Synthesis and Structure–Activity Relationships of Indazole - PDF Free Download (2023)

Article pubs.acs.org/jmc

Synthesis and Structure−Activity Relationships of Indazole Arylsulfonamides as Allosteric CC-Chemokine Receptor 4 (CCR4) Antagonists Panayiotis A. Procopiou,*,† John W. Barrett,§ Nicholas P. Barton,∥ Malcolm Begg,‡ David Clapham,⊥ Royston C. B. Copley,∥ Alison J. Ford,‡ Rebecca H. Graves,§ David A. Hall,‡ Ashley P. Hancock,† Alan P. Hill,∥ Heather Hobbs,† Simon T. Hodgson,† Coline Jumeaux,† Yannick M. L. Lacroix,† Afjal H. Miah,† Karen M. L. Morriss,† Deborah Needham,† Emma B. Sheriff,§ Robert J. Slack,‡ Claire E. Smith,§ Steven L. Sollis,† and Hugo Staton† †

Departments of Medicinal Chemistry, ‡Respiratory Biology, §Drug Metabolism and Pharmacokinetics, Respiratory CEDD, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, United Kingdom ∥ Computational and Structural Chemistry Department, Platform Technology & Science, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, United Kingdom ⊥ Exploratory Development Sciences, Platform Technology & Science, Research & Development, GlaxoSmithKline, Park Road, Ware, Hertfordshire SG12 ODP, United Kingdom S Supporting Information *

ABSTRACT: A series of indazole arylsulfonamides were synthesized and examined as human CCR4 antagonists. Methoxy- or hydroxyl- containing groups were the more potent indazole C4 substituents. Only small groups were tolerated at C5, C6, or C7, with the C6 analogues being preferred. The most potent N3-substituent was 5-chlorothiophene-2-sulfonamide. N1 meta-substituted benzyl groups possessing an α-amino-3-[(methylamino)acyl]− group were the most potent N1-substituents. Strongly basic amino groups had low oral absorption in vivo. Less basic analogues, such as morpholines, had good oral absorption; however, they also had high clearance. The most potent compound with high absorption in two species was analogue 6 (GSK2239633A), which was selected for further development. Aryl sulfonamide antagonists bind to CCR4 at an intracellular allosteric site denoted site II. X-ray diffraction studies on two indazole sulfonamide fragments suggested the presence of an important intramolecular interaction in the active conformation.

■

INTRODUCTION

as macrophage-derived chemokine (MDC), bind to the orthosteric binding site of CC-chemokine receptor 4 (CCR4).3 Upon exposure to allergen, dendritic cells within the lung (or other tissue) secrete MDC and TARC (also produced by endothelial cells) which can recruit Th2 cells from the circulation. The T cells can then migrate along this chemokine gradient to the dendritic cells. Upon maturation, the dendritic cells migrate from the inflamed tissue to local lymph nodes where the MDC and TARC which they produce may recruit further T cells to the inflammatory response. Elevated levels of TARC and MDC as well as accumulation of CCR4positive cells have been observed in lung biopsy samples from patients with atopic asthma following allergen challenge.6,7 Thus CCR4 antagonists represent a novel therapeutic intervention in diseases where CCR4 has a central role in pathogenesis, such as asthma, atopic dermatitis,8 allergic bronchopulmonary aspergillosis,9 cancer,10 the mosquitoborne tropical diseases, such as Dengue fever,11 and allergic

Chemokines are a group of small, basic proteins of 8−10 kDa, which together with their receptors mainly regulate the trafficking of leucocytes down a chemoattractant gradient.1,2 Chemokines possess four conserved cystein residues, and they are classified into four groups designated CC, CXC, C, and CX3C based on the arrangement of the first two conserved cysteine residues located at or near the N-terminus of each protein.3 Ten CC chemokine receptors have been identified so far and named as CC-chemokine receptor 1, 2, 3, etc. Most chemokine receptors recognize more than one chemokine and several chemokines bind to more than one receptor.1 CCR4 belongs to the 7-TM domain G-protein-coupled receptor family and is mainly expressed in T helper 2 (Th2) cells. The latter are a subset of CD4-positive T helper cells that produce interleukin (IL)-4, IL-5, and IL-13.4 Th2 cytokines in inflamed tissues lead to eosinophilia, high levels of serum IgE, and mast cell activation, all of which contribute to the pathogenesis of allergic diseases.5 CC Chemokine ligand 17 (CCL17), previously known as thymus activation-regulated chemokine (TARC), and CC chemokine ligand 22 (CCL22), also known © XXXX American Chemical Society

Received: October 24, 2012

A

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 1. Structures for some recently published CCR4 antagonists.

Scheme 1a

a Reagents and conditions: (a) NH2NH2·H2O, 1-butanol, reflux, 80−92%; (b) KOH, DMSO, 3-cyanobenzyl chloride, 52−84%; (c) 5-chloro-2thiophenesulfonyl chloride, pyridine, 29−87%; (d) 1 M LiAlH4 solution in ether, THF, 55−77%; (e) Ac2O, Et3N, DCM, 38−89%; (f) BBr3, DCM, 33%.

rhinitis.12 Progress in the discovery of small-molecule CCR4 antagonists as immunomodulatory agents was reviewed by Purandare and Somerville in 2006.13 A number of publications on CCR4 antagonists have appeared in the literature since the last review,14−23 and these antagonists appear to belong to two chemical categories (Figure 1). The first category includes lipophilic heteroarenes possessing basic amino groups such as Bristol Myers Squibb (BMS) compound 1,15 Astellas compound 2,18 and the Daiichi Sankyo compound 3.22 The AstraZeneca (AZ) 2,3-dichlorobenzenesulfonamide 424 and Ono 4-methylbenzenesulfonamide 525 belong to the second category of pyrazine arylsulfonamides. More recently, we have disclosed our own preliminary studies on a novel class of indazole arylsulfonamides including 6, which we believe is the first small-molecule clinical candidate targeting the CCR4 receptor.26

The AstraZeneca group have reported the discovery of a novel mechanism for antagonism of CCR4 and CCR5 receptors with a series of small-molecule pyrazine sulfonamides, which interact with an intracellular allosteric site on the receptor.27 The precise location of this allosteric binding site was not determined, however, it was suggested that it might be a generic site for chemokine receptors, or even more broadly for class A G-protein-coupled receptors. In our recent communication disclosing sulfonamide 6, we focused our investigations entirely on the variation of the indazole N1 substituent.26 In this publication, we present a fuller account of the structure−activity relationships (SAR) of the indazole arylsulfonamides covering the C4, C5, C6, C7, N3-sulfonamide, and additional N1-substitutions, together with extensive in vivo pharmacokinetic studies on 10 analogues, and shed light on the spatial requirement of the sulfonamide antagonists based on small-molecule X-ray diffraction studies. B

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 2a

Reagents and conditions: (a) NH2NH2·H2O, EtOH, 70 °C, 87%; (b) BOC2O, DMAP, Et3N, MeCN, 56%; (c) 5-chloro-2-thiophenesulfonyl chloride, DCM, pyridine, 56%; (d) TFA, DCM, 43%; (e) N-BOC-3-(hydroxymethyl)benzylamine (16), PPh3, iPrO2CNNCO2Pri, THF, 60 °C, 96%; (f) 2 M NaOH, MeOH, THF, 45 °C, 1 h; (g) TFA, DCM; (h) Ac2O, Et3N, DCM; (i) K2CO3, MeOH, 19%.

a

Scheme 3a

a

Reagents and conditions: (a) DIBAL-H, PhMe; (b) NaBH4, MeOH, 17−48%; (c) TFA, DCM, 66%; (d) Ac2O, Et3N, DCM; (e) 2 M aq NaOH, MeOH, 20 °C, 43%; (f) MeMgBr, THF, 30−77%.

■

CHEMISTRY The SAR investigations commenced with examination of the C4 substituted indazoles 7 outlined in Scheme 1. Reaction of the appropriately substituted 2-fluoro-benzonitrile 8a−d with hydrazine hydrate in n-butanol gave the amino indazoles 9a−d, which were selectively alkylated with 3-cyanobenzyl chloride in DMSO in the presence of powdered KOH to provide 10a−d. This procedure to introduce the N1 substituent was an improvement over our previous protocol which required the use of protecting group chemistry.26 Reaction of 10a−d with 5-

chloro-2-thiophenesulfonyl chloride in pyridine gave the sulfonamides 11a−d, the cyano group of which was reduced with LiAlH4 to give the benzylamines 12a−d. Acetylation of 12a−d with acetic anhydride gave the required amides 7a−d. On some occasions, 7 was accompanied by formation of a small amount of diacetylated products, which were converted back to 7 by trans-esterification with methanol over anhydrous potassium carbonate prior to workup. The phenolic analogue 7e was prepared from 12a by treatment with boron tribromide to give 12e (33%), followed by acetylation with acetic acid in C

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 4a

a

Reagents and conditions: (a) DAST, DCM, 5 h, 15%; (b) TFA, DCM, 54%; (c) Ac2O, Et3N, DCM; (d) 2 M aq NaOH, MeOH, 60 °C, 3 h, 36%.

Table 1

Mitsunobu reaction of 15 with N-BOC-3-(hydroxymethyl)benzylamine28 (16) gave 17 (96%), which was first hydrolyzed with NaOH to give the monosulfonamide 18, then deprotected with TFA, acetylated with Ac2O, and finally treated with potassium carbonate in MeOH to cleave overacetylated product, and provide 7f. The C4-hydroxymethyl analogue 7g was obtained by DIBAL-H reduction of the nitrile 17, followed by NaBH4 reduction of the resulting aldehyde 19 to give 20, cleavage of the N-BOC protecting group with TFA to give the benzylamine

the presence of HATU and diisopropylethylamine in DMF (69%). The 4-cyano analogue 7f was prepared by the route outlined in Scheme 2. Reaction of 3-fluoro-1,2-benzenedicarbonitrile (8f) with hydrazine hydrate in ethanol gave the 3-aminoindazole 9f (87%), which was then protected with di-tert-butyl dicarbonate in the presence of DMAP and Et3N in MeCN to give 13 in 56% yield. Treatment of 13 with excess 5-chloro-2thiophenesulfonyl chloride gave the bis-sulfonamide 14 in 56% yield, which was deprotected with TFA to give 15 in 43% yield. D

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

21, followed by acetylation with Ac2O, and finally hydrolysis with NaOH (Scheme 3). The C4-acetyl analogue 7h was prepared by methyl magnesium bromide addition to the nitrile 7f (77%), and the secondary alcohol 7i was prepared by NaBH4 reduction of 7h (48%). Addition of MeMgBr to ketone 7h gave the tertiary alcohol 7j (30%). Scheme 4 outlines the preparation of the C4-difluoromethyl analogue 7k which was prepared from the aldehyde 19 using diethylamino-sulfur trifluoride (DAST) in DCM to give 22 in 15% yield. The latter was deprotected with TFA to give the amine 23 (54%), which was then acetylated and deprotected to give 7k in 36% yield for the two steps. Further amide analogues of 6 containing either an ether (a− d) or amino (e−k) group were prepared by acylation of 12a with the appropriate carboxylic acid in DMF in the presence of N-[3-(dimethylamino)propyl]-N′-ethylcarbodiimide hydrochloride, N-hydroxybenztriazole hydrate (HOBT), and Nmethylmorpholine as base to give amides 24a−k (Table 1). The amino-acids (e−i) used in the amide coupling reactions were N-BOC protected and were subsequently deprotected with either 4 M HCl in dioxane or with TFA in DCM. The Nmethyl-morpholine enantiomers 24j and 24k were made from the racemic amino acid and separated by preparative chiral HPLC. The two enantiomers were labeled isomer A and B as their absolute configuration was not determined. Compound 24l was prepared from 12a and 3-oxetanone in the presence of sodium triacetoxyborohydride in THF in 94% yield. Compound 6 was prepared on a larger scale by reaction of 12a with 2-acetoxy-isobutyryl chloride in DCM and Et3N to give 25 in 77% yield, followed by trans-esterification of the acetate ester with methanol over anhydrous K2CO3 (71%) (Scheme 5).

EtOH gave 35 in 25%. The latter was reacted with 3chloromethyl benzonitrile in the presence of powdered KOH in DMSO to give 36, which was reduced with sodium borohydride in the presence of nickel(II) chloride and di-tertbutyl dicarbonate in THF and MeOH to give 37. Sulfonylation of 37 gave 38, which was then deprotected with TFA in DCM to give the amine 39. Acylation of 39 with AcOH or 2-acetoxyisobutyryl chloride and hydrolysis gave the two amides 33a and 33b, respectively. Scheme 8 outlines the synthesis of the 6-fluoro analogues 40a and 40b. Treatment of 3,5-difluoroanisole with α,αdichloromethyl methyl ether in the presence of TiCl4 gave the aldehyde 42, which was converted to the nitrile 43 with hydroxylamine-O-sulfonic acid and then cyclized to the indazole 44. Alkylation with 3-chloromethylbenzonitrile gave 45, which was converted to the sulfonamide 46 and then reduced with LiAlH4 to give the amine 47. The latter was acetylated to 40a and also converted to the amide 40b as previously. The 7-fluoro analogue 48 was obtained by the route outlined in Scheme 9. Treatment of 2,3-difluoro-6-methoxybenzonitrile 49 with hydrazine hydrate in N-methylpyrrolidinone (NMP) gave the indazole 50, which was alkylated with 3-chloromethylbenzonitrile as before to provide 51. The latter was reacted with 5-chloro-2-thiophenesulfonyl chloride to give 52, which was then reduced with LiAlH4 to 53 and subsequently acetylated with Ac2O to give 48. Scheme 10 shows the synthetic route by which analogues of 6 possessing different sulfonamide groups were synthesized. Benzonitrile 10a was reduced with LiAlH4 to give the diamine 54, which was selectively reacted with 2-acetoxy-isobutyryl chloride at the benzylamine group to give 55. The latter was reacted with the appropriate arylsulfonyl chloride and then trans-esterified with MeOH in anhydrous potassium carbonate to give the target sulfonamides 56a−d.

Scheme 5a

■

RESULTS AND DISCUSSION All compounds in Table 2 were tested as human CCR4 antagonists in vitro. Antagonist potency was determined by a [35S]-GTPγS radioligand competition functional assay using recombinant CCR4-expressing CHO cell membranes adhered to WGA-coated Leadseeker SPA beads in assay buffer (20 mM HEPES, 10 mM MgCl2, 100 mM NaCl, 0.05% BSA, 40 μg/mL saponin at pH 7.4) with output measured on a Wallac Microbeta Trilux scintillation counter.30 Another assay using human whole blood was used as a secondary screen to determine potency against the native receptor for the more potent compounds in the primary assay. The assay quantified cytoskeletal reorganization (formation of filamentous (F-) actin) which occurs in a variety of cells in response to chemoattractants and is a prelude to chemotaxis. This was achieved by staining the F-actin with a fluorescent derivative of phalloidin, which binds with high affinity and specificity to the interface between actin monomers in F-actin. The response was measured as an increase in the fluorescence intensity of the target cell population in a flow cytometer and was expressed as a pA2. In this assay, human CD4+ CCR4+ lymphocytes were identified by staining with antibodies to CD4 and CCR4. Ligand lipophilicity efficiency index (LLE) is a useful metric introduced by Leeson and Springthorpe for assessing the potency and lipophilicity of drug-like molecules during lead optimization.31 We have used the more recently introduced LLEAT, which combines lipophilicity, potency, and size and

a Reagents and conditions: (a) 2-acetoxy-isobutyryl chloride, Et3N, DCM, 77%; (b) K2CO3 (3 equiv), MeOH, 2 h, 20 °C, 71%.

The three pyridyl analogues of 7a, regioisomers 26, 27, and 28, were prepared from the bis-sulfonamide 2928 according to Scheme 6, which outlines only the synthesis of regioisomer 26. Thus Mitsunobu coupling of 29 with 2-(hydroxymethyl)-4pyridinecarbonitrile 3029 gave 31, which was reduced with LiAlH4 and hydrolyzed with aqueous NaOH in MeOH to give the benzylamine 32. The latter was acetylated with AcOH in the presence of HATU and Et3N to give 26. The other two regioisomers 27 and 28 were synthesized by an analogous method starting from 29 and the appropriate hydroxymethylpyridinecarbonitrile. The 5-fluoro-analogues 33a and 33b were prepared by the route shown in Scheme 7. Reaction of 3,6-difluoro-2methoxybenzonitrile (34) with hydrazine monohydrate in E

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 6a

Reagents and conditions: (a) PPh3, tBuO2CNNCO2But, THF, 80 °C, 88%; (b) LiAlH4, THF, 1 h; (c) 2 M NaOH, H2O, MeOH, 70 °C, 53%; (d) AcOH, HATU, Et3N, DCM, 59%. a

Scheme 7a

Reagents and conditions (a) NH2NH2·H2O, EtOH, 80 °C, 25%; (b) 3-chloromethylbenzonitrile, KOH, DMSO, −5 °C, 48%; (c) NaBH4, NiCl2, BOC2O, THF, MeOH, 65%; (d) 5-chloro-2-thiophenesulfonyl chloride, pyridine, DCM, 55%; (e) TFA, DCM, 45%; (f) (for 33a) AcOH, HATU, DIPEA, DMF, 27%; (g) (for 33b) 2-acetoxyisobutyryl chloride, pyridine, DCM; (h) KOH, H2O, MeOH, 49%.

a

compounds was measured and presented in Table 2. The test compounds were compared with two standards, compounds 1 and 5, together with our own sulfonamides 6 and 7a.26 We started our SAR investigations by focusing our attention on C4 substitutions with compounds 7b−7k. The compounds with LLEAT higher than 7a were the three hydroxyl compounds 7e, 7g, and 7i. All three compounds had higher affinity than 7a, ranging from 7.7 to 8.2, and lower log D by more than a log unit. In addition, the CLND solubilities of 7e and 7i were

makes comparisons with conventional ligand efficiency (LE) easier at the commonly used threshold value of 0.3 kcal/mol.32 In Table 2 the LLEAT, the chromatographic log D (chrom log D at pH 7.4) and the chemiluminescent nitrogen detection (CLND) kinetic solubility are included for all test compounds in this study.33 The high throughput CLND solubility assay involves addition of aqueous buffer to a test compound DMSO solution over a period of time until the compound precipitates. Finally, the pKa of the sulfonamide moiety for selected F

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 8a

Reagents and conditions: (a) Cl2CHOMe, TiCl4, DCM, 57%; (b) NH2OSO3H, H2O, 110 °C, 3 h, 95%; (c) NH2NH2·H2O, n-BuOH, 110 °C, 21%; (d) 3-chloromethylbenzonitrile, KOH, DMSO, 20%; (e) 5-chloro-2-thiophenesulfonyl chloride, pyridine, CHCl3, 45%; (f) LiAlH4, THF, 18%; (g) (for 40a) Ac2O, Et3N, CHCl3, 52%; (h) (for 40b) 2-acetoxyisobutyryl chloride, pyridine, DCM; (i) K2CO3, MeOH, 51%. a

Scheme 9a

Reagents and conditions. (a) NH2NH2·H2O, NMP, 150 °C; (b) 3-chloromethylbenzonitrile, KOH, DMSO, 56%; (c) 5-chloro-2-thiophenesulfonyl chloride, pyridine, DCM, 30%; (d) LiAlH4, THF, 36%; (e) Ac2O, Et3N, DCM, 34%.

a

strated that the pKa increased in the case of an ortho-methoxy group in the aniline ring.34 The pKa of all the C4 substituted analogues (7a−k) was measured and is shown in Table 2. Comparing 7a with 7b the MeO- group increased the pKa in line with Ludwig’s observations. Despite measurable differences in the acidity of the various analogues’ sulfonamide NH, there was no correlation between sulfonamide pKa and GTPγS affinity. Interestingly, the affinity of the two analogues that cannot form a hydrogen bond, namely the des-methoxy analogue 7b and the cyano analogue 7f, have the lowest affinity (6.6 and 6.5, respectively), whereas the three hydroxyl analogues 7e, 7g, and 7i, which are capable of hydrogen bonding, have the highest affinity (7.9, 8.2, and 7.7, respectively). We have therefore hypothesized that the effect of hydrogen-bonding would be to hold the analogues in a locked active conformation. In the absence of suitable crystals

higher than 7a. The higher affinity did not translate to the whole blood actin polymerization assay but remained at approximately the same level as 7a, with the highest potency of 6.3 for alcohol 7i. On the basis of these data, compounds 7e, 7g, and 7i were selected for further investigation. It was interesting to note that both the AZ and Ono compounds, 4 and 5, possessed an ortho-alkoxy substituent capable of hydrogen bonding with the sulfonamide NH forming a fivemembered ring. Our own candidate 6 had a C4 methoxy substituent capable of hydrogen bonding with the sulfonamide NH, forming a six-membered ring. The effect of hydrogen bonding to the acidic sulfonamide NH would be to alter its acidity, leading us to hypothesize that there might be a relationship between acidity and potency. The effects of ortho substituents in the dissociation of substituted N-phenyl benzenesulfonamides were reported by Ludwig, who demonG

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 10a

a

Reagents and conditions: (a) LiAlH4, THF, 50%; (b) 2-acetoxy-isobutyryl chloride, Et3N, DCM, 87%; (c) arylsulfonyl chloride, pyridine, DCM; (d) K2CO3, MeOH, 39−55%.

related cyclic ether amides 24c and 24d, were investigated. The LLEAT of all four compounds was inferior to 7a, whereas the chrom log D was similar or even higher than 7a. Interestingly, the CLND solubility of 24a, 24c, and 24d was slightly increased compared to 7a, whereas 24b was similar. In comparing the 3oxetane with the 2-oxetane analogues (24a vs 24b), the former had higher solubility and lower chrom log D, which might be due to the greater exposure of the oxygen atom. Another way of increasing the solubility would be the introduction of an ionizable group such as an amino group. All three amines, 24e, 24f, and 24g, had identical affinity, lower log D, and slightly increased CLND solubility. Furthermore, all three analogues had high whole blood potency, and indeed they were the most potent compounds in this assay, making them suitable for further investigation. It was interesting to investigate the reason that these amines were uniformly more potent in the whole blood assay, and we hypothesized that this might be due to lower plasma protein binding. The rapid equilibrium dialysis human plasma protein binding for a range of analogues is presented in Table 3. The three strongly basic analogues 24e, 24f, and 24g had the lowest plasma protein binding, indicating that the higher potency in the whole blood actin polymerization assay was related to the higher levels of unbound compound. The four less basic analogues 24h−k containing a morpholine moiety had very high affinity in the GTPγS assay. Compounds 24j and 24k, which are enantiomers, were found to have identical activity with a pIC50 value of 8.2, one of the highest values in this series. However, due to their high molecular weight, their LLEAT value was slightly reduced at 0.26. The solubility of analogues 24h−k was higher as expected, making these compounds worthy of further investigation. The basic analogue containing the 3-oxetane 24l was less potent with a LLEAT of 0.2 and low solubility, and it was therefore rejected. Another approach to increase the solubility of analogues of 7a was replacement of the phenyl ring of the benzylic substituent by a heterocyclic ring, such as pyridine, in analogues 26−28. These analogues had reduced lipophilicity and very good LLEAT, however, their potency was lower than 7a and they were not examined any further. Similarly introduction of

of 6 for an X-ray diffraction study, we have examined smaller fragments of 6 such as the N1-methyl analogues 57 and 58. Fragment 57 gave suitable crystals for a small-molecule X-ray diffraction study (Figure 2). In this structure, an intramolecular hydrogen bond between the sulfonamide and the C4-OMe group is clearly observed [N−H, 0.80(3) Å; H···O, 2.52(3) Å; N···O, 3.036(3) Å; and ∠N−H···O, 124(3)°]. A consequence of this hydrogen bond is to allow the plane normals to the thiophene ring and the indazole core to be approximately orthogonal [83.73(7)°]. Fragment 58 possesses a 3-thiophene substituent at C4, which removes the possibility of this intramolecular hydrogen bond. Figure 3 shows that the above two plane normals are now closer to being parallel [17.2(2)°]. A search of the Cambridge Structural Database (v 5.33, Nov 2011) for [5,6] bicyclic cores containing any combination of carbon, nitrogen, and oxygen atoms, and being substituted with a sulfonamide moiety and an oxygen atom at the equivalent 3and 4-position of the indazole ring system, yielded no hits, confirming the novel arrangement reported herein. Moreover, the GTPγS affinity of fragment 57 was 6.4, whereas that of 58 was 5.8. Replacement of the N1-Me substituent of 57 with a benzyl group leads to analogues with increased potency, whereas in the case of 58 lead to nonadditive effects. These data are consistent with the theory that disruption of the intramolecular hydrogen bond leads to an alternative conformation and a different binding mode, which in turn leads to reduced potency. Anisotropic atomic displacement ellipsoids for the nonhydrogen atoms are shown at the 50% probability level. Hydrogen atoms are displayed with an arbitrarily small radius. The intramolecular hydrogen bond is indicated by a dashed line. Anisotropic atomic displacement ellipsoids for the nonhydrogen atoms are shown at the 50% probability level. Hydrogen atoms are displayed with an arbitrarily small radius. In our preliminary communication, we have reported that amide and hydroxyl-amide groups at the N1 benzyl group increased both the potency and solubility.26 Recently, oxetanes were reported as groups that increase the solubility of compounds,35 therefore, oxetane amides 24a and 24b, and H

dx.doi.org/10.1021/jm301572h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 2. In Vitro Data pIC50 for GTPγS Binding Assay, Human Whole Blood F-Actin Polymerization pA2, Ligand Efficiency, Chrom log D, CLND Solubility, and pKa compd

GTPγS pIC50

LLE(AT)

chrom log D

CLNDa solubility (μg/mL)

hWB actin polymerization pA2 (n)

pKa

1 5 6 7a 7b 7c 7d 7e 7f 7g 7h 7i 7j 7k 24a 24b 24c 24d 24e 24f 24g 24h 24i 24j 24k 24l 26 27 28 33a 33b 40a 40b 48 56a 56b 56c 56d

8.26 ± 0.01 (150) 8.41 ± 0.02 (147) 7.83 ± 0.02 (113) 7.50 ± 0.04 (10) 6.6 ± 0.0 (4) 7.5 ± 0.1 (2) 7.1 ± 0.1 (2) 7.9 ± 0.2 (2) 6.5 ± 0.1 (4) 8.2 ± 0.1 (2) 7.0 ± 0.1 (2) 7.7 ± 0.1 (4) 7.0 ± 0.1 (2) 7.0 ± 0.1 (2) 7.7 ± 0.2 (4) 7.5 ± 0.1 (4) 7.4 ± 0.2 (4) 7.5 ± 0.2 (4) 7.8 ± 0.1 (2) 7.8 ± 0.2 (6) 7.8 ± 0.1 (10) 7.6 ± 0.2 (4) 7.8 ± 0.1 (6) 8.2 ± 0.1 (3) 8.2 ± 0.0 (3) 6.6 ± 0.2 (4) 6.8 ± 0.1 (4) 7.0 ± 0.1 (6) 6.9 ± 0.2 (2) 7.9 ± 0.1 (3) 7.4 ± 0.1 (2) 7.4 ± 0.2 (4) 7.8 ± 0.0 (14) 7.1 ± 0.1 (4) 6.4 ± 0.3 (4) 7.3 ± 0.2 (4) 6.4 ± 0.2 (4) 7.8 ± 0.2 (3)

0.25 0.27 0.27 0.28 0.24 0.25 0.26 0.32 0.26 0.35 0.27 0.31 0.26 0.24 0.25 0.27 0.23 0.26 0.27 0.28 0.24 0.27 0.28 0.26 0.26 0.20 0.32 0.33 0.32 0.29 0.25 0.26 0.26 0.25 0.19 0.24 0.19 0.25

2.9 2.9 4.3 4.1 3.3 3.8 3.0 2.9 2.5 2.6 4.8 2.7 2.9 3.7 4.0 4.5 4.9 4.4 3.1 3.3 3.3 3.9 3.9 4.4 4.4 4.3 2.6 2.5 3.0 4.1 4.1 4.1 4.2 4.0 5.1 4.5 4.7 4.5

≥145 193 ≥162 ≥124 ≥117 ≥226 ≥139 ≥166 ≥155 ≥113 57 ≥208 ≥159 ≥188 ≥173 124 ≥153 ≥193 191 ≥180 ≥220 ≥250 ≥212 ≥203 ≥180 122 ≥141 ≥161 ≥182 ≥199 ≥174 ≥161 ≥198 ≥177 76 ≥198 157 36

7.3 ± 0.3 (4) 6.6 ± 0.1 (24) 6.2 ± 0.1 (11) 6.0 ± 0.1 (7) ND 5.2 ± 0.2 (2) 5.3 ± 0.2 (2) 6.1 ± 0.1 (2) ND 6.0 ± 0.3 (2) ND 6.3 ± 0.1 (3) ND 5.1 ± 0.1 (2) 6.0 ± 0.2 (3) ND 6.1 ± 0.1 (2) 6.3 ± 0.1 (3) 6.7 ± 0.0 (5) 6.85 ± 0.04 (4) 6.9 ± 0.1 (4) 6.25 ± 0.15 (4) 7.1 ± 0.3 (4) 6.6 ± 0.2 (6) 6.6 ± 0.2 (6) ND