Visual System Development in Drosophila 
Flora N. Katz, Ph.D. — Assistant Investigator 
Dr. Katz is also Assistant Professor of Biochemistry at the University of Texas Southwestern Medical Center 
at Dallas. After her B.A. degree in biology from Kenyon College, she did wildlife research as a Thomas J. 
Watson Foundation Fellow. Following her Ph.D. degree in biology with Harvey Lodish at the Massachusetts 
Institute of Technology, she did research at the National Biology Institute in Bogor, fava, as a Henry luce 
Foundation Scholar. She then turned to research in neurobiology with Fric Kandel and James Schwartz 
at Columbia University and Lily and Yuh Nungjan at the University of California, San Francisco, under 
the auspices of the Helen Hay Whitney Foundation and the Damon Runyon- Walter Winchell Cancer Fund. 
VISION is accomplished through the integra- 
tion of signals received by the eye and com- 
municated to the brain. The neurons of the eye, 
or photoreceptor cells, elaborate long processes 
called axons that connect in specific patterns 
with central target areas in the brain to generate a 
map of the visual world. We are interested in un- 
derstanding the mechanisms by which photore- 
ceptor cells are specified and connect with their 
targets during development. 
The eye of the fruit fly Drosophila melanogas- 
ter has proved to be a particularly suitable sub- 
ject for studies of this nature. Unlike the mamma- 
lian eye, the fly eye is a compound structure 
composed of repeating units called ommatidia, 
which each contain eight photoreceptor cells. 
Six of these cells, Rl-6, send axon bundles into 
the optic lobe (a part of the brain specialized for 
visual processing) that terminate in the first syn- 
aptic zone, the lamina. The central two photore- 
ceptors, R7 and R8, send their axons past the lam- 
ina into a second zone, the medulla, where they 
form their connections in two distinct layers. Our 
eff'orts have focused on understanding the genera- 
tion of connections by the R7 and R8 cell types. 
To visualize the photoreceptor axons, we fill 
those originating from any selected population of 
ommatidia with a dye (or an enzyme that can be 
viewed by its reaction product) . The pathway and 
termination zone of each R7 and R8 axon can be 
scored individually, and the path followed dur- 
ing development can be reconstructed. By visual- 
izing the axons in this manner, those that have 
followed unusual routes or terminated in incor- 
rect locations can be identified. 
We have been studying a mutation, nac (neu- 
rally altered carbohydrate), that causes a cold- 
sensitive disruption of the diff'erentiation of the 
fly's photoreceptor cells and other selected neu- 
rons. We have shown that this mutation affects 
the expression of a neural-specific glycan in all 
neurons of the fly, although mutant flies are via- 
ble and show only selective defects in sensory 
behavior. When nac photoreceptor cells are 
filled with dye, the axons of R7 and R8 can be 
seen to enter the medulla correctly, but they 
show increasing defects as they approach their 
terminal zones. Some axons miss their targets al- 
together. Conversely, small groups of ectopic 
fibers have been seen to separate from the main 
axon bundles and take unorthodox routes across 
the medulla, nonetheless arriving in their correct 
terminal zones. These behaviors are likely to re- 
sult from both an altered terrain and an alteration 
on the surface of the photoreceptor axons. 
To identify additional mutations that might af- 
fect these processes more specifically, we com- 
bined our axon-filling technique with a genetic 
strategy. Mutations were generated by the in- 
sertion of a mobile piece of DNA, called a trans- 
posable element. When the element inserts into 
a gene, it inactivates it, causing a mutant 
phenotype. 
The expression of the mutated gene can be in- 
ferred from the expression of a reporter gene 
carried by the transposable element, which ap- 
pears to be regulated by the same controls that 
tell the endogenous gene where and when to be 
expressed. We first selected stocks of flies in 
which the reporter gene was expressed in the de- 
veloping photoreceptor cells or in their target 
cells in the optic lobe. We then asked if the inser- 
tion of the mobile element had caused a mutation 
that affected the diff'erentiation of the photore- 
ceptor cells or the ability of their axons to reach 
the correct target sites in the brain. Several inter- 
esting mutations have been isolated in this 
screen. 
The manner in which these mutations were 
generated — by the insertion of a marked piece of 
DNA — has furthered genetic and molecular stud- 
ies now in progress. In addition, we are analyzing 
these mutant phenotypes with independently 
generated markers that highlight various portions 
of the axon pathways to study the rules that allow 
the generation of a visual map. 
237 
