282 
Monitoring Stem Cell Research 
New human ES cells derived under these conditions would eliminate 
concerns about cross-species contamination by pathogens, but such 
cell lines could not, at present, be supported by federal funding in 
the United States. Growth on human feeder layers is a significant 
advance because of reduced safety concerns; nonetheless, the 
preparation of these feeders remains laborious and introduces a 
significant source of biological variability. The complete elimination 
of feeder layers and serum from human ES cell culture medium and 
their replacement by defined, cloned products remains an important 
goal for the future and is an active area of research for several 
groups. 
Genetic Modification of Human ES Cells 
Although the human genome project is essentially 
completed, we are ignorant about the function of most human genes. 
Human ES cells provide a powerful new model for identifying the 
function of any human gene, and this requires efficient methods for 
genetic modification of human ES cells. Genetic manipulation of 
human ES cells is essential to elucidate gene function; direct the 
differentiation of ES cells to specific lineages; purify desired 
differentiated cell types from mixed populations of ES cell 
derivatives; use the differentiated derivatives of ES cells as a vehicle 
for gene therapy; and modulate the immune response to transplanted 
ES cell derivatives. 
Transfection methods routinely used for mouse ES cells 
generally fail to transfect human ES cells efficiently, but there have 
now been several approaches developed for human ES cells. 
Transient [17] and stable [18] integration of plasmids into human ES 
cells can be accomplished through specific transfection reagents, the 
best reagents yielding stable (drug-selectable) transfection rates of 
about 10"^. Recently more labor-intensive, HIV-based, lentivirus 
vectors have been shovm to transduce human ES cells at an 
efficiency rate of over 90% [20, 21]. This should allow complex 
mixtures of genes to be screened for specific phenotypic effects by a 
process termed “expression cloning" [20]. 
Homologous recombination allows the defined modifications 
of specific genes in living cells [42, 43] and has been used 
extensively with mouse ES cells. However, the differences between 
mouse and human ES cells delayed the development of homologous 
recombination in human ES cells. Except for viral approaches, high 
stable transfection efficiencies in human ES cells have been difficult 
PRE -PUBLICATION VERSION 
