Inducing a pH-dependent conformational response by competitive binding to Zn2+ of a series of chiral ligands of disparate basicity

Molecules that change shape in response to environmental conditions are central to biological molecular communication devices and their synthetic chemical analogues. Here we report a molecular system in which a series of chiral anionic ligands of differing basicity are selectively protonated according to the pH of the medium. A cationic circular dichroism (CD) reporter complex responds to anion binding by selecting one of two alternative enantiomeric conformations. Exploiting the principle that less basic anions have, in general, weaker electrostatic interactions than more basic anions, a set of three chiral acids with large (>5 unit) pKa differences and differing configurations were sequentially deprotonated in acetonitrile by addition of base, allowing the most basic anion in the mixture at any time to bind to the reporter complex. A characteristic CD output resulted, which changed in sign as the next-most basic anion was revealed by the next deprotonation in the series. Four cycles of switching between three ligand-bound states were achieved with minimal changes in signal magnitude, by alternating addition of base and acid. The pH-dependent conformational response was used to transduce a signal by appending to the binding site a 2-aminoisobutyric acid (Aib) oligomer, whose M or P helical conformation depended on the chirality of the bound ligand, and was reported by a remote 13C-labelled NMR reporter group. The multicomponent system thus converts a pH signal into a programmable conformational response which induces a remote spectroscopic effect.

The optical purity of the salt was determined by first Cbz protection of the amine, then coupling to H-Phe-O t Bu (see below). This gave a mixture of diastereomers, the diastereomeric ratio of which could be quantified by integration of the 13 CH 3 peaks in the 13 C NMR spectrum. The d.r. was determined to be 80:20 in this way, therefore HBr.H-(R)-Aib*-OH was prepared in an 80:20 (R:S) ratio.

BL-28 Cbz-Aib*-OH
To a solution of HBr.H-(R)-Aib*-OH (505 mg, 2.73 mmol) and Na 2 CO 3 (868 mg, 8.19 mmol) in water (9.5 mL), a solution of benzyl chloroformate (0.78 mL, 5.46 mmol) in 1,4-dioxane (5.5 mL) was added at 0 °C over 15 min. The mixture was stirred at 0 °C for 1.5 h, before it was allowed to warm to room temperature and stirred for a further 20 h. The reaction mixture was washed with Et 2 O (2 x 15 mL), acidified with HCl (conc.) to pH=1, and extracted with EtOAc (2 x 20 mL). The combined organic phases were dried (MgSO 4 ), filtered and concentrated. The crude residue was purified by column chromatography (P.E.:EtOAc, 1:1) to give the title compound (751 mg, 81%) as a colourless solid. Spectroscopic data are consistent with those previously reported. 12

S3.2 Base titrations and formula for [θ]
An initial sample (c o = 0.25 mM) was prepared in a 1 mm path length quartz cuvette from Where θ λ is the ellipticity measured at wavelength λ, c is the concentration of Zn(BQPA) in the sample and l is the path length of the cuvette. In practice, it is more convenient to rewrite c in terms of Zn(BQPA) stock solution concentration: Where c' is the concentration of the Zn(BQPA) stock solution, V is the total volume of the sample and v is the volume of Zn(BQPA) stock solution used in the sample.

S3.3 Base titration to estimate concentration of (+)-TADDOL-PONHTf stock solution
A base titration was performed using a freshly prepared iii stock solution of (+)-TADDOL-PONHTf ( was the x-coordinate of the hypothetical signal maximum.
iii A precipitate formed within a couple of days in stock solutions of 5-H under ambient conditions. Storage in the freezer slowed, but did not halt, this process. repeated as desired, with the CD spectrum recorded after each addition of acid or base.

S3.5 Binding constant measurement titration
An initial sample (c o = 0.25 mM) was prepared in a 1 mm path length quartz cuvette from  (Fig. S2), to prove that any pHinduced changes in the observed CD spectrum were solely due to ligand binding or exchange at the Zn(1) receptor complex. All of these ligand species had molar ellipticities of around 0 at 240 nm, the characteristic output wavelength of Zn(1), so any structural changes in the ligands had negligible contribution to the overall CD spectrum of the reported CD switching systems. revealed an almost complete absence of a positive signal at 240 nm expected for the Boc-L-prolyl backbone (Fig. S3a). This was not due to a lack of binding since the negative lobe was still present at 232 nm. Hydroxamic acids function as bidentate metal chelators in nature, 14  The base titration was also performed with thioamide 6-H, but using triazabicyclodecene (TBD) as a more powerful base (pK a (MeCN) = 26.02, cf. 18.83 for NEt 3 ). 15 6-H gave a very weak positive signal at 240 nm upon addition of TBD (Fig. S3b). The signal magnitude reached a maximum of only 1000 deg dm 2 mol -1 at 1 eq. of base, only 13% of the maximum signal magnitude achieved with 1 eq. Boc-D-Pro-OH, which had the same chiral backbone. Further addition of TBD caused a reduction in signal magnitude, indicating that deprotonation of the ligand was likely complete around 1 eq. of base.  Acidity center C-OH