WO2015044928A1

WO2015044928A1 - Substituted indoles and their use as non-nucleoside reverse transcriptase inhibitors

Info

Publication number: WO2015044928A1
Application number: PCT/IB2014/064955
Authority: WO
Inventors: Stephen Christopher PELLY; Willem Arjen Lodewijk VAN OTTERLO; Ronel MULLER; Adriaan BASSON
Original assignee: Stellenbosch University
Priority date: 2013-09-30
Filing date: 2014-09-30
Publication date: 2015-04-02

Abstract

Compounds of Formula I or pharmaceutically acceptable salts thereof are claimed, wherein R₁ is an ester, amide or a heterocycle; R₂ is C or N; R₃ is a methoxy or ethoxy group, an alkyl group or a heterocyclic group; R₄ is a substituted or unsubstituted phenyl or heteroaromatic group; and R₅ is a halide or a nitrile. The use of these compounds as non-nucleoside reverse transcriptase inhibitors (NNRTIs) for inhibiting or treating HIV infection or AIDS is also described.

Description

SUBSTITUTED INDOLES AND THEIR USE AS NON-NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to South African provisional patent application number 2013/07447, which is incorporated by reference herein.

FIELD OF THE INVENTION

This invention relates to substituted indoles which can be used as HIV non-nucleoside reverse transcriptase inhibitors.

BACKGROUND TO THE INVENTION A cure for the Acquired Immunodeficiency Syndrome (AIDS) continues to remain elusive, despite more than thirty years of research into this disease. It is estimated that over 33 million people are infected with the human immunodeficiency virus (HIV), which is known to be the causative agent of AIDS. If left untreated, HIV sufferers become susceptible to AIDS, characterised by a drastic decline in the body's immune system response, the inevitable onset of opportunistic infections, and ultimately death.¹ The majority of these people are located in poorer developing areas, most significantly, sub-Saharan Africa.² Significant progress has been made in developing new therapeutic agents to control viral replication within HIV infected individuals, resulting in delaying the onset of AIDS. However, due to the high mutation rate of the virus, and the resulting emergence of resistance, there is a continuous need to develop new drugs to control the viral levels in HIV sufferers.

Currently there are six therapeutic targets utilized for anti-HIV treatment and within this group the enzyme reverse transcriptase (RT), which is crucial for the conversion of single stranded viral RNA into double stranded DNA, provides two opportunities for therapeutic intervention. The first of these is at the actual catalytic site, utilizing DNA base mimics which act as chain terminators. AZT, the first drug to be used in the treatment of HIV, falls into this category.³ Located just 10 A from the catalytic site is a hydrophobic and relatively small cleft known as the non-nucleoside inhibitor binding pocket (NNIBP), and this is the second opportunity for therapeutic intervention. The resulting allosteric effect of specific compounds binding in this pocket significantly inhibits the normal functioning of the RT enzyme. These compounds are commonly referred to as non-nucleoside reverse transcriptase inhibitors (NNRTIs) and have been found to be particularly effective in treating HIV, as the compounds are potent and show significantly less toxicity than their nucleoside counterparts.^4"5 Moreover, being small and somewhat non-polar molecules, they are the only anti-HIV drugs which are well suited to cross the blood brain barrier (BBB).⁶ This is particularly important in order to tackle viral reservoirs located across the BBB, not only to reduce viral levels overall but also to prevent complications such as the onset of AIDS dementia complex.⁷ Unlike drugs in the other therapeutic categories, the structural variance of NNRTIs is quite remarkable. Currently there are only five licensed NNRTIs: nevirapine, delavirdine, efavirez, etravirine and rilpivirine, the structures of which are shown below.

Efavirenz Etravirine Rilpivirine

EG*, = 2.3 nl EC_m = 2.1 riM EC_H0 = 1.3

Currently licensed NNRTIs and their potency values

Other NNRTIs have been described, such as the indolylarylsulfoxides which Williams ei al. reported in 1993 as having strong activity against HIV RT.¹⁴ Following this, the molecules evolved from sulfinyls to sulfonyls and eventually to sulfonamide compounds (shown below), which were potent inhibitors of HIV RT. The sulfonamide structures, however, were susceptible to high clearance in vivo, and so these structures were never pursued as drug candidates.^15"18

indolylarylsulfoxides indolylarylsulfones indolylarylsulfonamides

IC_S0 = 3 n

As a result of the high mutation rate of the HI virus and concomitant onset of resistance, and the limited number of effective drugs to control the virus, there is a continuing need to develop new therapeutic agents to control the viral levels in people living with HIV until such time as a cure is found.

SUMMARY OF THE INVENTION According to a first embodiment of the invention, there is provided a compound of formula I or a salt thereof

I

wherein

Ri is an ester, amide, or a heterocycle;

R₂ is C or N;

R₃ is a methoxy or ethoxy group, an alkyl group, or a heterocyclic group;

R₄ is a substituted or unsubstituted phenyl or heteroaromatic group; and

R₅ is a halogen or a nitrile.

When R is an ester, may be -C0₂C₂H₅ or -C0₂CH₃. When R_! is an amide, it may be primary, or secondary, or tertiary.

When R! is a heterocycle, Ri may be an isoxozole, an imidazole or an oxadiazole. R₃ may be a methoxy or ethoxy group, an alkyl group such as methyl or ethyl, or a heterocycle selected from a three or four membered oxygen-containing ring system, such as an epoxide or oxetane.

When R₂ is N, R₃ is typically an alkyl group such as methyl or ethyl, or is a methoxy or ethoxy group.

R₄ may be an unsubstituted or substituted phenyl or a heteroaromatic, such as pyridyl. Substituents, at any position on the ring, may be independently selected from methyl, ethyl, nitrile or halogens, such as CI, Br or F.

R₅ may be a halogen selected from F, Br and CI. More preferably, R₅ is CI.

The compound may be a compound of any one of formulae I I to XIV:

According to a second embodiment of the invention, there is provided a pharmaceutical composition comprising a compound of Formula I or a pharmaceutically acceptable salt thereof and an excipient, carrier or diluent. According to a third embodiment of the invention, there is provided a compound of Formula I or a pharmaceutical composition comprising a compound of Formula I for use in a method of treating or preventing HIV infection or AIDS.

According to a fourth embodiment of the invention, there is provided the use of a compound of Formula I in a method of preparing a medicament for use in a method of preventing or treating HIV infection or AIDS.

According to a fifth embodiment of the invention, there is provided a method of treating or preventing HIV infection or AIDS in a subject, the method comprising administering an effective amount of a compound of Formula I or a pharmaceutical composition substantially as described above to the subject.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows the 2-, 3- and 5-positions of the indole scaffold at which the compound of formula I can be substituted; Figure 2 shows a schematic representation of efavirenz bound within the NNRTI binding pocket showing a double hydrogen bonding interaction with Lys101 (A). The envisaged indole scaffold according to the invention similarly facilitates a double hydrogen bond interaction, and appropriate substitution at the 3-position to create Pi-Pi stacking opportunities with Tyr181 or Tyr188 (by conformational flexibility) (B);

Figure 3 shows that ethoxy or methoxy analogues are well accommodated in the small

Val179 pocket and result in better docking scores and binding energy values than the corresponding cyclopropyl derivatives. Molecular modelling indicates only a slight preference for the R enantiomer (orange) in comparison to the S enantiomer (white);

Figure 4 shows an ethyl derivative (A) and an amine derivative (B) as examples of a replacement compound for the methoxy derivative at the 3-position;

Figure 5 is a depiction of docking studies comparing the use of the hydrophobic space in the small Val179 pocket by a previously described cyclopropyl indole derivative (A) and a compound of the present invention (B);

Figure 6 is a depiction of molecular modelling studies showing that the ft-enantiomer (B) is only slightly better in terms of binding energy than the S-enantiomer (A);

Figure 7 shows the replacement of the ester with a suitable bioisostere which will maintain the H-bond interaction to Lys101 ;

Figure 8 shows various heterocycles as bioisosteres for the ester functionality. Some of these may have additional benefits such as conformational stability of the inhibitor (A) or electrostatic interactions with the water solvating the entrance to the binding pocket (B); and

Figure 9 is a chart showing comparison of activities of compounds of Formula II and XIV against a range of mutant HIV viral strains in a phenotypic assay, showing a decrease in potency expressed as fold change values. DETAILED DESCRIPTION OF THE INVENTION

Compounds of Formula I:

or pharmaceutically acceptable derivatives or salts thereof are described herein. Their use as non-nucleoside reverse transcriptase inhibitors (NNRTIs) for inhibiting HIV infection is also described. The derivative compounds will comprise functional groups at the 2-, 3- and/or 5-positions of the indole (Figure 1 ), with the following desirable features:

• having a hydrogen bond acceptor moiety at the 2-position to facilitate a second hydrogen bond to the backbone of Lys101 of the enzyme HIV reverse transcriptase (RT);

• being capable of forming a Pi-Pi stacking interaction at the 3-position with Tyr181 or Tyr188 of HIV RT, and filling the small Val179 pocket of HIV RT with a hydrophobic group

• having a halogen (CI, Br, I or F) or a nitrile group at the 5-position of the indole (it has been found that the potency is significantly enhanced by the presence of a halogen (especially CI or Br) or a nitrile group [18]). The compounds of formula I or salts or derivatives thereof will provide double hydrogen bonding interaction, thereby imparting a significant benefit in potency (as in the case of the known NNRTI efavirenz), whilst maintaining a level of conformational flexibility to ensure efficacy against resistant strains of the virus. Figure 2 shows a comparison of the binding of efavirenz to RT (A) and the envisaged binding of an example of a compound of formula I to RT (B).

Thus,

R, can be an ester, amide or a heterocycle;

R₂ can be C or N;

R₃ can be a methoxy or ethoxy group, an alkyl group or a heterocyclic group (and is not a cyclopropyl group);

R₄ can be a substituted or unsubstituted phenyl or heteroaromatic group; and

R₅ can be a halide or a nitrile. In one embodiment, the compound has formula II

II

where

R, is an ester;

R₂ is C;

R₃ is-OCH₃;

R₄ is phenyl; and

R₅ is CI.

Using molecular modelling, the applicant has previously designed cyclopropyl-indole derivatives as NNRTIs. Several of these compounds (e.g. 1 and 2) were found to be slightly more potent than nevirapine (IC₅₀ ~ 0.1 μΜ) in phenotypic assays, and are described in South African Patent No. 2012/09319, the contents of which are incorporated herein.

Previously described indole-based NNRTIs The applicant has now designed alternative indole-based compounds which do not have a cyclopropyl group. These have been shown to have improved potencies compared to the previously described compounds and have also been shown to be effective against resistant strains of the HI virus. At the back of the NNIBP (non-nucleoside inhibitor binding pocket) is a small hydrophobic pocket. It has generally been found that inhibitors with a suitable halogen substituent, or in some cases, a nitrile group, have significantly improved potency. This position corresponds to the 5-position on the indole system (R₅ in Formula I) and substituents at this position can therefore be selected from the group consisting of a nitrile, chloride, bromide or fluoride. In the compounds described below, R₅ is a chloride. The invention, however, is not intended to be limited to such compounds.

In initial efforts to improve upon the previously described cyclopropyl-indole compounds, the cyclopropyl moiety at the indole's 3-position (R₃ in Formula I) was retained (Figure 3A), as molecular modelling studies had indicated that this group was well suited to the small hydrophobic pocket located in the vicinity of Val179 (molecular modelling studies were carried out using Accelrys Discovery Studio 4.0. The RT receptor utilised was obtained from the Protein Databank with PDB ID 2RF2. Docking studies were performed using the CDocker protocol). Therefore, a suitable replacement for the ester located the indole's 2-position was sought (R_T in Formula I). Although the carbonyl of the ester functionality is well positioned to facilitate a second hydrogen bonding interaction to Lys1 01 , the hydrophobic ethyl chain is somewhat of a mismatch, as this protrudes out of the non-nucleoside inhibitor binding pocket (NNIBP) into the solvation region. Thus, a suitable polar group was incorporated at the end of the chain to interact with the water.

To this end, analogues 4-6 were synthesised (Scheme 1 ) by a two-step process of ester hydrolysis, followed by pyBOP mediated coupling of the resulting carboxylic acid 3 with the corresponding amine, alcohol or hydrazine.

Scheme 1 : Reagents and conditions: (a) KOH, EtOH, 4h, 80 °C, then HCI, 81%; (b)

PyBop, HOBt, HO(CH₂)₂OH, H₂N(CH₂)₂OH or NH₂NH₂, DIPEA, corresponding

amine or alcohol, DMF, 2h, rt, 60-85%

These compounds were then evaluated in a phenotypic assay. Given the favourable modelling results, coupled with the fact that the compounds exhibited more desirable cLogP values (with the exception of 6), it was surprisingly found that they were actually significantly less potent than their hydrophobic ethyl counterpart 1.

The binding contributions made by the cyclopropyl group and the aryl moiety attached at the 3-position of the indole were then investigated (R₄ and R₃ of Formula I, respectively). The effect of removing the hydrophobic group occupying the Val1 79 pocket was determined in a small SAR (structure-activity relationship) study. To this end, indole derivative 11 was prepared by acylation and defunctionalisation of the carbonyl 8 (Scheme 2). A deterioration in potency was observed, indicative that the cyclopropyl group had some beneficial role in the binding of the inhibitor.

Scheme 2: Reagents and conditions: (a) benzoyl chloride, AICI₃, DCE, 80°C, 18h, 86%;

(b) TFA, Et₃SiH, DMF, 4h, 0-30 °C, 32%; (c) Et₂NH, CH₂0, NaOH, 2h, 30 °C; (d)

Mel, KCN, MeOH, 18h, 30°C, 40% over two steps

The Pi-Pi stacking aromatic group needs to be considered when optimising the moiety at the 3-position of the compound. The applicant's modelling work showed that six membered aromatic ring systems are favoured over five membered rings, as the slightly larger ring system forms two vital interactions: one of these is the Pi-Pi stacking with Tyr181 (or Tyr188), and the second interaction is a sigma-Pi interaction with Trp229. This is a particularly important interaction as Trp229 is a conserved amino acid across the various mutant strains of the virus [5]. When the phenyl functionality was completely removed as in compound 12, a significant detrimental effect to the potency was observed.¹¹ This lends testimony to the importance of the Pi-Pi stacking of the aryl group to Tyr181.

As an extension of this study, the nitrile derivative 13 was synthesised by way of a Mannich reaction, followed by alkylation and substitution with a cyanide ion. This new structure lacks the functionality to occupy the small Val179 pocket and does not effectively interact with Tyr181. Therefore, 13 performed particularly poorly in an NNRTI phenotypic assay. In studying many different crystallised NNRTIs obtained from the Protein Databank, it is clear that the Val179 binding pocket is not only able to tolerate additional functionality on the phenyl ring, but may in fact be preferable, as it leads to more potent inhibitors.^{5, 12, 21 22} Any of the atoms of the phenyl ring on the scaffold of compounds of formula I of the present invention can thus be substituted. For example, one or more of the hydrogens at the 2-, 3,- 4- , 5- and/or 6-position could be substituted with a methyl, ethyl, nitrile or halide group (e.g. chloride, bromide or fluoride).

When synthesising the above analogues, the formation of the cyclopropyl group was found to be a problematic reaction, being modest yielding and difficult to reproduce. More easily attainable bioisosteres (structurally different groups or moieties that can form similar intermolecular interactions) at position R₃ in Formula I were therefore investigated. The most obvious change would be to switch to a dimethyl system as in compound 14 (Figure 3). However, molecular modelling studies indicated that the Val179 pocket would not be wide enough to accommodate both methyl groups, set about a somewhat more tetrahedral tertiary carbon in comparison to the cyclopropyl system 1. In a series of analogues evaluated by docking and binding energy calculations, the acyclic ethyl and methyl-ether analogues (15 and 16 respectively) appeared promising and in fact performed even better in terms of binding energy and docking score calculations than the original cyclopropyl derivative 1. An amine derivative (R₂)was also considered (Figure 4). Docking studies showed that the amine derivative binds particularly well within the NNRTI binding pocket and exhibits an interesting conformational flexibility, allowing Pi-Pi stacking of the aromatic group with either Tyr181 or Tyr188. This is particularly useful in that it could overcome the known Y181 C mutation, which normally abolishes the Pi-Pi stacking interaction of the inhibitor. The docking studies show that in the event of this mutation, a slight conformational change allows for Pi-Pi stacking with Tyr188, thereby maintaining this important interaction. Synthetically, the amine derivative could easily be obtained via a palladium catalysed amination reaction,¹⁹ or be derived from the azide derivative.²⁰ R₃ could also be an ethoxy or heterocycle, such as a 3- or 4- membered oxygen-containing ring system (e.g. an epoxide or oxetane).

The methyl ether derivative 16 (Formula II) was attained from the ketone 18 (Scheme 3), an intermediate in the synthetic strategy for obtaining the cyclopropyl system.¹¹ After protecting ketone 18 to afford 22, borohydride reduction, followed by acid-mediated substitution of the resulting hydroxyl functionality provided the protected derivative 31 in good yield. Finally, base mediated deprotection of the indole provided 16.

Scheme 3: Reagents and conditions: (a) benzoyl chloride or 3,5-dimethylbenzoyl chloride, AICI₃, DCE, 30 °C, 4h, 72-86%; -(b) p-TsCI, NaH, DMF, rt, 18h, 81-85% when R₃=Tosyl; Boc₂0, DIE A, DMAP, THF, rt, 5h, 82-84% when R₃=Boc; (c)

NaBH₄, EtOH, rt, 18h, 57-95%; (d) p-TsOH, MeOH or EtOH, rt, 18h, 38-87%; (e) 34, LDA, trifluoromethanesulfonyl chloride, THF, -78 X, 20min, 78%; (f) K₂C0₃, MeOH, reflux, 3h when R₃=Boc or KOH, THF, rt 18h, when R₃=Tosyl; (g ) K₃P0₄, MeOH or EtOH, 70 X), 4h, 90-99% R₃=Boc

Evaluation of racemic 16 in a phenotypic assay revealed that the methyl ether derivative was significantly more potent than the cyclopropyl derivative 1 , with an IC50 of 0.02 μΜ (Table 1 ). Moreover, 16 shows low toxicity (the CC₅₀ value was 30 μΜ). The potency of the compound may be due to the better occupation of the hydrophobic Val179 pocket by the methoxy moiety compared to similar compounds with other moieties at this position, such as the previously described cyclopropyl moiety (Figure 5).²³

Interestingly, the somewhat larger ethyl-ether derivative 36 was also potent in the phenotypic assay, but the lack of an alkyl group in this position (compound 30) had a serious detrimental effect to the potency. Table 1 : Evaluation of target compounds by phenotypic assay against wild type HIV-1 (IC₅₀) and cellular toxicity values (CC₅o)

IC₅₀ cc₅₀

Indole R₂ R₃

(μ ) (μ )

16 C0₂Et Ph OMe 0.02 25.3

30 C0₂Et Ph OH 1.15 22.7

36 C0₂Et Ph OEt 0.03 22.8

38 S0₂CF₃ Ph OMe 1.20 24.2

39 C0₂Et 3,5-dimethylphenyl OMe 0.03 36.5

40 H Ph OMe 2.43 67.0

41 H 3,5-dimethylphenyl OMe 0.30 67.7

Nevirapine 0.12 Molecular modelling studies suggested that both the R and S configurations of 15 and 16 would be well accommodated in the Val179 pocket with a slight shift of the phenyl substituent. The above assays were therefore performed on a racemic mixture, but it is possible that one enantiomer may be significantly more potent than the other (molecular modelling studies indicated that the f?-enantiomer is slightly favoured in terms of binding energy in comparison to the S-enantiomer (Figure 6)). Methods for separating enantiomers are well known in the art, and one way to separate the enantiomers would be to attach a suitable chiral auxiliary to the indole nitrogen, thereby creating diastereomers which, being energetically non-equivalent, should be separable by conventional chromatography. After separation of the diastereomers, removal of the chiral auxiliary should afford the separate enantiomers. Alternatively, a chiral reduction could be considered.

The ester functionality at the 2-position of the indole (P ) was identified as a potential problem. Although this functional group is stable in phenotypic assays, it is possible that it will be labile in vivo, hydrolysed to the corresponding carboxylate. SAR work conducted by the applicant on the synthesised carboxylate showed that it is a poor inhibitor of HIV.

Removing the ester group is unlikely to be a solution to this problem, as the carbonyl forms an important second hydrogen bonding interaction with Lys101 of the HIV enzyme reverse transcriptase (RT). The applicant's SAR data showed that removing this feature leads to a considerable decrease in potency. It is envisaged that replacing the ester with a suitable bioisostere (a functional group that is electronically similar to the ester in terms of being a hydrogen bond acceptor, and has similar spatial characteristics, so as not to perturb the binding of the methoxy and aromatic groups on the adjacent 3-position of the indole) will overcome this problem, whilst maintaining the desired H-bond interaction to Lys101 (Figure 7).

Preliminary modelling studies indicated that various heterocycles show promise as bioisosteres, such as the isoxazole derivative. Some heterocycles may even impart additional desired conformational stability to the inhibitor. For example, an imidazole heterocycle would fulfil the requirements of a suitable bioisostere for the Lys101 interaction (Figure 8A) and would also form an intramolecular hydrogen bond to the oxygen of the methoxy, stabilising the structure in the correct conformation for binding and thus lowering the entropic penalty of binding. More elaborate heterocycles, such as oxadiazoles, would also form the required hydrogen bond to Lys101 and have the additional benefit of a heteroatom pointing toward the outside of the binding pocket (Figure 8B). These moieties would be able to hydrogen bond to the water molecules solvating the entrance to the pocket. The most promising functional group modification, however, appeared to be a sulfone. A particular benefit of this group over the ester functionality was the improved angle of the hydrogen bond between the sulfone oxygen and the amide of Lys101. Synthesis of the sulfone 38 was begun with the more sparsely functionalised 5-chloroindole 17 (Scheme 3). A Friedal-Crafts acylation introduced the required phenacyl substituent at the 3-position of the indole, forming 20. Protection of the indole nitrogen, followed by reduction of the ketone 24 and then acid mediated substitution to introduce the methyl ether afforded the near final derivate 34. Directed ortho metallation, mediated by the tosyl protecting group then paved the way for the installation of the trifluoromethane sulfonyl functionality, providing 37. Finally, the key hydrogen bond donor indole nitrogen was once again revealed after deprotection, affording 38. In contrast to the results suggested by modelling, it was surprisingly found that compound 38 performed poorly, with an IC₅₀ value of 1 .2 μΜ (Table 1 ).

As compound 16 is the most potent compound in this series, derivatives thereof that would improve the performance against common resistant strains of the virus were investigated. 3,5-disubstituted phenyl rings and methyl, nitrile and chloro-substituents were considered. Analogue 39 was synthesised starting from ethyl 5-chloro-2-indolecarboxylate 7 and 3,5- dimethylbenzoyl chloride, itself readily obtainable from the corresponding acid (Scheme 3). The activities of compounds 16 and 39 were compared against a range of mutant strains in a phenotypic assay (Figure 9). Clearly evident was the fact that across all mutant strains the dimethyl analogue 39 was significantly more potent than the unsubstituted phenyl variant 16, although these two compounds performed very similarly against the wild type virus (16 was slightly more effective against the wild type virus). Of particular interest are the results of 16 and 39 when assayed against HIV-1 viruses with prevalent NNRTI resistance mutations. Lead compound 16 was found to be ineffective against the V106M mutation, which causes resistance to efavirenz. However, the dimethyl variant 39 remained effective. The Y181 C mutation, which renders nevirapine ineffective, was found to be problematic for both compounds, no doubt as a result of the dependence on the aryl π-π stacking with Tyr181 . Nevertheless, compound 39 was still significantly more potent than compound 16. The G190A mutation also causes resistance to nevirapine, but both compounds 16 and 39 remain effective, with some reduction in potency. Of great interest to the applicant was that when compounds 16 and 39 were evaluated against the most prevalent and particularly problematic front line mutation, K103N, indole 39 remained effective, displaying only slightly reduced potency, while intermediate and high levels of resistance were displayed towards compound 16 and nevirapine, respectively.

To evaluate the importance of the ester at the 2-position, analogues 40 and 41 , which lack the hydrogen bond acceptor functionality at the indole's 2-position, were synthesised (Scheme 3). These compounds were significantly less potent than their counterparts 16 and 39, highlighting the benefits of maintaining this interaction. References

[I ] D. G . Prajapati, R. Ramajayam , M. R. Yadav, R. Giridhar, The search for potent, small molecule NNRTIs: A review, Biorg. Med. Chem. 2009, 17, 5744-5762.

[2] R. K. Gupta, D. Pillay, HIV resistance and the developing world, Int. J. Antimicrob. Agents 2007, 29, 510-517.

[3] E. De Clercq, Anti-HIV drugs: 25 compounds approved within 25 years after the discovery of HIV,

Int. J. Antimicrob. Agents 2009, 33, 307-320.

[4] M.-P. de Bethune, Non-nucleoside reverse transcriptase inhibitors (NNRTIs), their discovery, development, and use in the treatment of HIV-1 infection: A review of the last 20 years (1 989—

2009), Antiviral Res. 2010, 85, 75-90.

[5] D. Li, P. Zhan, E. De Clercq, X. Liu, Strategies for the design of HIV-1 non-nucleoside reverse transcriptase inhibitors: lessons from the development of seven representative paradigms, J.

Med. Chem. 2012, 55, 3595-3613.

[6] S. M. Drake, NNRTIs— a new class of drugs for HIV, J. Antimicrob. Chemother. 2000, 45, 41 7-

420.

[7] S. A. Lipton, Neuronal injury associated with HIV-1 : approaches to treatment, Annu. Rev.

Pharmacool. Toxicol. 1998, 38, 1 59-1 77.

[8] K. Das, A. D. Clark, P. J. Lewi, J. Heeres, M. R. de Jonge, L. M. H. Koymans, H. M. Vinkers, F.

Daeyaert, D. W. Ludovici , M. J. Kukla, B. De Corte, R. W. Kavash, C. Y. Ho, H. Ye, M. A.

Lichtenstein, K. Andries, R. Pauwels, M.-P. de Bethune, P. L. Boyer, P. Clark, S. H. Hughes, P.

A. J. Janssen, E. Arnold, Roles of conformational and positional adaptability in structure-based design of TMC125-R1 65335 (etravirine) and related non-nucleoside reverse transcriptase inhibitors that are highly potent and effective against wild-type and drug-resistant HIV-1 variants,

J. Med. Chem . 2004, 47, 2550-2560.

[9] G. A. Freeman, C. W. Andrews, A. L. Hopkins, G. S. Lowell, L. T. Schaller, J. R. Cowan, S. S.

Gonzales, G. W. Koszalka, R. J. Hazen, L. R. Boone, R. G. Ferris, K. L. Creech, G . B. Roberts,

S. A. Short, K. Weaver, D. J. Reynolds, J. Milton, J. Ren, D. I . Stuart, D. K. Stammers, J. H.

Chan, Design of non-nucleoside inhibitors of HIV-1 reverse transcriptase with improved drug resistance properties. 2. , J. Med. Chem . 2004, 47, 5923-5936.

[1 0] A. L. Hopkins, J. Ren, J. Milton, R. J. Hazen, J. H. Chan, D. I . Stuart, D. K. Stammers, Design of non-nucleoside inhibitors of HIV-1 reverse transcriptase with improved drug resistance properties.

1 , J. Med. Chem. 2004, 47, 5912-5922.

[I I ] M. Hassam, A. E. Basson, D. C. Liotta, L. Morris, W. A. L. van Otterlo, S. C. Pelly, Novel cyclopropyl-indole derivatives as HIV non-nucleoside reverse transcriptase inhibitors, ACS Med. Chem. Lett. 2012, 3, 470-475.

[12] E. B. Lansdon, K. M. Brendza, M. Hung, R. Wang , S. Mukund, D. Jin, G. Birkus, N. Kutty, X. Liu, Crystal Structures of HIV-1 Reverse Transcriptase with Etravirine (TMC125) and Rilpivirine (TMC278) : Implications for Drug Design, J. Med. Chem. 201 0, 53, 4295-4299. [13] J. Ren, J. Milton, K. L. Weaver, S. A. Short, D. I. Stuart, D. K. Stammers, Structural basis for the resilience of efavirenz (DMP-266) to drug resistance mutations in HIV-1 reverse transcriptase, Structure 2000, 8, 1 089-1 094.

[14] T. M. Williams, T. M. Ciccarone, S. C. MacTough, C. S. Rooney, S. K. Balani, J. H. Condra, E. A.

Emini, M. E. Goldman, W. J. Greenlee, 5-Chloro-3-(phenylsulfonyl)indole-2-carboxamide: a novel, non-nucleoside inhibitor of HIV-1 reverse transcriptase, J. Med. Chem . 1993, 36, 1291 - 1294.

[1 5] R. Ragno, M. Artico, G. De Martino, G . La Regina, A. Coluccia, A. Di Pasquali, R. Silvestri, Docking and 3-D QSAR studies on indolyl aryl sulfones. binding mode exploration at the HIV-1 reverse transcriptase non-nucleoside binding site and design of highly active N-(2- hydroxyethyl)carboxamide and N-(2-hydroxyethyl)carbohydrazide derivatives, J. Med. Chem . 2005, 48, 21 3-223.

[1 6] R. Silvestri, M. Artico, G. De Martino, G. La Regina, R. Loddo, M. La Colla, P. La Colla, Simple, short peptide derivatives of a sulfonylindolecarboxamide (L-737, 126) active in vitro against HIV-1 wild type and variants carrying non-nucleoside reverse transcriptase inhibitor resistance mutations, J. Med. Chem . 2004, 47, 3892-3896.

[1 7] R. Silvestri, G. De Martino, G. La Regina, M. Artico, S. Massa, L. Vargiu, M. Mura, A. G. Loi, T.

Marceddu, P. La Colla, Novel indolyl aryl sulfones active against HIV-1 carrying NNRTI resistance mutations: synthesis and SAR studies, J. Med. Chem. 2003, 46, 2482-2493.

[1 8] Z. Zhao, S. E. Wolkenberg, M. Lu, V. Munshi, G. Moyer, M. Feng, A. V. Carella, L. T. Ecto, L. J.

Gabryelski, M.-T. Lai, S. G. Prasad, Y. Yan, G. B. McGaughey, M. D. Miller, C. W. Lindsley, G. D. Hartman, J. P. Vacca, T. M. Williams, Novel indole-3-sulfonamides as potent HIV non-nucleoside reverse transcriptase inhibitors (NNRTIs), Bioorg. Med . Chem. Lett. 2008, 1 8, 554-559.

[1 9] M. W. Hooper, M. Utsunomiya, J. F. Hartwig, Scope and Mechanism of Palladium-Catalyzed Amination of Five-Membered Heterocyclic Halides, The Journal of Organic Chemistry 2003, 68, 2861 -2873.

[20] D. Lubriks, I. Sokolovs, E. Suna, Indirect C-H Azidation of Heterocycles via Copper-Catalyzed Regioselective Fragmentation of Unsymmetrical A3-lodanes, J. Am. Chem. Soc. 2012, 134, 15436-1 5442.

[21 ] Z. K. Sweeney, S. F. Harris, N. Arora, H. Javanbakht, Y. Li, J. Fretland, J. P. Davidson, J. R.

Billedeau, S. K. Gleason, D. irschfeld, . . ennedy-Smith, T. Mir adegan, R. Roet , M. Smith, S. S erry, . M. Suh, . u, S. Tsing, A. . illase or, A. Paul, . Su, . eile , . . ang, A. S. hou, . A. ernelius, F.-J. Zhang, K. Klumpp, Design of Annulated Pyrazoles as Inhibitors of HIV-1 Reverse Transcriptase II , J. Med. Chem . 2008, 51 , 7449-7458.

[22] R. Corbau, J. Mori, C. Phillips, L. Fishburn, A. Martin, C. Mowbray, W. Panton, C. Smith- Burchnell, A. Thornberry, H. Ringrose, T. Knochel, S. Irving, M. Westby, A. Wood, M. Perros, Lersivirine, a nonnucleoside reverse transcriptase inhibitor with activity against drug-resistant human immunodeficiency virus type 1 , Antimicrob. Agents Chemother. 2010, 54, 4451 -4463.

[23] South African Patent No. 2012/0931 9. [24] Gupta, R. K.; Kohli, A.; McCormick, A. L; Towers, G. J.; Pillay, D.; Parry, C. M. AIDS 2010, 24, 1651 -1655.

[25] Parry, C. M. ; Kohli, A.; Boinett, C. J. ; Towers, G. J.; McCormick, A. L; Pillay, D. J. Virol. 2009, 83, 9094-9101 .

[26] Tim, M. J. Immunol. Methods 1983, 65, 55-63.

Claims

CLAIMS:

1 . A compound of formula I

or a salt thereof, wherein

R, is an ester, amide or a heterocycle;

R₂ is C or N;

R₃ is a methoxy or ethoxy group, an alkyl group, or a heterocyclic group;

R₄ is a substituted or unsubstituted phenyl or heteroaromatic group; and R₅ is a halogen or a nitrile.

2. A compound as claimed in claim 1 , wherein is an ester selected from -CO₂C₂H₅ and -CO₂CH₃.

3. A compound as claimed in claim 1 , wherein R, is a primary or secondary amide.

4. A compound as claimed in claim 1 , wherein is a heterocycle selected from an isoxozole, an imidazole and an oxadiazole.

5. A compound as claimed in any one of claims 1 to 4, wherein R₂ is C.

6. A compound as claimed in any one of claims 1 to 4, wherein R₂ is N.

7. A compound as claimed in any one of claims 1 to 6, wherein R₃ is a methoxy or ethoxy group.

8. A compound as claimed in any one of claims 1 to 6, wherein R₃ is a methyl or ethyl.

9. A compound as claimed in any one of claims 1 to 6, wherein R₃ is a three or four membered oxygen-containing heterocyclic group selected from an epoxide or an oxetane.

10. A compound as claimed in any one of claims 1 to 9, wherein R₄ is an unsubstituted phenyl.

1 1. A compound as claimed in any one of claims 1 to 9, wherein R₄ is a substituted phenyl with substituents at any position on the ring and the substituents are independently selected from the group consisting of methyl, ethyl, nitrile, chloride, bromide and iodide.

12. A compound as claimed in any one of claims 1 to 9, wherein R₄ is a pyridyl.

13. A compound as claimed in any one of claims 1 to 12, wherein R₅ is selected from F, Br and CI.

14. A compound as claimed in claim 13, wherein R₅ is chloride.

15. A compound as claimed in any one of claims 1 to 14, which is selected from:

A compound as claimed in claim 15, which is

A compound as claimed in claim 15, which is

18. A pharmaceutical composition comprising a compound of any one of claims 1 to 17 and an excipient, carrier or diluent.

19. A compound as claimed in any one of claims 1 to 17 or a pharmaceutical composition as claimed in claim 18 for use in treating or preventing HIV infection or AIDS. Use of a compound of any one of claims 1 to 17 in the manufacture of a medicament for preventing or treating HIV infection or AIDS.

A method of treating or preventing HIV infection or AIDS in a subject, the method comprising administering an effective amount of a compound of any one of claims 1 to 17 or a pharmaceutical composition of claim 18 to the subject.