity was way too far out for corporate minds, even after he produced a breakthrough now worth hundreds of millions of dollars. In the meantime, Cetus sold PCR to Hoffman-LaRoche for $300 million and consequently merged the company with Chiron Corporation of Emeryville, California, in December 1991.

Mullis, whose emancipation from dull conformity has always been evident, was born in 1944. As a high-school student in Columbia, South Carolina, he sent a frog up a mile and a half in a rocket. He started a poisons and explosives chemicals factory at Georgia Tech and invented a remote-control device to turn lights on and off with brain waves alone (flipping to a men's magazine center-fold triggers it). As a biochemistry major at Berkeley (from 1969 to 1972), he taught neurochemistry classes in hallucinogens, and at age 24 he published his first paper on the structure of the universe in Nature magazine. His adviser, Joe Nielands, recalls that Mullis's thesis was written in such a personal manner that it's still the best read in the department. Mullis was "very undisciplined and unruly," Nielands says--"a free spirit."

Interviewer Anthony Liversidge talked to Mullis at his La Jolla home overlooking the Pacific, where surfers dotted the afternoon rollers. After the interview, Mullis rushed to join them.

OMNI

Are you sexist?

Mullis

No, but a lot of people think so. I show pictures of naked women, that's why! I like naked women. I like them with their clothes on, too, but if you're gonna take a picture, you might as well take it with them naked. I show them with those Mandelbrots on them. This is my art! If you don't like it, close your eyes. I gave a lecture in Naples to a math department about how fractals were generated, interspersed with pictures of naked women with Mandelbrots. It went over real big. They wanted copies of my slides. They didn't call me sexist, though it's hard to be a sexist in Italy. But almost always someone comes up and says you can't show those kinds of pictures. It's not politically correct. I've cut down on it now because it causes me so much trouble. Women are some of my best friends and the people I confide in most. But sometimes the harder you try to say you sympathize with their problems, the more some people just sneer and say, "Sure."

OMNI

What are you working on now?

Mullis

I've made this breakthrough in using PCR to screen the blood supply. This normally takes about a day, exposes people to infection, and is a pain in the ass. My way takes the DNA out of a large volume of blood in 15 minutes. I just wrote a new patent. I tried the simplest thing possible and it worked. Pretty lucky, but lots of times I get real lucky. I took advantage of the fact that DNA is spider-webby. You get it out of cells by breaking them very gently. You pour the blood into a buffer with a detergent. Just don't jiggle the solution; let the DNA come out slowly so as not to break the pieces. Takes about five minutes. Then you pour the solution with the DNA molecules into a little filtration device. The only thing that stays on the filter is the DNA, because it's a long molecule, maybe a millimeter. The DNA catches like spaghetti in a colander. People said I'd never get the DNA off the filter. I tried for three weeks and finally popped it into the microwave. In a minute all the DNA came off the filter into the water! It's almost as if my fairy godmother said, "Pop it in the microwave."

The trick probably works because microwaves turn water molecules into little buzz saws by whirling them around. Where the DNA is pulled down into the holes [of the filter], it's twisted and stretched, and these little water molecules go whoosh! and rupture the bonds. Whatever it does, it works and works real fast.

OMNI

What inspired PCR?

Mullis

I wasn't developing a way to amplify DNA at all. It was like I was randomly putting Tinkertoys together and finally made a structure and said, "You know what? If I turn this toy wheel over there, that damn thing would wind string." Driving up to Mendocino and thinking about an experiment to look at one particular letter of the genetic code, I designed a system in my mind. As I repaired the things I thought could go wrong with it, suddenly I generated something that if I did it over and over again would be PCR. It would go 2, 4, 8, 16, 32 . . . in 30 cycles make as many base pairs from one little region as I had in the whole genome! That was the eureka point. I said holy shit! By putting the triphosphates [DNA building blocks] in there myself, I could do this process over and over and amplify the DNA.

I slammed on the brakes and stopped by the side of the road to calculate it out. Then I drove on, because my lab assistant was pissed. I said, "I have just come up with something incredible." She'd been asleep in the car. She was the only person in Mendocino that night who knew anything about biochemistry, but she didn't think it was any good.

A couple of miles down the road I stopped again. I realized I could use these bastards, the oligonucleotides [short pieces of DNA], and get the enzymes to reproduce as big a piece as I wanted to. They didn't have to be aimed at just one base pair. Hell, I could do a whole sequence. I realized you can cut the sequence out from a great big molecule. Pretty cool! Just cut, paste, and amplify. This is going to be a tool that's spread around the world!

I said, "If you get up and listen and help me with this, I'll put you on the invention." But she was just like the others at Cetus. She did not believe I could possibly have invented anything interesting because she knew me. She is not on the invention!

OMNI

How does PCR work, starting with one oligonucleotide?

Mullis

I have suggested dropping that clumsy word from the dictionary. An oligonucleotide is a short piece of several nucleotides, of single-stranded DNA. In PCR it acts as a primer, anchoring itself onto a long, single strand of DNA and getting elongated by the polymerase, the enzyme molecule. The polymerase copies the DNA by snatching these little monomers [DNA constituents] out of the solution and stuffing them in at the right place on the oligonucleotide. The polymerase copies down the information from the long strand. With PCR, the first copy you make has one end defined, so it can't get elongated. During the next cycle, the other end is closed off, too, so the polymerase just copies the target section of DNA you want. Then in cycles after that, only the defined DNA piece will be copied. It can be copied forever.

OMNI

So you end up with a pure sample of the DNA you're after?

Mullis

PCR detects a very, very small amount of some sequence interspersed in a whole bunch of similar sequences. Then PCR makes so much of the sequence, you end up with something that is almost all what you're interested in. It purifies as it amplifies because it only amplifies one thing. It's like a radio amplifying only one wavelength amid all that are coming in.

OMNI

You made history at Cetus. Why aren't you still its superstar?

Mullis

If Cetus had been more attentive to the needs of its inventors, I would have stayed and invented more things. At first it was a good environment. I used the Cetus computer to set up a lab to make oligonucleotides and then had nothing to do. My boss said, "Don't tell them you've got it beat. Just play." I started thinking of what we could do with all these oligonucleotides. I'd no real responsibilities for about two years, and just played. By the end, I had PCR.

OMNI

What went wrong?

Mullis

When they finally realized that someone among them had discovered something royally good, every sonofabitch administrator who wanted to make a name for himself suddenly decided he wanted to be my boss. There were wolves all around me. They all started proposing experiments for me to do, treating me like a grad student. By then I was working on what I thought would be the future direction of PCR. Nobody quite understood. They demanded I write down what I did and present it to a committee who'd decide if it was okay. I said, "I don't think that's necessary; I should be able to drift, okay?" They said no.

They put me over the flames--"You haven't done this control, that control." I said, "I've done it before and can probably do it again, but I am not a technician. If I have to do things your way I'll end up doing things just like you and that's all."

They should've said, "Okay, you just produced something that might make us a hundred million bucks, maybe a billion. What can we do to make life easier?" But they took the golden goose and cut its head off.

I was screaming for a year and a half about how important PCR was and no one was listening. They didn't expect an important breakthrough in genetics to come from an oligonucleotides lab. They didn't understand that important inventions almost always cross the lines of disciplines. You don't develop an invention by having one hundred guys working for five years to produce an invention. You have one guy who may even be flaky in his field and who jumps around and puts shit together in unlikely ways and sees something. It's hard to imagine even a good administrator having a sense of how it works. If he did, he'd be an inventor himself because it's more fun.

Most administrators work in sleazy ways, conferring in back rooms and coming in and acting as if the decision hasn't been made, and suddenly you think, What happened here? And they're off again to have another big meeting.

There's nothing on the agenda of the board meeting on "What We Have Done This Week for People Who Have No Legal Right to It"!

OMNI

You didn't even get a promotion as a result of your discovery?

Mullis

I didn't ever hang out in situations that lead to that. How do you do it--put a sign up in the bathroom? I used to hang out with the younger people in the company. A mistake. But I don't generally like people my age. Most of my good male friends are former boyfriends of my daughters.

OMNI

In other words, you're not interested in power?

Mullis

Over my own life, yes, but not over anybody else's. I had power. I ran the oligo lab and went up and said, "Let me out of this. I want to work by myself in the lab."

OMNI

But you helped Cetus fight Du Pont over the patent for PCR?

Mullis

I wouldn't have if Du Pont hadn't sued Cetus. I could've worked for either but reckoned I'd defend my patent. I felt the challenge was unreasonable. Nobody ever heard of what Khorana had done. It was unpublished because they didn't think it'd really work--because Khorana was publishing papers about once a month. It would have been a very interesting advance and would have changed the course of a lot of his work.

Mullis

In 1971 his lab was assembling synthetic pieces of DNA. They were trying to figure out a way of making more of what they had already made in pure form. They knew there was an enzyme that would copy DNA sequences if a single long strand sequence had another short, single-strand piece, with a primer, on it. The DNA polymerase would start at the end of the short strand and add nucleotides complementary to the long strand. They wondered if it were possible to get a piece of DNA, melt the two strands apart, put the short DNA primer on them, and get the enzyme to make two copies. The trouble was there was a competing reaction. If you melt a DNA molecule apart into two strands, as soon as you cool it they'll find each other and wrap up into one strand again.

Somebody of high authority, maybe Khorana himself, said that it wouldn't work for this reason. As far as I can tell, they abandoned those experiments and stuck to doing the strands separately in two tubes. Then at a meeting in Hawaii, Stan Cohen [Vanderbilt] and Herb Boyer [UCSF] invented recombinant DNA cloning in an all-night deli drinking beer. When word got back to MIT, they dropped all those experiments and pushed all the tubes to the back of the freezer. They solved their problems by cloning. So they never went back and tried to solve the separation problem.

OMNI

Cloning will make huge amounts of DNA, but unselectively, right?

Mullis

In molecular cloning, every 30 minutes or so the cell divides, making 2, 4, 8, 16 copies of an inserted DNA sequence. But every time it doubles, it copies everything, the whole DNA piece. There might be 5 million pairs of it there.

Cloning was a glitzy thing when it was discovered. It just blew people away. Everyone who could find a use for it dropped everything and did it. Now PCR has replaced it in many ways because it makes a small piece of DNA faster and easier. You don't get all the other stuff; you make a purified product. PCR is the first step that makes it easier to clone. It allows you to go in and clone a human gene by first isolating it, like a pair of tweezers going in and pulling it out. Once amplified up a millionfold, you can separate it and clone it up. All cloning of human genes today uses PCR in one way or another.

OMNI

What is the direction PCR's going in right now?

Mullis

It's so widely used by molecular biologists that its future direction is the future of molecular biology itself. It's like asking what is the future direction of the screwdriver--it's whatever people use screws for. PCR is to DNA what the screwdriver is to screws. For now, PCR's future is wherever anything is being done with DNA. There is a PCR machine in every DNA lab already.

OMNI

What neat things are PCR machines making possible?

Mullis

A rather complex one is a way of re-creating evolution. This will have major significance in designing pharmaceuticals. PCR enables you to re-create molecular evolution fast in the lab. You start with a mass of molecules and select the property you want. Then PCR pulls the molecules with that property away from the rest--one part in a trillion--and amplifies that. In the process, you introduce new mutations and select for those with the properties you want. Craig Turek at the University of Colorado has already used it to select for RNA molecules and is now using it to select for protein molecules.

OMNI

Why did you decide to call your cabin up in Mendocino the Institute for Further Study?

Mullis

A lot of papers end with the phrase "This result deserves further study," trying to get a grant. The IFS is a place where this can be done, where no study can be considered complete, where all results will be held pending further study, where no publications will be forthcoming, and all appointments are tentative. I am the provisional director awaiting study of the committee studying the provisional status of the director!

OMNI

Where are you going next?

Mullis

I have a great new job with General Atomics in San Diego. They make nuclear reactors for satellites and one of the two controlled fusion reactions in the world that work. Most of their stuff is classified. They are physicists, basically, who've decided to apply their sophisticated devices to create new biology technology. Molecular biology is, after all, a result of physicists going into biology. I'm supposed to look over their scientific proposals for research into problems they've identified and see if the proposed solution makes any sense. This is what I like most to do--look at basic ideas and have others work on ideas I suggest. I will come up with and review ideas.

OMNI

What are the potentials here?

Mullis

Frontiers of medicine are converging with the frontiers of chemistry. Aside from surgical procedures, developments have been coming mainly from chemistry for the last 15 years. Biology down underneath is just chemistry in a very, very sophisticated manifestation. Now we can approach biological problems with sophisticated instruments. We can study the structure of the microbe that causes measles, or whatever.

OMNI

Do you think that you will be able to come up with a second PCR-type breakthrough?

Mullis

My blood sampler may benefit some of their physics expertise and might be more valuable to the world than PCR. To identify the hundreds of compounds in a blood sample, I'll use a scanning probe microscope that can see atoms one at a time. The aspect I'm working on now is cluster coding molecules--like putting bar codes in a grocery store. The checker never has to look at the article you're buying but just has to pull it over the bar-code reader. I'm trying to do that for clinical chemistry to find a way to recognize the molecules in a blood sample without having to observe them directly. If you can put a bar code on each kind of molecule and find some way to read it, the problem is over.

OMNI

You'll put a blood sample in your machine and out will come a list of its constituents?

Mullis

All the things in it, and much more efficiently than before. A machine in hospitals gives the levels of some 12 chemicals in blood--sodium, calcium, potassium, and so on. But there's no such thing that will work for hundreds of proteins, really complex molecules. I'm going to try for 32 compounds at first. Then it should be no problem to do it for 64, or 128. That's as many things as anybody gives a darn about in blood. If it works, there's no reason it can't be done in a doctor's office. Commercially it would be enormous; you could take over all of clinical diagnostics, a 5-billion-dollar-a-year market. I have to figure out how to make it work or get someone else to. The concept is biologically valid, though technical problems need to be solved, just like with PCR.

This time I'm not going to hand it over to some company like Cetus without something saying it's mine. If anyone makes $300 million off it, I'm going to be part of that. Cetus has sold the rest of PCR they had not already sold for $300 million. This is the most money ever paid in history for a patent. You couldn't tell that from looking at my carpet in here that's not even clean. And it's not ermine, either!

OMNI

Couldn't they have given you a million dollars?

Mullis

I said,"Hey, why don't you make it $301 million? Send it in the mail; it will be good publicity for you?" But I'm going to do all right. You don't expect the world to take care of you. Absolutely no reason to think businessmen are going to behave like philosophers. Fair is not business, and business will grow, despite nasties and crummies in your tummy, as Dr. Seuss said. I don't have any resentment. You'd have to be neurotic to expect that business will be fair.

OMNI

Why didn't you charge Cetus to help defend the patent?

Mullis

I could have, but didn't realize it. When Du Pont started their suit, I could have demanded one million straight up front or go work for Du Pont. The lawyer told me this as we drove home from the courthouse after the verdict. You don't realize sometimes how important you are. I didn't take the case seriously. I didn't realize how fragile is the interaction between science and the law. A scientist has certain ways of thinking about how to establish proof that are not exactly the same as lawyers', and the response of the jury is a delicate kind of thing.

OMNI

How could you imagine believing PCR had been invented before you did invent it?

Mullis

When I told them in 1986 about PCR at the Symposium on Quantitative Biology held every June in Cold Spring Harbor--a big meeting of the elite in molecular biology--I got a standing ovation. It was an excited knot of people walking out of that room. Clearly something had been invented for the people who needed it. And nobody stood up to say, "That's cool, but we've known about it for years." It was, "Bravo! You have just changed our lives!" Even Khorana never said he'd invented it and hadn't told anybody about it. He'd have been an absolute fool in front of his colleagues if he'd said, "Sure we invented it in 1970; we just regretfully didn't publish it, and I'm sorry you guys had to wait 15 years for this little wimp from Cetus to come up with it."

The patent had my name on it. I would've felt real funny working for the other side, saying I really didn't invent it! I'd have had to say I didn't know it was invented before, or that all the elements were there, but I don't know why it wasn't invented before. Could have made a convincing case for the jury. But I knew I really had invented it for the first time. It is hard to invent it and forget it.

OMNI

When you become rich, will you work less hard?

Mullis

I will write. That's all. I am looking forward to sitting up in my place in Mendocino and spewing out all kinds of nonsense in my little computer. I've got a really nice science-fiction movie or two in my mind.

I've done a lot of fiction. The best fiction teaches people something of interest in a totally painless way. That's what I'd like to do more than anything else. But I don't want to be poor.

OMNI

What about your idea to sell celebrity DNA?

Mullis

A weird little thing, yes. We are going to try to use PCR to make certain pieces of DNA of celebrities. The idea is that teenagers might pay a little money to get a piece of jewelry, a bracelet, or whatever, containing the actual piece of amplified DNA of somebody like a rock star. We just have to get a little piece of skin, clip a nail or something from the person, prepare the DNA, copy it through PCR, and put it in a locket. You could say, "Here's a sequence from Mick Jagger." Something to do with his lips, say. The jewelry will look like something your Gypsy grandmother gave you, and in there will be a little speck of DNA.

OMNI

Who's to know that what you put in there isn't chicken DNA?

Mullis

The problem of authenticity is solved by the fact that I am Kary Mullis. Anyone from Taiwan who tried to pull it off might not get away with it. We will certify authenticity.

We'll also have a little collection of DNA of all primates showing evolution over the millennia, for the Nature Company, say. There are a lot of angles.

We've got a jeweler working on it now. You could stamp the DNA sequence on the jewelry, like stamping coins, so it looks really official. If we could get permission to use someone like Elvis Presley, we could do a gene of the month, and you could have a collection like stamps. We're doing this as fun. If we succeed, fine; if not, no one is going to go broke.

OMNI

How feasible is cloning a whole human today?

Mullis

If there weren't a law against it we could do it in an apartment. We'd have to get a woman who'd let us get some eggs. The mechanisms are fairly sophisticated on paper, but the actual manipulations you do are not hard at all. You just put a drop of this on that, do such-and-such, check for X, stick it back into some kind of thing, and grow it up. They make transgenetic mice now routinely. Anything you can do with a mouse, you can do with a person, if the person, the Catholic church, and the law allow it.

OMNI

Where do you see biology changing life most in the future?

Mullis

The biological weirdness will come in virtual reality. In the twenty-second century, people will experience realities so bizarre, we can hardly comprehend them. Once we crack the right code, we will take a little piece of something and lay it over your spinal cord, or maybe go in electronically through your finger to allow signals to be sent into your brain directly from a computer.

The machine will totally shut out all peripheral senses and take over, talking directly to your brain. You will sit at some console to control it. What you will experience will not be you anymore but a character in a videogame, and you'll really feel like it. Everything you ever hoped for you will now feel. In fact, you will be able to change into things you never imagined, because this setup will generate new kinds of sensations.

Many people will never go outside unless they have to. For entertainment, they will always get out their virtual reality equipment and dial up and visit a bar in Bangkok, say, or fly over Bangkok at 3,000 feet on their back. The images will include other individuals who can plug in and meet you in the bar, and you can have anything you want as your persona. You can come in as a perfectly functional male or female, or as an elephant. Some people such as the bartender will be paid to do this every day. He gets up, attaches himself to the system, and projects a bartender persona there. He may be a robot. Most customers won't be able to tell the difference: The reproduction will be so complete, you can see, touch, smell, and feel it. You can go home with somebody in the bar to some apartment in Bangkok and it will all be charged to your VISA card.

OMNI

How did you come to write your 1968 paper for Nature on the cosmological reversal of time?

Mullis

Courses in astrophysics gave me two competing descriptions of the universe: big-bang and steady-state theory. I thought both were sophomoric approaches, because each is based on the premise that you could actually step outside the universe and look at it. That's stupid. Relativity makes it clear you can't talk about the universe from outside. If you look at it from inside and assume relativity is part of what's going on--which neither theory did then--it makes sense.

I was at Berkeley and taking acid every week. That's what people did for entertainment: drink beer or go out into Tilden Park and take 500 micrograms of LSD and sit all day thinking about the universe, time going backwards and forward. Some mornings I'd wake up and think I ought to write that out. So I did.

Moving between fields is the way to be creative. Keep your fingers in a lot of pies. I do it because I'm curious. I'm the only person I know who goes into a poster session [at a scientific meeting] and stops at the first poster I have no idea what it's about. Find the poster you don't know anything about and look at it for a long time, and you might learn something totally different.



Home || Prime Time || Live Science || Machine Dreams || Project Open Book || SF-Fantasy-Horror
Continuum || Antimatter || Mind-Brain Lab || Interactive IQ || Gallery || OMNI Toons

Questions, comments and suggestions can be mailed to the webmaster.


Copyright (C) 1997 by Omni Publications International, Ltd. All Rights Reserved.