Comment on “Cutting Edge: Human CD4⁻CD8⁻ Thymocytes Express FOXP3 in the Absence of a TCR”

We read with great interest the report by Tuovinen and colleagues (1). in which the authors state that CD4⁻CD8⁻ double negative (DN) thymocytes express FOXP3 and do so in the absence of TCRαβ. While studying intrathymic regulatory T cells, i.e., CD4⁺CD8⁻ (CD4SP) FOXP3⁺CD25⁺ cells in patients with myasthenia gravis (2), we also examined thymuses of healthy adults and consistently observed concomitant FOXP3 and CD3 expression in DN thymocytes (Fig. 1⇓ and Table I⇓, which also include CD4SP characteristics as well as the data of Tuovinen and colleagues (1) for comparison). In fact, immunolabeling data demonstrate that FOXP3 is expressed by DN cells that are CD3^low (Fig. 1). This discrepancy cannot reflect the different specificity of mAb, i.e., anti-TCRαβ in Ref.1 and anti-CD3 in our work (2), because at the cell surface TCRαβ is associated with the CD3 complex. Differences due to thymus age are also unlikely, as the distribution of the major thymocyte subsets in our adult thymuses (CD4⁺CD8⁺, 73 ± 7.4; CD4SP, 13.1 ± 5; CD4⁻CD8⁺, 5.4 ± 1.8; DN, 5.1 ± 4.2) was similar to that of young thymuses (3). Finally, FOXP3⁺DN thymocytes in our thymuses are analogous to those described by the authors (1) in terms of CD25 (Table I) and CD127 expression (Fig. 1). The fact that FOXP3 is preferentially expressed by DN thymocytes with the TCR/CD3 complex at low density ties in with the staining profile of CD4SP thymocytes (Fig. 1C in Ref.1 and our own Fig. 1 accompanying this letter). We believe that the authors’ conclusion reflects the difficulty in discriminating TCRαβ-negative from TCRαβ-positive events due to a scarcely efficient reagent, so that TCRαβ^dimFOXP3⁺ cells are defined as TCRαβ⁻FOXP3⁺ cells. Interestingly, TCRαβ⁻CD3⁺ thymocytes were described as early as 1989 (4), but their existence was questioned shortly thereafter by the same authors on the grounds of unreliable immunoreactivity (5). In those articles (4, 5) TCRαβ was identified by clone WT31, currently commercialized by BD Biosciences and probably corresponding to that used by Tuovinen and colleagues (1). In conclusion, we do not dispute that DN thymocytes in young thymus (and in our experience also in adult thymus) express FOXP3, but we assert that these FOXP3⁺ DN thymocytes coexpress TCRαβ/CD3 at low intensity.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Representative flow cytometric analysis of CD3 and FOXP3 and CD127 and FOXP3 expression by CD4SP and DN thymocytes in two thymus samples. FOXP3 expression is restricted to CD3^dim cells and FOXP3 and CD127 are expressed in a mutually exclusive manner in both CD4SP and DN thymocytes. Numbers represent the relative frequency of cells in each quadrant.

• Copyright © 2008 by The American Association of Immunologists