The swashbuckler

The year was 1980, and Mark Davis, PhD, fresh out of graduate school, was hurling himself athwart one of the day’s biggest biomedical mysteries. At his new job at the National Institutes of Health, he had the youthful audacity to try to resolve this paradox:

The entire human genome consists of a mere 25,000 genes or so, yet the number of different shapes those all-important immune warriors called T cells can recognize is on the order of billions. How can such a relatively tiny number of genes provide all the protein-making recipes necessary for T cells to be able to identify — and spearhead attacks on — invading pathogens and cancer cells?

Fueled by curiosity and a competitive spirit, Davis and a small team under his command solved the first part of the puzzle of T-cell diversity: They identified the gene for one of two protein chains that make up the surface receptor overwhelmingly responsible for recognizing all those pathogens and cancer cells. And they figured out how it works. Within a year or two, after Davis had joined the faculty at Stanford (where he is now director of the school’s Institute for Immunity, Transplantation and Infection), he and his teammates nailed the gene for the other protein chain.

With the T-cell receptor gene in hand, scientists can now routinely sort, scrutinize, categorize and utilize T cells to learn about the immune system and work toward improving human health. Without it, they’d be in the position of a person trying to recognize words by the shapes of their constituent letters instead of by phonetics.

Young and crazy

When Davis began at the NIH, molecular biology was in an embryonic state and lacked the sophisticated and user-friendly equipment (such as high-speed gene sequencers) that today’s graduate students take for granted. But Davis, who earned his PhD at the California Institute of Technology, was receptive to the new DNA discovery tools that were coming into use.

“I had the good fortune of being young and impressionable,” says Davis, who is now the Burt and Marion Avery Family Professor in Stanford’s Department of Microbiology and Immunology. His early-adopter attitude at CalTech in the late 1970s placed him at the forefront of an earlier discovery concerning how the genes coding for antibodies — produced by another set of immune-system heavyweights called B cells — frequently reshuffle themselves during cell division. This hugely increases the diversity of antibodies the B cells produce.

However, Japanese scientist Susumu Tonegawa, PhD, got to the finish line first and ended up winning the 1987 Nobel Prize in Physiology or Medicine for his effort.

It turns out that B cells had much to teach Davis — and the entire field of immunology — about T cells. T cells and B cells, working in concert, deserve much of the credit for the vertebrate immune system’s knack of carefully picking bad guys of various stripes out of the lineup and attacking them. Together (along with the various molecules they produce and secrete), they comprise the highly selective so-called adaptive immune system. (A more primitive, loose-cannon arm of the body’s defensive armory, called the innate immune system, resides in all multicellular animals.)

While quite similar in many respects, B cells and T cells are more like fraternal than identical twins. B cells are specialized to find strange cells and strange substances circulating in the blood and lymph. T cells are geared toward inspecting our own cells for signs of harboring a virus or becoming cancerous.

So it’s not surprising that the two cell types differ fundamentally in the ways they recognize their respective targets. B cells’ antibodies recognize the three-dimensional surfaces of molecules. T cells recognize one-dimensional sequences of protein snippets, called peptides, on cell surfaces. All proteins in use in a cell eventually get broken down into peptides, which are transported to the cell surface and displayed in molecular jewel cases that evolution has optimized for efficient inspection by patrolling T cells.

Somehow, our inventory of B cells generates antibodies capable of recognizing and binding to a seemingly infinite number of differently shaped biological objects. Likewise, our bodies’ T-cell populations can recognize and respond to a vast range of different peptide sequences.

So how does the T cell do it?

Ask a B cell.

Game on

By the time Davis began his NIH postdoctoral fellowship, in the lab of prominent immunologist William Paul, MD, the field had grasped how B cells manage to generate so many different types of antibodies. By studying the genes that produce antibodies, immunologists had learned not only that these genes contain a number of “mix and match” component sequences capable of rearranging themselves when B cells divide, but that some stretches along those genes are particularly susceptible to mutation. This further boosts the variety of antibody shapes in the resulting overall B-cell population, with a corresponding increase in the range of different pathogenic features antibodies can latch onto.

Although many early “T-cell receptor” sightings had turned out to be mirages, multiple labs had converged on a likely candidate protein that was composed of two separate pieces or subunits, designated the alpha and beta chains. There were still many doubters, but the protein work fueled an intensive hunt for the genes that gave rise to these two protein subunits.

“Earlier thinking had been that a T cell must use part of the antibody gene to produce its receptor for features of foreign or altered cells. But this had been shown not to be true,” Davis says.

Davis, at that time almost the only person in his immunology department at NIH making intensive use of the new molecular-biology techniques, had an idea for how to fish the gene for the T-cell receptor out of the sea of DNA that composes our chromosomes: Check for differences in activity between T and B cells’ genes by subtracting activity levels of each gene in one cell type from those of the corresponding gene in the other type. “This approach had mainly been used to look for differences between widely different cell types, for example liver versus kidney cells,” he says. “T cells and B cells are very similar in appearance, and they interact with each other. They seemed to have a lot in common. But T-cell receptor genes are active only in T cells. I thought looking at the differences between B and T cells might yield a relatively small number of genes to look at.”

He also knew the T-cell receptor, which clearly has to sit on the T cell’s surface membrane to be useful, would wind up being membrane-bound. And he suspected that its gene might be capable of rearranging itself, shuffling its component sequences during cell division like the genes coding for antibodies ­­— so it would have to have highly variable as well as fixed stretches.

Paul was intrigued by his approach and made Davis the informal head of a small lab. Stephen Hedrick, PhD, a fellow postdoc and novice molecular biologist, wanted to learn the new technology and so joined the enterprise. By then Davis had already shown that 98 percent of the genes in B and T cells were similar or the same. Focusing on the 2 percent that were different might be where the T-cell receptor genes lay.

One night at a party of some CalTech buddies who had also come to work at the NIH, Davis spotted a research paper on a bookshelf, took a look, and read of a technique that would enable him to restrict his search to only those genes coding for membrane-bound proteins. And he had already mastered an existing workhorse method for checking for rearranging genes.

By early 1983, they had found 10 promising T-cell genes that both were active in T cells but not B cells and coded for membrane-bound proteins. In nine of them, Davis and Hedrick saw no gene rearrangement. Things looked grim, but they didn’t give up, hanging their hopes on the last one. It was giving them a hard time, though. Determining whether it was prone to rearrangement took several months.

“We knew we were in a race against time,” says Davis. “Other groups were closing in. Plus, I’d been hired by Stanford but was dragging out my stay at NIH because we were getting so close.” The breakthrough came that July Fourth weekend when Hedrick was taking his family to the zoo and Davis and Yueh-Hsiu Chien, PhD, his wife and colleague (now a professor of microbiology and immunology at Stanford), were driving to his dad’s home in Connecticut. “Steve wanted to stop in at the lab to nurse the ongoing experiment. He left his family in the car for what was supposed to be a minute or two,” says Davis. “He checked on that last gene, and kept his family waiting in the car for a long time.” This candidate gene practically screamed, “Look, Ma, I’m rearranging.” Hedrick called Davis.

“I told my dad, ‘Something big has happened. I’ve got to get back to the lab.’” And he and Chien returned to Bethesda that night, driven by the excitement of discovery and the need to beat the competition to the punch.

Soon afterward, they moved to Stanford and Davis set up his new lab, with Chien playing a major organizing role. In August 1983, Davis flew to China and delivered talks in Beijing and Shanghai about his T-cell work that the attending Chinese scientists (who were just emerging from the Cultural Revolution) had difficulty grasping, because what he was doing was so different from classical immunology. But a week later he skipped over to Kyoto, Japan, and finagled an unscheduled talk before the biggest international conference in immunology: the once-every-three-years International Congress. His report there was a sensation. The audience understood that Davis and his colleagues had bagged the gene coding for one of the two protein subunits composing the T-cell receptor, later identified as the beta chain. They were halfway home.

A little over half a year later, Davis and his colleagues finished writing up their beta-chain gene discovery for Nature, publishing two articles in March 1984.

The pursuit of the alpha-chain gene, like that of the beta-chain gene, took place against fierce competition, including from MIT’s Tonegawa, who actually managed to sweet-talk Davis into sharing his protocol for isolating T-cell-specific genes.

This time, Davis had a leg up. “We had a postdoc, Nick Gascoigne, from England whose father told him that if we could get our manuscript on a plane to London, he would have his office assistant drive to Heathrow Airport, take it straight to Nature’s London office and plop it on the editor’s desk. We had seven papers published in Nature in 1984. We’d load the manuscripts onto DHL flights at 7 p.m. For the first ones, it was only six days from submission to acceptance.

“Meanwhile, we were getting all these invitations to Japan. So we went to one meeting, and Tonegawa was walking around, looking proud. ‘We have the alpha chain,’ he told us. ‘Do you?’

“I told him, ‘No.’ He was happy about that,” Davis recalls.

During his talk, Tonegawa showed a slide depicting a number of short horizontal lines. Those lines represented sections of genes, and their relative positions constituted what, in those dark days before the advent of easy gene sequencing, passed for a “fingerprint” of that gene’s structure. The gene depicted in this slide was the one Tonegawa was so sure was the alpha-chain gene. On the plane home, Chien told Davis, “You know, that kind of looks like one of the genes we’ve been analyzing.”

“So I said, ‘OK, let’s assume that one is the alpha-chain gene, and work night and day to finish figuring out its structure and how it works,’ Davis recalls. “And we did.” So much so, in fact, that some colleagues, having passed their lab one night and noticed that the lights were out, told them later, “We thought it was a power failure.”

“It turned out to be the alpha chain,” Davis says. “We wrote yet another paper in a rush, and did another Heathrow run. A week later I get a call from the Nature editor saying that the journal had just received Tonegawa’s manuscript, and that he was very unhappy that ours got there first.’ I thought: This is divine justice!”

Their isolation of the gene coding for the alpha chain of the T-cell receptor was published in Nature that autumn. Tonegawa was quite magnanimous in defeat, says Davis.

Don’t look back

Since those knock-down, drag-out days, Davis has made a number of important contributions to immunology — for example, at Stanford in the 1990s he developed a breakthrough methodology that can determine, from a blood sample, whether a given individual’s immune system has previously encountered a particular molecular shape (characteristic, say, of a pathogen or tumor).

But in recent years, Davis has broadened his horizons. Notably, he has been grappling with the immunology of human beings. A huge amount of immunological research over the past 50 years has been conducted on mice. “We’ve cured cancer and autoimmune disease in mice a whole bunch of times,” he says. But mice and humans diverged from a common ancestor some 60 million years ago, and a mouse’s immune system can tell us only so much about our own.

Getting a handle on how real, live people’s immune systems respond requires research on real, live people. To this end, Davis was instrumental in securing tens of millions of dollars to establish a Stanford center dedicated to measuring thousands of variables in human blood samples in order to better characterize the human immune system in all its complexity.

“T-cell recognition is a huge part of immunity,” he acknowledges. “Knowing what T cells recognize and when they recognize them is crucial. But the problems of human immunity are much bigger than T-cell receptors. I’ve sunk 30 years into T-cell receptors. But I’m thinking, is that all I want to do? I think I’ve still got a few more good experiments left in me.”