UV resonance Raman spectroscopy of the supramolecular ligand guanidiniocarbonyl indole (GCI) with 244 nm laser excitation

Ultraviolet resonance Raman (UVRR) spectroscopy is a powerful vibrational spectroscopic technique for the label-free monitoring of molecular recognition of peptides or proteins with supramolecular ligands such as guanidiniocarbonyl pyrroles (GCPs). The use of UV laser excitation enables Raman binding studies of this class of supramolecular ligands at submillimolar concentrations in aqueous solution and provides a selective signal enhancement of the carboxylate binding site (CBS). A current limitation for the extension of this promising UVRR approach from peptides to proteins as binding partners for GCPs is the UV-excited autofluorescence from aromatic amino acids observed for laser excitation wavelengths >260 nm. These excitation wavelengths are in the electronic resonance with the GCP for achieving both a signal enhancement and the selectivity for monitoring the CBS, but the resulting UVRR spectrum overlaps with the UV-excited autofluorescence from the aromatic binding partners. This necessitates the use of a laser excitation <260 nm for spectrally separating the UVRR spectrum of the supramolecular ligand from the UV-excited autofluorescence of the peptide or protein. Here, we demonstrate the use of UVRR spectroscopy with 244 nm laser excitation for the characterization of GCP as well as guanidiniocarbonyl indole (GCI), a next generation supramolecular ligand for the recognition of carboxylates. For demonstrating the feasibility of the UVRR binding studies without an interference from the disturbing UV-excited autofluorescence, benzoic acid (BA) was chosen as an aromatic binding partner for GCI. We also present the UVRR results from the binding of GCI to the ubiquitous RGD sequence (arginylglycylaspartic acid) as a biologically relevant peptide. In the case of RGD, the more pronounced differences between the UVRR spectra of the free and complexed GCI (1:1 mixture) clearly indicate a stronger binding of GCI to RGD compared with BA. A tentative assignment of the experimentally observed changes upon molecular recognition is based on the results from density functional theory (DFT) calculations.


S1
Table S1 contains the theoretical vibrational spectrum of GCI ethyl amide in the single protonated form calculated at the B3LYP-D3/6-311++G(d,p) and B2PLYP-D3/G-311++G(d,p) level of theory employing the Gaussian 16 program package. 1 All normal modes together with their wavenumber values and Raman activities are listed.  Figure S1: UV-vis absorption spectra of GCI, RGD, BisTrisBuffer and a 1:1 GCI-RGD mixture.

General information
All solvents were distilled before use. Millipore water was obtained with a TKA MicroPure ultrapure water system. All other commercially available reagents were used as obtained unless otherwise specified. The reactions were monitored by TLC on silica gel plates (Macherey-Nagel POLYGRAM SIL G/UV254) and spots were visualized by UV light (254 nm and 366 nm). Reversed phase column chromatography was performed with an Armen Instrument Spot Flash Liquid Chromatography MPLC apparatus with RediSep C-18 Reversed Phase columns. Lyophilisation was done with a Christ Alpha 1-4 LD plus freeze dryer. The melting points were obtained with a Büchi Melting-Point B-540 apparatus with open end glass capillary tubes. The melting points are not corrected. The IR spectra were measured on a Varian 3100 FT-IR Excalibur Series. The low resolution ESI mass spectra were recorded with a Bruker amaZon SL and the high resolution ESI mass spectra with a Bruker maXis 4G UHR-TOF. Analytical HPLC was performed on a Dionex HPLC apparatus that consisted of a P680 pump, an ASI-100 automated sample injector and an UVD 340U photodiode array detector with a YMC ODS-AQ column (column size: 150 × 3.0 mm, particle size: 5 μm, pore size: 12 nm). The NMR spectra were measured with Bruker DMX 300, AV NEO 400, DRX 500 or AVHD 600 spectrometers. All measurements were recorded at room temperature using DMSO-d6 as solvent. The chemical shifts are relative to the signals of DMSO-d6 (δ 1 H = 2.50 ppm and δ 13 C = 39.5 ppm). The apparent coupling constants are given in hertz (Hz). The description of the fine structure means: s = singlet, br. s = broad singlet, d = doublet, t = triplet, m = multiplet.

Synthesis
The GCI building block I was synthesized starting from commercially available methyl 3-amino-4-iodobenzoate (A) following a synthesis strategy inspired and adjusted from a previous work 2 . The building block I was further functionalized with ethyl amine to achieve GCI ethyl amide 2.
The GCP ethyl amide 1 3 was synthesized starting from literature-known GCP building block K 4 .

N-tert-Boc-guanidine (F)
The reaction was performed as described in the literature 4 . A solution of t-Boc2O (E, 12.0 g, 55.0 mmol, 1 equiv) in acetonitrile (100 mL) was added very slowly over 8 h at 0 °C under vigorous stirring to a mixture of guanidinium chloride (D, 26.3 g, 275 mmol, 5 equiv) in an aqueous sodium hydroxide solution (12.0 g, 0.3 mol NaOH in 50 mL water). The resulting suspension was stirred at room temperature for additional 20 h. The acetonitrile was evaporated in vacuo and then 100 mL water was added. The aqueous suspension was extracted with ethyl acetate (3 times with 100 mL). The combined organic phases were washed with brine (3 times with 100 mL), dried (MgSO4), and evaporated in vacuo. The resulting white crystals were dried to yield 7.66 g (87%) of analytically pure guanidine F.

GCI ethyl amide 2
To a solution of Boc-GCI ethyl amide J (221 mg, 0.592 mmol, 1 equiv) in DCM/TFA (15 mL each), was added and the solution was stirred at room temperature for 3 h. The solvent was evaporated in vacuo to receive the off-white crude product, which was purified by flash chromatography (RP 18 MeOH/H2O + 0.1% TFA, 10% MeOH + 0.1% TFA to 100% MeOH + 0.1% TFA, gradient) and treated several times with 1 M HCl with respective solvent removal to give the chloride salt 2 (103 mg, 0.333 mmol, 56%) as white solid with a purity of 97% (HPLC).             Figure S21: HR-ESI mass spectrum of C (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to C. Figure S22: HR-ESI mass spectrum of F (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to F. Figure S23: HR-ESI mass spectrum of G (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to G. Figure S24: HR-ESI mass spectrum of H (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to H. Figure S25: HR-ESI mass spectrum of I (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to I. Figure S26: HR-ESI mass spectrum of J (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to J. Figure S27: HR-ESI mass spectrum of 2 (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to 2. Figure S28: HR-ESI mass spectrum of L (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to L. Figure S29: HR-ESI mass spectrum of 1 (positive ion mode, MeOH) and predicted mass spectrum of peaks which belongs to 1. Figure S30: Analytical HPLC (RP 18 MeOH/H2O + 0.1% TFA, 10% MeOH + 0.1% TFA to 100% MeOH + 0.1% TFA, gradient) 2. Figure S31: Analytical HPLC (RP 18 MeOH/H2O + 0.1% TFA, 10% MeOH + 0.1% TFA to 100% MeOH + 0.1% TFA, gradient) 1.