The eccentric scientist behind the ‘gold standard’ COVID-19 test

Bombastic biochemist Kary Mullis invented PCR, a tool that redefined genetic science, while driving in 1983. That was only the beginning.

Published 3 Mar 2021, 13:33 GMT
Photograph by Jim Wilson, The New York Times/Redux

Biochemist Kary Mullis says he was driving from the San Francisco's Bay Area to his cabin in Mendocino in 1983 when suddenly, like a bolt of lightning out of the California sky, he came up with a way to pinpoint a particular stretch of DNA and synthesise an enormous amount of copies.

“The simple technique would make as many copies as I wanted of any DNA sequence I chose, and everybody on Earth who cared about DNA would want to use it,” Mullis recounts in his colorfully titled 1998 memoir Dancing Naked in the Mind Field. “It would spread into every biology lab in the world. I would be famous. I would get the Nobel Prize.”

Mullis did indeed win the 1993 Nobel Prize in chemistry for inventing polymerase chain reaction, or PCR. These three letters have recently shot into the public consciousness because PCR is the basis of the most common, gold standard tests for the SARS-CoV-2 coronavirus. But that’s only the latest game-changing use of PCR. Since its debut, it’s been applied to tasks ranging from helping decode the human genome to saving coral reefs.

“If you're doing any sort of DNA studies, PCR is just the thing you do,” says pioneering genomics researcher Eric Green. “It's almost like saying, How do you use electricity?”

Mullis, who died in August 2019, recounts the origins of PCR in his memoir as the story of a larger-than-life genius who goes it alone and single-handedly invents a tool that kickstarts a new age of biology. But while it’s true Mullis had the original “eureka!” moment, there’s a lot more to the real history of PCR, including the other scientists who helped shape it into a biological powerhouse—sometimes in spite of Mullis’s difficult temperament.

Before PCR, studying DNA was tough. Lots of genetic information is packed into DNA molecules and isolating exactly the right small snippet to study was tricky. Even if a scientist could isolate a section of interest, the amount of material was often so minuscule that there just wasn’t much available for experiments.

To get around this, the state-of-the-art in the 1980s was DNA cloning. In this process, scientists put their desired genetic sequence into the genomes of bacteria, which then divided and replicated both themselves and the introduced genetic code. It’s a powerful but laborious process, which is why something simpler and faster would be such a windfall.

Genetics 101
What is a genome, and how are traits passed from generation to generation? Learn how pea plants helped launch the study of genetics and how the field of genetics research has evolved over time.

After that fateful weekend in his cabin, Mullis returned to work at Cetus Corporation in Emeryville, California. Cetus was one of the first biotechnology companies in the world, and the culture at the time was closer to what you might find today at a tech startup in Silicon Valley. There, various teams were playing with exciting new tools to clone genes and express proteins that could be used for medical applications.

Mullis’ main job at Cetus was to make little strips of genetic material for other scientists in the company to use in their experiments. While his bombastic personality led to personal conflicts with his colleagues, including fisticuffs on one occasion, his work was useful to Cetus.

It was “an unusual group of young scientists, and they tolerated Kary,” says Paul Rabinow, an anthropologist at the University of California, Berkeley, and author of Making PCR: A Story of Biotechnology.

In Rabinow’s telling, Mullis brought the PCR idea to his colleagues. The process was elegant and simple: Heat a DNA molecule to separate the double helix into two strands, and use each strand as a template for making a copy—much like how DNA unspools and copies itself inside our cells. Then you let the sample cool; this would normally cause the two DNA strands to click back into place, but you can hijack the process with lots of short stretches of DNA called primers—just the type of genetic fragments Mullis was working with for other projects.

Easily synthesised in a lab, these primers are designed to click on next to the targeted section of DNA and prevent the two original strands from coming back together. The places on the DNA strands where the primers attach then serve as landing pads for an enzyme called DNA polymerase. It marches down the exposed strand, snapping DNA building blocks known as nucleotides into the correct positions to reconstruct the complementary strands.

If you start with just one piece of DNA, you’ll have two copies of your target sequence after one PCR cycle. Each copy can again be unwound to make more templates. After just 30 cycles, you’ll have over a billion copies—all from one molecule of DNA.

Mullis was known for his eccentric ideas, many of which had basic biology mistakes according to his colleagues, so people initially either didn’t think it would work or didn’t care. But Mullis kept tinkering with the idea, and the following year he was able to bring them some experimental data that seemed to show the chain reaction was working. This caught the attention of several Cetus colleagues, especially biochemist Thomas White.

“I thought, Hmm, it could be bullshit, but he might actually be right,” White says. “And if he is right, it'll transform what we're trying to do here.”

White had had a soft spot for Mullis ever since they’d became close friends in graduate school at UC Berkeley. Mullis helped White rebuild his car engine and ordained White as a Universal Life minister. White returned the favour by presiding over Mullis’s wedding to his second wife. White had recruited Mullis to work at Cetus and ended up being his boss, helping diffuse tensions when Mullis’s ego would grate on coworkers. 

White asked Mullis to focus exclusively on getting PCR to work. By the end of 1984, White and other company leaders still didn’t think he had enough evidence, so the company kept adding experimental scientists to parallel his efforts. The skilled work of many colleagues—in particular Stephen Scharf, Fred Faloona, and Randall Saiki—finally yielded enough replicable data to declare PCR a success.

With the proof-of-concept demonstrated, getting a publication and, eventually, a patent became top priority. But Mullis kept putting off writing the paper. People had doubted him, White says, and procrastinating on the paper was his revenge. Frustrated by the wait, Saiki co-authored a 1985 paper in the journal Science about a test for sickle cell anaemia that included the first published description of PCR. However, that paper only hinted at its power as a standalone technique.

White pleaded with Mullis to finish his paper explaining PCR in detail, and Mullis eventually did and submitted it to Nature. It was rejected. Science passed on it as well. It ended up being published in 1987 in Methods in Enzymology.

“There is just no way the Human Genome Project could have been successful without PCR.”

Eric Green

By then, Mullis had left Cetus, aggrieved chiefly by the fact he wasn't the first author on the more prestigious Science paper. In parting, Cetus paid Mullis a few months’ salary and a £7,000 bonus, the largest the company had ever doled out to a scientist for an invention. Cetus retained the rights to the technology, and from then on out, Mullis’s contribution to the development of PCR was mostly popularising it—and himself—at invited speaking and consulting gigs.

In his lifetime, Mullis also denied that HIV causes AIDS, questioned human influence on climate change, gave talks featuring images of nude women, and made sexist remarks to journalists. White still reminisces about his unquestionable creativity, sharp wit, and good humour—but laments how the myth took over the man. “Mullis rejected all of his former friends and colleagues and just disparaged us,” he says. “The Nobel Prize went to his head.”

Even before Mullis left, other Cetus team members were working to make PCR truly lab-ready. Two problems still made the process clunky to perform. For starters, the heat necessary to perform a cycle was degrading that all-important DNA polymerase, the piece required to construct each DNA copy.

Before leaving, Mullis had proposed a solution: Use a polymerase from the microbes discovered in the boiling-hot pools of Yellowstone National Park. The thinking was that if these organisms can live and replicate at high temperatures, their DNA polymerases must be able to tolerate such extremes. So David Gelfand, another Cetus scientist, flew out to Wisconsin to meet microbiologist Thomas Brock. In the late 1960s, Brock had isolated a species of heat-loving bacteria named Thermus aquaticus from Yellowstone’s thermal pools. That species’ unique DNA polymerase ended up being exactly what was needed.

Meanwhile, Shirley Kwok, a scientist at Cetus, was applying PCR to study HIV but was getting annoyed with the process. Cycling the sample through different temperature regimes by hand was mind-numbingly tedious, and in her case, the work had to be done in a biocontainment facility wearing full personal protective equipment. That’s when a technician named Robert Watson modified a small in-house pipetting robot to manage the thermal cycling. Today, automated thermal cyclers based on the idea are standard in genetics laboratories around the world.

By the late 1980s, PCR was making waves in the scientific community, and in 1991, Cetus sold the PCR rights to what is now the healthcare giant Roche for £215 million. White ended up running the PCR division there, along with over a hundred Cetus scientists he took with him.

Since then, PCR usage has multiplied exponentially, with numerous adaptations for various applications. Medical diagnosis, forensics, food safety, crop development, even the search for the origin of humanity—the boundaries of all these fields and more were busted wide open with the power of PCR.

Genomics researcher Eric Green was finishing up an M.D.-Ph.D. at Washington University in St. Louis in the late 1980s when he first heard of PCR technology. A few years later, he figured out how to use PCR to map the human genome. He was soon tapped to do just that as part of the government-backed Human Genome Project.

“There is just no way the Human Genome Project could have been successful without PCR,” says Green, who is now the head of the National Human Genome Research Institute.

And of course, many of the COVID-19 tests being conducted today use PCR to amplify bits of the genetic code of the SARS-CoV-2 virus from swabbed samples, allowing the tests to detect its presence.

A particularly exciting path forward is simplifying the hardware so PCR can be used outside a laboratory. “These machines, really all they do is heat and cool a sample,” says geneticist and self-professed tinkerer Ezequiel Alvarez Saavedra. “So I figured people don't need to pay three [thousand], five thousand dollars or even more to get this.”

Together with neuroscientist Sebastian Kraves, Alvarez Saavedra started a company called miniPCR bio to create something simpler and cheaper. The key innovation, Alvarez Saavedra says, was switching the heating element from thermoelectric semiconductors to copper wires, similar to the lines that defrost a car windshield. This made the whole construction simpler and more energy efficient. Now you can order a PCR machine housed in a transparent container the size of a tissue box for less than a thousand dollars.

Alvarez Saavedra says that many of his customers are educators wanting to showcase the beauty of biology to students. PCR is “very simple, once someone shows it to you,” he says. “That’s the beauty—it’s very easy to understand.”

Amy Apprill, a marine microbial ecologist at the Woods Hole Oceanographic Institution, spends a lot of time in the U.S. Virgin Islands studying stony coral tissue loss disease, which leaves behind skeletons that look like bleached coral reefs. This devastating disease was first identified off the coast of Miami in 2014 and has since spread rapidly into the Caribbean. The cause has not been identified, but it could be some kind of bacteria.

“Because we’re looking at bacteria, which make up a really small component by biomass, we really need to rely on PCR,” Apprill says. PCR makes abundance out of scarcity, and scaling down the machine lets Apprill study her microbial harvest at a nearby AirBnB.

“It makes it all portable, and for us that's key,” Apprill says. “You just can’t afford to fly your whole lab to all your different field projects.”

And in 2015, miniPCR bio partnered with Boeing to organise a competition that would allow students to perform a DNA experiment in space using the miniaturised PCR machine. Anna-Sophia Boguraev, who was in high school at the time, entered a proposal and won, setting up the first PCR run in space as a proof of concept for future research.

“Since then, it's been used hundreds of times,” says Boguraev, who is currently in the Harvard/MIT M.D.-Ph.D. program. “The age of molecular biology in space has only accelerated.”

Today it’s clear that, despite their contentious relationships, Mullis and his colleagues made an amazing contribution to science—one that will likely inspire generations of researchers for decades to come.

Read More

Explore Nat Geo

  • Animals
  • Environment
  • History & Culture
  • Science
  • Travel
  • Photography
  • Space
  • Adventure
  • Video

About us


  • Magazines
  • Newsletter
  • Disney+

Follow us

Copyright © 1996-2015 National Geographic Society. Copyright © 2015-2016 National Geographic Partners, LLC. All rights reserved