In the Blood

That dedication netted him more than half a dozen publications, mostly in journals like the Journal of Biological Chemistry and Blood. And it carried Pieters into his current exploration of host–pathogen interactions, where he has identified several factors, both bacterial and human, that allow mycobacteria—which causes tuberculosis—to survive inside infected macrophages. This work on the molecular mechanisms of mycobacterial invasion constitutes “a very fundamental finding in cell biology,” says Bernhard Dobberstein of the University of Heidelberg, who was Pieters’s postdoctoral mentor at the European Molecular Biology Laboratory (EMBL). And it has spawned a second wave of publications, many in Nature, Science, and Cell.

After completing his bloody graduate work, Pieters headed to the EMBL in Heidelberg. There he joined Dobberstein, who was studying how Class II major histocompatibility complex (MHC) molecules make their way from the endoplasmic reticulum, where they’re synthesized, to the cell surface, where they present antigenic peptides (derived from, say, invading bacteria) to the immune system for recognition. These peptides can come from marauding microbes that are engulfed by immune cells and degraded in lysosomes; the resulting fragments get picked up by MHC molecules and carried to the plasma membrane, where they can alert passing T cells to the infection. “What wasn’t known,” says Pieters, “is how the MHCs end up capturing these peptides.”

So he and his colleagues set out to determine how and where these molecules meet. “He was so patient in dissecting the signal,” says Dobberstein. And he determined how a pair of amino acids on the so-called invariant chain of the MHC Class II complex regulates its intracellular transport. But that left the question of where these MHCs actually get loaded. Pieters tried to purify these loading stations. “But we never got anywhere,” he says. So Pieters brought the problem with him when he joined the staff at the Netherlands Cancer Institute in 1992. There, a colleague showed Pieters an electrophoresis device he’d built to separate different leukocytes from a sample of rabbit blood. Pieters decided to run his samples on the machine, and found that he could isolate pure, Class II–containing organelles. “We could show that MHC Class II molecules are transported there and that the antigens are loaded there,” he says. “So this compartment was where it happens”—results he published in Nature in 1994.

Then his research took an unexpected change in direction. Pieters moved his lab to the Basel Institute for Immunology (BII) in Switzerland, an outpost funded by Hoffmann–La Roche. His plan was to continue to dissect the components required to place peptides into MHC molecules and get them to the cell surface. “But so many labs were working on that,” says Pieters, that he started up a side project with a colleague who was working on using mycobacteria as a vaccine vector. Weakened strains of M. bovis are used to generate immunity to tuberculosis, so the researchers were thinking that if the bacteria could be genetically engineered to deliver additional peptides from HIV or other viruses, they might be able to act as a sort of “super vaccine,” producing immunity to TB and other diseases at the same time. The investigators would use Pieters’s experimental system to track the degradation and presentation of these recombinant bacteria and their antigenic payload.

But the antigens, it turned out, were never delivered. That’s because mycobacteria have figured out how to inhibit host-cell digestion—a trick that enables them to survive inside the macrophages that engulf them. There, the bacteria take up residence inside organelles called phagosomes, preventing these entry-level compartments from maturing into the lysosomes that would spell the invading bacteria’s doom. Pieters decided to find out how mycobacteria make that happen. “The clue was that this only works if you start out with live mycobacteria,” he says. “If you use dead mycobacteria, they immediately go to lysosomes where they are degraded.”

Pieters thus fed macrophages either live mycobacteria or dead ones. Then he separated their proteins on gels—over and over again—and searched for any discernable differences. “There was really only one protein that consistently showed up in fractions where we had live mycobacteria,” says Pieters, who dubbed the resulting molecule tryptophan and aspartate-containing coat protein (TACO). When macrophages are infected with live mycobacteria, TACO gets recruited to the site of uptake and adheres to the bacteria-containing phagosome. That TACO coat somehow keeps the phagosome from becoming a lysosome.

“I’m still stunned by the simplicity of the discovery,” says Fritz Melchers of the University of Basel, former director of the BII. Pieters had turned the observation that live and dead mycobacteria are processed differently into the identification of a host protein that the bacteria co-opt to promote their own survival. “It’s the simple discoveries that are usually the paradigm breakers,” he says. “They don’t happen too often in life, but we are blessed when they do.”

But how does TACO do its thing? Many insisted it had something to do with actin. That’s because TACO was found to be homologous to a protein in Dictyostelium called coronin, which was believed to modulate the assembly of actin filaments. To find out whether TACO—which was renamed coronin 1—also abets mycobacterial survival via a manipulation of actin, Pieters attempted to knock out the gene in mice. Which is when Hoffmann–La Roche announced the closing of the BII. “And the first people to leave were the mouse guys,” says Pieters.

In 2002, Pieters brought the project to the University of Basel’s Biozentrum, where he is currently a professor. It took another few years to generate a knockout mouse—the first made at the University of Basel facility. To their surprise, mice lacking coronin 1 looked absolutely fine. “They were fertile. They had no visible defects”—certainly nothing that could be attributed to problems with actin, says Pieters.

Around the same time, the group also succeeded in producing a line of macrophages lacking coronin 1. They, too, had no obvious flaws. “They were motile, they phagocytosed normally. Everything related to actin was normal,” says Pieters. But pathogenic mycobacteria taken up by coronin 1–depleted macrophages were quickly degraded—confirming that the bacteria need coronin 1 to stay alive. “We thought: this is great! We wrote up the paper, submitted it to several journals, and the general response was, ‘You must be doing something wrong. Because everybody knows that coronin 1 modulates F-actin, ‘” says Pieters. Without coronin 1, there must be something wrong with actin—or so the reviewers seemed to think.

The researchers looked at actin every way they could—and still couldn’t come up with a defect. But in looking more closely at their knockout mice, what they did discover is that the animals had no circulating T cells. It turns out that the loss of coronin 1 had destroyed the cells’ calcium-based signal transduction—a stimulatory system that T cells need to survive. “So this protein initially thought to be needed for bacterial survival is also needed for the survival of naïve T cells,” says postdoc Rajesh Jayachandran. Why this should be the case—and whether T-cell maintenance somehow plays into mycobacterial infection—is not known. But the findings took Pieters in yet another new direction. “I spent six years at an institute for immunology, but I never fully understood why people got so excited about T cells,” says Pieters. “But when we saw the depletion, I realized we had to try to understand it.”

Pieters is well known for following the findings wherever they lead. “I think Jean has twice switched the focus of the lab over the past 15 years, and has made significant contributions, with good papers, to each of these fields,” says former student and postdoc John Gatfield of Actelion Pharmaceuticals. “There are people who work on B cells their whole lives. But Jean takes the opportunity to try new things and he’s not deterred by never having done something before.”

But the T-cell findings also shed light on the mycobacterial work. Pieters was able to confirm that calcium signaling is also critical for mycobacterial survival: macrophages that lack coronin 1 and kill off their mycobacterial guests also show disrupted calcium-dependent signal transduction—results ultimately published in Cell in 2007. Coming up with a viable mechanism ultimately quelled the actin critiques, and the T-cell data—which bounced from journal to journal over the course of 2 years—finally appeared in Nature Immunology in 2008. “We spent that time revising and revising, and adding more data,” says former student Philipp Mueller. “Working through that was tough, but in the end we gained a lot of insight and the story was a better one for all the revisions.”

“All in all, it’s been a good balance between excitement and disappointment,” laughs Pieters, who continues to explore the mechanisms by which coronin 1 helps keep both mycobacteria and T cells alive. “At one point it was very difficult to motivate my lab to keep working on coronin 1—and I almost gave up myself,” he says. “But it’s been great fun to work on that project and to come up with things that we absolutely didn’t expect.”

And bringing fresh insights to the study of TB is a service in itself. “Mycobacteriology is a century-old field that’s rich in dogma, and I like that Jean is not afraid to go against the grain, to do what he thinks is interesting,” says HHMI investigator William Jacobs of the Albert Einstein College of Medicine. “He’s a very solid, very imaginative, very prolific cell biologist. At the end of the day, I’m very happy he’s in the field.”