Skip to content

Tissue-resident memory T cells (TRM) constitute an essential component of adaptive immunity. Unlike circulating memory lymphocytes, these cells do not recirculate: they remain as sentinels within tissues, poised to recognize their antigen and rapidly eliminate infected cells. By promptly secreting pro-inflammatory cytokines such as interferon gamma, TNF-α, and interleukin 2, they also orchestrate local secondary immune responses. The ability to induce these TRM, now established for numerous pathogens and viral vectors, therefore represents a major objective in vaccine design. Yet this property had never been assessed for recombinant adeno-associated viral vectors (rAAV), which are nonetheless attractive for vaccination owing to their excellent tolerability profile in humans and the considerable flexibility afforded by their serotypes and the choice of target tissue. These vectors are, however, weakly inflammatory and poorly transduce dendritic cells—two properties a priori unfavorable to the induction of T-cell responses.

To explore this question, the authors used a model vaccine based on an rAAV2/1 vector administered intradermally into the ear of C57BL/6 mice. The transgene, derived from ovalbumin and carrying additional epitopes, allowed monitoring of antigen-specific T-cell responses. Analysis of cutaneous and lymph node lymphocyte populations was carried out by flow cytometry, with tetramer staining and assessment of cytokine production after restimulation.

A single intradermal immunization was sufficient to induce fully functional cutaneous TRM at the injection site. Their optimal differentiation depended on local transgene expression as well as on help provided by CD4+ T cells. In contrast, transgene expression in dendritic cells proved dispensable: by blocking this expression in hematopoietic cells, the authors showed that the formation of specific cutaneous TRM was not impaired. This result indicates that cross-presentation alone of antigens produced by transduced non-hematopoietic cutaneous cells is both necessary and sufficient to prime these lymphocytes. This property is of particular interest for pseudotyped rAAV vectors, whose ability to transduce dendritic cells is even weaker than that of rAAV2/1.

As for CD4+ T-cell help, the authors suggest that it directly enhances the level of T-cell priming, thereby influencing the generation of the memory precursors from which both central and resident memory cells derive. Taken together, this work establishes that rAAV2/1 vectors can induce genuine cutaneous TRM and clarifies the underlying cellular mechanisms, providing a solid basis for further evaluating their protective potential and their use in vaccine strategies.