Cu(II) omplexes of cyclic pseudo-peptides: Carbonic anhydrase and phosphatase activity as possible metabolic functions of patellamides

Abstract

Ascidians like L. patella (in symbiosis with the cyanobacterium Prochleron) were found to produce cyclic pseudo-octapeptides. So far, the biological function of these macrocycles remains unclear. Their rigid structure provides a high preorganization for metal ion coordination and the accumulation of CuII in ascidians, together with recent findings in a number of studies that they preferably form dinuclear copper(II) complexes, suggest a metabolic role of copper(II) complexes of these macrocycles. For these reasons, the present thesis concentrates on the CuII solution chemistry of model systems of naturally occurring pseudo-octapeptides (patellamides and ascidiacyclamide) and their possible metabolic functions. The first two chapters give an introduction to the origin of the marine cyclic peptides and a brief overview of properties of the natural patellamides and their synthetic analogues, followed by a description of the aim of this thesis. The coordination chemistry of the (mono- and dinuclear) copper(II) complexes of the cyclic pseudo-peptides H4pat1-5 and H4LascA, synthetic analogues of the patellamides and ascidiacyclamide, is described in Chapter 3. A summary of the structural features and the so far studied metal coordination chemistry of the model systems is given at the beginning of the chapter (Chapter 3.1), followed by synthetic routes to the cyclic pseudo-peptides H4pat1-5 and H4LascA (Chapter 3.2). The investigations, leading to a comprehensive description with respect to the coordination chemistry involved, were based on the combination of spectroscopic methods (UV-vis-NIR and CD, Chapter 3.3) and completed by complex stability determination by ITC (Chapter 3.4). The stability constants of copper(II) complexes of the macrocycles H4pat1-5 have been measured by isothermal microcalorimetry (ITC). The measurements of all studied macrocycles show a high cooperative binding of two CuII ions. The overall stability constants are moderate (K≤106), but are in agreement with known stabilities of the natural ligands. The stability constants are slightly dependent on the ligand structure, and the one with side chain configuration identical to the natural products forms the most stable complexes (H4pat1). Despite the structural similarity, there are significant differences in the copper(II) coordination chemistry. The R*,S*,R*,S*-configuration of the isopropyl side chains of H4pat1 in combination with imidazole rings seems to rigidify the macrocycle. The high preorganization and the achirality leads to CD-inactive species. A number of observations and recent findings indicate the involvement of dicopper(II) complexes in fixation and hydrolysis of CO2. In Chapter 4 the dicopper(II) complexes of the six pseudo-octapeptides H4pat1-5 and H4LascA are shown to be efficient carbonic anhydrase model complexes with kcat between 1.7∙103 and 7.3∙103 s-1, determined by stopped-flow techniques. This means that there is an approximately 105-fold acceleration by the dicopper(II)-patellamide catalysts, which is only approximately two orders of magnitude slower than the rate of the enzymes. This is an interesting point because no copper-based natural carbonic anhydrases are known so far and no faster model systems have been described. The naturally observed R*,S*,R*,S*-configuration is shown to lead to more efficient catalysts than the S*,S*,S*,S*-isomers. The variation in rigidity seems to influence the reaction with CO2, as H4pat1 is the only cyclic pseudo-octapeptide studied, where no formation of a carbonato-bridged complexes could be observed. In addition, the catalytic efficiency also depends on the heterocyclic donor groups of the pseudo-octapeptides and, interestingly, the dicopper(II) complex of the ligand with four imidazole groups (H4pat1) is a more efficient catalyst than the close analogue of ascidiacyclamide with two thiazole and two oxazoline rings (H4LascA). The experimental observations indicate that the nucleophilic attack of a CuII-coordinated hydroxide at the CO2 carbon center is rate determining, i.e. formation of the catalyst-CO2 adduct and release of the product (carbonate or bicarbonate) are relatively fast processes. These observations also indicate that partial inhibition by product coordination can decrease the catalytic efficiency, where the macrocycles with natural configuration of the side chains of H4LascA and H4pat1 lead to little inhibition. In Chapter 5 another possible biological function of cyclic pseudo-octapeptides is presented. The efficient catalysis of CO2 hydrolysis indicated that other similar reactions might be possible, and phosphatase activity seemed to be a reasonable option. The dicopper(II) complexes of H4pat1 and H4pat2 have been tested with common phosphatase-substrates (BDNPP, DNPP). The maximum activity occurs around pH 7.2 and the kcat value of 4∙10-3 s-1 for H4pat1 as ligand is in a range found for rather reactive purple acid phosphatase (PAP) mimics. In contrast, H4pat2 as ligand shows an activity which is one order of magnitude lower than that of H4pat1. It appears that the conformational flexibility of the patellamide-dicopper(II) systems is the same reason for the lower activity of the dicopper(II) complex of H4pat2 as described above for CO2 hydrolysis (see Chapter 4). The observations indicate that a mechanism similar to that of PAP applies, and this is comparable to that for carbonic anhydrase.