Epitope specificity and clonality of EBV-specific CTLs used to treat posttransplant lymphoproliferative disease.

In a recent phase II clinical trial using banked allogeneic CTL lines to treat EBV-associated posttransplant lymphoproliferative disease, a response rate of 52% was recorded 6 mo posttreatment. Tumor response was associated with an increase in both CTL/recipient HLA matches and CD4(+) T cells within the infused CTL lines. The present study was undertaken to correlate tumor response with CTL specificity. The majority of CTL lines infused recognized EBV-encoded nuclear Ag-3 proteins, but CTL protein specificity itself did not correlate with tumor response. Specificity in conjunction with donor/recipient functional HLA matching as opposed to HLA matching alone, however, was important for tumor response. CTL receptor TCR beta-chain variable gene subfamilies were polyclonal, with no preferential use of a particular family. However, tumor response was improved in those receiving CTL lines with polyclonal vs clonal distribution for subfamilies 2, 3, and 9. Interestingly, in five of six tumors (five Hodgkin's-like and one Burkitt's-like posttransplant lymphoproliferative disease) with restricted viral gene expression a complete response was recorded, although in some cases the tumor cells did not express the proteins recognized by the infused CTL. Thus CTL were advantageous when functionally HLA matched but for certain tumor types complete responses occurred in the absence of detectable specific CTL/tumor recognition. We suggest that either the allogenic CTL contained small, undetectable, EBV-specific, HLA-matched T cell populations or perhaps they stimulated nonspecific inflammatory responses in vivo, which were beneficial for tumor regression. These observations should be considered when designing and implementing CTL therapies.

Allogeneic cytotoxic T-cell therapy for EBV-positive posttransplantation lymphoproliferative disease: results of a phase 2 multicenter clinical trial.

We present the results of a multicenter clinical trial using Epstein-Barr virus (EBV)-specific cytotoxic T lymphocytes (CTLs) generated from EBV-seropositive blood donors to treat patients with EBV-positive posttransplantation lymphoproliferative disease (PTLD) on the basis of the best HLA match and specific in vitro cytotoxicity. Thirty-three PTLD patients who had failed on conventional therapy were enrolled. No adverse effects of CTL infusions were observed and the response rate (complete or partial) in 33 patients was 64% at 5 weeks and 52% at 6 months. Fourteen patients achieved a complete remission, 3 showed a partial response, and 16 had no response at 6 months (5 died before completing treatment). At 5 weeks, there was a significant trend toward better responses with higher numbers of CD4(+) cells in infused CTL lines (P = .001) that were maintained at 6 months (P = .001). Patients receiving CTLs with closer HLA matching responded better at 6 months (P = .048). Female patients responded better than male patients, but the differences were not statistically significant. Our results show that allogeneic CTLs are a safe and rapid therapy for PTLD, bypassing the need to grow CTLs for individual patients. The response rate in this poor prognosis patient group is encouraging.

Crystal structure of tobacco etch virus protease shows the protein C terminus bound within the active site.

Tobacco etch virus (TEV) protease is a cysteine protease exhibiting stringent sequence specificity. The enzyme is widely used in biotechnology for the removal of the affinity tags from recombinant fusion proteins. Crystal structures of two TEV protease mutants as complexes with a substrate and a product peptide provided the first insight into the mechanism of substrate specificity of this enzyme. We now report a 2.7A crystal structure of a full-length inactive C151A mutant protein crystallised in the absence of peptide. The structure reveals the C terminus of the protease bound to the active site. In addition, we determined dissociation constants of TEV protease substrate and product peptides using isothermal titration calorimetry for various forms of this enzyme. Data suggest that TEV protease could be inhibited by the peptide product of autolysis. Separate modes of recognition for native substrates and the site of TEV protease self-cleavage are proposed.