Catalase
Created by Alexander Guendel
Aerobic organisms have adapted powerful anti-oxidant mechanisms to deal with the dangerous and inevitable radical by-products of respiration. The dumbbell shaped homotetramer enzyme catalase (eg 4BLC, derived from Bos tarus) is among the cell’s primary defenses. Localized in peroxisomes, this well-studied protein facilitates the dismutation of hydrogen peroxide with a blistering turnover rate of 800,000 molecules per second (Garrett & Grisham, 2008) – considered near the upper diffusion limit (Nicholls, Fita, & Loewen, 2000). Its pivotal role in protecting cells from reactive oxygen species (ROSs), high conservation amongst species, and ubiquitous expression among aerobic organisms has brought attention to it as early as 1900, and it was in fact one of the very first proteins ever crystalized (Nicholls et al., 2000). The four tetramers contain 506 residues each, which the bioinformatics analysis service ExPASy evaluates as having a total molecular weight of 230.3 kD and isoelectric point of 6.62 (Artimo et al., 2012).
Each monomeric subunit of beef liver catalase contains a catalytic center, as well as a protoporphyrin IX (heme) prosthetic group (Fig 1) and NADP ligand (Fig 2) (Reid et al., 1981). The quaternary structure is notable for subunits that bind tightly to one another through “arm exchange” of up to 25 overlapping residues per subunit. This interweaving is considered important due to the high reactivity of the hydrogen peroxide substrate (Goodsell, 2004; Nicholls et al., 2000). Each catalytic center has a channel leading to it, and is located at roughly the heme prosthetic groups approximately 2 nm below the protein’s surface and the same distance from the tetrameric center (Fita & Rossmann, 1985). Catalase’s 3D structure was solved to 0.25 nm as early as 1981, and has since been refined further (Fita, Silva, Murthy, & Rossmann, 1986). Four important domains come together in the monomeric units, each containing specific secondary structures necessary for the enzymatic function.
The amino terminal domain (A) contains the first 75 residues, and is important for the interlocking arm exchange that binds the monomeric units together. The residue Pro-69 is important in this interaction, to enhance stability and rigidity (Nicholls et al., 2000). The second domain (B) is the largest, containing residues 76-320 and the important heme ligand. Two similar four stranded anti-parallel sheets come together to make a large 8 stranded anti-parallel β barrel with several α helices intercalated within turns (Fita & Rossmann, 1985). The next domain (C) is called the “wrapping domain” which extends from residues 366-420. It is notable for containing the “essential helix” which has several important residues (eg Tyr-357 and Arg-353) expanded upon below (Fita & Rossmann, 1985). The last domain (D) contains the carboxy terminus, and is notable for 4 alpha helices (α10-13) lining the enzyme’s surface. Three of these helices are primarily composed of polar residues and directly contribute to the outermost portion of the funnel-like channel leading to the active site. The funnel becomes more hydrophobic as it descends into the protein (with residues such as Ala-116, Phe-163, Met-180, and Phe-184) and narrows to approximately 0.35 nm, which is thought to have a role in orienting the incoming hydrogen peroxide substrate for catalytic degradation (Fita & Rossmann, 1985).
The essential Tyr-357 residue located on the wrapping domain acts as the fifth heme Fe(III) ligand, and is important for keeping it in position in the binding pocket. The residue’s close proximity to the metal (0.19 nm) means that it is likely deprotonated, which is further supported by hydrogen bonding interactions of the phenolate and Arg-353 (Fita & Rossmann, 1985). Tyr-357 is also stabilized by the non-polar Pro-335, which sterically restricts the residue and maintains its conformation. His-74 is another critical residue, and is considered essential for catalase’s enzymatic function. Its importance is best illustrated through discussion of the reaction mechanism, which is a 2 step process outlined by the following equation:
(1) Enz(Por–FeIII) + H2O2 → CpdI(Por+–FeIV=O) + H2O
(2) CpdI(Por+–FeIV=O) + H2O2 → Enz(Por–FeIII) + H2O + O2
Equations 1 & 2 – Proposed reaction mechanism of hydrogen peroxide degradation by catalase. Enz(Por–FeIII) represents resting state of enzyme with Fe oxidation state indicated. CpdI(Por+–FeIV=O) indicates formation of compound I, an oxygen bound intermediate after degradation of 1 molecule of hydrogen peroxide (Nicholls et al., 2000).
As the hydrogen peroxide substrate approaches the active center, it is repositioned between the essential His-74 and Asn-147 by hydrogen bonding. One of the peroxide oxygens, O(2), is thus positioned proximal to the heme Fe(III) where it feels electrostatic attraction. The hydrogen bond formed with His-74 connects the proton associated with O(1) to the Nε of the imidazole side chain and polarizes the hydrogen peroxide molecule. These two interactions lower the O(2) proton pKa and lengthen the O(2)-hydrogen bond. The O(2) proton then transfers to O(1), causing further coordination of O(2) to the heme Fe leading to heterolytic breakage of the O(1)-O(2) bond. O(2) becomes part of the heme complex by double bonding and producing compound I (CpdI), and O(1) is able to dissociate from the Nε of His-74 with both protons as a water molecule thus completing part 1 of the reaction (Fita & Rossmann, 1985). At this stage, catalase in an alternative conformation (2CAG) from the bacterium Proteus mirabilis has been structurally determined by use of an inhibiting molecule, S-deoxymethionine (Gouet et al., 1996). This structure captures Fe in the middle of the 2 step reaction, while oxidized from Fe III to Fe IV. CpdI is restored to the original resting state Enz by reaction with another molecule of hydrogen peroxide. This mechanism proceeds as before, but with the evolution of molecular oxygen following coordination of the second substrate’s O(2) (Fita & Rossmann, 1985).
Also worth noting is the presence of a second ligand – NADP. While present in bovine liver catalase, NADP is not conserved among all catalases. For example, catalase isolated from Penicillium vitale lacks the nucleotide, but instead contains a flavodoxin-like domain in the same location (Melik-Adamyan et al., 1986). In either case however, the ligand is thought to protect the enzyme from ROSs (Nicholls et al., 2000).
Catalase has been captured in an alternative conformation in the bacterium Proteus mirabilis (2CAG). Direct comparison can be made to the catalytic center of Bos tarus liver catalase due to the high conservation of active site residues. However, a few residues elsewhere differ without immense impact on the enzymatic function. PSI-BLAST analysis is an iterative protein sequence similarity search method, which applies an algorithm to compare a queried protein against anything from a single protein to an entire database (Bhagwat & Aravind, 2007). It can be utilized in this case to show that despite sharing only an ancient ancestor, catalase from these bacteria and bovine species have similarity score (e-value) of 2·10-175 (where values less than 0.5 indicate close sequence similarity) indicating high conservation and homology (Altschul et al., 1997, 2005).
While comparison of primary structure can provide useful information regarding similarity and sequence homology, seemingly minor differences in sequence may result in vastly different conformation or function (Garrett & Grisham, 2008). Comparison of the tertiary structure can therefore provide insight into a dimension otherwise missed by BLAST techniques. The Dali server uses a sum of pairs method to make such a comparison of a given protein against a database (Holm, Kääriäinen, Rosenström, & Schenkel, 2008). Using it on catalase from B. tarus and the intermediate form from P. mirabilis reveals a z-score of 56.6, well above the threshold of 2 considered necessary to indicate structural similarity (Holm & Rosenstrom, 2010).