Appendix K. 
311 
also has relied on morphological and molecular indications of 
function, such as expression of specific enzymes. Since stem cells by 
definition have not yet taken on a specific differentiated function, 
their identification has relied primarily on use of cell surface markers, 
and only secondarily on production of differentiated products in 
Vcirious tissues. One stated goal has been to isolate a single putative 
adult stem cell, characterized fully by specific markers and molecular 
characteristics, and then to follow the differentiation of this single 
cell (and/or its progeny) to show that it indeed has multipot ent or 
pluripotent capabilities (clonogenic ability). For bone marrow stem 
cells, selection of putative adult stem cells has usually excluded 
typical markers for hematopoietic lineages (lin), CD45, CD38, with 
inclusion or exclusion of the hematopoietic marker CD34 and 
inclusion of the marker c-kit (GDI 17). Other proposed markers for 
adult stem cells are AC133-2 (CD133), which is found on many stem 
cell populations,^ and ClqRp, the receptor for complement molecule 
Clq,^ found on a subset of CD34'^^‘ human stem cells from bone 
marrow and umbilical cord blood. When transplanted into 
immunodeficient mice, ClqRp-positive human stem cells formed not 
only hematopoietic cells but also human hepatocytes. Other 
methods of isolation and identification include the ability of putative 
stem cells to exclude fluorescent dyes (rhodamine 123, Hoechst 
33342), allovvhng isolation by fluorescence-activated cell sorter 
(FACS) of a "side population" of cells within a tissue that have stem 
cell characteristics. Expression of the Bcrpl gene (ABCG2 gene in 
humans) is apparently responsible for this dye exclusion, and could 
provide a common molecular expression marker for stem cells^. A 
study of expressed genes from a single cell-derived colony of human 
mesenchymal stem cells identified transcripts from numerous cell 
lineages,® and a similar attempt at profiling the gene expression of 
human neural stem cell in culture v\hth leukemia inhibitory factor 
(LIF) has been done,"^ perhaps providing an expressed molecular 
milieu which could identify candidate stem cells. Attempts to 
determine the complete molecular signature of gene expression 
common to human and mouse stem cells have shov\m over 200 
common genes between hematopoietic and neural stem cells, with 
some considerable overlap with mouse embryonic stem cells as 
well.® The function of many of these genes is as yet unknown, but 
may provide distinctive markers for identification of adult stem cells 
in different tissues. 
However, dependence on particular markers for prospective 
identification and isolation of adult stem cells seems unreliable. In 
particular, the use of specific hematopoietic markers such as the 
PRE -PUBLICATION VERSION 
