Rihonuclease H and Genetic Replication 
Wayne A. Hendrickson, Ph.D. — Investigator 
Dr. Hendrickson is also Professor of Biochemistry and Molecular Biophysics at Columbia University Col- 
lege of Physicians and Surgeons. He did his doctoral studies in biophysics at the Johns Hopkins University 
and remained for a year of postdoctoral research with Warner Love before going to the Naval Research 
Laboratory for continued postdoctoral studies with Jerome Karle. He stayed on at NRL until he joined the 
faculty of Columbia University. His most recent honor is the Fritz Lipmann Award of the American Society 
for Biochemistry and Molecular Biology. 
RNA-DNA hybrids have a vital role in biology 
as intermediates of genetic replication. DNA 
synthesis in both eukaryotes and bacteria is initi- 
ated by short RNA primers that are hybridized 
with the template DNA. This primer RNA must be 
removed for DNA replication to be completed. 
Various nucleases can perform this digestion, but 
notably specialized to the task are the enzymes 
know^n as ribonuclease H (RNase H, where H 
stands for hybrid), which hydrolyze only RNA 
chains and then only when they occur in 
heteroduplexes. 
RNase H also participates in retroviral replica- 
tion. It is a constituent activity of the multifunc- 
tional enzyme reverse transcriptase and, as such, 
is indispensable for retroviral infectivity. During 
reverse transcription, the polymerase moiety of 
the transcriptase molecule first produces an RNA- 
DNA hybrid on the template viral RNA. RNase H 
then removes the genomic RNA to free the com- 
plementary DNA, which serves as the template for 
plus-strand synthesis. Finally, the resulting DNA 
duplex is integrated into the host genome. 
The RNase H domain constitutes the carboxyl- 
terminal third in reverse transcriptases of such 
retroviruses as human immunodeficiency virus 
(HIV) and Moloney murine leukemia virus 
(MMLV) . These portions of sequence are clearly 
homologous with the RNases H of Escherichia 
colt and yeast. In light of our ultimate interest in 
reverse transcriptase structure, we have under- 
taken a crystallographic study of E. coli RNase H. 
This work, performed in collaboration with Rob- 
ert Crouch, who discovered and extensively 
characterized this enzyme at the National Insti- 
tutes of Health, proves to be fascinating in its own 
right. 
MAD Structure of 
the Selenomethionyl Protein 
We have used a novel method to solve the crys- 
tal structure of E. coli RNase H. This involved 
producing the recombinant protein with seleno- 
methionine systematically replacing the four me- 
thionine residues in this 155-residue protein. 
The selenomethionyl protein is fully active, and 
it crystallizes isomorphously with the natural en- 
zyme. We took these crystals, grown by Wei Yang, 
to the Photon Factory synchrotron in Japan. 
There, in collaboration with Yoshinori Satow, we 
made measurements of multiwavelength anoma- 
lous diffraction (MAD) for use in the structure 
determination. This MAD phasing analysis pro- 
ceeded, in Wei Yang's hands, in a straightforward 
manner to an initial image at 2.2 A resolution. 
Presently the structure is refined at 1 .7 A resolu- 
tion with an R- value residual of 17 percent. 
The RNase H from E. coli is an a/fi protein with 
a tertiary folding that bears no significant similar- 
ity to other known structures. It is organized 
around a mixed central /3-sheet of five strands 
with four a-helices on one side and one on the 
other. Our model is virtually identical with one 
reported independently by another group work- 
ing with a different crystal form. 
Implications for Substrate Binding 
and Catalysis 
RNase H is an endonuclease that releases 5'- 
phosphate products upon hydrolysis of the RNA 
from RNA-DNA hybrid duplexes. The reaction re- 
quires divalent cations, preferentially Mg^"*" in the 
case of E. coli. These and other features distin- 
guish the catalytic mechanism of RNase H from 
that of the RNase A family. We can rather confi- 
dently identify the catalytic site in RNase H as 
being at the conjunction of seven evolutionarily 
conserved residues. At the center of this con- 
served group are a glutamate and two aspartate 
residues that site-directed mutations have shown 
to be absolutely required for activity. They form a 
carboxyl triad that appears to be a likely site for 
the binding of divalent cations. 
Several structural features appear to mark the 
substrate binding zone. One is a sulfate-binding 
site, which we propose to be occupied by a nu- 
cleotide phosphate in the enzyme-substrate com- 
plex. This site lies in a groove separated by a ridge 
from the carboxyl triad, which is 14 A away in 
another groove. The electrostatic potential sur- 
face of these grooves is complementary to that of 
191 
