THREE-DIMENSIONAL STRUCTURES OF BIOLOGICAL MACROMOLECULES 
JoHANN Deisenhofer, Ph.D., Investigator 
Dr. Deisenhofer's laboratory studies the three- 
dimensional structures of biological macromole- 
cules by the methods of x-ray crystallography. The 
general aim is to understand folding, structural sta- 
bility, and function. At the focus of interest are 
protein-protein interactions, the structure of 
membrane-spanning and membrane-associated pro- 
teins, photochemical energy conversion, energy 
transfer, and electron transfer reactions. 
Cytochrome P450 Enzymes 
Cytochrome P450s are heme proteins found in all 
eukaryotes and in some bacteria. In the presence of 
O2, NAD(P)H, and additional electron transfer pro- 
teins, they function as monooxygenases and catalyze 
the oxygenation of hydrocarbon compounds of both 
endogenous and exogenous origin. In collaboration 
with Dr. Julian A. Peterson and his colleagues (Uni- 
versity of Texas Southwestern Medical Center at 
Dallas), two P450 enzymes were crystallized: the 
hemoprotein domain of P4 5 03^.3 from Bacillus meg- 
aterium and P450t.erp from Pseudomonas putida. 
The former diffracts x-rays to 1.5-A resolution and 
the latter to ~ 2. 4-A resolution. The amino acid se- 
quence of either protein is less than 25% identical to 
that of P450cam, the only P450 enzyme of known 
three-dimensional structure. 
The structure of P4 505^.3 was solved by multiple 
isomorphous replacement and twofold noncrystal- 
lographic symmetry averaging. The current model 
has an R value of 0 . 2 6 at 2 . 5 -A resolution . The struc- 
ture analysis of P450yERp is hindered by a crystal unit 
cell dimension of 458 A, too large for the data col- 
lection facilities in the laboratory. Data from native 
crystals and from potential heavy-atom derivatives 
were collected at the Cornell synchrotron and are 
under analysis. 
DNA Photolyase from Escherichia coli 
DNA photolyase is a DNA repair enzyme that uses 
light energy to split carbon-carbon bonds between 
neighboring pyrimidine bases produced, e.g., by ul- 
traviolet irradiation. The enzyme is a complex of a 
polypeptide chain of 471 -amino acid residues and 
two chromophores: FAD (flavin-adenine dinucleo- 
tide) and 5,10-methenyltetrahydrofolate. FAD is ab- 
solutely required for enzyme activity, while the fo- 
late molecule acts as a "light-harvesting" group that 
transfers electronic excitation energy to the FAD. 
Dr. Aziz Sancar (University of North Carolina at Cha- 
pel Hill) has provided purified protein. 
Crystals of DNA photolyase, with two molecules 
related by local twofold symmetry in the triclinic 
unit cell, were obtained. Diffraction data to 2.2-A 
resolution were collected from native crystals. One 
promising candidate for a heavy- atom derivative has 
been identified, and a search for additional deriva- 
tives is under way. In parallel, the possibility of 
solving the structure by the MAD (multiwavelength 
anomalous dispersion) phasing method is being ex- 
plored. (This project is also supported by a grant 
from the Welch Foundation.) 
Synapsin I from Bovine Brain 
Synapsin I, a neuron-specific phosphoprotein, 
is a substrate for cAMP- and Ca^^/calmodulin- 
dependent protein kinases. It occurs in two closely 
related isoforms, called synapsin la and synapsin lb, 
with 706 and 670 amino acids, respectively. The 
difference between the isoforms is restricted to a 
region near the carboxyl terminus, beginning at resi- 
due 662. Synapsin I interacts with synaptic vesicles 
and actin filaments and plays an important role in 
the release of neurotransmitters. Its function ap- 
pears to be regulated by phosphorylation at three 
serine residues. 
Synapsin I was purified in collaboration with Dr. 
Thomas Siidhof (HHMI, University of Texas South- 
western Medical Center at Dallas). Small crystals 
that diffract x-rays to ~ 1 0-A resolution turned out 
to be built from a proteolytic fragment of synapsin I 
about half the size of the intact protein. Similar frag- 
ments expressed in E. coli could be crystallized 
with dramatically improved diffraction quality. Col- 
lection of diffraction data to 3-A resolution and a 
search for heavy- atom derivatives are in progress. 
Ribonuclease Inhibitor from Porcine Liver 
Ribonuclease inhibitor (Rnasin) from pig liver is 
an intracellular protein of 456 amino acids that 
forms extremely tight 1 : 1 complexes with ribonu- 
clease A and angiogenin. Its physiological role is as 
yet unclear. The amino acid sequence of Rnasin 
consists mainly of seven leucine-rich repeats of 59 
residues each. It also contains 30 cysteine residues, 
all of which appear to be in the reduced state. Simi- 
lar leucine-rich repeats occur in many other pro- 
teins in various numbers, but these proteins are 
functionally unrelated to Rnasin. 
Rnasin was purified from pig liver and crystal- 
lized. The tetragonal crystals diffract x-rays to ~3-A 
resolution and contain two molecules, related by 
468 
