cells implies the existence of a cell-cell signaling 
pathway, with subsequent processing of the signals 
to affect gene expression in the nucleus. 
The genetic evidence linking hh to this pathway is 
its requirement for the maintenance of en and wg 
expression. In situ hybridization to wild-type em- 
bryos demonstrated that the hh locus is expressed 
coincidently with en in posterior cells within each 
segment of the embryo. Similar studies with mutant 
embryos revealed that the establishment of hh ex- 
pression requires input from earlier-acting pair-rule 
genes, while its maintenance requires normal func- 
tion of other segment polarity genes, including en 
and wg. The hh gene thus functions in and is sensi- 
tive to positional signaling. Sequence analysis and 
in vitro translation studies suggest that the hh pro- 
tein product is targeted to the secretory pathway. All 
of these results are consistent with a role for hh in 
signaling the en cell identity to neighboring cells, a 
model supported by various studies, including ge- 
netic mosaics that demonstrate a noncell autono- 
mous requirement for hh in cuticle patterning. 
Future work will draw upon the existing genetic 
and molecular studies of other Drosophila segment 
polarity genes to help identify the hh protein recep- 
tor or target and to characterize the intracellular 
consequences of signaling. Studies of other segment 
polarity genes also suggest that hh homologues may 
exist in other species, including mammals. 
Segment Identity 
Following the subdivision of the embryo into seg- 
ments and the specification of positional identities 
within them, the homeotic genes function to specify 
the features that distinguish the segments. The pro- 
teins encoded by homeotic genes each contain the 
homeodomain, a 6l-amino acid segment associated 
with DNA-binding activity, and they generally ap- 
pear to be involved in control of gene expression at 
the level of transcription. The homeodomain has 
also been found in many other multicellular ani- 
mals. In vertebrates, this evolutionary conservation 
extends to the organization of clustered homeodo- 
main genes and to their sequentially ordered ex- 
pression along the body axis. These similarities sug- 
gest that some of the mechanisms of regional 
specification, and perhaps even some aspects of seg- 
mentation and segmental differentiation, are con- 
served between insects and vertebrates. 
Experiments from a number of laboratories sug- 
gest that the homeodomain itself is responsible for 
differences in functional specificity of Drosophila 
homeotic gene products. In an attempt to elucidate 
the mechanistic basis of functional specificity, Ste- 
phen Ekker, focusing initially upon the homeotic 
genes Deformed and Ultrabithorax, has shown that 
individual homeodomains bind DNA with similar 
but distinct base sequence preferences. These small 
differences in sequence recognition can be summed 
through cooperative binding to yield large overall 
differences in binding to multiple sites within a 
DNA region. 
More recently the Abdominal-B homeodomain 
was found to exhibit a DNA base sequence prefer- 
ence distinct from Deformed and Ultrabithorax. In 
contrast, Donald Jackson showed that the base se- 
quence preference of the Antennapedia homeodo- 
main is similar to that of Ultrabithorax. Because 
Antennapedia and Ultrabithorax specify distinct 
segmental differentiation pathways, this result sug- 
gests that differences in functional specificity can- 
not be accounted for solely by intrinsic DNA se- 
quence recognition properties. Other sources of 
functional difference are currently being investi- 
gated — in particular, the possibility of specific 
interactions with other proteins that act as DNA- 
binding partners or as links to other parts of the tran- 
scriptional apparatus. 
A continuing effort in Dr. Beachy's laboratory is 
the identification and characterization of genes regu- 
lated by homeotic gene products. Dr. Chin Chiang 
has identified a gene encoding a homeodomain pro- 
tein that is expressed in a specific subset of neuro- 
blasts and in the anterior spiracle precursor cells in 
the first thoracic segment. Embryos in which Ultra- 
bithorax function is deficient display a duplication 
of the spiracular precursor staining pattern in the 
second and third thoracic segments, and the other 
homeotic genes of the bithorax complex similarly 
appear to repress expression in the abdominal seg- 
ments. Dr. Chiang is currently analyzing the pheno- 
types of mutations at the locus and the molecular 
basis for the negative regulation of this novel ho- 
meodomain gene by the homeotic genes of the bi- 
thorax complex. 
Stephen Ekker was supported by a predoctoral fel- 
lowship from the March of Dimes Birth Defects 
Foundation; John Lee and Donald Jackson were sup- 
ported by training grants from the National Insti- 
tutes of Health. 
Dr. Beachy is also Assistant Professor in the De- 
partment of Molecular Biology and Genetics at 
the Johns Hopkins University School of Medicine. 
GENETICS 151 
