Biochimica et Biophysica Acta, 1041 (1990) 83-93 83 Elsevier

BBAPRO 33753

Interaction of aromatic donor molecules with manganese(III) reconstituted horseradish peroxidase: proton nuclear magnetic

resonance and optical difference spectroscopic studies

Anil Saxena, Sandeep Modi, Digambar V. Behere and Samaresh Mitra Chemical Physics Group, Tata Institute of Fundamental Research, Colaba, Bombay (India)

(Received 25 January 1990)

Key words: Manganese(III) reconstituted horseradish peroxidase; Aromatic donor molecule; NMR, 1H-; Optical difference spectroscopy

The interaction of aromatic donor molecules with manganese(HI) protoporphyrin-apohorseradish peroxidase complex [Mn(IH)HRP] was investigated by optical difference spectroscopy and relaxation rate measurements of tH resonances of aromatic donor molecules (at 500 MHz). pH dependence of substrate proton resonance line-widths indicated that the binding was facilitated by protouation of an amino acid residue (with a pK a of 6.1), which is presumably distal histidine. Dissociation constants were evaluated from both optical difference spectroscopy and tH-NMR relaxation measure- ments (pH 6.1). The dissociation constants of aromatic donor molecules were not affected by the presence of excess of 1-, C N - and SCN- . From competitive binding studies it was shown that all these aromatic donor molecules bind to Mn(IH)HRP at the same site, which is different from the binding site of I - , C N - and SCN-. Comparison of the dissociation constants between the different substrates suggests that hydrogen bonding of the donors with distal histidyl amino acid and hydrophobic interaction between the donors and active site contribute significantly towards the associating forces. Free energy, entropy and enthalpy changes associated with the Mn(][II)HRP-substrate equilibrium have been evaluated. These thermodynamic parameters were found to be all negative. Distances of the substrate protons

o

from the paramagnetic manganese ion of Mn(III)HRP were found to be in the range of 7.7 to 9.4 A. The K d values, the thermodynamic parameters and the distances of the bound aromatic donor protons from metal center in the case of Mn(III)HRP were found to be very similar as in the case of native Fe(III)HRP.

Introduction

Horseradish peroxidase (hydrogen peroxide:hydro- gen peroxide oxidoreductase, EC 1.11.1.7) is a plant hemoprotein. The enzyme catalyzes the oxidation of a variety of inorganic and organic substrates by hydrogen peroxide or related compounds [1]. The heme group in HRP is ferric protoporphyrin IX, which is bound to the protein through imidazole of histidine residue by Fe- N(His). The primary structure of HRP has been de- termined by the composition [2] and sequencing [3] of amino acids. However, little information is available on the secondary or tertiary structure though the enzyme

Abbreviations: Mn(III)HRP, manganese (III) protoporphyrin-apo- horseradish peroxidase complex; HRP, horseradish peroxidase; LPO, lactoperoxidase.

Correspondence: S. Mitra, Chemical Physics Group, Tata Institute of Fundamental Research, Homi Bhabha Road, Colaba, Bombay-400005, India.

has long been the subject of several ESR [4], MCD [5], resonance Raman [6] and NMR spectral studies [7]. Ligand binding studies [8] have also been used to char- acterize the heme pocket of HRP.

The mechanism of oxidation of substrates by hydro- gen peroxide catalyzed by native ferric peroxidase is now well established [1]. It involves the formation of two intermediates, called compound I and compound II, by reaction with hydrogen peroxide [1,9]. The oxida- tion of the substrate involves the initial binding of the substrate to the enzyme. The interaction of the oxidiz- able substrates with peroxidases, for example, HRP and LPO (lactoperoxidase) has been studied by optical dif- ference spectroscopy [10,11], NMR spectroscopy [11-13] and kinetic measurements [14,15]. We have recently studied the binding of aromatic donors to LPO to obtain information about the relative dispostion of the donors in the enzyme, the nature of the ionizable group in the heme crevice, strength of binding of these donors and distances of the donor protons relative to the Fe(III) centre. These donors were found to interact with

0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

84

the enzyme in 1:1 stoichiometry and bind to LPO or HRP away from the metal centre, the binding being stronger in the latter. The distances of the donor pro- tons from the iron centre were found to be in the range of 9.4 to 11.1 .~ in the case of LPO [16] and 7.0 to 9.2 .~ in the case of Fe(III)HRP [8,17].

Manganese porphyrin reconstituted hemeproteins have long been studied [18,19] due to the close relation- ship of manganese with iron. However, these studies have largely been confined to spectrophotometric and conforrnational studies pointing that the protein confor- mation is probably restored to the original [20]. In Mn(III)HRP one axial position of metal is occupied by water and the other by the imidazole group of histidine as in most hemeproteins [21]. The manganese proteins show mixed biological activity. Manganese recon- stituted cytochrome bs, hemoglobin and myoglobin do not show any activity because of the absence of ap- propriate redox potential and low spin state. However, in manganese reconstituted peroxidases some activity is observed as higher oxidation states, for example, Mn xv is accessible. As such, Fe(III)HRP and Mn(III)HRP might be similar in structure [22], and a study of their binding to exogenous substrates may throw some light on their structures.

The optical spectrum of the Mn(III) porphyrin- apohorseradish peroxidase complex, [Mn(III)HRP] does not show any change upon addition of even a large excess of substrates, such as I-, CN- and SCN- [19]. We have recently studied [21] them using 15N nuclear magnetic resonance relaxation measurements, which give K a = 42 mM and 156 mM at pH 4.0 for cyanide and thiocyanate binding, respectively. A comparison of this data with that of CN- and SCN- binding to native Fe(III)HRP [23] reveals that the binding of cyanide to [Mn(III)HRP] is weaker but the thiocyanate binding is almost similar in magnitude. The substrates (CN- and SCN-) bind to Mn(III)HRP away from the Mn ion in Mn(III)HRP as is the case for the SCN- in native Fe(III)HRP, but is different from CN- binding to Fe(III)HRP, since CN- binds to the iron centre of Fe(III)HRP and changes its spin state [24,25]. The 15N distance of thiocyanate and cyanide from the para- magnetic centre was found to be 6.9 and 6.6 ,~, respec-

o

tively, in [Mn(III)HRP] [21] compared to 6.8 A (SCN-) in Fe(III)HRP [23].

In the present study, we report results of our investi- gations on the interaction of different aromatic donor molecules with [Mn(III)HRP], using optical difference spectroscopy and proton magnetic resonance tech- niques. Dissociation constants and thermodynamic parameters for the binding of donors to the enzyme have been calculated. Distances of the donors from the manganese centre of the enzyme have also been de- termined. The results are compared with those of Fe(III)HRP and discussed in relation to hydrogen

bonding, hydrophobic interaction as well as ionization of distal amino acid residue.

Materials and Methods

All reagents used were of analytical grade. Deu- terium oxide (> 99.85%) was purchased from Aldrich. All solutions were made in double distilled water. Crude HRP (R z (A403//.4280) = 0.8) was purchased from Sigma and purified by the method of Shannon [26] to an R z value of 3.1. Mn(III)protoporphyrin IX-apohorseradish peroxidase complex [Mn(III)HRP] was prepared and purified by the method described earlier [21].

Optical difference spectroscopy Difference optical spectra (enzyme-donor complex

vs. enzyme) were obtained on a Shimadzu 2100 spectro- photometer equipped with a TCC-260 temperature con- troller. Quartz cells of 10 mm path length were used. All titrations were carried out at room temperature (23 ° C), unless mentioned otherwise, by adding 10-500 #1 of aliquots of the donors (50-250 mM) to the enzyme (10 /xM) solution (2 ml, 0.1 M phosphate buffer, pH 6.1) in sample cell and by diluting the enzyme by the same amount of buffer in the reference side. Dissociation constants were calculated using Eqn. 1 [16,17]:

1 [ K D ] I 1 a-~ = ~ ~ + a A ~ (1)

where AA and A A ~ are the changes in absorption at the observation wavelength at given and saturating donor concentrations, respectively. S O is the initial con- centration of the donor and K D is the dissociation constant of the enzyme-donor complex. K D and A A ~ can be evaluated from the slope and intercept of 1 / A A vs. 1 / S o plot.

N M R measurements 1H-NMR were recorded on 500 MHz Bruker AM

FTNMR spectrometer at 23 ° C. Samples were dissolved in 0.1 M phosphate buffer in 1:)20 (pH 6.1) to a final vol. of 0.4 ml. Spectra were obtained by accumulation of 40-200 transients at 16K data points in quadrature mode. Proton chemical shifts were referred to the pro- ton signal of trace HDO.

Line-width measurements. The line-width data were obtained from the spectra by fitting the donor proton resonances to the Lorentzian line shape. The Apl/2 (obs) is considered as the sum of the line-width due to the free and enzyme-bound fractions of the donor pro- ton resonances, assuming chemical shift difference to be negligible. Enzyme donor interaction is considered to take place between the protonated form of the enzyme and the donor, involving reaction of only one proton per molecule of the donor. The donors were assumed to

be present in the neutral form. Thus, the variation of the observed line-width as a function of pH is given by equation [16]:

APl/2(obs) = S 0 + K(1 + K a / H + ) (2)

where Av~/2 and AvF/2 denote the line-widths of the enzyme bound and free donor proton resonances, re- spectively. E 0 and S o represent the initial enzyme and donor concentrations, respectively. K a denotes the pro- tolytic dissociation constant of the enzyme. K denotes the dissociation constant for binding of donor to acidic state of the enzyme, and is related [27] to KD through K D = K(1 + KJH +).

Relaxation rate measurements. Both Mn(III)HRP and donor solutions were treated with Chelex-lO0 (Bio-Rad) to remove traces of free metal ions as described earlier [11,28]. Treated solutions were lyophilized and redis- solved in D20 for NMR studies. Inversion recovery method with 180 ° - z - 9 0 ° pulse was used to obtain the longitudinal relaxation time (Tlobs) as described earlier [11,16,21,23].

Calculation of K o from 1H-T1 NMR measurements The Tlobs Can be considered as the sum of the relaxa-

tion rates of the bound and free fractions of the donor molecule and is related to K o, Tlb and Tlf according to the equation given below, provided that only one donor molecule interacts with one molecule of the enzyme [11,16]:

[1 111 [1 111 rl 111 Eo r,o , =rD (3)

where Txb is the T 1 of the [Mn(III)HRP]-donor complex and Tlf is the T 1 of the free donor. The K D values for the binding of different donors to the enzyme can be obtained by linear least-squares fit of the data to the above equation.

Thermodynamic parameters The change in free energy upon binding of the donor

to [Mn(III)HRP] was obtained by following equation:

A G = R T In g D (4)

K D values at different temperatures were obtained from Eqn. 3 by carrying out T1 measurements (Tlobs. and Tu) at different temperatures. Change in the en- thalpy was evaluated from the slope of the Arrhenius plot [10,16] given by following equation:

l n (1 /KD) = -- - ~ ( l / T ) + Constant (5)

85

Change in entropy (AS) was calculated from Eqn. 6, assuming ACp was constant.

AG ffi A H - TA S (6)

Distance calculations using 1H-T1 measurements The relaxation rate Tlb is related to Tlm through

following equation [16,21,23]:

1 ,T,-llb 1 -- ,.r,-lld 1 = T l m + "rm (7)

If the chemical exchange is rapid as compared to the relaxation rate (l/Tim), em (lifetime of the enzyme- donor complex) can be neglected [16,21,29,30]. The diamagnetic component Tld has been shown to be negligibly small as compared to Tim [11,16,21,23]. Tlb obtained from Eqn. 3 may be regarded as Tim. The metal-proton distance for Mn +3 ( S = 2) in [Mn(III) HRP]/donor complex is obtained by the Solomon and Bloembergen equation [31,32] as follows:

[ / 3% 7% \ 3 ~/6 r(/~) = 7 6 3 i T e m | , , . ~ 2 _ 2 + , , . ~ 2 _ 2 / ~

[ ~ 1 + ~oi¢c I + ¢o~¢ c 1 ) (8)

The value of % was calculated from ratio of T2m//Tlm

as described earlier [11,16,21,33]. The transverse relaxa- tion time (T2b) of the [Mn(III)HRP) bound donor pro- ton was evaluated from the line-width of proton reso- nances of donors assuming the lines to be Lorentzian at different enzyme and donor concentrations and the chemical shift difference between the bound and free proton resonance to be negligible. T2b Was calculated using an equation similar to Eqn. 3. If the diamagnetic contribution is neglected T2b becomes T2m.

Competitive binding of aromatic donors in presence of inorganic substrates

Inorganic and organic substrates competing for bind- ing to the [Mn(III)HRP] at the same site as that of aromatic donor molecules affect the dissociation con- stant of the latter. The observed dissociation constant of the aromatic donor is related to the competitive inhibi- tor concentration [I] by the following expression [34]:

KD[I] KD(Obs ) ffi K D -I Ki (9)

where KD(Obs ) is the dissociation constant of the aromatic donor binding to the [Mn(III)HRP] in pres- ence of inhibitor, and K i the dissociation constant of inhibitor in absence of the aromatic donor. The values of Ki of the inhibitors were deduced from the intercept and slope of the straight line plot of KD(obs) vs. [I] of Eqn. 9.

86

Results

Optical spectroscopy of [Mn(III)HRP]donor complexes The structures and numbering scheme for hydrogen

atoms of the aromatic donor molecules used in this study are given in Fig. 1. The absorption spectrum of the Mn(III)HRP] in the Soret and visible regions alters only slightly in the presence of even a large excess of the aromatic donor molecules. This suggests that aromatic donors do not bind directly to the metal centre and perturb the electronic structure of the Mn(III) which remains essentially high spin with S = 2. Fig. 2 il- lustrates typical optical difference spectra between [Mn(III)HRP] and its complexes with p-cresol and re- sorcinol, respectively, in the range 365 to 450 nm at different degrees of saturation. The double reciprocal plot of 1/AA vs. 1 / S 0 for p-cresol and resorcinol titrations are shown in Fig. 3. Observation of the straight line shows the binding of one tool of donor per mol of [Mn(III)HRP] [16]. Dissociation constants (KD) of vari- ous donors were evaluated by a least-squares fit of the data to Eqn. 1 and are given in Table I.

Interaction of donor molecules studied by NMR Separate signals for the bound and free donor pro-

tons were not observed, indicating that the exchange between the bound and free ligand was fast on the NMR time scale. Fig. 4 shows that the line-width of resorcinol protons (H2, H4 and H6) varies linearly with the [Mn(III)HRP) concentration. The line-width of H2 proton increases from 5.2 to 22.5 Hz and of H4 or H6 proton from 4.3 to 17.8 Hz when the enzyme concentra- tion is increased from 0 to 40 /~M. This is consistent with fast ligand exchange.

<~ <3

0.04 A

/

011 360 380 400 420 440

(nm) ~ i i ; i I r i i i i i I i i i I i i q I i i t i i ~ I ~ r i

o.o4i B

0.02

-002

-00~ 380 400 420 440

(nm) Fig. 2. Optical difference spectra for Mn(lIl)HRP/resorcinol (A) and Mn(lll)HRP/p-cresol (B) complexes. Initial condition is same in both cases, i.e., 2 ml of Mn(III)HRP (10/tM) in 100 mM phosphate buffer (pH 6.1). Titrations were carried out by sequential addition of increas- ing concentrations of resorcinol/p-cresol to final concentrations of

1.5 mM and 0.35 raM, respectively.

< <3 0 0 0

OH OH OH OH OH

H H H H CH3

Phe nol Resorcinol o- Cresol m - C resol p- Cresol

NH 2 NH 2 NH 2 NH2

H H H CH 3 H H H H

H ~ ~ 4 3 H H ~ ~ 4 3 H H -CH 3 H 5 ~ ~ 4 3 H

H H H CH 3

Anil ine o -Toluidine m-Toluidine p- Toluidine

Fig. 1. Structures of various aromatic substrates showing the numbering of protons.

87

6( :

50

40

30

120

lO0

80

6 0

40

A

I I I I I I I 100 200 300 400 500 600 700

(M -1 ) + So

B

I I I l I 3 9 15 21 2 7

l (x 162 M-I) So

Fig. 3. Variation of change in absorbance (AA) at 430 nm with substrate concentration (So) for resorcinol (A) and p-cresol (B).

The line-widths of the bound donor proton reso- nances were also found to vary with the pH of the solution. Fig. 5 shows the effect of pH (between 3.8 to

20

16

N -1-

--~ 12

4 ¢

i I t I I I I I I0 20 30 4O

[Mn (111) H RP] (/.t.M)

Fig. 4. Variation of ~H-NMR line-width of H2, H4 and H6 protons of resorcinol with IVln(III)HRP concentration. Resorcinol concentration was 20.5 mM in 100 m M phosphate buffer, pH 6.1. Concentration of

Mn(III )HRP was changed from 0 to 40 ~tM.

10.0) on the line-widths of H2 and H4 protons of resorcinol in the presence of [Mn(III)HRP]. Deprotona- tion of an amino acid residue reduced the line-width of the H2 proton from 8.5 I-Iz at pH 3.8 to 4.9 Hz at pH

TABLE I

Dissociation constants (KD) and relaxation times (Tlb) for substrates binding to [Mn(II1)HRP] at 296 K and their comparison with Fe(III)HRP

Substrate K D (raM) b and Tlb (s) b ValUes obtained by N M R measurements K D (raM) from optical for each proton of donor difference spectroscopy

H2 H3 H 4 / C H 3 H5 H6 Mn-HRP Fe(III)HRP

Resorcinol K D 4.6 - 4.8 4.9 4.8 4.7:1:0.5 6.7 [10] c Tlb 0.017 - 0.011 0.023 0.011 4.0 [38]

Phenol a K D 8.2 . . . . 7.7 :t: 0.9 9.4 [10] T1b 0.010 . . . . 6.2 [38]

p-Cresol K D 2.8 3.0 3.2 3.0 2.8 2.0 + 0.4 3.5 [10] T1b 0.007 0.009 0.018 0.009 0.007 2.5 [17]

o-Cresol K D 4.8 4.1 4.1 4.1 4.9 4.5 + 0.3 Tlb 0.015 0.009 0.009 0.010 0.008

m-Cresol K D 5.5 5.2 5.5 4.8 4.8 5.1 + 0.5 Tlb 0.007 0.016 0.007 0.008 0.008

Aniline a K D 18.2 . . . . 12.1 :t: 3.8 19.0 [10] T~b 0.0109 . . . . 16.1 [17]

p-Toluidine K D 9.2 8.9 9.4 8.9 9.2 8.8 + 0.7 9.9 [10] Tlb 0.007 0.009 0.018 0.009 0.007 8.8 [17]

o-Toluidine K D 16 14 14 17 14 16 + 3 17.6 [38] T~b 0.014 0.008 0.008 0.009 0.008

m-Toluidine K D 20 17 20 17 17 19 + 3 Tlb 0.007 0.017 0.007 0.009 0.009

" Average values are given due to overlap of particular protons. b All Tlb values are accurate to +0.001 s except those for H2, H6 and CH 3 protons of p-toluidine and H3, H5 protons of p-cresol, where accuracy

is +0.002 s. The error in K D values (NMR) is 2-10%. c Numbers given in the square brackets refer to appropriate reference.

88

TABLE II

Temperature dependence of dissociation constants (KD) and relaxation times (TIb) for resorcinol binding to [Mn(IlI)HRP]

Error in Tlb ± 0.001.

Temperature (K)

K D (raM) and T l b (S) ValUeS at different temperatures

NMR spectroscopy Optical difference

H2 H4 H4 H6 spectroscopy

280 K D 2.5 ± 0.1 2.5 ± 0.3 2.4 ± 0.3 2.6 ± 0.2 2.5 ± 0.4 Tlb 0.014 0.009 0.019 0.009

296 K D 4.6 ± 0.2 4.8 ± 0.1 4.9 ± 0.6 4.8 ± 0.3 4.7 ± 0.5 Tlb 0.017 0.011 0.023 0.011

305 K D 6.6±0.5 6.9±0.8 6.1 ±0.3 7.1 ±0.2 6.8±0.4 Tlb 0.018 0.012 0.025 0.012

315 K D 9.1 ± 0.9 9.6 ± 0.5 8.5 ± 0.9 8.2 ± 0.6 10.1 ± 0.9 Tlb 0.021 0.014 0.028 0.014

TABLE III

Dissociation constants (KD) for resorcino~ phenol and aniline binding to [Mn(III)HRP] in presence of I -, SCN -, C N - p-cresol and p-toluidine

Substrate K D KD (raM) in presence of

(raM) I - SCN- C N - NO~- p-Cresol p-Toluidine

Resorcinol 4.7±0.3 5.1±0.2 5.9±0.8 4.9±0.6 5.9±0.6 2 ~ ± 2 3 57± 9 Phenol 8.0±0.4 8.0±0.2 8.9±0.6 9.3±0.9 8.4±0.8 328±39 97±10 Aniline 16 ±3 17 ± 2 19 ± 4 18 ±3 18 ± 2 656±71 194±23

10.0, and for H4 proton from 8.1 Hz at pH 3.8 to 4.5 Hz at pH 10.0. Least-squares fit of the data to Eqn. 2 yielded pK a ffi 6.1. The hydrogen bonding interaction between the substrate functional group and protonated amino acid residue (with pK a --6.1) therefore appears to facilitate the binding of the substrate to the enzyme. The good fit to Eqn. 2, also confirms involvement of one proton reaction in binding of the substrate to the enzyme. No effect of pH on the line-width of free substrate proton resonances was observed. The binding of the substrate appears to be specific in the pH range of 5 to 8.5 and is linked with the state of protonation of the amino acid residue with pK~ = 6.1. Since optimum pH for the donor binding is around 6.1, the relaxation rate measurements reported here were carded out at pH 6.1 in 0.1 M phosphate buffer solution in D20.

The Tlobs ValUeS were calculated from 1H-NMR spectra of donors in the presence and absence of [Mn(III)HRP]. A typical set of inversion recovery spec- tra for resorcinol binding to [Mn(III)HRP] is depicted in Fig. 6A. A plot of [ E 0 ( 1 / T l o b s - 1/Tlf) -1] vs. S O is shown in Fig. 6B which is linear (e.g., Eqn. 3) and again confirms the binding of only one tool of donor per tool of [Mn(III)HRP] [11,16]. All the substrates listed in Table I show such a linear fit. The values of Tlb and KI) listed in Table I were obtained by a least-squares fit of the data to Eqn. 3. The KD obtained by relaxation studies are in good agreement with those obtained by optical difference spectroscopy. Aniline values obtained

by these two methods differ slightly; however, the de- termination of K D in this case from the NMR method may involve some error due to the overlap of proton r e s o n a n c e s .

Spin-lattice relaxation rates (TI-~, and T~ 1) for re- sorcinol/[Mn(III)HRP] were measured at different tem- peratures (280 to 315 K). Table II lists the relaxation rates of resorcinol protons (Tlb) and the dissociation constants for resorcinol binding to [Mn(III)HRP] at different temperatures. The relaxation rate decreases with an increase in temperature, suggesting the presence

8

~ 7

0 v

5

I i I I I A I 4 5 6 7 8 9 I 0

pH

Fig. 5. Effect of the pH variation on the ~H-NMR fine-width of resorcinol protons in prmmce of M n ( I ~ . The solid lines are le~t-squares fit to Eqn. 2 for reaction involving one proton with pK a = 6.1, A~/2 = 35 Hz for H2 and AP~I 2 ~ 27 l-lx for H4 and H6

protons.

89

oo, - - v i 1/3 11 I o /WVW'

A oo ° ° ' - y ' qtfv"

o.,. o.,,7 j 0.,7 , y 'V'---

0 . 2 ' .,,t4 - - W " ' - ~

0 . 3 ~

o . s 5 "---

O H 0 . 4 ' -

-

/ " o , - - ; ; ~ \ _~ .,L~ , ,~ . / \ i ,rlr JL_

Fig. 6. (A) A 180°..'r-~ ° pulse ]tt-NMR spectra (501) MI-~) of resorcinol (22 ~ in p r e ~ c e of 0.6 mM I~(III)HRP (101) ~ M l ~ p h a t e buffer, pH 6.1). • values were varied from 0.008 to 10 s.

of fast exchange (Tlm ~" Tin) [29,30]. This is also con- sistent with the observed low binding affinity of re- sorcinol to [Mn(III)HRP] (K v --4.7 raM). Activation energy of 2.0 kcal/mol was calculated from the slope of the Arrhenius plot (log (1/Tlb) vs. 1 / T , Refs. 21,23,30, 33) which is in the range of 1-3 kcal/mol expected for ~m independent processes [29,35]. The temperature-de- pendence of the line-width of 1H-NMR resonance of the resorcinol/[Mn(III)HRP] system shows an increase in the line-width with a decrease in temperature (data not shown), which is also consistent with the fast chem- ical exchange [11,36] of this system.

Binding of aromatic donors to [Mn(III)HRP] was also studied in the presence of 0.1 M of iodide, thio- cyanate, cyanide and nitrate ions. The dissociation con- stants were evaluated from the analysis of Tlob, and are listed in Table III. The results show that the dissocia- tion constants of the aromatic donors are not affected by the presence of these ions. The optical difference spectra of various aromatic donors are slightly different (see Fig. 2) suggesting that these aromatic donors may not be binding at the same site. Therefore, the binding of aromatic donors to [Mn(III)HRP] was also studied in the presence of each other. The K v values for the competitive binding studies are given in Table III. As can be seen, the K v values for the binding of re- sorcinol, aniline and phenol in the presence of p-cresol and p-toluidine are markedly increased. Thus, these two donor molecules (p-cresol and p-toluidine) effectively compete with the other aromatic donors for binding to [Mn(III)HRP], and hence all the aromatic donors bind

tO the enzyme at the same site in spite of the difference in their o p t i c a l difference spectral patterns. To confirm this, dissociation constants for the binding of these donors to [Mn(III)HRP] were measured with varying concentrations of second substrate (p-cresol and~p- tolnidine). The K v values for resorcinol, phenol and aniline binding to [Mn(III)HRP] were found to increase linearly with the concentration of second substrate (p-

9O B "~ [xS]

x CH2] T ~ 6 0

,,=_ I-- \ 5 0

I t 4 0 [ H 4 , H 6 ]

t-,-

--- 30 O

t,iJ 20

10

i ~ i i i I I i I 4 e 12 16 20 24 2e 32 36

S O (X 103M )

Fig. 6. (B) Plot of [F.e(l/Tlob, -1~Tit) -11 vs. S e for resorcinol H2, H4 and H6 phemyl protons. Concentration of r'~orcinol was varied from 4 to 30 mM and of Mn(III)HRP was varied from 0.02 to L0

mM.

90

cOOl- A

2O0

~ 150 E E

o ~ v

, i I i O ~ l ~lO

[ 'p- C ref, ol] (M)

B

o.o 0.04 o.o, o.,o

r p - l 'oluidine] IMI

Fi& 7. Plot of Kv(obs) vs. inhibitor concentration (p -c reso l , (A) and p- to lu id ine , (B)) for binding of reso~inol (o), phenol (~) and aniline (rl).

cresol, Fig. 7A and p-toluidine, Fig. 7B) concentration. The apparent dissociation constant (Ki) of p-cresol was calculated from the intercept and the slope of the plot (Fig. 7A) for resorcinol, phenol and aniline binding in presence o f p-cresol at diferent concentrations, and were found to be 2.8, 3.0 and 3.1 mM, respectively. These values agree very well with the apparent dis*.ocia- tion constants calculated earlier from an independent experiment ( T a b l e I ) . T h e K i v a l u e s ( 9 . 2 , 8 . 1 a n d 9 . 6

m M ) for p-tohiidine were calculated similarly, which are also in g o o d agreement ( T a b l e I ) .

The distance between each proton of the donor and the M n + 3 ion was calculated using a v a l u e o f ~'c = 3 .

10 - l l s which has been determined from the ratio of T2m to Tim [11,16,21,33]. The value employed earlier [8,13,17] for other hemeproteins is 5 .10 - l l s, which is close to the present one. The distances were calculated for each proton of different donors and are listed in Table IV. The distances o f proton of donors from the metal ion in [Mn(III)HRP] fall in the range of 7.7 and 9.4 A.

Knowledge of the changes in enthalpy and entropy gives additional information about the nature of the enzyme-donor association. These thc~mod,ymmdc pa- rameters were obtained from the values of ~ t i o n constants determined at different temperatures by the

TABLE IV

Distances (~) of donor protons from Mn + J ion in [Mn(III)HRP] using ¢cffi3.10 -n s

Donor 1-12/CH3 H 3 / C H 3 H 4 / C H 3 H5 H6

Phenol 8.2 . . . . Resorc ino l 8.9 - 8.3 9.4 8.3

(9.1) (8.4) (9.2) (8.4) p-Cresol 7.7 8.0 9.0 8.0 7.7

(7.0) * (7.7) (8.5) (7.7) (7.0) o-Cr'-~eol 8.7 8.0 8.0 8.1 7.9 m-Creso l 7.8 8.8 7.8 7.9 7.9 Aniline 8.3 - - - -

p -To lu id ine 7.8 8 .0 9.1 8.0 7.8 o -To ln id i - e 8.6 7.9 7.9 8.0 7.9

m-Tohf id ine 7.7 8.9 7.7 8.0 8.0

* Values in pareathe~ are for Fe(IIDHRP from Ref. 8 for r e u ~ o l

and from Ref. 17 for p-cresol.

T A B L E V

Thermodynamic parm,neters for formation of [Mn(III)HRP] donor com- plexes

Values in parenthesis are for Fe(III)HRP from Ref. 38.

Donor ~G A H AS (kJ /mo l ) (kJ /mol ) ( l / n l o l / ° K )

~ o l - 13 - 27 -47 (-14) (-32) (-64)

Phenol - 12 - 21 - 30 p-Cresol - 15 - 28 - 44 o-Cresol - 13 - 25 - 4 0 m-Creso l - 13 - 21 - 27 Aniline - - 10 - 15 - 17

p-Toluidine - 12 - 20 - 27 o-Tolu id ine - 10 - 18 - 27 m-Toluidine - 10 - 17 - 24

91

T 1 and optical difference spectroscopic measurements (Table II). The thermodynamic parameters (AG, AH and aS) were obtained from Eqns. 4-6 and are listed in Table V. The AG, AH and a S values are negative for all the aromatic donor molecules studied here. The values for the Fe(III)HRP/resorcinol system [8,17] are included for comparison.

Discussion

Even though the optical spectrum of Mn(III)HRP shows little change on addition of aromatic donors, a wealth of information is obtained from the optical difference spectra of Mn(III)HRP and its complex with aromatic donors. The values of KD so obtained are comparable to Fe(III)HRP.

The cyanide ion binds very strongly to the ferric ion of Fe(III)HRP at the sixth coordination position and gives low spin (S = 1/2) iron species [24,25]. Kinetic studies of cyanide and fluoride binding to Fe(III)HRP have identified the presence of an ionizable acid residue on the enzyme with pKa--4.1, 6.4 and 10.8 [24]. Re- sidues with pKa values 4.1 and 6.4 have been attributed to propionic acid of heine moiety and distal histidine, respectively. We have recently shown [21] that the affin- ity of cyanide to peroxidase is lowered by substitution of ferric ion by manganese and that the cyanide ion does not bind to the manganese. The cyanide, thiocyanate and iodide ions have been shown [21] to bind to Mn(III)HRP away from the manganese centre at the same site. The binding of these ions is facilitated by protonation of an acidic residue of [Mn(III)HRP] with pK~ = 4.0. This residue has been assigned to the propionic acid group of the heine moiety [21,37] and on protonation of this residue, the salt bridge breaks and thus gives easy access to the entry of anion near heme. As shown in the present study, the presence of these ions does not affect the binding of aromatic donors (Table III). This clearly indicates that the binding site for aromatic donors is different from that of these inorganic ions. Also, the binding site for all the aromatic donors is the same. The binding of aromatic donors is facilitated by protonation of an ionizable group with pK~ = 6.1. The ionizable group with pKa---6.1 was at- tributed to histidyl imidazole residue in the heme distal pocket, since the pK a of the free histidine amino acid is very close to it (6.04, Refs. 11 and 36). This amino acid residue in the distal side of the heine crevice has also been implicated in the binding of aromatic donors to ferric lactoperoxidase [16].

The donors with functional groups capable of form- ing a hydrogen bond may form a hydrogen bond with the NH group of distal histidine. Thus alcohols such as phenol, resorcinol and cresol have relatively greater binding affinities to the enzyme as compared to anilines (Table I). This is consistent with the phenols being

better hydrogen bonding donors than aromatic amines. Resorcinol binds to the enzyme much more strongly as compared to phenol because of the presence of two hydrogen bonding sites. These results point to the im- portant contribution of hydrogen bonding towards complexation.

Among the alcohols, p-cresol forms a stronger en- zyme-donor complex compared to phenol. Similarly, p-toluidine forms a stronger complex than aniline. This may be due to the increased hydrophobicity introduced by an alkyl (CH3) group, indicating that the binding may also be hydrophobic in nature. A similar effect of the hydrogen bonding and hydrophobic interaction was observed on the binding of aromatic donors to Fe(III)HRP [17] and ferric lactoperoxidase [16]. Among the three isomers of cresol, the binding of the p-cresol is strongest followed by the o- and m-cresol. This may be explained on the basis of the inductive and hypercon- jugative effects of the methyl group. Both of these effects operate in o- and p-positions and increase the electron density at O atom. Similar effects operate in toluidine also. However, more work is required to clearly define the relative significance of these effects in the binding of donors to the enzyme. The distances of protons of donors from the metal ion in [Mn(III)HRP] fall in the range of 7.7 to 9.4 .~. These distances compare very well with the distances in Fe(III)HRP/ donor complexes (Table IV). The distances of the vari- ous protons of the bound donors are such that the donor molecules appear to be inclined to the mean porphyrin plane of [Mn(III)HRP]. Binding of these donors in a way that the aromatic ring is parallel to the porphyrin ring would require the distances of various protons of donors to differ by a much larger degree than observed (Table IV) [16,38].

The identical nature of the thermodynamic parame- ters for the binding of the phenols and aromatic amines to Mn(III)HRP, indicates that the mode of binding between the enzyme and these donor molecules is simi- lar in all complexes. It is rather surprising that even when large difference in the KI~ is observed, the ther- modynamic parameters are very similar. The specific nature of the binding of the aromatic donors to Mn(III)HRP is shown by negative AG, AS and AH values. This trend is similar to that observed in Fe(III )HRP/donor systems [38], but different from Fe(III)cytochrome P450/camphor complex ( A G - - 3 2 kJ/mol, AH = 0, AS = 108 J/ tool per K). In the latter, the binding of camphor was shown to be tighter with a spin state change from S = 1 /2 to S = 5 /2 [39]. It has been reported [40] that addition of n-propyl mercaptan or imidazole to metmyoglobin at pH 7.4 produces re- markable changes in both light absorbancy and electron paramagnetic resonance spectra due to a change in the spin state from high spin to low spin. However, addition of excess of resorcinol, phenol and aniline to

92

metmyoglobin does not show any change in the optical spectrum of metmyoglobin. The negative free energy and enthalpy changes for all the aromatic donors bind- ing to Mn(III)HRP energetically favour the donor as- sociation. Analysis of the Fe(III)HRP/aromatic donor equilibria has pointed out that besides hydrogen bond- ing and hydrophobic interaction, contribution of the qr-~r interaction between the ligating aromatic donor and an aromatic amino acid residue on the protein moiety with favourable stericity is appreciably signifi- cant [38]. Recent 1H-NMR study on the aromatic donor/Fe(III)HRP interaction supports these argu- ments [8]. Computer modelling studies [8] in the case of Fe(III)HRP indicate that the aromatic donors bind to the enzyme near the heine peripheral 8-CH3, Tyr-185 and Arg-183 residues, the phenyl ring of the donor being perpendicular to the heine plane. Since tyrosine is also present in a similar position in turnip peroxidase and cytochrome c peroxidase, it seems possible that the tyrosine may play an essential role in the peroxidative action. Several kinetic studies on the reaction of Fe(III)HRP compound I with a variety of substrates suggest involvement of distal histidine (His-42) [41]. The substrate (p-cresol) is oriented by hydrogen bond- hag to distal histidine (His-42) and as an electron is transferred from the substrate to the porphyrin cation radical, a proton is transferred along the hydrogen bond to His-42 [42]. This results in homolytic cleavage of H - O bond of p-cresol with formation of a free radical of p-cresol [43] and ferric HRP compound II. The oxyferryl group of HRP compound II is hydrogen bonded to the protonated His-42 residue [44], which again forms a suitable path for electron transfer from substrate to ferryl group of compound II.

Because of the similarities in the K D values, the thermodyBamic parameters and the distances of the bound donor protons for Mn(III)HRP/Donor and Fe(llI)HRP/Donor systems, we are tempted to assume that the donor in Mn(III)HRP is placed in a pocket consisting of Tyr-185, Arg-183 and 8-CH 3 by hydro- phobic bonds with the benzene ring of Tyr-185 and hydrogen bond with distal histidine. The mechanism for oxidation of the substrate, wherein the neutral substrate donates an electron to HRP-1 and simultaneously loses a proton [41,42] is the same for both cases, Fe(III)HRP and Mn(III)HRP. In both cases substrates bind to His-42. The presence o f water at the sixth position in the case of Mn(III)HRP [21] may be responsible for its low activity as compared to native Fe(III)HRP. These similarities in behavior of Mn(III)HRP and Fe(III)HRP toward substrates support the assumption that the sec- ondary structure of the protein in Mn(III)HRP is re- stored to the same extent as in Fe(III)HRP, and that the porphyrin rather tha~n the metal ion is responsible for the restoration of the conformational integrity of the protein.

Acknowledgements

This work was done at the 500 MHz National NMR facility. The cooperation of the facility is gratefully acknowledged. The authors also wish to thank Ms. Uma Mahashwari for her assistance in the measurements of optical difference spectra.

References

1 Dunford, H.B. and Stillman, J.S. (1976) Coord. Chem. Revs. 19, 187-251.

2 Welinder, K.G., SmiUie, L.B. and Schonbaum, G.R. (1972) Can. J. Biochemistry 50, 44-51.

3 Welinder, K.G. (1979) Eur. J. Biochemistry 96, 483-502. 4 Colvin, J.T., Rutter, R., Stapleton, H.J. and Hager, L.P. (1983) J.

Biophys. 41, 105-112. 5 Nozawa, T., Kobayashi, N., Hatano, M., Ueda, M. and Sog, ani~ M.

(1980) Biochem. Biophys. Acta 626, 282-290 and references therein.

6 Callahan, P.M. and Babcock, G.T. (1981) Biochemistry 20, 952- 958 and references therein.

7 Ropp, J.S.D., LaMar, G.N., Smith, ICM. and Langry, ICC. (1984) J. Amer. Chem. Soc. 106, 4 4 3 8 - A ~ and references therein.

8 Sakurada, J., Takahashi, S. and Hosoya, T. (1986) J. Biol. Chem. 261, 9657-9662.

9 George, P. (1952) Advan. Catal. 4, 367-372. 10 Hosoya, T., Sakurada, J., Kurokawa, C., Toyoda, R. and

Nakamura, S. (1989) Biochemistry 28, 2639-2644. 11 Modi, S., Behere, D.V. and Mitra, S. (1989) Biochemistry 28,

4689-4694. 12 Morishima, I. and Ogawa, S. (1979) J. Biol. Chem. 254, 2814-2820. 13 Burns, P.S., Williams, R.J.P. and Wright, P.E. (1975) J. Chen~ Soc.

(chem. Comm.) 795-796. 14 Cdtchlow, J.E. and Dunford, H.B. (1972) J. Biol. Chem. 247,

3703-3713. 15 Critchlow, J.E. and Dunford, H.B. (1972) J. Biol. Chem. 247,

3714-3725. 16 Modi, S., Behere, D.V. and Mitra, S. (1989) Biochim. Biophys.

Afta 996, 214-225. 17 Schejter, A., Lanir, A. and Epstein, N. (1976) Arch. Biochem.

Biophys. 174, 36-44. 18 Yonetani, T. and Asakura, T. (1969) J. Biol. Chem. 244, 4580-4588. 19 Boucher, L.C. (1972) Coord. Chem. Rev. 7, 289-329. 20 Fabry, T.L., Surio, C. and Javaherian, K. (1968) Biochim. Biophys.

Acta 160, 118-125. 21 Modi. S., Saxene, A., Behere, D.V. and Mitra, S. (1990) Biochim.

Biophys. Acta 1038, 164-171. 22 Lan8, G. (1970) Quart. Rev. Biophys. 3, 1-36. 23 Modi, S., Behe~e, D.V. and Mitra, S. (1989) J. Biol. Chem. 264,

19677-19684. 24 Ellis, W.D. and Dunford, H.B. (1968) Biochemistry 7, 2054-2062. 25 Behere, D.V., Gonzalez-Ve~'gara, E. and Goff, H.M. (1985) Bio-

china. Biophys. Acta 832, 319-325. 26 Shannon, L.M., Kay, E. and Ix.w, J.Y. (1966) J. Biol. Chem. 241,

2166-2172. 27 Schonbaum, G.R. (1973) J. Biol. Chem. 248, 502-511. 28 Willard, J.M., Davis, J3. and Wood, G.H. (1969) Biochemistry 8,

3137-3144. 29 Dwek, R.A. (1973) NMR in Biochemistry, Oxford University

Press, Oxford. 30 Jarori, G.K., Ray, B.D. and Rao, B.D.N. (1985) Biochemistry 24,

3487-3494. 31 Solomon, I. (1955) Phys. Rev. 99, 559-565. 32 Bloembergan, N. (1957) J. Chem. Phys. 27, 572-573.

33 Modi, S., Behere, D.V. and Mitra, S. (1990) J. Inorg. Biochem. 38, 17-25.

34 Aviram, I. (1981) Arch. Biochem. Biophys. 212, 483-490. 35 Dwek, R.A., Williams, R.J.P. and Xavier, A.V. (1974) in Metal

ions in Biological systems (Sigel, H., ed.), Vol. 4, pp. 62-204, Marcel Dekker, New York.

36 Sakurada, J., Takahashi, S., Shimizu, T., Hatano, M, Nakamura, S. and Hosoya, T. (1987) Biochemistry 26, 6478-6483.

37 Ugarova, N.N., Savitski, A.P. and Berezin, I.V. (1981) Biochim. Biophys. Acta 662, 210-219.

38 Paul, K.G. and Ohlsson, P.I. (1978) Acta Chim. Scand. B32, 395-404.

93

39 Hildebrandt, A.G., Returner, H. and Estabrook, R.W. (1968) Bio- chem. Biophys. Res. Commun. 30, 607-612.

40 Jefcoate, C.R.E. and Gaylor, J.L. (1969) Biochemistry 8, 3464- 3472.

41 Dunford, H.B. (1982) Adv. Inorganic Biochem. 4, 41-68. 42 Oertling, W.A. and Babcock, G.T. (1988) Biochemistry 27, 3331-

3338. 43 Shiga, T. and Imaizumi, K. (1975) Arch. Biochem. Biophys. 167,

469-479. 44 Hashimoto, S., Tatsuno, Y. and Kitagawa, T. (1986) Proc. Natl.

Acad. Sci. USA 83, 2417-2421.