(page 2 of 2)

This intimate association with the regulation of peptide binding suggested that the invariant chain might play a role in autoimmune disease—and prompted several questions: What does the invariant chain look like? How does its structure relate to this critical function - and is it a potential therapeutic target? Attempts to determine the invariant chain's structure were not successful—but they whetted Teyton's appetite for the powerful techniques of structural biology as a means of answering basic immunological questions.

There is a great deal of genetic diversity among individuals in the type of MHC-encoding gene expressed. At the time, it was known that people carrying the MHC-II gene known as "HLA-DQ" were much more likely to have Type 1 diabetes than those who did not. This suggested a new question to Teyton: Do these genetic differences translate into differences in protein structure that might ultimately affect function, thereby leading to the disease state?

Obtaining the structure of MHC molecules had already proven to be an elusive goal for many labs, due to the difficulty in producing large amounts of soluble, properly paired MHC-II molecules. Teyton solved this problem by engineering soluble mouse MHC-II molecules whose pairing was forced by the addition of leucine zipper peptide dimers at their COOH-terminus. Subsequent proteolytic removal of the leucine zipper left stable MHC-II dimers that bound peptide in the same manner as the native molecule.

A Picture is Worth a Thousand Words

This technical innovation made it possible to obtain large amounts of soluble protein molecules, enabling Teyton to grow crystals and to determine the crystal structure of mouse MHC-II. Teyton obtained structural data for two different mouse MHC molecules: the mouse MHC-II I-Ag7, which is expressed in the mouse model of Type 1 diabetes; and the mouse MHC-II I-Ad, which is not linked to diabetes.

"We had hypothesized that there were unusual structural features of I-Ag7 that accounted for its role in autoimmunity," Teyton said. "However, after solving the structure, we saw that the I-Ag7 is a normal class II MHC molecule in its structural features and in the way that it binds peptides."

Thus, many questions remain to be answered in the quest to understand the mechanisms of diabetic autoimmunity, such as: Is diabetes caused by a peptide-specific mechanism? The specific antigen recognized by auto-reactive T-cells has yet to be identified. Teyton's lab is part of a multi-center National Institutes of Health (NIH) project that is working on this question.

A second possibility is that the MHC-II molecule itself somehow leads to abnormal T-cell selection. Although there are no obvious structural differences between the MHC-II molecules that are linked to Type 1 diabetes and those that are not, there are potentially significant sequence differences between them. This sequence difference results in a net negative charge on the peptide-binding portion of the MHC-II not linked to autoimmune diabetes, while a neutral charge in this portion of the molecule is associated with disease susceptibility. This difference in charge could confer selectivity for peptides that cause diabetes, or could play a role in abnormal T-cell selection or activation.

To further explore this possibility, Teyton is conducting structural studies of the interaction of the diabetes-associated MHC-II and the T-cell receptor, through the use of interference microscopy, a new technique that allows the visualization of single molecules at the cell surface. This enables Teyton to examine the interactions between reconstituted T-cell receptors and MHC-II molecules—and to detect any differences in this interaction that are dependent on the type of MHC-II expressed. Ultimately, this will lead to a better understanding of the mechanisms of T-cell activation and the abnormalities associated with autoimmune responses.

Says Teyton, "If we can detect the structural basis for a problem, we will have a basis for specific clinical intervention. The fundamental question remains: How are T-cells activated, and can we specifically prevent the activation of particular subsets?"

Teyton states his goals with intensity, recalling the frustration of being a clinician who could not explain why his patients were suffering. But Teyton's commitment to figuring out how things work, combined with the advanced technological resources at his disposal, may someday help to alleviate that frustration for future physicians and patients.


1 | 2 |




Representation of the structure of an MHC molecule as found at the surface of infected cells. This structure was determined at the atomic level by x-ray diffraction and electron microscopy observation. Click to enlarge.