STRUCTURE AND DESIGN OF DNA-BINDING PROTEINS 
Carl O. Pabo, Ph.D., Investigator 
Dr. Pabo's research has focused on the structure 
and design of DNA-binding proteins. The laboratory 
is attempting to understand how proteins recognize 
specific sites on double-strand DNA and how the 
bound proteins regulate gene expression. During 
the past two years, Dr. Pabo's laboratory has re- 
vealed how some of the major families of regulatory 
proteins recognize their binding sites. This informa- 
tion will eventually be used to help design novel 
DNA-binding proteins for research, diagnosis, and 
therapy. 
Crystallographic Studies 
of Represser-Operator Interactions 
Prokaryotic repressors provide useful model sys- 
tems for the study of protein-DNA interactions, and 
Dr. Pabo's laboratory has been studying the repres- 
sor from bacteriophage X. Several years ago the labo- 
ratory solved the crystal structure of a complex that 
contains the DNA-binding domain of the repressor 
and a 20-base pair operator site. This structure 
showed how the prokaryotic helix-turn-helix 
(HTH) motif is used for DNA recognition. Crystallo- 
graphic refinement at high resolution has now re- 
vealed further details about the repressor-operator 
interactions and has shown how an extended amino- 
terminal arm makes additional contacts in the major 
groove. (The project described above was sup- 
ported by a grant from the National Institutes of 
Health.) 
Crystal Structures 
of Homeodomain-DNA Complexes 
The homeodomain is a conserved DNA-binding 
motif that is found in a large number of eukaryotic 
regulatory proteins. Although the intact proteins 
usually are much larger, the homeodomain itself 
contains about 60 amino acids. Several hundred ho- 
meodomains, which bind to closely related DNA 
sites, have been sequenced. They provide an inter- 
esting and important system for studying protein- 
DNA interactions. 
Over the past several years, Dr. Pabo's laboratory 
has determined the structures of two homeodo- 
main-DNA complexes. These crystal structures re- 
vealed that the homeodomain contains an extended 
amino-terminal arm and three a helices. Helices 2 
and 3 form an HTH unit that is closely related to the 
HTH unit found in a family of prokaryotic repres- 
sors. Although the overall fold of the HTH unit is 
conserved, helices 2 and 3 of the homeodomain are 
longer than the corresponding helices in the X re- 
pressor, and they dock against the DNA in a signifi- 
cantly different way. The homeodomain also has an 
extended amino-terminal arm that makes site- 
specific contacts in the minor groove. 
Even though the two homeodomains (one from 
yeast and one from Drosophild) have different 
amino acid sequences, the structures of the proteins 
and the complexes are quite similar. Dr. Pabo's re- 
search suggests that all homeodomains bind in fun- 
damentally similar ways, and a general model is pro- 
posed for homeodomain-DNA interactions. (The 
projects described above were supported by a grant 
from the National Institutes of Health.) 
Crystal Structures 
of Zinc Finger-DNA Complexes 
Another structural motif, the zinc finger domain, 
has also been found in a large number of eukaryotic 
DNA-binding proteins. These domains have about 
30 amino acids and contain conserved cysteine and 
histidine residues that bind to zinc. Dr. Nikola Pav- 
letich, a research associate in the laboratory, has de- 
termined the structure of two zinc finger-DNA 
complexes. 
Last year he determined the crystal structure of a 
zif268 complex that contains three zinc fingers and 
a consensus binding site. (The immediate-early 
mouse gene zif268 was characterized by Dr. Daniel 
Nathans [HHMI] and his colleagues at the Johns Hop- 
kins University.) This structure revealed that each 
finger docks against the DNA in a similar way and 
contacts a 3-base pair subsite. In each finger, resi- 
dues near the amino-terminal end of an a helix make 
critical contacts in the major groove. The repeating, 
modular arrangement seen in this complex sug- 
gested that it might be possible to mix and match 
different fingers to design DNA-binding proteins 
with new specificities. 
This year Dr. Pavletich determined the structure 
of a complex that contains the five zinc fingers from 
the human GLI protein. (This protein, which is am- 
plified in certain glioblastomas, was characterized 
by Dr. Bert Vogelstein and his colleagues at the 
Johns Hopkins University.) Although the overall ar- 
rangement of the fingers in GLI is similar to that 
discovered in zif, subtle variations in the position 
and orientation of the fingers allow a variety of new 
side chain-base interactions. Comparing the zif and 
GLI complexes helps explain how zinc fingers can 
be used to recognize so many different binding sites. 
480 
