MOLECULAR GENETICS OF VISUAL SYSTEM DEVELOPMENT IN DROSOPHILA 
S. Lawrence Zipursky, Ph.D., Associate Investigator 
Dr. Zipursky's laboratory studies mechanisms of 
neuronal development in the Drosophila visual sys- 
tem. These studies are largely directed toward un- 
derstanding the molecular and cellular mechanisms 
utilized to determine cell fates and to establish pre- 
cise patterns of synaptic connectivity. This system 
has proved amenable to a multidisciplinary ap- 
proach using genetic, biochemical, and histological 
analyses. 
Development of the Drosophila 
Visual System 
The Drosophila visual system comprises the com- 
pound eye, which receives light, and the optic gan- 
glia, which process visual information. The eye and 
the optic ganglia are embryologically distinct. Each 
is derived from a set of cells that are set aside sepa- 
rately as invaginations of the embryonic ectoderm. 
The cells within these two primordia proliferate 
during early stages of larval development, with pat- 
tern formation and neuronal differentiation com- 
mencing during later larval stages. Dr. Zipursky and 
his colleagues have been studying three aspects of 
visual system development: 1) an inductive interac- 
tion between two cells in the developing retina, 2) 
the determination of the optic lobe primordium and 
the regulation of cellular proliferation within it, and 
3) the establishment of the precise pattern of synap- 
tic connections made by photoreceptor neurons 
with their targets in the optic ganglia. 
Cell-Cell Interaction Regulates 
R7 Development 
Dr. Zipursky's laboratory has been studying an in- 
ductive interaction between the R8 photoreceptor 
neuron and a neighboring cell, the R7 precursor 
cell, in the developing compound eye. The com- 
pound eye comprises an array of some 800 identical 
units called ommatidia, each containing eight pho- 
toreceptor neurons (R1-R8 cells) and additional 
accessory nonneuronal cells. Lineage analysis 
has clearly established that interactions between 
cells play a critical role in regulating cell fate 
determination. 
Dr. Zipursky and his colleagues have studied the 
interaction between the protein products of the 
sevenless (sev) and bride of sevenless (boss) genes 
in controlling the development of the R7 neuron. 
Mutations in sev and boss lead to the R7 precursor 
cell assuming a nonneuronal fate. The sev gene en- 
codes a receptor tyrosine kinase and is required in 
the R7 precursor cell. In contrast, the boss gene, 
which also encodes a membrane protein, is ex- 
pressed solely in the R8 cell. 
Several lines of evidence indicate that boss is the 
ligand for the sev receptor. First, boss-expressing 
tissue culture cells bind specifically to sev-express- 
ing cells. Second, membranes from boss-expressing 
cells stimulate the sev tyrosine kinase. Third, boss is 
internalized into the developing R7 photoreceptor 
cell in a sev-dependent mechanism. Finally, ectopic 
expression of boss leads to the induction of R7 de- 
velopment in other sev-expressing cells. (These 
studies were supported by a grant from the National 
Institutes of Health.) 
An intriguing developmental issue is the mecha- 
nism by which only one sev-receptor-expressing 
cell responds to the boss inductive cue during nor- 
mal development. Through a series of experiments 
in which R7 development was assessed in different 
mutant backgrounds, and as a consequence of ec- 
topic expression of boss, it was shown that multi- 
ple mechanisms contribute to the specificity of 
induction. During normal development the sev- 
expressing cells that contact R8 fail to assume an R7 
cell fate as a result of their commitment to alterna- 
tive fates. Cells located one cell diameter away from 
R8 are competent to respond to the inductive cue, 
but fail to do so because boss is tethered to the sur- 
face of the R8 cell. 
Signal Transduction Downstream 
from the sev Receptor Tyrosine Kinase 
Dr. Gerald Rubin and his colleagues (HHMI, Uni- 
versity of California, Berkeley) have shown that the 
sev signal is mediated by a Drosophila homologue 
of ras. Several laboratories have shown in vertebrate 
systems that ras regulates the activity of the extracel- 
lular-regulated kinases 1 and 2 (ERKl and ERK2). 
Dr. Zipursky's laboratory has identified a Drosoph- 
ila homologue of ERK, designated ERK A, which is 
80% identical to its vertebrate counterparts and has 
similar biochemical characteristics. Studies using 
transgenic flies and classical genetic analysis are in 
progress to determine whether the Drosophila ERK 
A gene plays a role in signaling in the sev pathway. 
Control of Optic Ganglion Development 
Dr. Zipursky and his colleagues have been study- 
ing two genes that are required for early stages of 
optic ganglion development. The sine ocidis (so) 
and anachronism (ana) genes play critical roles in 
NEUROSCIENCE 451 
