DEVELOPMENTAL GENETICS 
Philippe Soriano, Ph.D., Assistant Investigator 
Research in Dr. Soriano's laboratory is focused 
on early development of the mouse embryo. The 
two main aspects of this research are 1) the study 
of cell lineages, during the egg cylinder stages and 
at gastrulation, and 2) the elucidation of the role of 
specific genes during development of the embryo, 
using insertional mutagenesis in transgenic mice. 
These projects involve both retroviral infection of 
embryos and embryonic stem (ES) cells and tar- 
geted mutagenesis of specific genes by homologous 
recombination in ES cells. Much of the work on 
mutagenesis is in close collaboration with Dr. Allan 
Bradley (Baylor College of Medicine). 
I. Retroviruses and Mouse Development. 
Classical approaches for the study of lineages, 
such as injecting dyes into individual cells, are of 
limited use in the mouse because of rapid dilution 
of the marker with successive cell divisions. Previ- 
ous work has shown that genetic mosaicism intro- 
duced into developing mouse embryos by retro- 
viruses could be used to obtain information on 
allocation of cells to specific lineages. These studies 
demonstrated a high level of cell mixing prior to 
gastrulation and an early determination of the 
germline. The aim of the studies that have been ini- 
tiated over the past year has been to obtain infor- 
mation on cell lineages in the mouse embryo dur- 
ing the egg cylinder stage and at gastrulation. 
A retroviral labeling system capable of identifying 
single cells and their progeny is required to address 
these issues. The bacterial (B-galactosidase gene has 
been chosen as a reporter gene because of the facil- 
ity to detect its activity. Because the viral promoter 
is not active in early embryos, the P-galactosidase 
gene has to be placed under the control of an inter- 
nal promoter, in this case the herpes simplex virus 
thymidine kinase promoter. Retroviral vectors thus 
expressing the reporter gene have been used to in- 
fect both preimplantation and postimplantation 
embryos. 
Because of the block to expression of the viral 
promoter and because the presence of the viral en- 
hancers has been thought to act as a negative regu- 
latory element in embryonal carcinoma (EC) cells, 
the effect of the enhancers on the efficiency of ex- 
pression from internal promoters has been tested 
in ES cells as a model system. These new retroviral 
vectors either contain sequences (Gen vectors). 
resulting in higher viral titers, or do not extend into 
the gag region (Zen vectors); all vectors include a 
bacterial supF gene to facilitate cloning of the pro- 
virus. Studies with these vectors have demonstrated 
that 1) deletion of the enhancers has no significant 
effect on viral titer; and 2) expression of the re- 
porter gene is modulated, at the most a few fold, by 
the presence or absence of the viral enhancers, in 
both Gen and Zen vectors, and that this effect is 
the same in fibroblasts, where the viral promoter 
is active, as in ES cells, where it is not. These re- 
sults argue against an active repression by viral se- 
quences of expression from an internal promoter in 
ES cells. The laboratory also has recently generated 
high titer, viral enhancer, and promoter-minus vec- 
tors that are being tested. 
II. Retroviral Insertional Mutagenesis. 
One approach to the study of the role of genes 
active in development is their inactivation by germ- 
line insertion of foreign DNA in transgenic mice. 
The approach is attractive, because the foreign DNA 
serves both as a means to disrupt the gene physi- 
cally and as a tag for the molecular cloning of the 
affected gene. Retroviruses appear particularly well 
suited for this purpose, since they do not cause any 
rearrangements of flanking sequences and can be 
engineered to allow for selective cloning of the 
flanking sequences. Experiments under way in the 
laboratory involve the introduction of retroviruses 
into the germline of mice to create new transgenic 
strains, both by infection of preimplantation em- 
bryos and by infection of ES cells. The use of ES 
cells allows screening of the cells, prior to their 
reintroduction into blastocysts, and breeding of 
the chimeras to test for contribution to the germ- 
line. The laboratory's first experiments with viral- 
infected ES cells have demonstrated a significant 
number of germline chimeras. 
To facilitate screening of mutation events. Dr. 
Soriano and his colleagues have devised an en- 
hancer trap and a gene trap system. In the en- 
hancer trap retroviral construct, lacking the viral 
enhancers, a P-galactosidase gene, with its initiator 
codon, has been placed downstream of the viral 
promoter and of a splice acceptor. The viral vector 
also contains a neo cassette, permitting selection in 
ES cells, and the supF gene to facilitate subsequent 
cloning. Almost no ES clones infected with this 
Continued 
303 
