If that is food for thought, it is also a speculative interpretation of what's reported in this week's online PNAS by Josh Syken and Carla Shatz of Harvard Medical School. They found that a piece of the T cell receptor, i.e., its β locus, is normally expressed in neurons of mouse brain, and that its expression pattern changes through development, with abundant expression in the cortex of adult mice.

This study continues the group's prior findings that MHC class 1 antigens and their co-subunit β2 microglobulin are expressed in CNS neurons of mammals, as is CD3ζ one component of the complex that makes up a functioning T cell receptor. Moreover, knockout mice of these components-created originally to study immunological problems-have immature projection patterns of optic nerves to their CNS target area, as well as alterations in their LTP and LDP responses to electrophysiological stimulation (Corriveau et al., 1998). This prompted the idea that components of the "immunological synapse," a complex protein-protein interaction hub that transmits signals between antigen-presenting cells and T lymphocytes, perhaps serves a second function in the dynamic pruning and rebuilding of synapses that is known to occur during development and is also implicated in adult learning.

In this study, the scientists searched for further MCH receptor components in brain. A typical T cell receptor for MHC1 proteins in the periphery contains an α and a β subunit. With in-situ hybridization, Syken found the β subunit in thalamus and various cortical subregions of newborn mice. The regional expression pattern changed throughout postnatal development, and robust expression in layer 6 of the cerebral cortex persisted into adulthood. Histochemistry and retrograde tracing techniques indicate that the TCRβ-expressing cells are indeed neurons and not glial cells, Syken and Shatz report.

The study brought out a striking difference between T lymphocyte and neural TCRβ expression. Neural expression appears to occur without genomic rearrangement. In immune cells, both T cell receptors and antibodies are built in an elaborate process of cutting and pasting pieces of DNA from separate V, D, and J regions, which are then spliced onto a C region. This gene reshuffling, a discovery which garnered a Nobel Prize, generates much of the diversity of B and T cells, but it appears not to be used here, the authors write. Instead, the brain-derived TCRβ transcripts were shorter than the ones in T cells, missing V and D regions and beginning just upstream of a J region. Perhaps, the authors speculate, TCR VDJ genes could rearrange in neurons, but don't do so because one of the enzymes required for the process, RAG2, appears not to be expressed in brain.

Syken and Shatz tried with immunoprecipitation and Western blots to detect the hypothetical protein derived from these TCRβ transcripts, but could not find it in mouse cortex and thalamus of various ages, though they did see it in spleen. Perhaps the protein is stable only in the presence of binding partners, or perhaps the RNA itself has a function, the authors speculate. Since TCRβ forms a receptor with CD3ζ in T cells, and CD3ζ knockout mice have projection errors in the target area of the optic nerve, the scientists checked whether TCRβ mutant mice had the same projection error. They did not, eliminating the possibility that TCRβ and CD3ζ together form a receptor for MCH1 in this brain area. In all other brain areas, TCRβ and CD3ζ expression do not even overlap, so the search for these molecules' binding partners is still wide open.-Gabrielle Strobel.

Reference:
Syken J, Shatz CJ. Expression of T cell receptor {beta}locus in central nervous system neurons. Proc Natl Acad Sci U S A. 2003 Oct 28;100(22):13048-13053. Epub 2003 Oct 20. Abstract