Developmental Genetics 
i 
Philippe M. Soriano, Ph.D., D.Sc. — Assistant Investigator 
Dr. Soriano is also Assistant Professor at the Institute for Molecular Genetics and Department of Cell Bi- 
ology, Baylor College of Medicine. He obtained his Ph.D. degree in biochemistry and his D.Sc. degree from 
the University of Paris. He did postdoctoral research in France, and then with Rudolf Jaenisch in Germany 
and at the Whitehead Institute for Biomedical Research of MIT, before joining the faculty at Baylor. He is 
a Pew Scholar in the biomedical sciences. 
THE major aim of my laboratory is to extend 
the understanding of early development of 
the mouse, both by studying cell lineages and by 
generating mutations that affect the embryo. 
In the study of cell lineages, we are particularly 
interested in determining the origin of germ cells 
and in establishing the time of allocation of cells 
to specific lineages once the embryo has been 
implanted in the uterus. In these studies, em- 
bryos are infected at various stages of develop- 
ment vv^ith retroviruses transducing the /3-galacto- 
sidase (|8gal) gene of Escherichia coli. The gene, 
used as a tracer, permits histochemical analysis of 
infected cells and their progeny, w^hich in turn 
permits the establishment of cell lineage rela- 
tionships between jSgal cells in a clone. 
Following implantation, the embryo under- 
goes a series of cleavages accompanied by exten- 
sive cell mixing. Gastrulation begins at day 7 of 
development with the formation of the primitive 
streak. We have observed frequent infection of 
postimplantation embryos with (8gal viruses ex- 
pressing the gene under the control of the viral 
promoter or of an internal phosphoglycerate ki- 
nase 1 promoter. Infection of embryos at day 7 
and 8 of development suggests that extensive cell 
mixing stops at this time, since embryos with sin- 
gle clusters of /3gal-expressing cells are readily 
observed. 
The second aspect of our research concerns 
the identification of developmentally regulated 
genes in the embryo. Despite the mapping of nu- 
merous mutations, both spontaneous or induced 
by radiation or chemicals, it has been difiicult to 
determine the molecular defect on mouse chro- 
mosomes that causes a given phenotype. For this 
reason, insertion mutagenesis, in which DNA in- 
troduced into the germline can cause a mutation 
by disrupting or affecting the expression of a 
gene, is an attractive research approach, with the 
transgene serving as a tag for the molecular clon- 
ing of the affected gene. The DNA can be intro- 
duced into the embryo by retroviral infection or 
direct microinjection into the zygote pronuclei, 
or by using embryonic stem (ES) cells, which can 
be infected or electroporated and then selected 
and introduced into embryos to colonize the 
germline. 
Over the past year we have developed a 
method for generating large numbers of muta- 
tions in mice. The method is useful because, on 
the average, only 1 out of 15 transgenic strains 
exhibits an overt mutant phenotype. To avoid ex- 
tensive breeding of useless strains, we devised a 
screen based on "promoter traps." A reporter 
gene is placed downstream of a DNA splice ac- 
ceptor, and the construct is then introduced into 
ES cells. Expression of the reporter gene can only 
originate from a flanking cellular promoter. 
Therefore, transgenic mice derived from selected 
ES cells can be used both to trace the activity of 
the tagged gene, by expression of the reporter, 
and to mutate the strain. 
We have used as reporters both the /3gal gene 
and a fusion protein encoding two enzymes, (8gal 
and neomycin phosphotransferase. This fusion 
protein, /Sgeo, allows direct selection for pro- 
moter trap events. Thirty- four transgenic lines 
have now been generated using a retroviral pro- 
moter trap vector, and they are being examined 
both for patterns of expression and for pheno- 
type. Among the first 13 strains analyzed, some 
have been found in which the promoter trap 
caused embryo death and some in which no overt 
phenotype is associated with the mutation. This 
suggests that a certain proportion of genes are 
unessential for normal development. To identify 
the mutated genes, we have started to clone the 
sites of viral integration. 
ES cells can be used also to select for mutations 
in specific genes. Efforts in the laboratory have 
focused initially on the gene encoding c-src, a 
proto-oncogene that is a tyrosine kinase. It is 
thought to play a role in the cell cycle and is 
highly expressed in neurons and platelets. Con- 
structs designed to delete the c-src activity were 
introduced into ES cells by electroporation, and 
homologous recombinant clones were isolated 
by screening with the polymerase chain reaction 
(PGR) technique. In a given construct, which in- 
cludes about 8 kilobases of homology, the fre- 
quency of such targeted events is about 1 in 1 00. 
Southern blot analysis has demonstrated that 
415 
