Direct mapping of MHC class II epitopes.

Despite technical improvements, the mapping of MHC class II epitopes within complex antigens by genetic or biochemical methods is still laborious and expensive. Here, we describe a simple and fast procedure to directly map T helper cell epitopes within known antigens by bacterial expression cloning. Short antigenic fragments, created by digestion of the coding sequence of the antigen with frequently cutting restriction enzymes, are randomly ligated to the coding sequence of GFP in a bacterial expression vector. Bacteria expressing antigen-GFP fusion proteins are then fed directly to MHC II+ antigen-presenting cells and probed with antigen-specific T cells. Bacterial colonies recognized by T cells are expanded, and the antigenic fragments identified by plasmid extraction and sequence analysis. This direct epitope identification (DEPI) approach offers several advantages. First, bacterial colonies expressing the antigen in frame with GFP are easily detectable by virtue of their green appearance and thereby reduce the screening effort significantly. Second, short antigenic peptides normally unstable in bacteria are highly expressed when fused to GFP. Third, the uniformly high level of expression of short antigenic peptides fused to GFP permits the identification of epitopes even within proteins which are difficult to express in bacteria. Furthermore, by fusing double-stranded oligonucleotides to the GFP gene, crucial amino acids within T cell epitopes may be defined. Thus, this method not only facilitates the identification of T cell epitopes, but also makes it possible to assess the role of individual amino acids for MHC binding or T cell recognition.

Epstein-Barr virus nuclear antigen 1 evades direct immune recognition by CD4+ T helper cells.

The Epstein-Barr virus (EBV) nuclear antigen 1 (EBNA1) is the only viral protein regularly expressed in EBV-associated malignancies. Immune recognition of EBNA1 by CD8+ T cells is prevented by an internal glycine-alanine repeat (GAr) which blocks proteasomal degradation. To test whether EBV-infected cells could be recognized by T helper cells, human CD4+ T cell clones specific for EBNA1 were isolated from latently EBV-infected individuals. These T cells, however, failed to recognize EBV-positive target cells. To investigate whether endogenous presentation of EBNA1 epitopes on MHC class II was prevented by the GAr domain, a mutant EBV strain with an EBNA1 lacking the GAr (EBNA1DeltaGA) was generated and used to establish an Epstein-Barr virus-immortalized lymphoblastoid B cell line (LCL). The EBNA1DeltaGA LCL were not recognized by the EBNA1-specific T cell clones either, indicating that the GAr domain does not mediate this effect. Immune recognition could be restored by overexpression of EBNA1, for which at least 60-fold higher levels of both EBNA1 or EBNA1DeltaGAr protein were required. These results demonstrate that EBNA1 evades direct recognition by CD4+ T helper cells, since its steady state level is below the threshold required for efficient presentation on MHC class II. These findings have important implications for the design of immunotherapeutic approaches to target EBV-positive malignancies.