Chimeras Derived from a P2Y14 Receptor Antagonist and UDP-Sugar Agonists for Potential Treatment of Inflammation

Abstract

Tethered glycoconjugates of a naphthalene- and piperidine-containing antagonist of the P2Y₁₄ receptor (PPTN) were synthesized, and their nM receptor binding affinity was determined using a fluorescent tracer in hP2Y₁₄R-expressing whole CHO cells. The rationale for preparing mono- and disaccharide conjugates of the antagonists was to explore the receptor binding site, which we know recognizes a glucose moiety on the native agonist (UDP-glucose), as well as enhance aqueous solubility and pharmacokinetics, including kidney excretion to potentially counteract sterile inflammation. Glycoconjugates with varied linker length, including PEG chains, were compared in hP2Y₁₄R binding, suggesting that an optimal affinity (IC₅₀, nM) in the piperidine series was achieved for triazolyl N-linked glucose conjugates having one (8a, MRS4872, 3.21) or two (7a, MRS4865, 2.40) methylene spacers. In comparison of different carbohydrate conjugates lacking a piperidine moiety but containing triazole spacers, optimal hP2Y₁₄R affinity (IC₅₀, nM) was achieved with N-linked glycosides of fucose 10f (6.19) and lactose 10h (1.88), and C-linked glucose 11a (5.30). Selected compounds were examined in mouse models of conditions known to be ameliorated by P2Y₁₄R antagonists. Two glycoconjugates that lacked a piperidine moiety, N-linked glucose derivative 10a and the isomeric C-linked glucose derivative 11a, were protective in a mouse model of allergic asthma. Piperidine-containing glucose conjugate 7a of intermediate linker length and corresponding glucuronide 7b (MRS4866) protected against neuropathic pain. Thus, glycoconjugation of a known antagonist scaffold has produced less hydrophobic P2Y₁₄R antagonists having substantial in vitro and in vivo activity.

Figure 1

Structures of known P2Y₁₄R ligands 1–3 (human (h) and mouse (m) P2Y₁₄R affinities obtained in a fluorescent antagonist binding assay in whole cells^,,, are shown) and the general formula of newly synthesized target compounds 4 and 5a–11a. The exact structures of the latter compounds are shown in Tables 1 and 2. Compound 2 is the prototypical antagonist PPTN, and 3 is MRS4738. Carbohydrate moieties are shown in red, and linker moieties in blue. A final compound series 12a–12c (Table 2) replaces the phenyl ring at the 4 position of the naphthyl moiety with a triazole.

Figure 2

Comparison of the proposed hP2Y₁₄R-bound poses proposed for agonist 1 (A) and rigidified antagonist 3 (B), as presented in Trujillo et al. and Wen et al.^,

Scheme 1. Preparation of 4, Analogue of 2, Modified with β-d-Glucose Attached Directly on the Naphthalene Moiety

Reagents and conditions: (a) CsCO₃, ACN, rt, 48 h, 77%; (b) 3 M NaOH/MeOH/THF (1:1:1), rt, overnight, 62%.

Scheme 2. Preparation of Analogues of 2 Modified with N-Linked β-d-Glucose Bridged with Alkyltriazolyl Linkers of Different Lengths, Including Preparation of an Azido-Sugar Intermediate 14 (A) and Tethered Conjugates 5a–8a Containing a Piperidine Moiety and Linked by Either a PEG-Containing (B) or a Straight Alkyl Chain (C)

Reagents and conditions: (a) NaN₃, DMF, rt, 2 h, 99%; (b) for 15 and 18: alkynyl bromide, K₂CO₃, DMF, 70 °C, 4 h; then H₂O/MeOH (2:1), LiOH, rt, 3 h, 30–75%; (c) 14, CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, quantitative; (d) 3 M NaOH/MeOH (1:2), rt, overnight, 47–79%; (e) 6-iodohex-1-yne (for 17) or propargyl bromide (for 19), K₂CO₃, DMF, rt, overnight; then H₂O/MeOH (2:1), LiOH, rt, 3 h, 75–89%.

Scheme 3. Preparation of Analogues of 2 Modified with N-Linked β-d-Glucuronide or Unsaturated Glucuronide Bridged with Alkyl-Triazolyl Linkers of Different Lengths, Including Preparation of an Azido-Sugar Intermediate 24 (A) and Tethered Conjugates 5b–8b and 5c–8c Containing a Piperidine Moiety and Linked by Either a PEG-Containing (B) or a Straight Alkyl Chain (C)

Reagents and conditions: (a) NaN₃, DMF, rt, 2 h, 91%; (b) 24, CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, 76%–quantitative; (c) 3 M NaOH/MeOH (1:2), rt, overnight, 22–42% b and 5–11% c.

Scheme 4. Preparation of Analogues of 2 in Which the Piperidine Moiety is Replaced by N-Linked 1H-1,2,3-Triazol-1-yl Glucose (10a), Glucuronide (10b), or Unsaturated Glucuronide (10c)

Reagents and conditions: (a) Na₂CO₃, Pd(PPh₃)₂Cl₂, THF-H₂O (4:1), 60 °C, overnight, 89%; (b) K₂CO₃, MeOH/DCM (1:1), rt, 2 h, quantitative; (c) 14 (for 33)/24 (for 34), CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, 41–51%; (d) 3 M NaOH/THF/MeOH (1:1:1), rt, overnight, 89% 10a, 74% 10b and 15% 10c.

Scheme 5. Preparation of Analogues of 2 in Which the Piperidine Moiety is Replaced by Various 1H-1,2,3-Triazol-1-yl-Linked Carbohydrates, Beyond Glucose

Reagents and conditions: (a) 3 M NaOH/THF/MeOH (1:1:1), rt, overnight, 80%; (b) CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, 57–80%; (c) a second method for 10g: tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA), CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (4:1), rt, overnight, 78%.

Scheme 6. Preparation of Analogues of 2 in Which the Piperidine Moiety is Replaced by C-Linked 1H-1,2,3-Triazol-4-yl-Linked Glucose (11a)

Reagents and conditions: (a) Na₂CO₃, Pd(PPh₃)₄, DME-H₂O (4:1), 80 °C, overnight, 39%; (b) CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, 91%; (c) 3 M NaOH/MeOH/THF (1:1:1), rt, overnight, 91%.

Scheme 7. Preparation of Analogues of 2 in Which the Phenylpiperidine Moiety is Replaced by Triazol-1-yl-Linked Glucose (12a), Glucuronide (12b), or Unsaturated Glucuronide (12c)

Reagents and conditions: (a) CuBr_2, H₂O, MeOH/THF (1:1), reflux, overnight, quantitative; (b) NIS, CuI, DMF/toluene (2:1), 110 °C, overnight, 89%; (c) trimethylsilylacetylene, [Pd(PPh₃)₂Cl₂], CuI, TEA, THF, rt, 24 h, 90% (for substrate 46); (d) K₂CO₃, MeOH/DCM (1:1), rt, 2 h, quantitative; (e) 14 (for 49)/24 (for 50), CuSO₄·5H₂O, sodium ascorbate, DMF/H₂O (9:1), 90 °C, overnight, 52–52%; (f) 3 M NaOH/MeOH/THF (1:1:1), rt, overnight, 90% 12a, 69% 12b and 16% 12c.

Figure 3

Representative fluorescence binding inhibition curves for three triazole-linked monosaccharide derivatives 10a, 10d, and 10g in CHO-hP2Y₁₄R cells (mean ± SEM, n = 3–5) (A). Comparison of affinity in high affinity series 5a–8a (extended piperidine moiety with an alkyl or PEG spacer that is triazole N-linked to the glucose moiety).

Figure 4

Probing the effects of glucose in the medium on the binding of fluorescent tracer 53, in whole hP2Y₁₄R-CHO cells determined using flow cytometry. (A) There is no effect on 53 binding at concentrations up to 10 mM d-glucose, which was chosen as the concentration for probing effect on glyco-conjugated antagonist ligand binding. At ≥ 0.2 M, a plateau at 42% of control binding was reached. (B) Inhibition curve by 8a in the presence of 10 mM d-glucose (IC₅₀ 82 nM).

Figure 5

Probing the effect of tunicamycin pretreatment of hP2Y₁₄R-CHO cells on cell viability (A) and on the binding of 11a using a fluorescent tracer 53, determined using flow cytometry in whole cells (B, C). (B) Compares 53 binding with and without tunicamycin pretreatment, and (C) shows the inhibition of fluorescent binding by antagonist 11a (IC₅₀ 2.0 nM without and IC₅₀ 4.5 nM with tunicamycin).

Figure 6

Activity of P2Y₁₄R antagonists (ref (2) and glycoconjugates 10a and 11a) in reducing BALF eosinophil content in a mouse model of asthma (dose: 20 μmol/kg, 9.8–12.2 mg/kg, i.p.); concentration: 2 μmol/mL (2 mM) in 10% DMSO, 30% PEG-400, 60% H₂O; administration: 10 μL/g of body weight. 7-Day sensitization with ovalbumin (OVA)/aspergillus (ASP) was followed by a single challenge (OVA, aerosol) at day 14. Antagonists were given 30 min prior to challenge. Harvesting of cells from the BALF and cell counting were on day 16. Methods used are reported in Karcz et al. and Wen et al. All groups are compared to ASP/OVA mice treated with vehicle alone using ordinary one-way ANOVA with Dunnett’s multiple comparison test. * P < 0.05, ** P < 0.01, and *** P < 0.001.

Figure 7

Activity of P2Y₁₄R antagonists in the CCI mouse model of chronic neuropathic pain. Two glycoconjugates, i.e., glucose conjugate 7a of intermediate linker length (A) and its corresponding glucuronide 7b (C), and several previously reported prodrugs were included for comparison. (B) shows the contralateral (control) paw. The antagonist was administered i.p. (10 μmol/kg) 7 days postsurgery. The activity of reference antagonist 2 and the two prodrugs (D, showing both prodrug moieties applied to each active drug) in the same model are found in Mufti et al. and Wen et al. Mean ± SD, for n = 3 (double prodrug of 3, MRS4738) and n = 4 (7a and double prodrug of 2). Two-tailed, two-way ANOVA with Dunnett’s comparisons: *P < 0.05 versus day 0 and ^†P < 0.05 versus day 7.