Search

Detection of the functional CD8+ CTL response usually requires in vitro restimulation. The differences between the CD8+ CTL repertoire in freshly isolated precursor cells and CD8+ CTL after short-term in vitro expansion have been generally assumed to be minimal, but have never been defined experimentally. Using staining with P18-I10/H-2Dd tetramers and monoclonal antibodies (mAb) against Vβ, we show the surprising result that there was significant skewing of the CD8+ CTL repertoire after just 7 days of stimulation. In contrast, we found that overnight incubation of precursor cells with peptide allows the functional assessment of CD8+ CTL (which cannot be detected ex vivo from freshly isolatedcells) without changing the absolute number of antigen-specific CTL as measured by tetramer staining or the repertoire of TCR analyzed with mAb. This study affords a better understanding of the differences between the ex vivo and in vitro stimulated CTL repertoire, and provides an approach to reveal a more faithful representation of the functional in vivo CTL response without skewing of the repertoire of T cells detected.

The ability of CD8+ cytotoxic T lymphocytes (CTL) to clear viral infections may be limited when high avidity CTL encounter supra-optimal antigen density on antigen-presenting cells (APC) and undergo antigen-dependent apoptosis of CTL (ADAC). Previously, we have shown ADAC in CD8+ populations to be Fas independent, TNF-alpha receptor 2 (TNFR2) mediated, caspase dependent, and accompanied by a decrease in Bcl-2. We now employ flow cytometry to follow ADAC within individual CD8+ cells to demonstrate that the intense TCR signal induced in high avidity CTL by supra-optimal antigen density results 8 - 16 h later in a caspase-independent TNFR2 down-modulation that is directly related to the stimulating APC antigen density and concludes in a rapid onset of apoptosis by 18 - 24 h. Individual CTL undergoing apoptosis exhibit a dramatic and concurrent: (1) positive staining with Annexin V and propidium iodide; (2) transformation to a smaller cell size characteristic of apoptosis; and (3) a nearly complete loss of Bcl-2, c-IAP1, and TRAF2. We conclude that the antigen-dependent apoptosis of CD8+ CTL occurs when a tandem TCR/TNFR2 signal initiates an abrupt and concordant onset of multiple apoptotic events.

Previously, we observed that high-avidity CTL are much more effective in vivo than low-avidity CTL in elimination of infected cells, but the mechanisms behind their superior activity remained unclear. In this study, we identify two complementary mechanisms: 1) high-avidity CTL lyse infected cells earlier in the course of a viral infection by recognizing lower Ag densities than those distinguished by low-avidity CTL and 2) they initiate lysis of target cells more rapidly at any given Ag density. Alternative mechanisms were excluded, including: 1) the possibility that low-avidity CTL might control virus given more time (virus levels remained as high at 6 days following transfer as at 3 days) and 2) that differences in efficacy might be correlated with homing ability. Furthermore, adoptive transfer of high- and low-avidity CTL into SCID mice demonstrated that transfer of a 10-fold greater amount of low-avidity CTL could only partially compensate for their decreased ability to eliminate infected cells. Thus, we conclude that high-avidity CTL exploit two complementary mechanisms that combine to prevent the spread of virus within the animal: earlier recognition of infected cells when little viral protein has been made and more rapid lysis of infected cells.

It was brought to our attention by Zoran Luicic and Milica Stojanovic, via Peter Gilkey, that some of the diagrams in our paper are not correct.The particular problems are the gluing diagrams for the pair of isospectral surfaces of genus 4, which occur on page 20. It is easy to check that the gluing diagrams given there give rise to a surface of the wrong genus. The problem arose because of carelessness in some of the identifications of some of the edges of the fundamental domain.

In this note, we will construct simple examples of isospectral surfaces. In what follows, we will use the term “surface” to mean a surface endowed with a Riemannian metric, while the term “Riemann surface” will be reserved for a surface endowed with a metric of constant curvature. We will show: THEOREM 1. There exist pairs of surfaces S1 and S2 of genus 3, such that S1 and S2 are isospectral but not isometric.THEOREM 2. There exist pairs of Riemann surfaces S1 and S2 of genus 4 and 6, which are isospectral but not isometric.THEOREM 3. There exist unoriented surfaces S1 and S2 of Euler characteristic X(S1) = X(S2) = — 6 which are isospectral but not isometric.