Iron protein active centers

The existence of this HiPIP-type structure was the starting point of an interesting development in the study of the iron-sulfur active center. X-ray crystallography showed that there is little difference between the 4Fe-4S cluster in a ferredoxin (E0 = —400 mV) and in HiPIP (E0 = +350 mV). This anomaly was elucidated by the so-called C state hypothesis of Carter et al. (7) (Figure 3), in which the existence of a super-reduced HiPIP and a super-oxidized ferredoxin was postulated. The super-reduced HiPIP was shown to exist by Cammack (8) utilizing 80% DMSO (dimethylsulfoxide) to distort the protein environment of HiPIP. In this case a super-reduced HiPIP with EPR signal similar to a... 

Fig. 3 Molecular model [Fe4S4(SCH3)4] of the iron-sulfur active center of ferredoxine proteins. The carbon, iron, sulfur, and hydrogen atoms are represented by spheres of different shades of gray, from dark to bright, respectively...

The protein from D. desulfuricans has been characterized by Mbss-bauer and EPR spectroscopy 224). The enzyme has a molecular mass of approximately 150 kDa (three different subunits 88, 29, and 16 kDa) and contains three different types of redox-active centers four c-type hemes, nonheme iron arranged as two [4Fe-4S] centers, and a molybdopterin site (Mo-bound to two MGD). Selenium was also chemically detected. The enzyme specific activity is 78 units per mg of protein. 

LOXs are proteins containing a single atom of nonheme iron in catalytic center, with the ferric enzyme in an active form. The free radical-mediated mechanism of LOX-catalyzed process may be presented as follows (see also Figure 26.1) ... 

Nickel-iron hydrogenases [NiFe] (Figure 8.2) are present in several bacteria. Their structure is known [22, 23] to be a heterodimeric protein formed by four subunits, three of which are small [Fe] and one contains the bimetallic active center consisting of a dimeric cluster formed by a six coordinated Fe linked to a pentacoordinated Ni (III) through two cysteine-S and a third ligand whose nature changes with the oxidation state of the metals in the reduced state it is a hydride, H, whereas in the oxidized state it may be either an oxo, 0, or a sulfide,... 

Mononuclear octahedral/trigonal bipyramidal iron centers are found in either the ferric or the ferrous oxidation state (Whittaker etal., 1984 Arciero et ai, 1983). Because the iron may participate directly in catalysis as either a Lewis acid or base, only one state is the active form for a given enzyme. Transient redox changes may occur during turnover, but the enzyme returns to its initial condition. In contrast the tetrahedral mononuclear iron proteins appear to function primarily as electron transfer agents and therefore change oxidation state with a single turnover. 

One type of the constituent metallocenters in the MoFe protein has the properties of a somewhat independent structural entity. This component, referred to as the FeMo cofactor (FeMo-co), was first identified by Shah and Brill (1977) as the stable metallocluster extracted from acid-denatured MoFe protein. The FeMo-co was able to fully activate a defective protein in the extracts of mutant strain UW45, a protein which subsequently was shown to contain the P clusters but not the EPR-active center. The isolated cofactor accounted for the total S = t system observed by EPR and Mdssbauer spectroscopies of the holo-MoFe protein (Rawlings et al., 1978). Elemental analysis indicated a composition of Mo Fee-8 Se-g for the cofactor, which, if there are two FeMo-co s per a2 2> accounts for all the molybdenum and approximately half the iron in active enzyme (Nelson etai, 1983). Although FeMo-co has been extensively studied [reviewed in Burgess (1990)] the structure remains enigmatic. To date, all attempts to crystallize the cofactor have failed. This is possibly due to the instability and resultant heterogeneity of the cofactor when removed from the protein. Also, there is a paucity of appropriate models for spectral comparison (see Coucouvanis, 1991, for a recent discussion). Final resolution of this elusive structure may require its determination as a component of the holoprotein. 

Several models have been proposed for the active center of iron and sulphur in Clostridial ferredoxin in which the cysteine residues in the peptide chain participate in the sulphur bridging. Fig 9 166). Unfortunately X-ray analysis of crystals of these proteins has not been completed. It is difficult to confirm that all the irons are clustered in a single linear array 167, 168). X-ray studies of another non-heme iron protein, the high potential iron protein, hipip, from chromatium, carried out by J. Kraut (personal communication), indicate that the four irons of this molecule may be clustered in a tetrahedral array. 

Cytochromes P450 are monooxygenases whose cosubstrates, often NADH or NADPH, deliver electrons to the active center heme via a separate flavoprotein and often via an iron-sulfur protein as well 476a b A typical reaction (Eq. 18-55) is the 11 (3-hydroxylation of a steroid, an essential step in the biosynthesis of steroid hormones (Fig. 22-11). The hydroxyl group is introduced without inversion of configuration. The same enzyme converts unsaturated derivatives to epoxides (Eq. 18-56), while other cytochromes P450... 

Meanwhile, Blumberg and Peisach (145) showed that the interaction between a low-spin ferrous atom and an adjacent free radical can give rise to a g= 1.94 EPR signal. Brintzinger, Palmer, and Sands (146) proposed the first two-iron model for the active center of a plant-type ferredoxin. Their model, which consisted of two spin-coupled, low-spin ferric atoms in the oxidized protein and one low-spin ferric and one low-spin ferrous atom in the reduced protein, explained much of the chemical data on the proteins. Later, they (Brintzinger, Palmer, and Sands, (147)) presented EPR data for a compound, bis-hexamethylbenzene, Fe(I), which demonstrated all the properties fo the g= 1.94 signal observed in the ferredoxins. 

The active center of the plant-type ferredoxins is nearly identical in every protein studied. The only differences in this active center are the presence and magnitude of the rhombic distortion of the symmetry for the ferrous iron in the reduced proteins. 

As already mentioned above, the class of non-heme iron proteins is inhomogeneous and therefore often divided into subclasses. A distinction can be made based on the ligands that coordinate to the iron center and concomitantly on the types of potential iron cofactors. In most catalytically active non-heme iron proteins, the metal ions are coordinated via nitrogen and oxygen ligands. These types of proteins will be discussed in this chapter. 

Alkaline phosphatases form a well-known class of proteins that perform quite interesting and complicated reactions. As previously reported, Zn enzymes, like carboxypeptidases, thermolysin, and carbonic anhydrases, consist of only one Zn atom per active center. Most of the alkaline phosphatases consist of two 96-kDa subunits, each containing two Zn and one Mg ion. The alkaline phosphatase from E. coli has been crystallized and described in full detail [4], and a mechanism has been proposed. Several enzymes in this category have been mentioned in recent years, some of them also containing different metal ions, such as iron and zinc, as in the purple acid phosphatase [5], It is likely that the detailed structure and mechanism of many more examples of enzymes that remove or add phosphate groups to proteins will become available in the next decade. 

The extent of CPO immobilized on the sol-gel was determined by the difference between the activity of the initial enzyme solution and that measured in cumulative washes. Based on the cumulative activity lost in six washes, a second preparation of the CPO-bound sol-gel contained 10, 24, and 55 mg of CPO/g of sol-gel for the 50-, 150-, and 200-A CPO sol-gels, respectively. In prior experiments, the total activity was measured and an estimated 80% of the bound CPO was active. The sol-gel immobilization is expected to limit the unfolding of the protein bound inside pores of the sol-gel. Thus, immobilization is expected to affect solvent stability and thermostability. Immobilization would probably not impact peroxide stability, since the mechanism of peroxide inactivation is associated with changes in the redox properties and oxidation state of the heme iron and the active center, which cannot be protected by immobilization. Experimental studies of immobilized CPO were therefore limited to temperature and solvent stability. 

Another class of heme proteins containing iron protoporphyrin as the active center includes enzymes such as cytochrome P-450 and horseradish peroxidase (HRP). The former is a monooxygenase enzyme (MW 50,000) that catalyzes hydroxylation reaction of substrates such as drugs, steroids and carcinogens ... 

These proteins contain non-heme iron and inorganic (acid-labile) sulfur in the active centers as 4Fe-4S or 2Fe-2S or, in the case of rubre-... 

Much of our recent knowledge of the active centers of iron-sulfur proteins has come from the synthesis work of Holm (4) on analog compounds with 4Fe-4S, 2Fe-2S, and IFe centers. The additional ability to extract the iron-sulfur clusters from the proteins themselves and reinsertion of these clusters into other proteins has led to some interesting experiments which, among others, have shown that the 4Fe-4S configuration is more stable than the 2Fe-2S configuration. 

In studying the evolution of iron-sulfur proteins, the requirements of the ferredoxins for specifically placed cysteines to bind the irons in the active center have been most useful. Figure 6 shows sequences of the ferredoxins from several obligate fermenting anaerobic bacteria, green... 

---