α-Substituted 2-(3-fluoro-4-methylsulfonamidophenyl)acetamides as potent TRPV1 antagonists

Abstract

A series of α-substituted acetamide derivatives of previously reported 2-(3-fluoro-4-methylsulfonamidophenyl)propanamide leads (1, 2) were investigated for antagonism of hTRPV1 activation by capsaicin. Compound 34, which possesses an α-m-tolyl substituent, showed highly potent and selective antagonism of capsaicin with K_i(CAP) = 0.1 nM. It thus reflected a 3-fold improvement in potency over parent 1. Docking analysis using our homology model indicated that the high potency of 34 might be attributed to a specific hydrophobic interaction of the m-tolyl group with the receptor.

The transient receptor potential vanilloid 1 channel (TRPV1) has emerged as a very promising therapeutic target, reflecting its central role in nociception and its involvement in a range of diseases.^1–3 Capsaicin⁴ and resiniferatoxin⁵ provided early structural leads for understanding the vanilloid pharmacophore, and it was soon recognized that appropriately modified derivatives were able to achieve antagonism.^6,7 Intense medicinal chemical efforts have now afforded a broad base of structures for the further development of TRPV1 antagonists.⁸ Although a crystal structure for TRPV1 is not available, homology modeling of TRPV1 represents a complementary tool to further refine insights into ligand–TRPV1 interactions.⁹ As the field has matured, it has become apparent that compounds may have different potencies or efficacies as antagonists for different modes of TRPV1 activation and may have different effects at the whole animal level.^10,11

Of particular importance, hyperthermia represents a common side effect of TRPV1 antagonists and the pattern of antagonistic activities for different agonists has been suggested to be related to the ability of an antagonist to cause hyperthermia.¹¹ Further, different antagonists are differentially affected by the signaling pathways that regulate TRPV1.¹² The availability of novel antagonists will be critical for optimizing such characteristics.

Previously, we have described a series of N-(6-trifluoromethylpyridin-3-yl)methyl 2-(3-fluoro-4-methylsulfon amidophenyl)pro panamides, originally designed based on a pharmacophoric combination approach, that showed potent hTRPV1 antagonism toward multiple activators.^13–17 The structure of capsaicin (CAP) has been divided into three pharmacophoric regions.⁴ Correspondingly, the antagonistic template was subdivided into the same three pharmacophoric regions, namely the A-region (3-fluoro-4-methylsulfon amidophenyl), the B-region (propanamide), and the C-region ((6-trifluoromethyl-pyridin-3-yl)methyl) (Fig. 1). The structure activity relationship (SAR) studies of the template were initiated by incorporating a variety of functional groups including amino,¹³ oxy,¹⁴ thio,¹⁵ alkyl¹⁶ and aryl¹⁷ groups into the 2-position in the pyridine C-region. In these series, multiple compounds showed highly potent and (S)-stereospecific antagonism of hTRPV1 activators including capsaicin, pH, heat (45 °C) and N-arachidonoyl dopamine (NADA). Their in vitro mechanism of action as TRPV1 antagonists was confirmed in vivo by their ability to block capsaicin-induced hypothermia. Most importantly, the selected compounds showed promising antinociceptive activity in neuropathic and inflammatory pain models.

Figure 1.

Lead TRPV1 antagonists.

In an effort to optimize the properties of the antagonistic template, our focus was next directed to the propanamide (or α-methylacetamide) B-region. The modeling analysis of the B-region using our established hTRPV1 homology model^9,13 indicated that the amide group made a hydrogen bond with Tyr511 and also contributed to the appropriate positioning of the C-region for hydrophobic interactions. In order to gain better positioning of the C-region with the receptor through α-substitution in the B-region, with consequent enhancement of activity, we have extensively investigated the structure activity relationships of α-substituted acetamide derivatives for hTRPV1 antagonism. Compounds 1 and 2, previously described as highly potent and stereoselective antagonists, were selected as the parent compounds for this optimization of the B-region (Fig. 1).^13,15 Here, we describe incorporation of various alkyl, dialkyl and aryl groups at the α-position in the B-region of 1 and 2, with 4-methylpiperidinyl and cyclohexylthio groups at the 2-position in the pyridine C-region, respectively, and we have evaluated their antagonism of CAP stimulation of hTRPV1 expressed in CHO cells.

The syntheses of α-substituted acetamide derivatives were accomplished by one of four different synthetic routes (Schemes 1–4). For the synthesis of A/B-regions of the α-ethyl derivative 21, the nucleophilic substitution of 2,4-difluoronitrobenzene 3 by diethyl malonate afforded the adduct 4,¹⁸ which was alkylated with ethyl iodide to provide 5. Further elaboration by a known protocol provided the a-ethyl acetic acid 7 (Scheme 1). α-Substituted 2-phenylacetic acids as the A/B-regions of compounds 22–43 were prepared efficiently by vicarious nucleophilic substitution¹⁹ from the corresponding ethyl 2-substituted 2-chloroacetates 8, respectively, using the method previously reported for the synthesis of parent compounds 1 and 2 (Scheme 2). For the synthesis of the A/B-regions of the α,α′-dialkyl derivatives 44–47, the α-methyl ethyl ester 13²⁰ was alkylated with the corresponding alkyl iodides to provide the dialkyl intermediates 14, which were converted to the respective α,α′-dialkylated acetic acids 15 (Scheme 3). For the synthesis of the A/B-regions of the cyclopentyl derivatives 48–49, the ethyl 2-(3-fluorophenyl)acetate 16 was nitrated and then alkylated with 1,4-dibromobutane. Further three-step elaboration provided the α-cyclopentyl acetic acid 18 (Scheme 4). The prepared α-substituted 2-phenylacetic acids (7, 12, 15, 18) as A/B-regions were coupled with the two C-regions, 19¹³ and 20¹⁵, to provide the final compounds 21–49, respectively (Scheme 5).

Scheme 1.

Synthesis of the A/B-region for 21. Reagents and conditions: (a) diethyl malonate, NaH, DMSO, 15 °C, 52%; (b) EtI, TBAF, DMF, rt, overnight, 65%; (c) Pd/C, H₂, EtOH, 80%; (d) MsCl, pyridine, CH₂Cl₂, 0 °C–rt, 1 h, 86%; (e) 2 N NaOH, EtOH, reflux, 8 h, 99%.

Scheme 4.

Synthesis of the A/B-region for 48–49. Reagents and conditions: (a) HNO₃, H₂SO₄, 0 °C, 2 h; (b) NaH, 1,4-dibromobutane, THF, rt, 30 min; (c) Pd/C, H₂, EtOH, 1 h, 92%; (d) MsCl, pyridine, CH₂Cl₂, 0 °C–rt, 1 h, 82%; (e) LiOH, THF–H₂O, 45 °C, 2 h.

Scheme 2.

Synthesis of the A/B-region for 22–43. Reagents and conditions: (a) 1-fluoro-2-nitrobenzene, KOtBu, DMF, −5 to 0 °C; (b) H₂, Pd/C, EtOH–EtOAc (1:1); (c) MsCl, pyridine, CH₂Cl₂, 0 °C–rt; (d) LiOH, THF–H₂O (2:1), reflux-rt.

Scheme 3.

Synthesis of the A/B-region for 44–47. Reagents and conditions: (a) NaH, MeI (or EtI), DMF, 0 °C; (b) Pd/C, H₂, EtOH, 1 h, 82–90%; (c) MsCl, pyridine, CH₂Cl₂, 0 °C–rt, 1 h, 78–89%; (d) LiOH, THF–H₂O, 45 °C, 2 h.

Scheme 5.

Syntheses of α-substituted derivatives. Reagents and conditions: (a) RCO₂H (7, 12, 15, 18), TBTU, HOBt, DIPEA, THF–DMF, rt, 70–90%

The synthesized TRPV1 ligands were evaluated in vitro for antagonism as measured by inhibition of activation by capsaicin (100 nM, CAP). The assays were conducted using a fluorometric imaging plate reader (FLIPR) with human TRPV1 heterologously expressed in Chinese hamster ovary (CHO) cells.¹³ The results are summarized in Table 1, together with the potencies of the parent antagonists 1 and 2 as racemates with K_i(CAP) = 0.3 and 0.9 nM, respectively.^13,15

Table 1.

In vitro hTRPV1 antagonistic activities for α-substituted acetamide derivatives

X		Ki(CAP) (nM)		Ki(CAP) (nM)
	1	0.3	2	0.9
	21	1.6		NS
	22	88.7	23	103
	24	31.1	25	WE
	26	1.3	27	62.5
	28	2.4	29	48.7
	30	3.4	31	37.6
	32	42.9	33	WE
	34	0.1	35	56.4
	36	1.1	37	7.5
	38	65.9	39	WE
	40	49.3	41	WE
	42	WE	43	40.3
	44	49	45	65.7
	46	124	47	WE
	48	WE	49	WE

WE: weakly active, NS: not synthesize.

First, we examined the SAR of the α-alkyl derivatives (21–25). The α-ethyl derivative 21, the one-carbon elongated surrogate of 1, showed 5-fold less antagonistic potency compared to 1. However, a further increase in size, such as cyclopentyl (22, 23) and cyclohexyl (24, 25) groups, led to a dramatic loss in antagonism (>100-fold) compared to the parents 1 and 2.

Next, the SAR of the α-phenyl derivatives (26–37) was examined. The α-phenyl derivatives 26 and 27 showed a moderate or a substantial reduction, respectively, in antagonism compared to the parents but still exhibited potent antagonism compared to the cyclohexyl derivatives (24, 25) despite of a comparable size of the α-substituent. We therefore further investigated other phenyl derivatives substituted with small size groups such as fluoro and methyl. α-Fluorophenyl derivatives (28–31) displayed comparable potencies to the α-phenyl derivatives (26–27), respectively. However, there was a significant difference in potency with the α-tolyl derivatives (32–37) depending on the methyl position in the α-tolyl group as well as on the 2-substituent in the pyridine C-region. Whereas the α-o-tolyl derivatives (32, 33) exhibited significant reductions (>140 fold) in potency, the α-p-tolyl derivatives (36, 37) displayed moderate loss (3–8 fold) in potency compared to the parents 1 and 2, respectively. Surprisingly, the α-m-tolyl derivative 34 with a 4-methylpiperidinyl group showed exceptionally excellent antagonism with K_i(CAP) = 0.1 nM, which was even 3-fold more potent than the parent 1. Contrary to this finding, the α-m-tolyl derivative 35 with a cyclohexylthio group was found to be 60-fold less potent than the parent 2. The results indicated that the 2-substituent in the pyridine C-region contributed to the appropriate orientation of α-substituent in the B-region for optimal binding with the receptor. Compound 34 was found to be the only antagonist in this series which showed better potency than the corresponding parent and was the most potent antagonist in the series of 2-(3-fluoro-4-methylsulfonamidophenyl)propanamide antagonists reported so far.

Next we investigated the SAR of α-benzyl derivatives (38–43). Unfortunately, all of them showed much weaker potencies than the corresponding α-phenyl derivatives, indicating that α-benzyl substituents appear to provide unfavorable steric interactions with the receptor.

Finally, we examined α,α′-dialkyl derivatives (44–49). They were found to be weak antagonist. The SAR analysis indicated that their antagonistic potencies became progressively poorer as the size of α,α′-dialkyl groups increased.

In order to examine its in vitro activity as an antagonist for multiple hTRPV1 activators, compound 34, the most potent antagonist in this series, was evaluated for antagonism of TRPV1 activation by capsaicin, pH, heat (45 °C) and N-arachidonoyl dopamine (NADA). Inhibitory potencies were compared to the respective values for the parent 1 (Table 2). Whereas compound 34 showed 3-fold better potency for antagonism of CAP, it exhibited 5–15 fold weaker potency toward the other activators compared to 1.

Table 2.

In vitro hTRPV1 antagonistic activities of 34 for multiple activators

Activators, parameter	1	34
CAP (f) Ki (nM)	0.3 (±0.07)	0.1 (±0.03)
pH, IC50 (nM)	15.8 (±3.68)	86.7 (±10.6)
Heat 45 °C, IC50 (nM)	2.56 (±0.45)	36.2 (±5.58)
NADA (f) Ki (nM)	0.02 (±0.01)	0.25 (±0.05)

Using our hTRPV1 model¹³ built based on our rTRPV1 homology model⁹, we performed a flexible docking study of compound 34. Its predicted binding interactions are shown in Figure 2.²¹ Since the active stereoisomer of the propanamide B-region in this template was found to be the (S)-configuration in all cases in our previous reports, we performed the docking study with the (S)-34.

Figure 2.

Docking result of (S)-34 in the hTRPV1 model. (A) Binding mode and interactions of 34 at the binding site of hTRPV1. The key interacting residues are labeled and displayed as capped-stick with their carbon atoms in white color. The helices are colored by gray and the neighboring monomer helices are shown in line ribbon. Compound 34 is depicted in ball-and-stick with the carbon atoms in magenta color. The van der Waals surface representation of 34 is colored by the lipophilic potential property. Hydrogen bonds are shown in black dashed lines, and non-polar hydrogens are not shown for clarity. (B) The Fast Connolly surface representation of hTRPV1 and the van der Waals surface representation of the docked 34. The molecular surface of hTRPV1 was generated by MOLCAD and presented with the lipophilic potential property. For clarity, the surface of hTRPV1 is Z-clipped and that of 34 is in magenta color. (C) 2-D representation of the interactions between 34 and hTRPV1. Hydrophobic, hydrogen bonding, and π–π stacking interactions are marked in light brown, red, and blue, respectively.

The A-region, 3-fluoro-4-methylsulfonamidophenyl group, occupied the deep bottom hole and formed hydrophobic interactions with Tyr511, Ile564, Ile569, and Met572. The sulfonamide S=O participated in hydrogen bonding with the backbone amide of Lys571, and the phenyl ring engaged in π-π stacking with the phenyl ring of Tyr511. In the B-region, the amide group formed hydrogen bonds with the side chains of Tyr511 and Lys571, contributing to the appropriate positioning of the C-region for its hydrophobic interactions. In addition, the m-tolyl ring as an α-substituent nicely fitted into the hydrophobic pocket composed of Tyr554, Tyr555, Ile564, and Ile569, contributing for the high potency of 34. In the C-region, the 3-trifluoromethyl pyridine ring extended toward the upper hydrophobic area consisting of Met514 and Tyr511. Furthermore, the 4-methylpiperidine ring was involved in hydrophobic interactions with Thr550, Tyr554, and Phe587 of the adjacent monomer.

In conclusion, the structure activity relationship for hTRPV1 antagonism by α-substituted acetamide derivatives in the B-region of the 2-(3-fluoro-4-methylsulfonamidophenyl) propanamide template was investigated. Steric repulsion of the α-substituent emerged as a key determinant of antagonistic potency. In this series, compound 34 showed excellent antagonism with K_i(CAP) = 0.1 nM. Compound 34 was thus 3-fold more potent than the parent 1, which we previously had described as potent antagonist, and compound 34 is the most potent antagonist reported using this antagonistic template. Detailed in vitro analysis of antagonism by 34 of the response to multiple TRPV1 activators indicated that 34 exhibited preferential antagonism for capsaicin, with weaker potency for the other activators. The docking study of 34 indicated that its high potency might be attributed to a specific hydrophobic interaction of the m-tolyl group with the receptor.