Tải bản đầy đủ - 0 (trang)
1 Dioxygen Activation at Heme-Iron Centers: Lessons from Cytochrome P450cam

1 Dioxygen Activation at Heme-Iron Centers: Lessons from Cytochrome P450cam

Tải bản đầy đủ - 0trang

5 Transition Metal Complexes and the Activation of Dioxygen



O



137



O

= RH



ROH =



H



OH

H



H



OH2

FeIII

S



R



RH



OH

1



FeIII



FeIII



S



S

H2O2



8



2

e-



RH



RH



O



Peroxide

Shunt



FeIV

S

Cpd 1



FeII



Auto

Oxidation



H+



S



7



3



O2



H2O



RH



HO

H+



O



O

FeIII



FeIII



S

Cpd 0

6



RH

O



O

H+



S



RH

O

FeII



e-



4



S

5



Figure 2 Mechanism of O2 activation by cytochrome P450cam (adapted from [12]).



conserved Asp251 and Thr252 residues, serves to protonate the distal oxygen

atom of the hydroperoxide ligand, pulling away the resulting water molecule

[12]. The strengthening of the Fe-O bond coupled with the simultaneous formation/loss of water has been termed the “push-pull” mechanism [18] and results in

the cleavage of the O-O bond and the formation of Cpd I (7).

Similar structural features are found in catalase and peroxidase enzymes, which

form Cpd I by cleavage of the O-O bond in H2O2 (Figure 3). In both of these

enzymes deprotonation of the proximal O-atom of the H2O2 ligand is performed by

a distal histidine residue upon coordination. This proton is transferred to the distal

O-atom, and the resulting water molecule is pulled away by either an asparagine or

an arginine residue in the case of catalase (Figure 3a) or the peroxidases (Figure 3b),

respectively [19]. In contrast, the active sites of heme-O2 carriers lack the push-pull



138



Yee and Tolman



machinery for O-O bond cleavage. For example, in the case of oxy-myoglobin

(oxyMb) (Figure 3c) a non-hydrogen bonded neutral histidine residue serves as a

much weaker axial ligand, and the distal pocket lacks the polar residues necessary

to pull the O-O bond apart [19].



a



b



H

N

N



N

H



H

Asn



O



OH



H

N



N

H



H



O



c



H

N



H

Arg+



O



O



O



O



O



N III N

Fe

N

N



N III N

Fe

N

N



N III N

Fe

N

N



O



O



OH



O



OH



N



OH



O



O



OH



N



O



OH



NH



N

H

O



O

Asp



Catalase



Peroxidase



oxyMb



Figure 3 Active site interactions in catalase (a), the peroxidases (b), and oxyMb (c).



Over the past four decades significant progress has been made in characterizing

the various intermediates shown in Figure 2 for a number of enzymatic systems, yet

many questions still remain. In addition to being of fundamental interest, a better

understanding of how nature is able to harness the oxidizing power of O2 and H2O2

will enable the development of greener, more efficient catalysts for industrial

purposes. With these goals in mind, chemists have turned to synthetic models.



2.2



Iron-Porphyrin Complexes as Heme Models



Generally, iron-porphyrin models have been developed with the dual purpose

of (1) providing a better understanding of the spectroscopic signatures of the

enzymatic systems and consequently the electronic nature of these intermediates,

and (2) reproducing the reactivity of these systems, both as a means of understanding the structure-function relationship for these systems and for the purpose of

catalyst development. The following sections focus first on the structural and

spectroscopic features of these systems, beginning with models that bind O2 and

proceeding in order of the O2 reduction process described above in Figure 2.

Discussion of the reactivity of these models with specific regard to O-O bond

cleavage and C-H bond activation will follow.



5 Transition Metal Complexes and the Activation of Dioxygen



2.2.1



139



Fe(III)-Superoxo Intermediates



The first step in the activation of O2 by iron-porphyrin complexes involves binding

to the Fe(II) ion. Initial attempts to prepare such 1:1 Fe-O2 adducts were met with

difficulty due to the propensity of these early models to undergo auto-oxidation via

the intermediacy of (μ-η1:η1-peroxo)diiron intermediates [10]. Success in model

studies was achieved by inhibiting dimerization, often through the use of sterically

hindered porphyrin rings, and using aprotic organic solvents as well as low temperatures in order to mitigate auto-oxidation pathways. Generally, oxygenation of a

high-spin (S ¼ 2) Fe(II) precursor yields a low-spin (S ¼ 0) six coordinate

oxy-heme complex [10, 20].

Building upon the early, now classic work in this area [10], recent studies have

provided new insights. For example, in work aimed at providing benchmark

information for studies of models of cytochrome c oxidase (Section 2.3.1), reversible O2 binding to the Fe(II) precursor of 9 was observed at À80  C (Figure 4)

[21]. Manometry confirmed the formation of a 1:1 O2/Fe adduct in tetrahydrofuran;

however, in non-coordinating solvents the reversible formation of a μ-peroxo dimer

was observed, as supported by a 1:2 O2/Fe stoichiometry [21]. More recently,

compounds 10–12 have also been synthesized in which the axial solvent molecule

has been replaced by a pendant imidazole or pyridyl ligand [22, 23]. Compounds 9,

10, and 12 have been assigned as low-spin Fe(III)-superoxide species, primarily on

the basis of their 1H NMR and UV-vis spectra, as well as resonance Raman

(rR) data in the case of 9 [22]. The assignment of 11 was inferred based on its

conversion to an Fe(III)-peroxo intermediate upon one electron reduction [23].

O

Ar



O



Ar



N III N

Fe

N

N

Ar



O



F

Ar



Ar



Ar =



O



Ar



F



Ar



O



F



N III N

Fe

N

N



Ar =



N



HN



F

O



9

10



O

Ar



O

O



O



Ar

11: Ar =



N III N

Fe

N

N

Ar



N



Ar

HN

F



O



HN



Ar



O



N III N

Fe

N

N

Ar



N



N



R

N



12: Ar =

F



R'

13

a: R = Me, R' = H

b: R = H, R' = Me

c: R = H, R' = Et



Figure 4 Representative examples of synthetic oxy-heme adducts.



O

NH

Ar =



140



Yee and Tolman



Crystallographic characterization of synthetic oxyheme compounds has

remained elusive, and to date, compounds 13a–c (termed the “picket-fence

porphyrin”) are the only crystallographically characterized examples. The structure

of 13b was first solved in 1978, but was plagued by severe disorder of the O2 ligand

as well as high thermal motion [24, 25]. Very recently, data for this structure,

along with those of 13a,c, was recollected at low temperature (100–300 K) [26].

Important features of these structures include a bent η1-coordination of the O2

ligand to the iron center and disorder of the O2 ligand over 4 positions, the

occupancies of which are temperature-dependent [24–26]. At low temperatures

(80 K), a 6.2 tilt in the Fe-O bond relative to the normal of the porphyrin plane

is resolved for 13c, which results in an O-O bond length of approximately 1.28 Å.

This distance is consistent with values of ν(O-O) and with bond lengths proposed by

DFT calculations [27–29]. Taken together, the crystallographic data are in satisfying agreement with that of oxyMb [30].



a



b



c



O

O

Ar



Ar



O



N II N

Fe

N

N

Ar



N



Ar



Fe(II) (S = 0), O2 (S = 0)



Ar



O



O



HN



N III N

Fe

N

N

Ar



N



O



O

O



HN



N



Ar



Fe(III) (S = 1/2), O2 (S = 1/2)



HN



N II N

Fe

N

N

Ar



N



Ar



O



N

N



Fe(II) (S = 1), O2 (S = 1)



Figure 5 Proposed Fe-O2 bonding descriptions for oxy-heme species, shown for compounds

13a–c (adapted from [35]).



Despite the availability of crystal structures and evidence from rR experiments,

the nature of the Fe-O2 bonding in oxy-heme species is still not fully understood.

Three main descriptions have persisted since the 1960’s and these are summarized

in Figure 5 [31–33]. In models b and c, anti-ferromagnetic coupling is proposed to

account for the observed diamagnetism of these species. Recent theoretical studies

support model b for oxyMb [34]. The results of these studies indicate a

π-interaction for the Fe-O2 moiety resulting from single electron transfer from the

iron center to the O2 ligand. Even more recently, iron L-edge X-ray absorption

studies on 13b have also revealed a significant π-interaction which the authors

interpret as involving donation from the π* orbital of an OÀ

2 ligand into the empty dπ

orbital of the iron center [35]. However, comparison of the data to a set of Fe(II) and

Fe(III) standards show the oxidation state of the iron center to be most consistent

with an S ẳ 0 Fe(II) configuration. Moăssbauer spectra, which are characterized by



5 Transition Metal Complexes and the Activation of Dioxygen



141



large quadrupole splittings and small isomer shifts, also support an Fe(II) center

with a strong π-interaction to the O2 ligand [36]. Taken collectively, these results

point towards an oxidation state of the iron center that is intermediate with respect

to low-spin Fe(II) and Fe(III), which results from significant charge-transfer

facilitated by a strong π-interaction for the Fe-O2 moiety.



2.2.2



Fe(III)-Peroxo Intermediates



The second step in O2 activation by iron-porphyrins is the one-electron reduction of

the initial Fe-O2 adduct formed upon O2 binding, which yields species best formulated as Fe(III)-O22 À [12, 37, 38]. Alternatively, treatment of the Fe(II) precursor



with OÀ

2 can also generate Fe(III)-O2 adducts [39]. Some representative examples

are shown in Figure 6.

F



O

Ar



a: Ar =



O Ar



F



c: Ar =



F



R



R



N III N

Fe

N

N



F



Ar



O



R



b: Ar =



14



15

R = CH2CH3



d: Ar =



R

R



R

Ar



R



N III N

Fe

N

N



F



F



O



R



F



OO



O

N O



K



O

O



O



O Ar



N III_ N

Fe

N

N

Ar



Ar



Ar



t Bu



Mes



N III N

Fe

N

N

Mes



16



O



N



CO2Me

Ar =



HN



O



O

N



Ar =

17



tBu



Figure 6 Representative examples of Fe(III)-O22 À adducts.



Early studies performed on compounds 14a and 15 led researchers to formulate

these species as side-on bound (η2-peroxo)Fe(III) species on the basis of spectroscopic evidence from EPR, Moăssbauer, and magnetic measurements, which confirmed the high-spin Fe(III) state of the iron center [39, 40]. Further characterization

of the O-O and Fe-O bonding by IR, rR, and EXAFS spectroscopy supported a

symmetrically bound peroxo ligand [39, 41]. These data have served as benchmarks



142



Yee and Tolman



for the characterization of other model compounds (14b–d, 16) (Figure 6), which

all show similar spectroscopic features [42–44]. In a more recent report, compound

16 was found to be in equilibrium with an Fe(II)-OÀ

2 form. Preparation of this

species was achieved by the treatment of the Fe(II) precursor with two equivalents

of OÀ

2 [42]. Kinetic and equilibrium studies showed the reaction between the

Fe(II) precursor and OÀ

2 to be fast, with the equilibrium strongly in favor of the

bound species. New experimental and theoretical evidence suggests that 16 actually

exists in equilibrium with its (η1-superoxo)Fe(II)(DMSO) isomer rather than a

resonance form of 16 [45].

The high-spin state and side-on binding of these previous examples are in

contrast to what has been observed in heme enzymes [38]. Recently, the first

example of a low-spin (η1-peroxo)Fe(III) intermediate (17) (Figure 6) was prepared

which utilized an imidazole-tailed hang-man porphyrin motif to sterically force the

end-on binding of the O22 À ligand [38]. The synthesis of this species was achieved

by oxygenation of the Fe(II) precursor followed by one electron reduction with

cobaltocene at À70  C. Spectroscopic comparison of the end-on and side-on

imidazole-tailed Fe(III)-O22 À species revealed some significant differences for the

end-on adduct. These differences include a blue shifted Soret band, as well as a

low-spin Fe(III) EPR spectrum and higher-energy Fe-O stretching frequencies for

the end-on complex. Interestingly, both species were suitable precursors for the

formation of the corresponding Fe(III)-OOH species, suggesting the possibility that

protonation of the η2 adduct involves the initial transition to the η1 conformation

[23, 38].



2.2.3



Fe(III)-Hydroperoxo (Alkylperoxo) Intermediates



In the P450 and peroxidase enzymes, as well as in catalase, an Fe(III)hydroperoxo intermediate (Cpd 0, 6 in Figure 2) is the precursor to Cpd I (7 in

Figure 2), forming the reactive Fe(IV)-oxo π* cation-radical upon O-O bond

heterolysis. Additionally, Fe(III)-OOR intermediates have been proposed as

being competent oxidants themselves [46–48]. However, literature reports of

well-characterized Fe(III)-hydroperoxo (alkylperoxo) models are uncommon,

owing to their general instability in solution (even at low temperatures) and the

rapid formation of Fe(IV)-oxo/Fe(IV)-oxo π* cation-radical species. Synthetic

methods targeting Fe(III)-hydroperoxo (alkylperoxo) complexes generally

involve the addition of HOOR (R ¼ alkyl, acyl, or H) to an Fe(III) precursor at

low temperatures (< À60  C) in organic solvents (Figure 7). These reactions are

often performed under basic conditions to generate ROOÀ to yield either a fiveor six-coordinate product. It has also been demonstrated that reduction of an

Fe(II)-superoxide followed by protonation can yield a hydroperoxide intermediate, although these examples are rare [23, 54].



5 Transition Metal Complexes and the Activation of Dioxygen



O

N

N



143



RO

O



X



O

N



FeII



e-, H+



N

N



N



N



FeIII



N



ROOH

-H+, -X-



N

N



N



FeIII



N



R = alkyl, acyl, or H



H+

O

N

N



O



O

N



FeIII



N



e-



N

N



O

N



FeIII



N



Figure 7 Synthetic routes to porphyrin Fe-OOR adducts.



Early model compound studies observed putative acyl and alkylperoxo

intermediates in reactions leading to Cpd I type species [49]. However, conclusive

characterization of these intermediates was lacking. These studies were able to

show that the observed species were five-coordinate and high-spin on the basis of

their electronic absorption, EPR, and 1H NMR spectra [50–52]. Compounds

18a–19 (Figure 8) were the first six-coordinate examples [53–55] and differed

from their five-coordinate counterparts insofar as they displayed EPR spectra

characterized by a low g dispersion, consistent with a low-spin Fe(III) center and

similar to spectra reported for hydroperoxo (alkylperoxo) adducts of hemoglobin,

horseradish peroxidase, and bleomycin [56–58].

Axial ligand exchange was also demonstrated in the case of complex 20

(Figure 8), which was prepared by the addition of imidazole to a solution of the

hydroxide precursor [59].

Notwithstanding the above data, until fairly recently conclusive characterization

of the O-O and Fe-O bonding in Fe(III)-hydroperoxo (alkylperoxo) species was

lacking. The characterization of complexes 21a and 21b marked the first reports of

reliable rR parameters for an Fe(III)-hydroperoxide intermediate [23, 38]. Bands

assigned to ν(Fe-O) and ν(O-O) were observed for both complexes near 570 and

810 cmÀ1, respectively, consistent with an η1 binding mode for the HOOÀ ligand.

Additionally, the EPR spectra of these complexes exhibited the low g-dispersion

characteristic of previously reported six-coordinate Fe(III)-hydroperoxide

intermediates.

Interestingly, formation of 21a (Figure 8) required the presence of the pendant

imidazole, underscoring the importance of axial ligation in these systems. A role of

the axial ligand has been proposed to be to enhance the nucleophilicity of the O22 À

moiety. This has been recently demonstrated for a non-heme Mn(III)-peroxo

complex, for which the authors postulate the axial ligand to facilitate the conversion between the η2 and η1 conformations [60]. It was also shown for the



144



Yee and Tolman

RO



R'



N III N

Fe

N

N



Ph



R'



Mes



O



OH



HO



Mes



N



Mes



R'



O



O



Mes



N III N

Fe

N

N

Mes



19

R = CH2CH3



N III N

Fe

N

N

Mes



R'

R'



R'



18

a: R = H, X = OHb: R = t Bu, X = OMec: R = nBu, X = OMeHO



RO



R'



N III N

Fe

N

N



Ph



X



Ph



HO

R' O



Ph



O



N

NH

20



t Bu



Mes



N III N

Fe

N

N



Ar



Mes



HN



N



Mes



CO2Me

Ar =



HN



O



O



O

N



N



21a



21b



tBu



HO

Ar



O



HO



Ar



N III N

Fe

N

N



Ar



Ar



O



Ar



N III N

Fe

N

N



Ar



Ar

22a



Ar



OH

22b



F

F

Ar =

F



F

F



Figure 8 Well characterized examples of Fe-OOR complexes.



electron-deficient heme complex [(tpp)FeIII(O22 À )] (tpp ¼ 5,10,15,20-tetrakis(pentafluorophenyl)porphyrin) that epoxidation of the olefin 2-methyl-1,4naphthoquinone only proceeds in the presence of 20 % dimethyl sulfoxide, which

the authors attributed to increased axial coordination at the Fe(III) center [61].

EPR studies have also demonstrated that changes in spin-state tend to accompany coordination of an additional axial ligand for these systems, with five- and

six-coordinate species often being in high- and low-spin states, respectively [55].

These differences can have significant mechanistic implications for O-O bond

cleavage. Product analysis for the decomposition of 22a and 22b (Figure 8) by

O-O bond cleavage revealed different mechanisms for each complex. Invoking a

thermodynamic argument, the authors reasoned that 22a, being in the high-spin

state, favors the homolytic pathway which proceeds via a high-spin transition state

to yield a high-spin Fe(IV)-oxo product, minimizing the energetic cost associated

with changes in spin. Similarly, 22b is low-spin and favors the heterolytic pathway

yielding an Fe(IV)-oxo π*-cation radical species as the final product [62].



5 Transition Metal Complexes and the Activation of Dioxygen



2.2.4



145



Fe(IV)-Oxo Intermediates



Synthetic (porphyrin)Fe(IV)-oxo complexes have long been sought in efforts to

understand and harness the oxidation chemistry of enzymes such as the cytochrome

P450’s. Over the past three to four decades, thorough spectroscopic characterization has provided a good understanding of the electronic nature and structure of

these species. Representative examples of these complexes are shown below in

Figure 9. A brief overview of this early work will be given here as a preface to the

discussion of the reactivity of these species which has largely been the focus of

more recent efforts.

O



Ar



Ar



N IV N

Fe

N

N



Ar



O

Ar



N IV N

Fe

N

N



Ar

Ar



Ar



Ar

23



24



Cl

b: Ar =



a: Ar =



c: Ar =



Cl



Cl

F



Cl

d: Ar =



O

Ar



F



Ar



Ar



B



Ar



N IV N

Fe

N

N

Ar



25

a: B = N-methyl imidazole

b: B = pyridine

c: B = piperadine



F



Ar



O

Ar



F



e: Ar =



Cl

Cl



N IV N

Fe

N

N



F



O

Ar



Ar



N IV N

Fe

N

N



Ar



N



Ar



Ar

N



27



26



Ar =



Ar =

Ar =



Figure 9 Some examples of the most fully characterized Fe(IV)-oxo π* cation-radical and

Fe(IV)-oxo intermediates.



Compound 23a (Figure 9) is the earliest and most studied examples of a Cpd I

model, which was shown by Moăssbauer and EPR studies to be an S ¼ 3/2 system,

in which a triplet Fe(IV) center is ferromagnetically coupled (J ¼ 43 cmÀ1)



146



Yee and Tolman



to an a2u porphyrin radical [63, 64]. Electron-withdrawing aryl rings (23b–e) have

been shown to weaken ferromagnetic interaction and even result in a change in

symmetry of the porphyrin radical (23e) from a2u to a1u as confirmed by 1H NMR

and EPR studies [65, 66]. Electron-withdrawing substituents have also been

shown to prolong the lifetimes of these intermediates with respect to oxidative

degradation, allowing for the study of these compounds at temperatures well

above À50  C [62]. In contrast to their meso-substituted counterparts, β-substituted

Fe(IV)-oxo π* cation-radicals (24a–e) (Figure 9) have been characterized as

a1u porphyrin radicals weakly coupled with the Fe(IV) triplet state [65–67].

Interestingly, the electronic differences of the porphyrin systems in 23 and 24 do

not significantly affect the Fe-O bond strength as reflected by their similar ν(Fe-O)

values [68]. The Fe-O bonding has been characterized by EXAFS and rR, which

have shown the Fe-O bond distance to be approximately 1.6 Å with values of

ν(Fe-O) in the range of 800–850 cmÀ1 [69, 70]. These latter values are also

sensitive to axial coordination as well as electron-withdrawing aryl substituents,

which tend to result in negative shifts of ν(Fe-O) [63, 71–74].

Fe(IV)-oxo intermediates containing neutral porphyrin rings have also been

prepared (25a–c, 26, and 27) (Figure 9), and were first observed in the decomposition of (η1:η2-peroxo)diiron intermediates [75, 76]. More recently, electrochemical preparation of 27 has been achieved by the one-electron oxidation of

the Fe(III)-OH precursor [77]. Many structural similarities exist between these

species and their Cpd I cousins, particularly with regard to their Fe-O stretching

frequencies. Thus, the oxidation state of the porphyrin ring does not significantly

affect the Fe-O bond strength [6].



2.2.5



Reactivity of Iron-Porphyrin Intermediates



Factors Affecting O-O Bond Cleavage: Heterolysis versus Homolysis

In Nature and in a number of synthetic systems, Fe(IV)-oxo intermediates are the

O-O bond cleavage products of Fe(III)-OOR species. As such, significant efforts

have been made to find out what factors determine the mode of O-O bond scission.

Product analysis studies carried out in the late 1990’s and early 2000’s demonstrated

that electron-deficient porphyrin systems and protic/acidic reaction conditions

favored O-O bond heterolysis, while the opposite was true for O-O bond homolysis

[78–80]. The nature of the substituent of the hydroperoxide oxidant also was shown

to influence the mechanism of cleavage, with electron-withdrawing and electrondonating groups favoring heterolytic and homolytic cleavage, respectively [81].

A number of subsequent studies on the reactivity of Fe(IV)-oxo species have

indirectly confirmed these results [82–84].

Two more recent studies [85, 86] have sought to further characterize the nature

of the observed pH dependence of these systems. While it had been postulated that a

mechanistic changeover was the result of acid-base effects or changes in speciation

of the reacting Fe(III) complexes, it was found that the observed reactivity



5 Transition Metal Complexes and the Activation of Dioxygen



147



correlates well with the pH dependence of E ’(FeIV/III) and E(P+/P)

(P ẳ porphyrin). These results were interpreted to be consistent with the notion

that pH-dependent redox equilibria may in fact mask the true identity of the

oxidation products, especially if product analysis is the method of identification

[85]. Interestingly, further investigation using rapid-scan UV-vis experiments

performed at low temperature and under conditions of excess oxidant only showed

formation of the Fe(IV)-oxo π* cation-radical in the pH ranges studied (pH ¼ 6.3–

11.4), indicating that, for the complexes studied, O-O bond cleavage only proceeds

via heterolysis when meta-chlorobenzoic acid (m-CPBA) is the oxidant [86].

O



O

Ar

OH

Cl



Ar



N IV N

Fe

N

N



Ar



Ar



Cl



RO-



O

O

Ar



O



RO



Ar



N III N

Fe

N

N



Ar



Ar



Ar

Ar



Ar



Ar

O

Ar =



O



N III N

Fe

N

N



O



S OO



Ar



-RO



Ar



N IV N

Fe

N

N



Ar



Ar



Figure 10 Proposed mechanism of O-O bond cleavage and formation of Fe(IV)-oxo intermediate

under basic conditions.



Under basic conditions (pH >8, Figure 10) further reaction of the Fe(IV)-oxo π*

cation-radical was observed to generate the one electron reduced Fe(IV)-oxo

species as well as regenerate some of the starting Fe(III) precursor. These findings

are consistent with the fact that under basic conditions (pH >7) the Fe(IV)-oxo

form is electrochemically more stable than the Fe(IV)-oxo π* cation-radical

[85]. The authors propose that under basic conditions, reaction of the Fe(IV)-oxo

π* cation-radical with RỒ (R¼H, CH3) likely generates a reduced Fe(III)-OOR

species which can then undergo homolysis to form the observed Fe(IV)-oxo

intermediate or lose ROOÀ to reform the starting Fe(III) precursor [86]. These

findings highlight the importance of the nature of the oxidant as well as the reaction

conditions in dictating the mechanism of O-O bond cleavage.



Alkane Hydroxylation by Fe(IV)-Oxo Intermediates

Alkane hydroxylation by Fe(IV)-oxo π* cation-radical species has been well

studied, although few models have come close to reproducing the reactivity towards

C-H bonds observed in the cytochrome P450 enzymes. Recently, compound 29



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

1 Dioxygen Activation at Heme-Iron Centers: Lessons from Cytochrome P450cam

Tải bản đầy đủ ngay(0 tr)

×