the formation of the CNS longitudinal connectives, 
and defects in muscle development. Dr. Rydel 
and Dr. Luis Alonso Garcia are currently generat- 
ing new mutants in lama, conducting a mosaic 
analysis of its function during wing morphogenesis, 
and searching for mutations in the laminin B2 
subunit. 
B. Cadherins. Dr. Paul Mahoney used the polymer- 
ase chain reaction (PGR) method to clone the genes 
that encode two different cadherins (calcium- 
dependent cell adhesion molecule) in Drosophila, 
the first cadherins identified outside of the chor- 
dates. Both of these Drosophila cadherins appear 
to be E-cadherin-like molecules, in that both are 
expressed at high levels in the embryonic ectoderm 
and in larval imaginal disks. With the help of Dr. 
Peter Bryant (University of Galifornia at Irvine), Dr. 
Mahoney has shown that the first cadherin is en- 
coded by the fat gene; mutations in xhc fat gene 
lead to a hyperplastic, cell proliferation phenotype 
of imaginal disks. Hilary Glark has been characteriz- 
ing the second cadherin and has identified candi- 
dates for mutations in this gene. Others are search- 
ing for additional members of the cadherin family 
in Drosophila, with a focus on identification of 
cadherins expressed during neuronal development. 
C. Immunoglobulin superfamily An immunologi- 
cal approach was used to identify four different sur- 
face glycoproteins, fasciclin I, II, and III and neu- 
roglian, which are expressed on subsets of axon 
fascicles and glia during embryonic development. 
Drs. Allan Harrelson, Allan Bieber, Peter Snow, and 
Michael Hortsch showed that fasciclin II and neu- 
roglian are members of the immunoglobulin super- 
family and are highly related to a series of verte- 
brate neural cell adhesion molecules, several of 
which appear to function as homophilic adhesion 
molecules. Fasciclin II is more highly related to 
N-CAM, and neuroglian is more highly related 
to LI. Fasciclin II has five immunoglobulin (Ig)- 
like domains and two fibronectin (FN) type III do- 
mains, whereas neuroglian has five Ig-like domains 
and four FN type III repeats. Dr. Bieber has identi- 
fied a lethal, null mutation in the neuroglian gene 
and with Dr. Hortsch is beginning a genetic analy- 
sis of its function both in vivo and in transfected 
cell lines. 
The amalgam gene encodes another Ig super- 
family molecule expressed in the developing Dro- 
sophila nervous system. Since joining Dr. 
Goodman's laboratory. Dr. Mark Seeger has contin- 
ued a genetic analysis of amalgam function during 
Drosophila development and has shown that the 
amalgam protein is secreted but yet is found associ- 
ated with neuronal membranes, suggesting the 
presence of amalgam receptors and/or binding 
proteins. 
D. Novel cell adhesion molecules. The other two 
proteins (fasciclin I and III) uncovered by the im- 
munological screen for molecules expressed on 
subsets of axon pathways are unrelated thus far to 
anything currently in the data bank. To begin to 
test their function, Drs. Snow and Bieber and Dr. 
Tom Elkins used molecular genetic techniques to 
induce the expression of fasciclin on the surface of 
Drosophila S2 cells in cell culture. Using standard 
aggregation assays and biochemical methods, these 
workers were able to show that both fasciclin I and 
III are homophilic adhesion molecules that appear 
to define new classes of adhesion molecules. Dr. 
Elkins has generated viable, null mutations in the 
genes encoding both of these novel neural cell ad- 
hesion molecules and is conducting a detailed ge- 
netic analysis of their function. Studies in progress 
(in collaboration with Dr. Martin Heisenberg in 
Germany) suggest that both mutants may have in- 
teresting behavioral phenotypes: fasciclin I mu- 
tants are uncoordinated in their leg movements, 
whereas fasciclin III mutants appear to have ab- 
normal optimotor orientation and responses. 
E. Involvement of cell adhesion molecules in sig- 
nal transduction. To begin to unravel the potential 
interactions of neural cell adhesion molecules with 
different second messenger systems. Dr. Elkins has 
examined the phenotypes of embryos doubly mu- 
tant for one of several adhesion molecules 
(fasciclin I, fasciclin III, or neuroglian) and one of 
two tyrosine kinases known to be expressed at high 
levels in developing neurons (the Drosophila ho- 
mologues of the src and abl cellular oncogenes). In 
collaboration with Dr. Michael Hoffman (University 
of Wisconsin), Dr. Elkins has discovered a powerful 
interaction betwe.cnfasciclin I and abl. Whereas at 
a gross level the CNS develops relatively normally 
in either fasciclin I or abl mutant embryos, in dou- 
ble mutants, CNS development is highly abnormal. 
Cell determination and initial axon outgrowth ap- 
pear unaffected. However, they fail to make the 
proper pathway choices; the major outcome is that 
the two large commissural pathways in each seg- 
ment do not form, and the longitudinal connectives 
are often abnormal. 
Continued 
488 
