1 The emergence of different resistance mechanisms towards

1

The emergence of different resistance mechanisms towards nucleoside

inhibitors is explained by the properties of the wild type HIV-1 reverse

transcriptase

Catherine Isel§, Chantal Ehresmann, Philippe Walter, Bernard Ehresmann and Roland Marquet

UPR9002 du CNRS, IBMC, 15 rue René Descartes, 67084 Strasbourg cedex, France

§To whom correspondence should be addressed: Catherine Isel

Tel: 00 33 (0)3 88 41 70 40/ Fax : 00 33 (0)3 88 60 22 18 / [email protected]

Copyright 2001 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on October 19, 2001 as Manuscript M108352200 by guest on A

pril 7, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

2

Nucleoside reverse transcriptase inhibitors (NRTIs) represent one of the main drug families

used against AIDS. Once incorporated in DNA, they act as chain terminators, due to the lack of

a 3' hydroxyl group. As for the other anti-HIV-1 drugs, their efficiency is limited by the

emergence of resistant viral strains. Unexpectedly, previous studies indicated that resistance

towards NRTIs is achieved via two distinct and generally exclusive mechanisms. Resistance

mutations either decrease the efficiency of NRTIs incorporation, or increase their excision from

the extended primer. To understand the emergence of different resistance mechanisms

towards a single inhibitor class, we compared the incorporation and the pyrophosphorolysis of

several NRTIs using wild-type reverse transcriptase (WT RT). We found that the efficiency of

discrimination or excision by pyrophosphorolysis in the presence of nucleotides of a given

NRTI is a key determinant in the emergence of one or the other resistance pathway. Indeed, our

results suggest that the pathway by which RT become resistant towards a given NRTI can be

predicted by studying the inhibition of WT RT, because the resistance mutations do not confer

new properties to the mutant enzyme, but rather exacerbate pre-existing properties of the WT

enzyme. They also help to understand the low cross-resistance towards d4T observed with the

AZT-resistant RT.

by guest on April 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

3

Introduction

Reverse transcription is a key event in the replication cycle of retroviruses. The virally encoded

reverse transcriptase (RT), an RNA- and DNA dependent DNA polymerase, converts the viral (+) RNA

genome into a double stranded DNA (1) that will integrate into the host genome. Reverse

transcriptase (RT) of human immunodeficiency virus type 1 (HIV-1), the causative agent of the

acquired immunodeficiency syndrome (AIDS), is the target for two families of therapeutic: non

nucleoside inhibitors (NNRTIs) and nucleoside analogues (NRTIs).

Amongst the group of nucleosides used in multi-therapies, two compounds are analogues of

thymidine, 3’-azido-3’ deoxythymidine (AZT or zidovudine) and 2’,3’didehydro-2’,3’–dideoxythymidine

(d4T or stavudine), two are cytidine analogues , 2’,3’ dideoxycytidine (ddC or zalcitabine) and β-L-(-)-

2’,3’-dideoxy-3’-thyacytidine (3TC or lamivudine), one is an adenosine analogue, 2’, 3’-dideoxyinosine

(ddI or didanosine), and one is a guanosine analogue, abacavir. Those nucleoside analogues are

metabolically activated by host cellular kinases to their corresponding triphosphate forms (for review

see (2)) which are incorporated into the DNA by HIV-1 RT. The base moiety of didanosine and

abacavir are also modified during this process, generating ddATP and ddGTP, respectively (for an

overview, see http://www.niaid.nih.gov/daids/dtpdb/fdadrug.htm). Due to the lack of a 3’ OH group on

the ribose ring, NRTIs act as chain terminators by blocking further elongation of the nascent DNA,

leading to inhibition of viral replication. The efficiency of a nucleoside as an inhibitor and hence the

effectiveness of the therapy depends on i) the cellular uptake of the compound and its activation into

the triphosphate form ii) the incorporation of the analogue into the DNA and iii) the removal of the

incorporated chain terminator.

Prolonged use of NRTIs in clinical treatment and in cell culture of HIV-1 invariably give rise to

resistant viruses bearing substitutions in the pol gene. Patients treated with AZT display a set of up to

6 mutations in the pol gene involving M41L/D67N/K70R/L210W/T215F or T215Y/K219Q which confer


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

4

a >100-fold AZT resistance to HIV-1. When d4T is given as the only drug to patients naive to AZT, RT

bearing a V75T substitution is selected in 10% of the cases (3). Notably, the most important resistance

mutations associated to treatment with d4T are in fact AZT-associated resistance mutations (4).

Resistance to ddC, ddI and 3TC are conferred by single mutations in the pol gene, which map to

K65R (5,6), L74V (7), located in the finger subdomain of RT, and M184V (8-10), in the catalytic site,

respectively.

In principle, the resistance mutations could act either by i) decreasing the incorporation

efficiency of the triphosphate form of the NRTI or ii) increasing the removal of the incorporated NRTI.

In fact, the two latter mechanisms are now well documented.

Retroviral RTs lack 3’-exonuclease proof-reading activitiy (11) but are capable of

pyrophosphorolysis, the reversal reaction of polymerisation, releasing an unblocked, extensible DNA

chain and the analogue triphosphate (12,13). It has been reported that AZT resistance mutations lead

to enhanced excision of AZT from the nascent DNA strand by pyrophosphorolysis (14-16), although

this observation was not confirmed by other groups (17,18). More recently, it was shown that the AZT-

resistant enzyme, can efficiently unblock AZT-terminated primers by transfer of the chain terminator to

a nucleoside triphosphate, most likely ATP in vivo, in a reaction similar to pyrophosphorolysis (16,18-

20). In addition, it has been shown that AZT-resistant RT binds AZTMP-terminated primers more

tightly than WT RT does (21). Moreover, removal by ATP-lysis of d4TMP from the primer terminus is

more efficient with AZT-resistant enzyme as compared to WT RT, but only when the concentration of

the next incoming nucleotide is low (19).

Most of the resistance mutations associated with the other nucleoside analogues cluster into a

different mechanism, whereby the acquisition of the mutation interferes with nucleoside incorporation

into the DNA. Resistance to 3TC, due to M184V/I mutations in the RT, most likely involves steric

hindrance, decreasing either binding of the analogue (22,23) or the rate of incorporation (24). Recent


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

5

structural data show that the incoming 3TCTP is able to form a close complex with the primer/template

(P/T):RT which is however less stable than in the presence of the normal nucleotide (25). In addition, it

was shown that the 3TC-resistant enzyme has very poor pyrophosphorolysis and ATP-lysis properties

on 3TCMP- or AZTMP-terminated primers (16). Similarly, mutations K65R (26) and L74V (27) also

confer resistance to RT by decreasing the incorporation efficiency of NRTIs. Finally, the molecular

mechanism underlying d4T resistance has very recently been investigated, showing that the V75T

mutation changes both nucleotide selectivity and repair of d4TMP-terminated DNA chains by PPi, but

not ATP (28), unlike the situation with the AZT-resistant enzyme.

In order to understand the emergence of different resistance mechanisms towards a single

class of inhibitors, we performed a study on AZTTP, d4TTP, ddCTP, 3TCTP and ddATP, the

metabolised products of inhibitors used in the clinic, using WT HIV-1 RT. We conducted a comparative

study of the incorporation and the pyrophosphorolysis, in the presence or absence of the next

incoming nucleotide, of these NRTIs. We found that the efficiency of NRTI discrimination or excision

by pyrophosphorolysis in the presence of the next incoming nucleotide by WT RT was a key

determinant in evolution towards one or the other resistance pathways. Indeed, our results suggest

that the pathway by which RT will evolve to become resistant towards a given inhibitor can be

predicted by studying the inhibition of WT RT by this inhibitor. In addition, comparison of our results

with those published by others (6,14-16,18-20,24,27-29) using resistant RTs indicate that the

resistance mutations do not confer new properties to the mutant enzyme, but rather exacerbate pre-

existing properties of the WT enzyme towards a given inhibitor. They also help to understand the

existence of low cross-resistance towards d4T observed with the AZT-resistant RT (30).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

6

MATERIALS AND METHODS

Templates, primers and RTs

Viral RNA encompassing nucleotides 1-311 of HIV genomic RNA (Mal isolate) was used as a

template. It was synthesised by in vitro transcription and purified as previously described (31). We also

used as a template a 38-mer oligodeoxyribonucleotide (DPBS20) carrying, in addition to the primer

binding site (PBS) sequence, 20 nucleotides of the HIV-1 Mal sequence upstream of the PBS.

Natural tRNA3Lys, used as a primer, was purified from beef liver as previously described (32).

After dephosphorylation with calf intestine phosphatase, it was labelled at the 5’ end with [γ-32ATP] and

phage T4 polynucleotide kinase. An 18-mer oligodeoxyribonucleotide (ODN) complementary to the

PBS, labelled at the 5’ end with phage T4 polynucleotide kinase, was also used as a primer.

Chimerical primers, consisting of analogue-terminated-ODN primers were prepared as follows: 5’ end-

labelled ODN was hybridised to DPBS20 in 100 mM NaCl for 20 min at 70°C prior to extension at

37°C for 1 hour with 7.5 µM RNase H(-) HIV-1 RT using the appropriate set of dNTP and chain

terminator (AZTTP, d4TTP, ddTTP, 3TCTP, ddCTP and ddATP), at 0.2 mM each, in 50 mM Tris-HCl

(pH 8), 50 mM KCl, 6 mM MgCl2, 1 mM DTE, to obtain the expected analogue-terminated primer. After

proteinase K treatment and phenol/chloroform extraction, the chimerical primers were ethanol

precipitated and purified on 15% denaturing polyacrylamide gels.

Plasmids used for production of WT and RNase H(-) HIV-1 RT, bearing the E478Q mutation

were kindly provided to us by Dr. Torsten Unge (Uppsala, Sweden), together with the protocols for

protein over expression and purification.

For primer/template (P/T) formation, the primers were quantitatively heat-annealed with RNA

or DNA templates at a 1:2.5 ratio, as described previously (33).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

7

RT assays

Minus strand strong-stop DNA synthesis. Ten nM of P/T (tRNA3Lys /1-311 RNA or ODN/1-311

RNA) were pre-incubated at 37°C for 4 min with 20 nM HIV-1 RT (WT or RNase H (-)) in 50 mM Tris-

HCl (pH 8), 50 mM KCl, 6 mM MgCl2, 1 mM DTE. Reverse transcription was initiated by addition of 50

µM of each dNTP, in the presence or absence of 5 µM chain terminator (AZTTP, d4TTP, ddTTP,

3TCTP, ddCTP and ddATP) and 150 µM PPi. The reaction was stopped at various times ranging from

1 min to 3 hours by addition of one volume of formamide containing 50 mM EDTA, and the reaction

products were analysed on 8% denaturing polyacrylamide gels and quantified with a BioImager BAS

2000 (Fuji).

+1 rescue of DNA synthesis from analogue-terminated primers. Ten nM of analogue-

terminated primer/template were pre-incubated at 37°C for 4 min with 200 nM of RNase H(-) HIV-1

RT. Reactions were initiated by addition of 150 µM PPi, 50 µM of the dNTP corresponding to the

analogue present at the end of the chain and 50 µM of the next complementary ddNTP, allowing

synthesis of the +1 product with respect to the analogue. Reactions were stopped at various times

from 30 s to 3 hours and the products analysed as described above.

Primer unblocking by pyrophosphorolysis. Pyrophosphorolysis reactions were performed

essentially as described above, except that the reaction was initiated by addition of 150 µM PPi to the

P/T/RT complex, in the absence of nucleotides.

Electrophoretic Mobility Shift Assay

The 5’ end-labelled, chain-terminated primer (ODN)/template (DPBS20) were incubated for 10

min at room temperature with a 25 fold excess of RNase H(-) RT and increasing concentrations of the

next incoming dNTP (Fig. 2), in 40 mM Hepes (pH 7.5), 20 mM MgCl2, 60 mM KCl, 1 mM DTE, 2.5%

glycerol and 100 µl/ml Bovine Serum Albumin. The reaction mixture was further incubated at 37°C for


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

8

5 min, after the addition of KCl (to 100 mM final) and poly(rA)/oligo(dT) to 0.3 OD260 unit/ml. The

mixture was then cooled on ice prior to analysis on a 6% non-denaturing polyacrylamide gel. Both the

gel and the migration buffer contained 45 mM Tris-borate (pH 8.3) and 50 mM KCl.

Quantification and curve fitting.

For all experiments, quantification of the radioactivity and curve fitting were performed with

MacBAS 2.5 (Fuji) and IGOR Pro 3.1 (WaveMetrics, Inc.) softwares, respectively.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

9

RESULTS

A number a studies compared the relative incorporation of NRTIs and their natural

counterpart. Indeed, it has been shown that wild-type (WT) HIV-1 RT barely discriminated between

dTTP and AZTTP (34,35) or d4TTP (36). On the contrary, ddCTP was incorporated less efficiently

than dCTP (22), and incorporation of 3TCTP was even worse (22,24). Finally, ddATP was also shown

to be a moderately efficient substrate for WT HIV-1 RT (37). However, these data were obtained using

different experimental conditions and P/T complexes and can hardly be used to compare the inhibitory

efficiency of these nucleotide analogues. In addition, to our knowledge, no detailed pre-steady state

kinetics of the pyrophophorolysis of NRTIs has been performed, and thus, the available data

(16,18,19) are hardly amenable to quantitative comparison. Therefore, we undertook a comparative

study of the incorporation and pyrophosphorolysis of the available NRTIs by WT HIV-1 RT, with the

aim of understanding the emergence of two distinct resistance mechanisms for this class of RT

inhibitors.

Minus strong-stop DNA synthesis in the presence of nucleoside analogues

First, we performed (-) strand strong-stop DNA synthesis experiments using as a template a

viral RNA encompassing the first 311 nts of the 5’ end of the HIV-1 Mal genome and as a primer either

the natural tRNA3Lys or a DNA oligodeoxyribonucleotide complementary to the PBS (ODN). Using ODN

as a primer allows DNA synthesis to start immediately in the processive elongation mode (38,39). In

the absence of nucleoside analogues and PPi, when tRNA3Lys was used as a primer (Fig. 1A, left

panel), reverse transcription proceeded as previously described (38). During the initiation phase, i.e.

the addition of the first 6 nucleotides to the tRNA3Lys, the +3 and +5 products accumulated at short

incubation times (1 min) but disappeared after prolonged incubation. Quantification of the gels


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

10

indicated that, after 1 to 3 hours of incubation, the amount of (-) strand strong-stop product

represented 14% of the total radioactivity.

When 5 µM of a given inhibitor was added to the reaction, taking it at 1/10 the dNTP

concentration, the inhibition rate varied (Fig. 1B, C, D, E and F, left panels). Calculation of the

percentage of final product in the presence of each analogue, relative to no inhibitors (Table I),

allowed us to classify the nucleoside analogues according to their efficiency of inhibition : AZTTP was

the most effective inhibitor, followed by ddCTP, d4TTP, ddATP and 3TCTP. The results followed the

same trend when ODN was used as a primer for reverse transcription (data not shown). Overall, these

results are qualitatively in keeping with the previous studies discussed above, except for ddCTP,

which was expected to be a significantly weaker inhibitor than d4T (22,36).

As expected, addition of the T-analogues to the reaction induced arrest of reverse

transcription where As were present in the template, e.g. at positions +2 and +5, or at positions 157,

155, 150, 147 (compare Fig. 1A, left panel, to Fig. 1B and C, left panels). C-analogues (3TC and ddC)

did not seem to be incorporated efficiently at position +1. However, they were undoubtedly

incorporated at position +4 and at the other expected sites on the template (Fig. 1D and E, left

panels). The same holds true for ddATP, which was incorporated at position +6 and further along,

where Ts are present in the template (Fig. 1F, left panel).

Addition of physiological concentrations of PPi (150 µΜ) (40) did not significantly affect the

synthesis of (-) strand strong-stop DNA in the absence of inhibitor (Fig. 1A, right panel), the amount of

this product representing 12% of the total radioactivity after 1 to 3 hours of incubation. At short

incubation times, the pauses of RT were more pronounced. However, theses pauses rapidly

disappeared with increased incubation times.

When PPi was added to the reaction in presence of the nucleoside analogues, synthesis of

longer DNA chains, including the (-) strand strong-stop product, was partially restored, with various


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

11

efficiencies depending on the inhibitors (Fig. 1B, C, D, E and F, right panels). The highest level of DNA

repair observed was with the AZT/PPi combination. The rescue ratio, calculated as the ratio between

the percentage of (-) strand strong-stop DNA in the absence and presence of PPi for a given inhibitor

was 6 fold (Table I). When ODN was used as a primer, the rescue ratio was 30 fold (data not shown),

in agreement with previously published results showing that repair was not occurring during the

initiation step of reverse transcription (35). Consistently, the intensity of the band at position +2, within

the initiation complex, did not diminish with the addition of PPi to the reaction.

Using the same type of calculation, we determined the rescue ratio for the other nucleoside

analogues used in this assay (Table I), allowing us to asses their ability to be removed from the end of

the pr imer in the presence of the four natural dNTPs as fo l lows:

AZTTP>>d4TTP>ddCTP=3TCTP=ddATP. The same conclusions were drawn from experiments

performed with ODN as primer (data not shown).

+1 extension of T analogue-terminated primers in the presence of PPi

As the incorporation efficiency during (-) strand strong stop DNA differed among the various

NRTIs, the efficiency of primer unblocking and extension in the presence of PPi could not strictly be

compared in these experiments. To further address the issue of unblocking of analogue-terminated

DNA chains, and in order to reduce the complexity of the reaction to a minimum by monitoring the

rescue of DNA synthesis at only one position, we quantitatively prepared three T-analogue-terminated

primers, ODN-dC-AZTMP, ODN-dC-d4TMP and ODN-dC-dTMP (Fig. 2). Those primers were

hybridised to either 1-311 viral RNA or a DNA oligodeoxyribonucleotide, DPBS20 (see the Material

and Methods section) acting as a template. Hybrids were pre-incubated with HIV-1 RT and the repair

of T-analogue-terminated primers was initiated by addition of PPi, dTTP, the correct nucleotide to be


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

12

incorporated in place of AZTMP, d4TMP, or ddTMP, and ddGTP, corresponding to the next

complementary ddNTP, to stop the reaction at the following position on the template (Fig. 2).

The time course experiment presented in Fig. 3A shows that WT HIV-1 RT repaired

considerably more efficiently AZTMP- than d4TMP- or ddTMP-terminated primers. Indeed, synthesis

of the ODN-dC-dT-ddG product reached nearly completion (93%) after 1 hour incubation in the first

case, (Fig. 2A and B) while repair was much less efficient when ddTMP or d4TMP were the chain

terminators (44% and 23% after 1 hour incubation, respectively) (Fig. 3A and B). The experimental

data presented in Fig. 3A were fit to a first order rate equation and the rate constants of the repair

reaction were summarised in Table II. Interestingly, the repair rate constant of AZTMP-terminated

primers was 4 times higher than the one observed with the other T-analogue terminated primers. The

same results were obtained with the two types of hybrids described above.

Comparison of the pyrophosphorolysis rate of T-analogue terminated primers

In order to test whether diminished rates of repair of d4TMP- and ddTMP-terminated primers

as compared to AZTMP-terminated DNA chains were due to the presence of the next incoming

nucleotide, we compared the rate of pyrophosphorolysis of primers terminated by the chain

terminators cited above or by dTMP. The P/T hybrids were prepared as described previously, pre-

incubated with HIV-1 RT and the reaction was initiated by addition of PPi (Fig.2). As shown in Fig. 4A,

pyrophosphorolysis conducted in the absence of nucleotides revealed no important differences

between the four T-analogue-terminated primers. Quantification of the data and curve fitting (Fig. 4B)

further proved this statement, since the pyrophosphorolysis rate constants were all in the same range

(Table II). Remarkably, the T-analogues were removed as efficiently as the unmodified parent

nucleotide.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

13

Since the rate of T-analogue incorporation (34-36,41) is 4 to 5 orders of magnitude faster than

the overall repair reaction rates listed in the first column of Table II, the rate limiting step of the repair

reaction must be pyrophosphorolysis in the presence of the next incoming nucleotide (19). Thus,

differences between the two columns of Table II correspond to the effects of the next incoming

nucleotide on pyrophosphorolysis. Our results indicate that primer unblocking, allowing subsequent

repair, was inhibited about ten fold by the presence of the next incoming nucleotide in the case of

d4TMP- and ddTMP- terminated DNA chains, but only a two fold when AZT was used as inhibitor.

Repair and pyrophosphorolysis of the C- and A-analogue terminated primers

We next used the same experimental strategy to investigate the repair and unblocking of C-

and A-analogue-terminated primers (Fig. 2). As shown in Table II, the initial rate constants for the

combined removal reaction and further extension of C-analogue terminated primers were similar for

3TC and ddC-terminated primers. These values were slightly lower than the ones obtained previously

with d4T and ddT.

In addition, the initial rate constants for pyrophosphorolysis were also in the same range when

comparing the two C-analogues. Noticeably, these rates were 5-8 fold lower as compared to dCMP-

terminated primers and around 10 fold lower than those measured in the case of T-analogues.

Contrary to what we observed with the T-analogues, we did not detect any significant inhibition of the

pyrophosphorolysis of the C-analogues in the presence of the next incoming nucleotide (Table II).

The last family of inhibitors we studied were A-analogues. ddATP is the activated triphosphate

form of ddI, given as such to the patients. Surprisingly, repair of ddAMP-terminated primers was hardly

detectable (Table II) even though pyrophosphorolysis was taking place at a rate higher than the one

obtained for ddCTP-terminated primers (Table II). Pyrophosphorolysis of a dAMP-terminated primer


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

14

was also measured (Table II). Similarly to the situation with dCTP, the rate constant we found for that

reaction was also lower than the one obtained with dTTP.

The difference between T analogue- and other dNMP-terminated primers that we point out in

this pyrophosphorolysis assay could either be due to an intrinsic difference in the pyrophosphorolysis

rate linked to the nature of the nucleotide, or could reflect a context dependent phenomenon.

Addressing this issue would need the construction of mutant templates that would permit the different

end of the primers to be replaced in the same sequence context. This is of interest but was not the aim

of our study.

Efficiency of P/T:RT:dNTP complex formation

The results presented so far, together with similar experiments performed by others using

resistant RTs (14-16,18,19), suggest that the efficiency of a NRTI depends not only on its efficiency to

be incorporated and removed from the end of the DNA chain, but also on the sensitivity of the removal

reaction to the inhibition by the next incoming nucleotide. The latter observation is biologically

relevant, since the “incoming dNTP” used in our experiments is representative for the pool of dNTP

present in cells. It was suggested that analogue-terminated primers can follow two distinct pathways

(18): either the next incoming nucleotide does not bind efficiently and the removal reaction via

pyrophosphorolysis or ATP-lysis will be favoured, or, on the contrary, the repair reaction is impaired by

the next incoming nucleotide binding efficiently in a quaternary P/T:RT:dNTP complex, resulting in the

formation of a so called dead-end complex (DEC) (18,19,42). These DEC, formed at low salt

concentration, remain stable in a “higher” salt buffer (100-150 mM KCl) and can be visualised on non-

denaturing polyacrylamide gels. The increased stability is due to the closed conformation of the

polymerase, where the fingers of HIV-1 RT fold onto the P/T, once the incoming nucleotide is bound

(43).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

15

According to our results, the primer rescue pathway should be favoured when AZTTP is the

chain terminator, with AZTMP-terminated P/T:RT:dGTP complexes being unstable whereas ddAMP-

terminated complexes should be amongst the most stable ones. To test this hypothesis, formation of

DEC was investigated for the three T-analogue terminated P/Ts. Incubation of HIV-1 RT with the

different hybrids and with increasing concentrations of dGTP, the next complementary nucleotide (Fig

2) showed that DEC was formed less readily by AZTMP-terminated primers as compared to the two

other T-analogue-terminated primers (Fig. 5A). Quantification of these data revealed that the most

stable complex was obtained with d4TMP as a chain terminator, followed by ddTMP and AZTMP (Fig.

5B). These results are in agreement with the +1 rescue experiments previously described. We next

tested the ability of ddAMP-terminated P/T to form the closed ternary complex in the presence of

dGTP. As predicted, this complex was highly stable (Fig. 5B): at the highest dGTP concentrations

tested, the same amount of complex was obtained as compared to d4T, but at low concentrations of

incoming dGTP (10 µM), the ddAMP-terminated P/T complex was significantly more stable. Finally,

our results also showed that a ddCMP-terminated P/T formed a weakly stable DEC, in keeping with

the absence of inhibition of the pyrophosphorolysis of the corresponding primer by the next incoming

nucleotide. Unexpectedly, we were unable to detect any DEC formation when the primer was

terminated by 3TCMP.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

16

DISCUSSION

NRTIs represent the largest familly of drugs approved for treatment of AIDS. Even though

multi-therapy treatments combining NRTIs, NNRTIs and protease inhibitors are relatively efficient,

they fail to totally eradicate the virus from the body of infected patients (44). The major problem is the

emergence of viral strains resistant to the inhibitors as a result of point mutations in the gene encoding

the viral RT and protease. Numerous studies using recombinant RTs bearing the NRTI resistance

mutations have been performed. Their general goal was to test whether the in vitro assays with the

mutant RTs could account for the resistance observed in vivo and in cell culture, and hence, to

elucidate the resistance mechanism(s). From these studies, two distinct resistance pathways towards

NRTIs emerged. Some RT mutations lead to enhanced excision of the analogue (principally AZT, and

to a lesser extent d4T) from the nascent DNA strand (14-16,18-20,28); other mutations (mainly single

point mutations) reduce the efficiency of incorporation of the nucleotide into the DNA (24-27). The

goal of the present study was different. Our aim was to understand why two different resistance

mechanisms emerged against a single class of inhibitors. More specifically, we wanted to know

whether the mechanism by which RT becomes resistant towards a particular NRTI could be predicted

from in vitro studies using WT RT. Therefore, we compared the incorporation and the

pyrophosphorolysis, by WT RT, of several NRTIs used in the clinic.

We first compared the efficiency of inhibition of NRTIs in an in vitro assay of (-) strand strong

stop DNA synthesis. In the absence of PPi, the efficiency of DNA synthesis inhibition was

AZT>>ddC>d4T>ddA>3TC. Interestingly, with the exception of ddC, the NRTIs for which resistance

arise via a decreased incorporation are those that are the less efficiently incorporated in our assay. In

agreement with our data, Arts et al. (45) also found that ddC efficiently inhibits (-) strand strong stop

DNA synthesis. However, pre-steady state kinetics showed that WT HIV-1 RT very efficiently

discriminates against ddCTP on a DNA template (22). Therefore, it appears that if WT RT efficiently


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

17

discriminates a nucleoside analogue from its natural counterpart, resistance is very likely to evolve by

further increasing this discrimination. Obviously, the mutation(s) increasing the negative discrimination

might differ from one analogue to another and cannot be predicted from the simple experiments

described here. On the contrary, AZT, which is the most efficient NRTI, because WT RT hardly

discriminates between AZTTP and dTTP (34,35,41), gives rise to another resistance mechanism.

By comparing the inhibition of (-) strand strong stop DNA synthesis by NRTIs in the absence

and presence of physiological concentrations of PPi (40), we could evaluate the efficacy of the rescue

provided by pyrophosphorolysis. Interestingly, in this assay, WT RT repaired efficiently the AZTMP-

terminated primers and moderately the d4TMP-terminated primers, while those terminated by other

NRTIs were very poorly repaired. Thus, the NRTIs that select resistance mutations enhancing the

removal of the nucleoside analogue from the primer termini (AZT and d4T) (14-16,18-20,28) are

precisely those that can be efficiently excised by WT RT by pyrophosphorolysis. Indeed, our study

indicates that AZT and d4T are the only NRTIs used in the clinic that were pyrophosphorolysed as

efficiently as their natural counterpart. In other words, if WT RT efficiently excised an incorporated

NRTI in the presence of the complete pool of natural dNTPs, it will select resistance mutations that

enhance repair of the blocked primers, rather than increasing counter-selection of this substrate.

Thus, our results strongly suggest that the mechanism by which HIV-1 will become resistant to

a given NRTI can be predicted from in vitro inhibition studies conducted with WT RT in the absence

and presence of PPi. This finding implies that resistance mutations exacerbate pre-existing properties

of WT RT, rather than conferring new properties to the mutant polymerase.

A particular case is that of d4T, which is at the border between the two resistance

mechanisms. Our results, as well as previous data, indicate that WT RT very poorly discriminates

against d4T (36), and that d4TMP is poorly excised from the primer termini by the WT polymerase in

the presence of rather high concentrations (50 µM) of dNTP (19). Strikingly, resistance against d4T is


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

18

mediated either by mutations enhancing excision (M41L/D67N/K70R/L210W/T215F or T215Y/K219Q)

(4) or by a mutation enhancing discrimination (L75V) (28). Interestingly, both resistance mechanisms

proved rather inefficient (<2 and 8 fold, respectively), as compared to the resistance provided by the

M41L/D67N/K70R/L210W/T215F or T215Y/K219Q mutations towards AZT (180 fold), or by the

M184V mutation towards 3TC (>100 fold) (http://resdb.lanl.gov/Resist_DB/default.htm). Once again,

these observations are in keeping with our proposal that resistance mutations only increase the pre-

existing capabilities of WT RT rather than creating new ones. Starting with poor excision and

discrimination capabilities of WT RT towards d4T, mutations can only provide limited resistance. On

the contrary, high basal discrimination (against 3TC) or excision (of terminal AZTMP) by the WT RT

allow mutations to confer a high level of resistance.

Our results also help to understand the cross-resistance between d4T and AZT. It had been

previously concluded, from phenotypic assays obtained from cell culture, that AZT-resistant HIV-1 is

not cross-resistant to d4T (46,47). However, d4T treatment selects for AZT resistance mutations (for

review, see (48)), and prior exposure of patients to AZT reduces the efficiency of subsequent

treatment with d4T (30). However, our results, as well as other (19), showed that the excision

mechanism, which is enhanced by the AZT-resistance mutations, might be largely underestimated in

the phenotypic assays performed at high nucleotide concentration, while this mechanism might be

efficient in cells with low dNTP pools.

Parniak and co-workers suggested that the M41L/D67N/K70R/L210W/T215F or T215Y/K219Q

mutations provide resistance towards AZT by increasing pyrophosphorolysis. More recently, Scott and

co-workers demonstrated that the resistant RT is able to unblock the primer strand by transferring the

terminal AZTMP to ATP, whereas this reaction is inefficient with WT RT (18,20). However, we observe

a remarkable parallel between our study of the PPi-mediated primer unblocking by WT RT, and the

ATP-mediated unblocking by the AZT-resistant RT studied by Scott and co-workers (18-20). Indeed,


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

19

they showed that, in the absence of the next incoming nucleotide, ATP-mediated unblocking is

efficient with both AZT- and d4T-terminated primers, but that addition of the next nucleotide strongly

inhibits removal of d4TMP (19). Similarly, ATP-lysis of terminal ddAMP by the AZT-resistant RT was

efficiently inhibited by DEC formation (18). We observed exactly the same pattern of inhibition of

pyrophosphorolysis by the incoming nucleotide in the presence of WT RT. All together, these results

showed that the AZT-resistance mutations, while favouring the use of the most abundant substrate

(ATP, instead of PPi), essentially do not affect the fundamental features of the unblocking reaction.

Interestingly, molecular modelling, based on the crystal structure of a ternary P/T:RT:incoming

dNTP complex (43), also indicated that the specificity of the excision reaction for AZTMP-terminated

primers is not due to the mutations that confer resistance, but depends instead on the structure

around the RT polymerase active site (49). Based on their model, Boyer et al. proposed that the P/T

complex is in equilibrium between polymerisation (P) and nucleotide (N) sites of RT. PPi or ATP

mediated excision could take place only when the P/T is in the N site. In the case of AZTMP-

terminated primers, the bulky azido group on the ribose ring makes a steric clash with the catalytic

Mg2+ bound to D185 when in the P site, thus favouring binding in the N site, and hence nucleotide

excision (49). This model correctly accounts for our pyrophosphorolysis data obtained in the presence

of the next nucleotide. However, it would predict that pyrophosphorolysis of AZTMP should also be

more efficient than that of d4TMP or ddTMP in the absence of nucleotides, contrary to our

observations. In addition, since the next complementary nucleotide and PPi (or ATP) compete for

binding to the same Mg ions in the catalytic site, (43), one would expect that all excision reactions

should be, at least partially, inhibited by the next incoming nucleotide. However, we observed that

pyrophosphorolysis of terminal ddCMP and 3TCMP took place essentially at the same rate in the

absence and presence of the next nucleotide. All these data show that although the excision

mechanism is being progressively better understood, the phenomenon is not fully elucidated yet.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

20

Understanding the behaviour of WT HIV-1 RT towards the existing inhibitors provides

important insights into the mechanism by which mutations selected during drug therapy can give rise

to drug resistant forms of the enzyme. Such experiments are important in order to predict the evolution

towards resistance and the possible cross-resistance. They are also indispensable for the evaluation

of new inhibitors and for rational design of novel nucleosides.

Acknowledgements

This work was supported by a grant from the “Agence Nationale de la Recherche contre le Sida” and a

“Jeunes Equipes” grant from the Centre National de la Recherche Scientifique to R.M. We thank M.

Rigourd for helpfull discussions, T. Unge who kindly provided us the clones for HIV-1 RT and G. Bec

and G. Keith for purification of tRNA3Lys.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

21

REFERENCES

1. Skalka, A. M., and Goff, S. P. (eds) (1993) Reverse Transcriptase, Cold Spring Harbor

Laboratory Press, N.Y.

2. Guedj, R. (1999) Virologie 3, 17-28

3. Coakley, E. P., Gillis, J. M., and Hammer, S. M. (2000) Aids 14(2), F9-15.

4. Lin, P. F., Gonzalez, C. J., Griffith, B., Friedland, G., Calvez, V., Ferchal, F., Schinazi, R. F.,

Shepp, D. H., Ashraf, A. B., Wainberg, M. A., Soriano, V., Mellors, J. W., and Colonno, R. J.

(1999) Antivir Ther 4(1), 21-8

5. Gu, Z., Gao, Q., Fang, H., Salomon, H., Parniak, M. A., Goldberg, E., Cameron, J., and

Wainberg, M. A. (1994) Antimicrob Agents Chemother 38(2), 275-81.

6. Zhang, D., Caliendo, A. M., Eron, J. J., DeVore, K. M., Kaplan, J. C., Hirsch, M. S., and

D'Aquila, R. T. (1994) Antimicrob Agents Chemother 38(2), 282-7.

7. St. Clair, M. H., Martin, J. L., Tudor-Williams, G., Bach, M. C., Vavro, C. L., King, D. M.,

Kellam, P., Kemp, S. D., and Larder, B. A. (1991) Science 253(5027), 1557-1559

8. Gao, Q., Gu, Z., Parniak, M. A., Cameron, J., Cammack, N., Boucher, C., and Wainberg, M. A.

(1993) Antimicrob Agents Chemother 37(6), 1390-2.

9. Schinazi, R. F., Lloyd, R. M., Jr., Nguyen, M. H., Cannon, D. L., McMillan, A., Ilksoy, N., Chu,

C. K., Liotta, D. C., Bazmi, H. Z., and Mellors, J. W. (1993) Antimicrob Agents Chemother

37(4), 875-81.

10. Tisdale, M., Kemp, S. D., Parry, N. R., and Larder, B. A. (1993) Proc Natl Acad Sci U S A

90(12), 5653-6.

11. Roberts, J. D., Bebenek, K., and Kunkel, T. A. (1988) Science 242(4882), 1171-3.

12. Hsieh, J. C., Zinnen, S., and Modrich, P. (1993) J Biol Chem 268(33), 24607-13


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

22

13. Reardon, J. E. (1993) J Biol Chem 268(12), 8743-51

14. Arion, D., Kaushik, N., McCormick, S., Borkow, G., and Parniak, M. A. (1998) Biochemistry

37(45), 15908-17

15. Arion, D., Sluis-Cremer, N., and Parniak, M. A. (2000) J Biol Chem 275(13), 9251-5.

16. Götte, M., Arion, D., Parniak, M. A., and Wainberg, M. A. (2000) J Virol 74(8), 3579-85

17. Carroll, S. S., Geib, J., Olsen, D. B., Stahlhut, M., Shafer, J. A., and Kuo, L. C. (1994)

Biochemistry 33(8), 2113-20

18. Meyer, P. R., Matsuura, S. E., Mian, A. M., So, A. G., and Scott, W. A. (1999) Mol. Cell 4(1),

35-43

19. Meyer, P. R., Matsuura, S. E., Schinazi, R. F., So, A. G., and Scott, W. A. (2000) Antimicrob

Agents Chemother 44(12), 3465-72.

20. Meyer, P. R., Matsuura, S. E., So, A. G., and Scott, W. A. (1998) Proc Natl Acad Sci USA

95(23), 13471-6

21. Canard, B., Sarfati, S. R., and Richardson, C. C. (1998) J Biol Chem 273(23), 14596-604

22. Feng, J. Y., and Anderson, K. S. (1999) Biochemistry 38(1), 55-63.

23. Wilson, J. E., Aulabaugh, A., Caligan, B., McPherson, S., Wakefield, J. K., Jablonski, S.,

Morrow, C. D., Reardon, J. E., and Furman, P. A. (1996) J Biol Chem 271(23), 13656-62.

24. Krebs, R., Immendorfer, U., Thrall, S. H., Wohrl, B. M., and Goody, R. S. (1997) Biochemistry

36(33), 10292-300

25. Gao, H. Q., Boyer, P. L., Sarafianos, S. G., Arnold, E., and Hughes, S. H. (2000) J Mol Biol

300(2), 403-18.

26. Gu, Z., Arts, E. J., Parniak, M. A., and Wainberg, M. A. (1995) Proc Natl Acad Sci U S A 92(7),

2760-4.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

23

27. Martin, J. L., Wilson, J. E., Haynes, R. L., and Furman, P. A. (1993) Proc Natl Acad Sci U S A

90(13), 6135-9.

28. Selmi, B., Boretto, J., Navarro, J. M., Sire, J., Longhi, S., Guerreiro, C., Mulard, L., Sarfati, S.,

and Canard, B. (2001) J Biol Chem 276(17), 13965-74.

29. Gu, Z., Quan, Y., Li, Z., Arts, E. J., and Wainberg, M. A. (1995) J Biol Chem 270(52), 31046-

51

30. Katlama, C., Valantin, M. A., Matheron, S., Coutellier, A., Calvez, V., Descamps, D., Longuet,

C., Bonmarchand, M., Tubiana, R., De Sa, M., Lancar, R., Agut, H., Brun-Vezinet, F., and

Costagliola, D. (1998) Ann Intern Med 129(7), 525-31.

31. Marquet, R., Baudin, F., Gabus, C., Darlix, J. L., Mougel, M., Ehresmann, C., and Ehresmann,

B. (1991) Nucleic Acids Res. 19(9), 2349-2357

32. Bénas, P., Bec, G., Keith, G., Marquet, R., Ehresmann, C., Ehresmann, B., and Dumas, P.

(2000) RNA 6(10), 1347-1355

33. Isel, C., Marquet, R., Keith, G., Ehresmann, C., and Ehresmann, B. (1993) J Biol Chem

268(34), 25269-25272

34. Kerr, S. G., and Anderson, K. S. (1997) Biochemistry 36(46), 14064-70

35. Rigourd, M., Lanchy, J. M., Le Grice, S. F., Ehresmann, B., Ehresmann, C., and Marquet, R.

(2000) J Biol Chem 275(35), 26944-51.

36. Vaccaro, J. A., Parnell, K. M., Terezakis, S. A., and Anderson, K. S. (2000) Antimicrob Agents

Chemother 44(1), 217-21.

37. Ueno, T., and Mitsuya, H. (1997) Biochemistry 36(5), 1092-9.

38. Isel, C., Lanchy, J. M., Le Grice, S. F. J., Ehresmann, C., Ehresmann, B., and Marquet, R.

(1996) EMBO J 15, 917-924


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

24

39. Lanchy, J. M., Ehresmann, C., Le Grice, S. F. J., Ehresmann, B., and Marquet, R. (1996)

EMBO J 15, 7178-7187

40. Barshop, B. A., Adamson, D. T., Vellom, D. C., Rosen, F., Epstein, B. L., and Seegmiller, J. E.

(1991) Anal Biochem 197(1), 266-72.

41. Vaccaro, J. A., and Anderson, K. S. (1998) Biochemistry 37(40), 14189-94

42. Tong, W., Lu, C. D., Sharma, S. K., Matsuura, S., So, A. G., and Scott, W. A. (1997)

Biochemistry 36(19), 5749-57.

43. Huang, H., Chopra, R., Verdine, G. L., and Harrison, S. C. (1998) Science 282, 1669-1675

44. Finzi, D., Hermankova, M., Pierson, T., Carruth, L. M., Buck, C., Chaisson, R. E., Quinn, T. C.,

Chadwick, K., Margolick, J., Brookmeyer, R., Gallant, J., Markowitz, M., Ho, D. D., Richman,

D. D., and Siliciano, R. F. (1997) Science 278(5341), 1295-300.

45. Arts, E. J., Marois, J. P., Gu, Z., Le Grice, S. F., and Wainberg, M. A. (1996) J Virol 70(2),

712-20.

46. Larder, B. A., Chesebro, B., and Richman, D. D. (1990) Antimicrob Agents Chemother 34(3),

436-41.

47. Lin, P. F., Samanta, H., Rose, R. E., Patick, A. K., Trimble, J., Bechtold, C. M., Revie, D. R.,

Khan, N. C., Federici, M. E., Li, H., and et al. (1994) J Infect Dis 170(5), 1157-64.

48. Schinazi, R. F., Larder, B. A., and Mellors, J. W. (2000) Int Antiviral News 8, 65-91

49. Boyer, P. L., Sarafianos, S. G., Arnold, E., and Hughes, S. H. (2001) J Virol 75(10), 4832-42.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

25

FOOTNOTES

The costs of publication of this article were defrayed in part by the payment of page charges. This

article must therefore be hereby marked “advertisment” in accordance with 18 U.S.C. Section 1734

solely to indicate this fact.

The abbreviations used are : AIDS, acquired immunodeficiency syndrome ; RT, reverse transriptase ;

HIV-1, human immunodeficiency type 1; NNRTI, non nucleoside inhibitor; NRTI, nucleoside inhibitor;

AZT, 3’-azido-3’ deoxythymidine; d4T, 2’,3’didehydro-2’,3’–dideoxythymidine; ddC, 2’,3’

dideoxycytidine; 3TC β-L-(-)-2’,3’-dideoxy-3’-thyacytidine; ddI, 2’, 3’-dideoxyinosine; AZTTP 3’-azido-3’

deoxythymidine 5’-triphosphate; d4TTP, 2’,3’didehydro-2’,3’–dideoxythymidine 5’-triphosphate; dTTP,

thymidine 5’-triphosphate; ddTTP, 3’-deoxythymidine 5’-triphosphate; dGTP, deoxyguanosine 5’-

triphosphate; 3TCTP, β-L-(-)-2’,3’-dideoxy-3’-thyacytidine 5’-triphosphate; ddCTP, 2’,3’ dideoxycytidine

5’-triphosphate; ddATP, 2’,3’ dideoxyadenosine 5’-triphosphate; PPi, inorganic pyrophosphate; P/T,

primer:template; AZTMP, 3’-azido-3’ deoxythymidine 5’-monophosphate; d4TMP, 2’,3’didehydro-

2’,3’–dideoxythymidine 5’-monophosphate; ddCMP, 2’,3’ dideoxycytidine 5’-monophosphate; 3TCMP,

β-L-(-)-2’,3’-dideoxy-3’-thyacytidine 5’-monophosphate; ddAMP, 2’,3’ dideoxyadenosine 5’-

monophosphate; PBS, Primer Binding Site; WT, wild type; dNTP, deoxynucleotide triphosphate; DTE,

; DEC, dead-end complex; P site, polymerisation site; N site, Nucleotide binding site.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

26

Legends to figures and tables

Figure 1: Minus strand strong-stop DNA synthesis in the presence of various NRTIs and PPi.

Ten nM of tRNA3Lys/1-311 viral RNA were pre-incubated with 20 nM of HIV-1 RT and the

polymerisation reaction was initiated by the addition of a mixture of the four dNTPs (50 µM each) in

the absence or presence of PPi (150 µM) and 5 µM AZTTP (B) , d4TTP (C) , 3TCTP (D) , ddCTTP (E)

and ddATP (F). The reaction was stopped after 1 min, 30 min, 1 hour and 3 hours.

Figure 2 : Experimental strategy. (A) Only part of the sequence of the 1-311 HIV-1 RNA used as a

template is shown. The PBS corresponds to nucleotides 179-196 of this RNA. Various analogue-

terminated primers, listed underneath, were purified prior to their use. (B) Binary complexes,

containing purified analogue-terminated primers hybridised to the RNA template were pre-incubated

with HIV-1 RT and used either in a primer-rescue experiment or in a direct pyrophosphorolysis

experiment, depending on the substrates added to the reaction mixture (PPi + dNTPs/ddNTP or PPi

alone, respectively).

Figure 3: PPi-dependent rescue of T-analogue terminated primers. (A) Ten nM of ODN-dC-

AZTMP(or ODN-dC-d4TMP, or ODN-dC-ddTMP)/1-311 viral RNA were incubated with 200 nM of HIV-

1 RT, and the reaction was initiated by the addition of 150 µM PPi, 50 µM dTTP and 50 µM ddGTP in

order to allow synthesis of the +1 product with respect to the NRTI-terminated primer. The reaction

was stopped after 15, 30, 60 s, 4, 10, 20, 30 and 60 min. (B) Quantification of the non-extended and

+1 extended products allowed calculation of the percentage of rescue. Experimental data were fit to

equation ODN A e k text+[ ] = • −( )− •3 1 , where [ODN+3] is the concentration of T-analogue


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

27

terminated primer extended by one nucleotide, A is the amplitude of the reaction, and kext is the

apparent repair rate constant. The curves shown correspond to fits with kextAZTMP = 1.88 10-3 s-1, kextd4TMP

= 0.46 10-3 s-1 and kextddTMP = 0.46 10-3 s-1.

Figure 4: Pyrophosphorolysis of T-analogue terminated primers. (A) Ten nM of ODN-dC-AZTMP

(or ODN-dC-d4TMP, or ODN-dC-ddTMP, or ODN-dC-dTMP)/1-311 viral RNA were incubated with 200

nM of HIV-1 RT, the reaction was initiated by the addition of 150 µM PPi and stopped after 6, 12, 18,

24, 30, 36, 42, 48, 54, 60 s, 3, 5,10 20, 30 and 60 min. (B) Quantification of the initial amount of primer

and the amount of product that has been pyrophosphorolysed allowed calculation of the percentage of

pyrophosphorolysis. Experimental data were fit to equation %pyro A ek tpyro= • −( )− •1 , where A is the

amplitude of the reaction and kpyro is the apparent rate constant of the pyrophosphorolysis reaction.

The curves shown correspond to fits with kpyroAZTMP = 3.35 10-3 s-1, kpyrod4TMP = 5.4 10-3 s-1, kpyroddTMP =

4.82 10-3 s-1 and kpyrodTMP = 4.2 10-3 s-1.

Figure 5: Stable complex formation between HIV-1 RT, chain-terminated P/T and the next

incoming nucleotide. (A) Dead-end complex (DEC) formation with AZTMP-, d4TMP- and ddTMP-

terminated P/Ts. The labelled T-analogue-terminated P/T (8 nM) were incubated with 200 nM of HIV-1

RT and dGTP, the next complementary nucleotide, in increasing concentrations ranging from 0.01 to

1000 µM. After 10 min of complex formation at room temperature, the salt concentration was

increased to 100 mM KCl and an unlabeled chase substrate, poly(rA)/oligo(dT), was added, for a

further 5 min incubation at 37°C. The free P/T and the dead-end P/T:RT:dGTP complexes were

separated by electrophoresis on a 6% non-denaturing polyacrylamide gel. (B) Quantification of the

data obtained in (A) and data obtained on ddAMP- and ddCMP-terminated P/T:RT:dNTP complexes.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

28

The amount of free P/T and DEC were quantified and the percentage of DEC formed was plotted as a

function of the concentration of the incoming nucleotide.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

29

Table I : Efficiency of inhibition of (-) strand strong-stop DNA synthesis and rescue ratio by

pyrophosphorolysis.

% of final product

relative to no

inhibitor

Rescue ratio

AZT 2.5%

AZT/PPi 15%

6x

d4T 8.4%

d4T/PPi 18%

2.14x

3TC 36%

3TC/PPi 48.6%

1.35x

ddC 4.6%

ddC/PPi 7%

1.52x

ddA 20%

ddA/PPi 24%

1.2x

The amount of (-) strand strong-stop DNA versus the total radioactivity was quantified in the absence

and presence of the various NRTIs and PPi. The percentage of final product relative to no inhibitor

was the ratio between the amount of (-) strand strong-stop DNA in the presence and absence of

inhibitor. The rescue ratio was calculated as the ratio between the percentage of final product in the

presence and in the absence of PPi, for a given inhibitor.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

30

Table II: Rate constants for PPi-dependent rescue and pyrophosphorolysis of analogue-

terminated P/Ts.

Primer/templateRepair

Kext (10-3) (s-1) ± SD

Pyrophosphorolysis

Kpyro (10-3) (s-1) ± SD

AZTMP-terminated 1.88 ± 0.24 3.35± 0.05

d4TMP-terminated 0.46 ± 0.27 5.4 ± 0.4

ddTMP-terminated 0.46 ± 0.12 4.82 ±0.28

dTMP-terminated 4.2 ± 0.4

3TCMP-terminated 0.33 ± 0.1 0.32 ± 0.07

ddCMP-terminated 0.35 ± 0.12 0.49 ±0.1

dCMP-terminated 2.5 ± 1.5

ddAMP-terminated not detected 0.69 ± 0.06

dAMP-terminated 1.32 ± 0.25

SD: Standard Deviation

kext is the rate of PPi dependent +1 rescue of analogue terminated primers. kpyro is the rate of

pyrophosphorolysis of analogue-terminated primers, in the absence of nucleotides. The fit were

performed as described in Fig. 3 and 4 , except that data points up to 3 hours were collected to ensure

that a plateau was reached in all cases.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

3h0

3h0

PP

i

A.

3h3h

00

d4Td4T

+P

Pi

A-rich loop

C.

03h

03h

ddCddC

+P

Pi

E.

3h3h

00

AZ

TA

ZT

+P

Pi

A-rich loop

B.

A174

A157

A155

A150

A147

3h

2h

00 3T

C+

PP

i3T

C

D.

G162

G160

G159

G156

+5

+3

tRN

A3 Lys

primer

(-) sstopD

NA

3H3H

00

ddA+

PP

iddA

F.

U173

U171

+6

Fig

ure 1, Isel et al.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

+PPi

+PPi+dCTP+ddTTP

UCUAGCAGUGGCGCCCGAACAGGGAC

C-ACCGCGGGCTTGTCCCTG ddC-ACCGCGGGCTTGTCCCTG 3TC-ACCGCGGGCTTGTCCCTG

AZT-CACCGCGGGCTTGTCCCTG d4T-CACCGCGGGCTTGTCCCTG ddT-CACCGCGGGCTTGTCCCTG T-CACCGCGGGCTTGTCCCTG

ddA-TCGTCACCGCGGGCTTGTCCCTG A-TCGTCACCGCGGGCTTGTCCCTG

ODN-3TCMPODN-ddCMPODN-dCMP

ODN-dC-AZTMPODN-dC-d4TMPODN-dC-ddTMPODN-dC-dTMP

ODN-dC-dT-dG-dC-dT-ddAMPODN-dC-dT-dG-dC-dT-dAMP

+6 +2 +1

+PPi

+PPi+dTTP+ddGTP

+PPi

+PPi+dATP+ddGTP

5' 3' Template

B. Analogue-terminated primer/1-311 viral RNA

+ HIV-1 RT (200 nM)

4 min 37°C

150 mM PPiPPi + dNTP corresponding tothe analogue+ next complementary ddNTP

Pyrophosphorolysis+1 rescue of DNA synthesis

Primer Binding SiteA.

Figure 2, Isel et al.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

PPi-dependent rescue of T-analogue terminated primers

0 1H

ODN-dC-AZTMP ODN-dC-d4TMP ODN-dC-ddTMP

0 1H 0 1H

+1

A.

B.

100

80

60

40

20

0

[% O

DN

+3

]

40003000200010000time (sec)

AZT d4T ddT



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

0 1H 0 1H0 1H


Pyrophosphorolysis of T-analogue terminated primers

A.

B.

100

80

60

40

20

0

[% o

f py

roph

osph

orol

ysis

]

40003000200010000time (sec)

AZT d4T dT ddT



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

B.

60

50

40

30

20

10

% o

f com

ple

x

0.01 0.1 1 10 100 1000

conc. incoming nucleotide (µM)

AZT

d4T ddT

ddA

ddC

A.

1000

500

250

100

50100.01

1000

500

250

100

50100.01

1000

500

250

100

50100.01


P/T

P/T/RT0 0 0

µM dGTP



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

MarquetCatherine Isel, Chantal Ehresmann, Philippe Walter, Bernard Ehresmann and Roland

explained by the properties of the wild type HIV-1 reverse transcriptaseThe emergence of different resistance mechanisms towards nucleoside inhibitors is

published online October 19, 2001J. Biol. Chem.

10.1074/jbc.M108352200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M108352200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M108352200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2001/10/19/jbc.M108352200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2001/10/19/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2001/10/19/jbc.M108352200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

Documents

1 The emergence of different resistance mechanisms towards