receptors and binding proteins in the early embryo. 
Transgenic mice carrying a response element for RA 
(RARE) in front of a neutral promoter-/«cZ con- 
struct have provided in vivo evidence that there is 
an active RA transcriptional response in the embryo. 
The transgene shows no expression until the late- 
gastrula/early-neural-plate stage and is then ex- 
pressed only in the posterior half of the embryo. As 
development proceeds the anterior boundary of 
transgene expression regresses in concert with the 
establishment of the anterior boundaries of 3' 
members of the Hox-2 gene family, believed to be 
important in defining regional identity along the 
body axis. 
Further evidence linking RA and Hox gene ex- 
pression comes from studies on the effects of RA 
treatment in vivo on Hox gene expression in em- 
bryos. Treatment of gastrulating mouse embryos 
with excess RA causes an anterior shift of Hox- 2.9 
and Hox- 2.8 expression within four hours of treat- 
ment. More-posterior (5') Hox-2 genes are not af- 
fected. The response of the Hox-2 genes at different 
stages is complex, but the general picture seems to 
be that anterior (3') Hox-2 genes whose boundaries 
end up in the hindbrain or beginning of the spinal 
cord can be shifted more anteriorly by RA in the 
neural tube until they reach their full anterior ex- 
tent of expression. After this they are resistant to RA 
treatment. Combined with the known teratogenic 
effects of RA on hindbrain and in vitro evidence 
from other laboratories of differential sensitivity of 
Hox genes to RA along the 3'-to-5' direction of the 
cluster, this suggests that RA may be a modulator of 
the boundaries of Hox gene expression. The source 
of RA in the embryo is still unknown but may be the 
node at the anterior end of the primitive streak that 
first appears concomitant with the onset of the RARE 
/flcZ transgene expression. 
Gene Trap Screen in ES Cells 
In collaboration with Dr. Alexandra Joyner 
(HHMI, International Research Scholar), a large- 
scale screen is being performed for new genes ex- 
pressed in spatially restricted domains around gas- 
trulation, using gene-trap approaches. The gene trap 
vector contains the Escherichia coli lacZ gene as a 
reporter with a splice acceptor but no promoter up- 
stream. Insertion of this vector into a gene in the 
right orientation and reading frame will disrupt the 
host gene and allow a fusion transcript and protein 
to be produced between the host gene and lacZ. 
After introduction of this vector into embryonic 
stem cells, clones selected for lacZ expression are 
injected into blastocysts. Sets of chimeras from indi- 
vidual clones are screened for lacZ expression at 
gastrulation. Those showing a pattern of interest can 
be taken through the germline to study the possible 
mutant phenotype, and exons in the host gene can 
be cloned from /acZ-containing cDNAs. This is thus 
an efficient means of identifying and mutating 
mouse developmentally regulated genes. A screen 
of 300 such lines for expression in chimeras is al- 
most complete, and a number of interesting patterns 
have been observed. One line of particular interest 
for possible involvement in anterior-posterior pat- 
terning shows /«cZ expression in the node and head 
process derivatives. Cloning and characterization of 
this gene is under way. 
Dr. Rossant is a Senior Scientist at the Samuel 
Lunenfeld Research Institute of Mount Sinai Hos- 
pital, Toronto, and Professor of Molecular and 
Medical Genetics at the University of Toronto. 
Articles 
Forrester, L.M., Bernstein, A., Rossant, J., and 
Nagy, A. 1991- Long-term reconstitution of the 
mouse hematopoietic system by embryonic stem 
cell-derived fetal liver. Proc Natl Acad Sci USA 
88:7514-7517. 
Motro, B., van der Kooy, D., Rossant, J., Reith, A., 
and Bernstein, A. 1991. Contiguous patterns of 
c-^/Yand steel expression: analysis of mutations at 
the TFand 5/ loci. Development 113:1 207-1 22 1 . 
Rossant, J. 1991. Gene disruption in mammals. 
Curr Opin Genet Dev 1:236-240. 
Rossant, J., and Hopkins, N. 1992. Of fin and fur: 
mutational analysis of vertebrate embryonic devel- 
opment. Genes Dev 6:\-\i. 
Rossant, J., Zirngibl, R., Cado, D., Shago, M., and 
Giguere, V. 1991. Expression of a retinoic acid 
response element-^5p/acZ transgene defines spe- 
cific domains of transcriptional activity during 
mouse embryogenesis. Genes Def 5: 1 333-1 344. 
INTERNATIONAL RESEARCH SCHOLARS 531 
