quence, the homeobox domain, that encodes a 
protein DNA binding motif. On the basis of DNA se- 
quence similarity, a set of homeobox-containing 
genes, the box genes, have been isolated in the 
mouse. The function of these genes is not known. 
However, the embryonic expression patterns of 
these genes imply roles in establishing positional 
information during development. Twenty-five 
mouse homeobox-containing Antennapedia genes 
have been isolated and shown to be contained in 
four linkage groups: hoxl, hox2, hox3, and hox5- 
PNS was used to disrupt seven of these genes in ES 
cells. These cells have, in turn, been used to gener- 
ate mouse chimeras and will be evaluated for their 
ability to transmit the mutant allele to their prog- 
eny. Targeted disruption of these genes should not 
only reveal the phenotypes associated with the in- 
activation of the individual genes but, also through 
epistasis and molecular analysis, help define the de- 
velopmental network controlling early mouse mor- 
phogenesis. 
IV mf-related Genes. 
In addition to disrupting genes that encode tran- 
scription factors that activate the progression of the 
developmental program of the mouse, the labora- 
tory is also focusing on genes that mediate cell-cell 
interactions that feed back information to the pro- 
gram concerning the localized developmental prog- 
ress achieved within the embryo at any given stage. 
The m?-related genes are excellent candidates for 
this second class of genes. These genes were first 
identified as sequences activated in mammary tu- 
mors of mice by the nearby insertion of the mouse 
mammary tumor virus. The protein products show 
sequence similarities to growth factors. In situ hy- 
bridization analyses reveal diverse but highly re- 
stricted patterns of expression during develop- 
ment. Four int genes have been identified: int-1 to 
int-4. The int-2 homologue in Drosophila, wing- 
less, participates in establishing cell identity within 
segments via cell-cell interactions by indirectly 
modulating the activity or quantity of transcription 
factors involved in the developmental program. 
Transplantation experiments in chick embryos sug- 
gest a role for the int-2 product in inducing neigh- 
boring cells to progress along a specific differentia- 
tion pathway The above properties are consistent 
with a role for int-1 and int-2 in cell-cell interac- 
tions via unidentified receptors. At the same time, 
they may indirectly modulate the activities of tran- 
scription factors and thereby direct the develop- 
mental program along specific pathways. PNS was 
used to disrupt the endogenous int-1 and int-2 
genes in ES cells. These cell lines were also used 
to generate mouse chimeras, which will be evalu- 
ated for transmission of the mutant alleles to their 
progeny. 
The power of gene targeting is that the experi- 
menter chooses both which gene to mutate and 
how to mutate it. The precision afforded by gene 
targeting should allow the formulation of genetic 
questions with sufficient clarity to yield informative 
answers. 
Dr. Capecchi is also Professor of Biology at the 
University of Utah and Professor of Human Genet- 
ics at the University of Utah School of Medicine. 
PUBLICATIONS 
Books and Chapters of Books 
Capecchi, M.R. 1989. Creating mice with specific mutations by gene targeting. In Molecular Genetics of Early 
Drosophila and Mouse Development (Capecchi, M.R., Ed.). Cold Spring Harbor, NY: Cold Spring Harbor. 
Articles 
Capecchi, M.R. 1989. Altering the genome by homologous recombination. Science 244:1288-1292. 
Capecchi, M.R. 1989. The new mouse genetics: altering the genome by gene targeting. Trends Genet 5:70- 
76. 
Mansour, S.L., Thomas, K.R., and Capecchi, M.R. 1988. Disruption of the proto-oncogene int-2 in mouse em- 
bryo-derived stem cells: a general strategy for targeting mutations to non-selectable genes. Nature 
336:348-352. 
168 
