Human AC133 antigen, also called CD133, was recently identified as a hematopoietic stem cell marker. However, the molecular structure and function of this protein has remained unclear. Here we cloned and identified a novel isoform of AC133, which we named AC133-2. In comparison to the reported AC133 cDNA, which is referred to herein as AC133-1, a small exon of 27 nucleotides is deleted in AC133-2 by alternative mRNA splicing. Similar to the previously characterized AC133 antigen, recombinant AC133-2 expressed in 293 cells was glycosylated and transported to plasma membrane. AC133-2 mRNA was found predominant in a variety of human fetal tissue, adult tissues, and several carcinomas. In contrast, AC133-1 mRNA was more prominent in fetal brain and adult skeletal muscle but was not detected in fetal liver and kidney, adult pancreas, kidney, and placenta, suggesting different roles for the two isoforms in fetal development and mature organ homeostasis. Here, we demonstrate that AC133-2 is the isoform expressed on hematopoietic stem cells derived from fetal liver, bone marrow, and peripheral blood. The results indicate that AC133-2, not AC133-1, has been the cell surface antigen recognized by anti-AC133 monoclonal antibodies that are used for isolation of hematopoietic stem cells. To further investigate its expression in other stem cell populations, we found that AC133-2 co-expressed with beta(1) integrin in the basal layer of human neonatal epidermis. AC133-2(+)/beta(1) integrin(+) cells proliferated and differentiated in culture, which coincided with a loss of AC133-2 and gain in a terminal differentiation marker involucrin. Taken together, these results suggest that AC133-2 is expressed in multiple stem cell niches and may provide a means to isolate specific stem cell subpopulations from human tissues.

Infantile hemangioma is a benign tumor of endothelial cells (ECs) that is well known to pediatricians because of its frequency and sometimes disturbingly rapid growth. The primary defect(s) that cause hemangioma have not yet been elucidated, but recent studies have revealed that hemangiomas are composed of clonal ECs that exhibit abnormal properties in vitro. These studies support the hypothesis that hemangiomas arise when a somatic mutation occurs in a single endothelial progenitor, leading to dysregulated activity of one or more genes that control EC growth. Aberrant endothelial growth may in turn alter patterns of gene expression in nearby cells--in the overlying epidermis, for example--and account for observed extrinsic alterations that may influence hemangiogenesis. Unraveling the cellular and molecular basis of hemangioma is likely to yield critical insights into the mechanisms of EC proliferation and regression that may be applicable to other cardiovascular disorders.