Since starting her own laboratory at The Scripps Research Institute in 1991, immunologist Wendy Havran has been searching for the answer to a single question: What activates gamma-delta T cells? These immune cells make up a small proportion of the T cells in blood and lymphoid organs but are abundant in body barrier tissues, residing permanently in the skin of mice and humans. They act as rapid responders, recognizing tissue damage and secreting growth factors and other signaling molecules that alert...
“During resting conditions when there is no damage, the skin gamma-delta T cells . . . have these dendrites that they extend and retract, touching their epithelial cell neighbors to survey for any damage or disease,” Havran explains. When there is damage, the T cells gather, migrate to the damaged site, and begin to repair the tissue. Identifying what co-receptors bound to these T-cell receptors sense when a wound occurs has been a major focus of Havran’s lab. “My first grant from the NIH was to look for antigens that would activate the skin’s gamma-delta T cells. Several other groups have actively looked for these too, and so far, no one has been able to find them,” Havran says. But she’s not going to quit looking.
Havran was born in Houston, Texas, in 1955. Her father was an engineer, her mother an elementary school teacher. Her parents often took her and her two younger sisters to the Houston Museum of Natural Science and several US National Parks, where the family would camp for several weeks each summer. Her parents also took her and her sisters to the Houston Grand Opera. “Music continues to be an important part of my life,” Havran says.
She entered Duke University as an undergraduate in 1973 assuming she would pursue a career in medicine. At the beginning of her sophomore year, her advisor encouraged her to do research, and not long after that she joined the lab of hematologist Gerald Logue. “The lab and the department was mostly MDs and MD/PhDs, and they really adopted me and decided that I should have exposure to the medicine side as well. They would come grab me and take me along to see interesting patient cases and had me attend their weekly clinical and research meetings. The research meeting was at eight in the morning on Saturday, prime sleep time for an undergrad, but I was dedicated and showed up every Saturday,” Havran recalls.
When she graduated from Duke in 1977, Havran stayed in Logue’s lab as a technician, trying to decide if she wanted to pursue a research career.
A new assistant immunology professor, John Cambier, moved in next door to Logue’s lab, and the two labs began a collaboration headed up by Havran, who coordinated joint experiments between Cambier’s and Logue’s teams. It was Havran’s first exposure to immunology and led to her first publication.
“It just clicked, and there was no going back,” she says. “I wanted to understand how the immune system worked.” The Duke Comprehensive Cancer Center received funding for a fluorescence-activated cell sorter (FACS), then a relatively new device, for categorizing populations of cells, distinguishing cell subpopulations in a clinical or research sample, and diagnosing blood disorders. It was the first and only such machine in the southeastern part of the US at the time.
Cambier supervised the FACS and recruited Havran to operate the machine, overseeing the experiments of others in the university and helping them analyze their FACS data. She did some of her own research as well, characterizing the functional biomarkers found on B-cell populations to distinguish mature from immature cells.
Immersed in immunology
After two years, Havran was ready to focus solely on her own research. She applied to only one graduate program, at the University of Chicago, and chose to work in Frank Fitch’s T-cell immunology lab, which was the first to isolate T cells and make T-cell clones that could propagate in cell culture for months. Fitch was also an expert in making monoclonal antibodies, an invaluable skill that Havran learned directly from him.
As a graduate student, Havran worked to determine the roles of CD4 and CD8 surface markers on T cells by fusing mouse T-cell clones that expressed each marker and selecting the clones that expressed both. At the time, Havran was the only woman in the lab, but that did not bother her as much as the Chicago winters. “I defended after four years, in November, because I wanted to get out before the next winter started,” she says. In her first Nature paper, published in January 1987, she and Fitch showed that only the CD8-expressing cells caused damaged or cancerous cells to disintegrate when the T-cell receptors bound antigens.
Not long before the paper came out, Havran moved to James Allison’s lab to do a postdoc. She had gotten to know Allison as a graduate student, having spent a week in his lab—then at one of MD Anderson Cancer Center’s sites outside of Austin, Texas—learning how to biochemically analyze the antibodies she made against T-cell receptors. She reconnected with Allison at an American Association of Immunologists meeting in 1986, and he invited her to come visit the lab, which had recently moved to Berkeley, California.
These cells were unique because T cells typically each express a unique T-cell receptor that recognizes a unique antigen, but these cells all expressed the same T-cell receptor, so they were basically clones.
“I had been really decisive in my career—I only applied to Duke for college, and then only a single lab for both graduate school and my postdoc,” says Havran. She moved to California in 1986—at a time when T-cell immunology was booming, she recalls. She collaborated with Roger Tsien, who would win a Nobel Prize in 2008, to find out if lymphocytes in the thymus gland that co-express CD4 and CD8, which scientists thought were non-active cells on their way to dying, had any immune function. Tsien had developed calcium indicator dyes, and Havran tested whether the cells’ calcium levels could change, which would indicate that they were serving some immune purpose. Rather than dying cells, she found that the CD8 and CD4-expressing T cells were a functional, intermediate cell type that can undergo a selection process to develop into mature T cells.
During her postdoc, Havran also started studying gamma-delta T cells, which Allison and his colleagues had first described in 1986. A group of dermatologists at UT Southwestern had discovered a population of immune cells expressing the Thy1 protein marker with dendritic cell properties in the mouse epidermis and wanted Allison’s lab to determine if they were T cells or not. Havran characterized the cells, showing that they were in fact T cells, but with unique gamma and delta chains, rather than the typical alpha and beta chains found in T-cell populations circulating in the blood, spleen, and lymph nodes. The gamma- and delta-chained cells had previously been missed because they represent only 1 to 2 percent of lymphoid T cells. But, Havran notes, gamma-delta T cells are “actually very abundant in epithelial barrier tissues such as the skin and intestine, where they can be over 50 percent of the resident T cells.”
The work, which required Havran to generate one of the first antibodies against the gamma-delta T-cell receptor, was published in Nature in 1990. The cells, Havran found, migrate from the embryo’s developing thymus to the skin, where they take up residence and remain throughout life. “Dermatologists were convinced that there were no T cells in the skin, so this finding was very unexpected,” Havran says. “These cells were unique because T cells typically each express a unique T-cell receptor that recognizes a unique antigen, but these cells all expressed the same T-cell receptor, so they were basically clones.”
Havran checked the antibody results with the still relatively new PCR technology to confirm a lack of diversity in the skin gamma-delta T-cell receptor.
But even with this evidence, the scientific community was not convinced. “A lot of people didn’t believe our work when we published it. There was a dermatology group that did similar studies and found, in contrast to our results, that this T-cell population had diverse sets of T-cell receptors. They would stand up when I presented my results and tell me that I was wrong.”
But Havran’s careful work was eventually shown to be correct: The other group ended up discovering that their samples were contaminated with cells from the dermis and peripheral blood, which is why the researchers were detecting a diverse set of receptors.
Migrating to independence
As a postdoc, Havran had received a prestigious Lucille P. Markey Scholar in Biomedical Science grant, which provided funding for the last two years of her postdoc, as well as five years of funding in her first faculty position. Havran used this grant, which had only been in existence for 15 years at the time and is awarded to only eight PhDs and eight MD/PhDs in the US every year, to start her own lab in 1991 at Scripps, where she focused on gamma-delta T cells.
She hit the ground running, doing many of the experiments herself. She had read a Nature paper describing a novel growth factor found in fibroblasts called keratinocyte growth factor (KGF). It was essential for tissue repair in the skin. In a series of experiments, Havran and then-postdoc Richard Boismenu reported in Science that this growth factor is upregulated by gamma-delta T cells in the mouse skin upon injury to the epidermis. “This ended up being a foundational experiment for the lab because it was the first suggestion that these skin resident T cells contributed to wound healing,” Havran explains.
Then in 2002, in another Science paper, Havran, then-postdoc Julie Jameson, and a team of scientists established a role for these skin-resident immune cells in wound healing. They found that wound healing was slowed in mice that lacked gamma-delta T cells in the skin, and the delay was at least in part due to a lack of KGF expression. The lab has also investigated gamma-delta T cells in the intestine, where the cells protect the intestinal lining through a similar production of KGF.
Following up on the mouse work, a visiting dermatologist from France, Antoine Toulon, set up a collaboration with plastic surgeons, and, using their non-diseased and chronic and acute wound skin samples, showed for the first time, that human skin-resident gamma-delta T cells also function in wound healing.
Researchers had previously shown that receptors on the surface of alpha-beta T cells detect foreign microbial peptides when they are linked to major histocompatibility complex (MHC) molecules on the surface of infected cells. But gamma-delta T cells don’t recognize MHC-linked peptides, nor do they express typical cell surface glycoprotein T-cell co-receptors such as CD4, CD8, or CD28, which facilitate the recognition of foreign peptides, as alpha-beta T cells do.
In 2011, Havran’s lab made a soluble version of the mouse skin gamma-delta T-cell receptor—a T-cell receptor tetramer linked to a fluorescent marker that can be easily detected using microscopy and that should bind to the unknown activating ligand. The team found that the mysterious ligands for the T-cell receptor are undetectable in resting, undamaged tissue, yet are rapidly expressed by keratinocytes in mice at the site of a wound in the epidermis.
“Using microscopy, we could see the T cells being activated,” Havran says. “We thought that we could give the soluble receptor complexed with the ligand to the mass spectrometry facility, and they would tell us what the ligand was.” Based on the mass spectrometry analysis, the ligand doesn’t appear to be a peptide, even though “it’s been the dogma that T cells recognize peptide antigens,” Havran notes. “Since these experiments, we have changed our view completely. Alpha-beta T cells recognize foreign peptides from bacteria and viruses, but skin gamma-delta T cells seem to function differently.”
Havran’s working hypothesis is that the ligand is a molecule that is not normally expressed by skin cells but becomes upregulated during damage and is perhaps shuttled to the cell surface. “The molecule might be a glycosylated protein or a sugar that is a stress signal that doesn’t need to be rapidly processed by antigen-presenting cells like in the case of alpha-beta T cells,” she explains.
Researchers in her lab have identified other molecules that are recognized by the cells’ receptors, including a costimulatory junctional adhesion molecule-like (JAML) protein, which spurs these skin immune cells to facilitate wound healing. The lab has also identified gamma-delta T cells in the intestine, where the cells protect the intestinal lining from physical damage. Havran has plans to translate the work to the clinic to facilitate wound healing. One way to do this would be to use the JAML molecule as part of a degradable hydrogel system to stimulate skin healing. In the meantime, Havran and her colleagues continue to investigate how gamma-delta T cells work, in hopes the answers may also offer hope for healing chronic skin wounds.