Cell Interactions in Neurogenesis 
lar and genetic analysis of E( spl ) revealed that 
the locus is defined by several transcription units. 
Focusing mostly on one of those transcripts, we 
have show^n that it codes for a protein with strik- 
ing homology to the bovine /8-transducin, known 
to be involved in signal transduction. We found 
that'point mutations in the protein are capable of 
dramatic synergistic interactions with mutations 
affecting the intracellular region of Notch. We 
were rather surprised to find the subcellular lo- 
calization of the E(spl) transducin homologous 
protein in the nucleus, mastermind, which also 
interacts with Notch, seems to code for a nuclear 
protein as well. This raises the intriguing possibil- 
ity that Notch communicates with the nucleus 
relatively directly. 
While examining identified interactions be- 
tween Notch and other genes, we are studying 
two more, Serrate and slit. The Serrate locus 
captured our attention by virtue of the synergistic 
genetic interactions we noted between certain al- 
leles of Serrate and Notch. We were particularly 
intrigued to find, after cloning Serrate, that it en- 
codes a protein product of 1,404 amino acids 
with a single transmembrane domain and 1 4 EGF- 
like repeats in the extracellular region. Thus 
Serrate represents another member of the group 
of EGF-containing loci in Drosophila and pro- 
vides us with the reasonable working hypothesis 
that Notch and Serrate may interact at the protein 
level. 
To gain more insight into the role of the EGF 
sequence motif in extracellular interactions and 
morphogenetic events, we have undertaken a mo- 
lecular and biochemical characterization of slit. 
This gene codes for a secreted protein with se- 
quences homologous to EGF and to a superfamily 
of extracellular matrix-binding glycoproteins. 
Its embryonic localization, mutant phenotype, 
and sequence homology suggest that it mediates 
interaction among glial cells, axons, and the ex- 
tracellular environment. The interactions in- 
clude those required for developing cells that are 
necessary for axon pathway formation and 
selection. 
In conclusion, work in the past year extended 
our knowledge of the molecular rules underlying 
certain early neural differentiation events. Analy- 
sis indicates that we have identified elements of a 
specific signal transduction mechanism control- 
ling the correct segregation of neuroblasts and 
involving extracellular signals together with cy- 
toplasmic and nuclear components. Moreover, it 
seems clear that this interaction mechanism is not 
exclusively involved in early neurogenesis but is 
used in various developmental times and tissues 
and results in the fine tuning of the differentia- 
tion of certain tissues, including that of the ner- 
vous system. 
In addition, the work carried out this year de- 
fined new questions and directions. Most signifi- 
cantly, we have started to examine the analogies 
that may exist between the invertebrate experi- 
mental model and vertebrates. This will not only 
allow for interesting comparisons, but it will also 
permit us to exploit the experimental advantages 
offered by the sophisticated vertebrate tissue cul- 
ture systems. We will be able to address a variety 
of cell biological questions related to the molecu- 
lar mechanism within which Notch and the other 
neurogenic genes are integrated. As a first step we 
have isolated the human Notch homologue and 
found it to be remarkably similar to its inverte- 
brate counterpart. 
20 
