Introduction: Dendritic cells (DCs) are the master regulators of the adaptive immune response. In previous experiments, the double-loading of DCs with matched sets of MHC class I and class II repertoires (i.e. mRNA and cell lysate derived from the same source) appeared to enhance T-cell responses by maximizing the availability of T-cell help. T-cell help, provided by accessory CD4+ lymphocytes, elevates DC immunocompetence in response to antigen-specific interactions between the T-cell receptor and the peptide antigen/MHC class II complex. While T-cell help is specific and aids in the priming of only cognate CD8+ effectors, the mechanism governing this specificity is not understood. Here, we have explored the hypothesis that the specificity of T-cell help may be initiated by the DC itself upon detection of an antigen-dependent signal: the loading of the DC with matched sets of class I and class II determinants.

Methods: To examine this hypothesis, we doubly-loaded DCs with either matched sets of antigenic determinants derived from acute myelogenous leukemia (AML) mRNA and lysate preparations, or, mismatched sets of antigenic preparations derived from AML products, the human TF-1a erythroblastic cell line, or the murine FBMD-1 stromal cell line. Following loading and maturation, we analyzed DCs for markers of immunocompetence such as IL-12 secretion and CD83 surface expression. Using specific siRNA oligonucleotides, we also examined a possible role for the intracellular CD63 tetraspanin in this process. DCs were generated by CD14 magnetic selection of apheresis products and six days of culture in GM-CSF and IL-4. Analysis of immature DCs at this time point by flow cytometry typically showed a CD3+ content equivalent to or less than that of the isotype control. DCs were then loaded with either matched or mismatched sets of determinants (as described above) and matured.

Results: Matched-loaded DCs exhibited a 4-fold increase in IL-12 secretion over unloaded DCs while mismatch-loaded DCs showed only a 0.7-fold increase (similar to controls) (p = 0.0086). Moreover, matched-loaded DCs demonstrated a 21% increase in CD83 surface expression over unloaded/singly-loaded controls while mismatch-loaded DCs showed only a 6% increase (statistically identical to controls) (p = 0.009). In a single-antigen model system, electroporation of CD63 siRNA could reduce IL-12 secretion from matched-loaded DCs by 60% in comparison to matched-loaded DCs electroporated with non-targeting siRNAs. Secretion of non-Th-1 cytokines (i.e. IL-10) was unaffected.

Conclusions: It appears that DC immunocompetence may be upregulated in a cell-autonomous, antigen-dependent fashion. Such upregulation is induced by the loading of DCs with matched sets of MHC class I and class II antigenic determinants and does not occur if DCs are loaded with mismatched determinants. CD63 appears to be involved in this process. CD63 is a member of the tetraspanin family of integral membrane proteins, molecules that facilitate the interaction of membrane and intracellular signaling complexes. In DCs, CD63 is localized exclusively to lysosomal exosomes, sites at which MHC class II, MHC class I, and phagocytosed antigens also co-localize. The data are suggestive of a cross-licensing model by which T-cell help might first be solicited by DCs loaded with matched antigenic determinants. Permission granted by a receptive DC, the helper T-cell might then license DC priming of CD8+ responses. Confirmatory studies are in progress.

Author notes

Corresponding author