(Li etal., 2005)

a Organism abbreviations are as follows: Ap=Aeropyrum pemi.x; Ax=Amphibacillus xylanus; Bt=Bos taunts', Cf=Crithidia fasciculata; Ec=Escherichia coli; Hi=Haemophilus influenzae; Hp=Helicobacter pylori; Hs=Homo sapiens; Mt=Mycobacterium tuberculosis', Pf=Plasmodium falciparum', Pt=Populus trichocatpa; Pv=Plasmodium viva.x; Py=Plasmodium yoelii', Rn=Rattus notvegicus; Sc=Saccharomyces cerevisiae; Sp=Streptococcus pneumoniae', St=Salmonella typhimutium; Tc=Trypanosoma b The redox state of CP is given as well as the residue numbers of CP and, for 2-Cys Prxs, CR.

0 The conformation of the active site is indicated as FF for fully-folded and LU with subscripts for the kinds of local unfolding seen in various subfamilies (see Figure 5).

d This disulfide form shows only the stable B-type dimer, but the protein is believed to be a BA decamer in the reduced state. e A concatameric interaction of the dodecamers is believed to be an artifact of crystallization. f A Cys —Ser mutant of CP mimics the reduced state.

g Originally described as a monomer when published by the authors but later acknowledged as A-type dimer (Evrard et al., 2004). h The glutaredoxin domains interact to make the protein a dimer of dimers.

1 MiAhpE does not clearly fit into any of the designated subfamilies and so has been set apart.

' The authors described the structures as an («2)4 octamer, but we suspect (see text) the octamer is an artifact of high protein concentration. Also, the FF form in entry 1XXU is as seen for other subfamilies, but that in 1XVW is slightly different.


As first recognized in 1994 (Chae et al., 1994) Prxs are a widely distributed family of peroxide reducing enzymes that evidence suggests have evolved from an ancestor protein having the thioredoxin fold (Copley et al., 2004). All of the known Prx sequences share recognizable similarities, including an absolutely conserved Cys residue (called the peroxidatic Cys) that is involved directly in the reduction of the substrate hydroperoxides. As outlined in Chapter 2, the known Prx sequences can be organized into five major subfamilies, each constituting a group of proteins that are more similar to each other than to the other Prxs. This grouping based on sequence similarity is most useful here because the level of structural similarity observed between two homologous proteins is generally related to their level of sequence similarity (Chothia and Lesk,1986). For the sake of consistency, we will here use the nomenclature introduced in Chapter 2, with the five subfamilies being referred to as Prxl, Prx6, Prx5, Tpx and BCP (shortened from BCP/PrxQ). In terms of the Prxs from humans, subfamily Prxl contains human PrxI, II, III and IV, subfamily Prx6 contains PrxVI, and subfamily Prx5 contains PrxV. Subfamily Tpx contains only bacterial Prxs and subfamily BCP contains bacterial and plant (PrxQ) Prxs. Subfamilies Prxl and Prx6 are listed next to each other as they are similar enough to each other that in some reports they are grouped into a single subfamily (e.g. Copley et al., 2004).

It is wise to be cautious about assigning a particular Prx to a subfamily just based on the common name of the enzyme, because many individual Prxs were named based on their activities before it was known which ones were most similar to each other. For example, within the Prxl subfamily individual enzymes have a variety of common names ranging as widely as PrxI, PrxII, PrxIII, PrxIV, Tsal, PrxA, PrxB, Tpxl, and AhpC. In terms of labels based on mechanism, all "typical 2-cys" Prxs are in the Prxl and Prx6 subfamilies, while "atypical 2-Cys Prxs", and "l-Cys Prxs" are not associated with any family in particular, but are distributed among a variety of families (see section 6.2 below).


A combination of structural and enzymatic studies has revealed that all Prxs have in common a catalytic cycle that includes a crucial conformational step as well as (at least) three chemical steps (Figure l). Throughout this Chapter, the Cys that directly reduces peroxide will be referred to as the peroxidatic Cys, using SP to designate the sulfur atom of the Cys side chain and using CP to designate the residue. Similarly, the resolving thiol, the thiol that forms a disulfide with CP, will be designated by SR for the sulfur atom and CR for the residue if it is a Cys.

As seen in Figure l, the catalytic cycle begins with the peroxide substrate (either an alkyl hydroperoxide or hydrogen peroxide) entering the fully-folded substrate binding pocket and reacting with the peroxidatic Cys (CP) at the base of this pocket. In chemical step l, the peroxide substrate is reduced to its corresponding alcohol and

Figure 1. The universal catalytic cycle of Prxs. The three main chemical steps of (1) peroxidation, (2) resolution, and (3) recycling are shown along with an explicit local unfolding step required for the resolution reaction. Sp and SR designate the sulfur atoms of the peroxidatic and resolving thiols, respectively. The fully-folded and locally-unfolded enzyme conformations are designated as FF and LU, respectively. See the text for further details

Figure 1. The universal catalytic cycle of Prxs. The three main chemical steps of (1) peroxidation, (2) resolution, and (3) recycling are shown along with an explicit local unfolding step required for the resolution reaction. Sp and SR designate the sulfur atoms of the peroxidatic and resolving thiols, respectively. The fully-folded and locally-unfolded enzyme conformations are designated as FF and LU, respectively. See the text for further details

CP becomes oxidized to the sulfenic acid form (SPOH). Resolution (step 2) occurs when a free thiol (SRH) attacks the SPOH to release water and form a disulfide. This attacking thiol, whether present on the same or another subunit ot the Prx, is referred to as the resolving thiol, as it resolves a potential block of the catalytic cycle resulting from the poor accessibility of CP by the bulky natural substrate. Because in the fully-folded enzyme CP is located in a protected active site pocket, resolution cannot occur without a conformational change that involves (at a minimum) the local unfolding of the active site pocket so as to make the CP side chain much more accessible. It is expected that the locally-unfolded and fully-folded conformations of the protein are in a dynamic equilibrium, governed by the equilibrium constant Klu that may differ for different Prxs and for the various redox states of each Prx. Because disulfide formation involves the adduction to CP of a large group, the disulfide forms of Prxs cannot adopt the fully-folded conformation, but remain locked into a locally-unfolded conformation. The reaction cycle is completed when the disulfide form is recycled to regenerate the peroxidatic and resolving thiols (step 3), and the Prx is freed to again adopt the fully-folded peroxidatic active site. In principle, recycling may involve protein or small molecule thiols. For many Prxs this step is known to involve a thioredoxin-like dithiol containing protein or domain (see Chapter 4).

While it is not part of the normal productive catalytic cycle, in competition with the resolution reaction is an overoxidation reaction (Figure 1). In this side reaction, the fully-folded SPOH form reacts with a second molecule of peroxide to form a sulfinic acid (SPO2H) and in certain Prxs this can further react with a third peroxide substrate to yield a terminally oxidized sulfonic acid (SPO3H) form.

As discussed by Sarma et al. (2005), the terminal state for a given Prx appears to be governed by details of the active site geometry. In any case, neither of these "overoxidized" forms can be readily converted to a disulfide and thus represent inactive forms of the enzyme, although the SPO2H form of certain eukaryotic Prxs is thought to be physiologically relevant in peroxide signal transduction (Wood et al., 2003; Immenschuh et al., 2005; Kang et al., 2005; Chapters 14 & 15) and can be resurrected to SPOH in an ATP dependent reaction (Biteau et al., 2003; Woo et al., 2003; Chang et al., 2004). The structural studies summarized in the next section reveal not only representative fully-folded and locally unfolded structures for various Prx subfamilies, but also interesting variations in quaternary structure that add complexity to the structure-function relations.


Since the first Prx crystal structure was reported in 1998 (Choi et al., 1998), the field has rapidly matured so that as of July 2006, as summarized in Table 1, 35 crystal structures of Prxs are available in the Protein Data Bank (Berman et al., 2000). Although three Prxs from the Prx5 subfamily have been analyzed by NMR to the point of making resonance assignments (Trivelli et al., 2003; Bouillac et al., 2004; Echalier et al., 2005), no complete NMR-derived structures are in the protein Data Bank. The 35 available structures represent the wild type and/or mutant forms of 25 distinct Prxs, including at least one representative from each Prx subfamily: eleven from subfamily Prxl, three from subfamily Prx6, four from subfamily Prx5, five from subfamily Tpx, and two from subfamily BCP. Eight of the structures, some of which are derived from structural genomics projects, have not yet been described in a publication in the original literature. In terms of the redox state of the peroxidatic Cys residue, all possibilities have been seen from SH, SOH, SO2H, SO3H and SS, although in only three cases, those of AhpC from Salmonella typhimurium (subfamily Prxl), human PrxV (subfamily Prxl), and a BCP from Aeropyrum pernix have both SH and SS states been observed for the same protein.


At the topology level, all Prxs have core tertiary structures that are highly spatially conserved (Figure 2a) with variations in loop lengths and conformations and N- and C-terminal extensions. When schematized, the core structure can be seen to include 7 P-strands and 5 a-helices, which are organized as a central 5-stranded antiparallel P-sheet, including strands P5-P4-P3-P6-P7, with one face of the sheet covered by pi-p2-a1 and a4 and the other face of the sheet covered by a2, a3 and a5 (Figure 2b). Because strand P5 has some interaction with strand P1, the central sheet is sometimes referred to as a single 7-stranded sheet rather than a 5-stranded

Figure 2. The Prx fold. (a) An overlay of all 19 fully-folded Prx structures indicating the conservation of the core of the fold. Colored by mobility with deep blue representing the least mobile portions of the chain and bright red representing the most mobile portions. (b) Stereoview of a representative fully-folded Prx (PDB code 1HD2) labeled to identify the common core a-helices (red), and (-strands (blue) that are conserved among all Prx proteins. The peroxidatic cysteine in the first turn of helix a2 is shown as a ball and stick with Sp in mustard yellow sheet plus an additional 2-stranded (-hairpin. In the fully-folded conformation of Prxs (as shown in Figure 2) the CP-residue is always located in the first turn of helix a2, and the unraveling of the first turn or two of this helix appears to be a universal feature of local unfolding.

In crystal structures, in addition to the coordinates, temperature factors (or B-factors) are derived for each atom. These values give information about the level of mobility of the structure, with larger values implying more disordered regions. In Figure 2a, the coloring of the Prx structure indicates the level of order, with a color gradient extending from the less mobile portions being blue to the most mobile portions being red. Figure 2a makes it very clear that surface loops are in general the most mobile parts of the structure, and these are also the regions that vary most in conformation and in the presence of insertions and deletions.

► irvTt "P TnnnrtfyyyygBTrBTf





---------------APUíVGtiAlPAVfcVífcG--------------EPGNKVílLAJiLrft—GRKGVLFGV^iAFU'íGcSK'í'ftLPGfVEClAfiALÍW-KGVQ Ó.i




--------------ILNVGAiAPUrrj ,fll5------------aNtí0^Wtl'lMGiHG~-AJSNV^yFFPWVí,X^ÍCÜ-Gf:iil3CLiíDHUJEFBN--aDSeV

fi3 (15 a4










Was this article helpful?

0 0

Post a comment