DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 109–110 109 Introduction of ·-hydroxymethylserine residues in a peptide sequence results in the strongest peptidic, albumin-like, copper(II) chelator known to date Piotr M�ynarz,a Wojciech Bal,a Teresa Kowalik-Jankowska,a Marcin Stasiak,b Miros�aw T. Leplawyb and Henryk Koz�owski a* a Faculty of Chemistry, University of Wroc�aw, F. Joliot-Curie st. 14, 50-383 Wroc�aw, Poland. E-mail: henrykoz@wchuwr.chem.uni.wroc.pl b Institute of Organic Chemistry, Technical University, Zÿ eromski st. 116, 90-924 £ódz�, Poland Received 26th October 1998, Accepted 24th November 1998 A tripeptide amide HmS–HmS–His–NH2 is the strongest peptidic CuII chelator known to date, due to the steric shielding of the chelate plane as well as electronic effects.a-Hydroxymethylserine (HmS) is a non-proteinaceous amino acid, found as the N-terminal residue in antibiotic peptides, antrimicin1 and cirratiomycin.2 It diVers from serine by having another –CH2OH function at the a carbon.This substitution generates specific constraints on the conformational freedom of a peptide containing HmS, which is the likely reason for its existence. From the co-ordination point of view, the presence of two alcoholic functions in HmS enhances its binding abilities. The direct involvement of alcoholic functions in co-ordination was found for oxovanadium(IV) and copper(II) complexes of the HmS amino acid.3 Indirect, conformational phenomena also contributed to the stabilisation of particular complex species in di- and tri-peptides containing HmS residues.4,5 Peptides containing the N-terminal sequence Xaa–Yaa–His exhibit a particular aYnity towards CuII and NiII, resulting from the formation of three fused chelate rings and a flat fournitrogen (4N) co-ordination sphere around the metal ion.6,7 We have previously shown that the stability of such complexes can be influenced by conformational and electronic eVects resulting from substitutions of amino acids Xaa and Yaa.Substitutions of non-bonding Gly with bulkier Val and Ile provided a stability gain of two orders of magnitude, and the introduction of a positive Arg residue in position 1 was even more eVective.8,9 The study presented in this communication was aimed at finding out whether the presence of HmS residues can augment the binding capabilities of Xaa–Yaa–His peptides. The peptides, H–HmS–HmS–His–OH (1) and H–HmS– HmS–His–NH2 (2), were prepared using optimised methodology for incorporation of HmS into the peptide chain.10,11 Their co-ordination to CuII was studied by potentiometry and spectroscopy (UV/VIS, CD, EPR) in conditions analogous to those applied previously.5 Table 1 contains the stability constants (log b values) and spectroscopic parameters of complexes formed.Fig. 1 presents species distribution diagrams for these complexes, derived from potentiometric and spectroscopic measurements.Table 2 Table 1 Stability constants and spectroscopic characterisation of complexes formed by 1 and 2 UV/VISb CDb EPR Species log b a l (e) l (De) A|| c g|| HmS–HmS–His 1 HL H2L H3L CuL CuH21L CuH22L CuH23L 7.140(1) 13.176(1) 15.848(2) 7.66(1) 4.223(3) 0.064(2) 210.41(2) 510 510 (108) d (114) d 564 487 307 563 487 308 (20.31) d (10.73) d (10.80) e (20.32) d (10.74) d (10.79) e 210 210 2.17 2.17 HmS–HmS–His–NH2 2 HL H2L CuH21L CuH22L CuH23L 6.636(3) 12.322(3) 4.09(4) 1.271(7) 210.15(3) 510 510 (99) d (110) d 559 484 319 287 559 483 323 288 (20.26) d (10.69) d (10.06) f (20.31) g (20.29) d (10.74) d (10.06) f (10.39) g 207 210 2.18 2.18 a b(CuHiL) = [CuHiL]/{[Cu21][H1]i[L]}. Standard errors on the last digits are included in parentheses.Three titrations were performed for each system. b UV/VIS and CD units: l/nm, e/dm3 mol21 cm21, De/dm3 mol21 cm21. c EPR unit: A||/G. d d–d transition. e Nim aA CuII and N2 aA CuII CT transitions.f Nim aA CuII CT transition. g N2 aA CuII CT transition.110 J. Chem. Soc., Dalton Trans., 1999, 109–110 provides protonation-corrected stability constants for a range of complexes of Xaa–Yaa–His peptides that allow one to compare their metal binding capabilities directly. Spectroscopic parameters indicate that the binding mode in the 4N complex CuH22L, which predominates at pH 4–10 for both ligands, is identical to that of complexes of Gly–Gly–His and other Xaa– Yaa–His peptides.It involves the N-terminal amine (HmS-1), the amide nitrogens of HmS-2 and His-3 and the N-3 nitrogen of His-3 imidazole. There is no evidence for the direct involvement of alcoholic groups of HmS in the binding. The potential C-terminal donors, carboxyl in 1, or amide in 2, do not participate in CuII co-ordination as well. The dramatic stability gain of 3.3 log units vs. Gly–Gly–His, seen for the CuH22L species (Table 2), is likely to originate from the partial shielding of the CuII binding site from the bulk of solution both from above and below the co-ordination plane.Fig. 1 Superimposed species distributions for CuII and 1 (? ? ?) or 2 (——), calculated for CuII concentrations of 1 × 1023 mol dm23 and ligand concentrations of 1.2 × 1023 mol dm23. Fig. 2 CD spectra of CuH22L complexes of 1 (? ? ?) or 2 (——), recorded at pH 7.0. Concentrations used were 1.88 × 1023 mol dm23 (CuII) and 2.28 × 1023 mol dm23 (1), and 1.75 × 1023 mol dm23 (CuII) and 2.1 × 1023 mol dm23 (2).Table 2 Comparison of log *K values for the 4N complexes of X–X–His peptides with CuII Peptide Gly–Gly–His b Gly–Gly–His c Gly–Gly–His–OMec Gly–Gly–His–Gly–Gly c Gly–Gly–hist d Arg–Thr–His–Gly–Asn–NH2 e Arg–Thr–His–Gly–Asn–(15) f HmS–HmS–Hisg HmS–HmS–His–NH2 g log *Ka 216.43 216.14 214.97 214.59 217.14 214.24 213.13 213.12 211.04 a log *K = log b(CuH22L) 2 log b(H2L). b Ref. 6. c Ref. 12, 37 8C. d Ref. 14, hist stands for histamine. e Ref. 9. f Ref. 9, pentadecapeptide. g This paper. In this way, the access of water molecules to metal ion-bound amide nitrogens is limited, and the dissociation reaction is slowed. Complexes of Xaa–Yaa–His peptides composed of L-amino acids can provide such shielding only at one side. The NMR-derived solution structure of the NiII complex of Val– Ile–His–Asn, exhibiting this phenomenon, correlates with the stability gain of 2 log units.7 Quite surprisingly though, the amidation of the carboxylic function in 2 results in a further hundredfold increase of complex stability (to 5.4 log units vs.Gly–Gly–His), making it the strongest peptidic CuII chelator known to date. There is no reference data for appropriate amides (e.g. Gly–Gly–His–NH2) in the literature, but the stability constants for Gly–Gly–His–OMe and Gly–Gly–His–Gly– Gly indicate that the carboxylate charge neutralisation by means of esterification or peptide chain extension may increase the complex stability by ca.one log unit.12 The much bigger eVect seen in our complexes presumably results from more specific interactions. The only major spectroscopic diVerence between the complexes of 1 and 2 is the sign of the amide nitrogen to CuII CT, positive with 1, and negative with 2 (Fig. 2). Complexes of other Xaa–Yaa–His peptides studied previously 7–9 and of human and bovine serum albumins, sharing a similar metal binding site,13 exhibited a positive CT band. HmS residues are not chiral, so the sign of the CT band is governed by the conformation of the 6-membered chelate ring between the His amide and the imidazole nitrogens.However, any major conformational change in this ring ought to be reflected in the d–d bands, while this region of the spectra is practically identical for both complexes, indicating the unchanged metal ion environment. At this point we conclude that there is a specific interaction in the CuH22L complex of 2, which is probably related to its extremely high stability, but the reasons for this eVect remain to be elucidated.Further studies of the complexes presented above and of their nickel counterparts are currently in progress in our laboratory and will be reported soon. Acknowledgements This work was financially supported by the Polish State Committee for Scientific Research (KBN 3T09A-10514 and 3T09A- 11108), within the framework of the COST D8/0018/98 programme. References 1 N.Shimada, K. Morimoto, H. Naganawa, T. Takita, M. Hamada, K. Maeda, T. Takeuchi and H. Umezawa, J. Antibiot., 1981, 34, 1613. 2 T. Shiroza, N. Ebisawa, K. Furihata, T. Endo, H. Seto and N. Otake, Agric. Biol. Chem., 1982, 46, 865. 3 T. Kowalik-Jankowska, H. Koz�owski, K. Kocio�ek, M. T. Leplawy and G. Micera, Transition Met. Chem., 1995, 20, 23. 4 T. Kowalik-Jankowska, H. Koz�owski, M. Stasiak and M. T. Leplawy, J. Coord. Chem., 1996, 40, 113. 5 T. Kowalik-Jankowska, M. Stasiak, M. T. Leplawy and H. Koz�owski, J. Inorg. Biochem., 1997, 66, 193. 6 R. W. Hay, M. M. Hassan and C. You-Quan, J. Inorg. Biochem., 1993, 52, 17. 7 W. Bal, M. I. Djuran, D. E. Margerum, E. T. Gray, Jr., M. A. Mazid, R. T. Tom, E. Nieboer and P. J. Sadler, J. Chem. Soc., Chem. Commun., 1994, 1889. 8 W. Bal, G. N. Chmurny, B. D. Hilton, P. J. Sadler and A. Tucker, J. Am. Chem. Soc., 1996, 118, 4727. 9 W. Bal, M. Jez· owska-Bojczuk and K. S. Kasprzak, Chem. Res. Toxicol., 1997, 10, 906. 10 M. Stasiak, W. M. Wolf and M. T. Leplawy, J. Pept. Sci., 1998, 4, 46. 11 M. Stasiak and M. T. Leplawy, Lett. Pept. Sci., 1998, 5, 449. 12 R. Agarwal and D. Perrin, J. Chem. Soc., Dalton Trans., 1977, 53. 13 W. Bal, J. Christodoulou, P. J. Sadler and A. Tucker, J. Inorg. Biochem., 1998, 70, 33. 14 T. Gajda, B. Henry, A. Aubry and J.-J. Delpuech, Inorg. Chem., 1996, 35, 586. Communication