Structural Basis for the Resilience of Efavirenz (DMP-266) to Drug Resistance Mutations in HIV-1 Reverse Transcriptase

Structure, Vol. 8, 1089–1094, October, 2000, 2000 Elsevier Science Ltd. All rights reserved.

PII S0969-2126(00)00513-X

Structural Basis for the Resilience of Efavirenz (DMP-266) to Drug Resistance Mutations in HIV-1 Reverse Transcriptase Jingshan Ren,* John Milton,† Kurt L. Weaver,‡ Steven A. Short,‡ David I. Stuart,*§k and David K. Stammers*§k * Structural Biology Division The Wellcome Trust Centre for Human Genetics University of Oxford Oxford OX3 7BN United Kingdom †Glaxo Wellcome R&D Stevenage SG1 2NY United Kingdom ‡Glaxo Wellcome North Carolina 27709 §Oxford Centre for Molecular Sciences Oxford OX1 3QT United Kingdom

Summary Background: Efavirenz is a second-generation non-nucleoside inhibitor of HIV-1 reverse transcriptase (RT) that has recently been approved for use against HIV-1 infection. Compared with first-generation drugs such as nevirapine, efavirenz shows greater resilience to drug resistance mutations within HIV-1 RT. In order to understand the basis for this resilience at the molecular level and to help the design of further-improved anti-AIDS drugs, we have determined crystal structures of efavirenz and nevirapine with wild-type RT and the clinically important K103N mutant. Results: The relatively compact efavirenz molecule binds, as expected, within the non-nucleoside inhibitor binding pocket of RT. There are significant rearrangements of the drug binding site within the mutant RT compared with the wild-type enzyme. These changes, which lead to the repositioning of the inhibitor, are not seen in the interaction with the first-generation drug nevirapine. Conclusions: The repositioning of efavirenz within the drug binding pocket of the mutant RT, together with conformational rearrangements in the protein, could represent a general mechanism whereby certain second-generation non-nucleoside inhibitors are able to reduce the effect of drug-resistance mutations on binding potency.

step in achieving this is an understanding of drug-resistance mechanisms, as such knowledge will contribute to the rational design of new anti-HIV therapies [5]. One class of compounds increasingly used in combination therapy together with earlier approved nucleoside analog reverse transcriptase (RT) inhibitors (NRTIs) and protease inhibitors (PIs) is the non-nucleoside reverse-transcriptase inhibitors (NNRTIs). These compounds are noncompetitive inhibitors of HIV-1 RT that bind at an allosteric site some 10 A˚ from the polymerase active site [6–8], causing a distortion of the catalytic aspartate triad [9]. It has been shown that NNRTIs can bind more tightly in the presence of substrates [10], and X-ray crystallographic studies of binary complexes of RT with NNRTIs fully explain data on the location and effects of resistance mutations within RT [11–13]. The first-generation NNRTIs, such as the currently marketed drugs nevirapine (Figure 1b) and delavirdine, can show orders of magnitude decrease in binding affinity as a result of many single point mutations in RT [14]. The so-called second-generation NNRTIs such as the benzoxazin-2-one, efavirenz (DMP266) [15] (Figure 1a), certain carboxanilides [16], and quinoxalines [17] have a more favorable resistance profile, showing smaller losses of activity against many common drug-resistance mutations. A wide range of mutations giving resistance to various NNRTIs have been reported from in vitro and in vivo studies [14]. In clinical trials and therapeutic use of NNRTIs, the most commonly observed mutation within RT is Lys103→Asn (K103N). The K103N mutant shows a 6-fold weaker binding of efavirenz when compared with wild-type RT [15]; in contrast, the loss of binding for nevirapine by this mutant RT is 40-fold. In tissue culture the loss of activity for virus containing the K103N mutation seems to be greater [15]. We have determined the structure of efavirenz in complexes with both wild-type and K103N mutant RTs, as well as the structure of nevirapine in complex with the K103N mutant. This has allowed a detailed comparison of these examples of firstand second-generation drugs, leading to an understanding of some structural factors that confer resilience to drug-resistance mutations. Results

The introduction of highly active antiretroviral therapy (HAART) involving the use of drug combinations to treat HIV-1 infection has resulted in a significant reduction in the death rate from AIDS for patients receiving such treatment [1,2]. However, there are continuing problems of drug resistance, including the emergence of multi-drug-resistant strains of HIV-1 [3,4]. Further novel drugs active against resistant HIV-1 are therefore required in the continuing effort to combat AIDS. An important

We have determined the structures of efavirenz in complex with wild-type and the K103N mutant RTs to 2.5 A˚ and 2.9 A˚ resolution, respectively, and the complex of nevirapine with K103N RT to 2.9 A˚. Details of the structure determinations are given in Table 1. Omit electron density maps clearly define the position of the inhibitors, the conformation of residue 103, and certain sidechain movements (Figure 2). The structure of the wild-type RT in complex with nevirapine at 2.2 A˚ resolution has been reported previously [7]. The position of efavirenz in the inhibitor binding pocket of wild-type HIV-1 RT is shown in Figure 3. As expected, many of the inhibitor’s interactions with RT involve a series of hydrophobic contacts. Thus, the cyclopropyl–propynyl group is positioned in the top sub-pocket surrounded by the aromatic sidechains of Tyr181, Tyr188, Trp229, and Phe227. The benzo-

k To whom correspondence should be addressed (e-mail: daves@strubi. ox.ac.uk; [email protected]).

Key words: drug resistance; efavirenz; HIV-1; nevirapine; reverse transcriptase

Introduction

Structure 1090

Figure 1. Structure of RT Inhibitors The structures of (a) efavirenz (DMP-266) and (b) nevirapine.

xazin-2-one ring is sandwiched between the sidechains of Leu100 and Val106, while also making edge-on contacts with Tyr318 and Val179. A prominent non-hydrophobic contact is the hydrogen bond formed between the benzoxazin-2-one NH and the mainchain carbonyl oxygen of Lys101. There is a van der Waals contact between the CG carbon of the sidechain of Lys103 and the nitrogen of the benzoxazin-2-one ring of efavirenz. In Figure 4 we compare the binding of efavirenz with the first-generation NNRTI nevirapine and the second-generation NNRTI UC-781 [11]. The cyclopropyl group of efavirenz overlaps the dimethylallyl group of UC-781 at the top of the NNRTI pocket. The benzoxazin-2-one ring of efavirenz sits parallel with the carboxanilide ring of UC-781, with less than 0.9 A˚

displacement for the two ring systems as well as the chlorine and oxygen/sulfur substituents. Several structural features of efavirenz also overlap with nevirapine (Figure 4); the trifluoromethyl group occupies the same space as the cyclopropyl group of nevirapine, and the propynyl–cyclopropyl group partially overlaps one pyridine ring. Analysis of the structure of the complex of efavirenz with the K103N mutant clearly shows the mutated sidechain and reveals a significant conformational rearrangement within the drug binding pocket when compared with the wild-type complex (Figure 5a). The greater bulk of the amide group of asparagine compared with the corresponding CG methylene group of lysine results in the efavirenz being pushed deeper into the pocket. This in turn makes the sidechain of Tyr181 undergo a flip to a “down” orientation such that its -OH group is displaced more than 8 A˚ from its position in the wild-type enzyme. This movement pushes the sidechain of Glu138 away from the pocket, breaking a hydrogen-bonding interaction with Lys101; this lysine then moves toward the mutated sidechain at residue 103. The movement of efavirenz deeper into the pocket perturbs the mainchain of ␤9 and ␤10, and it also effects sidechain movements of Trp229 and Leu234. The maximum displacements of efavirenz itself within the binding site are only 0.5 A˚ for the carbonyl oxygen and 0.7 A˚ for the cyclopropyl group; however, this rearrangement is sufficient to alter some of the contacts between efavirenz and the NNRTI site. The mutated sidechain at 103 (asparagine in place of lysine) is in position to interact via its OD1 atom with the efavirenz ring nitrogen.

Table 1. Crystallographic Structure Determination Statistics Data Collection Details Data set Data collection site Image plate Wavelength (A˚) Collimation (mm) Unit cell dimensions (a,b,c in A˚) Resolution range (A˚) Observations Unique reflections Completeness (%) Reflections with F/␴(F) ⬎ 3 Rmergeb

WT-efavirenz SRS PX7.2 MAR 18cm 1.488 0.20 139.8, 115.0, 65.4 (cell form F)e 30.0–2.5 87,374 31,707 85.2 25,268 0.062

K103N-efavirenz KEK BL-6A Fuji BAS III 1.000 0.10 140.3, 109.7, 74.4 (cell form C)e 30.0–2.9 83,108 23,338 89.2 15,980 0.118

K103N-nevirapine KEK BL-6A Fuji BAS III 1.000 0.10 139.7, 109.8, 72.9 (cell form C)e 30.0–2.9 81,694 24,586 94.9 16,930 0.151

2.59–2.5 2,682 69.7 973

3.0–2.9 2,006 75.3 445

3.0–2.9 2,146 84.5 559

30.0–2.5 30,166/1,541 0.218/0.301 0.214 7,614/21/113 0.008 1.3 65/70/43/50 3.1

30.0–2.9 22,214/1,124a 0.205/0.287 0.203 7,757/21/– 0.008 1.4 79/86/72/– 4.3

30.0–2.9 22,603/1,150 0.215/0.281 0.209 7,820/20/– 0.010 1.6 64/69/55/– 3.3

Outer Resolution Shell Resolution range (A˚) Unique reflections Completeness (%) Reflections with F/␴(F) ⬎ 3 Refinement Statistics Resolution range (A˚) Number of reflections (working/test) R-factorc (Rwork/Rfree) R-factorc (all data) Number of atoms (protein/inhibitor/water) Rms bond length deviation (A˚) Rms bond angle deviation (⬚) Mean B-factor (A˚2)d Rms backbone B-factor deviation (A˚2)

Positional restraints were applied to all atoms lying greater than 25 A˚ from the C␣ atom of residue 188 due to smaller number of observations. Rmerge ⫽ ⌺|I ⫺ ⬍I⬎|/⌺⬍I⬎ c R factor ⫽ ⌺|Fo ⫺ Fc|/⌺Fo d mean B factor for main chain, side chain, inhibitor, and water atoms, respectively. e Different crystal forms as described previously [30]. a

b

Although a hydrogen bond looks chemically possible, at a length of 3.7 A˚ it can only be weak. The nevirapine–RT(K103N) complex (Figure 5b), in contrast, shows very little rearrangement from the nevirapine–wild-type RT complex. Thus, the Tyr181 sidechain remains in the wildtype conformation, and there is minimal movement of the nevirapine itself. Discussion We suggest that there are at least three factors responsible for the improved properties of the second-generation NNRTIs such as efavirenz compared with the first-generation drugs such as nevirapine. Firstly, efavirenz shows distinctive features in its interaction with HIV-1 RT that help to explain its improved resilience to certain drug-resistance mutations. Thus, the smaller propynyl– cyclopropyl group of efavirenz compared with the structurally equivalent aromatic pyridine ring in nevirapine means that contacts with the readily mutable Tyr181 and Tyr188 sidechains are more limited and, hence, mutation to smaller nonaromatic sidechains such as cysteine has little effect on efavirenz binding. An additional significant feature of efavirenz binding to HIV-1 RT is the presence of a hydrogen bond to the mainchain C⫽O of residue 101, which is not present in the nevirapine complex. Such an interaction is less easily disrupted by mutation than is a sidechain interaction. Interestingly, another second-generation NNRTI, UC-781, shares both of these features with efavirenz, although they are achieved by structurally distinct means [11]. The second factor was first detected from structural studies of a series of carboxanilide NNRTIs with both first- and second-generation properties [11]. These studies indicated that the overall size of the compound in relation to its binding site might determine if the compound demonstrates first- or second-generation characteristics. We can now rationalize this from our current studies of the binding of efavirenz to wild-type and K103N mutant RTs as well as comparison with the nevirapine–RT(K103N) complex. We suggest that the significant structural rearrangements, involving both sidechain and drug-molecule movements, that we observe in the NNRTI pocket in the case of efavirenz binding to the RT(K103N) are characteristic of such compounds, enabling them to adapt to a mutated drug site. This leads to the third factor, which is the other side of the coin: a smaller drug can reposition itself within the pocket, whereas nevirapine’s bulky fused ring system remains relatively rigidly in place and no such rearrangements are possible. The three factors we have delineated are clearly interlinked. Thus, the relatively minor contacts of efavirenz with Tyr181 mean that this is not crucial for tight binding of this inhibitor, and hence rearrangement to what is a less-favorable conformation for binding first-generation inhibitors [18] does not significantly compromise its potency. From our studies of other NNRTIs binding to HIV-1 RT we have shown that there are alternative means than those described here for efavirenz by which the effects of the mutations at Tyr181 and Tyr188 on drug binding can be ameliorated.

Figure 2. Omit Maps Simulated-annealing omit electron density maps for the inhibitors and certain sidechains: (a) efavirenz–RT(wild-type); (b) efavirenz–RT(K103N), Asn103, and Tyr181; (c) nevirapine–RT(K103N) and Asn103. The maps are contoured at 4␴.

Structure 1092

Figure 3. Stereoview of Efavirenz in Its Binding Site on HIV-1 RT (Wild Type) Efavirenz is shown with the following atom colors: carbon, dark gray; nitrogen, blue; oxygen, red; fluorine, yellow; and chlorine, green. Protein sidechains are shown in a cyan skeletal representation, with the protein backbone shown as thin sticks colored gray. The hydrogen bond to the mainchain of Lys103 is indicated by a broken yellow rod.

Thus, for the emivirine analog GCA-186, introduction of 3,5dimethyl substitution of its benzyl ring allows greater interaction with the highly conserved Trp229 and less contact with Tyr181 and Tyr188 [19]. This effect is also observed with S-1153, which additionally has more extensive hydrogen bonding to the protein mainchain than does efavirenz [20]. The K103N mutation within HIV-1 RT seems to be a particular problem in the clinical use of NNRTIs, as this is the most frequently encountered mutation and gives wide cross-resistance to many NNRTIs [14]. One contributing factor to this could be the ability of the mutant asparagine residue at 103 to stabilize the unliganded RT structure. From model-building studies we have noted that the sidechain of Asn103 might be hydrogen bonded to the -OH of Tyr188 [9]. This could thus provide resistance to a wide range of NNRTIs by giving a reduced rate of association of the inhibitor–protein complex. Our structure of efavirenz in complex with RT(K103N) presents some possible clues as to how to counteract this effect. There is a potential hydrogen bond between the benzoxazin-2-one ring nitrogen of efavirenz and the OD1 atom of the Asn103 sidechain; however, the distance is rather long for a strong interaction (3.7 A˚). Clearly, analogs in which this distance were shortened could potentially interact more strongly with the mutant than the wild-type RT. Such targeting of the RT(K103N) enzyme should allow this mutation to be positively selected against. An analysis of the structural basis of the resilience of secondgeneration NNRTIs to drug-resistance mutations within HIV-1 RT has thrown up new ideas for the design of anti-HIV-1 drugs.

The development of such drugs will be of vital importance in continuing the fight against AIDS. Biological Implications The emergence of HIV-1 strains that are resistant to anti-retroviral drugs is a significant problem in the continuing effort to treat AIDS, and further new drugs are therefore required. A primary target for anti-retroviral drugs used in AIDS chemotherapy is the HIV-1 reverse transcriptase (RT), and knowledge of its three-dimensional structure is being used in the design of new compounds. The non-nucleoside inhibitors of HIV-1 RT (NNRTIs) are a drug class increasingly used in combination therapy. First-generation NNRTI drugs such as nevirapine are characterized by significant loss of binding activity in the presence of single point mutations within RT. In contrast, so-called second-generation NNRTI drugs, such as efavirenz, are more resilient to the presence of resistance mutations. It has been observed from clinical trials of a wide range of NNRTIs that the change of Lys103 to asparagine is the most commonly selected resistance mutation in HIV-1 RT. This work is aimed at determining structural mechanisms that give rise to the properties of efavirenz, a second-generation NNRTI, in comparison with nevirapine. The crystal structures of efavirenz and nevirapine in complexes with wild-type and K103N HIV-1 RT help to explain the difference in resilience properties to mutations between first- and second-generation compounds. The small cyclopropyl group of efavirenz (compared with a bulky aromatic ring of nevirapine) has few interac-

Figure 4. Comparison of Four NNRTIs Stereodiagram showing the conformational overlap of NNRTIs as they are observed bound to wild-type HIV-1 RT: efavirenz (red), UC-781 (blue) Cl-TIBO (yellow), and nevirapine (cyan). The superimposition was carried out using the C␣ atoms of amino acid residues around the NNRTI pocket to compensate for different crystal forms and domain rearrangements. Residues used included 94– 118, 156–215, and 225–243 from the palm domain, 317–319 from the connection domain of p66, and 137–139 from the fingers domain of p51.

Efavirenz Binding to HIV-1 RT 1093

Figure 5. Drug Binding Sites on HIV-1 RT Stereoviews of (a) efavirenz and (b) nevirapine in their binding sites on HIV-1 RT (wild-type and K103N). In each case the drug molecule and the 103 residue sidechain are colored orange for the wild-type complex and cyan for the K103N mutant complex. Other sidechains and mainchain regions are colored dark gray for wild-type and light gray for the K103N mutant.

tions with the two highly mutable tyrosine sidechains at positions 181 and 188, and it is therefore less dependent on these interactions. Additionally, there is a mainchain hydrogen bond for efavirenz but not for nevirapine, an interaction that is less easily disrupted by mutation of sidechains. The ability of the smaller efavirenz molecule to rearrange in the context of the mutated K103N binding site compared with nevirapine’s bulkier, rigid structure is a further factor in explaining the different binding properties of these two compounds to mutant RT. The principles derived here will be of value in the design of novel NNRTIs with improved resilience against drug-resistant RTs.

[28]). Model rebuilding was carried out with FRODO [29] on an Evans and Sutherland ESV.

Experimental Procedures

References

Crystallization and Data Collection The crystals of complexes of RTs with efavirenz were grown and soaked in 50% PEG 3400 using methods described previously [21, 22] X-ray data for the RT(wild-type)–efavirenz crystals were collected at 100 K at SRS Daresbury, UK, using 1.5⬚ oscillations with exposure times of 90 s. X-ray data for the complex of RT(K103N)–efavirenz and RT(K103N)– nevirapine were collected at the Photon Factory, KEK, Japan, at 16⬚C using a Weissenberg camera [23]. Data frames of 3.5⬚ with a coupling constant 1.5⬚/mm were collected with an exposure time of 140 s. The collimator, crystal enclosure, and camera cassette were helium filled. Indexing and integration of data images were carried out with DENZO, and data were merged with SCALEPACK [24]. X-ray data statistics are given in Table 1. Structure Solution and Refinement The structures were solved by molecular replacement using the RT-9-ClTIBO complex (Protein Data Bank accession code 1rev [25]) for RT(wildtype)–efavirenz, and RT-1051U91 complex (1rth [7]) for RT(K103N)–efavirenz and RT(K103N)–nevirapine complexes, as described previously [7], including the use of anisotropic scaling [26]. Rigid-body refinement, positional, simulated annealing, and individual B-factor refinement with bulk-solvent correction were performed with the programs X-PLOR [27] and CNS (version 0.4

Acknowledgments We would like to thank the staff of the SRS, Daresbury, UK and the Photon Factory, Tsukuba, Japan. D. I. S. is a member of the TARA project. We thank the UK MRC for long-term funding of the RT work with grants to D. K. S. and D. I. S. Received: May 15, 2000 Revised: August 15, 2000 Accepted: August 16, 2000

1. Brettle, R.P., et al., and Gore, S. (1998). Combination therapy for HIV: the effect on inpatient activity, morbidity and mortality of a cohort of patients. Int. J. STD AIDS 9, 80–87. 2. Mocroft, A., et al., and Lundgren, J.D. (1996). Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet 352, 1725–1730. 3. Iversen, A.K., et al., and Merigan, T.C. (1996). Multidrug-resistant human immunodeficiency virus type 1 strains resulting from combination antiretroviral therapy. J. Virol. 70, 1086–1090. 4. Larder, B.A., et al., and Pauwels, R. (1999). A family of insertion mutations between codons 67 and 70 of human immunodeficiency virus type 1 reverse transcriptase confer multinucleoside analog resistance. Antimicrob. Agents Chemother. 43, 1961–1967. 5. Huang, H., Chopra, R., Verdine, G.L., and Harrison, S.C. (1998). Structure of a covalently trapped catalytic complex of HIV-1 reverse transcriptase: implications for drug resistance. Science 282, 1669–1675. 6. Kohlstaedt, L.A., Wang, J., Friedman, J.M., Rice, P.A., and Steitz, T.A. (1992). Crystal structure at 3.5 A˚ resolution of HIV-1 reverse transcriptase complexed with an inhibitor. Science 256, 1783–1790. 7. Ren, J., et al., and Stammers, D. (1995). High resolution structures of HIV-1 RT from four RT-inhibitor complexes. Nat. Struct. Biol. 2, 293–302. 8. Ding, J., et al., and Arnold, E. (1995). Structure of HIV-1 RT/TIBO R

Structure 1094

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24. 25.

26.

27. 28.

29. 30.

86183 complex reveals similarity in the binding of diverse nonnucleoside inhibitors. Nat. Struct. Biol. 2, 407–415. Esnouf, R., Ren, J., Ross, C., Jones, Y., Stammers, D., and Stuart, D. (1995). Mechanism of inhibition of HIV-1 reverse transcriptase by nonnucleoside inhibitors. Nat. Struct. Biol. 2, 303–308. Spence, R.A., Kati, W.M., Anderson, K.S., and Johnson, K.A. (1995). Mechanism of inhibition of HIV-1 reverse transcriptase by nonnucleoside inhibitors. Science 267, 988–993. Ren, J., et al., and Stammers, D.K. (1998). Crystal structures of HIV-1 reverse transcriptase in complex with carboxanilide derivatives. Biochemistry 37, 14394–14403. Ren, J., Esnouf, R.M., Hopkins, A.L., Stuart, D.I., and Stammers, D.K. (1999). Crystallographic analysis of the binding modes of thiozolo-isoindolinone non-nucleoside inhibitors to HIV-1 reverse transcriptase and comparison with modelling studies. J. Med. Chem. 42, 3845–3851. Ren, J., et al., and Stammers, D.K. (2000). Phenylethylthiazolylthiourea (PETT) non-nucleoside inhibitors of HIV-1 and HIV-2 reverse transcriptases: structural and biochemical analyses. J. Biol. Chem. 275, 5633–5639. Schinazi, R.F., Larder, B.A., and Mellors, J.W. (1997). Mutations in retroviral genes associated with drug resistance. International Antiviral News 5, 129–135. Young, S.D., et al., and Emini, E.A. (1995). L-743, 726 (DMP-266): a novel, highly potent nonnucleoside inhibitor of the human immunodeficiency virus type 1 reverse transcriptase. Antimicrob. Agents Chemother. 39, 2602–2605. Balzarini, J., et al., and Karlsson., A. (1996). Highly favorable antiviral activity and resistance profile of the novel thiocarboxanilide pentenyloxy ether derivatives UC-781 and UC-82 as inhibitors of human immunodeficiency virus type 1 replication. Mol. Pharmacol. 50, 394–401. Kleim, J.-P., et al., and Paessens, A. (1993). Activity of a novel quinoxaline derivative against human immunodeficiency virus type 1 reverse transcriptase and viral replication. Antimicrob. Agents Chemother. 37, 1659–1664. Hopkins, A.L., et al., and Stuart, D.I. (1996). Complexes of HIV-1 reverse transcriptase with inhibitors of the HEPT series reveal conformational changes relevant to the design of potent non-nucleoside inhibitors. J. Med. Chem. 39, 1589–1600. Hopkins, A.L., et al., and Stammers, D.K. (1999). Design of MKC-442 (Emivirine) analogues with improved activity against drug resistant HIV mutants. J. Med. Chem. 42, 4500–4505. Ren, J., et al., and Stammers, D.K. (2000). Binding of the second generation non-nucleoside inhibitor S-1153 to HIV-1 RT involves extensive main chain hydrogen bonding. J. Biol. Chem. 275, 14316–14320. Stammers, D.K., et al., and Stuart, D.I. (1994). Crystals of HIV-1 reverse transcriptase diffracting to 2.2 A˚ resolution. J. Mol. Biol. 242, 586–588. Ren, J., et al., and Stammers, D.K. (1998). 3⬘-azido-3⬘-deoxythymidine drug resistance mutations in HIV-1 reverse transcriptase can induce long range conformational changes. Proc. Natl. Acad. Sci. USA 95, 9518–9523. Sakabe, N. (1991). X-ray diffraction data collection system for modern protein crystallography with a Weissenberg camera and an imaging plate using synchrotron radiation. Nucleic Instr. Methods Phys. Res. 303, 448–463. Otwinowski, Z., and Minor, W. (1996). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. Ren, J., et al., and Stuart, D. (1995). The structure of HIV-1 reverse transcriptase complexed with 9-chloro-TIBO: lessons for inhibitor design. Structure 3, 915–926. Esnouf, R.M., et al., and Stuart, D.I. (1997). Unique features in the structure of the complex between HIV-1 reverse transcriptase and the bis(heteroaryl)piperazine U-90152 explain resistance mutations for this nonnucleoside inhibitor. Proc. Natl. Acad. Sci. USA 94, 3984–3989. Bru¨nger, A.T. (1992). X-PLOR (Version 3.0) Manual. (New Haven, CT: Yale University Press). Bru¨nger, A.T., et al., and Warren, G.L. (1998). Crystallography and NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr. D 54, 905–921. Jones, T.A. (1985). Interactive computer graphics: FRODO. Methods Enzymol. 115, 157–171. Esnouf, R.M., et al., and Stuart, D.I. (1998). Continuous and discontinuous changes in the unit cell of HIV-1 reverse transcriptase crystals on dehydration. Acta. Crystallogr. D 54, 938–954.

Protein Data Bank ID Codes The coordinates have been deposited in the Protein Data Bank (PDB) with accession codes 1fk9, 1fko, and 1fkp for RT(wild type)–efavirenz, RT(K103N)–efavirenz, and RT(K103N)–nevirapine, respectively.