In Vivo Permanent Marking of Cells with RAG1 Activation History Demonstrates Only a Minor Contribution of the Lymphoid Pathway to Dendritic Cell Development.

Dendritic cells (DCs) have been shown to arise from both myeloid and lymphoid pathways, by evaluating the in vivo reconstituting potential of myeloid- and lymphoid-committed progenitor populations. It has been suggested that these “myeloid” or “lymphoid” DCs may constitute independent DC compartments with different functions. However, evaluation of DC development after conventional adoptive transfer experiments may not correctly represent normal DC development including lineage contribution toward the formation of the DC pool. In order to accurately evaluate the distribution and the function of DCs originated from each lineage, it is ideal to mark their origin in vivo in steady-state hematopoiesis. By crossing RAG1-Cre knockin with yellow fluorescence protein (YFP)-floxed reporter lines, we developed a mouse line in which cells with a history of RAG activation should be permanently marked with YFP as a result of lymphoid commitment. The YFP expression started at the common lymphoid progenitor (CLP) stage (~35%), and the percentages of YFP+ cells progressively increased as they differentiate into mature lymphocytes. More than 99.5% of mature T and B cells were marked by YFP, while <0.5% of granulocyte/monocyte progenitors (GMPs) or granulocytes were YFP+, indicating that this system efficiently and exclusively marks cells of lymphoid origin. We found that only 4.1, 4.6, and 9.3% of DCs were YFP+ in the spleen, the lymph nodes, and the thymus, respectively. This is the formal evidence that the vast majority of DCs in steady-state hematopoiesis originate from stages that have committed to the myeloid lineage, even in the thymus. We then profiled the gene expression of the YFP+ and YFP- DCs at a genome wide level by DNA microarray analysis. Unexpectedly, the gene expression profiles of myeloid and lymphoid-derived DCs are virtually identical. These results collectively suggest that myeloid and lymphoid DCs might use a common developmental program that can be activated even after myeloid or lymphoid commitment. Thus, the DC developmental program is unique, which should be independent of either myeloid or lymphoid commitment program.