Abstract T cell therapy for the treatment of malignant diseases is based on the lenti- or retroviral introduction of an exogenous receptor in peripheral blood T cells. The exogenous receptor is either antibody based or T cell receptor (TCR) based. Chimeric antigen receptors (CAR) are antibody based receptors that can redirect T cells against membrane antigens expressed by malignant cells. CD19-specific CARs were reported to be very effective in the treatment of CD19+ acute leukemias. To redirect T cells based on cytoplasmic antigens, transduction of a TCR is required. However, this approach still faces technical problems, esp. interference of the endogenous TCR chains may cause loss of avidity and possibly induction of autoimmunity. We here present an alternative strategy, in which, not mature T cells but CD34+ hematopoietic precursor cells are transduced and subsequently differentiated to mature T cells after introduction of a wild type TCR or of a fusion TCR:CD3ζ with or without costimulator signal. When Wilms tumor 1 (WT1)/HLA-A2-specific T cell receptor α and β chain is introduced in CD34+ cells derived from human thymus, cord blood or adult mobilized precursor cells and subsequently induced to differentiate to T cells on OP9 stromal cells expressing Delta-like ligand 1(OP9-DL1) in the presence of stem cell factor, flt3 ligand and interleukin 7, massive proliferation is observed while the cells differentiate to CD4+CD8+double positive (DP) transduced TCR+ immature cells. Few mature T cells are generated in these cultures, but after addition of the specific peptide to HLA-A2+ cultures, DP cells rapidly differentiate to phenotypically mature naïve CD8 single positive T cells. Upon activation, these T cells specifically lyse WT1/HLA-A2 cell lines and produce interferon-γ. Microarray expression analysis revealed these culture-generated T cells to be similar to TCR-transduced peripheral blood T cells, except for 1) the expression of only one TCR α and β chain by the in vitro generated T cells and 2) the underexpression of costimulatory/inhibitory molecules such as CD28, CTLA-4 and PD-L1. The absence of CD28 on the cell membrane was confirmed by flow cytometry. Since it was shown that CD28 signaling is essential for in vivo functionality using CARs, we next generated fusion TCR constructs of a gp100/HLA-A2-specific TCR and the signaling cassettes of CD3ζ and CD28.The following constructs were introduced in CD34+ cells: wild type TCR, TCR:ζ or TCR:CD28ζ α and β chains. The α and β chain double-transduced cells were subsequently cultured on OP9-DL1 in the absence of the specific antigen. It was observed that TCR:ζ transduced precursors proliferated significantly less than wild type TCR transduced cells, but the majority of the cells differentiated towards DP TCR:ζ+ cells, which upon addition of the specific antigen differentiated to phenotypically mature T cells. TCR:CD28ζ transduced cells proliferated least of all and spontaneously matured to functional double negative T cells without passing through the DP stage. These observations are compatible with data obtained in mice showing that strong TCR activation during thymocyte differentiation inhibits the generation of DP cells. In all of these cultures, endogenous TCR rearrangements were suppressed, which resulted in single receptor tumoricidal cells. Functional analysis of these various cell populations showed similar proliferation on T cell growth factors and specific cytolytic activity of gp100+ HLA-A2+ tumor lines. However, the TCR:CD28ζ transduced cells produced significantly higher levels of TNFα and interferon-γ and were the only ones that produced interleukin-2 upon specific stimulation. In conclusion, we have shown that high numbers of polyfunctional single receptor TCR:CD28ζ+ cells can be generated in vitro from clinically relevant stem cell sources. These cells produce interleukin-2, TNFα and interferon-γ and specifically kill gp100/HLA-A2+ tumor cell lines. Disclosures No relevant conflicts of interest to declare.