- Email: [email protected]
Fluorescence and the structure of proteins XXII. Fluorescence of aminotyrosyl residues formed in ribonuclease A
Bioch~m~ca et Blophyslca Acta, 439 (1976) 4~0-4%
© Elsevier Scientific Publishing Company, Amsterdam - - Printed in The Netherlands t]BA 37394 F L U O R E S C E N C E A N D T H E S T R U C T U R E OF P R O T E I N S XXII. F L U O R E S C E N C E RIBONUCLEASE A
OF A M I N O T Y R O S Y L
RESIDUES FORMED
IN
ROGER L. SEAGLE and ROBERT W. COWGILL* Department of Biochemistry, The Bowman Gray School of Medicine, Wake Forest University, WinstonSalem, N.C. 27103 (U.S.A.)
(Received January 14th, 1976)
S UMMARY 1. Tyrosyl residues on ribonuclease A were nitrated with tetranitromethane and then reduced to aminotyrosyl residues. By variation of reaction conditions and degree of exposure of tyrosyl residues it was possible to convert from 1 to all 6 tyrosyl to aminotyrosyl residues. 2. At the lower levels of 1-3 aminated tyrosyl residues/molecule the change in conformation seemed minor and 70 ~ of the enzymatic activity was retained. When the three buried tyrosyl residues or all six residues were aminated only 5 ~o of the enzymatic activity was retained. 3. Titration data, susceptibility to urea denaturation, and fluorescence characteristics indicated that some of the aminotyrosyl residues were buried in the interior and others were exposed on the surface of the protein. On the basis of the activation/ emission wavelengths it was possible to distinguish buried (288/320 nm) and exposed (288/365-395 nm) aminotyrosyl residues as well as exposed tyrosyl residues (275305 nm). 4. The modification of specific tyrosyl residues on a protein to aminotyrosyl residues appears to have some promise for observation of changes in environment of the residues that accompany various conformational changes by monitoring the fluorescence.
INTRODUCTION Chemical modification of tyrosyl residues on proteins to 3-aminotyrosyl residues alters their fluorescence properties. This appears to be a promising way to modify tyrosyl residues at specific sites in the protein because reaction conditions are mild, the aminated phenolic group remains uncharged, and the amino group is small. It should then be possible to selectively monitor changes in the environment of those residues by fluorescence. Earlier papers from this laboratory described the fluorescence _
° Author to whom correspondence should be addressed.
471 of 3-aminotyrosine [1] and aminotyrosyl residues in simple peptides and c~-helical fibrous proteins [2]. The present paper extends this study to aminotyrosyl residues in different regions of the globular protein ribonuclease A (ribonucleate pyrimidinenucleotide-2'-transferase EC 2.7.7.16). RNAase A has six tyrosyl residues, three of which are exposed to the solvent (Nos. 73, 76, and 115) and three (Nos. 25, 92, and 97) that are buried in the hydrophobic interior of the protein [3]. q-he three exposed tyrosyl residues exhibit fluorescence with activation and emission maxima A/E= 275/305 nm while the three buried tyrosyl residues are nonfluorescent [4]. Beaven and Gratzer [5] have reported a procedure whereby one, two, or all three exposed residues can be nitrated by reaction of the native protein with tetranitromethane. Following this procedure, it was possible to perform the nitrations and subsequently to reduce these exposed residues to the 3-aminotyrosyl state. By initial protection of the three exposed tyrosyl residues by acetylation, it was possible to expose and aminate the three buried tyrosyl residues. Finally, it was also possible to aminate all six tyrosyl residues on the protein, q-he fluorescence characteristics of these various aminotyrosyl residues were determined and interpreted on the basis of the presumed loci of the residues on the protein surface or buried in the interior. Enzymatic activity, solvent perturbation of exposed aminotyrosyl residues, and changes in fluorescence emission maxima upon denaturation were all employed in the latter evaluation. MATERIALS AND METHODS Bovine pancreatic ribonuclease A, type X-A, was the product of the Sigma Chemical Co. The protein was stored at 0 °C in 0.2 M phosphate buffer solution at pH 6.4. This company also supplied 3-amino-L-tyrosine dihydrochloride, tetranitromethane and cytidine 2',3'-cyclicphosphate. Urea of the Ultra-Pure grade was the product of the Schwarz-Mann Chemical Co.; sodium hydrosulfite and dimethylacetamide were supplied by Fisher Scientific Co. All fluorescence measurements were made as previously described and are expressed as fluorescence efficiency relative to 3-amino-L-tyrosine as a standard (Rr~H2 -Tyr) [1, 2]. Activity of RNAase A was determined by the spectrophotometric assay of Crook et al. [6] as modified by Cowgill [4]. Amino acid analysis. After complete hydrolysis of the RNAase A with HC1, the amino acids were analyzed by the gas chromatographic procedure of Gehrke et al. [7]. The amino acid mixture for standardization of the elution pattern was supplemented with an authentic sample of 3-aminotyrosine. q-he latter appeared as a distinct peak between lysine and arginine. Nitration of the tyrosyl residues of RNAase was accomplished by a modification of the method of Beaven and Gratzer [5] that eliminated acidification of the nitration mixture. The protein was treated with an excess of tetranitromethane for thirty minutes at pH 9.0 and was then placed on a short Sephadex G-25 gel chromatography column at the same pH to terminate the reaction by separation of the protein from residual tetranitromethane. Nitration of one tyrosyl residue was accomplished by the use of a fifteen-fold molar excess of tetranitromethane while nitration of two and three exposed residues required a ninety-fold and two hundred-fold excess, respectively. The nitro derivatives were reduced to the 3-aminotyrosyl derivatives by treatment with sodium hydrosulfite [8]. The solution then was dialyzed at 0 °C against
472 0.05 M Tris. HCI at pH 8.0. Absorption spectra of the products gave no evidence of chromophoric biproducts at least in the 270-370 nm region of interest in these studies. Amination of the three buried tyrosyl residues was accomplished by first acetylatirtg the exposed residues [9]. The degree of acetylation was determined by the decrease in absorbance at 276 nm. After acetylation of all three exposed residues the buried tyrosyl residues were exposed by denaturation of the protein with eight molar urea [10], and the latter residues were aminated according to the procedures described above. Then the acetylated tyrosyl residues were deacetylated by treatment with hydroxylamine [9]. All six tyrosyl residues were most successfully aminated by first acetylating the three exposed residues by first acetylating the three exposed residues and aminating the buried tyrosyl residues as above. Then the three acetylated tyrosyl residues were deacetylated and aminated by the procedures described above. RESULTS Tyrosyl residues of RNAase A were aminated by the various procedures described above for modification of specific exposed tyrosine residues, just the three buried tyrosine residues or all six tyrosine residues. The properties of the preparations are summarized in Table I. The extent of amination for these preparations was determined by the absorbance of the anainotyrosyl residues [8]. The reliability of this method was demonstrated by a chemical amino acid analysis of the RN(2 NHzTyr)~x preparation in Table I. The latter analysis showed the presence of 1.7 aminotyrosine and 4.3 tyrosine per molecule of RNAase A and this value for aminotyrosine compares well with the value of 1.79 residues reported in Table I. In confirmation of the report of Beaven and Gratzer [5], we were able to modify an average of one or more exposed tyrosine residues by employing limited amounts of tetranitromethane under mild conditions (see Materials and Methods section for details). ~I-lae loci of these aminated tyrosyl residues are indicated in Table I on the basis of the presumed availability of the various tyrosine residues for reaction under the conditions for nitration and on the assumption the modified protein retains the native conformation. The degree of enzymatic activity is one criterion of retention of the native conformation and it was observed that this activity decreased with the extent of amination as shown in Table I. These percentages of the original enzymatic activity do not distinguish between the possibility that a portion of the protein is fully active and the remainder is inactive and a second possibility that all molecules of the protein are equivalent and of equally diminished activity. However, it is apparent that conformational changes occurred and therefore the designation of exposed or buried aminctyrosyl residues is uncertain. Other criteria for designating the various residues as buried or exposed will be considered below. The absorption spectra in acid, neutral and alkaline pH regions for these aminotyrosyl residues were similar to those reported earlier for aminotyrosyl residues in peptides and proteins [8]. Fluorescence properties of the aminated tyrosyl residues are compiled in Table II. A number of these properties are common to all of the preparations and some of these properties will be described for RN(1 NH2-Tyr)ex as a representative example. At neutral pH the fluorescence spectrum contained a single peak as shown in Fig. 1. For RN(1 NHz-Tyr)~, the emission maximum was excep-
473 TABLE I CHARACTERISTICS OF RNAase A AFTER VARIOUS DEGREES OF AMINATION OF TYROSYL RESI DUES Preparation symbol Aminated tyrosine residues
Enzymatic activity ( ~ of activity of unmodified RNAase A)
Aminotyrosine Position* molecule RN(I RN(2 RN(3 RN(3 RN(6
NHz-Tyr)e~ NHz-Tyr)~ NHz-Tyr)~x N Hz-Tyr)b~,r NHz-Tyr)
1.04 1.79 3.00 3.04 5.96
Exposed Exposed Exposed Buried Some buried Some exposed
70 60 72 5 5
* See text for the basis of designating residues as buried or exposed. t i o n a l l y high (395 nm) c o m p a r e d to the other p r e p a r a t i o n s in T a b l e II, the value o f 320 n m for RN(3 NHz-Tyr)b,~ was u n u s u a l l y low and the significance o f these variations will be discussed later. One exception to the a p p e a r a n c e o f a single emission peak at n e u t r a l p H occurred for R N ( 6 N H z - T y r ) as shown in Fig. 1. This m o r e c o m p l e x s p e c t r u m is a t t r i b u t e d to emission f r o m b o t h buried a m i n o t y r o s y l residues at 320 n m and exposed residues at 385 n m as also will be discussed later. A n acidic fluorometric t i t r a t i o n was p e r f o r m e d with RN(1 N H z - T y r ) ~ . As the p H decreased b e l o w p H 5 the emission at 395 n m decreased and a new emission p e a k began to a p p e a r at 310 nm. The a p p e a r a n c e o f a second emission p e a k at 310 n m was not due to the fluorescence o f u n a m i n a t e d tyrosyl residues because the wavelength o f a c t i v a t i o n at 288 n m was selected to specifically activate the a m i n o t y r o s y l residue. The fluorescence emission o f the unmodified tyrosyl residues was m o n i t o r e d at AlE = 275/305 n m a n d n o change occurred over the range o f p H 4-8. Therefore, the a p p e a r a n c e o f a new emission p e a k b e l o w p H 5 when the activation was at 288 nm was due to the cationic f o r m o f the 3 - a m i n o t y r o s y l residue. (A large increase in emission at A / E = 275/305 n m occurred at p H 3, because d e n a t u r a t i o n o f the p r o t e i n TABLE I l FLUORESCENCE PROPERTIES OF THE AMINATED PREPARATIONS OF RNAase A The relative fluorescence efficiency (RNHz_~r~.r)at pH 8.0 was compared to the value for L-3-aminotyrosine as a standard at the same pH. _
Preparation
Activation and emission maxima (nm) pH 8
RNHz_xyr Fluorometric titrations 3-NH2 group
A/E
Phenolic OH group
pK
AlE
pK
288/395 288/385 288/365 288/320
10.05 10.15 10.05 11.0
_
RN(1 NH2-Tyr)cx RN(2 NH~-Tyr)ex RN(3 NH2-Tyr)cx RN(3 NH2-Tyr)bur RN(6 NHz-Tyr) .
_
288/395 288/385 288/365 288/320 288/320 + 385
1.55 0.96 0.67 1.68
i~ 288/310 288/310 288/320 288/320
4.70 4.75 4.20 4.40
474
l'-~\
1.4
//
,-\
.~
1.2
~
I.0
~\
,
~ °-~I
/
'" _.,
X~
,
©'4f 0.2 __~/
280
3~o
340 370 Wovelengl.h in nm
400
430
460
Fig. 1. Fluorescence emission s p e c t r u m at p H 7 for RN(1 NH2-Tyr)ex ( ( - - -- --). Activation was at 288 n m .
) and RN(6 NH2-Tyr)
occurred and exposure of the buried tyrosyl residues was responsible for the increase in fluorescence [11]. This change was readily resolved from changes in the aminotyrosyl residue by proper choice of A/E values.) In more acidic solution, the emission peak at A/E = 288/310 nm became the dominant peak with the 395 nm peak appearing as a shoulder. These fluorometric titration curves are similar to those reported for 3-aminotyrosine [1] and aminotyrosyl residues in peptides [2]. The p K of 4.70 from the fluorometric titration curve is comparable to the values for the other aminotyrosyl residues in Table II and the value of 4.4 for the amino group of 3-aminotyrosine as reported earlier [1]. An alkaline fluorometric titration also was performed with RN(1 NH2-Tyr)ex over the range of p H 8-12. In this p H range the 3 l0 nm emission peak of the cationic 3-NH3 + form was not present so that only the 395 nm emission was monitored. As the p H increased, the emission at 395 nm decreased. This decrease in emission was identical to that seen with the helical proteins [2]. The titration curve yielded a p K value of 10.05 for the phenolic hydroxyl group of the aminated tyrosyl residue which is identical to that determined spectrophotometrically for 3-NH2 tyrosine [1]. Table II also contains values for the average fluorescence yield of aminotyrosyl residues in each preparation expressed as RNH2_TYr values. Several efforts were made to determine whether or not the aminotyrosyl residues in the preparations remained buried or exposed as indicated in Table I. Disruption of the conformation by urea would be expected to produce changes in fluorescence of the aminotyrosyl residues that might be interpretable in terms of their change in environment. Fig. 2 shows the effect of urea on the fluorescence of these residues that are presumed to be in both exposed and buried sites. For RN(1 NH2Tyr)ex the fluorescence remained at 395 nm and the increase in emission was similar to that seen for 3-aminotyrosine. This increase in fluorescence is attributed to a general enhancement of fluorescence by alteration in the nature of the solvent at the higher concentrations of urea [11], and such as monotonic rise would be expected if the aminotyrosyl residue was exposed throughout the range of urea concentration. In the case of RN(3 NH2-Tyr)bur no such increase was noted at low levels of urea but there was a marked enhancement of fluorescence at 6-7 M urea that was accompanied by a shift of the emission maximum from 320 nm in the absence of urea to 350 nm in
475 i
~E 2~)(
J
x
> ._ _o c~
1.00
~
~
;
;
Ureo (M)
Fig. 2. Effect of urea on the fluorescence o33-aminotyrosine (©), RN(I NH2-Tyr)~x( x ), RN(3 NH2Tyr)bur (~) and RN(6 NH2-Tyr) ([~) in 0.05 M Mo~s at pH 7.5. 8 M urea. Native RNAase A becomes denatured in this same region of 6-8 M urea [10] and exhibits enhanced fluorescence from its tyrosyl residues as they undergo a change from buried to exposed positions [11]. RN(6 NH2-Tyr) showed enhanced fluorescence at 3-5 M urea and in this case the spectrum shown in Fig. 1 shifted to a single peak at 385 nm. Both of these alterations are consistent with a shift from buried aminotyrosyl residues (320 nm) to exposed residues (350-385 nm) upon change in conformation of these aminated preparations by the urea. The fluorescence emission of RN(3 NH2-Tyr)bur o f RNH2_Ty r ~--- 1.68 was much lower than expected for residues buried in a nonpolar environment for it had been observed in the earlier studies that fluorescence was enhanced some ten fold in nonpolar solvents [1, 2]. It appears probable that the much lower fluorescence observed for the presumably buried aminotyrosyl residues in the present preparation isattributable to quenching of fluorescence in the protein interior. It has been shown that buried tyrosyl residues in RNAase A are nonfluorescent because they are hydrogenbonded to carbonyl groups, either of the peptide bond or the side-chains of asparagine, aspartic acid, glutamine and glutamic acid residues. To test this possibility, the effect of dimethylacetamide on fluorescence of o-aminophenol was tested in n-hexane. The data conformed to a linear Stern-Volmer plot for fluorescence quenching which was shown for the analogous quenching of phenol fluorescence to be due to formation of a static hydrogen-bonded complex [12]. F r o m this data, the association constant would be 124 M -1 for a honfluorescent hydrogen-bonded complex formed between o-aminophenol and dimethylacetamide. This value is comparable with values ranging from 67 M -1 to 285 M - t obtained in a similar fashion for other derivatives of phenol [12]. DISCUSSION Aminotyrosyl residues were formed on RNAase A by reactions that are known to yield these derivatives in proteins [2, 5, 8]. The identity of the products as amino-
476 tyrosyl residues was confirmed by their spectrophotometric and fluorometric properties, pK values for ionization of the 3-NH3 + group and phenolic hydroxyl group. and detection of the expected amount of 3-aminotyrosine upon complete hydrolysis. The principle purpose of this study was to compare the fluorescence characteristics of aminotyrosyl residues buried in the protein interior and those on the protein surface and exposed to the aqueous solution. The protein RNAase A was selected for this purpose because of the presence of three buried and three exposed tyrosyl residues as well as the absence of tryptophan which fluoresces in the same region as aminotyrosyl residues. Brief treatment of RNAase A at pH 9 with a 15 times excess of tetranitromethane at room temperature, followed by reduction of the nitrotyrosyl residue is a mild procedure that would be expected to affect only exposed tyrosyl residues on the native molecule. The properties of the product, RN(I NH2Tyr)ox, are in accord with this supposition for the product retained 70~i~ of the original enzymatic activity and displayed properties attributable to an exposed residue. That is, the 3-NHz group and phenolic hydroxyl groups titrated with normal pK values, the emission maximum was high and the fluorescence was perturbed by low concentrations of urea (Fig. 2). The emission maximum at 395 nm is exceptionally high as compared to values of 350-370 nm observed for aminotyrosyl residues in peptides and on the surface of fibrous proteins [2]. Because the amination was performed under limited conditions of time and reactant concentration and because the fluorescence is atypically high, it appears that one specific tyrosyl residue on the surface is especially susceptible to nitration. The emission at 395 nm from this aminotyrosyl residue suggests that it is located in some unusual environment. Perhaps these latter conditions also account for its susceptibility to nitration. The other two products of nitration and reduction under conditions where the protein would be expected to remain native also should have reaction restricted to the exposed tyrosyl residues. These products RN(2 NHz-Tyr)ex and RN(3 NHz-Tyr)ex had their aminotyrosyl residues exposed as judged by the criteria discussed above. The fluorescence properties summarized in Table II show progressive shifts of the emission maxima and fluorescence efficiency as amination becomes more extensive. The shift of emission maxima from 395 nm for RN(1 NH2-Tyr)~x to 365 nm for RN(3 NHz-Tyr)~= and the observed broadening of the spectrum would be expected if the observed spectrum for the latter were a summation of one residue of 395 nm emission and a second and third residue of 350-360 nm emission. There was a progressive decrease in fluorescence efficiency from Rs,~2 "ryr 1.55 for RN(I NHz-Tyr)~ to 0.67 for RN(3 NH2-Tyr)~x, and this may indicate that energy transfers between residues and energy sink effects decreased the fluorescence efficiency. These effects are known to affect tyrosine fluorescence in RNAase A [11 ] but they are too complex for analysis in the present situation. The fluorescence from buried*aminotyrosyl residues is believed to occur in RN(3 NH2-Tyr)~,~ which was formed by blockage of the exposed tyrosyl residues by acetylation, exposure and amination of the formerly buried tyrosyl residues and then deacetylation of the nonaminated tyrosyl residues. If the native conformation was approached after return of the protein to a benign environment the three aminated tyrosyl residues should be buried, however, the retention of only 5 o//oof the original enzymatic activity indicates that the conformation was not the same. Even though the conformation was altered sufficiently to abolish enzymatic activity, the amino-
477 tyrosyl residues appear to be buried in the protein for the following reasons. The fluorometric titration gave a pK = 11.0 which is higher than the value of 10.1 for the exposed aminotyrosyl residues in Table II. This behavior is analogous to the titration behavior of native RNAase A [13] for the buried tyrosyl residues do not titrate until the pH exceeds I 1.5. The aminotyrosyl residues also appear to be buried and thus protected from perturbation of their fluorescence by addition of urea to the solution as shown by the constant fluorescence in the region of 0-5 M urea in Fig. 2. At higher concentration of urea in the region of 6-8 M the increase in fluorescence probably is a result of expansion of the protein into a random coil with all residues exposed. Once again, RNAase A exhibits similar behavior at the same region of urea concentration [10, 11 ] and this suggests that RN(3 NHz-Tyr)bur, although of altered conformation, is still about as stable to urea denaturation as the original RNAase A. The transformation of RN(3 NH2-Tyr)bur over the region 6-8 M urea was accompanied by a shift of emission from 320 nm to 350 nm which would be expected for exposure of aminotyrosyl residues because the earlier studies [1, 2] with model compounds indicated that fluorescence occurs at 320 nm in a nonpolar environment and at 350-370 nm in aqueous solution. The fluorescence efficiency of these model aminophenols in nonpolar solvents was about ten fold higher than in aqueous solution but a high RNH 2 Vyr value for RN(3 NHz-Tyr)b,r was not observed (Table II). Instead, the low fluorescence and enhancement upon denaturation in urea solution appears analogous to the behavior of buried tyrosyl residues in RNAase A. In the latter case, the nonfluorescence of buried tyrosyl residues was due to the formation of hydrogenbonds in the protein interior as mentioned in Results. The demonstration that dimethylacetamide quenches fluorescence of o-aminophenol in hexane indicates that aminotyrosyl residues in the protein interior also may form hydrogen-bonds with carbonyl groups and become nonfluorescent. The completely aminated product RN(6 NHz-Tyr) displayed fluorescence characteristics of both buried and exposed aminotyrosyl residues. J-he emission peaks at both 320 nm and 385 nm in Fig. 1 appear to arise from the two types of residues and the loss of the 320 nm peak upon denaturation in urea is consistent with this supposition. The disruption of conformation occurred at 3-5 M urea and indicates less stability for this preparation which is not surprising in view of the extensive modification of residues. In conclusion, this study of aminated tyrosyl residues on RNAase indicates that it should be possible to determine from the fluorescence characteristics the location of such residues in buried or exposed loci on proteins and changes in location upon change of conformation. Contrary to the situation with tyrosine whose fluorescence maximum is unchanged by the environment of the residue, the emission from the aminotyrosyl residue is sensitive to the environment in a fashion that is analogous to that for tryptophanyl residues. It should be noted that the range of emission maxima for aminotyrosyl residues of 320-370 nm as well as the optimal activation at 288 nm permits one to distinguish fluorescence of aminotyrosyl residues from fluorescence of tyrosyl residues ( A l E - - 276/308 nm). However, if tryptophan is present in the protein its fluorescence ( A / E = 280/320-350 nm) would obscure the weaker fluorescence of the aminotyrosyl residues.
478 REFERENCES 1 Cowgill, R. W. (1971) Photochem. Photobiol. 13, 183-194 2 Seagle, R. L. and Cowgill, R. W. (1976) Biochim. Biophys. Acta 439, 4 6 1 4 6 9 3 Wyckoff, H. W., Hartman, K. D., Allewell, N. M., lnagami, T., Johnson, L. N. and Richards, F. M. (1967) J. Biol. Chem 242, 3984-3988 4 Cowgill, R. W. (1966) Biochim. Biophys. Acta 120, 189-195 5 Beaven, G. and Gratzer, W. (1968) Biochim. Biophys. Acta 168, 456-462 6 Crook, E. M., Mathias, A. P. and Rabin, B. R. (1960) Biochem. J. 74, 234-238 7 Gehrke, C. W., Kuo, K. and Zumwalt, R. W. (1971) J. Chromatogr. 57, 209-217 8 Sokolovsky, M., Riordan, J. F. and Vallee, B. L. (1967) Biochem. Biophys. Res. Commun. 27, 20-25 9 Riordan, J. F., Wacker, W. E. C. and Vallee, B. L. (1965) Biochemistry 4, 1758-1765 10 Nelson, C. A. and Hummel, J. P. (1962) J. Biol. Chem. 237, 1567-1574 11 Cowgill, R. W. (1966) Biochim. Biophys. Acta 120, 196-211 12 Cowgill, R. W. (1965) Biochim. Biophys. Acta 109, 536-543 13 Tanford, C., Hauenstein, J. D. and Rands, D. G. (1956) J. Am. Chem. Soc. 77, 6409-6413
© Elsevier Scientific Publishing Company, Amsterdam - - Printed in The Netherlands t]BA 37394 F L U O R E S C E N C E A N D T H E S T R U C T U R E OF P R O T E I N S XXII. F L U O R E S C E N C E RIBONUCLEASE A
OF A M I N O T Y R O S Y L
RESIDUES FORMED
IN
ROGER L. SEAGLE and ROBERT W. COWGILL* Department of Biochemistry, The Bowman Gray School of Medicine, Wake Forest University, WinstonSalem, N.C. 27103 (U.S.A.)
(Received January 14th, 1976)
S UMMARY 1. Tyrosyl residues on ribonuclease A were nitrated with tetranitromethane and then reduced to aminotyrosyl residues. By variation of reaction conditions and degree of exposure of tyrosyl residues it was possible to convert from 1 to all 6 tyrosyl to aminotyrosyl residues. 2. At the lower levels of 1-3 aminated tyrosyl residues/molecule the change in conformation seemed minor and 70 ~ of the enzymatic activity was retained. When the three buried tyrosyl residues or all six residues were aminated only 5 ~o of the enzymatic activity was retained. 3. Titration data, susceptibility to urea denaturation, and fluorescence characteristics indicated that some of the aminotyrosyl residues were buried in the interior and others were exposed on the surface of the protein. On the basis of the activation/ emission wavelengths it was possible to distinguish buried (288/320 nm) and exposed (288/365-395 nm) aminotyrosyl residues as well as exposed tyrosyl residues (275305 nm). 4. The modification of specific tyrosyl residues on a protein to aminotyrosyl residues appears to have some promise for observation of changes in environment of the residues that accompany various conformational changes by monitoring the fluorescence.
INTRODUCTION Chemical modification of tyrosyl residues on proteins to 3-aminotyrosyl residues alters their fluorescence properties. This appears to be a promising way to modify tyrosyl residues at specific sites in the protein because reaction conditions are mild, the aminated phenolic group remains uncharged, and the amino group is small. It should then be possible to selectively monitor changes in the environment of those residues by fluorescence. Earlier papers from this laboratory described the fluorescence _
° Author to whom correspondence should be addressed.
471 of 3-aminotyrosine [1] and aminotyrosyl residues in simple peptides and c~-helical fibrous proteins [2]. The present paper extends this study to aminotyrosyl residues in different regions of the globular protein ribonuclease A (ribonucleate pyrimidinenucleotide-2'-transferase EC 2.7.7.16). RNAase A has six tyrosyl residues, three of which are exposed to the solvent (Nos. 73, 76, and 115) and three (Nos. 25, 92, and 97) that are buried in the hydrophobic interior of the protein [3]. q-he three exposed tyrosyl residues exhibit fluorescence with activation and emission maxima A/E= 275/305 nm while the three buried tyrosyl residues are nonfluorescent [4]. Beaven and Gratzer [5] have reported a procedure whereby one, two, or all three exposed residues can be nitrated by reaction of the native protein with tetranitromethane. Following this procedure, it was possible to perform the nitrations and subsequently to reduce these exposed residues to the 3-aminotyrosyl state. By initial protection of the three exposed tyrosyl residues by acetylation, it was possible to expose and aminate the three buried tyrosyl residues. Finally, it was also possible to aminate all six tyrosyl residues on the protein, q-he fluorescence characteristics of these various aminotyrosyl residues were determined and interpreted on the basis of the presumed loci of the residues on the protein surface or buried in the interior. Enzymatic activity, solvent perturbation of exposed aminotyrosyl residues, and changes in fluorescence emission maxima upon denaturation were all employed in the latter evaluation. MATERIALS AND METHODS Bovine pancreatic ribonuclease A, type X-A, was the product of the Sigma Chemical Co. The protein was stored at 0 °C in 0.2 M phosphate buffer solution at pH 6.4. This company also supplied 3-amino-L-tyrosine dihydrochloride, tetranitromethane and cytidine 2',3'-cyclicphosphate. Urea of the Ultra-Pure grade was the product of the Schwarz-Mann Chemical Co.; sodium hydrosulfite and dimethylacetamide were supplied by Fisher Scientific Co. All fluorescence measurements were made as previously described and are expressed as fluorescence efficiency relative to 3-amino-L-tyrosine as a standard (Rr~H2 -Tyr) [1, 2]. Activity of RNAase A was determined by the spectrophotometric assay of Crook et al. [6] as modified by Cowgill [4]. Amino acid analysis. After complete hydrolysis of the RNAase A with HC1, the amino acids were analyzed by the gas chromatographic procedure of Gehrke et al. [7]. The amino acid mixture for standardization of the elution pattern was supplemented with an authentic sample of 3-aminotyrosine. q-he latter appeared as a distinct peak between lysine and arginine. Nitration of the tyrosyl residues of RNAase was accomplished by a modification of the method of Beaven and Gratzer [5] that eliminated acidification of the nitration mixture. The protein was treated with an excess of tetranitromethane for thirty minutes at pH 9.0 and was then placed on a short Sephadex G-25 gel chromatography column at the same pH to terminate the reaction by separation of the protein from residual tetranitromethane. Nitration of one tyrosyl residue was accomplished by the use of a fifteen-fold molar excess of tetranitromethane while nitration of two and three exposed residues required a ninety-fold and two hundred-fold excess, respectively. The nitro derivatives were reduced to the 3-aminotyrosyl derivatives by treatment with sodium hydrosulfite [8]. The solution then was dialyzed at 0 °C against
472 0.05 M Tris. HCI at pH 8.0. Absorption spectra of the products gave no evidence of chromophoric biproducts at least in the 270-370 nm region of interest in these studies. Amination of the three buried tyrosyl residues was accomplished by first acetylatirtg the exposed residues [9]. The degree of acetylation was determined by the decrease in absorbance at 276 nm. After acetylation of all three exposed residues the buried tyrosyl residues were exposed by denaturation of the protein with eight molar urea [10], and the latter residues were aminated according to the procedures described above. Then the acetylated tyrosyl residues were deacetylated by treatment with hydroxylamine [9]. All six tyrosyl residues were most successfully aminated by first acetylating the three exposed residues by first acetylating the three exposed residues and aminating the buried tyrosyl residues as above. Then the three acetylated tyrosyl residues were deacetylated and aminated by the procedures described above. RESULTS Tyrosyl residues of RNAase A were aminated by the various procedures described above for modification of specific exposed tyrosine residues, just the three buried tyrosine residues or all six tyrosine residues. The properties of the preparations are summarized in Table I. The extent of amination for these preparations was determined by the absorbance of the anainotyrosyl residues [8]. The reliability of this method was demonstrated by a chemical amino acid analysis of the RN(2 NHzTyr)~x preparation in Table I. The latter analysis showed the presence of 1.7 aminotyrosine and 4.3 tyrosine per molecule of RNAase A and this value for aminotyrosine compares well with the value of 1.79 residues reported in Table I. In confirmation of the report of Beaven and Gratzer [5], we were able to modify an average of one or more exposed tyrosine residues by employing limited amounts of tetranitromethane under mild conditions (see Materials and Methods section for details). ~I-lae loci of these aminated tyrosyl residues are indicated in Table I on the basis of the presumed availability of the various tyrosine residues for reaction under the conditions for nitration and on the assumption the modified protein retains the native conformation. The degree of enzymatic activity is one criterion of retention of the native conformation and it was observed that this activity decreased with the extent of amination as shown in Table I. These percentages of the original enzymatic activity do not distinguish between the possibility that a portion of the protein is fully active and the remainder is inactive and a second possibility that all molecules of the protein are equivalent and of equally diminished activity. However, it is apparent that conformational changes occurred and therefore the designation of exposed or buried aminctyrosyl residues is uncertain. Other criteria for designating the various residues as buried or exposed will be considered below. The absorption spectra in acid, neutral and alkaline pH regions for these aminotyrosyl residues were similar to those reported earlier for aminotyrosyl residues in peptides and proteins [8]. Fluorescence properties of the aminated tyrosyl residues are compiled in Table II. A number of these properties are common to all of the preparations and some of these properties will be described for RN(1 NH2-Tyr)ex as a representative example. At neutral pH the fluorescence spectrum contained a single peak as shown in Fig. 1. For RN(1 NHz-Tyr)~, the emission maximum was excep-
473 TABLE I CHARACTERISTICS OF RNAase A AFTER VARIOUS DEGREES OF AMINATION OF TYROSYL RESI DUES Preparation symbol Aminated tyrosine residues
Enzymatic activity ( ~ of activity of unmodified RNAase A)
Aminotyrosine Position* molecule RN(I RN(2 RN(3 RN(3 RN(6
NHz-Tyr)e~ NHz-Tyr)~ NHz-Tyr)~x N Hz-Tyr)b~,r NHz-Tyr)
1.04 1.79 3.00 3.04 5.96
Exposed Exposed Exposed Buried Some buried Some exposed
70 60 72 5 5
* See text for the basis of designating residues as buried or exposed. t i o n a l l y high (395 nm) c o m p a r e d to the other p r e p a r a t i o n s in T a b l e II, the value o f 320 n m for RN(3 NHz-Tyr)b,~ was u n u s u a l l y low and the significance o f these variations will be discussed later. One exception to the a p p e a r a n c e o f a single emission peak at n e u t r a l p H occurred for R N ( 6 N H z - T y r ) as shown in Fig. 1. This m o r e c o m p l e x s p e c t r u m is a t t r i b u t e d to emission f r o m b o t h buried a m i n o t y r o s y l residues at 320 n m and exposed residues at 385 n m as also will be discussed later. A n acidic fluorometric t i t r a t i o n was p e r f o r m e d with RN(1 N H z - T y r ) ~ . As the p H decreased b e l o w p H 5 the emission at 395 n m decreased and a new emission p e a k began to a p p e a r at 310 nm. The a p p e a r a n c e o f a second emission p e a k at 310 n m was not due to the fluorescence o f u n a m i n a t e d tyrosyl residues because the wavelength o f a c t i v a t i o n at 288 n m was selected to specifically activate the a m i n o t y r o s y l residue. The fluorescence emission o f the unmodified tyrosyl residues was m o n i t o r e d at AlE = 275/305 n m a n d n o change occurred over the range o f p H 4-8. Therefore, the a p p e a r a n c e o f a new emission p e a k b e l o w p H 5 when the activation was at 288 nm was due to the cationic f o r m o f the 3 - a m i n o t y r o s y l residue. (A large increase in emission at A / E = 275/305 n m occurred at p H 3, because d e n a t u r a t i o n o f the p r o t e i n TABLE I l FLUORESCENCE PROPERTIES OF THE AMINATED PREPARATIONS OF RNAase A The relative fluorescence efficiency (RNHz_~r~.r)at pH 8.0 was compared to the value for L-3-aminotyrosine as a standard at the same pH. _
Preparation
Activation and emission maxima (nm) pH 8
RNHz_xyr Fluorometric titrations 3-NH2 group
A/E
Phenolic OH group
pK
AlE
pK
288/395 288/385 288/365 288/320
10.05 10.15 10.05 11.0
_
RN(1 NH2-Tyr)cx RN(2 NH~-Tyr)ex RN(3 NH2-Tyr)cx RN(3 NH2-Tyr)bur RN(6 NHz-Tyr) .
_
288/395 288/385 288/365 288/320 288/320 + 385
1.55 0.96 0.67 1.68
i~ 288/310 288/310 288/320 288/320
4.70 4.75 4.20 4.40
474
l'-~\
1.4
//
,-\
.~
1.2
~
I.0
~\
,
~ °-~I
/
'" _.,
X~
,
©'4f 0.2 __~/
280
3~o
340 370 Wovelengl.h in nm
400
430
460
Fig. 1. Fluorescence emission s p e c t r u m at p H 7 for RN(1 NH2-Tyr)ex ( ( - - -- --). Activation was at 288 n m .
) and RN(6 NH2-Tyr)
occurred and exposure of the buried tyrosyl residues was responsible for the increase in fluorescence [11]. This change was readily resolved from changes in the aminotyrosyl residue by proper choice of A/E values.) In more acidic solution, the emission peak at A/E = 288/310 nm became the dominant peak with the 395 nm peak appearing as a shoulder. These fluorometric titration curves are similar to those reported for 3-aminotyrosine [1] and aminotyrosyl residues in peptides [2]. The p K of 4.70 from the fluorometric titration curve is comparable to the values for the other aminotyrosyl residues in Table II and the value of 4.4 for the amino group of 3-aminotyrosine as reported earlier [1]. An alkaline fluorometric titration also was performed with RN(1 NH2-Tyr)ex over the range of p H 8-12. In this p H range the 3 l0 nm emission peak of the cationic 3-NH3 + form was not present so that only the 395 nm emission was monitored. As the p H increased, the emission at 395 nm decreased. This decrease in emission was identical to that seen with the helical proteins [2]. The titration curve yielded a p K value of 10.05 for the phenolic hydroxyl group of the aminated tyrosyl residue which is identical to that determined spectrophotometrically for 3-NH2 tyrosine [1]. Table II also contains values for the average fluorescence yield of aminotyrosyl residues in each preparation expressed as RNH2_TYr values. Several efforts were made to determine whether or not the aminotyrosyl residues in the preparations remained buried or exposed as indicated in Table I. Disruption of the conformation by urea would be expected to produce changes in fluorescence of the aminotyrosyl residues that might be interpretable in terms of their change in environment. Fig. 2 shows the effect of urea on the fluorescence of these residues that are presumed to be in both exposed and buried sites. For RN(1 NH2Tyr)ex the fluorescence remained at 395 nm and the increase in emission was similar to that seen for 3-aminotyrosine. This increase in fluorescence is attributed to a general enhancement of fluorescence by alteration in the nature of the solvent at the higher concentrations of urea [11], and such as monotonic rise would be expected if the aminotyrosyl residue was exposed throughout the range of urea concentration. In the case of RN(3 NH2-Tyr)bur no such increase was noted at low levels of urea but there was a marked enhancement of fluorescence at 6-7 M urea that was accompanied by a shift of the emission maximum from 320 nm in the absence of urea to 350 nm in
475 i
~E 2~)(
J
x
> ._ _o c~
1.00
~
~
;
;
Ureo (M)
Fig. 2. Effect of urea on the fluorescence o33-aminotyrosine (©), RN(I NH2-Tyr)~x( x ), RN(3 NH2Tyr)bur (~) and RN(6 NH2-Tyr) ([~) in 0.05 M Mo~s at pH 7.5. 8 M urea. Native RNAase A becomes denatured in this same region of 6-8 M urea [10] and exhibits enhanced fluorescence from its tyrosyl residues as they undergo a change from buried to exposed positions [11]. RN(6 NH2-Tyr) showed enhanced fluorescence at 3-5 M urea and in this case the spectrum shown in Fig. 1 shifted to a single peak at 385 nm. Both of these alterations are consistent with a shift from buried aminotyrosyl residues (320 nm) to exposed residues (350-385 nm) upon change in conformation of these aminated preparations by the urea. The fluorescence emission of RN(3 NH2-Tyr)bur o f RNH2_Ty r ~--- 1.68 was much lower than expected for residues buried in a nonpolar environment for it had been observed in the earlier studies that fluorescence was enhanced some ten fold in nonpolar solvents [1, 2]. It appears probable that the much lower fluorescence observed for the presumably buried aminotyrosyl residues in the present preparation isattributable to quenching of fluorescence in the protein interior. It has been shown that buried tyrosyl residues in RNAase A are nonfluorescent because they are hydrogenbonded to carbonyl groups, either of the peptide bond or the side-chains of asparagine, aspartic acid, glutamine and glutamic acid residues. To test this possibility, the effect of dimethylacetamide on fluorescence of o-aminophenol was tested in n-hexane. The data conformed to a linear Stern-Volmer plot for fluorescence quenching which was shown for the analogous quenching of phenol fluorescence to be due to formation of a static hydrogen-bonded complex [12]. F r o m this data, the association constant would be 124 M -1 for a honfluorescent hydrogen-bonded complex formed between o-aminophenol and dimethylacetamide. This value is comparable with values ranging from 67 M -1 to 285 M - t obtained in a similar fashion for other derivatives of phenol [12]. DISCUSSION Aminotyrosyl residues were formed on RNAase A by reactions that are known to yield these derivatives in proteins [2, 5, 8]. The identity of the products as amino-
476 tyrosyl residues was confirmed by their spectrophotometric and fluorometric properties, pK values for ionization of the 3-NH3 + group and phenolic hydroxyl group. and detection of the expected amount of 3-aminotyrosine upon complete hydrolysis. The principle purpose of this study was to compare the fluorescence characteristics of aminotyrosyl residues buried in the protein interior and those on the protein surface and exposed to the aqueous solution. The protein RNAase A was selected for this purpose because of the presence of three buried and three exposed tyrosyl residues as well as the absence of tryptophan which fluoresces in the same region as aminotyrosyl residues. Brief treatment of RNAase A at pH 9 with a 15 times excess of tetranitromethane at room temperature, followed by reduction of the nitrotyrosyl residue is a mild procedure that would be expected to affect only exposed tyrosyl residues on the native molecule. The properties of the product, RN(I NH2Tyr)ox, are in accord with this supposition for the product retained 70~i~ of the original enzymatic activity and displayed properties attributable to an exposed residue. That is, the 3-NHz group and phenolic hydroxyl groups titrated with normal pK values, the emission maximum was high and the fluorescence was perturbed by low concentrations of urea (Fig. 2). The emission maximum at 395 nm is exceptionally high as compared to values of 350-370 nm observed for aminotyrosyl residues in peptides and on the surface of fibrous proteins [2]. Because the amination was performed under limited conditions of time and reactant concentration and because the fluorescence is atypically high, it appears that one specific tyrosyl residue on the surface is especially susceptible to nitration. The emission at 395 nm from this aminotyrosyl residue suggests that it is located in some unusual environment. Perhaps these latter conditions also account for its susceptibility to nitration. The other two products of nitration and reduction under conditions where the protein would be expected to remain native also should have reaction restricted to the exposed tyrosyl residues. These products RN(2 NHz-Tyr)ex and RN(3 NHz-Tyr)ex had their aminotyrosyl residues exposed as judged by the criteria discussed above. The fluorescence properties summarized in Table II show progressive shifts of the emission maxima and fluorescence efficiency as amination becomes more extensive. The shift of emission maxima from 395 nm for RN(1 NH2-Tyr)~x to 365 nm for RN(3 NHz-Tyr)~= and the observed broadening of the spectrum would be expected if the observed spectrum for the latter were a summation of one residue of 395 nm emission and a second and third residue of 350-360 nm emission. There was a progressive decrease in fluorescence efficiency from Rs,~2 "ryr 1.55 for RN(I NHz-Tyr)~ to 0.67 for RN(3 NH2-Tyr)~x, and this may indicate that energy transfers between residues and energy sink effects decreased the fluorescence efficiency. These effects are known to affect tyrosine fluorescence in RNAase A [11 ] but they are too complex for analysis in the present situation. The fluorescence from buried*aminotyrosyl residues is believed to occur in RN(3 NH2-Tyr)~,~ which was formed by blockage of the exposed tyrosyl residues by acetylation, exposure and amination of the formerly buried tyrosyl residues and then deacetylation of the nonaminated tyrosyl residues. If the native conformation was approached after return of the protein to a benign environment the three aminated tyrosyl residues should be buried, however, the retention of only 5 o//oof the original enzymatic activity indicates that the conformation was not the same. Even though the conformation was altered sufficiently to abolish enzymatic activity, the amino-
477 tyrosyl residues appear to be buried in the protein for the following reasons. The fluorometric titration gave a pK = 11.0 which is higher than the value of 10.1 for the exposed aminotyrosyl residues in Table II. This behavior is analogous to the titration behavior of native RNAase A [13] for the buried tyrosyl residues do not titrate until the pH exceeds I 1.5. The aminotyrosyl residues also appear to be buried and thus protected from perturbation of their fluorescence by addition of urea to the solution as shown by the constant fluorescence in the region of 0-5 M urea in Fig. 2. At higher concentration of urea in the region of 6-8 M the increase in fluorescence probably is a result of expansion of the protein into a random coil with all residues exposed. Once again, RNAase A exhibits similar behavior at the same region of urea concentration [10, 11 ] and this suggests that RN(3 NHz-Tyr)bur, although of altered conformation, is still about as stable to urea denaturation as the original RNAase A. The transformation of RN(3 NH2-Tyr)bur over the region 6-8 M urea was accompanied by a shift of emission from 320 nm to 350 nm which would be expected for exposure of aminotyrosyl residues because the earlier studies [1, 2] with model compounds indicated that fluorescence occurs at 320 nm in a nonpolar environment and at 350-370 nm in aqueous solution. The fluorescence efficiency of these model aminophenols in nonpolar solvents was about ten fold higher than in aqueous solution but a high RNH 2 Vyr value for RN(3 NHz-Tyr)b,r was not observed (Table II). Instead, the low fluorescence and enhancement upon denaturation in urea solution appears analogous to the behavior of buried tyrosyl residues in RNAase A. In the latter case, the nonfluorescence of buried tyrosyl residues was due to the formation of hydrogenbonds in the protein interior as mentioned in Results. The demonstration that dimethylacetamide quenches fluorescence of o-aminophenol in hexane indicates that aminotyrosyl residues in the protein interior also may form hydrogen-bonds with carbonyl groups and become nonfluorescent. The completely aminated product RN(6 NHz-Tyr) displayed fluorescence characteristics of both buried and exposed aminotyrosyl residues. J-he emission peaks at both 320 nm and 385 nm in Fig. 1 appear to arise from the two types of residues and the loss of the 320 nm peak upon denaturation in urea is consistent with this supposition. The disruption of conformation occurred at 3-5 M urea and indicates less stability for this preparation which is not surprising in view of the extensive modification of residues. In conclusion, this study of aminated tyrosyl residues on RNAase indicates that it should be possible to determine from the fluorescence characteristics the location of such residues in buried or exposed loci on proteins and changes in location upon change of conformation. Contrary to the situation with tyrosine whose fluorescence maximum is unchanged by the environment of the residue, the emission from the aminotyrosyl residue is sensitive to the environment in a fashion that is analogous to that for tryptophanyl residues. It should be noted that the range of emission maxima for aminotyrosyl residues of 320-370 nm as well as the optimal activation at 288 nm permits one to distinguish fluorescence of aminotyrosyl residues from fluorescence of tyrosyl residues ( A l E - - 276/308 nm). However, if tryptophan is present in the protein its fluorescence ( A / E = 280/320-350 nm) would obscure the weaker fluorescence of the aminotyrosyl residues.
478 REFERENCES 1 Cowgill, R. W. (1971) Photochem. Photobiol. 13, 183-194 2 Seagle, R. L. and Cowgill, R. W. (1976) Biochim. Biophys. Acta 439, 4 6 1 4 6 9 3 Wyckoff, H. W., Hartman, K. D., Allewell, N. M., lnagami, T., Johnson, L. N. and Richards, F. M. (1967) J. Biol. Chem 242, 3984-3988 4 Cowgill, R. W. (1966) Biochim. Biophys. Acta 120, 189-195 5 Beaven, G. and Gratzer, W. (1968) Biochim. Biophys. Acta 168, 456-462 6 Crook, E. M., Mathias, A. P. and Rabin, B. R. (1960) Biochem. J. 74, 234-238 7 Gehrke, C. W., Kuo, K. and Zumwalt, R. W. (1971) J. Chromatogr. 57, 209-217 8 Sokolovsky, M., Riordan, J. F. and Vallee, B. L. (1967) Biochem. Biophys. Res. Commun. 27, 20-25 9 Riordan, J. F., Wacker, W. E. C. and Vallee, B. L. (1965) Biochemistry 4, 1758-1765 10 Nelson, C. A. and Hummel, J. P. (1962) J. Biol. Chem. 237, 1567-1574 11 Cowgill, R. W. (1966) Biochim. Biophys. Acta 120, 196-211 12 Cowgill, R. W. (1965) Biochim. Biophys. Acta 109, 536-543 13 Tanford, C., Hauenstein, J. D. and Rands, D. G. (1956) J. Am. Chem. Soc. 77, 6409-6413