RNA probes of steric effects in active sites: High flexibility of HIV-1 reverse transcriptase

Article abstract of DOI:10.1021/ja072791b

As part of viral replication, the HIV-1 reverse transcriptase (HIV-RT) makes a DNA copy of the RNA genome of the virus. It is a mutagenic polymerase, which leads to the rapid development of resistance in patients being treated with antiviral drugs. To stu

Full text of DOI:10.1021/ja072791b

Published on Web 08/14/2007
RNA Probes of Steric Effects in Active Sites: High Flexibility of HIV-1
Reverse Transcriptase
Adam P. Silverman and Eric T. Kool*
Department of Chemistry, Stanford UniVersity, Stanford, California 94305-5080
Received April 21, 2007; E-mail: kool@stanford.edu
As part of the replication of the HIV-1 virus, the HIV reverse
transcriptase enzyme (HIV-RT) makes a DNA copy of the existing
HIV-1 RNA single-stranded genome. The enzyme has been widely
studied because of its biomedical importance as a target for drugs
1
that treat AIDS. It is a highly mutagenic polymerase, a property
that leads to rapid development of drug resistance during treatment
2
of the disease. It has been proposed that one could make use of
this property by employing nucleotide analogues that increase the
3
mutagenicity beyond which the virus can tolerate. Study of the
origins of mutagenicity in this enzyme could contribute new
information on strategies for treatment of this disease, as well as
basic understanding of the mechanisms of replication and evolution.
Recent studies have suggested that polymerase enzymes can use
steric mechanisms to enforce high fidelity of DNA base pairing
during DNA synthesis. In such mechanisms, polymerases tightly
Figure 1. Structures of nonpolar uridine analogues having gradually
increased size. Space-filling models of bases (with methyl in place of ribose)
are shown below, with electrostatic potentials mapped on the surfaces (PM3
calculations, Spartan, Wavefunction Inc.).
4
surround the incipient base pair, using rigid structure to enforce
correct base pair size and shape. Conversely, enzymes that process
DNA with low selectivity might present a sterically more flexible
active site.5 Although recent studies have begun to assess the
performed with a 28mer/23mer or 28mer/24mer template-primer
duplexes, respectively (see SI). For incorporation experiments, the
unnatural base was positioned in the template to pair with an
incoming natural dNTP. Radiolabeled products of single nucleotide
addition were resolved from unreacted primer by denaturing
polyacrylamide gel electrophoresis, and the amounts of incorpora-
tion were quantified by autoradiography. The kinetics data are given
in detail in Tables S1 and S2 (SI) and are shown graphically in
Figures 2 and 3. For comparison, we show data with a related DNA
steric series for a previously studied DNA polymerase;6a the current
study is the first with an RNA-dependent enzyme, and it is of
fundamental interest to compare the RNA-dependent response with
the available literature data for other classes of polymerases having
varied flexibility and varied fidelity.
,6
6
rigidity of DNA polymerases, no such studies have been directed
to RNA-templated DNA synthesis as carried out by reverse
transcriptases. Here we describe the development of a set of
ribonucleoside analogues as tools for probing RNA steric effects
in sub-angstrom increments and their application in systematic
studies of the active site flexibility of HIV-RT. We report that this
enzyme is markedly more flexible than other nucleic-acid-
synthesizing enzymes studied to date.
The variably sized RNA probe nucleosides (Figure 1) were
designed to be shaped similarly to uridine but to have increasing
size over a small 0.8 Å range. The analogues rH and rF were
reported previously,7 and we prepared rL and rB to complete the
variable set (see Supporting Information (SI) for experimental
details). We attempted to prepare an analogous diiodo-substituted
rI nucleoside from the dibromo case using a procedure analogous
to that used in preparation of a related diiodo deoxynucleoside
,8
The experiments involving HIV-RT with RNA templates showed
much lower size dependence than previous experiments with DNA
polymerases. All of the uracil analogues selectively directed
preferential incorporation of dATP over the other nucleoside
triphosphates. As for the effect of size, the smallest analogue, rH,
was less well tolerated (by a factor of 35) than other analogues in
the template (Figure 2A), but there was otherwise little or no
difference in dATP incorporation efficiencies opposite analogues
ranging from the difluoro, dichloro, to dibromo. This is in marked
contrast to high-fidelity DNA polymerases.6 For example, in
previous studies, the Klenow fragment of E. coli DNA Pol I (Kf)
displayed a clear preference for a dichloro-substituted analogue of
thymine (Figure 2B) and showed a much larger 182-fold drop in
9
analogue. However, varied attempts to use a copper-catalyzed
halogen-exchange reaction10 resulted in conversion of a single
aromatic bromine to iodine, while the other bromine was unreactive.
This led us to limit the RNA probe substituent variation to the series
H < F < Cl < Br. The 5′-OH groups of the nucleoside analogues
were protected using dimethoxytrityl chloride, and the 2′-OH groups
were protected using TOM or TBDMS groups. Subsequently, 3′-
O-phosphoramidite derivatives of the series were prepared for use
in the automated solid-phase synthesis of RNAs. Mass spectrometry
data confirmed intact incorporation of the new nucleoside analogues
into RNA oligomers.
6
a
efficiency for decreasing size to the dihydrogen case. Similarly,
while HIV-RT here exhibits no drop in efficiency on increasing
size beyond the difluoro or dichloro cases, Kf showed a 14-fold
decrease from dichloro to dibromo.
To test the functional flexibility of the HIV-1 reverse tran-
scriptase, we prepared 28mer RNA template strands containing the
set of analogues rH, rF, rL, and rB and measured their ability to be
reverse transcribed into DNA strands. Steady-state single nucleotide
incorporation and single nucleotide extension experiments were
Remarkably, the sensitivity to size increases for HIV-RT
measured here is even less than that of a low-fidelity repair
5
polymerase, Dpo4. The latter enzyme showed a 6-fold drop in
10626
9
J. AM. CHEM. SOC. 2007, 129, 10626-10627
10.1021/ja072791b CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. Effects of incoming nucleotide base size on base pair synthesis
by HIV-RT. Data are steady-state efficiencies for insertion of size-varied
dNTP analogues opposite natural RNA bases. (A) Incorporation efficiencies
of the size-varied nucleoside triphosphates. (B) Comparison of the effect
of varying dNTP base size on efficiencies of HIV-RT (black) and the
Klenow fragment of DNA Pol I (gray).6a
Figure 2. Flexibility of HIV-RT, as shown by effects of varying RNA
template base size on nucleotide incorporation. Data are steady-state
efficiencies for insertion of natural dNTPs opposite unnatural bases in the
template, with natural U data shown for comparison. Note log scales for
efficiency; empty columns indicate no observable incorporation. (A) Plot
showing incorporation efficiencies of the four nucleoside triphosphates
opposite the size-varied RNA series. (B) Comparison of varying base size
data for HIV-RT (black) and the Klenow fragment of DNA Pol I (Kf;
active site. In general, the development of a set of systematically
varied nucleotide analogues that retain biological activity may be
broadly useful in probing steric effects in varied subfields of RNA
biology.
6
a
gray). The latter is a higher efficiency enzyme; for ease of comparison,
3
we divided its efficiencies by 10 .
efficiency on increasing size from dichlorotoluene to a dibromo
analogue, whereas HIV-RT shows no change within experimental
error. Thus the data are consistent with the HIV-RT active site for
incipient pair formation being sterically quite flexible as compared
with other polymerase enzymes. We suggest that this flexibility is
likely to be a strong contributor to the mutagenic properties of this
enzyme (and thus of the HIV virus as a whole).
Acknowledgment. This work was supported by the U.S.
National Institutes of Health (GM067201). A.P.S. acknowledges
an NSF Graduate Fellowship and a Lieberman Fellowship.
Supporting Information Available: Details of nucleoside synthe-
sis, RNA synthesis, and enzyme kinetics. This material is available
free of charge via the Internet at http://pubs.acs.org.
To test whether this flexibility is observed on both sides of a
base pair during RNA-to-DNA reverse transcription, we carried
out similar experiments with a set of variably sized incoming
deoxynucleoside triphosphates, using natural RNA bases in the
template RNA. The dNTP series contains bases (dH, dF, dL, dB,
References
(
(
1) Rausch, J. W.; Le Grice, S. F. Curr. HIV Res. 2007, 5, 11-22.
2) (a) Roberts, J. D.; Bebenek, K.; Kunkel, T. A. Science 1988, 242, 1171-
1173. (b) Geretti, A. M. AIDS ReV. 2006, 8, 210-220.
6a,9
(3) Anderson, J. P.; Daifuku, R.; Loeb, L. A. Annu. ReV. Microbiol. 2004,
dI) analogous to the RNA base analogues described here. Steady-
state kinetics data (see SI) are graphed in Figure 3. The results
again showed remarkable tolerance to increasing base size. For
example, increasing size from dichloro to diiodo gives a drop of
only 3.3-fold in efficiency (Figure 3), whereas the Kf DNA
polymerase gave a 22-fold drop for the same size increase.
Moreover, fidelity of HIV-RT pairing of U(T) analogues with A
5
8, 183-205.
(4) (a) Echols, H.; Goodman, M. F. Annu. ReV. Biochem. 1991, 60, 477-
11. (b) Beard, W. A.; Wilson, S. H. Chem. Biol. 1998, 5, R7-R13. (c)
Kunkel, T. A.; Bebenek, K. Annu. ReV. Biochem. 2000, 69, 497-529.
d) Kool, E. T. Annu. ReV. Biochem. 2002, 71, 191-219.
(5) Mizukami, S.; Kim, T. W.; Helquist, S. A.; Kool, E. T. Biochemistry 2006,
5
(
45, 2772-2778.
(
6) (a) Kim, T. W.; Delaney, J. C.; Essigmann, J. M.; Kool, E. T. Proc. Natl.
Acad. Sci. U.S.A. 2005, 102, 15803-15808. (b) Kim, T. W.; Brieba, L.
G.; Ellenberger, T.; Kool, E. T. J. Biol. Chem. 2006, 281, 2289-2295.
(relative to mispairings with U,C,G) was relatively low for the
(
c) Strerath, M.; Cramer, J.; Restle, T.; Marx, A. J. Am. Chem. Soc. 2002,
analogue series (see Figures 2A and 3), whereas it is an order of
magnitude higher with the Kf enzyme.6a Our results suggest that
124, 11230-11231.
(7) Peracchi, A.; Matulic-Adamic, J.; Wang, S.; Beigelman, L.; Herschlag,
D. RNA 1998, 4, 1332-1346.
the mutagenicity and flexibility of the enzyme are directly related.
In summary, we report the development of a set of variably sized
uridine analogues, and we have used them to show that the HIV
reverse transcriptase is an unusually flexible polymerase enzyme.
Interestingly, the RNA template-dependent synthesis by this enzyme
(
8) (a) Guckian, K. M. Ph.D Thesis, University of Rochester, 1999. (b) Parsch,
J.; Engels, J. W. J. Am. Chem. Soc. 2002, 124, 5664-5665. (c) Somoza,
A.; Chelliserrykattil, J.; Kool, E. T. Angew. Chem., Int. Ed. 2006, 45,
4994-4997.
(9) Kim, T. W.; Kool, E. T. Org. Lett. 2004, 6, 3949-3952.
10) Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 14844-14845.
11) (a) Yu, H.; Goodman, M. F. J. Biol. Chem. 1992, 267, 10888-10896.
(b) H u¨ bner, A.; Kruhoffer, M.; Grosse, F.; Krauss, G. J. Mol. Biol. 1992,
223, 595-600.
(
(
is known to proceed with lower fidelity than the DNA-dependent
activity;11 thus it would be of interest in the future to compare steric
responses of these two distinct activities that proceed in the same
JA072791B
J. AM. CHEM. SOC.
9
VOL. 129, NO. 35, 2007 10627