the 600 A structures, and to determine how 600 A 
structures pack to form the final 1,300 A structure. 
This new level of clarity is undoubtedly due in part 
to the use of the cooled CCD to record the tilt data 
directly and digitally. Current efforts focus on inter- 
preting this reconstruction, recording more data, 
and improving data collection, reconstruction, and 
"display. 
C. Three-dimensional optical microscopy and 
chromosome structure. The three-dimensional op- 
tical microscopic studies pioneered by Drs. Agard 
and Sedat have been used as a powerful control for 
EM sample preparation methods and also to inves- 
tigate the dynamic behavior of chromosomes 
throughout the cell cycle. Three-dimensional reso- 
lution in the light microscope is now sufficient to 
trace the path of chromosomes in diploid nuclei 
from early prophase through late telophase. The 
most exciting results have come fi-om a com- 
bination of three-dimensional in vivo microscopy 
with the higher resolution data available from 
fixed specimens. These data suggest that the com- 
plex process of chromosome condensation is ac- 
tually nucleated at discrete sites on the nuclear en- 
velope. Furthermore, chromosome compaction 
that occurs during the prophase-metaphase transi- 
tion begins at the centromere and spreads in a 
wave-like manner out to the teleomeres. This work 
has led to the first reliable data on the three-dimen- 
sional spatial arrangement of diploid chromosomes 
within the nucleus. Recently developed three-di- 
mensional in situ hybridization methods will be 
used to determine what sequences lie at the nucle- 
ation points. 
D. Use of antibodies to combine structure with 
function. Staining of Drosophila nuclear division 
cycle 14 embryos with an antilamin monoclonal an- 
tibody produces a noncontinuous staining pattern 
of interlocking fibers in which a significant fraction 
of the nuclear surface is not occupied by lamins. 
Similar staining patterns have been observed in 
HeLa and Drosophila Kc cells. These unanticipated 
patterns suggest that the extremely large oocytes 
may be anomalous. Analysis of the spatial distribu- 
tion of topoisomerase II staining indicates that at 
high resolution, topoisomerase II is not colocalized 
with the chromosomes but instead resides largely 
on structures that seem to wind around the chro- 
mosomes. Although still preliminary, this result has 
striking implications for models of chromosome 
and nuclear matrix structure. 
II. Structural Basis of Enzyme Specificity. 
One fundamental function of an enzyme is to be 
specific, that is, to limit the number of substrates 
on which it can act. Since catalysis derives from the 
enzyme's ability to stabilize the transition state of a 
reaction, specificity is a consequence of selective 
binding of the transition states for preferred sub- 
strates. Dr. Agard has chosen a-lytic protease as an 
ideal model system to investigate structural and en- 
ergetic aspects of specificity In addition to there 
being a wide range of substrates and inhibitors 
available, its binding pocket is made of the side 
chains of three amino acids (Metl92, Met213, 
Val217A), providing a large volume that could be 
experimentally manipulated. 
The enzyme from the soil bacterium Lysobacter 
enzymogenes has been cloned and expressed at us- 
able levels in Escherichia coli (100-200 mg/101 fer- 
menter run). Through a collaboration with Dr. 
Charles Kettner (DuPont), Dr. Agard and his col- 
leagues have been able to obtain a large number of 
tight-binding peptide boronic acid inhibitors. These 
inhibitors were used to determine high-resolution 
crystal structures for more than a dozen of the in- 
hibitor-enzyme complexes using the native enzyme. 
These structures have provided insights into the 
structure of the transition state and the importance 
of substrate hydrogen bonding for its stabilization, 
as well as basic information on steric exclusion and 
specificity. 
By mutation, Dr. Agard and his co-workers have 
been able to alter dramatically the pattern of sub- 
strate specificity while maintaining or increasing en- 
zyme activity (activity toward Phe substrates was in- 
creased by nearly six orders of magnitude). 
Structural analyses of the first few mutants as free 
enzymes and as complexes have provided insights 
into the mechanism of specificity and indicated the 
crucial role that protein flexibility plays in selectiv- 
ity. Current efforts involve mapping the detailed en- 
ergetics of protein flexibility through further muta- 
genesis, kinetic, and crystallographic analyses. More 
than 25 structures have been determined. A collab- 
oration with Dr. Peter Kollman (University of Cali- 
fornia at San Francisco) has been initiated to apply 
theoretical methods to this complex problem. 
III. Structural and Biochemical Probes of Folding of 
a-Lytic Protease. 
An unexpected benefit of choosing a-lytic prote- 
ase was revealed when the gene was cloned and 
Continued 
566 
