seed the bone marrow space within two weeks of 
transplant) and the long-term repopulating cells 
(LTR, >6 months post-transplant). 
At this time the only reliable method for analyzing 
the earliest hematopoietic precursors — the stem 
cells — is bone marrow transplantation. MRA and 
LTR stem cell clones (20/22) integrated single- 
vector types rather than representatives of the entire 
mixture in the culture. Southern analysis showed 
that multiple copies of that vector integrated into 
each precursor. These studies provide evidence for 
gene transfer and growth of the precursors in a re- 
stricted culture microenvironment. 
A new culture system that apparently supports 
both lymphoid and myeloid precursor growth is 
now being evaluated, using this gene marking 
method. Since most efforts to transfer therapeutic 
genes into stem cells use retrovirus vectors, the find- 
ing that cell interactions are required for efficient 
vector transduction has important implications for 
human gene therapy methods. 
Dr. Belmont's laboratory has also tested the po- 
tential for gene transfer in human hematopoietic 
stem cells. As a preliminary to critical in vivo tests, 
human long-term culture assays were exploited to 
measure the transduction of primitive human pre- 
cursors — long-term culture-initiating cells (LTCIC). 
Diff^erential vector marking was used to establish 
competitive long-term culture assays. These results 
highlighted the importance of stromal cell support 
for the LTCIC. 
Transduction efficiencies of 60-70% were 
achieved under optimal conditions — i.e., cultiva- 
tion of the target cells on preestablished stroma dur- 
ing the period of transduction. In contrast to the 
murine system, transduction efficiency could not be 
improved by stimulation with leukemia inhibitory 
factor (LIF), stem cell factor (SCF), interleukin-3 
(IL-3), or IL-6. The procedures developed in these 
studies are very promising for use in clinical proto- 
cols and will be incorporated into studies being con- 
ducted by Dr. Albert Deisseroth (M.D. Anderson 
Cancer Center, Houston) . (These experiments were 
funded in part by a grant from the National Institute 
of Allergy and Infectious Diseases, National Insti- 
tutes of Health.) 
Another way to approach the problem of stem cell 
growth is to examine the earliest embryologic speci- 
fication of the precursor cells. Hematopoiesis first 
occurs in the yolk sac of developing embryos. In the 
mouse, blood islands develop from the extraem- 
bryonic mesoderm at day 7.5-8.0. This process can 
be modeled in culture, using embryonic stem (ES) 
cells. Under proper conditions ES cells form cystic 
embryoid bodies, up to 90% of which show active 
hematopoiesis. Dr. Belmont's laboratory has re- 
cently detected cells with cell surface markers char- 
acteristic of primitive hematopoietic precursors in 
the embryoid bodies. The laboratory hopes to use 
this new culture system to identify genes important 
in lineage determination. 
One additional approach is also being pursued — 
the identification of the growth factor receptors ex- 
pressed by the stem cells. A systematic search for 
novel receptors and their ligands might lead to clari- 
fication of the molecular basis for growth control of 
these stem cells. 
Using PCR, Dr. Belmont and his colleagues identi- 
fied a new member of the receptor tyrosine kinase 
family by its expression in cells enriched for stem 
cell activity. Determination of its complete se- 
quence is under way. Functional studies using the 
ES cell system should help to clarify its potential 
function in stem cells. 
X-linked Agammaglobulinemia 
Dr. Belmont's group is attempting to clone the 
gene mutated in Bruton's X-Iinked agammaglobu- 
linemia (XLA), one of the commoner human pri- 
mary immune deficiencies. Affected males fail to 
produce mature B lymphocytes as a result of a cell 
autonomous block in the transition of pre-B cells to 
more-mature forms. This condition provides a clini- 
cally important model for genetic control of lym- 
phoid precursor differentiation. 
XLA has been linked to Xq22. Flanking markers 
have placed it in an interval of ~3-5 cM. An anony- 
mous marker, DXS178, shows no recombination 
with the disease locus in more than 100 meioses. In 
collaboration with Dr. Mary Ellen Conley (St. Jude 
Children's Research Hospital, Memphis), the labora- 
tory is using a combination of linkage analysis, yeast 
artificial chromosome (YAC) contig construction, 
and pulsed-field gel electrophoresis (PFGE) map- 
ping to identify candidate genes. An incomplete 
YAC contig spanning ~3 Mb around DXS178 has 
been constructed. One short tandem repeat (STR) 
polymorphism has been identified. This detects 1 1 
alleles (heterozygosity ~70%). 
End probes from the YACs have been used in 
PFGE. At least three CpG islands have been identi- 
fied in the area around DXS178. Lineage-specific 
differential methylation has been observed in two of 
the CpG islands, suggesting the possibility of regu- 
lated expression within the cell types of interest. 
Cloning of these putative genes is under way. 
An important aid to linkage analysis and clinical 
genetic counseling is accurate carrier detection. Car- 
rier females are clinically normal, but their B cells 
show exclusive use of the nonmutant X chromo- 
GENETICS 159 
