Panning for T-cell Gold

A COMPLICATED UNION:Image redrawn from Ann Rev Biochem, 72:717–42, 2003Antigens bound by major histocompatibility complex (MHC) molecules interact with T-cell receptors (TCRs) at the heart of the interface for T-cell recognition. But, many other players are involved. Massive polyvalency within this confined space may compensate for low-affinity receptors, giving added sensitivity.In some ways, science resembles gold mining: Eager practitioners congregate around rich veins of discovery whil

By | July 19, 2004


Image redrawn from Ann Rev Biochem, 72:717–42, 2003

Antigens bound by major histocompatibility complex (MHC) molecules interact with T-cell receptors (TCRs) at the heart of the interface for T-cell recognition. But, many other players are involved. Massive polyvalency within this confined space may compensate for low-affinity receptors, giving added sensitivity.

In some ways, science resembles gold mining: Eager practitioners congregate around rich veins of discovery while leaving less-promising territory nearly deserted. Yet even neglected are as can produce a mother lode.

So it has been with cell-cell recognition. Despite its importance in almost every aspect of development, organ function, and in the hematopoietic and nervous systems, the molecular basis of cellular recognition has failed to incite a scientific gold rush, largely due to the daunting technical difficulties in working with membranes, membrane proteins, and their cytoskeletal underpinnings.

Scientific prospectors, though, are migrating to the field in ever-greater numbers, thanks largely to a spate of new technologies, particularly video-fluorescence and multiphoton microscopy. These new tools have reinvigorated efforts to understand the molecular basis of immune-cell recognition and are illuminating some surprising new phenomena. They are also freeing us from the shackles of "population biochemistry," which assumes that all cells in a dish will behave exactly the same way in response to a particular stimulus.

We now know that, even though immune-cell recognition proteins are low-affinity receptors, they are very sensitive, requiring just one or a few antigens to jumpstart a response. For such a specialized system to work, a highly organized multiprotein space must be established at the cell-cell interface. Further work just may allow us to capture it all on videotape.


My lab focuses on immune recognition by T lymphocytes, those key mediators of cellular immunity without whom very little happens in terms of an antigen-specific immune response. The two main functions of T cells are to reward B cells that have made antibodies to the "correct" antigens by secreting growth and differentiation factors, and to punish "evildoers," the cells that are expressing aberrant or foreign antigens. Different types of T cells carry out these functions – CD4+ helpers reward, and CD8+ killers punish – but both continually probe the surfaces of many normal cells for the particular foreign peptide-major histocompatibility complex molecule (MHC) that a given T cell will recognize.


Courtesy of Mark Davis

T-cell recognition once was thought to involve just the T-cell receptor (TCR) and an antigen-MHC complex on an antigen-presenting cell (APC). Now we know that many molecules influence T-cell recognition (see figure), thus upholding Delbruck's law, which says that no matter how complicated you might think a biological phenomenon is, it always turns out to be more so. Just as eukaryotic transcription has exploded from simplistic regulatory models à la the lac operon to highly choreographed productions involving scores of proteins, cellular recognition seems to require dozens of cell-surface molecules and probably many more cytoplasmic players operating in a highly organized two-dimensional space known as the immunological synapse.

One clear characteristic of these surface molecules is that they have strikingly weaker affinities (1–100 micromolar) than receptors that snatch ligands from three-dimensional environments (in the low nanomolar to picomolar range).1 And yet these surface molecules make no apparent sacrifice in specificity, being just as sensitive to minor ligand changes as much higher-affinity molecules are.

This illustrates the unique environment in which surface molecules operate, where diffusion is largely confined to two dimensions and massive polyvalency and confined space offset low affinity. It also seems important for transient cellular interactions to employ molecular bonds that can easily be broken, as demonstrated by Deborah Leckband and colleagues,2 who showed that multiple cell-surface interactions stronger than about 10 micromolar could not be disrupted without damaging vesicle membranes.


It has long been asserted that T cells are very sensitive to antigen, but published reports estimate that anywhere from one to 400 peptides per APC are necessary to fully activate at least some cells. Recently my laboratory developed a procedure to visualize individually labeled peptides bound to MHC molecules on cell surfaces and thus determine exactly how many ligands a T cell was in contact with, and then using video microscopy, follow the response.34 Surprisingly, we found that four different T-cell lines, representing both helper and killer cells, were all able to detect even one molecule of antigen-MHC, though it took about 10 peptides to elicit maximal calcium release and form a stable synapse.

This finding could not have been made using population-based assays. Large numbers of cells, even in a clonal population, won't respond to the same stimulus in exactly the same way at the same time. Up until now, we have had to hope that they did, but I predict that single-cell imaging will set the new standards. After all, what good is careful quantitation if it's off by 400-fold?

One interesting aspect of the T-cell sensitivity data is the parallel with sensory organs. Lymphocytes and NK cells can be viewed as cell-sized sensory organs, continuously sampling the internal environment for things that don't belong there or for cellular stress or aberrations. Just as rod cells in the eye can detect even a single photon, cytotoxic T cells can kill on the advice of only three peptide-MHC ligands.4 This redundancy could be a way to eliminate cytocidal "accidents" due to peptide "noise."


Much of the current interest in lymphocyte and NK-cell biology stems from the discovery of what is now called the immunological synapse, by teams led by Abraham Kupfer,5 Michael Dustin,6 and ourselves.7 For many years work on lymphocyte activation had described the aggregation of antigen receptor clusters as "caps" at the T-cell-APC interface. The breakthrough came with high-resolution 3-D fluorescence microscopy and antibody staining as well as video microscopy using T cells and artificial bilayers,56 which revealed a complex organization.

In these synapses, the T-cell antigen receptor clusters in the middle of the interface, surrounded by the key lymphocyte integrin LFA-1. Opposite this on the adjacent cell surface are MHC molecules, including those carrying the antigenic peptide. Around these and directly opposite the LFA-1 on the T cell is a ring of ICAM-1, its ligand. Many other surface and subsurface molecules seem to be taking up specific positions in this tableau, as does the microtubule organizing center just below. The interface is thus a highly organized structure, and not the simple clustering of molecules it was once thought to be.

How does the synapse form? Stimulation through both the T-cell receptor and other T-cell surface molecules (CD28 and LFA-1) triggers the active transport of TCRs from all around the cell into the interface,8 forming the central cluster of receptors. This TCR movement sweeps across the adjacent B-cell area and gathers a selection of MHC molecules opposite it. Why LFA-1-ICAM-1 stays on the periphery of this complex is unknown, but it could be due to the relatively large size of those molecules, compared to TCR and the MHCs.

The purpose of the synapse is still a matter of debate. But one function seems to be directed secretion, akin to its neuronal counterpart. The synapse likely serves to avoid the delivery of secreted factors to the wrong cells, leading to cell death or, perhaps more detrimentally, the inadvertent activation of self-reactive lymphocytes, resulting in autoimmunity. Also at least one report suggests that synapse formation may have a dampening effect, at least at high peptide concentrations.9

Answers to these questions will likely emerge as single-cell techniques develop. But thus far, we have made considerable progress in understanding the molecular basis of cellular recognition in lymphocytes, particularly T cells. We have a good understanding of who the players are, what they do, and where they are in the beginning, middle, and end of the story. We have also discovered several novel phenomena: the immunological synapse, active TCR transport to the interface, the extreme sensitivity of antigen detection, and so forth, all of which deserve explanation. While this is a beginning, I suspect that there is still much to be extracted from this rich vein of cellular biology.

Mark M. Davis is a Howard Hughes Medical Institute Investigator in the Department of Microbiology and Immunology at Stanford University. His work focuses on T-cell receptor molecules and the mechanisms of T-cell activation.

Mark Davis can be contacted at

Article Extras

Related Articles

Feature | Vaccines: Victims of Their Own Success?

Technology | Cancer Immunology Inches Forward

Profession | Ongoing Battle over Transgenic Mice

Popular Now

  1. Opinion: Why I Published in a Predatory Journal
    News & Opinion Opinion: Why I Published in a Predatory Journal

    My “colleagues” and I at the fictitious Arthur Vandelay Urological Research Institute were surprised to find our bogus “uromycitisis” case report swiftly accepted, with only minor revisions requested.

  2. Consilience, Episode 3: Cancer, Obscured
  3. Genetic Analysis Reveals the Evolutionary History of Dogs
  4. Record-Setting Corn Grows 45 Feet Tall