Photoswitchable Diazocine Derivative for Adenosine A3 Receptor Activation in Psoriasis

Abstract

Incorporating photoisomerizable moieties within drugs offers the possibility of rapid and reversible light-dependent switching between active and inactive configurations. Here, we developed a photoswitchable adenosine A₃ receptor (A₃R) agonist that confers optical control on this G protein-coupled receptor through noninvasive topical skin irradiation in an animal model of psoriasis. This was achieved by covalently bonding an adenosine-5'-methyluronamide moiety to a diazocine photochrome, whose singular photoswitching properties facilitated repeated interconversion between a thermally stable, biologically inactive Z agonist form and a photoinduced, pharmacologically active E configuration. As a result, our photoswitchable agonist allowed the precise modulation of A₃R function both in vitro and in vivo, which led to a clear light-controlled pharmacotherapeutic effect on mouse skin lesions. This breakthrough not only demonstrates the potential of diazocine photoswitches for in vivo photopharmacology but also paves the way for the development of new strategies for skin-related diseases that require localized and temporally controlled drug action.

Figure 1

Chemical structures of Piclidenoson, MRS7344, and MRS7787.

Figure 2

Docking of MRS7787 stereoisomers to a human A₃R model. (A) Docking pose of IB-MECA (cyan) at the hA₃R model obtained through induced fit docking. (B–C) Docking pose of an E-MRS7787 (orange) and Z-MRS7787 (green), respectively, at the optimized hA₃R model. Hydrogen bonds are represented by yellow dashed lines, and clashes by orange dashed lines.

Figure 3

Chemical synthesis of MRS7787 (4). (A) Conditions and reagents: (i) 4 N HCl in dioxane, CH₂Cl₂, rt, 1 h; (ii) 2, DIPEA, 2-propanol, reflux, overnight. (B) Photoisomerization process between the Z and E states of MRS7787.

Figure 4

Photochemical characterization of MRS7787. (A) Variation of the absorption spectra of Z-MRS7787 in PBS:DMSO 98:2 upon irradiation at 405 nm (30 s increments at 6.5 mW cm^–2) and until a photostationary state (PSS_Z–E) is obtained (c_MRS7787 = 71 μM). (B) Variation of the absorption spectra of the PSS_Z–E of MRS7787 in PBS:DMSO 98:2 upon illumination at 532 nm (60 s at 10 mW cm^–2) (c_MRS7787 = 71 μM). For sake of comparison, the spectrum of the initial Z-MRS7787 compound is shown, which matches the spectrum of the PSS_E–Z achieved after irradiation at 532 nm. (C) Absorbance variation at 391 (λ_{max,Z-MRS7787}) and 470 nm (λ_{max,E-MRS7787}) upon 10 consecutive cycles of Z–E photoisomerization of MRS7787 in PBS:DMSO 98:2 under sequential irradiation at 405 nm (30 s at 6.5 mW cm^–2; violet) and 532 nm (60 s at 10 mW cm^–2; green) (c_MRS7787 = 71 μM).

Figure 5

Photomodulation of the intrinsic activity of MRS7787 in living cells. HEK-293 cells expressing A₁R, A₃R, A_2AR, and A_2BR were challenged with increasing concentrations of MRS7787 irradiated with 420 nm (blue symbols) or 520 nm (green symbols) before cAMP accumulation was determined (see Supporting Information). In Gi stimulation experiments (A), forskolin (direct activator of adenylate cyclase) treatment induced maximum cAMP accumulation (black dashed line) and treatment with MRS5698 and CPA (N⁶-cyclopentyladenosine, selective A₁R agonist) induced the maximum A₃R- and A₁R-mediated inhibition of cAMP accumulation (respectively green dashed line). In Gs stimulation experiments (B) treatment with CGS21680 and NECA induced maximum stimulation of cAMP accumulation in cells expressing A_2AR and A_2BR, respectively (green dashed line). Treatment with SCH442416 (selective A_2AR antagonist) and PSB603 (selective A_2BR antagonist) determined the blockade of A_2AR, and A_2BR, respectively. Data are expressed as the mean ± SEM of three independent experiments performed in quadruplicate. ***P < 0.0001, F_(6,238) = 80 when all nonlinear regression EC₅₀ and E_max values from were compared.

Figure 6

MRS7787 photoactivates A₃R in a mouse model of psoriasis. (A) The temporal drug treatment of the IL-23-induced psoriasis model is explained in Figure S5 (Supporting Information). IL-23 treated (i.d., intradermal) mice were administered (i.p., intraperitoneal) with vehicle (red triangles) or MRS5698 (1 mg/kg; black triangles). IL-23 administered animals were treated with MRS7787 (1 mg/kg) in the absence (circles) or presence (squares) of MRS1523 before both ears were irradiated with 420 nm (blue line) or 520 nm (green line) for 8 min. Ear thickness was measured in millimeters (mm) and shown as mean ± S.E.M., n = 7–13 mice per group. ****P < 0.0001 one-way ANOVA with Dunnett’s posthoc test comparing to MRS5698. (B) Representative H&E-stained ear sections of IL-23-treated mice from the indicated experimental group. Scale bar = 500 μm.