Our first interest was in investigating to which extent the catalytic efficiency of peptide A was affected by small structural changes, in particular regarding the distance between the anionic tail and the His-residue. For that reason, two analogues of A were synthesized with the His-residue immediately attached to the DDD–fragment (B) or separated by two amino acids (C). In all sequences, also, a Trp-residue was added as a fluorescent probe to facilitate detection of peptide binding to the nanoparticles. Peptide D was added to study whether the Trp-spacer in the original peptide A was of importance. All peptides were synthesized on solid support using Fmoc-chemistry and, after cleavage, purified by RP-HPLC and characterized by UPLC and MS (see Supporting Information).

The complexation of peptides A–D on the monolayer surface of Au MPC 1 was studied by fluorescence titrations as described before under identical experimental conditions ([R-N(CH₃)₃⁺] = 60 ± 10 μM; H₂O:CH₃CN = 90:10; [HEPES] = 10 mM; pH = 7.0; T = 37 °C) [21]. Also previously studied, peptide A was added in order to have a data set performed with the same batch of nanoparticles [21]. The resulting binding isotherms (Figure 2a) reveal that all peptides bind the cationic monolayer under saturation conditions. Surface saturation concentrations (SSCs) were determined as reported before and are listed in Table 1[21]. All peptides have similar SSCs in the range between 7.8 (D) and 9.5 (A) [23]. Subsequently, kinetic studies were performed to determine the efficiency of the catalytic systems in accelerating the hydrolysis of the substrate I (Figure 2b). Kinetic experiments were performed by adding the substrate (10 μM) to a solution of Au MPC 1 and the peptides A–D in a H₂O:CH₃CN (9:1) mixture buffered at pH 7.0 at 37 °C. Peptides A–D were present at concentrations corresponding to 50% of their respective surface saturation concentrations, since we had previously shown that these conditions were favorable for catalysis (see above) [21]. From the kinetic profiles pseudo-first order rate constants, k_obs, were obtained by fitting the experimental data to an appropriate model. In order to permit a direct comparison between the different peptides, the k_obs-values were corrected for the peptide-concentration to yield the second order rates constants, k₂ (Table 1). It is noted that in the absence of Au MPC 1, none of the peptides displayed any significant activity (data not shown). Analysis of the data shows that the distance between the His-residue and the anionic tail is not crucial with peptides A and C, having nearly identical rate constants (k_2,A = 2.1 × 10³ M⁻¹·s⁻¹ and k_2,C = 1.7 × 10³ M⁻¹·s⁻¹). In fact, peptide B, which does not even have a spacer between the His-residue and the DDD-tail, has a second-order rate constant that is the lowest (k_2,B = 0.8 × 10³ M⁻¹·s⁻¹). Rather, the sequence order of the catalytic peptide seems of more importance, i.e., the presence of a Trp-residue preceding the His-residue (HisTrp as in A and C compared to TrpHis in B and D). Previous studies have recently demonstrated that hydrophobic units (such as Trp) present in the probes can be involved in binding to the monolayer, because of interactions with the hydrophobic interior part of the monolayer [24,25]. The obtained data suggest that such an interaction may favor catalytic activity by affecting the relative position of the His-residue compared to the cationic head groups. It is noted, though, that the observed effect on catalytic activity is relatively small.

We next proceeded with the synthesis and study of a series of peptides aimed at determining whether the catalytic activity of the His-residue could be affected by flanking residues. For that purpose, peptides E–H were synthesized with the consensus sequence Ac-XHXWDDD-OH (Figure 1). The His-residue was positioned between Leu (E)-, Phe (F)-, Ser (G)- and Tyr (H)-residues in order to explore different local chemical environments. All peptides were synthesized on solid support using Fmoc-chemistry and, after cleavage, purified by RP-HPLC and characterized by UPLC and MS (see Supporting Information). As before, the complexation of peptides E–H on the monolayer surface of Au MPC 1 was studied by fluorescence titrations (Figure 3a). From the obtained curves, it is immediately evident that the chemical nature of the catalytic tail significantly affects the SSC (Table 2). The highest SSC (15.9 ± 0.2 μM) is observed for peptide H containing two Tyr-residues. Peptides E and F have intermediate SSC values of 8.6 ± 0.2 μM and 9.1 ± 0.2 μM, respectively, whereas peptide G with polar Ser-residues has a significantly lower SSC (4.4 ± 0.2 μM ). The shallow curve observed for peptide G is a sign of the lower affinity of this peptide for the monolayer surface. Kinetic experiments were performed as described earlier using substrate I (Figure 3b), and the results are given in Table 2. The obtained data (k²-value, i.e., corrected for the peptide concentration) clearly show an effect of the peptide sequence on catalytic activity. In particular, a higher activity is observed for the His-residue in peptide G surrounded by Ser-residues. Tentatively, this might be explained by an enhanced stabilization of the polar transition state of the reaction by the polar peptide sequence. One could argue that the lower SSC of peptide G causes an enhanced activity of the system, because of a lesser inhibition of the intrinsic catalytic activity of Au MPC 1. However, the intrinsic activity of Au MPC 1 (k_obs = 1.0 × 10⁻³ s⁻¹, see below) is too low to explain the observed difference.

A series of substrates were tested in order to assess the relation between substrate structure and catalytic efficiency. Apart from the reference substrate used so far (I), substrates II–VI were selected. All substrates are p-nitrophenyl activated carboxylic acids, which ensures reactivity and permits a straightforward measurement of the reaction kinetics by UV-Vis spectroscopy. In addition, the transesterification reaction of II can also be studied by fluorescence, since hydrolysis removes the p-nitrophenyl group, which is a quencher of the Trp-fluorescence. Both substrates II and III provide information on the importance of the side chain substituent. In addition, both the d- and l-enantiomers of III (III_L and III_D) were studied in order to detect possible enantioselectivity. Substrate IV carries an acetyl group instead of the CBz-group and, finally, substrates V and VI are simple, non-peptidic substrates of different polarity. Catalytic peptide A was used as the reference catalyst.

First, the background hydrolysis rate was measured for each new substrate by measuring the increase in absorption at 400 nm as a function of time of a solution of the respective substrate (10 μM) in a 9:1 H₂O:CH₃CN mixture buffered at pH = 7.0 at 37 °C ([HEPES] = 10 mM) (Figure 4a). The concentration of 10 μM is set to avoid solubility problems. None of the substrates showed significant degrees of hydrolysis over a period of 1 hour, except for substrate IV, which was completely hydrolyzed within 30 min (k_obs = 2.5 × 10⁻³·s⁻¹). For the other substrates, first-order rate constants were determined by fixing the known end value of 10 μM (Table 3). The addition of catalyst A (4.8 μM) without nanoparticles did not significantly affect the hydrolysis rate of the substrates (Figure 4b). On the other hand, similar as in our previous studies [21], in all cases the addition of Au MPC 1 by itself caused a rate acceleration (except for substrates V and VI), but this intrinsic contribution could be suppressed by saturating the monolayer surface with catalytically inert peptide Ac-WDDD-OH (11 μM) (Table 3) (Figure 4c). The latter situation serves as a reference system to assess the actual rate acceleration originating from the catalytic peptide A on the surface of Au MPC 1 (Figure 4d). The addition of catalyst A (4.8 μM corresponding to 50% of the SSC) in the presence of Au MPC 1 resulted for all substrates in a significant rate acceleration (Table 3).

First of all, the data obtained for substrate IV show that the CBz-protecting group serves to prevent a rapid auto-hydrolysis of the substrate. However, even for substrate IV, a catalytic effect of Au MPC 1•A is observed (k_obs,1•A = 4.2 × 10⁻³·s⁻¹ against k_obs,_1•WDDD = 2.5 × 10⁻³·s⁻¹), although relatively small. The observation that the strongest rate enhancement is observed for substrates I–III suggests that the CBz-group plays a role in binding of the substrate to the monolayer. Indeed, the rate acceleration observed for substrate IV lacking the CBz-group is much lower even if just Au MPC 1 is added. Although structurally very simple, the small substrates V and VI are interesting, because their hydrolysis is not catalyzed at all by Au MPC 1 alone, suggesting a much lower affinity of these substrates for the monolayer. In the presence of the catalytic system, Au MPC 1•A, a rate acceleration is observed, but the absolute rates are still much lower compared to the larger apolar substrates I–IV. Nonetheless, also in this case, the effect is stronger on substrate VI, which has a larger hydrophobic tail compared to III. In summary, the catalytic activity of Au MPC 1•A is not restricted to substrate I, but appears to be general for substrates with a significant hydrophobic component. The identical rate constants obtained for enantiomeric substrates III_L and III_D show that the system is not enantioselective at all.

More information on the catalytic system was obtained from a detailed study of substrate II. As stated before, the hydrolysis kinetics of substrate II can be followed simultaneously by UV-Vis and fluorescence spectroscopy, because hydrolysis results in the formation of a chromogenic (p-nitrophenolate) and fluorogenic (CBz-Tryp-OH) product. Thus, the hydrolysis of substrate II was followed simultaneously by monitoring the fluorescence intensity at 360 nm (CBz-Tryp-OH) and the UV absorption at 400 nm (p-nitrophenol). Four catalytic cycles were performed by sequentially adding 10 μM of substrate II. As illustrated in Figure 5a, the kinetic profiles give a perfect match, which, together with the observed turnover of the system, is definitive proof of the catalytic activity of Au MPC 1•A. The slightly lower rate constants obtained from the fluorescence trace may be a result of product binding to the monolayer. In fact, it is observed that the rate constants show a constant drop after each cycle. This is indicative of the accumulation of an inhibitor in the system, which disfavors formation of the catalytic system composed of Au MPC 1, catalytic peptide A and substrate II. The most likely candidate is the carboxylic acid, CBz-Tryp-OH, as it combines a negative charge with a hydrophobic unit. Indeed, a direct fluorescence titration of CBz-Tryp-OH to Au MPC 1 yielded a curve that deviates from linearity in the 0–10 μM concentration regime, which indicates some weak affinity for the monolayer (Figure 5b). Most likely, the accumulation of the reaction product CBz-Tryp-OH in the system affects the equilibrium between Au MPC 1 and substrate II, as the latter is the weakly bound component in the system.