Abstract: The effect of iodide on the tryptophyl fluorescence of model compounds and of lysozyme was studied in order to evaluate the factors that determine the use of iodide as a selective quencher of the fluorescence of tryptophyl side chains of proteins exposed to solvent. The results with the model compounds indicate the involvement of a collisional quenching mechanism due to the agreement with the Stern-Volmer law and the proportionality of the quenching constant with To7 for indole-3-acetamide. Bimolecular rate constants, k a , calculated from measured quenching constants using available lifetime data are equal to, greater than, or less than 4-6 X lo9 M-' sec-l for uncharged, positively charged, and negaI n a preliminary study it was shown that a large fraction of the tryptophyi fluorescence of lysozyme in aqueous solution was quenched by low concentrations of iodide ion (Lehrer, lY67). It was concluded from a study of the magnitude of the quenching of fluorescence and the character of the difference fluorescence spectrum produced in the presence and absence of substrate that the fluorescence of tryptophyls exposed to solvent and located in the substrate binding site was preferentially quenched by iodide. It appeared that this technique, which can be called solute perturbation of protein fluorescence, could be used as a probe of fluorophor exposure in proteins in a manner analogous to the technique of solvent perturbation of protein absorption (Herskovits and Laskowski, 1960; Laskowski, 1966). * From the Department of Muscle Research, Boston Biomedical Research Institute, Boston, Massachusetts 021 14, and from the Department of Neurology, Harvard Medical School, Boston, Massuchusetts 02115. Receired April 22, 1971. This work was supported by grants from the National Institutes of Health (AM 11677 and HE 0581 1) and the iMass'ichuserts Heart Association (516). tively charged tryptophyl compounds, respectively. A modified version of the Stern-Volmer law was calculated for a fluorophor population with different quantum yields and quenching constants. This formulation allows the calculation of the effective quenching constant from the intercept and the slope at low iodide concentration of a F o ] M cs. l/(I-) plot. Data obtained for lysozyme indicate that for the native protein about one-half the tryptophyl fluorescence is accessible at pH 5.3 whereas all of the tryptophyl fluorescence is accessible in 6 M G d n . HCI. Information regarding the presence of charged groups near tryptophyl side chains was obtained for lysozyme by studying the dependence of the quenching on pH. More recently, studies by other workers have ~ised bromate (Winkler, 1969) and iodide (Arrio er al., 1970) to quench extrinsic fluorescence (Teale and Badley, 1970). Oxygen has also been used as a quencher of pyrenebutyric acid bound to proteins (Vaughan and Weber, 1970). Burstein (1968a) has also independently studied the quenching of tryptophyl fluorescence in model compounds by iodide. In order to learn more about the quenching mechanism and the factors which determine fluorophor exposure, various tryptophyl model compounds and a model protein, lysozyme. were used in the present study. The results of the model compound study provide evidence for a mechanism that follows the classical Stern-Volmer law (1919), predominantly involving collisional quenching, and illustrate the importance of local charge and solvent viscosity. The quenching of lysozyme fluorescence by iodide also appears to follow a similar mechanism because of the agreement obtained with a inodified version of the Stern-Volmer law which was calculated for a heterogeneous distribution of fluorophors in a protein. Effective Stern-Volmer quenching constants and values for the fractional accessible fluorescence were obtained for lyso3254 B I O C H E M I S T R Y , V O L . 1 0 , N O . 1 7 , 1 9 7 1 I O D I D E Q U E N C H I N G O F P R O T E I N F L U O R E S C E N C E zyme in 6 M Gdn.HCI, 'S M urea, and in aqueous solution at different pH's using the modified Stern--Volmer law. Values obtained are consistent with information regarding accessibility obtained by other methods. Experimental Section Muteriais. The following high-purity compounds were used as obtained from Mann Research Laboratories, New York, N. Y. : indole-3-acetic acid, indole-3-propionic acid, indole-3-butyric acid, indole-3-acetamide, N-Ac-L-TrpNH?, L-TrpOEt, Gdn . HCI, and urea. L-Trp (Cyclo Chemical Corp., Los Angeles), KI , Na&03, citric acid, and NaCl (Fisher Scientific Co., Freehold, N. J.) were all of high purity and used as obtained. Indole (Fisher) and skatole (3-methylindole) (Mann) were recrystallized from methanol containing Norit A (Matheson Coleman & Bell, Rutherford, N. J.). Hepes buffer was used as obtained from Calbiochem (Los Angeles). Poly(Glug9Trp1) and poly(Lysg7Trp3) were high molecular weight random sequence copolymers kindly supplied by Dr. G. Fasman. Lysozyme from two different sources were used (twice crystallized from Worthington Biochemical Corp., Freehold, N. J., and six-times crystallized from Miles Laboratories, Elkhart, Ind.). Both preparations gave similar results. Ac3Glcn was kindly supplied by Dr. J. Rupley and glycol chitin was obtained from Miles Laboratories. Methods. Quenching measurements at constant pH were made on five solutions of a given material containing increasing amounts of K I (0-0.2 M). These were prepared by diluting stock solutions of the model compound, of KI, of NaC1, and of buffer, into volumetric flasks. NaCl was used to keep the ionic strength constant. Stock solutions of the indole compounds were used within a few days of preparation and kept in the dark at 0-5" overnight. A small amount of SO3?(ca. M) was added to the iodide solution to prevent 1 3 formation. This was necessary because Isabsorbs in the wavelength region of tryptophyl fluorescence (filter effect) and because of possible chemical reaction. The solutions were equilibrated at 25 O before the measurements. Stock solutions of lysozyme were routinely filtered through a Millipore filter (HAWP 0.45 p ) before use. pH titrations were performed in the absence and presence of iodide by adding small quantities of 0.5 M HC1 to the solution in the cuvet, which contained 2 mM Hepes and 2 mM citrate, originally pH 8, then measuring the pH and fluorescence. pH was measured with a Radiometer PHM4c meter standardized at pH 4 and 7. Fluorescence spectra and intensities were measured by exciting a t 280 nm or longer. In most cases no corrections for iodide absorption were necessary. The fluorescence of a reference (usually the 0.2 M NaC1-0.0 M K I solution) was measured just before measuring the fluorescence of each solution in order t o correct for any exciting lamp fluctuation. Fluorescence measurements were made with either an Aminco-Bowman spectrofluorometer or an instrument that employs two Jarrell-Ash 0.25-m monochromators, an EM1 9601 B photomultiplier, and either a high-pressure 200-W mercury lamp or a 150-W high-pressure xenon lamp. Low temperatures were obtained with a refrigerated water circulator attached to the sample housing. The temperature was measured by inserting a calibrated thermistor into the sample solution. Abbreviations used are: Gdn . HCI, guanidine hydrochloride; Trp, tryptophyl or tryptophan; Hepes, N-2-hydroxyethylpiperazine-N'2-ethanesulfonic acid ; Ac3GIcii. tri-N-acetyl-D-glucosamirie. The activity of lysozyme was determined by the method of Hamaguchi et a/ . (1960). The decrease in viscosity with time caused by hydrolysis of glycol chitin (2 mg/ml) by lysozyme (0.02 mg/ml) in the presence of 0.2 M NaCl or 0.2 M KI in 2 m M citrate (pH 5.5) is the basis of this method. The specific viscosity of glycol chitin solutions in Cannon viscometers at 25" was measured with time after a small volume of lysozyme was added. The slope of the approximately linear viscosity decrease between 1 and 10 min was used as a measure of activity. The optical rotatory dispersion and circular dichroism spectra of lysozyme (0.95 mg/ml) in 0.2 M NaCl or in 0.2 M KI , 2 mM citrate (pH 5.2) were measured in a 1-cm cell with a Jasco spectropolarimeter. The absorbance of Iprevented measurements below 265 nm. Difference spectra were either measured with a Cary 15 or a Beckman DK spectrophotometer using mixing cells (Pyrocell, Inc., N. Y . ) . The total absorption over the wavelengths scanned was always below 2.2. The low-temperature studies were performed with a Beckman D K using a refrigerated sample holder.