Photoisomerization of Azobenzene-Extended Charybdotoxin for the Optical Control of Kv1.2 Potassium Channel Activity

Abstract

Natural peptides from animal venoms effectively modulate ion channel activity. While photoswitches regulate small compound pharmacology, their application to natural peptides rich in disulfide bridges and active on ion channels is novel due to larger pharmacophores. A pilot study integrating azobenzene photoswitches into charybdotoxin (ChTx), known for blocking potassium channels is initiated. Two click-chemistry-compatible azobenzene are synthesized differing in length and amide orientation (Az₁ & Az₂). Az₁ is grafted onto ChTx at various amino acid positions using L-azidohomoalanine mutation. ChTx monomers outperformed dimers, particularly with azobenzene at position 14, by exhibiting optimal photoswitching activity. In the cis configuration, Az₁ altered ChTx's pharmacophore, reducing potassium channel blockage, while conversely, Az₂ increased ChTx potency. This study pioneers photoswitch application to complex peptides, leveraging structure-activity relationships. Successful integration depends on precise azobenzene positioning and chemical grafting guided by SAR insights. This advancement underscores the adaptability of photoswitch technology to intricate peptide structures, offering new avenues for pharmacological modulation.

Keywords: Azobenzenes; Click chemistry; Ion channels; Photopharmacology; Photoswitches.

Figure 1

Chemical properties of a click chemistry‐compatible azobenzene (Az₁). a) Synthesis pathway of Az₁. b) Isomerization reaction of Az₁ promoted by UV light (trans to cis) and visible light (cis to trans) irradiation at a 9.5 mW cm⁻² light intensity in 100% DMSO. Amide bonds are all in the trans configuration. c) Photostationary states of Az₁ in the dark, after illumination at 365 or 340 nm and upon back‐switching by illumination at 435 nm as assessed by HPLC. d) Cycles of illuminations to probe the Az₁ fatigue.

Figure 2

Strategy for the positioning of Ah substitution on ChTx. a) Primary structure of ChTx illustrating the disulfide bridging pattern and the critical residues of the pharmacophore (in red). Residues in yellow and green are defined as in (b) depending on location compared to the pharmacophore. b) 3D‐structure of K_v1.2 channel with docked ChTx on the ionic pore in top and side views. The pharmacophore of ChTx is highlighted in red. Yellow residues of ChTx are in close proximity to K_v1.2 structures, while green residues are those that are the most distant from the channel amino acid residues. The position of the residues to be substituted individually by Ah are illustrated (T⁹ and W¹⁴ from the yellow category, closest to the channel structure, and E¹² and R¹⁹ from the green category).

Figure 3

Production and purification of the ChTx‐Ah¹⁴‐Az₁ monomer and the ChTx‐Ah¹⁴‐Az₁‐Ah¹⁴‐ChTx dimer. a) Schematic 3D representation of the monomer and the dimer. The pharmacophore is given in red with K²⁷ being the residue that plunges into the K_v1.2 channel pore. The pharmacophore comprises T²³, R²⁵, G²⁶, M²⁸, N²⁹, R³⁴ and Y³⁶ besides K²⁷. The size and length of the Az₁ linker in the trans configuration are proportionate. b) Click chemistry conjugation between ChTx‐Ah and Az₁. c) Left panel: elution profile and [M + 5H]⁵⁺ MS (inset) of ChTx‐Ah¹⁴‐Az₁ monomer after click chemistry conjugation in the 0.33 Az₁/1 ChTx‐Ah¹⁴ stochiometric ratio. A small fraction of the cis isomer is also visible. Right panel: elution profile and [M + 9H]⁹⁺ MS of ChTx‐Ah¹⁴‐Az₁‐Ah¹⁴‐ChTx dimer after click chemistry conjugation in the 3 Az₁/1 ChTx‐Ah¹⁴ stochiometric ratio. The cis isomer is not detectable possibly because less stable in the case of the dimer.

Figure 4

Isomerization properties of the ChTx‐Ah¹⁴‐Az₁ monomer and ChTx‐Ah¹⁴‐Az₁‐Ah¹⁴‐ChTx dimer. a) the kinetics of isomerization of the ChTx‐Ah¹⁴‐Az₁ monomer, schematized on the left panel. Purple color represents cis isomer, while green represents trans isomer. b) Kinetics of the isomerization. Half‐isomerization times (t_1/2) were extracted. c) Stability of the cis conformation of the ChTx‐Ah¹⁴‐Az₁ monomer over a period of 24 h. d) Photostationary states of ChTx‐Ah¹⁴‐Az₁ in the dark, after illumination at 365 nm and upon back‐switching by illumination at 435 nm, as assessed by HPLC. The y‐axis is the absorbance at 214 nm in arbitrary units. e–h) as for a–d) but for the ChTx‐Ah¹⁴‐Az₁‐Ah¹⁴‐ChTx dimer. The absorbance for the dimer in (e) is about half the absorbance value observed in (a) for the monomer because a similar concentration of peptide was used in both conditions which implies half a concentration of Az₁ in the dimer. In (h) the exact percentage of cis in dark and PSS₄₃₅ conditions cannot be determined because of the proximity of the elution peaks. All peptide/Az₁ conjugates are dissolved in water.

Figure 5

Structural consequences of Az₁ cycloaddition onto ChTx‐Ah¹⁴ and trans → cis isomerization. a) Comparison of ChTx‐Ah¹⁴ and ChTx‐Ah¹⁴‐Az₁ 3D structures. Variations in lateral chain positions were observed for Q¹⁸, S²⁴, R²⁵, G²⁶, K²⁷, Y³⁶ and S³⁷ upon Az₁ attachment (in yellow, right panel). The position of the pharmacophore is shown for comparison on ChTx‐Ah¹⁴ alone (red color, left panel). b) 1D ¹H‐NMR spectra illustrating the time‐dependent evolution of d‐Y³⁶ and the e‐Y³⁶ signals upon illumination at 365 nm of ChTx‐Ah¹⁴‐Az₁, a signature of the cis conformation, and upon illumination at 435 nm or upon back thermal relaxation over 24 h in the dark. The incomplete back thermal relaxation is coherent with the data in Figure 4c. c) Peptide models illustrating the Az₁ reorganization at the surface of ChTx‐Ah¹⁴‐Az₁ upon trans to cis switching. In the cis configuration of Az₁, the free end orients itself toward Y³⁶ (purple color in cis). F² position is not affected by the isomerization. Drawn by ChimeraX.

Figure 6

Functional impact of Az₁ isomerization at the surface of the ChTx‐Ah‐Az₁ monomers on K_v1.2 currents. Purple traces are for trans conformation and green for cis. a) Comparison of K_v1.2 current block by 100 nm ChTx‐Ah¹⁴‐Az₁ in trans‐state vs cis‐favored state. Peak currents are measured by convention. b) Concentration‐response curves for ChTx‐Ah¹⁴‐Az₁ in the two isomerization states. For ChTx‐Ah¹⁴‐Az₁ in cis‐favored configuration, two IC₅₀ values are measured with 22.5 nm (n = 64; 49% trans) and 757 nm (n = 64, 51% cis). In all‐trans configuration, the IC₅₀ = 22.5 nm (n = 78). For these experiments, illumination at 365 or 435 nm was maintained before and during the application of the peptides to minimize the impact of channel interaction on unwanted isomerization.

Figure 7

The new Az₂ photoswitch provides dynamic modulation of K_v1.2 currents. a) Synthesis pathway of Az₂. b) Isomerization reaction of Az₂ promoted by illumination at 9.5 mW cm⁻² (λ = 385 nm for trans to cis and λ = 490 nm for cis to trans) in 100% DMSO. Left panel: traces at 10 and 30 s are superimposed with the trace at 60 s. Right panel: trace at 60 s is superimposed with trace at 90 s. c) Cycles of illuminations to probe the Az₂ fatigue. d) Stability of the cis conformation of Az₂ over a period of 24 h. e) Concentration‐response curves for ChTx‐Ah¹⁴‐Az₂ in the two isomerization states. Measured IC₅₀ values are 37.6 nm (n = 41; nH = 0.9; trans) and 8.7 nm (n = 40; nH = 0.6; cis). f) Average K_v1.2 current amplitude variation (n = 10) induced by cycles of irradiation at 385 nm (purple) and 490 nm (green). g) Representative K_v1.2 current traces at 10 nm in the dark and at time points as indicated in (f). Scale bars: 500 ms/1 nA. h) Quantification of the changes in K_v1.2 current amplitude for the individual cells as a function of wavelength in the presence of 10 nm ChTx‐Ah¹⁴‐Az₂ (n = 10 cells studied).