Catalase subunit structure

Fig. 8. Comparison of the subunit structures of a small subunit catalase (BLC) (A) and a large subunit catalase (HPII) (B). Segments including the N-terminal domain, the beta barrel core, the wrapping domain, the alpha helical domain, and the C-terminal domain (HPII) are indicated and are described in Section V,A.

The heme of catalases is deeply buried within the core of the catalase subunit. Protoheme IX or heme b is found in all small-subunit catalases so far characterized. The two large-subunit enzymes HPII and PVC have been characterized biochemically, spectrally, and structurally 91) as containing heme d in which ring HI is oxidized to a cis-hydroxyspirolactone. Heme b is initially bound to both enzymes during assembly, and it is subsequently oxidized by the catalase itself during the early rounds of catalysis 92). 

The diversity among catalases, evident in the variety of subunit sizes, the number of quaternary structures, the different heme prosthetic groups, and the variety of sequence groups, enables them to be organized in four main groups the classic monofunctional enzymes (type A), the catalase-peroxidases (type B), the nonheme catalases (type C), and miscellaneous proteins with minor catalatic activities (type D). 

Within this common core structure, modifications have been identified that provide catalases with further unique properties. The large-subunit enzymes have extensions at both the amino and carboxyl ends, the latter having a flavodoxinlike structure, a unique His-Tyr bond, a protected cysteine, and a modified heme. NADPH binding and an oxidized methionine are found in small subunit enzymes. Identification and assignment of roles to channels providing access to and egress from the deeply buried heme have recently become the focus of study. Analysis of the structure of catalase HPII of E. coli has been facilitated by the construction of more than 75 mutants (Table I). 

The tertiary structure of small subunit enzymes can be subdivided into four distinct regions, and the C-terminal or flavodoxin domain of the large subunit enzymes becomes a fifth region. These are indicated in Fig. 8 for clarity. The first region is the amino terminal arm (Fig. 8), which extends 50 or more residues from the amino terminus almost to the essential histidine residue (to residue 53 in PMC, 60 in PVC, 73 in BLC, and 127 in HPII). There is very little structural similarity in the N-terminal region and, in the case of HPII, the structure of the terminal 27 residues is not even defined and they do not appear in the crystal structure. Within the N-terminal arm is a 20-residue helix, helix a2 in HPII, which is the first secondary structure element common to all catalases. The presence of helix al varies among catalases, and there is no sequence or location equivalence even when it is present. 

Bergdoll et al. 87) have proposed that some proteins, including catalase, exhibit arm exchange or an interaction of one subunit with an adjacent subimit to stabilize quaternary structure, and that this... 

Fig. 12. A hypothetical folding and assembly pathway for catalases. In A secondary and tertiary folding first occurs in the individual subunits to form the 3-barrel (p), wrapping domain (W), a-helical segment (a), and fiavodoxin domain (F, only in HPII). In proceeding to B, heme is bound to each of the subunits, and this may serve as a catalyst for the rapid association of the i -related subunits to form the structure in C. In proceeding to D, Q-related subunits associate, resulting in the N-terminal arms being overlapped as the C-terminal portions fold back on themselves to form the fully folded structure shown in E. Only two subunits are shown in the progression from C to E, but a simultaneous folding must be occurring in the associated dimer. The fully folded tetramer is shown in two orientations.

Fig. 13. Active site residues in a small-subunit catalase BLC (A) and a large-subunit catalase HPIl (B). The active site residues are labeled, and hydrogen bonds are shown between the serine (113 in BLC and 167 in HPll) and the essential histidine (74 in BLC and 128 in HPll). A single water is shown hydrogen bonded to the histidine. The equivalent water in BLC is located by analogy to the position of the water in HPll. The unusual covalent bond between the N of His392 and the C of Tyr415 in HPll is evident on the proximal side of the heme in B. The flipped orientations of the hemes are evident in a comparison of the two structures, as is the eis-hydroxyspirolactone structure of heme d in B.

Have we exhausted catalases as a source of information about protein structure and the catalatic mechanisms The answer is clearly no. With each structure reported comes new information, often including structural modifications seemingly unique to catalases and with roles that remain to be explained. Despite a deeply buried active site, catalases exhibit one of the fastest turnover rates determined. This presents the as yet unanswered question of how substrate can access the active site while products are simultaneously exhausted with a potential turnover rate of up to 10 per second. The complex folding pathway that produces the intricate interwoven arrangement of subunits also remains to be fully clarified. 

Crystal structures of manganese catalases (in the (111)2 oxidation state) from Lactobacillus plantarum,its azide-inhibited complex, " and from Thermus thermophilus have been determined. There are differences between the structures that may reflect distinct biological functions for the two enzymes, the L. plantarum enzyme functions only as a catalase, while the T. thermo-philus enzyme may function as a catalase/peroxidase. The active sites are conserved in the two enzymes and are shown schematically in Figure 32. Each subunit contains an Mu2 active site,... 

X-ray diffraction studies at 3.0 A resolution have indicated the presence of two neighbouring manganese atoms (3.6-0.3 A) in each subunit (Barynin et al., 1986). Unlike the iron catalases, the Mn-catalases contain no heme and the active centres have an unusual structure. The subunit skeleton is a bundle of four nearly parallel helices, two of which are ca. 40 A long. There are similarities to the structure of apoferritin, however, the polypeptide chain in Mn-catalase is twice that of apoferritin. 

Three Mn catalases have been purified and characterized, and all appear to have similar Mn structures (17). The Mn stoichiometry is ca. 2 Mn/subunit, suggesting a dinuclear Mn site. The optical spectrum of the as-isolated enzyme has a broad weak absorption band at ca. 450-550 nm in addition to the protein absorption at higher energies. This spectrum is similar to those observed for Mn(III) superoxide dismutase and for a variety of Mn(III) model complexes, thus implying that at least some of the Mn in Mn catalase is present as Mn(III). In particular, the absorption maximum at ca. 500 nm is similar in energy and intensity to the transitions seen for oxo-carboxylato-bridged Mn dimers, suggesting that a similar core structure may be seen for Mn catalase (18). 

These are difficult enzymes to work with and only recently have crystal structures become available for two catalase-peroxidases Haloarcula marismortui (HMCP) and Burkholderia pseudomallei (BpKatG). A typical subunit is approximately 80 kDa in molecular mass, with a single heme b prosthetic group. The primary structure of each subunit can be divided into two distinct domains, N terminal and C terminal. The N-terminal domain contains the heme and active site, while the C-terminal domain does not contain a heme binding motif and its function remains unclear. The clear sequence similarity between the two domains suggests gene duplication and fusion. Curiously, despite many years of study, the actual in vivo peroxidatic substrate of the catalase-peroxidases has not been identified. 

Most catalases consist of four subunits of 60,000 mol each and contain one ferri-protoporphyrin IX molecule per subunit. With few exceptions, catalases are found in all but anaerobic organisms. The crystal structure of beef liver catalase has been determined to 2.5 pm resolution ", and the primary sequence is also known. ... 

Both in structure and in functions, catalase is akin to peroxidase. Catalase is the most effective among all known catalysts for the decomposition of hydrogen peroxide (k = 10 s ) and it displays, moreover, a peroxidase activity. The catalase molecule with a molecular mass of 225,000 is comprised of four subunits, each of which contains a hematin group. Hydrogen peroxide forms three compounds with catalase. Decomposition of H2O2 or ROOH occurs as a result of interaction between compound I of the catalase and the second substrate molecule. Not only hydroperoxides but other hydrogen donors, for instance, ethanol, are capable of reacting with compound I. 

---