Identification of a Clinical RT Backbone that Improves Replication of an NNRTIResistant Mutant by In

1
Identification of a Clinical RT Backbone that
Improves Replication of an NNRTI-Resistant Mutant
by Increasing the Rate of Polymerization R. A.
Domaoal, C. Dykes, R. A. Bambara, and L. M.
Demeter University of Rochester School of
Medicine and Dentistry, Rochester, NY
ABSTRACT Background We previously identified a
clinical reverse transcriptase (RT) sequence
containing P236L (8-26) which, when placed into
NL4-3, resulted in virus with improved
replication kinetics compared to NL4-3 with a
P236L/NL4-3 RT (Dykes et al. Virology
2001285193). We demonstrated that P236L/NL4-3
RT slows steady-state rates of both
polymerase-independent and polymerasedependent
modes of RNase H cleavage (Gerondelis et al. J
Virol 1999735803). We also recently found that
P236L/NL4-3 RT slows the turnover rate (kss)
during single nucleotide incorporation with a
DNADNA primertemplate, under pre-steady state
conditions (Domaoal et al. 10th CROI, 612). For
wild-type RT, kss reflects the rate of product
dissociation, and is normally the rate-limiting
step in polymerization. No reductions were seen
in kpol (the maximal rate of nucleotide
incorporation) or Kd (affinity) for dGTP by P236L
RT. We wanted to determine whether improvements
in either RNase H or polymerase function
accounted for the partial improvement in
P236L/8-26 replication relative to P236L/NL4-3.
Methods Wild-type NL4-3, P236L/NL4-3, and
P236L/8-26 RTs were expressed in E. coli and
purified. Polymerase-dependent RNase H activity
was assayed using a 5-end- 32P-labelled 41
nucleotide (nt) RNA annealed to a shorter DNA,
with the DNA 3-end recessed relative to the RNA
5-end. Polymerase-independent RNase H activity
was assayed using the same RNA hybridized to a
longer DNA, with the RNA 5-end recessed relative
to the DNA 3-end. RNase H reactions were
carried out in the absence of dNTPs and RT input
was normalized for steady-state polymerization
specific activity on a homopolymeric RNADNA
templateprimer. kss during single nucleotide
incorporation of dGTP was measured under burst
conditions (substrate excess) with a DNADNA
templateprimer, using a KinTek quench-flow
apparatus. Results Polymerase-dependent and
polymerase-independent RNase H cleavages by
P236L/NL4-3 and P236L/8-26 RTs were similarly
slowed relative to wild-type RT. The turnover
rate for dGTP incorporation (ksssd) was
0.160.018/sec for wild-type, 0.040.018/sec for
P236L/NL4-3, and 0.210.018/sec for P236L/8-26
RTs. Conclusions A clinical RT backbone that
partially compensates for the replication defect
of P236L also improves its reduced turnover rate
(kss) during polymerization. We postulate that
this significantly contributes to the ability of
this clinical RT sequence to compensate for
P236Ls replication defect. These studies
demonstrate that pre-steady state kinetics can
identify the underlying biochemical mechanisms
leading to modulation of drug resistance
mutations by clinical RT backbones. In addition,
these studies indicate that the polymerization
abnormalities identified for P236L/NL4-3 RT
contribute significantly to the reduction in
replication efficiency conferred by this mutant
in cell culture. We believe these abnormalities
account for the infrequent occurrence of this
mutant during delavirdine therapy, despite its
high level of drug resistance, and that such
studies can be used to better understand
resistance patterns during clinical failure of
other non-nucleoside RT inhibitors.
P236L
Relative Replication of NL4-3 with Clinical RT
Sequences
Patient Clone NNRTI-R mutns p24 content _at_ d10,
relative to wt 3 3-1 K103NY181C 1.31 3-16
P236L 0.42 3-23 P236L 0.11 8 8-4
K103N 0.21 8-15
K103N 0.05 8-26 P236L 0.76
Product (nM)
rate-limiting step of catalysis, occurs w/in
msec (burst rate)
The relative replication kinetics of K103N
and P236L are reversed in this clinical RT
background
overall rate-limiting step of polymerization,
takes seconds to occur (steady-state rate)
Coding changes in 8-26 RT relative to NL4-3
Pre-steady state burst of dGTP incorporation into
a DNADNA templateprimer. P236L has a slowed
kss (steady-state rate). This rate is normal for
the 8-26 (P236L) RT.
V531I
C162S, K174T
P236L
R356K,M357R, K358R
P468S
T200A
Q102K
V245M
I435V
L491A
Fingers
Palm
Thumb
RNase H
Connection
INTRODUCTION Several potential mutations can
confer resistance to NNRTIs, yet only a
restricted subset of mutations occurs clinically.
The most common mutation that confers NNRTI
resistance in clinical isolates is K103N .
Surprisingly, some uncommonly occurring mutations
confer high levels of drug resistance, suggesting
that other factors account for their low
prevalence in clinical isolates. Patients
receiving delavirdine monotherapy developed K103N
most commonly. Less than 10 of the patients had
P236L, which confers high levels of delavirdine
resistance . Similarly, V106A and G190S confer
high levels of nevirapine and efavirenz
resistance, respectively, yet are much less
common in clinical isolates than K103N. Our
studies have shown that K103N in an NL4-3
background replicates better compared to P236L
and V106A. For P236L, this replication defect
relative to K103N persists in the presence of all
but very high levels of delavirdine. This
finding suggests that viral replication fitness
in addition to the level of drug resistance can
influence the likelihood of a drug resistant
mutant emerging during treatment failure. Our
previous studies have shown that the NNRTI
resistance mutations P236L, V106A, K103N, and
Y181C specifically alter RNase H cleavage
relative to RNA-dependent DNA polymerization,
under steady state conditions (Gerondelis et al.
J Virology 735803-5813, 1999 and Archer et al. J
Virology 748390-8401, 2000). The P236L and
V106A mutants, which have reduced replication
fitness relative to both wild-type and K103N, had
more extensive reductions in RNase H activity,
affecting both polymerization-dependent and
-independent modes of cleavage. K103N, which had
significantly improved replication fitness
relative to P236L or V106A, affected
polymerization-dependent mode of RNase H cleavage
only. These results suggested that reductions in
RNase H activity have important effects on HIV
replication fitness. No significant effects
on the processivity of RNA- or DNA-dependent DNA
polymerization were seen in our original studies
of NNRTI-resistant mutants. This was surprising,
given the close proximity of these mutations to
the polymerase active site. In order to more
definitively determine whether these mutations
have any effects on polymerization, we studied
the pre-steady state kinetics of single
nucleotide incorporation by wild-type and mutant
RTs. These studies have demonstrated that P236L
has a slowed turnover rate (kss) during single
nucleotide incorporation on a DNADNA
templateprimer (Domaoal et al., 10th CROI,
abstract 612). Thus, this mutant has
abnormalities in both DNA-dependent DNA
polymerization and RNase H cleavage. We
previously identified a clinical RT sequence,
(8-26) which contains the P236L mutation, but
which confers improved replication kinetics
relative to pNL4-3 with P236L in the absence of
drug. This RT sequence has 24 amino acid changes
relative to P236L/NL4-3. We wanted to determine
what biochemical changes were associated with the
ability of this clinical RT backbone to
compensate for the replication defect conferred
by P236L.
L517V, S519N
T369V, A376V, W410L
Kd (P/T) P/T Dissociation Constant (nM) Kd
(dNTP) dNTP Dissociation Constant (uM) kpol
Maximum Rate of Incorporation (s-1) kss
Steady-state rate of polymerization (s-1) kss in
wild-type RT is determined by the rate of
dissociation from product
A288S,E297A, G335D
E138A, I142T
G550K, A554S

DNA 3'-end-directed RNase H (Polymerase-dependent)
16
21
100
80

• SUMMARY
• The P236L mutant in an NL4-3 RT background
  demonstrates a 4-fold slowing in kss, the
  turnover rate during DNA-dependent
  polymerization. This rate is the rate-limiting
  step of polymerization, and therefore would be
  expected to slow the overall rate of
  polymerization by the P236L RT.
• The P236L mutant is also slowed in both
  polymerase-dependent and -independent modes of
  RNase H cleavage.
• NL4-3 containing the P236L/8-26 clinical RT
  replicates more rapidly than NL4-3 with a
  P236L/NL4-3 RT. K103N in 2 RT clones from the
  same patient replicated more slowly than
  P236L/NL4-3 RT.
• The P236L/8-26 clinical RT demonstrates no
  improvement in rates of RNase H cleavages
  relative to P236L/NL4-3 RT.
• The P236L/8-26 clinical RT has an increase in the
  turnover rate (kss) of dGTP incorporation into a
  DNADNA templateprimer, relative to P236L NL4-3
  RT.
• We postulate that the improved replication
  capacity conferred by the P236L-containing 8-26
  RT is due to an improvement in the turnover rate
  of polymerization. The most plausible
  explanation for this is an increase in the rate
  of product dissociation during polymerization.
  The moderate reduction in replication capacity of
  the 8-26 RT-containing virus relative to
  wild-type virus is most likely due to the
  residual abnormalities in RNase H cleavage.
• Pre-steady state kinetic studies can be used
  successfully to characterize the biochemical
  mechanisms underlying changes in replication
  capacity in clinical RT backbones. These studies
  also confirm the importance of the slowed kss of
  single nucleotide incorporation in contributing
  to the reduced replication kinetics of this
  NNRTI-resistant mutant.

60
40
of substrate remaining
4
3
2
1
RNA 5'-end-directed RNase H (Polymerase-independen
t)
METHODS RT expression and purification. Wild-type
and mutant RTs were expressed in E.coli using the
pRSET expression vector. p66 and p51 subunits
were expressed separately. Proteins were
purified by metal affinity and ion exchange
chromatography to gt95 homogeneity. Steady-state
RNase H cleavage. Polymerase-dependent (DNA
3'-end-directed) RNase H activity was assayed
using a 5-end- 32P-labelled 41 nucleotide (nt)
RNA annealed to a shorter DNA, with the DNA
3-end recessed relative to the RNA 5-end.
Polymerase-independent (RNA 5'-end-directed)
RNase H activity was assayed using the same RNA
hybridized to a longer DNA, with the RNA 5-end
recessed relative to the DNA 3-end. RNase H
reactions were carried out in the absence of
dNTPs and RT input was normalized for
steady-state polymerization specific activity on
a homopolymeric RNADNA templateprimer. Active
Site Titration and Determination of Turnover Rate
of Single Nucleotide Incorporation (kss). Rapid
chemical quench experiments were performed using
a KinTek Instruments Model RQF-3 rapid quench
flow machine. A pre-incubated complex of 100nM RT
and 300nM 5end labeled DNADNA primertemplate
was rapidly mixed with an equal volume of 10mM
Mg2 100?M dGTP at times ranging from 0.025 to
5.50 sec (all concentrations represent the final
concentration after mixing). Reactions were
quenched with 0.5M EDTA. Products were analyzed
by denaturing PAGE and quantitated by
PhosphorImaging. Data were fitted to a burst
equation product A1 exp(-k1t) ksst,
where A represents the amplitude of the burst,
which corresponds to the concentration of active
enzyme. Time is represented by t, the
first-order rate constant for dNTP incorporation
is represented by k1, and the steady state rate
constant is represented by kss. The mean
amplitude and standard deviations were calculated
from 3 independent experiments.
100
80
60

• PLANNED STUDIES
• Compare the effects of P236L, K103N, and 8-26 RTs
  on the turnover rate of single nucleotide
  incorporation using an RNADNA templateprimer.
• Study the effects of the RT backbone from patient
  8 on K103N polymerization and RNase H cleavage
  to determine why replication kinetics of this
  patients K103N-containing clinical RTs are
  slowed.
• Extend these studies to clinical RTs containing
  G190S and K103NL100I, which both have reduced
  replication fitness in an NL4-3 backbone.

40
4
3
2
1
Time (min)