ARCHIVES OF RIOCHEMISTRY AND BIOPHYSICS

Vol. 276, No. 2, February 1, pp. 527-530, 1990

Enzymatic Generation of Triplet Acetone by Deglycosylated Horseradish Peroxidase

Eduardo Silva,*,’ Ana M. Edwards,* and Adelaide Faljoni-Alariot

*Facultad de &mica, Pontificia Universidad Catdlica de Chile, Casilla 6177, Santiago, Chile; and TDepartment of Biochemistry, Znstituto de Q&mica, Universidade de Scio Paulo, C. P. 20780, 01498 SGo Paula, S. P. Brazil

Received July 7, 1989, and in revised form September 27, 1989

A study of the effects of deglycosylation of horserad- ish peroxidase on protein conformation, as well as on its catalytic activity of oxidation of isobutyraldehyde or its enol form to triplet acetone and formic acid, was performed. The loss of carbohydrates leads to struc- tural modifications of this enzyme. This is confirmed by a change in the circular dichroism spectrum, an in- crease in tryptophan’s environment polarity, and a loss of the chiral specificity toward D- and rJ-tryptophan. Deglycosylation does not destroy either the peptide backbone or the amino acid residues and does not affect

erate an electronically excited triplet carbonyl com- pound when acting as an oxidase upon appropriate sub- strates (4). It was thus of interest to compare the behavior of native HRP to that of deglycosylated HRP with regard to the generation of-and emission from- these excited states. The reaction selected for this study was the well-known HRP-catalyzed oxidation of isobu- tyraldehyde to triplet acetone and formic acid:

(CH:&CH-C<’ = (CH,),C=C(OH [l] H H

the heme group content of the protein. The rates of oxy- gen uptake and light emission observed when horserad- HRP + H,O, + HRP - compound ish peroxidase oxidizes isobutyraldehyde or the tri- methylsilyl enol ether form of the latter are reduced /

OH

HRP-compound I + (CH&C=C\ -+ when the enzyme is 70% deglycosylated. Concomi- H tantly, the acting deglycosylated enzyme becomes inac- tivated during the course of the reaction. It appears that the carbohydrate moiety plays an important role in the HRP-compound II + (CH:J& - (

protection of the peroxidase from damaging effects in- duced by triplet acetone and in the stabilization of the

/ OH

three-dimensional structure of this enzyme. CC’ 1~90 HRP-compound II + (CH,),C=C\ --f Academic Press, Inc. H

It is well established that horseradish peroxidase (HRP)’ is a glycoprotein, the carbohydrate moieties be- ing important for both the solubility and the efficiency of the enzyme (1).

This protein consists of 308 amino acids, one heme group, and eight neutral carbohydrate side chains at- tached to asparagine residues. The amino acid sequence and the carbohydrate composition are known (2,3). One of the novel features of peroxidase is its capacity to gen-

’ To whom correspondence should be addressed. ’ Abbreviations used: CD, circular dichroism; DBAS, 9,10-dibro-

moanthracene-2-sulfonate ion; HRP, horseradish peroxidase; TF- MSA, trifluoromethanesulfonic acid.

000:~-9861/90 $3.00 Copyright (~1 1990 hy Academic Press, Inc. All rights of reproduction in any form reserved

[21

I0 \H ‘31

HRP + (CH&C - C;; H [41

I0 HRI’

(CH,),C-C\H + O2 + “CHs -CO-CH,

+ HCOOH, [ 51

where the superscript 3 stands for triplet acetone. Since the true substrate in this reaction is the enol

form of the aldehyde (5) the trimethylsilyl enol ether of isobutyraldehyde was preferentially used as substrate.

MATERIALS AND METHODS

Trifluoromethanesulfonic acid (TFMSA) and anisole were obtained from Aldrich Chemical Co. and used without further purification.

527

528 SILVA, EDWARDS, AND FALJONI-ALAR10

TABLE I

Carbohydrate Content, Fluorescence Emission Maxima (X,, = 295 nm), and Fluorescence Quantum Yields of Native Peroxidase and of Peroxidase Previously

Treated with Trifluoromethanesulfonic Acid

TFMSA incubation % time (h) Carbohydrate

0 18.0 330 0.003 1 9.7 335 0.016 2 6.9 335 0.032 3 5.5 335 0.033

Horseradish peroxidase (Type VI) and urea (grade of molecular biol- ogy reagent) were purchased from Sigma Chemical Co. Isobutyralde- hyde (Carlo Erba) was purified by distillation under nitrogen. The tri- methylsilyl enol ether of isobutyraldehyde was prepared according to House et al. (6, 7) and purified by fractional distillation at reduced pressure. 9,10-Dibromoanthracene-2-sulfonate (sodium salt) was pre- pared according to Battegay and Brand (8).

The deglycosylation was performed according to Edge et al. (9). Quantitative carbohydrate analysis was achieved by gas-liquid chro- matography of the alditol acetates (10). Deglycosylated HRP is par- tially soluble in water. Solutions of HRP (native and deglycosylated) were initially prepared in a small volume of 5 M urea. The final urea concentration was 0.02 M. A control of native HRP dissolved in the same buffer was used in all the experiments. Initial solubilization of the native enzyme in 5 M urea had no effect on the activity of the final mixture. The standard reaction mixture for generating triplet acetone enzymatically was prepared as follows: 10 ~1 of a solution of 0.5 mM

horseradish peroxidase in 5 M urea was added to a mixture of 20 ~1 of 0.1 M EDTA, 40 pl of 3.2 X 10m3M H202, 1 ml of H,O, and 1.8 ml of 1.0 M phosphate buffer, pH 7.4. The reaction was initiated by the addition of 0.1 ml of a 2.2 M stock solution of isobutyraldehyde in ethanol. When the substrate was the enol form, the reaction mixture was as follows: 10 ~1 of a solution of 0.5 mM horseradish peroxidase in 5 M urea was added to a mixture of 20 ~1 of 0.1 mM EDTA, 40 ~1 of 3.2 X 10m3M Hz02, and 2.8 ml of 0.2 M phosphate buffer, pH 7.4. Initiation of reaction was with 0.1 ml of 0.7 M stock solution of the enol form of isobutyraldehyde in ethanol. TFMSA-treated HRP remains soluble after dilution in the reaction mixtures.

Chemiluminescence was measured in a Hamamatsu TV Photon Counter C-767. Absorption spectra were recorded on a Zeiss DMR-10 spectrophotometer. Oxygen consumption was measured with a Yellow Springs Model 53 oxygen monitor. Circular dichroism spectra were measured on a Cary 60 spectropolarimeter equipped with the Model 6002 CD attachment. For these spectra, solutions of HRP (native and deglycosylated) 1 mg/ml in 5 M urea, path length of 1 cm, were used. Amino acid analyses were carried out using a Beckman 120-C amino acid analyzer. Samples were hydrolyzed in sealed and evacuated tubes containing 6 N HCl and 4% thioglycolic acid, at 105°C for 22 h (11).

RESULTS AND DISCUSSION

The carbohydrate moiety of peroxidase was removed by incubation with TFMSA. It is a well-known fact that this treatment does not affect the peptide backbone (9, 12). Table I compares some of the properties of native HRP and of TFMSA-treated peroxidase. Upon incuba- tion with TFMSA, there is a 70% decrease in the carbo- hydrate content from an initial value of 18% (2) down to 5.5% after 3 h of incubation. Amino acid analysis shows

that there has been no modification in any amino acid residues.

In order to estimate possible conformational changes

induced by the deglycosylation procedure, the emission spectra of the Trp as inner probe in the different samples were measured. The single Trp residue (Trp-117) (2) of native HRP exhibits a fluorescence emission band with a maximum at 330 nm, a typical value of a Trp residue located in the interior of the protein structure (13). It is known that the environment of the Trp residue in a pro- tein has a direct influence on the wavelength of maxi- mum emission of fluorescence (A,,,), with a shift to longer wavelengths as the Trp residues become more ex- posed to the solvent (14, 15). After TFMSA treatment of native HRP there is a 5-nm shift in the X,,, to longer wavelengths. The quantum yield of tryptophan fluores- cence of native HRP is very low (4 = 0.003) due to effi- cient, energy transfer (4 = 0.85) between tryptophan and the heme group in the enzyme (16). From absorption measurements at 280 and 403 nm it was established that the procedure used to remove the carbohydrates does not modify the heme group content of the enzyme in any sig- nificant way. The higher quantum yields of Trp fluores- cence (Table I) must be a consequence of a change in the distance or relative orientation of these two chromo- phores. The values shown in Table I thus imply that the deglycosylated form of HRP undergoes some type of structural change. Figure 1 compares the CD spectra of native and deglycosylated peroxidase. Due to difficulties in dissolving 1 mg/ml of TFMSA-treated HRP, both samples were dissolved and measured in 5 M urea. Under these conditions native peroxidase retains its character- istic CD bands; but clearly the Soret and Tryptophan bands as well as other CD active bands disappear for the TFMSA-treated HRP. This fact can be attributed to the

16~10~ ~

t

32 x lo3

7 7 w

-16 x xl03

300 LOO 500

A (nm)

FIG. 1. Circular dichroic absorption spectrum of native (-) and deglycosylated peroxidase (. .) in 5 M urea and of native peroxidase (- - -) in 0.05 phosphate buffer pH 7.0. The left ordinate applies to the spectra below 250 nm.

DEGLYCOSYLATED HORSERADISH PEROXIDASE AND TRIPLET ACETONE 529

5.0 Y lo7 R

i

5.0 Y loL

10 x107

1.0 x loL

20 40

MINUTES

FIG. 2. The time dependence of the emission observed in the oxida- tion of the trymethylsilyl enol ether of isobutyraldehyde promoted by native peroxidase (-; ordinates at left), and by deglycosylated peroxi- dase (- -; ordinates at right).

denaturation effect of the media, which is favored by partial deglycosylation of the enzyme.

The HRP action over substrates such as isobutyralde- hyde or its enolic form can be experimentally followed by oxygen consumption and light emission due to the formation of triplet acetone (9,18).

The light emission for native HRP in the presence of 80 mM isobutyraldehyde was 2.4 X lo5 counts (total inte- grated emission). When the enzyme was replaced by its deglycosylated form a drastic decrease was produced, since an integrated emission of 5.1 X lo3 counts was ob- tained.

Using the much more reactive enol form of the sub- strate (25 mM), the emission of the deglycosylated HRP was appreciable, yet still smaller than that with the na- tive enzyme (Fig. 2). When a free hemin concentration equal to that in HRP was used light emission was not observed, nor was it when the substrates were used in the absence of enzyme.

It has been reported that the hemin can generate trip- let acetone which does not emit, since it would be efi- ciently quenched by the molecular oxygen in the media (17). The observed emission for the deglycosylated HRP can be explained only if the triplet acetone is enzymati- tally generated and is protected by the protein moiety as happens in the native enzyme (4, 18). When oxygen uptake was determined by using 80 mM isobutyralde- hyde the values were very low, a consumption of 2.99 X lo-*’ pmol O2 min-’ being obtained for the native en- zyme and only 3.6 X lo-” pmol 0, min-’ for the deglyco- sylated form (initial rates in both cases).

Figure 3 shows the oxygen uptake kinetics for the na-

enolic form of the substrate. The initial rate of oxygen uptake for the TFMSA-treated HRP was appreciable but smaller than that with the native HRP, and rapidly levelled off. The rapid drop in the rate of oxygen uptake early in the deglycosylated HRP-catalyzed reaction could be due to enzyme inactivation by triplet acetone.

The emission of the enol/HzOz/TFMSA-treated HRP system was not altered by added t-BuOOH, suggesting that hydroperoxide radicals are not involved in the inac- tivation of the enzyme.

Addition of 9,10-dibromoanthracene-2-sulfonate ion and tryptophan to the enol/H,O,/TFMSA-treated HRP system showed an increase in oxygen consumption (Fig. 3). This would indicate that the enzymatic activity is less affected during the course of the reaction. The effect found can be explained as a consequence of a diminished damage produced by triplet acetone, since DBAS and Trp are efficient quenchers of triplet acetone. Also shown is the rate of oxygen uptake when an equivalent amount of free hemin is used, which is lower than that observed for TFMSA-treated HRP. This should be due to lack of enzyme catalysis in the case of the hemin. Un- der the experimental conditions self-oxidation of the substrate was not observed in either of its two forms. In the case of native HRP (which did not show damage by triplet acetone) DBAS and Trp do not affect the oxygen consumption kinetics.

As in the case of triplet acetone generated by catalysis with native HRP (4) the emission from the TFMSA- treated HRP system is intensified by energy transfer to the 9,10-dibromoanthracene-2-sulfonate ion. A double- reciprocal plot of the effect of this ion concentration

80

70

‘.

i

-i , i L I 1 I

3 6 9

MINUTES

J

FIG. 3. Oxygen uptake during the oxidation of the trimethylsily- lenol ether of isobutyraldehyde catalyzed by deglycosylatedperoxidase (-). Also shown is the effect of 0.3 mM L-tryptophan (---) and 0.01 mM 9,10-dibromoanthracene-2-sulfonate ion (. .). For comparative purposes also shown is oxygen consumption with the free hemine

tive peroxidase and its deglycosylated form using the I ~ (- - - -) and with the peroxidase in the native form (-. -. ).

SILVA, EDWARDS, AND FALJONI-ALAR10 530

2.0

IO T

1.5

1.0 L

1.0 x 10.‘ 2.0 x d 3.0 x 10-L

[TRYPTOPHAN], M

FIG. 4. Stern-Volmer plot for the quenching of triplet acetone by D- (-) and L-tryptophan (- -). Triplet acetone was generated by oxi- dation of the trimethylsilyl enol ether of isobutyraldehyde catalyzed by deglycosylated peroxidase and monitored by energy transfer to the fluorescent state of the 9,10-dibromoanthracene-2-sulfonate ion.

upon the intensity of the sensitized emission provides a Y-intercept/slope ratio (&T) of 6 X lo5 M-’ (not

shown). The fact that this value is similar to that ob- tained with the native enzyme (2 X lo5 Me’) (19) calls for comment. In the native enzyme, the excited species are protected from oxygen collisions (19). In the TFMSA- treated enzyme, however, the structure has been greatly altered, such that the excited species should be much more exposed to dissolved oxygen. This also can be in- ferred from the fact that the ratio of emission intensity to the rate of 0, uptake decreases by one order of magni- tude upon going from the native peroxidase-catalyzed reaction to the TFMSA-treated enzyme-catalyzed reac- tion (initial stages before inactivation). It would there- fore appear that a compensating effect is operative in determining the kETr values. Due to quenching of triplet acetone by oxygen a lower value of kETr would be ex- pected due to a lower T value. The fact that a threefold increase was observed indicates an increase in kET. We suggest that transfer to 9,10-dibromoanthracene-2-sul- fonate ion is facilitated in the case of TFMSA-treated HRP, presumably because this acceptor can approach (or bind) more closely to the site where the excited spe- cies is generated.

As another criterion to evaluate the effects of remov- ing the carbohydrate moiety, the quenching by D- and L-tryptophan was studied. It was found that TFMSA- treated HRP no longer exhibits chiral discrimination in the quenching by D- and L-tryptophan as shown in Fig. 4. It is known that triplet acetone generated within the native enzyme shows differential quenching by these stereoisomers (20). This difference is probably a conse-

quence of the conformational changes observed. Since the TFMSA-treated enzyme would have a less compact conformation than native HRP, both stereoisomers will show similar accessibility to the reactive species.

The results shown in this work allow us to state that the carbohydrate moiety plays an important role at the structural level. This explains the conformational changes observed when the carbohydrate is removed.

The deglycosylated enzyme had a lower oxygen con- sumption and light emission with respect to that ob- served on native HRP, and also showed a depressed pro- tection toward triplet acetone generated by the enzyme itself (Figs. 2 and 3). This allows us to conclude that the carbohydrate moiety in native HRP plays an essential protective role against the deactivating action of the ex- cited species produced by the enzyme.

ACKNOWLEDGMENTS

The authors express their deep gratitude to Professor Giuseppe Ci- lento for his interest in this work and for many helpful discussions, and to Professors Elsa Abuin (University of Santiago de Chile) and Frank H. Quina (University of S&o Paulo) for critical reading of the manuscript. This work was supported by grants from FINEP (Rio de Janeiro), FAPESP (Sio Paulo), CNPq (Brasilia), FONDECYT (Chile), DIUC (P. Universidad Catolica de Chile), the Volkswagen Foundation (Hannover, FRG), and GTZ (Eschborn, FRG).

REFERENCES

1. Tadros, L. K., Andrei, D., and Motas, C. (1983) Reu. Roum. Bio- chim.20,293-300.

2. Welinder, K. G. (1976) FEBS Lett. 72, 19-23. 3. Clarke, J., and Shanon, L. M. (1976) Biochim. Biophys. Acta 427,

428-442.

4. Cilento, G. (1984) Pure Appl. Chen. 56,1179-1190. 5. Adam, W., Baader, W. J., and Cilento, G. (1986) Biochim. Biophys.

Acta 881,330-336. 6. House, H. O., Czuba, L. J., Gall, M., and Olmstead, H. D. (1969)

J. Org. Chem. 34,2324-2336. 7. House, H. O., Gall, M., and Olmstead, H. D. (1971) J. Org. Chem.

36,2361-2371.

8. Battegay, M., and Brandt, P. (1923) Bull. Sot. Chim. Fr. 33,1667- 1678.

9. Edge, A. S. B., Faltynek, C. R., Hof, L., Reichert, L. E., Jr., and Weber, P. (1981) Anal. Biochem. 118,131-137.

10. Kim, J. H., Shome, B., Liao, T., and Pierce, J. G. (1967) Anal. Biochem. 20,258-274.

11. Matsubara, H., Sasaki, R. M. (1969) Biochem. Biophys. Res. Com- mun. 35,175-181.

12. Harrison, R. A. (1983) Biochemistry 22,5001-5007. 13. Burstein, E. A., Vedenkina, N. S., and Ivkova, M. N. (1973) Photo-

them. Photobiol. 18,263-279. 14. Sun, M., and Song, P.-S. (1977) Photochem. Photobiol. 25,3-g. 15. Cowgill, R. W. (1967) Biochim. Biophys. Acta 133,6-18. 16. Brunet, J. E., Gonzalez, G. A., and Sotomayor, C. P. (1983) Photo-

them. Photobiol. 38,253-254. 17. Augusto, O., and Bechara, E. J. H. (1980) Biochim. Biophys. Acta

63 1,203-209. 18. Cilento, G. (1988) Photochem. Photobiol. 48,361-368. 19. Bechara, E. J. H., Oliveira, 0. M. M. F., Durin, N., Baptista, R. C.,

and Cilento, G. (1979) Photochem. Photobiol. 30, 101-110. 20. Rivas-Suarez, E., and Cilento, G. (1981) Biochemistry 20, 7329-

7333.