New Antivirals and Drug Resistance

Home , Brecanavir, Neuraminidase inhibitor

ANRV378-BI78-05 ARI 29 April 2009 10:11

Peter M. Colman

The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia 3050; email: [email protected] by University of Texas - Austin on 10/05/09. For personal use only. Annu. Rev. Biochem. 2009. 78:95–118 Key Words First published online as a Review in Advance on Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org antibody, HIV, inﬂuenza March 2, 2009

The Annual Review of Biochemistry is online at Abstract biochem.annualreviews.org Progress in the discovery of new antiviral medicines is tempered by This article’s doi: the rapidity with which drug-resistant variants emerge. A review of 10.1146/annurev.biochem.78.082207.084029 the resistance-suppressing properties of four classes of antivirals is Copyright c 2009 by Annual Reviews. presented: inﬂuenza virus neuraminidase inhibitors, HIV protease in- All rights reserved hibitors, antibodies, and protein-based fusion inhibitors. The analysis 0066-4154/09/0707-0095$20.00 supports the hypothesis that the more similar the drug is to the target’s natural ligands, the higher the barrier to resistance. However, other fac- tors, such as entropy compensation and solvent anchoring, might also be exploited for improved drug design.

95 ANRV378-BI78-05 ARI 29 April 2009 10:11

an early example of structure-based drug dis- Contents covery. Two drugs are approved, a third is in clinical evaluation, and a fourth remains unde- INTRODUCTION AND SCOPE.... 96 veloped. Even though influenza is most com- INFLUENZA VIRUS monly an acute infection, clinical drug resis- NEURAMINIDASE ...... 96 tance is observed. The two approved drugs have Neuraminidase Inhibitors ...... 97 very different resistance characteristics that can Resistance ...... 99 be traced to the detail of their interaction with The N1 Subtype ...... 101 the conserved catalytic machinery. THE HIV PROTEASE...... 102 The neuraminidase of influenza virus (1) is Inhibitors and Drugs ...... 102 a homotetrameric glycoprotein anchored by a Prodrugs ...... 105 fibrous stalk in the viral membrane. Its pri- Off-Target Resistance ...... 106 mary role in the infectious cycle is to liber- BIOLOGICALS AS ANTIVIRALS . . . 107 ate progeny from infected cells, but some roles Neutralizing Antibodies ...... 107 in mobilizing the virus through mucous secre- Inhibitors of Viral Membrane tions has also been described. The enzyme lib- Fusion...... 109 erates N-acetyl neuraminic acid (Neu5Ac) from SUMMARY ...... 112 α2-3 or α2-6 linkages to galactose, thereby de- stroying the receptor for the virus. All of the enzymatic and antigenic properties of the pro- INTRODUCTION AND SCOPE tein are associated with the tetrameric globular head, which is liberated when virus is treated Studies over the past one to two decades have with proteolytic enzymes. Ten serologically dis- yielded many examples of new drugs and drug tinct neuraminidases have been characterized candidates for treatment or prevention of vi- thus far, N1 through N9 associated with type A ral infections. Most of these discoveries have influenza, and type B neuraminidase from type been aided, if not initiated, by knowledge of B influenza viruses. The crystal structures of the three-dimensional (3D) structure of the tar- representatives of the N2 (2, 3), N9 (4), and get. A large body of structural data accompa- type B (5) neuraminidases provided the plat- nies these new medicines and informs an anal- form for the discovery of the first examples of ysis of one of the major issues that bedevils the the neuraminidase inhibitor drug class. Struc- management of infectious disease, namely the tures of compounds discussed below are de- rapid emergence of drug resistance. The differ- picted in Figure 1. by University of Texas - Austin on 10/05/09. For personal use only. ential capacity of drugs to suppress the selec- The sequence and structural invariance of tion of resistant virus is a property of both the the enzyme active site, located on the axis of drug and its target. The discovery and clinical Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org a β-propeller fold, held the promise for drugs experience of neuraminidase inhibitors for ineffective against not only all antigenic strains fluenza and selected drugs for HIV are reviewed of a particular subtype but also all subtypes and in the context of drug resistance. Experience type B (1). The catalytic site has four invagi- with protein-based antivirals, including emerg- nations accommodating substituents at C2,C4, ing structure-based approaches to vaccine de- C5, and C6 on the substrate (6) (Figures 1a and sign for HIV, are also discussed in this setting. 2a). A cluster of three arginyl residues (R118,

R292, R371) encircles the C2-carboxylate of the substrate, which binds in a twist-boat conﬁgu- INFLUENZA VIRUS ration. A tryptophan residue (W178) engages NEURAMINIDASE the C5-acetyl moiety. A network of hydrogen The neuraminidase inhibitors for treatment bond donors and acceptors engage the C6- and prevention of inﬂuenza virus infection are glycerol and C4-hydroxyl, as commonly found

96 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

in carbohydrate-binding proteins and enzymes. a Neu5Ac b Neu5Ac2en Notable among these residues is E276, which forms hydrogen bonds to the C8 and C9 hydrox- O HO O HO yls from the glycerol moiety of the substrate. 4 3 OH O O 5 The catalytic nucleophile is believed to be N 2 1 N H - H - Y406. These amino acids are “preset” for catal- 6 O O OO ysis, because binding of Neu5Ac (or the tran- HO 7 H HO H sition state analog Neu5Ac2en) (Figure 1b)to 8 OH OH N2, N9, and subtype B neuraminidases causes HO 9 HO no conformational change in the enzyme. Such conditions are likely to be optimal for designing drugs that are truly substrate or transition state c 4-Amino-Neu5Ac2en d Zanamivir mimetics. H2N +H N NH O 3 2 + Neuraminidase Inhibitors O O HN N O H Replacement of the C4-hydroxyl moiety of OO- N H - Neu5Ac2en with amino- or guanidino sub- HO H O O stituents (Figures 1c,d and 2b), designed to OH HO H optimize chemical and shape complementarity HO OH to the enzyme, resulted in approximately 100- HO and 10,000-fold improvements, respectively, in binding potency (7). Zanamivir (4-guanidino- Neu5Ac2en) binds in the active site almost ex- e Diethyl-carboxamide f Oseltamivir carboxylate analog of part c actly as predicted, but the guanidinium is not hydrogen bonded to E119, rather it forms an +H N +H N O 3 O 3 ionic contact with it through a stacking in- O O teraction (8). Although E119 is conserved in N N H H wild inﬂuenza viruses, it is evidently not a cat- OO- O- alytic residue because different amino acids are O O found in its place in bacterial and mammalian N neuraminidases. Intranasal or inhaled dosing of zanamivir is effective in treatment of inﬂuenza Figure 1 by University of Texas - Austin on 10/05/09. For personal use only. in experimental animals (7) and in man (9), but Chemical structures of neuraminidase ligands. it is poorly distributed when taken orally, even

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org as an ester prodrug. glycerol replacement (Figure 1e) is illustrated Various attempts to improve the pharma- (Figure 2c). One arm of the diethyl moiety of cological properties of zanamivir focused on the carboxamide side chain buries itself in a replacing the glycerol moiety with hydropho- hydrophobic pocket created by the movement bic substituents linked through a carboxamide of E276. However, amino acid sequence dif- (10, 11). This study revealed that, although ferences between type A and B neuraminidases the catalytic machinery of the type A and B conspire to present a higher energy barrier to neuraminidases appeared almost identical, sub- this modest structural transition in type B than tle but signiﬁcant differences exist in the im- in N2 or N9 neuraminidases. A series of car- mediately surrounding structure. These differ- boxamide analogs of Neu5Ac2en show simi- ences manifest themselves as selective binding lar selectivity for type A viruses over type B, of Neu5Ac2en analogs, in this case preferen- making them unattractive clinical candidates tially to type A (12). An example of such a (12).

www.annualreviews.org • New Antivirals and Drug Resistance 97 ANRV378-BI78-05 ARI 29 April 2009 10:11

R224

H274 D151 E276 E119

R292 R118 R371

R224 R224 E276

E276

by University of Texas - Austin on 10/05/09. For personal use only. Figure 2

(a) Neu5Ac binds to neuraminidase in a twist-boat conﬁguration. The C2-carboxylate is surrounded (left to right) by R292, R371, and R118. E119 and D151 are close to the C4-hydroxyl of the sugar. The

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org C6-glycerol-binding site includes E276 and R224. H274 is at the far left. The enzyme structure does not change when Neu5Ac binds [Protein Data Bank (PDB) code 1MWE] (124). Molecular ﬁgures prepared with PyMOL (125). (b) Zanamivir bound to N9 neuraminidase (PDB code 1NNC). Neu5Ac2en is a transition state analog, and zanamivir is the 4-guanidino derivative of Neu5Ac2en. The protein surface is shaded gray. (c) A carboxamide analog of 4-amino Neu5Ac2en introduces a change in the conformation of the active site at E276, which is now engaged with R224, to create a hydrophobic pocket for the “deep” ethyl group (PDB code 2QWJ). (d ) Oseltamivir carboxylate induces the same conformational change seen in c (PDB code 2QWK).

A more radical redesign of the pyran to a ester by host enzymes. The pentyl ether (which carbocyclic scaffold (Figure 1f ) led to the dis- now replaces the glycerol) induces similar but covery of the orally active neuraminidase in- subtly different conformational changes in the hibitor oseltamivir (13). The active ingredient is enzyme to those observed in the Neu5Ac2en generated by the removal of the carboxylic acid carboxamides (Figure 2d ), but now binding

98 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

to type B neuraminidases is less compromised selected in tissue culture by zanamivir, are and leaves open a therapeutic window. A more E119A and E119D. The glycine and alanine comprehensive account of the discovery of mutants cause loss of binding of zanamivir by the neuraminidase inhibitor class is available up to 1000-fold, and the aspartic acid mutant (14). by up to 2500-fold (18). A resistant virus selected with one of the carboxamide analogs (19) revealed a mutation in Resistance one of the three argininyl residues that char- The earliest attempts to culture neuraminidase acterize all known neuraminidases, R292. Sub- inhibitor-resistant influenza viruses with stitution to K292 results in some loss of en- Neu5Ac2en analogs resulted in viruses with zyme activity but a large (250- to 1000-fold) loss off-target mutations in the receptor-binding of binding potency to carboxamide analogs of site of the hemagglutinin (15), revealing Neu5Ac2en. Structural studies of carboxamides an essential balance between the affinity of bound to the R292K variant reveal that E276 the hemagglutinin for the receptor and the does change conformation to allow binding into efficiency of destruction of the receptor by the the hydrophobic pocket (20). The R292K vari- neuraminidase. In separate experiments, variant most likely acquires resistance by introduc- ants in the catalytic site of the neuraminidase ing a barrier to the structural transition through were observed (16). Zanamivir selected viruses which E276 engages R224 and creates the novel with E119G, rationalized by the absence of the hydrophobic binding pocket (Figure 3a). That ionic interaction with the novel 4-substituent barrier is a hydrogen bond between E276 and of the drug. E119G has the same enzyme K292, an interaction revealed in the crystal activity as the wild type but is less stable than structure of the unliganded R292K variant and wild type (17). Other variants at that position, not accessible in the wild-type structure.

ab Carboxamide Wild type Oseltamivir carboxylate Mutant by University of Texas - Austin on 10/05/09. For personal use only. Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org

Figure 3 (a) Overlays of the carboxamide of Figure 2c with oseltamivir carboxylate bound to the R292K mutant N9 neuraminidase. Binding to both compounds is compromised by the mutation. A structural rationale is that reduction of carboxamide binding is due to the penalty of inducing the conformational change, whereas for oseltamivir carboxylate, it is due to the failure to create the hydrophobic binding site (PDB codes 2QWG, 2QWH). (b) Overlay of oseltamivir carboxylate bound to wild-type and H274Y mutant N1 neuraminidase. The larger tyrosyl residue in the mutant seems to prevent the movement of E276 that is needed for high-afﬁnity binding (PDB codes 2HU4 and 3C10).

www.annualreviews.org • New Antivirals and Drug Resistance 99 ANRV378-BI78-05 ARI 29 April 2009 10:11

The loss of inhibitory potency toward less widely used in the treatment of seasonal R292K N9 neuraminidase of a panel of influenza than has oseltamivir. neuraminidase inhibitors, including Neu5Ac, Both zanamivir (oral inhalation) and os- Neu5Ac2en, 4-amino-Neu5Ac2en, zanamivir, eltamivir (oral) are prescribed for acute 5- two Neu5Ac2en carboxamides, and oseltamivir day treatment of infected patients. Analysis of carboxylate, is greater the less the inhibitor re- matched pairs of virus from 41 patients treated sembles the substrate (20). The outlier in simi- with zanamivir did not reveal any examples of larity to the substrate is oseltamivir carboxylate, drug-resistant virus (22). Similar studies, with for which a 6500-fold loss of binding was deter- much larger patient numbers, reveal that two mined in the structural background of N9. This percent of oseltamivir-treated individuals shed analysis supports the principle that drugs that drug-resistant virus (23). That number is as more closely resemble natural substrates or lig- high as 18% in one pediatric study wherein the ands are better equipped to suppress the emer- dosing is thought to have been suboptimal (24). gence of a drug-resistant virus. Suppression of The R292K mutation is frequently seen in these such a virus demands a minimal loss of in- studies. Other mutations detected in these pa- hibitory activity toward the mutant and a min- tients are E119V and N294S. The appearance imal loss of biological activity by the mutant. of drug-resistant viruses in these treated pa- A drug-resistant virus must be able to maintain tients does not noticeably compromise the effi- its capacity to bind to and (if necessary) pro- cacy of treatment, although it may render pro- cess its natural substrates while discriminating phylaxis to immediate contacts ineffective. This between them and the drug. may be the way in which amantadine-resistant Interestingly, oseltamivir carboxylate does influenza viruses have taken hold (25, 26). not induce the conformational switch in E276 In the N1 antigenic background, the H274Y in R292K mutant N9 neuraminidases (20), un- mutant is selected by oseltamivir and reduces like the analogous carboxamides (Figure 3a). sensitivity to the drug by several hundredfold Oseltamivir carboxylate does not properly en- (27). Unlike a number of other active-site vari- gage with the R292K enzyme, the pentyl ether ants, the H274Y virus is transmissible in fer- resting “uncomfortably” over the hydrophilic rets (28). Furthermore, this mutant was found binding site used by the glycerol moiety of the in early 2008 in H1N1 human influenza viruses substrate. It may be that the same subtleties that in Europe (29). Of 437 viruses tested, 59 were allow oseltamivir carboxylate to retain binding resistant to oseltamivir by assay or by identifi- efficacy to wild-type type B neuraminidases are cation of H274Y, yet there is no evidence that at play. In any event, loss of binding potency of any of the patients from whom samples were ob- by University of Texas - Austin on 10/05/09. For personal use only. oseltamivir carboxylate to the N2 subtype neu- tained had either been treated with oseltamivir raminidases has been estimated to be as high as or exposed to individuals who had. This ob-

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org 30,000-fold (21). servation is causing concern lest this muta- Although experiments in tissue culture are tion become associated with an avian H5N1 informative, selection pressures in infected pa- strain and render stockpiled oseltamivir inef- tients on drug therapy are very different and fective against an H5N1 pandemic avian in- can lead to quite different outcomes compared ﬂuenza. The molecular basis for resistance to to in vitro studies. Different dosing regimes and H274Y in N1 neuraminidase was described fol- routes of administration result in different lev- lowing a structural study of the mutant en- els of drug at the site of infection (21a, 21b), and zyme bound to both zanamivir and oseltamivir natural clearance mechanisms may be at play in carboxylate (30). Y274 in the mutant prevents clearing even drug-resistant viruses. A head-to- E276 from adopting the necessary “switched” head comparison of zanamivir and oseltamivir conformation demanded by the pentyl ether is compromised because zanamivir has been far side chain, whereas substrate and zanamivir are

100 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

still accommodated (Figure 3b). The 265-fold development in resistance is the H274Y muta- loss of potency in H274Y (N1) for oseltamivir tion in H1N1 viruses from patients apparently carboxylate is due to a 10-fold slower asso- naive to any inhibitor. ciation rate and a 25-fold faster dissociation rate. In comparison, the kinetics for binding to zanamivir are essentially the same as for The N1 Subtype the wild-type H274 (30). Another N1 mutant, Neuraminidase subtypes commonly found in N294S, is also resistant to oseltamivir carboxy- human influenza viruses are N1, N2, and type late by virtue of hydrogen bonding to E276 B. Other subtypes occur in wild and domestic and restraining the necessary conformational bird populations. H5N1 viruses are the cause of change associated with drug binding (30). the contemporary highly pathogenic avian in- Baseline sensitivity of type B viruses to both fluenza, a virus which has prompted many gov- zanamivir and oseltamivir is lower than for ernments around the world to stockpile neu- type A viruses, and the emergence in Japan raminidase inhibitors as a first line of defense of wild influenza B strains with further re- against its spread in humans. The 3D structure duced sensitivity to neuraminidase inhibitors is of the N1 subtype is therefore of great interest, of some concern (31, 32). These viruses have even though there were no grounds (on the ba- been isolated from patients not undergoing sis of its amino acid sequence) for believing that treatment with neuraminidase inhibitors, and it would depart from the structures displayed by transmission between individuals seems highly N2, N9, and type B neuraminidase. likely. Three such mutants were reported with However, the recently described structure of

the following mean IC50 (nM) values against the N1 neuraminidase subtype, together with zanamivir and oseltamivir, respectively, D198N that of the N4 and N8 subtypes, reveals a new (50, 240), I222T (25, 480), and S250G (190, subsite within the enzyme active site (34). As a 50). One additional mutant, G402S (50, 280), result of a novel fold for a short loop segment of appeared during treatment with oseltamivir. the structure, amino acids 147–152 in N2 num-

Wild-type sensitivity (IC50) in this study was bering, the pocket adjacent to the C4-hydroxyl 6 nM for zanamivir and 72 nM for oseltamivir. moiety of Neu5Ac is further opened up, invit- Oseltamivir was some 90 times more widely ing exploration by medicinal chemists. Amino used in Japan than zanamivir in 2004/2005, and acid residues affected by the altered loop con- it is possible that this has led to the altered drug formation include E119 and D151; the former sensitivity in the communities studied (31). is one of the amino acids targeted by both the Testing the sensitivity of clade 1 and clade guanidinium group of zanamivir and the amino by University of Texas - Austin on 10/05/09. For personal use only. 2 H5N1 neuraminidases to neuraminidase in- group of oseltamivir carboxylate, and the lat- hibitors has revealed that clade 2 viruses isolated ter is a putative catalytic residue whose pre-

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org from Indonesia in 2005 are six- to sevenfold less cise function is unknown. The signature amino sensitive to oseltamivir than clade 1 viruses (33). acid sequence in N1, N4, and N8 that produces It is important to determine if this difference this novel architecture has not been identiﬁed extends to clade 2 viruses isolated elsewehere but seems to include residues beyond the loop and to identify its underlying molecular basis, itself. which remains equivocal on the basis of current When inhibitors were introduced to N1 or sequence analysis. No difference in sensitivity N8, either at low inhibitor concentration or to zanamivir was detected in this study. with short incubation time, binding was ob- A comparison of resistance properties of os- served in the active site with no change in

eltamivir and zanamivir must recognize the vast conformation, i.e., with the enlarged C4 sub- excess of use of the one over the other. A nu- site still intact (34). However, higher concen- merical excess of reports of drug resistance to trations of ligand caused the loop to close in oseltamivir is expected. The most troubling on the catalytic site, yielding a structure that

www.annualreviews.org • New Antivirals and Drug Resistance 101 ANRV378-BI78-05 ARI 29 April 2009 10:11

appears indistinguishable from those reported diverse, including at the P1 (four occurrences earlier for N2, N9, and type B neuraminidase of F, three of L, and one of Y, M, and N) and inhibitor complexes. The inhibitory concentra- P1 (three P, two F, W, Y, L, M, and A) sites tions of zanamivir and oseltamivir carboxylate (42). Even a palindromic sequence of amino for N1 are not appreciably weaker than the acids is stereochemically asymmetric, and as- ∼1 nM values observed for N2 and N9, sug- sociation of the protease dimer with its var- gesting that the conformational rearrangement ious substrates is inherently asymmetric (43).

associated with closure of the C4 subsite in N1 Furthermore, access of substrate to the ac- comes with little energetic penalty. No function tive site requires a movement of the so-called

has yet been ascribed to the enlarged C4 sub- flaps, which are open and loosely structured in site of N1. It seems unlikely to be of catalytic the apoenzyme (39, 44) and close over bound significance because of the structural similar- substrates and inhibitors (Figures 4 and 5a,b). ity of all complexes of influenza neuraminidases These features contrast starkly with the in- with Neu5Ac2en. Other possible functions for fluenza neuraminidase wherein the specificity an enlarged catalytic site are altered substrate for terminal Neu5Ac is paramount and little or specificity for the enzyme and even a role in re- no structural rearrangement of the enzyme oc- ceptor binding, as in the parainfluenza viruses curs during binding. Thus, in addition to the wherein a single active site in the HN protein catalytic machinery, functional domains of the supports both receptor-binding and catalytic protease include those involved in its essential functions (35). No current data support any of dynamic properties and in its dimerization, and these possibilities. these represent legitimate, if difficult, drug tar- In the event that the new site lacks any bi- gets. A further distinguishing feature of the pro- ological function, targeting it for new antiviral tease from the neuraminidase is that its sub- discovery may be of limited therapeutic benefit strates are also the products of a viral gene and should resistant viruses rapidly emerge. Nev- hence susceptible to selection pressure. Indeed, ertheless, exploiting this feature in the design protease drug-resistant variants with mutations of new neuraminidase inhibitors would further in both the enzyme active site and in the sub- test the hypothesis that greater similarity be- strate have been described. tween a drug and a natural ligand leads to better suppression of resistant virus. Inhibitors and Drugs All but one of the currently registered HIV THE HIV PROTEASE protease inhibitors are peptidomimetics (37). by University of Texas - Austin on 10/05/09. For personal use only. Although enzymes are generally viewed as fa- A vast literature has accumulated on drug vorable drug targets, the protease of HIV poses resistance-associated mutations, revealing that

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org special problems owing to the dynamic proper- substitutions in the substrate-binding cleft are ties of the proteins that control access of sub- often associated with substitutions that de- strates to the active site. Nevertheless, protease termine movement of the flaps, the former inhibitors have become an integral component compromising substrate (and drug) binding of antiretroviral therapy (36, 37). and the latter providing compensating en- The cleavage of the Gag and Pol polypro- hanced catalytic efficiency. Thus, the double teins of HIV by the viral protease is an es- mutant V82T/I84V directly impacts substrate sential step in maturation of the virus. The (Figure 5a) and inhibitor (45) (Figure 5b) in- protease is a symmetric dimer with a pair of teractions in the P1 and P1 sites, and the dou- active-site aspartyl residues (D25, D25) in close ble mutant M46I/L63P is rate enhancing for proximity to the symmetry axes and the scis- many of the natural cleavage events (46). These sile bond of the substrate (38–41). The 10 nat- four mutations alone confer broad resistance ural polyprotein cleavage sites are remarkably to the first generation of protease inhibitors.

102 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

a Amprenavir b Tipranavir

NH CF OH 2 3 H OH H O N N N O S O S N OO OO O O

c Darunavir d Brecanavir

NH OH 2 OH O H H H H O N N O N N O O S O S OO H H O H H O OO O O N O S

e TMC-126 f R=H, GS-8373; R=Et, GS-8374

OH O OH O H H H O N N H O N N O S O S OO H H O H H O OO O O OR O P OR O

Figure 4 Chemical structures of protease inhibitors.

Computational studies support the notion that emergence of future drug-resistant strains. A the compensatory mutation M46I alters the dy- structural study of six of the natural substrates, by University of Texas - Austin on 10/05/09. For personal use only. namic properties of the ﬂaps (46a). The struc- complexed with an inactive (D25N) mutant tural basis for other compensatory mutants protease, has led to the idea that speciﬁcity is de-

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org (e.g., L90M, which is spatially adjacent to the termined primarily by substrate shape (42) and catalytic aspartate D25) is more subtle. In that that the envelope defined by this common shape case, a computational simulation (46b) fails to should guide the design of future inhibitors (49, capture the experimentally determined poise of 50). Analysis of the loss of inhibitory potency of nelfinavir bound to the mutant enzyme (46c). five first-generation protease inhibitors to 13 Currently, mutations at some 15 residues are different protease mutants shows some corre- characterized in the Stanford HIV database (47, lation with the volume of the inhibitor lying 48) as major resistance mutations for one or outside the substrate envelope (50). This study, more of the approved protease inhibitor drugs. which formally takes no account of the chemical Attention has recently become focused similarity between the inhibitor and the sub- on strategies for improving the performance strate, is for five peptidomimetic drugs, each of the drug class both at inhibiting known with a noncleavable hydroxyethylene linkage in drug-resistant strains and at suppressing the place of the scissile peptide bond (Figure 5b).

www.annualreviews.org • New Antivirals and Drug Resistance 103 ANRV378-BI78-05 ARI 29 April 2009 10:11

P1' pocket residues a Catalytic HIV protease dimer aspartates

ARVLAEAM P1 pocket residues

bcProtease surface

Protease

GS-8373

Core of peptide (VLAE) Amprenavir

Figure 5 ( ) The HIV protease dimer showing the two catalytic aspartates (mutated to N25 in this structure), the P1, by University of Texas - Austin on 10/05/09. For personal use only. a and P1 pocket residues 82 and 84, and the peptide substrate, ARVLAEAM (PDB code 1F7A). (b) Overlay of amprenavir bound to the protease (PDB code 1HPV) with the core of the peptide (VLAE) from panel a. (c) GS-8373, a darunavir analog, modifed with a phosphonate at P1, bound to the protease (trace in green), Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org illustrating the exposure of the phosphonate at the protease surface (shaded green) (PDB code 2I4D).

Thus, chemical similarity (as well as shape simi- Unlike other inhibitors, binding of tipranavir to larity) to substrate is apparent in all these drugs, the protease is almost entirely entropy driven and a strong chemical similarity to substrate has (H -0.7 Kcal/mol; -TS -13.9 Kcal/mol) previously been qualitatively scored to ampre- (54). The mechanism by which it maintains navir (Figure 4a) (51), the drug favored by the binding afﬁnity to four different drug-resistant substrate envelope analysis. mutants, including one selected by tipranavir Tipranavir (Figure 4b), the only nonpep- itself, has been analyzed crystallographically tidomimetic compound (52) currently licensed, and thermodynamically, showing that the en- also selects resistant variants, albeit slowly tropic losses that accompany binding to certain and requiring multiple sequence changes (53). mutants are usually either offset by enthalpic

104 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

gains or, at worst, associated with minimal performance of darunavir in the clinic in sup- enthalpic losses. In contrast, in cases where pressing drug resistance. measurements are available for peptidomimetic Brecanavir (Figure 4d ) is also a derivative of inhibitors, loss of binding to drug-resistant amprenavir/darunavir. It has a P1 site extension mutants is largely accounted for by enthalpic terminating with a thiazole, which structurally losses. Although the structural correlates of this closely overlays with the phosphonate of GS- anomalous behavior of tipranavir remain poorly 8374 (see below) (59). Brecanavir also retains understood, tipranavir binding to the enzyme potency against a broad spectrum of protease is characterized by seven direct (as opposed inhibitor-resistant viruses. In vitro brecanavir to water-mediated) hydrogen bonds to “con- selects mutants in the protease active site, al- served” active-site elements (54). The rather though high-level resistance was only observed different resistance profile of tipranavir to other in the context of A28S, a mutation that reduces inhibitors advocates its use in combination ther- viral replication efficacy (60). Despite this en- apies. There is not yet any evidence to suggest couraging profile with respect to potency and that tipranavir as a single agent is any more or resistance, the development of brecanavir was less effective in suppressing resistance than are halted in December 2006 because of formula- the peptidomimetics. tion problems, another reminder of the many The second generation of peptidomimetic competing properties that are sought in drug protease inhibitors is exemplified by darunavir candidates. (Figure 4c), an analog of amprenavir modified in the P2 site (55, 56). Structural studies of darunavir and amprenavir bound to Prodrugs wild-type protease and to a triple mutant One way to improve the therapeutic profile of (L63P/V82T/I84V) reveal additional interac- a drug is to alter its distribution in favor of tions between darunavir and the enzyme; these the site of action of the drug target, in this interactions explain its 100-fold tighter binding case inside an infected cell. Increasing the ef- than amprenavir to wild-type protease. Against fective concentration of the drug might sup- the triple mutant, darunavir retains a 33-fold press mutants that would otherwise be viable. advantage in potency over amprenavir. One fea- Thus, attachment to a protease inhibitor scaf- ture of the interactions between darunavir and fold of a charged phosphonate, masked by a the protease is the extent to which backbone chemical moiety that is selectively removed by atoms of the protease are engaged, including intracellular enzymes, should improve the re- by the novel bis-tetrahydrofuranylurethane P2 tention of the compound inside cells (61). Pur- by University of Texas - Austin on 10/05/09. For personal use only. moiety (57). Because of the widely conserved suit of this strategy, based on derivitizing the P1 conformation of the protein backbone across position of TMC-126 (Figure 4e) (an analog

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org diverse protease mutants, a strategy of targeting of amprenavir and darunavir), led to the unex- the backbone promises to suppress drug resis- pected finding that a phosphonate-containing tance. Backbone hydrogen bonds to substrate inhibitor, GS-8373 (Figure 4f ), has an im- are a feature of the six substrate complexes de- proved binding profile to a broad range of scribed above (42), and in this sense, target- inhibitor-resistant protease variants. Crystallo- ing the backbone demands substrate similarity. graphic analysis reveals that the phosphonate However, darunavir does make an H-bond to moiety of GS-8373 projects into the solvent the backbone amide of D30, an interaction not (Figure 5c), on a track adjacent to but not shared with substrates. Darunavir is marginally overlapping with the path used by substrates, more effective in inhibiting flap mutants (at po- where it is loosely ordered with few if any spe- sitions 48, 50, and 54) than is saquinavir (58). cific interactions with the protease. The di- It will be especially interesting to observe the ethyl ester prodrug (GS-8374) (Figure 4f )is

www.annualreviews.org • New Antivirals and Drug Resistance 105 ANRV378-BI78-05 ARI 29 April 2009 10:11

an equally potent inhibitor in vitro, supporting dynamic incorporation of substrate into the ac- the conclusion that the phosphonate makes no tive site, in which case resistant mutants may be specific interactions with the protease. The per- unviable. Its near neighbors on the protein are formance of the two compounds in antiviral as- G52, F53, and P81, and mutation at G52 or P81 says is as predicted: The charged phosphonate has not yet been reported (47). If these residues is inactive, and the neutral form shows equal are somehow functionally important, then the or better activity to TMC-126. Isothermal approach for HIV holds great promise. Exem- titration calorimetry of GS-8373, GS-8374, plifying the generalization of this approach to TMC-126, and amprenavir to two mutant other targets will be watched with great interest. proteases (M46I/I47V/I50V and I84V/L90M) showed that the phosphonates retain activity to the mutants, whereas the nonphospho- Off-Target Resistance nates suffer a 10- to 40-fold loss of activity. A Off-target resistance to protease inhibitors has similar trend was observed for phosphonate- also been reported. Mutations in cleavage sites derivatized atazanavir. In all cases studied, the of the Gag polyprotein have previously been calorimetric data show enthalpic losses in the viewed as “rescue mutants” in the context of binding interaction to mutants compared with protease mutations that directly affect drug wild type, but entropic compensation for these binding (63). However, some studies have de- losses is always greater in the phosphonate- tected resistance to protease inhibitors with- containing inhibitors thereby minimizing the out concomitant mutations in the protease that loss of binding potency to the mutants. A de- affect drug binding (64), suggesting off-target tailed structure analysis shows that the poise of mutations. More recently in vitro selection with the TMC-126 core is identical in wild type and the protease inhibitor RO033-4649 has re- in the I84V/L90M mutant. In contrast, a small sulted in drug-resistant variants possessing Gag but significant difference is observed for GS- mutations with no protease mutations (65); the 8374 in wild type compared to mutant, suggest- Gag mutants were associated with increased ing it is more adaptable to its binding site than processing efficiency for one of the intermedi- is TMC-126 and could be considered anchored ate products (NCp6). It was also shown that the by the solvent (61). NC/p1 cleavage site mutant A431V could con- This approach is suggested as a way for- fer fourfold resistance to ritonavir. Evidently, ward for suppression of drug-resistant virus increasing the susceptibility of the polyprotein (61, 62). Attempts to culture virus in the to protease processing is a novel and bona fide presence of various phosphonate-containing mechanism by which HIV acquires resistance by University of Texas - Austin on 10/05/09. For personal use only. protease inhibitors were unsuccessful, but to protease inhibitors (65). Longer-term cul- parallel studies with TMC-126 generated an turing of virus with GS-8374 led to a 15-fold

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org L10F/A28S/M46I/I50V mutant. These results loss of drug sensitivity owing to mutations in seem at odds with the substrate similarity the Gag polyprotein. The only protease mu- principle for suppression of drug resistance. tant detected was K41R, a mutation outside The phosphonate moiety fails the chemical- the peptide-binding tunnel but close to the similarity-to-substrate test and occupies a peptide’s entry and exit points. However, this volume beyond the substrate envelope, as mutation alone does not alter drug sensitivity. defined above (50). Of course, the natural The observed 15-fold loss of efficacy could be substrate is the polyprotein, which does extend mapped entirely to the Gag mutations (66). beyond the conventionally defined substrate- One unmistakable trend in current pro- binding cleft. Perhaps the solvent-anchored tease inhibitor drug discovery is toward com- moiety is mimicking these distal elements of pounds with improved resistance properties. the substrate and/or targeting regions of the Molecules with increased substrate similarity, protease structure that are essential for the both in shape and chemistry, have improved

106 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

performance in inhibiting extant drug-resistant Functionally, essential sites of viruses ap- strains and in suppressing the emergence of new pear susceptible to antibody surveillance, but ones (50, 51). Nevertheless, inhibitors, such a number of mechanisms operate to avoid de- as GS-8374 (61) and tipranavir (54), that dis- tection. In the case of the influenza antigens, play obvious departures from substrateness also the enzyme site of the neuraminidase presents score well in resistance properties. It remains a surface that is smaller than a typical footprint to be seen whether GS-8374 is additionally tar- of an antibody-binding site (2, 75). Carbohy- geting some other essential functionality of the drate masking and conformational gymnastics protease, such as its dynamic properties. The feature as protection mechanisms of HIV gp160 appearance of off-target resistance may pose an (76). Even when a functional site is exposed, a ultimate barrier to the degree to which a pro- thought experiment suggests that only a spe- tease inhibitor can truly suppress the selection cial subset of antibodies, i.e., those that see the of drug-resistant virus. site in the same way as does the natural ligand, would be truly broadly neutralizing (77). Vaccination against influenza virus requires BIOLOGICALS AS ANTIVIRALS annual matching of vaccine strains to circu- There are few examples of registered protein- lating viral strains. Despite an improved un- based antivirals, but more are expected in fu- derstanding of global circulation and evolution ture. Here the resistance-suppressing prospects of type A influenza viruses (78), mismatches of such antiviral agents are considered. still occur (79). For other viral diseases (small- pox, measles, mumps, hepatitis B), vaccination is extraordinarily successful. For HIV, the Neutralizing Antibodies large number of recorded strains suggests that Neutralizing antibodies, raised through vaccination to raise neutralizing antibody will vaccination or prior infection, are highly fail to protect against nonhomologous strains. successful antiviral agents. Of course, viral However, the characterization of a number of strain variation can give rise to resistance to broadly neutralizing antibodies for HIV (80) antibody. Following the early structural studies has raised the prospect of attempting to elicit of the structure and antigenicity of influenza just such a response via vaccination. Four such virus antigens (2, 67) and human rhinoviruses antibodies have now been well characterized: (68), a large body of data has accumulated b12 binds to the CD4-binding site on gp120, around the question of antigenic escape from 2G12 binds a cluster of oligomannose residues, neutralizing antibody. Landmark studies, using and 2F5 and 4E10 bind to membrane proximal by University of Texas - Austin on 10/05/09. For personal use only. monoclonal antibodies to select “monoclonal” sites on gp41. The International AIDS Vaccine variants (69), showed that single amino acid Initiative has established a Neutralizing Anti-

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org sequence changes sufﬁced to abolish binding body Consortium, which is pursuing the crystal of the antigen to the antibody, and subsequent structures of antibody complexes with compo- structural studies showed that the structural nents of gp160 as a basis for rational vaccine consequences of these single amino acid design (81). Already, crystal structures for these sequence changes were indeed only local (70, four broadly neutralizing antibodies have been 71). Later, it was realized that certain amino determined. acid sequence substitutions could be tolerated Antibody b12 is one of the few full-length within the antibody-antigen interface (72), immunoglobulin structures to have yielded to that two different antibodies could engage a crystallographic analysis. It has an elongated common epitope in chemically unrelated ways CDR H3 loop, which is thought to insert into (73), and that mutations in an epitope could the recessed CD4-binding site on gp120 (82). A have quite different effects on the binding of structure of the Fab in complex with a peptide two antibodies sharing that epitope (74). designed to mimic the discontinuous epitope

www.annualreviews.org • New Antivirals and Drug Resistance 107 ANRV378-BI78-05 ARI 29 April 2009 10:11

of b12 disproved the design hypothesis, indi- The recent NIAID HIV Vaccine Summit in cating how difficult is the design of immuno- March 2008 listed as two of its nine highest re- gens (83). Antibody 2G12 has a unique domain search priorities the determination of the 3D swap generating a dimeric structure with two structure of the envelope trimer and finding normal antigen-binding sites, (each comprising a way to elicit broadly neutralizing antibody one VL and one VH domain) flanking one un- (91). usual binding site at the interface of the two VH In combating a virus where strain variation domains (84). Cocrystallization with mannose- arises readily through error-prone replication, containing carbohydrates confirms that the an antibody that is capable of binding to all normal sites each bind one oligosaccharide and functional strains of the virus would need to that the unusual site binds two. Oligomannose have the very special properties of binding ex- occurs on gp160 of most HIV strains. Struc- clusively to a functional site and engaging that tural studies of antibody 2F5 complexed with site in a chemically similar way to the functional cognate peptides from the membrane proximal binding partner of the virus (77). This goal region of gp41 (85, 86) reveal that the apex of might be achieved with an appropriate poly- the CDR H3 loop of the antibody is formed clonal response to an epitope representing, say, by four hydrophobic residues, L, F, V, and I. the receptor-binding site on a virus or the viral These amino acids play little part in binding fusion machinery. the peptide, and they are exposed in a way that Currently only one monoclonal antibody is may promote association with a membrane- registered as an antiviral agent. Palivizumab tethered peptide. Nonconservative mutation of (Synagis®) is a humanized mouse monoclonal the phenylalanine residue in the apex decreases antibody approved for antiviral prophylaxis in binding to both the peptide and to gp41 by up pediatric populations susceptible to respira- to 10-fold, but severely (100-fold) compromises tory syncitial virus (RSV). Antibody-resistant the neutralizing effect of 2F5 against otherwise viruses have been selected in tissue culture and sensitive HIV strains (87). Antibody 4E10 also then tested for prophylactic efficacy in cotton has a long CDR H3 loop, which extends be- rats (92). Three variants in the RSV Fusion (F) yond the contacting interface with a 13-residue protein were selected in vitro, K272M, K272Q, helical peptide (88). Here, the apical CDR H3 and N268I. The structure of the RSV F protein residue is W,again so oriented that it could con- is known only by homology with the paramyx- tact the viral membrane when 4E10 binds to ovirus F proteins (93). Interestingly, in both the virus. Crystal structures of 4E10 Fab, bound to pre- and postfusion structures of the F pro- three related designed peptides, have led to a tein (94–96), these amino acids are located in by University of Texas - Austin on 10/05/09. For personal use only. better definition of the epitope and of ways of the surface between two helical segments. In- stabilizing its structure for incorporation into a fection of animals with N268I virus was still

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org candidate immunogen (89). Collectively, these blocked by palivizumab, but with K272Q, the structural data provide a good launching pad for virus was not. The F protein mutant K272Q designing structures that might elicit broadly was also selected in vivo (97), and in tissue cul- neutralizing antibodies. ture, the K272M virus grows less well than the However, the critically important structure N268I virus (98). An afﬁnity-matured form of of the natural immunogen, gp160, still remains palivizumab is in development. It is not known unknown. The Neutralizing Antibody Consor- whether clinical failures of palivizumab prohy- tium’s strategy is to extend the panel of known laxis in premature infants (99) were the result broadly neutralizing antibodies and character- of infection by antibody-resistant RSV strains. ize their epitopes, determine the 3D structure There is considerable interest in the use of functional HIV glycoproteins with and with- of monoclonal antibody therapy for hepatitis out these antibodies bound, and use this infor- C virus, especially in liver transplant patients mation to rationally design immunogens (90). (99a). These developments stand to beneﬁt

108 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

from experiences in other settings of viral with an identical sequence on the virus. Such a infection. drug seems to meet the requirement of closely resembling the natural substrate, and yet the barrier to emergence of enfuvirtide-resistant Inhibitors of Viral Membrane Fusion virus in the clinic is low (107). The earliest Current understanding of the molecular basis studies of resistance to enfuvirtide in tissue cul- for the fusion of a viral membrane with a tar- ture (108) revealed drug-resistant mutations in get cell membrane stems from the description HR1, in the peptide sequence G36-I37-V38, of the structure of the inﬂuenza virus hemag- but no compensating mutations were observed glutinin in its pre- and postfusion conforma- in HR2, despite the fact that G36 and V38 face tions (100). In one other system, the paramyx- residues K144 and N145 in the outer HR2 helix oviruses, comparable information is available (106). Other studies suggest that compensating on the structures of the pre- and postfusion mutations might occur elsewhere in the enve- forms of the fusion protein (96, 101). For HIV, lope protein (109). Resistant variants selected a structure for gp160 remains elusive, but nu- in the clinic map to this same region (36–45) of merous studies of the postfusion conformation HR1 (107). Common resistance mutations are of gp41 have informed the exploitation of the G36 to D, S, E, and V and V38 to A, G, E, fusion machinery as a drug target. Early studies and M. Loss of binding of enfuvirtide to corre- on the antiviral properties of peptides, spanning sponding mutants displayed in the HR1 region putative helical regions of HIV gp41(102–104), of gp41 correlates with loss of antiviral activity have their basis in the observation that mem- (110), supporting the proposed mode of action brane fusion follows the formation of a struc- (Figure 6c). ture often referred to as a six-helix bundle (105, However, other studies show that enfuvir- 106). (Figure 6a). This helical bundle is be- tide only weakly inhibits formation of the six- lieved to be formed via a pathway in which one helix bundle, at least when compared with pep- intermediate is a trimeric coiled-coil structure tides representing different segments of HR2 comprising helical region 1 (or HR1), which (111), suggesting that the mode of action of en- can associate with the target membrane via a fuvirtide may be more complex than previously “fusion peptide” at its N-terminal end. A sec- thought. The structure of the HR1 trimeric ond helical region (HR2) from the C terminus coiled coil presents a prominent pocket-shaped of the fusion protein engages with the trimeric feature into which amino acids W117, W120, coiled coil in an antiparallel complex such that D121, and I124 from HR2 are inserted when its C terminus, inserted in the viral membrane, the helical bundle is formed (Figure 6c). Such by University of Texas - Austin on 10/05/09. For personal use only. is at the same end of the helical bundle as the a cavity is an enticing target for drug discovery, target membrane. Thus, the putative interme- and several attempts are underway to exploit

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org diate structure, the trimeric coiled coil of HR1 it, including the use of D-peptides (112, 113). peptides, has been viewed as a potential drug A discussion of organic antagonists of fusion target for disruption of the fusion-competent is beyond the scope of this article. The cav- conformer. ity on the HR1 trimer is toward its C-terminal Enfuvirtide (T-20) is the ﬁrst drug to vali- end (near G61), and the residues of HR2 that date this approach. It is an oligopeptide, com- insert into it are not present in enfuvirtide prising residues 127–162 of the HIV fusion (Figure 6b). Considering the HR1 coiled coil protein gp41 (Figure 6b). This sequence over- as the drug target, at its N-terminal end are laps with HR2 (residues 117–154) as deﬁned the mutations conferring resistance to enfu- in structural studies (106), and its mode of ac- virtide, and at its C-teminal end is the cavity tion is generally believed to be prevention of the (Figure 6c). A study of the antiviral proper- formation of the helical bundle by competition ties of a number of overlapping HR2 peptides

www.annualreviews.org • New Antivirals and Drug Resistance 109 ANRV378-BI78-05 ARI 29 April 2009 10:11

a HR2

HR1 HR1

HR1 HR2

HR2 b

77 30 HR1 KLYREVALIRAQLQKIGWVTLQLLHQQAEIARLLNNQQQVIGSLLQRA — fusion-peptide

117 154 HR2 WMEWDREINNYTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWFNISNWLWYIR Links HR2 to membrane anchor

Enfuvirtide T-20 YTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWF T-1249 WQEWEQKITALLEQAQIQQEKNEYELQKLDKWASLWEWF Literature T-2635 TTWEAWDRAIAEYAARIEALIRAAQEQQEKNEAALREL peptides SFT WIEWEREISNYTNQIYEILTESQNQQDRNEKDLLE

G36 mutation site

V38 mutation site by University of Texas - Austin on 10/05/09. For personal use only.

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org Figure 6 (a) The postfusion state of the assembly of helical regions one and two (HR1 and HR2) in gp41 (PDB code 1ENV). The trimeric nature of the assembly is illustrated by its coloring ( green, cyan, and magenta), and the ribbon drawing of the magenta subunit is overlayed with the atomic structure in stick format. The N-terminal end of HR1 and the C-terminal end of HR2 are to the right as in panel b.(b) The ﬁrst and second lines are the amino acid sequences of HR1 and HR2, with their approximate structural alignment from PDB code 1ENV. (c) The six-helix bundle of gp41 with surfaces colored as in (a) and with HR2 of the magenta subunit shown as a coil and highlighting ( yellow surface) the enfuvirtide-resistant mutation sites G36 and V38. Also shown in stick format is the pocket-binding motif at the N-terminal end of HR2 (PDB code 1ENV).

including 127–162 (enfuvirtide), 122–157, and HR2 helix has distinct domains of interaction 117–152 has recently shown that the middle se- with HR1, including with the cavity at one end quence is some 100-fold less potent than either and with lipids or the fusion peptide at the other of the other two (111). This suggests that the end.

110 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

T-1249 is a second generation peptide, again target with peptides related to HR1 have been derived from amino acid sequences comprising frustrated by poor solubility of the relevant se- and adjacent to HR2 (114). The sequence is a quences. In isolation, HR1 is expected to exist chimera derived from HIV-1, HIV-2, and SIV as a trimeric coiled coil. Commencing with an and displays a pocket-binding motif (see above) HR1 sequence including residues 29 through at its N-terminal end, although the location of 79 (T-865), an engineered peptide, which in- this motif is displaced by one heptad from its po- cludes the core “a” and “d” residues of the hep- sition in the gp41 sequence (Figure 6b). Drug- tad from HR1 of RSV, has improved stability resistant virus has been selected both in vitro and antiviral activity (118). Passaging virus in and in vivo, displaying HR1 mutants (residues the presence of a related HR1 trimeric sequence 37 and 38) that are cross-resistant to T-20 selected variants with mutations in HR2, sup- (115). A subsequent study in patients already porting the model for the mode of action of treated with enfuvirtide generated mutants in the HR1 peptides. Once again, it will be inter- both HR1 and HR2 (116). Clinical develop- esting to see whether the use of nonnative se- ment of T-1249 has been discontinued (117). quences in HR1 lowers the barrier for drug re- A third-generation peptide, T-2635 sistance to emerge. A novel way to target HR2 (Figure 6b), has also been reported with a and to overcome some of the solubility issues span similar to the HR2 helix but containing with HR1 trimers is with the peptide known engineered sequences to enhance helicity and as 5-helix (121). This construct contains the se- stability (118). Curiously, V38 mutants selected quence HR1-HR2-HR1-HR2-HR1 connected with enfuvirtide or T-1249 remain susceptible with linkers and assembles into a 5-helix bun- (some even more susceptible than wild type) dle (122), presenting a binding site for one HR2 to T-2635 (119), even though the sequences helix (Figure 6c). 5-helix is inhibitory of a num- on the peptides adjacent to V38 are essentially ber of HIV variants at nanomolar levels (121). identical. The structural basis for these ob- Kinetic studies suggest that, during viral infec- servations is unknown. Perhaps the modes of tion, the HR2 helix is accessible for only a few interaction of enfuvirtide, T-1249, and T-2635 seconds, underscoring the importance of the as- with HR1 are not identical. Perhaps the sociation kinetics of any agent that seeks to cap- additional interaction energy deriving from the ture it (123). The resistance profile of 5-helix is pocket-binding motif at the N-terminal end of unknown. T-2635 allows efficacious binding even in the The early structural insights into viral mem- face of V38 mutations. It will be interesting to brane fusion are the foundation for the fusion- see whether the nonnative sequences in T-2635 inhibitor class of HIV drugs. The pattern of by University of Texas - Austin on 10/05/09. For personal use only. give rise to a novel panel of resistant viruses. drug-resistant mutations generally fits a model Sifuvirtide, another third-generation peptide with the mode of action through disruption of

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org (Figure 6b), is similar in span and sequence formation of the fusion-competent conformer to T-2635 and contains engineered elements of gp41. Nevertheless, a molecular interpreta- for stability, potency, and pharmacokinetics tion of the enfuvirtide-selected resistance mu- (120). It lacks the C-terminal decapeptide of tations still lacks a sound structural basis. No enfuvirtide, which may explain its failure to examples of structures of drug bound to its interact with lipid membranes. Like T-2635, target, nor to a drug-resistant target, are cur- it prevents replication of enfuvirtide-resistant rently available. That aside, we remain con- HIV strains, and it was safe and well-tolerated fronted with the fact that the native sequence in a Phase 1a study (120). Its proﬁle in sup- from HR2 and the adjacent membrane proxi- pressing the emergence of drug-resistant virus mal region represented in the drug select for is yet unknown. resistance. Features of enfuvirtide that are de- A complementary drug target is HR2 it- partures from the “natural ligand” include that self, but until recently, attempts to exploit this it is monomeric, is not covalently associated

www.annualreviews.org • New Antivirals and Drug Resistance 111 ANRV378-BI78-05 ARI 29 April 2009 10:11

with HR1, and is not representative of the entire second-order preference for the target might binding surface to the HR1 coiled coil. Future be that its function should be directed at host analysis of the second- and third-generation molecules and not at virus-encoded molecules, HR2-based peptides and of the emerging op- which are themselves subject to selection. portunities with HR1-based peptides will fur- Minimizing the chemical differences between ther inform approaches to raising the resistance the target’s natural ligands and the drug will barrier to this important drug class. raise the barrier to the emergence of mutated targets that can distinguish between the two. SUMMARY Multicomponent therapy will continue as Target sites for antivirals should be func- a prominent strategy for suppressing the tional sites. The failure of antibody-mediated emergence of drug-resistant variants, but in- immunity to combat viral strain variation is creased attention will be paid to optimizing usually a failure of the antibody to target the resistance-suppressing characteristics of the a functional site. Notwithstanding the enor- individual components. The principles dis- mous impact that protease inhibitors have made cussed are clearly not restricted to antiviral drug as part of the anti-HIV armamentarium, a discovery.

SUMMARY POINTS 1. The barrier to emergent resistance is a property of the drug and its target. 2. Very subtle chemical changes in the drug can radically alter that barrier. 3. The drug-binding site is best conﬁned to functionally essential elements. 4. Stereochemical similarity between the drug and natural ligands is desirable.

FUTURE ISSUES 1. Further review is needed of extant resistance data for generic trends to inform new drug discovery. 2. Monitoring the performance of new drugs with purportedly higher barriers to resistance by University of Texas - Austin on 10/05/09. For personal use only. should continue.

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org DISCLOSURE STATEMENT The author owns shares in Biota Holdings Ltd., which receives royalties from GlaxoSmithKline on sales of Relenza.

ACKNOWLEDGMENTS I thank Brian Smith for helpful discussions and the NHMRC (Australia) for support.

LITERATURE CITED 1. Colman PM. 1994. Inﬂuenza virus neuraminidase: structure, antibodies, and inhibitors. Protein Sci. 3:1687–96

112 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

2. Colman PM, Varghese JN, Laver WG. 1983. Structure of the catalytic and antigenic sites in influenza virus neuraminidase. Nature 303:41–44 3. Varghese JN, Laver WG, Colman PM. 1983. Structure of the influenza virus glycoprotein antigen neuraminidase at 2.9 A˚ resolution. Nature 303:35–40 4. Baker AT, Varghese JN, Laver WG, Air GM, Colman PM. 1987. Three-dimensional structure of neuraminidase of subtype N9 from an avian influenza virus. Proteins 2:111–17 5. Burmeister WP, Ruigrok RW, Cusack S. 1992. The 2.2 A˚ resolution crystal structure of influenza B neuraminidase and its complex with sialic acid. EMBO J. 11:49–56 6. Varghese JN, McKimm-Breschkin JL, Caldwell JB, Kortt AA, Colman PM. 1992. The structure of the complex between influenza virus neuraminidase and sialic acid, the viral receptor. Proteins 14:327–32 7. von Itzstein M, Wu WY, Kok GB, Pegg MS, Dyason JC, et al. 1993. Rational design of potent sialidase- based inhibitors of influenza virus replication. Nature 363:418–23 8. Varghese JN, Epa VC, Colman PM. 1995. Three-dimensional structure of the complex of 4-guanidino- Neu5Ac2en and influenza virus neuraminidase. Protein Sci. 4:1081–87 9. Hayden FG, Treanor JJ, Betts RF, Lobo M, Esinhart JD, Hussey EK. 1996. Safety and efficacy of the neuraminidase inhibitor GG167 in experimental human influenza. JAMA 275:295–99 10. Sollis S, Howes P, Smith P, Cherry P, Bethell R. 1996. Novel inhibitors of influenza sialidase related to GG167. Synthesis of 4-amino and guanidino-4H-pyran-2-carboxylic acid-6-propylamides; selective inhibitors of influenza A sialidase. Bioorg. Med. Chem. Lett. 6:2931–36 11. Smith PW, Whittington AR, Sollis SL, Howes PD, Taylor NR. 1997. Novel inhibitors of influenza sialidases related to zanamivir. Heterocyclic replacements of the glycerol sidechain. Bioorg. Med. Chem. Lett. 7:2239–42 12. Taylor NR, Cleasby A, Singh O, Skarzynski T, Wonacott AJ, et al. 1998. Dihydropyrancarboxamides related to zanamivir: a new series of inhibitors of influenza virus sialidases. 2. Crystallographic and molecular modeling study of complexes of 4-amino-4H-pyran-6-carboxamides and sialidase from influenza virus types A and B. J. Med. Chem. 41:798–807 13. Kim CU, Lew W, Williams MA, Liu H, Zhang L, et al. 1997. Influenza neuraminidase inhibitors possessing a novel hydrophobic interaction in the enzyme active site: design, synthesis, and structural analysis of carbocyclic sialic acid analogues with potent anti-influenza activity. J. Am. Chem. Soc. 119:681– 90 14. Colman P. 2006. Anti-influenza drugs from neuraminidase inhibitors. In Structure-Based Drug Discovery: An Overview, ed. RE Hubbard, pp. 193–218. Cambridge, UK: RSC Publ. 15. McKimm-Breschkin JL, Blick TJ, Sahasrabudhe A, Tiong T, Marshall D, et al. 1996. Generation and characterization of variants of NWS/G70C influenza virus after in vitro passage in 4-amino-Neu5Ac2en and 4-guanidino-Neu5Ac2en. Antimicrob. Agents Chemother. 40:40–46

by University of Texas - Austin on 10/05/09. For personal use only. 16. Blick TJ, Tiong T, Sahasrabudhe A, Varghese JN, Colman PM, et al. 1995. Generation and characterization of an inﬂuenza virus neuraminidase variant with decreased sensitivity to the neuraminidase-speciﬁc inhibitor 4-guanidino-Neu5Ac2en. Virology 214:475–84

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org 17. McKimm-Breschkin JL, McDonald M, Blick TJ, Colman PM. 1996. Mutation in the influenza virus neuraminidase gene resulting in decreased sensitivity to the neuraminidase inhibitor 4-guanidino- Neu5Ac2en leads to instability of the enzyme. Virology 225:240–42 18. McKimm-Breschkin JL. 2000. Resistance of influenza viruses to neuraminidase inhibitors–a review. Antivir. Res. 47:1–17 19. McKimm-Breschkin JL, Sahasrabudhe A, Blick TJ, McDonald M, Colman PM, et al. 1998. Mutations in a conserved residue in the influenza virus neuraminidase active site decreases sensitivity to Neu5Ac2en- derived inhibitors. J. Virol. 72:2456–62 20. Varghese JN, Smith PW, Sollis SL, Blick TJ, Sahasrabudhe A, et al. 1998. Drug design against a shifting target: a structural basis for resistance to inhibitors in a variant of influenza virus neuraminidase. Structure 6:735–46 21. Tai CY, Escarpe PA, Sidwell RW, Williams MA, Lew W, et al. 1998. Characterization of human influenza virus variants selected in vitro in the presence of the neuraminidase inhibitor GS 4071. Antimicrob. Agents Chemother. 42:3234–41

www.annualreviews.org • New Antivirals and Drug Resistance 113 ANRV378-BI78-05 ARI 29 April 2009 10:11

21a. Cass LM, Efthymiopoulos C, Bye A. 1999. Pharmacokinetics of zanamivir after intravenous, oral inhaled or intranasal administration to healthy volunteers. Clin. Pharmacokinet. 36(Suppl. 1):1–11 21b. He G, Massarella J, Ward P. 1999. Clinical pharmacokinetics of the prodrug oseltamivir and its active metabolite Ro 64-0802. Clin. Pharmacokinet. 37:471–84 22. Barnett JM, Cadman A, Gor D, Dempsey M, Walters M, et al. 2000. Zanamivir susceptibility monitoring and characterization of influenza virus clinical isolates obtained during phase II clinical efficacy studies. Antimicrob. Agents Chemother. 44:78–87 23. Jackson HC, Roberts N, Wang ZM, Belshe R. 2000. Management of influenza: use of new antivirals and resistance in perspective. Clin. Drug Investig. 20:447–54 24. Kiso M, Mitamura K, Sakai-Tagawa Y, Shiraishi K, Kawakami C, et al. 2004. Resistant influenza A viruses in children treated with oseltamivir: descriptive study. Lancet 364:759–65 25. Bright RA, Medina MJ, Xu X, Perez-Oronoz G, Wallis TR, et al. 2005. Incidence of adamantane resistance among influenza A (H3N2) viruses isolated worldwide from 1994 to 2005: a cause for concern. Lancet 366:1175–81 26. Bright RA, Shay DK, Shu B, Cox NJ, Klimov AI. 2006. Adamantane resistance among influenza A viruses isolated early during the 2005–2006 influenza season in the United States. JAMA 295:891–94 27. Gubareva LV, Webster RG, Hayden FG. 2001. Comparison of the activities of zanamivir, oseltamivir, and RWJ-270201 against clinical isolates of influenza virus and neuraminidase inhibitor-resistant variants. Antimicrob. Agents Chemother. 45:3403–8 28. Herlocher ML, Truscon R, Elias S, Yen HL, Roberts NA, et al. 2004. Influenza viruses resistant to the antiviral drug oseltamivir: transmission studies in ferrets. J. Infect. Dis. 190:1627–30 29. Lackenby A, Hungnes O, Dudman SG, Meijer A, Paget WJ, et al. 2008. Emergence of resistance to oseltamivir among influenza A(H1N1) viruses in Europe. Eurosurveillance 13. Online http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=8026 30. Collins PJ, Haire LF, Lin YP, Liu J, Russell RJ, et al. 2008. Crystal structures of oseltamivir-resistant influenza virus neuraminidase mutants. Nature 453:1258–61 31. Hatakeyama S, Sugaya N, Ito M, Yamazaki M, Ichikawa M, et al. 2007. Emergence of influenza B viruses with reduced sensitivity to neuraminidase inhibitors. JAMA 297:1435–42 32. Moscona A, McKimm-Breschkin J. 2007. News about influenza B drug resistance that cannot be ignored. JAMA 297:1492–93 33. McKimm-Breschkin JL, Selleck PW, Usman TB, Johnson MA. 2007. Reduced sensitivity of influenza A (H5N1) to oseltamivir. Emerg. Infect. Dis. 13:1354–57 34. Russell RJ, Haire LF, Stevens DJ, Collins PJ, Lin YP, et al. 2006. The structure of H5N1 avian influenza neuraminidase suggests new opportunities for drug design. Nature 443:45–49 35. Lawrence MC, Borg NA, Streltsov VA, Pilling PA, Epa VC, et al. 2004. Structure of the haemagglutinin-

by University of Texas - Austin on 10/05/09. For personal use only. neuraminidase from human parainﬂuenza virus type III. J. Mol. Biol. 335:1343–57 36. Wlodawer A. 2002. Rational approach to AIDS drug design through structural biology. Annu. Rev. Med. 53:595–614

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org 37. De Clercq E. 2007. The design of drugs for HIV and HCV. Nat. Rev. Drug Discov. 6:1001–18 38. Navia MA, Fitzgerald PM, McKeever BM, Leu CT, Heimbach JC, et al. 1989. Three-dimensional structure of aspartyl protease from human immunodeficiency virus HIV-1. Nature 337:615–20 39. Wlodawer A, Miller M, Jaskolski M, Sathyanarayana BK, Baldwin E, et al. 1989. Conserved folding in retroviral proteases: crystal structure of a synthetic HIV-1 protease. Science 245:616–21 40. Lapatto R, Blundell T, Hemmings A, Overington J, Wilderspin A, et al. 1989. X-ray analysis of HIV-1 proteinase at 2.7 A˚ resolution confirms structural homology among retroviral enzymes. Nature 342:299– 302 41. Wlodawer A, Erickson JW. 1993. Structure-based inhibitors of HIV-1 protease. Annu. Rev. Biochem. 62:543–85 42. Prabu-Jeyabalan M, Nalivaika E, Schiffer CA. 2002. Substrate shape determines specificity of recognition for HIV-1 protease: analysis of crystal structures of six substrate complexes. Structure 10:369–81 43. Prabu-Jeyabalan M, Nalivaika E, Schiffer CA. 2000. How does a symmetric dimer recognize an asymmetric substrate? A substrate complex of HIV-1 protease. J. Mol. Biol. 301:1207–20

114 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

44. Heaslet H, Rosenfeld R, Giffin M, Lin YC, Tam K, et al. 2007. Conformational flexibility in the flap domains of ligand-free HIV protease. Acta Crystallogr. D Biol. Crystallogr. 63:866–75 45. Kim EE, Baker CT, Dwyer MD, Murcko MA, Rao BG, et al. 1995. Crystal structure of HIV-1 protease in complex with Vx-478, a potent and orally bioavailable inhibitor of the enzyme. J. Am. Chem. Soc. 117:1181–82 46. Schock HB, Garsky VM, Kuo LC. 1996. Mutational anatomy of an HIV-1 protease variant conferring cross-resistance to protease inhibitors in clinical trials. Compensatory modulations of binding and activity. J. Biol. Chem. 271:31957–63 46a. Piana S, Carloni P, Rothlisberger U. 2002. Drug resistance in HIV-1 protease: flexibility-assisted mechanism of compensatory mutations. Protein Sci. 11:2393–2402 46b. Ode H, Neya S, Hata M, Sugiura W, Hoshino T. 2006. Computational simulations of HIV-1 proteases multidrug resistance due to nonactive site mutation L90M. J. Am. Chem. Soc. 128:7887–94 46c. Kozisek M, Bray J, Rezacova P, Saskova K, Brynda J, et al. 2007. Molecular analysis of the HIV-1 resistance development: enzymatic activities, crystal structures, and thermodynamics of nelfinavir-resistant HIV protease mutants. J. Mol. Biol. 374:1005–16 47. Shafer RW, Rhee SY, Pillay D, Miller V, Sandstrom P, et al. 2007. HIV-1 protease and reverse tran- scriptase mutations for drug resistance surveillance. AIDS 21:215–23 48. Shafer RW, Schapiro JM. 2008. HIV-1 drug resistance mutations: an updated framework for the second decade of HAART. AIDS Rev. 10:67–84 49. Prabu-Jeyabalan M, Nalivaika EA, King NM, Schiffer CA. 2003. Viability of a drug-resistant human immunodeficiency virus type 1 protease variant: structural insights for better antiviral therapy. J. Virol. 77:1306–15 50. Chellappan S, Kairys V, Fernandes MX, Schiffer C, Gilson MK. 2007. Evaluation of the substrate envelope hypothesis for inhibitors of HIV-1 protease. Proteins 68:561–67 51. Colman PM, Smith BJ. 2003. Specificity and promiscuity in protein-ligand and protein-protein interactions. Aust. J. Chem. 56:763–67 52. Thaisrivongs S, Skulnick HI, TurnerSR, Strohbach JW, TommasiRA, et al. 1996. Structure-based design of HIV protease inhibitors: sulfonamide-containing 5,6-dihydro-4-hydroxy-2-pyrones as nonpeptidic inhibitors. J. Med. Chem. 39:4349–53 53. Doyon L, Tremblay S, Bourgon L, Wardrop E, Cordingley MG. 2005. Selection and characterization of HIV-1 showing reduced susceptibility to the nonpeptidic protease inhibitor tipranavir. Antivir. Res. 68:27–35 54. Muzammil S, Armstrong AA, Kang LW, Jakalian A, Bonneau PR, et al. 2007. Unique thermodynamic response of tipranavir to human immunodeficiency virus type 1 protease drug resistance mutations. J. Virol. 81:5144–54

by University of Texas - Austin on 10/05/09. For personal use only. 55. King NM, Prabu-Jeyabalan M, Nalivaika EA, Wigerinck P, de Bethune MP, Schiffer CA. 2004. Structural and thermodynamic basis for the binding of TMC114, a next-generation human immunodeﬁciency virus type 1 protease inhibitor. J. Virol. 78:12012–21

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org 56. Surleraux DL, Tahri A, Verschueren WG, Pille GM, de Kock HA, et al. 2005. Discovery and selection of TMC114, a next generation HIV-1 protease inhibitor. J. Med. Chem. 48:1813–22 57. Ghosh AK, Chapsal BD, Weber IT, Mitsuya H. 2008. Design of HIV protease inhibitors targeting protein backbone: an effective strategy for combating drug resistance. Acc. Chem. Res. 41:78–86 58. Liu F, Kovalevsky AY, Tie Y, Ghosh AK, Harrison RW, Weber IT. 2008. Effect of flap mutations on structure of HIV-1 protease and inhibition by saquinavir and darunavir. J. Mol. Biol. 381:102–15 59. Miller JF, Andrews CW, Brieger M, Furfine ES, Hale MR, et al. 2006. Ultra-potent P1 modified arylsulfonamide HIV protease inhibitors: the discovery of GW0385. Bioorg. Med. Chem. Lett. 16:1788– 94 60. YatesPJ, Hazen R, St Clair M, Boone L, Tisdale M, Elston RC. 2006. In vitro development of resistance to human immunodeficiency virus protease inhibitor GW640385. Antimicrob. Agents Chemother. 50:1092– 95 61. Cihlar T, He GX, Liu X, Chen JM, Hatada M, et al. 2006. Suppression of HIV-1 protease inhibitor resistance by phosphonate-mediated solvent anchoring. J. Mol. Biol. 363:635–47

www.annualreviews.org • New Antivirals and Drug Resistance 115 ANRV378-BI78-05 ARI 29 April 2009 10:11

62. Callebaut C, Stray KM, TsaiL, Xu L, He G-X, et al. 2007. GS-8374, a novel phosphonate HIV protease inhibitor with potent in vitro antiretroviral activity, low metabolic toxicity, and favorable resistance profile. Antivir. Res. 74:A27 63. Maguire MF, Guinea R, Griffin P, Macmanus S, Elston RC, et al. 2002. Changes in human immunodeficiency virus type 1 Gag at positions L449 and P453 are linked to I50V protease mutants in vivo and cause reduction of sensitivity to amprenavir and improved viral fitness in vitro. J. Virol. 76:7398–406 64. De Meyer S, Azijn H, Surleraux D, Jochmans D, Tahri A, et al. 2005. TMC114, a novel human immunodeficiency virus type 1 protease inhibitor active against protease inhibitor-resistant viruses, including a broad range of clinical isolates. Antimicrob. Agents Chemother. 49:2314–21 65. Nijhuis M, van Maarseveen NM, Lastere S, Schipper P, Coakley E, et al. 2007. A novel substrate-based HIV-1 protease inhibitor drug resistance mechanism. PLoS Med. 4:e36 66. Callebaut C, Stray KM, TsaiL, Xu L, Lee W, Cihlar T. 2007. In vitro HIV-1 resistance selection to GS- 8374, a novel phosphonate protease inhibitor: comparison with Lopinavir, Atazanavir and Darunavir. Antivir. Ther. 12:S18 67. Wiley DC, Wilson IA, Skehel JJ. 1981. Structural identification of the antibody-binding sites of Hong Kong influenza haemagglutinin and their involvement in antigenic variation. Nature 289:373–78 68. Rossmann MG, Arnold E, Erickson JW, Frankenberger EA, Griffith JP, et al. 1985. Structure of a human common cold virus and functional relationship to other picornaviruses. Nature 317:145–53 69. Gerhard W, Yewdell J, Frankel ME, Webster R. 1981. Antigenic structure of influenza virus haemagglutinin defined by hybridoma antibodies. Nature 290:713–17 70. Knossow M, Daniels RS, Douglas AR, Skehel JJ, Wiley DC. 1984. Three-dimensional structure of an antigenic mutant of the influenza virus haemagglutinin. Nature 311:678–80 71. Varghese JN, Webster RG, Laver WG, Colman PM. 1988. Structure of an escape mutant of glycoprotein N2 neuraminidase of influenza virus A/Tokyo/3/67 at 3 A. J. Mol. Biol. 200:201–3 72. Tulip WR, Varghese JN, Webster RG, Laver WG, Colman PM. 1992. Crystal structures of two mutant neuraminidase-antibody complexes with amino acid substitutions in the interface. J. Mol. Biol. 227:149– 59 73. Malby RL, Tulip WR, Harley VR, McKimm-Breschkin JL, Laver WG, et al. 1994. The structure of a complex between the NC10 antibody and influenza virus neuraminidase and comparison with the overlapping binding site of the NC41 antibody. Structure 2:733–46 74. Dall’Acqua W, Goldman ER, Eisenstein E, Mariuzza RA. 1996. A mutational analysis of the binding of two different proteins to the same antibody. Biochemistry 35:9667–76 75. Colman PM, Laver WG, Varghese JN, Baker AT, Tulloch PA, et al. 1987. Three-dimensional structure of a complex of antibody with influenza virus neuraminidase. Nature 326:358–63 76. Kwong PD. 2005. Human immunodeficiency virus: refolding the envelope. Nature 433:815–16 by University of Texas - Austin on 10/05/09. For personal use only. 77. Colman PM. 1997. Virus versus antibody. Structure 5:591–93 78. Russell CA, Jones TC, Barr IG, Cox NJ, Garten RJ, et al. 2008. The global circulation of seasonal influenza A (H3N2) viruses. 320:340–46

Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org Science 79. Salzberg S. 2008. The contents of the syringe. Nature 454:160–61 80. Binley JM, Wrin T, Korber B, Zwick MB, Wang M, et al. 2004. Comprehensive cross-clade neutralization analysis of a panel of antihuman immunodeﬁciency virus type 1 monoclonal antibodies. J. Virol. 78:13232–52 81. Burton DR, Desrosiers RC, Doms RW, Koff WC, Kwong PD, et al. 2004. HIV vaccine design and the neutralizing antibody problem. Nat. Immunol. 5:233–36 82. Saphire EO, Parren PW, Pantophlet R, Zwick MB, Morris GM, et al. 2001. Crystal structure of a neutralizing human IGG against HIV-1: a template for vaccine design. Science 293:1155–59 83. Saphire EO, Montero M, Menendez A, van Houten NE, Irving MB, et al. 2007. Structure of a high- afﬁnity “mimotope” peptide bound to HIV-1-neutralizing antibody b12 explains its inability to elicit gp120 cross-reactive antibodies. J. Mol. Biol. 369:696–709 84. Calarese DA, Scanlan CN, Zwick MB, Deechongkit S, Mimura Y, et al. 2003. Antibody domain exchange is an immunological solution to carbohydrate cluster recognition. Science 300:2065–71

116 Colman ANRV378-BI78-05 ARI 29 April 2009 10:11

85. Ofek G, Tang M, Sambor A, Katinger H, Mascola JR, et al. 2004. Structure and mechanistic analysis of the antihuman immunodeficiency virus type 1 antibody 2F5 in complex with its gp41 epitope. J. Virol. 78:10724–37 86. Pai EF, Klein MH, Chong P, Pedyczak A. 2000. Int. Patent No. WO-00/61618 87. Zwick MB, Komori HK, Stanfield RL, Church S, Wang M, et al. 2004. The long third complementarity- determining region of the heavy chain is important in the activity of the broadly neutralizing antihuman immunodeficiency virus type 1 antibody 2F5. J. Virol. 78:3155–61 88. Cardoso RM, Zwick MB, Stanfield RL, Kunert R, Binley JM, et al. 2005. Broadly neutralizing anti-HIV antibody 4E10 recognizes a helical conformation of a highly conserved fusion-associated motif in gp41. Immunity 22:163–73 89. Cardoso RM, Brunel FM, Ferguson S, Zwick M, Burton DR, et al. 2007. Structural basis of enhanced binding of extended and helically constrained peptide epitopes of the broadly neutralizing HIV-1 antibody 4E10. J. Mol. Biol. 365:1533–44 90. Walker BD, Burton DR. 2008. Toward an AIDS vaccine. Science 320:760–64 91. Fauci AS, Johnston MI, Dieffenbach CW, Burton DR, Hammer SM, et al. 2008. HIV vaccine research: the way forward. Science 321:530–32 92. Zhao X, Chen FP, Megaw AG, Sullender WM. 2004. Variable resistance to palivizumab in cotton rats by respiratory syncytial virus mutants. J. Infect. Dis. 190:1941–46 93. Smith BJ, Lawrence MC, Colman PM. 2002. Modelling the structure of the fusion protein from human respiratory syncytial virus. Protein Eng. 15:365–71 94. Chen L, Gorman JJ, McKimm-Breschkin J, Lawrence LJ, Tulloch PA, et al. 2001. The structure of the fusion glycoprotein of Newcastle disease virus suggests a novel paradigm for the molecular mechanism of membrane fusion. Structure 9:255–66 95. Yin HS, Paterson RG, Wen X, Lamb RA, Jardetzky TS. 2005. Structure of the uncleaved ectodomain of the paramyxovirus (hPIV3) fusion protein. Proc. Natl. Acad. Sci. USA 102:9288–93 96. Yin HS, Wen X, Paterson RG, Lamb RA, Jardetzky TS. 2006. Structure of the parainfluenza virus 5 F protein in its metastable, prefusion conformation. Nature 439:38–44 97. Zhao X, Sullender WM. 2005. In vivo selection of respiratory syncytial viruses resistant to palivizumab. J. Virol. 79:3962–68 98. Zhao X, Liu E, Chen FP, Sullender WM. 2006. In vitro and in vivo fitness of respiratory syncytial virus monoclonal antibody escape mutants. J. Virol. 80:11651–57 99. Simon A, Ammann RA, Wilkesmann A, Eis-Hubinger AM, Schildgen O, et al. 2007. Respiratory syncytial virus infection in 406 hospitalized premature infants: results from a prospective German multicentre database. Eur. J. Pediatr. 166:1273–83 99a. Keck ZY, Machida K, Lai MM, Ball JK, Patel AH, Foung SK. 2008. Therapeutic control of hepatitis C virus: the role of neutralizing antibodies. Curr. Top. Microbiol. Immunol. 317:1–38

by University of Texas - Austin on 10/05/09. For personal use only. 100. Skehel JJ, Wiley DC. 2000. Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin. Annu. Rev. Biochem. 69:531–69 101. Lamb RA, Jardetzky TS. 2007. Structural basis of viral invasion: lessons from paramyxovirus F. Curr. Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org Opin. Struct. Biol. 17:427–36 102. Wild C, Oas T, McDanal C, Bolognesi D, Matthews T. 1992. A synthetic peptide inhibitor of human immunodeficiency virus replication: correlation between solution structure and viral inhibition. Proc. Natl. Acad. Sci. USA 89:10537–41 103. Wild CT, Shugars DC, Greenwell TK, McDanal CB, Matthews TJ. 1994. Peptides corresponding to a predictive α-helical domain of human immunodeficiency virus type 1 gp41 are potent inhibitors of virus infection. Proc. Natl. Acad. Sci. USA 91:9770–74 104. Jiang S, Lin K, Strick N, Neurath AR. 1993. HIV-1 inhibition by a peptide. Nature 365:113 105. Chan DC, Fass D, Berger JM, Kim PS. 1997. Core structure of gp41 from the HIV envelope glycoprotein. Cell 89:263–73 106. Weissenhorn W, Dessen A, Harrison SC, Skehel JJ, Wiley DC. 1997. Atomic structure of the ectodomain from HIV-1 gp41. Nature 387:426–30 107. Lu J, Deeks SG, Hoh R, Beatty G, Kuritzkes BA, et al. 2006. Rapid emergence of enfuvirtide resistance in HIV-1-infected patients: results of a clonal analysis. J. Acquir. Immune Defic. Syndr. 43:60–64

www.annualreviews.org • New Antivirals and Drug Resistance 117 ANRV378-BI78-05 ARI 29 April 2009 10:11

108. Rimsky LT, Shugars DC, Matthews TJ. 1998. Determinants of human immunodeficiency virus type 1 resistance to gp41-derived inhibitory peptides. J. Virol. 72:986–93 109. Labrosse B, Morand-Joubert L, Goubard A, Rochas S, Labernardiere JL, et al. 2006. Role of the envelope genetic context in the development of enfuvirtide resistance in human immunodeficiency virus type 1- infected patients. J. Virol. 80:8807–19 110. Mink M, Mosier SM, Janumpalli S, Davison D, Jin L, et al. 2005. Impact of human immunodeficiency virus type 1 gp41 amino acid substitutions selected during enfuvirtide treatment on gp41 binding and antiviral potency of enfuvirtide in vitro. J. Virol. 79:12447–54 111. Liu S, Jing W, Cheung B, Lu H, Sun J, et al. 2007. HIV gp41 C-terminal heptad repeat contains multi- functional domains. Relation to mechanisms of action of anti-HIV peptides. J. Biol. Chem. 282:9612–20 112. Eckert DM, Malashkevich VN, Hong LH, Carr PA, Kim PS. 1999. Inhibiting HIV-1 entry: discovery of d-peptide inhibitors that target the gp41 coiled-coil pocket. Cell 99:103–15 113. Welch BD, VanDemark AP, Heroux A, Hill CP, Kay MS. 2007. Potent D-peptide inhibitors of HIV-1 entry. Proc. Natl. Acad. Sci. USA 104:16828–33 114. Chinnadurai R, Rajan D, Munch J, Kirchhoff F. 2007. Human immunodeficiency virus type 1 variants resistant to first- and second-version fusion inhibitors and cytopathic in ex vivo human lymphoid tissue. J. Virol. 81:6563–72 115. Eron JJ, Gulick RM, Bartlett JA, Merigan T, Arduino R, et al. 2004. Short-term safety and antiretroviral activity of T-1249, a second-generation fusion inhibitor of HIV. J. Infect. Dis. 189:1075–83 116. Melby T, Demasi R, Cammack N, Miralles GD, Greenberg ML. 2007. Evolution of genotypic and phe- notypic resistance during chronic treatment with the fusion inhibitor T-1249. AIDS Res. Hum. Retrovir. 23:1366–73 117. Martin-Carbonero L. 2004. Discontinuation of the clinical development of fusion inhibitor T-1249. AIDS Rev. 6:61 118. Dwyer JJ, Wilson KL, Martin K, Seedorff JE, Hasan A, et al. 2008. Design of an engineered N-terminal HIV-1 gp41 trimer with enhanced stability and potency. Protein Sci. 17:633–43 119. Eggink D, Baldwin CE, Deng Y, Langedijk JP, Lu M, et al. 2008. Selection of T1249-resistant human immunodeficiency virus type 1 variants. J. Virol. 82:6678–88 120. He Y, Xiao Y, Song H, Liang Q, Ju D, et al. 2008. Design and evaluation of sifuvirtide, a novel HIV-1 fusion inhibitor. J. Biol. Chem. 283:11126–34 121. Root MJ, Kay MS, Kim PS. 2001. Protein design of an HIV-1 entry inhibitor. Science 291:884–88 122. Luftig MA, Mattu M, Di Giovine P, Geleziunas R, Hrin R, et al. 2006. Structural basis for HIV-1 neutralization by a gp41 fusion intermediate-directed antibody. Nat. Struct. Mol. Biol. 13:740–47 123. Steger HK, Root MJ. 2006. Kinetic dependence to HIV-1 entry inhibition. J. Biol. Chem. 281:25813–21 124. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, et al. 2000. The Protein Data Bank. Nucleic Acids Res. 28:235–42 by University of Texas - Austin on 10/05/09. For personal use only. 125. DeLano WL. 2007. The PyMOL molecular graphics system. Palo Alto, CA: DeLano Sci. http:// www.pymol.org/ Annu. Rev. Biochem. 2009.78:95-118. Downloaded from arjournals.annualreviews.org

118 Colman