Plxdc2 Marks Hematopoietic Stem Cells and Th2 Cytokine-Producing Innate Lymphocytes in Adult Bone Marrow

Abstract 1278

Hematopoietic stem cells (HSCs) have been highly enriched using combinations of more than 10 surface markers. However the simple method using a few positive markers is preferable to identify HSCs location in tissue section. We performed a stringent comparative gene expression profiling analysis to find genes preferentially expressed in the HSC population, and identified a total of 63 genes that are highly expressed in HSC among various hematopoietic cell population.

In order to find HSC-specific marker we focused on genes encoding cell surface protein, and found that plexin domain containing 2 (Plxdc2) is highly expressed in CD34^—c-Kit⁺Sca-1⁺Lineage⁻(CD34⁻KSL) HSC population using Plxdc2::GFP knock-in mice. Only 0.2% of whole bone marrow cells were Plxdc2⁺, and competitive repopulation assay clearly showed that all HSCs are included in the Plxdc2⁺ fraction. These results identify Plxdc2 as a new marker of HSCs. Plxdc2⁺ population contain not only HSCs but uncharacterized c-Kit^low/−Sca-1⁺Lineage⁻cells. To further purify HSCs, we investigated the additional positive marker. Throughout the screening of various known HSC-related marker, CD150 was selected. CD150 is already recognized as a positive HSC marker (Kiel, et al. Cell 2005). The Plxdc2⁺CD150⁺ fraction represented only 0.1%±0.002% in whole bone marrow, and 6% in c-Kit⁺Sca-1⁺Lineage⁻ cells, respectively. To test whether the combination of Plxdc2 and CD150 with or without other markers can highly enrich long-term HSCs, we competitively reconstituted irradiated mice with single Plxdc2⁺CD150⁺ cells or single Plxdc2⁺CD150⁺c-Kit⁺Sca-1⁺Lineage⁻ cells. One out of every 4.6 Plxdc2⁺CD150⁺ cells (22%), and one out of 2.2 Plxdc2⁺CD150⁺c-Kit⁺Sca-1⁺Lineage⁻ cells (44%) engrafted and gave long-term multi-lineage reconstitution. The simple combination of Plxdc2 and CD150 significantly increased HSC purity. In addition, we found robust levels of PLXDC2 transcripts in purified human cord blood CD34⁺ HSCs.

Next, we attempted to characterize the another Plxdc2⁺ fraction which is c-Kit^low/−Sca-1⁺Lineage⁻. Multicolor flowcytometric analysis revealed that Plxdc2⁺c-Kit^low/−Sca-1⁺Lineage⁻ cells uniformly express CD45, IL7Rα, Thy-1.2, CD27, T1/ST2 (IL1RL1, a subunit of IL33R) and CD25. These cell surface phenotype indicated that this population is probably of lymphoid lineage. However, culturing Plxdc2⁺ c-Kit ^low/−Sca-1⁺Lineage⁻ cells on OP9-DL1, which supports the development of T-cell progenitors to mature T-cells, did not induce T-cell differentiation. Plxdc2⁺c-Kit^low/−Sca-1⁺Lineage⁻cells also did not differentiate into B cells when co-cultured with OP9 stroma cell line. Furthermore Plxdc2⁺c-Kit^low/−Sca-1⁺Lineage⁻ cells produce IL-5 and IL-13 in response to IL-33 or a combination of IL-2 and IL-25. These characteristics resemble that of “natural helper (NH) cells”, a recently identified cell population capable of producing large amounts of Th2 cytokines in fat-associated lymphoid clusters (Moro, et al. Nature 2010). Immunohistochemical staining of bone section to detect HSCs, and functional analyses to clarify why Plxdc2 specifically express in HSCs and bone marrow “NH cells” using Plxdc2-deficient mice are our ongoing tasks.