Photochemical generation and reactions of heme cation radicals in heme proteins

Vol. 159, No. 2, 1989 March 15, 1989

BIOCHEMICAL

PHOTOCHEMICAL

GENERATION

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 472-476

AND REACTIONS HEME PROTEINS

OF HEME CATION RADICALS IN

Edmond Magner and George McLendon Department of Chemistry, University of Rochester, Rochester, NY 14627 Received

December

20,

1988

A novel approach is described for generating reactive oxidizing centers in heme proteins, with zinc hemoglobin (Zn Hb) and zinc cytochrome c (Zn cyt c) used as examples. The reaction of 3Zn* Hb with [Cdn(NH3)5 C112+,and of 3Zn* cyt c with methyl viologen arc described. In the case of Zn Hb the cation radical produced decays with a rate constant of k3 = 2400st. Using this value the rate of the reaction

can be calculated to be 4500~~.

0 1989

Academic

Press,

Inc

The factors which control the rates of electron transfer between redox proteins are incompletely understood (1). Since many of these reactions arc quite rapid, techniques such as pulse radioIysis and flash photolysis (2) have been applied to the study of protein electron transfer. Combining metal substitution techniques in heme proteins with laser flash methods has proved particularly valuable(3). For example, pioneering studies by Hoffman et al. demonstrated that rapid (k = 100s1) long distance (25A)electron transfer in ap hFe hemoglobin hybrids occurs on excitation of the zinc porphyrin chromophore (3(a)). This basic approach has been applied to a variety of metal substituted proteins, leading to some understanding of how electron transfer rates depend on such key parameters as distance and reaction free energy (2,3). However, the utility of flash photolysis studies of zinc substituted heme proteins has been limited by the short lifetime of the reactive intermediate, the zinc porphyrin cation radical (Zn par+.), formed. For example, for the Zn/Fe hemoglobin hybrids already mentioned (3(a)), the reaction sequence can be described as: a

Zn*

Fe’39

kl

)

a Zn+.

k2

+

,a

P

a Z-I+. a Zn+.

P

Fe(m) Fe(m)

k3

P

P

Fe.0

Fe0

Fe00

k, << k, I k, + protein oxidation pmdllcts Since k2 >> kt, no appreciable concentration of the reactive intermediate builds up. In order to fully characterize the kinetics for such systems,it is necessary to devise techniques to P

0006-291X039 $1.50 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

*

azn

412

P

Vol. 159, No. 2, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

independently generate the zinc porphyrin radical cation. The reactivity of such intermediates are of particular interest not only because of their relevance to enzymatic intermediates, but also for the insight they may provide into the pathways for electron transfer in proteins. We now report a simple approach to generate these short lived reaction intermediates to permit detailed kinetic studies. This method depends on oxidative quenching of the reactive triplet excited state of the zinc porphyrin by weak oxidants such as [Cfl(NH3)5C1]2+, or methyl viologen. The approach is demonstrated to work well for the representative zinc substituted proteins, zinc hemoglobin and zinc cytochrome c. Using this method, the previously inaccessible value of k2 for ZniFe hemoglobin can be calculated.

Zinc substituted hemoglobins were prepared according to the procedures of Simolo (4). Zinc cytochrome c was prepared by literature methods (7). Potassium phosphate buffer was used throughout. Methyl viologen was purchased from G.F. Smith. All manipulation were carried out in the dark, unless the solutions had been thoroughly degassed, as zinc porphyrins are light sensitive. Degassing was accomplished by passing a stream of prepurified nitrogen over the surface of a slowly stirring solution of protein for one hour. The solution was then transferred to a glove box and diluted to give solutions which were 3mM in hemoglobin tetramer. The flash photolysis apparatus has been described in detail previously (5). In brief, the sample is excited by an NdNAG laser (532nm), with an ions pulse (ca 80 m.I). The monitor lamp, a PIJ “ultra stable” 100 W Xe short arc, is perpendicular to excitation, to minimize scattered light. Monitor illumination is controlled by a fast (uniblitz) shelter to minimize steady state irradiation. The signal is recorded on a fast (6 stage) 1P 28 photomultiplier tube, amplified, and input into a Techtronix 7912 digitizer. The digitizer signal is transferred to an IBM PC for storage and analysis. This computer, via an AID also controls the laser Q switch and the fast shutter. “A differential amplifier was used to offset the monitoring beam. [C0(NlI3)5Cl]Cla was prepared from cobalt dichloride by the method of Willard and Hall (6). The error in the lifetimes reported is +lO%. It was noticed that a considerable baseline shift accompanied the reaction of C@ with the triplet state of zinc hemoglobin, so the number of laser shots was kept to a minimum Signal averaging was employed only when the same result was obtained as with a single shot.

Excitation of the Zn(II) porphyrin chromophore in zinc hemoglobin or zinc cytochrome c produces long lived triplet excited states ( &a = 465 nm, 3k= 45s’ for ZnHb(S), 3k = 71s’ for Zn cyt c (9)). As anticipated from the photochemistry of simple Zn porphyrins (lo), the Zn heme protein triplet states can be quenched by reaction with weak oxidants (eg: [Cdn(NH3)C1]2+; methyl viologen, Mp+) to produce the zinc porphyrin radical cation (Zn par’) and the reduced electron acceptor (COn or MV+.). When [Cfl(NH3)&1]2+ is used as the oxidant, the [Co”(NH3)5Cl]+ 2+, thereby trapping the zinc porphyrin radical. The methyl product rapidly aquates to [COn(H20)63 viologen radical is stable, but the recombination reaction between the two radicals formed is relatively slow (ms timescale), allowing for the study of reactions which occur on the submillisecond time scale. 473

Vol. 159, No. 2, 1989

Figure triplet Figure as in

BIOCHEMICAL

1. Stern-Volmer state by Mv2+. 2. Second Figure 1.

order

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

plot for PH = 72

the dynamic quenching (1OmM phosphate).

plot

the

for

disappearance

of the %n*Hb T = 296OC. of

M@.

Conditions

Figure 1 showsa Stern-Voh-nerplot for the quenchingof zinc cytochrome c by methyl viologen. A quenchingrate constantof 9.0 x 106M-‘s-l is obtained,which is in good agreement with the value obtainedby Vanderkooi (9). The appearanceof a transientabsorbanceat 605nm (which is characteristicof the methyl viologen cation radical) indicatesthat electron transferis the mechanismof quenching. Addition of methyl viologen to give a final concentrationof 3.6mM ensuresrapid decay of 3Zn* cyt c (ca. 100~s). By monitoring at 605nm the rate of the back reaction can be easily observed. Figure 2 showsa secondorder plot for the recombinationreaction Zn+. cyt c + MV’.

-

Zn cyt c + MV*+

A rate constantof 1.7 x lO*M-‘s-l is obtained. This reaction provides a convenient meansof measuringthe back electron transferreaction in hemeproteins. The triplet excited stateof hemoglobin,sZnHb,decayswith increasingcobalt concentrationup to ca. 2 mM where the rate of decay, k&s, is 6 x lo6 M-W (figure 3). Beyond this concentration, static quenching occurs, and the rate reachesa limit of ca. 1.3 x 104s-1. At the Soret wavelength, the lifetime of the photoproduct, asmeasuredby the Soret recovery, reachesa limiting value of 2200 s-l, which is independentof cobalt concentrationover the rangestudied(0.2 to 7 mM). The difference between the triplet kinetics monitoredat 475 andthe Soret recovery at 424 nm suggeststhat the triplet is rapidly quenchedto form the cation radical. The radical cation then undergoesfurther reaction, previously observedby Hoffman (3(a)), reforming the zinc porphyrin ground state, with a rate constantks. This processmay involve oxidation of an adjacenttryptophan residue(11). At the isobesticpoint for 3ZnHb*LGHb, the cation showsa well-defined bleach (A0.D > 0.2 absorbance units), with an initial rate constantof 2400~~ (figure 4, R = -.98 for a singleexponential fit), which is consistentwith the resultsof the Soret experiment, Tryptophan is tentatively assignedas the “sacrificial donor” asphotolysis of Zn Hb in the presenceof [Co(NH& Cl]*+ producesan irreversible lossof ca. 75% of the tryptophan emission. In order to further characterðe zinc radical cation, its transientabsorptionspectrumwasobtained. Transient absorbances were measuredfor solutionswhich were ca. 5 mM in [Co(NH3)5Cl]Cl2, ensuringrapid formation of the radical cation. At t = 0 the spectrumresemblesthat of the previosly publishedspectrumof sZn*Hb (8): showingbleachingof the Soret, cxand p bandsand having two broad peaksat ca. 470 and 474

BIOCHEMICAL

Vol. 159, No. 2, 1989

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

A0.D.

Figure Figure %%$%b

3. 1.

Quenching Decay (B).

of

of Zn%b

3Zn*Hb (A)

by at

[Co(NH3)5Cl]2+. the

isosbestic

Conditions (436.5

nm)

as point

in for

680 nm. At 150 ps the triplet hasalmostcompletely decayed,and the spectrumshouldbc just that of the cation radical. The spectrumshowsa sharppeak at 475 nm, but only two bleaches,at the Soret and at = 550 nm. There is a broad peak at 660 nm. This spectrumresemblesthat of the zinc uroporphyrin cation studiedby CarapellucciandMauzerall(l2). uponreduction further complicatethe spectrum.

Changesin the spectrumof Corn

With the triplet quenchingkinetics characterizedfor zinc hemoglobin, it becomespossible to characterize the reaction chemistry of photo products suchas az~+$zFem(CN). This species waspostulatedasan intermediatein Hoffman’s studiesof Zn/Fe hemoglobinphotochemistry (3(a)), In earlier studies,the intermediatewas not directly observed,nor its reactionsmonitored, sincethe rate of decay is much faster than direct formation (kz + ks >> k2). By using [CO(~)(NH~)~C~]~+quenchingto generateZn po+ it is possibleto directly measurethe rate constant, k3. Combining this k3 value with the previously measuredvalue of ered= ks/(kz + ks) = 0.37 (2(a)) provides a measurementof the constant k2 = 45OOs-l for “hole transfer” from Zn pot+. to the Fe0 porphyrin in the neighboring subunit. An independentmeasurementof the rate constantk2 is obtainedfrom studiesof the oxidative quenchingby methyl viologen of the reducedhybrid hemoglobin. Methyl viologen is chosensince it cannot oxidise the Fen subunit, unlike COm. The cation radical species,CQ~+&~~~, which is producedon quenchingof the triplet excited statedecayswith a rate constant, kobs= k2 + ks = 65OOs-1.Using the value of 2400~~ forks, k2 is calculated to be 4100st which is in good agreementwith the value determinedabove.

Discusslpn The rate of back electrontransfer from a &+. to PFQQis significantly faster than the forward rate (k = 100s1)even though AG is -0.2V lessfavorable. Similarly, using the measured quantum yield for cytochrome b5 reduction by %nHb*, ( ==O.OS) (3(b)), the rate of forward 475

Vol. 159, No. 2, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

electron transfer is k = 2 x 103s-1.while the rate of back-electrontransfer is 5 x 104s-l. The difference betweenthe ratesin the cr,~pF%b systemvs. the ZnHbh systemcan be explained (3) by the difference in separationbetweenthe donor and acceptorsites(-8 8, for the Hb/bs and 20 A for the a2Z@)PFe@Qhybrid). However, the ordering of rate constantsk2 >> kr is lessreadily understood,sinceboth involve equivalent distancesand similarfree energies. With zinc cytochromec the rate of back electrontransferis not significantly different from the forward rate, even though there is a large difference in AG. However the rate of back electron transfer is probably maskedby diffusion. This experimentdemonstratesthat it is possibleto generatethe zinc radical cation. This makesit possibleto study andcomparethe forward and reverseratesof electrontransfer in protein/proteincomplexessuchascyotochromec/cytochrome c peroxidase. In zinc hemoglobin,a reasonableexplanationfor the relative rate enhancementof k2 over kr is that ke involves “hole” transfer while kt involves “electron” transfer. It is possiblethat such differencesreflect particularly effective superexchange via “hole” conduction states(analogousto “valence bands”)of the protein medium. Such an explanationhasbeenadvancedby Miller (13) to explain otherwiseunexpectedlyrapid ratesof long distanceelectron transferbetweenTMPD+. and various electron donors,and hasbeenelaboratedin somedetail by Hopfield andBeratan (14). This explanation shouldbe regardedasspeculative,pendingmore detailedtests. In conclusion,a meansof generatingzinc porphyrin radical cationsin hemeproteinshas beendemonstrated.Such a methodmakesit possibleto estimatethe rate of back electron transfer asin c&~Hb where a rate of 45OOs-1 is obtained. Acknowledements This researchwas supportedby NIH GM33881 and in part by the NSF. We would like to thank Ken Simolo for providing the hemoglobinhybrid.

Referaus 1 (a) Marcus, R.A.; Sutin, N. (1985) B&hem. Biophys. Acta 811,265-322. (b) Guarr, T.F.; McLendon, G.L. (1985) Coord. Chem. Rev. 68, l-52. 2 (a) Peterson-Kennedy, S.E.; McGourty, J.L.; Kalweit, J.A.; Hoffman, B.M. (1986) J. Am. Chem. Sot. 108, 1739-1746. (b) McLendon, G.L.; Miller, J.R. (1985) J. Am. Chem. Sot. 107,7811-7816. (c) Bechtold, R.; Kuehn, C.; Lepre, C.;Isied, S. (1986) Nature 322, 286-288. (d) Mayo, S.L.; Ellis, W.R.; Crutchley, R.J.; Gray, H.B. (1986) Science 233,948-952. 3 (a) McGourty, J.L.; Blough, N.V.; Hoffman, B.M. (1983) J. Am. Chem. Soc.l05,4470- 4472. (b) Simolo, K.P.; McLendon, G.L.; Mauk, M.R.; Mauk, A.G. (1984) J. Am. Chem. Sot. 106,5012-5013. 4 (a) Simolo, K.P.; Stucky, G.; Chen, S.; Bailey, M.; Scholes,C.P.; McLendon, G. (1985) J. Am. Chem. Soc.107.2865. (b) Simolo, K.P. Ph.D. Thesis, University of Rochester, 1986. 5 Strauch, S. Ph. D. Thesis,University of Rochester, 1986. 6 Willard, J.; Hall, (1922) J. Am. Chem. Sot. 44,222O. 7 (a)Vanderkooi, J.M.; Erecinska, M. (1975) Eur. J. B&hem. 60,199-207 (b) Magner, E.; McLendon, G.L. to be published. 8 Zemel, H.; Hoffman, B.M. (1981) J. Am. Chem. Sot. 103, 1192-1201. 9 Horie, T.; Maniara, G.; Vanderkooi (1984) FEBS Letters 177,287-290. 10 Dolphin, D.; Felton, R.H. (1974) Act. Chem. Res. 7,26-32. 11 Uyeda., M. ; Peisach,J. (1981) Biochemistry 20,2028-2035. 12 Mauzerall, D.; Carapelucci, P. (1975) Annal. New York Acad. Sci., 24, 197. 13 Miller, J.R.; Beitz, J. (1979) J. Chem. Phys. 7 1,4579-4595. 14 Beratan, D.; Hopfield, J. (1984) J. Am. Chem. Soc.106, 1584-1594. 476