FLUORESCENCE QUENCHING zyxwvutsrq OF SUBTILISIN ENZYMES Quenching of the Tyrosyl and Tryptophyl Fluorescence of Subtilisins Carlsberg and Novo by Iodidet Michael F. Brown,t Siraj Omar, Rodger A. Raubach, and Thomas Schleich* ABSTRACT: The tyrosyl and tryptophyl fluorescence of di- isopropylphosphorylsubtilisins Carlsberg and Novo, respec- tively, is quenched efficiently by zyxwvutsrq I- but is not significantly affected by Cs+. The I- quenching data were analyzed using a modified Stern-Volmer treatment (Lehrer, zyxwvutsrq S. S. (1971), Biochemistry IO, 3254), yielding values for the effective fraction of accessible protein fluorescence of 90-95 and 88- 92% for the tyrosyl and tryptophyl emission of diisopropyl- phosphorylsubtilisins Carlsberg and Novo, respectively. Similar values were obtained at pH 5 and 7. The effective collisional quenching constant depends on pH in a manner suggesting the participation of protein surface charge in the quenching mechanism. Significant singlet energy transfer (efficiency zyxwvutsrqpo = 0.52) from tyrosyl to tryptophyl residues was A m i n o acid substitutions in homologous proteins can be used to advantage in spectroscopic studies which employ particular residues as intrinsic probes of structural detail. On the basis of circular dichroism (CD)' and secondary structure predictions, the details of polypeptide chain folding, internal side chain packing, and hydrogen bonding are thought to be conserved among the homologous subtilisins, with the excep- tion of a sequence region containing the Pro-56 deletion in subtilisin Carlsberg (M. F. Brown and T. Schleich, manuscript in preparation). Although the crystal structure of subtilisin Carlsberg is not known, subtilisins Novo2 and Carlsberg have been studied in solution using CD (Myers and Glazer, 197 1 ; Brown and Schleich, 1975; M. F. Brown and T. Schleich, manuscript in preparation), infrared spectroscopy (Johansen and Ottesen, 1974), fluorescence spectroscopy (Longworth, 1971; Schlessinger et al., 1975), and solvent perturbation techniques (Myers and Glazer, 197 1; Herskovits and Fuchs, 1972). These studies support extensive conformational homology among the subtilisins, although differences in the environments of the aromatic residues are evident. In this paper we present a study of the I- quenching of the tyrosyl and tryptophyl fluorescence of inactivated subtilisins + From the Division of Natural Sciences, University of California, Santa Cruz, California 95064. zyxwvutsrqpo Receioed September zyxwvutsrqp 16, 1976. This work was supported by Grant No. BMS 75-17114 from the National Science Foundation. This work constitutes part of a thesis submitted in partial fulfillment of the requirements for the Ph.D. degree of M.F.B. at the University of California, Santa Cruz. * Present address: Department of Biophysical Chemistry, Biocenter of the University of Basel, CH-4056 Basel, Switzerland. I Abbreviations used: CD, circular dichroism; Dip-F, diisopropyl fluorophosphate; Dip, diisopropylphosphoryl; Gdn-HCI, guanidine hy- drochloride; PhCH2S02F, phenylmethanesulfonyl fluoride; PhCH2S02, phenylmethanesulfonyl; N-Ac-~-Tyr-"z, N-acetyl-L-tyrosineamide; UV, ultraviolet. Subtilisins Novo and BPN' are identical enzymes (Olaiten et al., 1968; Robertuset al., 1971; Drenth et al., 1972). Weuse thesamesubtilisin Novo when referring to this enzyme. inferred from the excitation spectra of diisopropylphosphor- ylsubtilisin Novo. The very low efficiency of energy transfer to Trp- 1 13 in zyxwv diisopropylphosphorylsubtilisin Carlsberg suggests that TrplO5 and Trp-241 are the acceptors of tyrosyl emission in the homologous Novo enzyme. The unusually low quantum yield of Trp-113 in diisopropylphosphorylsubtilisin Carlsberg together with the tryptophyl fluorescence quenching behavior of the Novo enzyme suggests that this residue is "buried" and inaccessible to quenching in both enzymes. The tyrosyl quenching behavior of diisopropylphosphorylsubtil- isin Carlsberg is consistent with the high degree of solvent exposure of aromatic residues evident in the x-ray model of subtilisin Novo. Carlsberg and Novo, respectively. The use of polarizable salts such as I- and Cs+ to estimate the accessibility of protein fluorophores to solvent is well established (Lehrer, 1967; Burstein, 1968a,b; Lehrer, 1971). The presence of tryptophyl and tyrosyl substitutions among subtilisin enzymes makes them attractive for study by fluorescence techniques. In particular, only one of the three tryptophyl residues of subtilisin Novo (Trp- 1 13) is retained in subtilisin Carlsberg (Trp- 106 and Trp-241 are substituted by Gly and Leu, respectively), so that structural homology affecting the environment of Trp- 1 13 can be studied from the relative fluorescence behavior of the two enzymes. The fluorescence results are compared with both the solvent perturbation experiments and the x-ray model of subtilisin Novo. Materials and Methods Subtilisins Novo and Carlsberg were inactivated with di- isopropyl fluorophosphate (Dip-F) or phenylmethanesulfonyl fluoride (PhCH2S02F) and purified as described previously (Brown and Schleich, 1975). N-Acetyl-L-tyrosineamide (N-Ac-L-T~PNH~) was obtained from Sigma Chemical Co., St. Louis, Mo. Reagent grade salts were used. The KI solutions contained M NarS203 to prevent oxidation and forma- tion of 13- (Lehrer, 1971). Fluorescence spectra were obtained using a Hitachi Perkin-Elmer MFP-2A fluorimeter operated in the direct mode at an ambient temperature of 22-23 "C. The emission and excitation monochromators were calibrated using a solution of anthracene in benzene. Spectra were generally recorded from 250 nm to the secondary scattering peak or zy 500 nm. Relative quantum yields were determined by integration of the uncorrected fluorescence spectra with a planimeter. Results Fluorescence Quenching zyxw of Model Compounds. The con- tribution of collisional quenching to the deactivation rate of an excited state fluorophore is described by the Stern-Volmer (1919) law: BIOCHEMISTRY, VOL. 16, NO. 5, 1977 987