| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
St. Jude Children’s Research Hospital, Memphis, Tennessee 38105, USA;
* Yale University School of Medicine, New Haven, Connecticut, USA; and
University of Texas, Southwestern Medical Center, Dallas, Texas, USA
2Correspondence: St. Jude Children’s Research Hospital, 332 N. Lauderdale St., Memphis, TN 38105, USA. E-mail: william.evans{at}stjude.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: thiopurines • thioguanylated DNA • viral RNase H
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. The combined use of HIV RT and protease
inhibitors has markedly improved the treatment of HIV infected
patients, yet resistance of HIV-1 to anti-retroviral therapy remains a
major cause of treatment failure (4
, 5)
.
The thiopurines mercaptopurine (MP) and thioguanine (TG) are widely
used as antileukemic agents, targeting proliferating malignant cells
with high DNA replication rates (6
, 7)
.
2'-Deoxy-6-thioguanosine 5'-triphosphate (dGSTP),
the active metabolite of MP and TG, is known to be a good substrate for
human DNA polymerases 
, 
, and 
, with Km’s
similar to those of natural substrates (8)
. Incorporation
of dGS into DNA leads to notable alterations in
DNA properties, affecting interactions with several DNA binding
proteins (9
10
11
12)
. Thiopurines have also been shown to have
anti-viral effects against herpes simplex virus (13)
,
influenza viruses (14)
, and SV40 (15)
, but
anti-HIV effects have not been recognized and the mechanism(s)
underlying the known anti-viral activities has not previously been
elucidated.
Recently, we demonstrated that incorporation of deoxythioguanosine into
the DNA strand of DNA-RNA heteroduplexes abrogates RNA hydrolysis by
bacterial and mammalian RNase H (16)
. These findings led
us to postulate that thiopurine treatment of cells could reduce viral
RNA cleavage, thereby inhibiting early stages of HIV replication. Here
we report that deoxythioguanosine triphosphate is incorporated into DNA
by HIV-1 RT, abrogating RNA cleavage by HIV-1 RNase H and producing
significant anti-HIV effects in human lymphocytes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P] ATP was obtained from Amersham
(Piscataway, NJ). The RNA oligonucleotide size markers were prepared by
alkaline hydrolysis with piperidine (17)
. Fresh human
blood seronegative for HIV and hepatitis B virus (HBV) was obtained
from the American Red Cross (Baltimore, MD). The lymphotropic clinical
isolate HIV-1ROJO was obtained from Southern
Research Institute Co., Frederick, MD. Phytohemagglutinin was obtained
from Sigma (St. Louis, MO) and recombinant IL-2 was from Amgen (San
Francisco, CA). Tritiated thymidine triphosphate was obtained from New
England Nuclear (Boston, MA).
Preparation of dGSTP
dGSTP was prepared from
5'-DMT-3'-Bz-N2-phenylacetyl-2'-deoxyguanosine (Sigma) by consecutive
treatment with thioacetic acid according to the previously published
method (18)
and subsequent phosphorylation with
2-chloro-4H-1,3,2-benzodioxaphosphorin-4-one (Fluka, Buchs,
Switzerland) (19)
. Deblocked product was dissolved in 2 M
LiClO4 and precipitated with acetone. The final
purification was achieved by ion-exchange chromatography on a column
(1x30 cm) packed with Toyopearl DEAE-650M (Supelco, Bellefonte, PA) in
a gradient of LiClO4 (0–0.2M) in 10 mM
Tris-HClO4 pH 7.0. The presence of a 6-thio group
was confirmed by UV spectrum (
max=340 nm) and the presence of a
triphosphate group was confirmed by activity of the product in the
polymerization reaction catalyzed by Klenow fragment (9)
.
RNA template and DNA primer preparations
The DNA oligonucleotides (18 and 20 mer primers) were
synthesized by the Yale DNA synthesis facility and 5' radiolabeled with
T4 polynucleotide kinase (New England Biolabs, Beverly, MA). The RNA
oligonucleotide (49 mer, Fig. 1
) was synthesized by in vitro transcription (MEGAscripTM, Ambion) and
5' radiolabeled by dephosphorylation with shrimp alkaline phosphatase
(Amersham), followed by phosphorylation with T4 polynucleotide kinase.
Purity of the oligonucleotides was estimated by polyacrylamide gel
electrophoresis (20% PAGE, 8 M urea).
|
The heteroduplexes R49/D20 and R49/D18 (Fig. 1)
were formed by
annealing 
1:1.2 molar ratio of DNA primer and the corresponding RNA
template at 90°C for 4 min and 50°C for 30 min. Oligonucleotide
concentrations were determined spectrophotometrically. The following
molar extinction coefficients (
260) were used
for nucleotides: pA = 15400; pT = 9300; pU = 8800;
pC = 7300, pG = 11700; for pdGS, a
molar extinction coefficient (340) of 24800
was used (20)
. The heteroduplexes were analyzed by
nondenaturing PAGE (15%).
Preparation of HIV-1 RT
HIV-1 RT was purified as described previously (21
, 22)
. The protein concentration of purified HIV-1 RT was measured
spectrophotometrically at 280 nm, using an extinction coefficient
280 = 260,450
M-1cm-1. The
concentration of active RT was determined as described previously with
presteady-state burst experiments (21)
, which gave burst
amplitudes of 40%; the experiments described here were performed using
the corrected active site concentration.
Kinetic analysis for single nucleotide incorporation into R49/D20
primer template by HIV-1 RT
Rapid chemical quench experiments were performed as described
previously with a KinTek Instruments Model RQF-3 rapid quench-flow
apparatus at 37°C (21
, 22)
. Unless noted otherwise, all
concentrations refer to the final concentration after mixing. In
presteady-state burst experiments, HIV-1 RT (100 nM) and primer
32P-labeled R49/D20 heteroduplex (300 nM) were
preincubated on ice for 5 min in 50 mM Tris-HCl (pH 7.8), 50 mM NaCl.
Polymerization was initiated at 37°C by the rapid addition of either
dGTP (5–200 µM) or dGSTP (10–400 µM) in
buffer containing 10 mM MgCl2 and the reactions
were quenched at the indicated times with 0.3 M EDTA. Products were
analyzed by PAGE.
Processive polymerization by HIV-1 RT on R49/D18 primer template
The processive polymerization was studied under steady-state
condition in the presence of either dGTP or
dGSTP, along with the other three dNTPs using
primer 32P-labeled R49/D18 heteroduplex (Fig. 1)
.
The reaction conditions were similar to the single nucleotide
incorporation studies, except for the following modifications: 10 nM
(active site) HIV-1 RT and 1000 nM R49/D18 heteroduplexes used at a low
concentration of dNTP (10 µM) or 500 nM HIV-1 RT and 100 nM R49/D18
with a high concentration of dNTP including dGSTP
(100 µM) and 5 µM dGTP. An aliquot of the reaction mixture was
removed and manually quenched with 0.3 M EDTA at each time point. The
reaction samples were analyzed as described below in Modeling and
Parameter Estimation.
Effect of dGS inserts in DNA strand on HIV-1 RT RNase
H activity
The processive polymerization was carried out under steady-state
conditions in the presence of either dGTP or
dGSTP with the other three dNTPs using
32P-labeled RNA template of R49/D18 heteroduplex
(Fig. 1)
. A aliquots of the reaction mixture was manually quenched at
each time interval, and the reaction samples were analyzed as follows.
Modeling and parameter estimation
All samples from HIV-1 RT activity assays were analyzed by gel
electrophoresis (20% PAGE, 8 M urea). The reaction products were
quantified using a Bio-Rad GS525 Molecular Imager (Bio-Rad, Hercules,
CA) and Molecular Imager software (Bio-Rad). Data were fitted by
nonlinear regression using the program KaleidaGraph (version 3.09,
Synergy Software). Data from burst experiments were fitted to a burst
equation: [product] =
A[1-exp(-kobsdt)+ksst],
where A represents the amplitude of the burst that
correlates with the concentration of enzyme in active form,
kobsd is the observed first-order rate
constant for dNTP incorporation, and
kss is the observed steady-state rate
constant. The dissociation constant, Kd, for
dNTP binding to the complex of RT and 20/49 heteroduplex is calculated
by fitting the data into the following hyperbolic equation:
kobsd =
(kpolx[dNTP])/(Kd+[dNTP]),
where kpol is the maximum rate of dNTP
incorporation, [dNTP] is the corresponding concentration of dNTP, and
Kd is the equilibrium dissociation constant for the
interaction of dNTP with the enzyme–DNA complex. The mathematical
model was fit to data on the cleavage of R49/D49G or
R49/D49GS heteroduplexes using maximum likelihood
fitting techniques via the ADAPT II program (23)
.
Transduction of HeLa cells with HIV-based gene transfer vector
HeLa cells were treated with MP or TG freshly dissolved in 1:1
sterile H2O:DMSO to make a 10 mM stock solution
(final concentration in experiments 0.1–10 µM). Cells were treated
with MP or TG added either 16–20 h before transduction or
simultaneously with the vector, essentially as described previously
(24
, 25)
. Briefly, HeLa cells
(2x104 per well) were plated on 6-well tissue
culture dishes the day before transduction (MOI=0.2 and 0.6) with
concentrated RtatpEGFP vector supernatant. Cells were harvested 36 h after transduction and analyzed by flow cytometry using a FACScalibur
instrument (BD). Data were collected and analyzed using CellQuest
software (BD). Percent inhibition was determined as the difference
between transduction with and without the addition of MP or TG at the
two different vector concentrations indicated above. To ensure that
EGFP expression was due to effective transduction with the lentivirus
vector, samples of HeLa cells were also incubated with 5–50 µM
zidovudine (AZT) for 30 min before addition of vector. Under these
conditions, AZT reduced transduction efficiency by 95% (data not
shown) (24
, 25)
.
Anti-HIV activity of MP and TG
Experiments to assess the antiviral and cytotoxic effects of
thiopurines in human peripheral blood mononuclear cells (PBMC) infected
with HIV-1ROJO were performed by Southern
Research Institute, as follows. Fresh human PBMC were isolated from
blood drawn from HIV and HBV seronegative donors, as determined by the
American Red Cross. PBMC were isolated by Ficoll-Hypaque density
centrifugation and infected with a clinical isolate
HIV-1ROJO at an MOI of 0.1, either simultaneously
or 20 h after treatment with six semi-log dilutions of thioguanine
(0–200 µM). Antiviral activity and cellular toxicity were assessed 6
days postinfection by reverse transcriptase activity assay
(26)
and
2,3-bis[2-methoxy-4-nitro-5-silfophenil]2H-tetrazolium-5-carboxanilide
staining (27)
. The difference between the anti-HIV effect
and cytotoxicity was evaluated by the Wilcoxon matched pairs test.
Analysis of drug combination assay was performed as described
(27)
with statistical evaluations performed according to
the methods of Prichard and Shipman (28)
. Five
concentrations of AZT at 3.125 nM, 6.25 nM, 12.5 nM, 25 nM, and 50 nM
were tested in all possible combinations with nine concentrations of TG
(0.195 µM, 0.391 µM, 0.781 µM, 1.563 µM,3.125 µM, 6.26
µM,12.5 µM, 25 µM, 50 µM). Effects of the drug combinations
were calculated based on the activity of each compound when tested
alone. Experimental data were analyzed by the stringent statistical
means by assuming the compounds inhibited HIV replication by acting at
the same site (mutually exclusive model). Results of the combination
assays are presented 3 dimensionally at each combination concentration,
yielding a surface of activity extending above (synergy) or below
(antagonism) the plane of additivity. The volume of the surface is
calculated and expressed as a synergy volume
(µM2 %) calculated at the 95% confidence
interval.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. A biphasic burst of product formation of DNA 21 mer was observed:
kobsd = 19.3 ± 0.8
s-1 for dGTP and
kobsd = 4.8 ± 0.4
s-1 for dGSTP. The
concentration dependence of the polymerization rate for incorporation
of dGTP and dGSTP into a DNA strand of R49/D20 is
shown in Fig. 2
, indicating that dGSTP is a reasonably good
substrate for HIV-1 RT.
|
|
Processive polymerization by HIV-1 RT on R49/D18 heteroduplex
The processive primer elongation using dGSTP
vs. dGTP in R49/D18 heteroduplex (Fig. 1)
was studied under
steady-state conditions at concentrations of dNTP ranging from 10 µM
to 100 µM. For the natural substrate dGTP, the majority of elongated
products were in the fully extended form (DNA 49 mer) after 40 min of
incubation. In the presence of the dGSTP analog,
however, formation of the fully extended product was significantly
less: most of the product formed corresponded to a DNA 20 mer (data not
shown). Processive elongation of the 18 mer DNA primers occurred with
similar efficiency in the presence of 100 µM
dGSTP or 5 µM dGTP (Fig. 3
). The presence of dGS in the DNA strand was
confirmed by enzymatic hydrolysis to nucleosides, followed by HPLC
analysis, as we have previously described (16)
. The
nucleoside ratio in D49GS
(dA/dG/dT/dC/dGS) was 11/6/11/12/3, close to the
theoretically calculated ratio of 11/6/12/15/5.
|
RNA degradation within DNA/RNA heteroduplex by RNase H activity of
HIV-1 RT
The cleavage of RNA within the DNA/RNA heteroduplex by the
RNase H activity of HIV-1 RT was characterized in
5'-32P-RNA-labeled R49/D49 or
5'-32P-RNA-labeled
R49/D49GS heteroduplexes after RT
polymerization with dGTP or dGSTP (Fig. 4
). Incorporation of dGS into
R49/DGS49 by HIV-1 RT polymerization (Fig. 1)
substantially altered the short RNA cleavage pattern compared with the
natural heteroduplex D49/R49. The linear steady-state rate parameters
(k49) for cleavage of the R49 template
and formation of R17/18 fragments documented that the rates of
hydrolysis of R49 were comparable for both heteroduplexes R49/D49
(k49=0.2546
min-1) and R49/DGS49
(k49=0.1811
min-1). However, incorporation of
dGS in the proviral cDNA strand significantly
changed the cleavage rate of 18/17 mers
(k18/17=0.7780
min-1 for R49/D49 vs.
k18/17=0.2649
min-1 for R49/DGS49,
P<0.001 by t test; Fig. 4
, bottom panel).
|
Inhibition of RtatpEGFP transduction in the presence of MP or TG
Incubation of HeLa cells with MP or TG for 16–20 h before
addition of the HIV-based vector, with EGFP expression analyzed 36 h later, resulted in 50% inhibition by 1.0 µM MP or 0.1–1 µM TG
at MOI = 0.2 (Fig. 5B
, D
). Inhibition was also evident at a higher level of
infection (MOI=0.6), but less than that observed at the lower level of
infection (inhibition for 1.0 µM TG=32% and for 10 µM MP=43%;
Fig. 5C
, E
).
|
Anti-HIV effects in cell cultures
Significant anti-HIV effects (decrease in RT activity) were
evident in HIV-1-infected peripheral blood lymphocytes pretreated with
thioguanine (Fig. 6
, filled symbols: IC50=0.0346 µM,
IC95=15.4 µM). TG was significantly less toxic
to donor lymphocytes, as depicted in Fig. 6
(open symbols:
TC50>200 µM). The difference between the
anti-retroviral activity and cytotoxicity curves (Fig. 6)
was
statistically significant (Wilcoxon matched pairs test,
P <0.000005).
|
The activity of TG in combination with AZT revealed evidence of at least additive interactions across the entire range of concentration tested, with evidence of synergy at low concentrations (synergy volume=19 µM2 %). There was no evidence of synergetic cytotoxicity in uninfected PBMCs at all combined concentrations of TG and AZT examined. In fact, TG was less toxic to host cells when given with AZT than without (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, we provide the first
evidence that dGsTP is incorporated into viral
DNA by the polymerase function of HIV-1 RT. The diagnostic burst phase
observed in presteady-state experiments indicates that HIV-1 RT
recognizes and incorporates dGSTP in a manner
similar to that of natural dNTPs, whereas binding affinity
(Kd) of dGSTP (110±30 µM)
for the enzyme-DNA complex is threefold weaker than that of dGTP
(35±4 µM). Experiments with 5'-32P-labeled DNA
primers showed that both dGTP and dGSTP were
incorporated into DNA of R49/D20 heteroduplex, although the maximum
rate of incorporation (kpol) and the
incorporation efficacy
(kpol/Kd) were
lower for dGSTP (Table 1)
. The single nucleotide
incorporation studies of dGTP vs. dGSTP indicate
that the rate of incorporation for 5 µM dGTP is similar to the rate
for 100 µM dGSTP (Fig. 2)
. This is consistent
with DNA synthesis stalling at the site where the next correct incoming
dNTP is dGSTP opposite a template cytosine.
Accumulation of the full-length cDNA product (D49) occurred in the
presence of either dGTP or dGSTP (Fig. 3)
,
clearly demonstrating that similar to bacterial and mammalian DNA
polymerases (8)
, the RNA-dependent DNA polymerase of HIV-1
efficiently uses dGSTP for viral cDNA synthesis.
Cleavage of viral RNA by the RNase H function of HIV-1 RT, essential
for early stages of HIV replication, is accomplished through two
different modes (29
, 30)
: one polymerase dependent and the
other polymerase independent (31)
. In our model
(5'-P32-labeled 49 mer oligoribonucleotide
template), the primary polymerase-dependent RNaseH cleavage results in
formation of 17 and 18 mer RNA fragments (Figs. 1
and 4)
. The secondary
RNase H cleavage (29)
leads to further degradation of the
17–18 mer RNA fragments to shorter RNA fragments (15–9 mer RNA, Fig. 4
). The linear steady-state rate parameters
(k49) for cleavage of the R49 template
document that the rates of hydrolysis of R49 by the primary
polymerase-dependent RNaseH cleavage are comparable for both
heteroduplexes R49/D49 and R49/DGS49, because
there is no dGS incorporated close to this
position (Fig. 1)
. Shorter cleavage products are thereafter formed due
to the exonuclease activity of RNase H (second mode), which acts
progressively toward the 5'-end of the RNA (30)
. The
cleavage of 18/17 mers for R49/DGS49 was
threefold slower than for R49/D49 (Fig. 4
, bottom panel). The pattern
of short hydrolysis products was markedly altered at the sites opposite
dGS in the DNA strand of the heteroduplex (no RNA
fragments shorter that 12 mer was found for
R49/DGS49), indicating that hydrolysis of the RNA
strand of the heteroduplex by exonuclease activity of HIV-1 RNase H is
largely abolished at positions close to dGS
inserts. This is consistent with our previous findings that the rate of
hydrolysis of a heteroduplex by bacterial or mammalian RNase H is
decreased within 2–3 nucleotides of the dGS
incorporated into the deoxy strand of the heteroduplex
(16)
. The effect of MP and TG on the initial phases of the
HIV replication cycle (using RtatpEGFP, Fig. 5
) indicate that the
targets of anti-HIV effects of TG are events occurring no later than
transcription of viral RNA from the proviral DNA, consistent with the
data documenting inhibition of RNA cleavage by HIV RNase H. In contrast
to existing RT-targeting drugs that directly inhibit the polymerase
function or bind to the RNase H-catalyzing center of RT, incorporation
of dGS into the minus strand of viral DNA creates
poor RNase H substrates, thereby abrogating subsequent hydrolysis of
RNA in the RNA/GSDNA hybrid, representing a novel
strategy for inhibiting retroviral replication.
As shown in Fig. 6
, anti-HIV effects were evident in PBMC treated with
thioguanine 20 h before viral infection, reaching a maximum of
98% inhibition at >10 µM TG (filled symbols:
IC50=0.0346 µM, IC95=15.4
µM). Cytotoxicity of TG on donor lymphocytes (Fig. 6
,
TC50>200 µM) revealed a statistically
significant difference between the anti-HIV and cytotoxicity curves
(P < 0.000005). These concentrations are well within
the usual plasma and cerebrospinal fluid concentrations achieved after
oral and intravenous administration of MP in humans (0.2–4 µM)
(32
, 33)
. In slowly dividing cells, the most active DNA
synthesis is likely that of proviral cDNA; thus, the major
thioguanylated DNA would be viral DNA, consistent with greater
antiviral activity than cytotoxicity. TG treatment simultaneously with
infection of PBMC (i.e., 0 h pretreatment with TG) also had
antiviral effects (IC50=1.48 µM, data not
shown); the higher IC50 is consistent with the
requirement for metabolic activation of TG before its effects are
evident. When evaluated in combination with AZT, there was evidence of
at least additive interactions across the entire concentration range
tested, with evidence of synergy at TG = 0.195 µM and AZT 3.25
nM (synergy volume=19 µM2 %). Evidence of
additive antiviral activity and the absence of additive cytotoxicity of
TG and AZT to host lymphocytes indicates that TG warrants further
evaluation as a component of antiviral chemotherapy.
These experiments establish that incorporation of
dGSTP affects both steps of viral
replication—DNA polymerization and RNA cleavage; therefore,
thiopurines represent a new class of agents with anti-retroviral
activity. The currently available non-nucleosides inhibitors of HIV-1
RT interfere with RNase H (34)
and polymerase
(35)
functions by binding a specific region of the p66
subunit of HIV-1 RT in a hydrophobic pocket of the palm
domain, and mutations in this binding site have been shown
to confer HIV-1 resistance to these agents. Because thiopurines inhibit
RNaseH function by modifying its DNA-RNA substrate, and not by directly
binding HIV RT, its anti-HIV activity should not be impaired by these
HIV-RT mutations. Thus, the present work establishes that thiopurine
medications, which have been in clinical use for almost 50 years,
possess novel anti-retroviral effects that may complement contemporary
agents used to treat HIV infected patients.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication March 9, 2001. Accepted for publication May 15, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Bohon and C. R. de los Santos Structural effect of the anticancer agent 6-thioguanine on duplex DNA Nucleic Acids Res., February 15, 2003; 31(4): 1331 - 1338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Mansky, E. Le Rouzic, S. Benichou, and L. C. Gajary Influence of Reverse Transcriptase Variants, Drugs, and Vpr on Human Immunodeficiency Virus Type 1 Mutant Frequencies J. Virol., February 1, 2003; 77(3): 2071 - 2080. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Somerville, E. Y. Krynetski, N. F. Krynetskaia, R. D. Beger, W. Zhang, C. A. Marhefka, W. E. Evans, and R. W. Kriwacki Structure and Dynamics of Thioguanine-modified Duplex DNA J. Biol. Chem., January 3, 2003; 278(2): 1005 - 1011. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. L. Carroll, D. Bhojwani, D.-J. Min, E. Raetz, M. Relling, S. Davies, J. R. Downing, C. L. Willman, and J. C. Reed Pediatric Acute Lymphoblastic Leukemia Hematology, January 1, 2003; 2003(1): 102 - 131. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) concentrations were
fitted to a hyperbolic equation. For dGTP (•), the hyperbola (solid
line) yield a Kd value of 35 ± 4 µM for
dGTP dissociation and a maximum rate of incorporation
(kpol) of 24 ± 1 s-1. For
dGSTP (
) and DGS49(
)
formation (bottom panel).
