Studies of the regulatory function of L2a in mouse CD8 gene expression

The TCR coreceptors CD4 and CD8 are crucial for thymocyte development and effector function of T cells. L2a was identified as a cis-acting DNA element putatively involved in CD8 expression. The L2a element has the properties of a nuclear matrix attachment region (MAR). It interacts with two MAR-binding proteins, SATB1 and CDP/Cux, through separated AT-rich regions, L and S. L2a mutants with an increased inter-LS region have decreased CDP/Cux binding, suggesting that both sites are required for binding at the same time. Upon binding of SATB1, these L2a mutants display altered DNase I hypersensitivity (DH) in the inter-LS region. A palindromic DNA 12-mer proximal to the S site was found to alter interactions between L2a and its binding proteins, and two 12-mer binding proteins have been identified. Transgenic studies suggested that L2a is potential silencer for regulating CD8 expression. Transgenes driven by the L2a-containing DH cluster II and an enhancer E8I showed no reporter expression in thymic subsets or in peripheral splenocytes or in intraepitheal lymphocytes (IELs). Deletion of L2a resulted in robust reporter expression, even in the DP population. A small fraction (1~5%) of the L2a-containing transgenic CD8SP thymocytes and peripheral T cells “escaped” L2a-silencing, suggesting that compensatory mechanisms can overcome silencing during transition to the CD8SP stage. Crossing this transgene onto a SATB1 knockdown background decreased the escape rate, indicating that SATB1 is involved in re-starting silenced CD8 expression. Knock-in studies were carried out to further investigate the function of L2a. The M1 mutant knock-in mice, which have altered binding sites that abolish SATB1 interaction, showed no significant changes in CD8 expression. Knock-in mice in which the entire L2a element was deleted (ΔL2a) showed modestly increased CD8 levels in CD8SP thymocytes, peripheral CD8 T cells, and IELs. These effects are indicative of the consequences of losing a potential CD8 silencer, but their modest magnitudes suggest that other compensatory mechanisms suppress L2a function in the germline. Finally, targeted deletion of L2a resulted in significantly decreased CD8αα expression on splenic dendritic cells, implicating an unsuspected regulatory role for L2a in the lineage development of this myeloid sub-population.