In vitro Suppression of K65R Reverse Transcriptase-mediated Tenofovirand. Adefovir-5' Diphosphate Resistance Conferred by the Borano- ACCEPTED

Transcription

1 AAC Accepts, published online ahead of print on July 00 Antimicrob. Agents Chemother. doi:./aac Copyright 00, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved In vitro Suppression of KR Reverse Transcriptase-mediated Tenofovirand Adefovir-' Diphosphate Resistance Conferred by the Borano- phosphonate Derivatives Running Title : BH -PMEApp and BH -PMPApp suppress KR RT resistance. Antoine Frangeul*, Karine Barral*, Karine Alvarez and Bruno Canard From the Centre National de la Recherche Scientifique and Universités d'aix-marseille I et II, UMR 0, Architecture et Fonction des Macromolécules Biologiques, Ecole Supérieure d'ingénieurs de Luminy-Case, 1 avenue de Luminy, 1 Marseille cedex, France. * These authors contributed equally Address correspondence to: Bruno Canard, AFMB-CNRS-ESIL, Case, 1 avenue de Luminy, 1 Marseille Cedex, France, Tel:+-1--, Fax:+-1--, Bruno.Canard@afmb.univ-mrs.fr 1

2 ABSTRACT The -[-(boranophosphonomethoxy)ethyl]adenine diphosphate (BH -PMEApp) and (R)--[-(boranophosphonomethoxy)propyl]adenine diphosphate (BH -PMPApp) described here represent the first nucleoside phosphonates modified on their -phosphate, that act as efficient substrates for the Human Immunodeficiency Virus Reverse Transcriptase (HIV-1 RT). These analogues were synthesized and evaluated for their in vitro activity against wild- type, KR, and RA RT. BH -PMEApp and BH -PMPApp exhibit the same inhibition properties as their non-borano analogues on wild-type (WT) RT. However, KR RT was found hypersensitive to BH -PMEApp, and as sensitive as WT RT to BH -PMPApp. Moreover, the presence of the borano group restores incorporation of the analogue by RA HIV-RT, the latter being nearly inactive with regular nucleotides. The BH -mediated suppression of HIV-1 RT resistance, formerly described with nucleoside -( -P-borano)- triphosphate analogues is thus also conserved at the phosphonate level. The present results show that an -phosphate modification is also possible and interesting for phosphonate analogues, a result that might find application in the search of control of HIV-RT-mediated 1 drug resistance.

3 INTRODUCTION The reverse transcriptase (RT) 1 of Human Immunodeficiency Virus type 1 (HIV-1), plays a key role in the viral life cycle and is therefore an important target for antiretroviral drugs such as nucleoside and nucleotide RT inhibitors (NRTIs). After intracellular phosphorylation, these analogues are selectively incorporated into the viral DNA, leading to the observed antiviral effect through premature termination of viral DNA synthesis. Although effective at reducing viral loads in HIV-1 infected patients, NRTIs use has resulted in the development of mutations in RT that contribute to HIV-1 drug resistance. Tenofovir disoproxil fumarate, the oral prodrug of tenofovir [(R)--(- phosphonomethoxypropyl)adenine, (R)-PMPA] () has been approved for the treatment of HIV-1 infection in 001. Tenofovir is an acyclic nucleoside phosphonate (ANP) analogue of damp that requires two phosphorylation steps by cellular kinases to become the active metabolite tenofovir diphosphate (PMPApp), and act as a DNA chain terminator. There is considerable interest in phosphonate mimics because of their long intracellular half-life () and their capacity to circumvent the rate-limiting first phosphorylation step (1). The lysine to arginine substitution at residue (KR) in HIV-1 RT is amongst the most relevant resistance mutation () selected by tenofovir in vitro and in vivo. This mutation is also selected by other nucleosides (, 1,, ) and adefovir (), conferring reduced susceptibilities to these drugs. Although tenofovir does not select for thymidine associated mutations (TAMs), the presence of specific combinations of TAMs resulted in reduced susceptibility to tenofovir in vivo (1). Several mechanisms are known to contribute to decreased drug susceptibility. One mechanism explaining the decreased susceptibility of KR RT towards tenofovir is the decreased ability of the KR RT to bind to and incorporate tenofovir diphosphate into proviral DNA, as measured by a.-fold increase in Ki (). Indeed, the KR substitution affects the interaction between the enzyme and the

4 triphosphate moieties of NRTIs and might therefore alter the analogue binding specificity or the phosphodiester bond formation kinetics (). Based on the knowledge of the mechanisms of resistance of tenofovir (, 1,, ) and adefovir diphosphate (), we wanted to explore the drug susceptibility in vitro of (R)-- [-(boranophosphonomethoxy)propyl]adenine diphosphate (BH -PMPApp) and PMEApp analogues as the borano modification has been described to overcome resistance. Indeed, mutant RT resistant to the cognate non-borano inhibitor recover sensitivity upon the presence of an -boranophosphate group into the triphosphate form of clinically relevant compounds, such as AZT, dt or dideoxyadenosine (1,,, 1). The presence of the borano group does not influence the binding affinity of the analogue for the RT active site, but specifically either provides or restores a high incorporation rate of the analogue by wild-type (WT) or mutant HIV-1 RTs, respectively (1, ). The focus of the present study is thus to investigate in vitro the effect on susceptibilities of HIV-1 WT and mutant RTs to the borano modification introduced at the position of acyclic phosphonate diphosphates. We have explored the in vitro activity of -boranophosphonate diphosphates BH -PMEApp and BH -PMPApp (Fig.1) against WT RT, but also against both KR and RA mutants. The arginine side chain contacts the phosphate of the incoming nucleotide. Consequently, RA RT is severely impaired to perform processive DNA polymerization (1, 1, 0, ). We have investigated on the ability of the -boranophosphonate diphosphates BH -PMEApp and BH -PMPApp to restore processive DNA polymerization capability of RA RT, such as in the case of boranophosphate analogues (). Pre-steady state kinetic is the method of choice to measure the substrate efficiency of a nucleotide analogue for RT. The efficiency of incorporation of a nucleotide into DNA is given by the ratio k pol /K d, k pol being the rate constant for the creation of the phosphodiester bond and K d the equilibrium binding (or affinity) constant of the nucleotide for RT (1).

5 Discrimination of an ANP relative to its natural counterpart dntp is reflected by the efficiency of incorporation of the ANP into DNA relative to that of the natural substrate. Although this has been done for various NRTIs, such constants have not been determined for BH -ANPs. In this paper, we have made use of pre-steady state kinetics to determine the affinity constant K d as well as the catalytic constant k pol of formation of the phosphodiester bond for datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp substrates by either wild-type or KR RTs. Lastly, and as in the case of boranophosphate analogues, we observe improvement of catalysis by the presence of the borano group in phosphonate analogues using RA RT.

6 MATERIALS AND METHODS Chemistry All air-sensitive reactions were performed in oven-dried glassware under argon using extra-dry solvents from Aldrich. The 1 H, 1 C and 1 P NMR spectra were recorded with a BRUKER AMX 0 MHz and the B NMR spectrum was recorded with a BRUKER AMX 00 MHz. Chemical shifts are expressed in ppm and coupling constants (J) are in hertz (s = singlet, bs = broad singulet, d = doublet, dd = double doublet, t = triplet, dt = double triplet, td = triple doublet, qd = quadruple doublet, m = multiplet, dm = double multiplet). FAB Mass Spectra (MS) and High Resolution Mass Spectra (HRMS) were obtained on a JEOL SX mass spectrometer using a cesium ion source and a glycerol/thioglycerol matrix (GT). Analytical high-performance liquid chromatography (HPLC) analyses were carried out on a Waters Associates unit equipped with a model 00E multisolvent delivery system, a model 00E controller system, a model Rheodyne sample injector, a 1 photodiode array detector and an in-line degazer AF. Samples were eluted using a linear gradient of 0.0M triethylammonium bicarbonate buffer in 0% water (ph.) (buffer A) to a 0.0M triethylammonium bicarbonate buffer (TEAB) in 0% acetonitrile (buffer B), programmed over a 0 min period with a flow rate of 1 ml/min and detection at 0 nm. All solvents were of HPLC grade and filtered prior to use. A 1M solution of triethylammonium bicarbonate buffer was prepared by adding dry ice to a 1M triethylamine solution until the ph reached.. Triethylammonium bicarbonate solutions were made fresh by dissolving reagent-grade triethylammonium bicarbonate in HPLC-grade water prior to filtration. Analytical reversephase chromatography was carried out on a. 0 mm Source 1RPC column. Preparative purifications of a-boranophosphonate derivatives were achieved on an ÄKTAprime FPLC (Amersham) using a reverse-phase Source 0RPC column (1 0 mm) and a linear gradient of 0.0M triethylammonium bicarbonate buffer in 0% water (ph

7 ) (buffer A) to a 0.0M triethylammonium bicarbonate buffer in 0% acetonitrile (buffer B), programmed over a h period with a flow rate of ml/min and detection at nm. Boranophosphonate nucleosides BH -PMEA and BH -PMPA were synthesized following the procedure of Barral et al. (). To prepare '-diphosphate analogues, the Hoard and Ott s procedure (1) was adapted as follows. The boranophosphonates BH -PMEA and BH -PMPA were treated with an excess of 1,1 -carbonyldiimidazole (CDI) in DMF and stirred for h. The activated intermediates were not isolated. Unreacted CDI was decomposed with anhydrous methanol, before bis(tri-n-butylammonium) pyrophosphate was added. The phosphorylation was essentially complete within h and the products were purified by reversed-phase chromatography, followed by conversion to the sodium salt to conduce to BH -PMEApp and BH -PMPApp with a and % yield, respectively. Non-borano phosphonate diphosphates PMEApp and PMPApp, which serve as reference compounds, were synthesized as described in the literature (1). -[-(Diphosphorylboranophosphonomethoxy)ethyl]adenine (BH -PMEApp). The - [-(Boranophosphonomethoxy)ethyl]adenine (1 mg, 0.0 mmol) was dissolved in DMF ( ml) and treated with 1,1 -carbonyldiimidazole ( mg, 0. mmol). The resulting mixture was stirred at room temperature for h. Excess CDI was decomposed by addition of anhydrous methanol ( µl) and stirring was continued for 0 min. Anhydrous tri-nbutylamine ( µl) and tributylammonium pyrophosphate (0 µl of a 0.M solution in DMF) were added and the mixture was stirred at room temperature for days. The reaction was stopped by the addition of ml of cold water. The solvent was removed under vacuum and the residue was purified by reversed-phase column chromatography (linear gradient 0-0% B). The appropriate fractions were collected, evaporated to dryness and lyophilized. The residue was dissolved in water and passed through a Dowex 0WX (Na + form) column to give BH -PMEApp as trisodium salt (. mg, %) ; HPLC purity (> %) ; 1 H NMR

8 (D O) :. (s, H, H-, H-),.1 (t, J =. Hz, H, CH N),. (t, J =. Hz, H, CH O),. (m, H, CH P), 0.- (-) 0. (q, J = 0 Hz, H, BH ). 1 P NMR (D O) :.0, -.1, HRMS (TOF, ES - ) calcd for C H 1 N O P B (M) - 0.0, found 0.0. (R)--[-(Pyrophosphoroxyboranophosphonomethoxy)propyl]adenine (BH - PMPApp). The (R)--[-(Boranophosphonomethoxy)propyl]adenine ( mg, 0.0 mmol) was dissolved in DMF ( ml) and treated with 1,1 -carbonyldiimidazole (0 mg, 0. mmol). The resulting mixture was stirred at room temperature for h. Excess CDI was decomposed by addition of anhydrous methanol ( µl) and stirring was continued for 0 min. Anhydrous tri-n-butylamine ( µl) and tributylammonium pyrophosphate ( µl of a 0.M solution in DMF) were added and the mixture was stirred at room temperature for days. The reaction was stopped by the addition of ml of cold water. The solvent was removed under vacuum and the residue was purified by reversed-phase column chromatography (linear gradient 0-0% B). The appropriate fractions were collected, evaporated to dryness and lyophilized. The residue was dissolved in water and passed through a Dowex 0WX (Na + form) column to give BH -PMPApp as trisodium salt (1. mg, %) ; HPLC purity (> %) ; 1 H NMR (D O) :. (d, J =. Hz, 1H, H-),. (d, J =. Hz, 1H, H-),.0 (dd, J = 1.0 Hz and J =. Hz, 1H, CH a N),. (m, 1H, CH b N),. (m, 1H, CHO),.1 (m, 1H, CH a P),.0 (dd, J = 1.0 Hz and J =.0 Hz, 1H, CH b P), 0. (d, J =. Hz, CH ), 0.-(-) 0.0 (q, J = Hz, H, BH ). 1 P NMR (D O) :., -., HRMS (TOF, ES - ) calcd for C H 1 N O P B (M) -.0, found.0. Reverse transcriptase assays HIV-RT Plasmid Constructions, Enzyme Preparations and Reagents. The WT RT bacterial expression gene construct prtb was used to obtain KR and RA RT as described previously (). All constructs were verified by DNA sequencing. The recombinant RTs were co-expressed with HIV-1 protease in Escherichia coli in order to obtain p/p1

9 heterodimers, which were later purified using affinity chromatography. All enzymes were quantified by active-site titration before biochemical studies. DNA oligonucleotides were obtained from Life Technologies, UK. Oligonucleotides were - P-labeled using T polynucleotide kinase (New England Biolabs, MA). [ - P]ATP was purchased from Amersham Biosciences. Drug Susceptibility Assays Using Recombinant HIV-1 RTs. Standard RT activity was assayed using 0 µg activated calf thymus DNA per ml. To determine IC 0 values for PMEApp, PMPApp, BH -PMEApp and BH -PMPApp reactions were performed with nm HIV-1 RT and µm of each dntp containing 0 µci per mmol of [ H]dATP, for 1 min with increasing amounts of inhibitor. Each aliquot was spotted in duplicate on DE1 ion- exchange paper discs, and the discs were washed three times for min in 0. M ammonium formate, ph.0, two times in ethanol, and dried. The radioactivity bound to the filters was determined by liquid scintillation counting. Values of IC 0 are the average from at least three independent experiments. Pre-Steady State Kinetics of Single Nucleotide Incorporation Into DNA by RT. Presteady state kinetics were performed using datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp, in conjunction with WT and KR RTs. Rapid quench experiments were performed with a Kintek instrument Model RQF- using reaction times ranging from ms to 0 s. All indicated concentrations are final. The primer DNA/DNA oligonucleotides used for the rapid reaction were a -labeled 1-mer primer ( -ATA CTT TAA CCA TAT GTA TCC- ) annealed to a 1-mer template 1A-RT ( -AAA AAA AAA TGG ATA CAT ATG GTT AAA GTA T- ). For natural nucleotides, the reaction was performed by mixing a solution containing 0 nm (active sites) of HIV-1 RT bound to 0 nm of primer/template in RT buffer (0 mm Tris-HCl, ph.0, 0 mm KCl, 0.0 % Triton X-0), and a variable concentration of dntp in mm MgCl. Reactions involving ANPs were conducted with

10 excess concentrations of enzyme (00 nm) over primer/template duplex (0 nm). These conditions were chosen to eliminate the influence of the enzyme turnover rate (k ss ) that interferes in the measurements of low incorporation rates. Products of reactions were analyzed using sequencing gel electrophoresis (1 % acrylamide, M urea in TBE buffer), and quantified using photo-stimulated plates and FujiImager. The formation of product (P) over time was fitted with a burst equation: (P) = A.(1-exp(-(k app.t)) + k ss.t) (Eq. 1) where A is the amplitude of the burst, k app is the apparent kinetic constant of formation of the phosphodiester bond, and k ss is the enzyme turnover rate, i.e., the kinetic constant of the steady-state linear phase. The dependence of k app on dntp concentration is described by the hyperbolic equation: k app = k pol.[dntp]/( K d + [dntp]) (Eq. ) where K d and k pol are the equilibrium constant and the catalytic rate constant of the dntp for RT, respectively. K d and k pol were determined from curve-fitting using Kaleidagraph (Synergy Software, PA). DNA Polymerization rate assays by RTs. The rate of polymerization was measured using a -labeled oligo(da) 1 annealed to a poly(ru) 1, in conjunction with WT and RA RTs. Extension products were analyzed in a gel assay. The primer-template ( nm) was incubated with 0 nm RT in RT buffer at C. The reaction was initiated by the addition of 00 µm of nucleotide (datp) or analogues (PMEApp, PMPApp, BH -PMEApp and BH - PMPApp) in mm MgCl and quenched at various time by 0. M EDTA. % of elongated primer were determined from the product profile analysis using a FujiImager. Average polymerization rates were calculated for datp by measuring the amount of the most abundant band product divided by time, to yield an apparent polymerization rate in nt.s -1.

11 RESULTS Enzyme susceptibility assays using steady-state kinetics. In order to examine the relative inhibitory capacity of ANPs, we tested the susceptibility of WT and KR RTs to PMEApp, PMPApp, BH -PMEApp and BH -PMPApp. Results are summarized in Table 1. We observe that BH -PMEApp (IC 0 = 1 µm) is a slightly less efficient inhibitor than PMEApp (IC 0 =. µm), whereas BH -PMPApp (IC 0 = 1. µm) is slightly better than PMPApp (IC 0 = µm). The BH -PMEApp and BH -PMPApp are thus active in vitro against purified WT RT. As described before (, ), we observe that the KR mutation induces a to -fold resistance in vitro to both PMEApp (IC 0 = µm) and PMPApp (IC 0 = µm). Remarkably, however, susceptibility assays performed on KR RT show that the potency of both BH -PMEApp (IC 0 = 1 µm) and BH -PMPApp (IC 0 = 1 µm) is similar to that of WT RT, reversing apparent resistance. In the case of BH -PMEApp, KR is 0.-fold resistant (i.e., more sensitive) compared to WT RT. In the case of BH -PMPApp, KR is not significantly resistant (1.-fold) compared to WT. We conclude that boranophosphonates analogues are better inhibitor than their non-borano counterparts on KR RT. Single incorporation of PMEApp, PMPApp, BH -PMEApp and BH -PMPApp into DNA by WT RT. Using pre-steady state kinetics for a single nucleotide incorporation into DNA, we measured the nucleotide initial binding affinity (K d ) and the subsequent burst-rate of catalysis (k pol ). The ratio (k pol / K d ) defines the nucleotide efficiency. For a given RT, comparing k pol / K d values between a natural dntp and its corresponding analogue is a convenient manner to assess selectivity for the natural nucleotide, or discrimination of the analogue. For the sake of clarity, we will refer only to discrimination -not selectivity- of the analogue relative to its natural counterpart. Comparing discrimination between RTs defines in vitro resistance at the enzymatic level. The kinetic constants of datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp are reported in Table.

12 Firstly, it is striking that BH -PMPApp is a better stubstrate that PMPApp. Indeed, discrimination relative to the natural datp substrates decreases from 1.- to.-fold for PMPApp and BH -PMPApp, respectively. Perhaps the most apparent basis for this improvement of BH -PMPA is that it originates mainly from a >-fold k pol effect (from to s -1 ). This not the case for PMEApp and BH -PMEApp, and this parallels steady-state data satisfactorily (see above). Single incorporation of PMEApp, PMPApp, BH -PMEApp and BH -PMPApp into DNA by KR RT. The same type of assays were performed in the case of the resistant KR variant of RT. Strikingly, in this case, both PMEApp and PMPApp discriminations are decreased upon the presence of the BH - group (1.- to.-fold, and. to -fold, respectively). As a consequence, resistance is also suppressed in both cases, and a.-fold hypersensitivity (or 0.-fold resistance) is even observed in the case of BH -PMEApp. In the same vein as for WT RT, the presence of the -borano group has a larger effect on the burst rate of catalysis (k pol ) than the initial binding affinity (K d ). In the case of PMPApp, the presence of the BH - group results in a./0. = 1-fold increase in k pol. In conclusion, both compounds BH -PMEApp and BH -PMPApp inhibit KR RT efficiently, and are better inhibitor than their respective PMEApp and PMPApp counterparts. Comparison between WT and KR RTs for PMEApp, PMPApp, BH -PMEApp and BH -PMPApp. Comparing discrimination between RTs defines in vitro resistance at the enzymatic level. Expectedly, as shown in Table, the KR RT variant is resistant to PMEApp and PMPApp (. and.-fold, respectively). The resistance is due to a decrease of the catalytic rate constant, as described before (1). The increased discrimination of PMPApp compared to datp brought by the KR mutation accounts for the phenotypic loss of susceptibility of the mutated virus to the drug in infected cell cultures (,, ). But in the presence of the -borano analogue, the resistance factor of KR is decreased down to 1-fold 1

13 (i.e., inexistent) for BH -PMPApp or even hypersensitive, i.e. 0.-fold, for BH -PMEApp. This decrease of resistance results from a 1- and 1.-fold increase in k pol, respectively. We conclude that, remarkably, the presence of the -borano group renders KR RT hypersensitive to the BH -PMEApp analogue. In the case of BH -PMPApp, the presence of the -borano suppresses the KR RT-mediated resistance observed with PMPApp. Polymerase activity of the active-site mutant RA RT with ANPs and BH -ANPs. A polymerization rate assay was used in conjunction with WT or RA RT to measure the incorporation rate of datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp along a poly(ru) 1 RNA template during the early polymerization events (Fig. ). In average, we observe that WT RT incorporates datp at a.-fold higher rate than RA RT, i.e., 0. nt.s - 1 and 0.1 nt.s -1 for WT and RA, respectively, as described before (0). Product profile analysis show that these values represent and % respectively of the elongated primer for a reaction time of seconds (lanes and 1, respectively). Under these same experimental conditions, PMEApp and PMPApp chain-terminated DNA products by WT RT represent % and % (lanes 1 and, respectively). These product yields are similar to those observed when the -borano group is present on their -phosphate (BH -PMEApp and BH -PMPApp, and % (lanes and, respectively). Comparatively, RA RT incorporates very poorly PMEApp and PMPApp (1. and 0.%, lanes and 1, respectively) but these low polymerization rates are improved dramatically using their -borano analogues BH - PMEApp and BH -PMPApp ( and %, lanes 1 and 1, respectively). We conclude that the -borano group stimulates catalysis, indeed. It provides the nucleotide with intrinsic chemical properties able to circumvent the absence of the key catalytic amino acid R, as described before for -boranophosphate nucleotides (). Single incorporation of PMPApp and BH -PMPApp into DNA by WT and RA RTs using a single 00 µm concentration of analogue. The study of the restoration of the catalytic 1

14 rate constant upon the presence of a borano group was refined in the case of compounds PMPApp and BH -PMPApp using pre-steady state kinetics. As shown in Fig., the apparent kinetic constant of formation of the phosphodiester bond (k app ) increases.-fold for the BH - PMPApp borano analogue (1 s -1 ) compared to PMPApp using WT RT (. s -1 ). The effect is even greater in the case of RA RT where k app for BH -PMPApp (0. s -1 ) increases more than -fold compared to PMPApp (0.0 s -1 ). We conclude that, in a seemingly similar fashion as in the case of -boranophosphate nucleosides, the presence of the -borano group increases the incorporation efficiency in WT RT, and restores the catalytic properties of RA RT. 1

15 DISCUSSION We describe here the synthesis of the first -boranophosphonate diphosphate nucleosides in which a borane (BH - ) group substitutes one nonbridging oxygen atom of the -phosphonate moiety. Compounds BH -PMEA and BH -PMPA previously synthesized are inactive against HIV-1 in infected cell cultures, however, due to their rapid degradation in an extracellular medium mimic (). This suggests that chemical modification ("vectorisation") should be considered to prevent compound catabolisation and promote cell penetration. These BH -PMEApp and BH -PMPApp analogues could then potentially become potent resistance- suppressors useful in the clinic. The compounds described here present interesting properties, however. Enzyme susceptibility assays of BH -PMEApp and BH -PMPApp on HIV-1 RTs show that these compounds inhibit WT RT in the same range as that of PMEApp and PMPApp. Furthermore, boranophosphonate analogues are better inhibitors than their unmodified PMEApp and PMPApp counterparts using KR RT. Pre-steady state kinetics of single nucleotide incorporation of BH -PMEApp and BH -PMPApp on HIV-1 RT allow a closer look on the reaction. The presence of the -borano group has a beneficial effect on the burst rate of catalysis (k pol ), albeit in the case of BH -PMPApp only. HIV-1 viruses encoding a KR mutant RT are selected in vivo by tenofovir, and KR seems to reduce the incorporation of tenofovir into DNA (). KR HIV-1 RT has a reduced pre-steady-state rate of dntp incorporation (1), suggestion that the reduced incorporation of tenofovir comes at a cost to incorporation of natural dntp. Studies on KR RT reveal that both compounds BH -PMEApp and BH -PMPApp inhibit KR RT efficiently, and are better inhibitors than PMEApp and PMPApp, respectively. Therefore, KR RT is not resistant anymore to BH -PMPApp compared to PMPApp and is hypersensitive to BH -PMEApp compared to PMEApp. The presence of the borano group 1

16 does not influence the binding of the analogue to the RT active site, but provides (or restores) a high catalytic rate constant (k pol ) of incorporation of the analogue specifically, as described previously on boranophosphates (1). The active-site mutant RA RT is severely impaired in its ability to perform processive DNA polymerization. The RA RT shows a defect in eliminating the pyrophosphate produced at the active-site upon incorporation of the nucleotide into DNA (1, 1, 0, ). However, the presence of borano group on the -phosphate of '- deoxynucleotides restore the processive DNA polymerization capability of RA RT (). It was of interest to verify if this property would also be observed using -boranophosphonate diphosphate BH -PMEApp and BH -PMPApp. WT RT incorporates BH -PMEApp and BH - PMPApp in the same manner as non-boranophosphonate diphosphates. Comparatively, RA RT incorporates very poorly PMEApp and PMPApp but these low polymerization rates are improved dramatically using their -borano analogues BH -PMEApp and BH -PMPApp. More precisely, the apparent kinetic constant of formation of the phosphodiester bond (k app ) with RT is improved for the BH -PMPApp borano analogue compared to PMPApp. This improvement is even greater in the case of RA RT where the difference of k app for BH - PMPApp and PMPApp is increased. The presence of the -borano group increases the catalytic rate of WT RT, and restores the catalytic properties of RA RT. The borano modification provides the nucleotide with intrinsic chemical properties able to circumvent the presence of the key catalytic amino acid R, as described before for -boranophosphate nucleotides () and shows a capability to compensate a mutation inducing a defect in the nucleotide incorporation rate. In conclusion, these novel nucleotide analogues demonstrate that not only the phosphate of phosphonate analogues can be modified without jeopardizing incorporation properties, but that this modification could be useful to circumvent HIV RT-mediated drug 1

17 resistance. It would be interesting to assess the use of other -phosphonate modifications, such as thio- or seleno-phosphonates, not only for their incorporation efficiency but also to improve the stability of analogue-terminated DNA chains. 1

18 REFERENCES Alvarez, K., J. Deval, B. Selmi, K. Barral, J. Boretto, C. Guerreiro, L. Mulard, S. Sarfati, and B. Canard. 00. Borano-nucleotides : new analogues to circumvent HIV-1RT-mediated nucleoside drugs resistance. Nucleosides, Nucleotides & Nucleic Acids :1-.. Arion, D., G. Borkow, Z. Gu, M. A. Wainberg, and M. A. Parniak. 1. The KR mutation confers increased DNA polymerase processivity to HIV-1 reverse transcriptase. J Biol Chem 1:10-.. Barral, K., S. Priet, J. Sire, J. Neyts, J. Balzarini, B. Canard, and K. Alvarez. 00. Synthesis, in vitro antiviral evaluation, and stability studies of novel alpha- borano-nucleotide analogues of -[-(phosphonomethoxy)ethyl]adenine and (R)--[- (phosphonomethoxy)propyl]adenine. J Med Chem :-0.. Bazmi, H. Z., J. L. Hammond, S. C. Cavalcanti, C. K. Chu, R. F. Schinazi, and J. W. Mellors In vitro selection of mutations in the human immunodeficiency virus type 1 reverse transcriptase that decrease susceptibility to (-)-beta-d-dioxolane- guanosine and suppress resistance to '-azido-'-deoxythymidine. Antimicrob Agents Chemother :1-.. Boretto, J., S. Longhi, J. M. Navarro, B. Selmi, J. Sire, and B. Canard An integrated system to study multiply substituted human immunodeficiency virus type 1 reverse transcriptase. Anal Biochem :1-.. De Clercq, E Antiviral drugs : current state of the art. Journal of Clinical Virology :-.. Deval, J., K. Alvarez, B. Selmi, M. Bermond, J. Boretto, C. Guerreiro, L. Mulard, and B. Canard. 00. Mechanistic insights into the suppression of drug 1

19 resistance by human immunodeficiency virus type 1 reverse transcriptase using alphaboranophosphate nucleoside analogs. Journal of Biological Chemistry 0:-.. Deval, J., B. Selmi, J. Boretto, M. P. Egloff, C. Guerreiro, S. Sarfati, and B. Canard. 00. The molecular mechanism of multidrug resistance by the QM human immunodeficiency virus type 1 reverse transcriptase and its suppression using alpha-boranophosphate nucleotide analogues. Journal of Biological Chemistry :0-.. Deval, J., K. L. White, M. D. Miller, N. T. Parkin, J. Courcambeck, P. Halfon, B. Selmi, J. Boretto, and B. Canard. 00. Mechanistic basis for reduced viral and enzymatic fitness of HIV-1 reverse transcriptase containing both KR and M1V mutations. J Biol Chem :0-1.. Engel, R. 1. Phosphonates as analogues of natural phosphates. Chemical Reviews :-.. Foli, A., K. M. Sogocio, B. Anderson, M. Kavlick, M. W. Saville, M. A. Wainberg, Z. Gu, J. M. Cherrington, H. Mitsuya, and R. Yarchoan. 1. In vitro selection and molecular characterization of human immunodeficiency virus type 1 with reduced sensitivity to -[-(phosphonomethoxy)ethyl]adenine (PMEA). Antiviral Res : Gu, Z., Q. Gao, H. Fang, H. Salomon, M. A. Parniak, E. Goldberg, J. Cameron, and M. A. Wainberg. 1. Identification of a mutation at codon in the IKKK motif of reverse transcriptase that encodes human immunodeficiency virus resistance to ','-dideoxycytidine and ','-dideoxy-'-thiacytidine. Antimicrob Agents Chemother : Hoard, D. E., and D. G. Ott. 1. Conversion of mono- and oligodeoxyribonucleotides to '-triphosphates. Journal of the American Chemical Society :1-1. 1

20 Holy, A., and I. Rosenberg. 1. Synthesis of -(- phosphonylmethoxyethyl)adenine and related compounds. Collect. Czech. Chem. Commun. : Kati, W. M., K. A. Johnson, L. F. Jerva, and K. S. Anderson. 1. Mechanism and fidelity of HIV reverse transcriptase. J Biol Chem :-. 1. Lewis, D. A., K. Bebenek, W. A. Beard, S. H. Wilson, and T. A. Kunkel. 1. Uniquely altered DNA replication fidelity conferred by an amino acid change in the nucleotide binding pocket of human immunodeficiency virus type 1 reverse transcriptase. J Biol Chem : Margot, N. A., E. Isaacson, I. McGowan, A. K. Cheng, R. T. Schooley, and M. D. Miller. 00. Genotypic and phenotypic analyses of HIV-1 in antiretroviral- experienced patients treated with tenofovir DF. Aids 1: Meyer, P., B. Schneider, S. Sarfati, D. Deville-Bonne, C. Guerreiro, J. Boretto, J. Janin, M. Veron, and B. Canard Structural basis for activation of alphaboranophosphate nucleotide analogues targeting drug-resistant reverse transcriptase. EMBO Journal 1: Robbins, B. L., R. V. Srinivas, C. Kim, N. Bischofberger, and A. Fridland. 1. Anti-human immunodeficiency virus activity and cellular metabolism of a potential prodrug of the acyclic nucleoside phosphonate -R-(- phosphonomethoxypropyl)adenine (PMPA), Bis(isopropyloxymethylcarbonyl)PMPA. Antimicrob Agents Chemother : Sarafianos, S. G., V. N. Pandey, N. Kaushik, and M. J. Modak. 1. Site-directed mutagenesis of arginine of HIV-1 reverse transcriptase. Catalytic role and inhibitor sensitivity. J Biol Chem 0:1-. 0

21 Selmi, B., J. Boretto, S. R. Sarfati, C. Guerreiro, and B. Canard Mechanism-based suppression of dideoxynucleotide resistance by KR human immunodeficiency virus reverse transcriptase using an alpha-boranophosphate nucleoside analogue. Journal of Biological Chemistry :-.. Sluis-Cremer, N., D. Arion, N. Kaushik, H. Lim, and M. A. Parniak Mutational analysis of Lys of HIV-1 reverse transcriptase. Biochem J Pt 1:-.. Tisdale, M., T. Alnadaf, and D. Cousens. 1. Combination of mutations in human immunodeficiency virus type 1 reverse transcriptase required for resistance to the carbocyclic nucleoside 1U. Antimicrob Agents Chemother 1:-.. Tuske, S., S. G. Sarafianos, A. D. Clark, Jr., J. Ding, L. K. Naeger, K. L. White, M. D. Miller, C. S. Gibbs, P. L. Boyer, P. Clark, G. Wang, B. L. Gaffney, R. A. Jones, D. M. Jerina, S. H. Hughes, and E. Arnold. 00. Structures of HIV-1 RT- DNA complexes before and after incorporation of the anti-aids drug tenofovir. Nat Struct Mol Biol :-.. Wainberg, M. A., M. D. Miller, Y. Quan, H. Salomon, A. S. Mulato, P. D. Lamy, N. A. Margot, K. E. Anton, and J. M. Cherrington. 1. In vitro selection and characterization of HIV-1 with reduced susceptibility to PMPA. Antivir Ther :-.. White, K. L., N. A. Margot, T. Wrin, C. J. Petropoulos, M. D. Miller, and L. K. Naeger. 00. Molecular mechanisms of resistance to human immunodeficiency virus type 1 with reverse transcriptase mutations KR and KR+M1V and their effects on enzyme function and viral replication capacity. Antimicrob Agents Chemother :-. 1

22 ACKNOWLEDGMENTS The authors wish to thank Dr A. Holy for kindly providing PMEA and PMPA derivatives, and Joëlle Boretto for excellent technical assistance. This work was supported by the Agence Nationale de la Recherche contre le SIDA (ANRS) and by Ensemble Contre le SIDA (ESC-SIDACTION). Dr. K. Barral is supported by a post-doctoral fellowship from the Agence Nationale de Recherche sur le SIDA (ANRS). 1 The abbreviations used are: ANP, acyclic nucleoside phosphonate; BH, borano group; BH -PMEApp, -[-(boranophosphonomethoxy)ethyl]adenine diphosphate; BH - PMPApp, (R)--[-(boranophosphonomethoxy)propyl]adenine diphosphate; damp, - deoxyadenosine -monophosphate; datp, -deoxyadenosine -triphosphate; HIV-1, Human Immunodeficiency Virus type 1; NRTI, nucleoside reverse transcriptase inhibitor; PMEApp, adefovir diphosphate; PMPApp, tenofovir diphosphate; RT, reverse transcriptase; WT, wild-type. 1

23 FIGURE LEGENDS Fig. 1: Chemical structures of -boranophosphonate diphosphates BH -PMEApp, BH -PMPApp, and their non-borano phosphonate diphosphates PMEApp and PMPApp. Fig. : Polymerization rate assay using datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp. WT and RA RTs were assayed in the first seconds of the primer extension reaction on a primer oligo(da) 1 annealed to a template poly(ru) 1 with datp or analogues. The figure shows an autoradiograph of the reaction products separated using denaturing del electrophoresis (see "Materials and Methods"). Fig. : Kinetics of a single nucleotide incorporation into DNA using PMPApp and BH -PMPApp in conjunction with WT and RA RTs. The latter were assayed in the first seconds of the primer extension reaction on a 1-mer primer annealed to a 1-mer template allowing incorporation of adenine nucleotide analogue into a -mer product (see "Materials and Methods"). 0 1

24 Table 1: Drug susceptibilities of WT RT for PMEApp, PMPApp, BH -PMEApp and BH - PMPApp. IC 0 (µm) a RT PMEApp BH -PMEApp PMPApp BH -PMPApp WT. ±1. 1 ±1. ±1 1. ±1. KR ± (x) 1 ±1. (x0.) ± (x) 1 ±1. (x1.) a IC 0 were conducted with recombinant RTs assayed on activated calf thymus DNA (see "Materials and Methods"), and averaged (± S.D.) for at least three independents experiments. Table : Pre-steady state kinetic constants of datp, PMEApp, PMPApp, BH -PMEApp and BH -PMPApp and incorporation by HIV-1 RTs. datp PMEApp WT RT BH - PMEApp PMPApp BH - PMPApp datp PMEApp KR RT BH - PMEApp PMPApp k pol (s -1 ) a 0. b b. K d (µm) a... b b BH - PMPApp k pol / K d (s -1.µM -1 ) Discrimination c Resistance d a K d and k pol were determined as described under Materials and methods. Standard deviations (SD) were < 0%, for three independent experiments. b Value from Deval et al. (). c Discrimination is the ratio of [k pol / K d (nucleotide)] / [k pol / K d (nucleotide analogue)]. d Resistance is the ratio of [Discrimination (mutant)] / [Discrimination (WT)].

25

26

27