We present an overview of metallophosphoesterases, highlighting aspects of their biochemistry, structure and function. Despite the high degree of structural similarity these enzymes are versatile in terms of substrate utilization, and in many cases, their precise biological roles remain enigmatic.

INTRODUCTION

Phosphoesters occupy a central role in biology. Their presence in diverse biomolecules such as nucleic acids, proteins and lipids illustrates the success of this kind of chemistry during evolution. The abundant availability of inorganic phosphate, the trivalency of phosphate anions allowing the formation of mono-, di- and tri-esters, high solubility of phosphates in water and the range of pKa values for the three ionizable groups in phosphates allowing negative charge at physiological pH have been proposed as likely reasons for the prevalence of phosphoesters in biomolecules [1]. Equally important is the relative stability of several biologically relevant phosphoesters under physiological conditions, preventing spontaneous hydrolysis. In the case of DNA, for example, it has been suggested that the half-life for the spontaneous hydrolysis of phosphodiester bonds may be of the order of 30000000 years [2]. Similarly, monoesters such as phosphoserine are estimated to have half-lives of the order of 1010 years [3]. This high stability of phosphoester bonds can, however, pose a problem for biological systems, which strive for reversibility. It is therefore not surprising that phosphoesterases, enzymes that catalyse phosphoester hydrolysis, are also ubiquitous. Indeed, it has been argued that protein phosphatases, one such enzyme class, evolved independently several times during the evolution of eukaryotes, underlining the importance of phosphoesterases in biological systems [4].

In the 1990s, it was realized that common sequence motifs (GD and GNH) were identifiable in some (but not all) seemingly diverse phosphoesterases [5,6]. These included proteins with varied substrate specificities from phylogenetically distant organisms such as protein phosphatases from bacteriophage λ and mammals, bacterial cyclic nucleotide phosphodiesterases and nucleotidases and PAPs (purple acid phosphatases) from plants [5,6]. We now know that a substantial proportion of identified and predicted phosphoesterases belong to this enzyme superfamily, referred to as calcineurin-like metallophosphoesterases, named after the calcineurin phosphatase [also known as protein phosphatase 2B (PP2B)], one of the best characterized members of this family. These enzymes not only share catalytically important sequence motifs, but also possess a remarkably similar fold and active-site organization, including binding sites for two metal ions.

In the present review, we attempt to discuss the available biochemical, structural and biological information on this superfamily of enzymes (henceforth referred to as MPEs for metallophosphoesterases). Rather than cataloguing the different members of this family, we present them as variations on a conserved structural feature, i.e. the MPE core fold, which has undergone modification to serve diverse functions ranging from hydrolysis of nucleotides, polynucleotides, phospholipids and dephosphorylation of proteins to non-catalytic scaffolding functions. Despite this functional diversity, however, several common themes emerge which include catalytic promiscuity and the ability of MPEs to serve as scaffolds for recruiting other proteins, enabled by the extreme plasticity of the overall MPE fold and active-site architecture. In some cases, different biological roles of these proteins are also achieved by combining the MPE domain with other functional domains, either on the same polypeptide chain or as homo- or hetero-oligomeric complexes.

MPE SUPERFAMILY OF PROTEINS: PHYLOGENETIC RELATIONSHIPS AND DIVERSE FUNCTIONS

The MPE superfamily includes proteins from viruses, bacteria, archaea and fungi, as well as higher eukaryotes. For a comprehensive list of characterized and predicted MPEs, the reader may refer to the Pfam database (PF00149) [7]. Although the overall sequence similarity between these proteins can be as low as 20–30%, MPEs are identified easily on the basis of the presence of a motif comprising five blocks of conserved residues that form the active site of MPEs and allow metal co-ordination (Figure 1). However, there are noteworthy exceptions. For example, the eukaryotic Vps29 (vacuolar protein sorting 29) proteins lack one or more of these residues, but retain the overall MPE fold [8]; this and other examples are discussed below.

Conserved sequence motif in MPEs

Figure 1
Conserved sequence motif in MPEs

Upper panel: alignment of the five sequence blocks (marked I–V) characteristic of MPEs from some representative enzymes (M.tu, Mycobacterium tuberculosis; S. pom, Schizosaccharomyces pombe; H. sap, Homo sapiens; Lambda, bacteriophage λ). Conserved residues are highlighted. Lower panel: schematic representation of the MPE motif showing conserved residues and some variations seen in these residues. In both panels, residues in yellow co-ordinate metal α (MI), whereas those in blue co-ordinate metal β (MII). A conserved aspartate residue co-ordinates both metals and is coloured green. Other conserved residues are highlighted in orange. A conserved glycine residue in block I and tyrosine residue in block II found only in PAPs are highlighted in grey in the upper panel.

Figure 1
Conserved sequence motif in MPEs

Upper panel: alignment of the five sequence blocks (marked I–V) characteristic of MPEs from some representative enzymes (M.tu, Mycobacterium tuberculosis; S. pom, Schizosaccharomyces pombe; H. sap, Homo sapiens; Lambda, bacteriophage λ). Conserved residues are highlighted. Lower panel: schematic representation of the MPE motif showing conserved residues and some variations seen in these residues. In both panels, residues in yellow co-ordinate metal α (MI), whereas those in blue co-ordinate metal β (MII). A conserved aspartate residue co-ordinates both metals and is coloured green. Other conserved residues are highlighted in orange. A conserved glycine residue in block I and tyrosine residue in block II found only in PAPs are highlighted in grey in the upper panel.

The cladogram in Figure 2, built using the family's seed alignment (source: Pfam [7]), shows the relationship between various members of the MPE superfamily. On the basis of sequence similarity to characterized enzymes and branching patterns, proteins grouped together probably have similar functions. It can be seen that several branches of the cladogram do not possess any characterized members, precluding a functional assignment for them (Figure 2). Our inability to predict functionally relevant substrates for these enzymes on the basis of their sequences alone is because, first, sequence signatures related to substrate binding and recognition are not yet clearly identified for MPEs, and, secondly, several members of this enzyme superfamily can utilize multiple substrates. These may range from monoesters to diesters and even triesters of phosphate. This promiscuity in substrate utilization is best exemplified by the GpdQ MPE from Enterobacter aerogenes that can utilize all three types of substrates [9,10]. Therefore much of the biochemical characterization of these enzymes relies on the use of chromogenic synthetic substrates such as bis(p-nitrophenyl) phosphate (bis-pNPP) and p-nitrophenyl phosphate (pNPP) that represent phosphodiester and phosphomonoesters respectively. Some members of the MPE superfamily also hydrolyse bonds other than phosphoesters such as phosphonate (exemplified by the Rv0805 [11] and metallophosphoesterase domain-containing 2 (MPPED2) [12] proteins) and pyrophosphate (LpxH [13], diphosphonucleotide phosphatase/phosphodiesterase (PPD1) [14] and ApaH [1517]) without large deviation in sequence relative to the MPE consensus.

Phylogenetic relationships between the MPE domains of some MPEs

Figure 2
Phylogenetic relationships between the MPE domains of some MPEs

The cladogram was built with the seed alignment of the MPE superfamily from Pfam (PF00149; containing 324 sequences) using the neighbour-joining algorithm in MEGA 6.0. Branches were collapsed and assigned functions on the basis of available information on characterized enzymes. Possible biochemical activities are indicated next to each branch wherever known. Branches of the cladogram that did not possess even a single characterized enzyme are not annotated.

Figure 2
Phylogenetic relationships between the MPE domains of some MPEs

The cladogram was built with the seed alignment of the MPE superfamily from Pfam (PF00149; containing 324 sequences) using the neighbour-joining algorithm in MEGA 6.0. Branches were collapsed and assigned functions on the basis of available information on characterized enzymes. Possible biochemical activities are indicated next to each branch wherever known. Branches of the cladogram that did not possess even a single characterized enzyme are not annotated.

Since the cladogram in Figure 2 is built using only the core MPE domain, it does not take into account additional domains or structural elements found in several enzymes. Nevertheless, several proteins with similar function and from phylogenetically diverse organisms do branch together (Figure 2). A brief description of these is provided below.

Nucleases

Eukaryotic and archaeal orthologues of the Mre11 DNA-phosphodiesterases are closely related to SbcD enzymes, their bacterial counterparts, in the MPE cladogram. These proteins are known to function as part of the ‘MR’ complex, consisting of Mre11 and Rad50 in eukaryotes, and SbcD and SbcC in bacteria [18]. Whereas the roles of the MR complex in eukaryotes are to repair double-stranded breaks during meiosis and telomere maintenance, in bacteria the MR complex repairs ‘misfolded’ DNA hairpin structures during replication [18]. Given its critical roles in ensuring genome integrity, it was not surprising that deletion of the mre11 gene in mice resulted in embryonic lethality [19,20], whereas hypomorphic mutations in humans are associated with cancer-prone disease conditions such as ataxia-telangiectasia-like disease (ATLD) [2123].

It is important to note that another class of MPEs that utilize nucleic acids as substrates, i.e. orthologues of yeast Dbr1, do not branch with Mre11-like proteins, but form a separate clade (Figure 2). Dbr1 enzymes are involved in the processing of RNA, specifically 2′5′-phosphodiester ends found in lariat intron RNAs [24,25]. In the absence of Dbr1, these intermediates were found to accumulate in Schizosaccharomyces pombe cells, and significantly reduced growth rate [26]. Dbr1-like enzymes are restricted to eukaryotes. Interestingly, a substitution (of aspartate to cysteine) in sequence block I is found in all Dbr1-like MPEs from yeast to humans (Figure 1 and Table 1). Hence, unlike the Mre11/SbcD proteins that appear to have evolved early in evolution, Dbr1 may have evolved independently to recognize and hydrolyse nucleic acid substrates in a common eukaryotic ancestor, concomitant with the appearance of introns in genes.

Table 1
Active-site residues in MPEs

Summary of the active-site residues found in some well-characterized MPEs highlighting deviation from the MPE motif consensus sequence. Abbreviations for the enzymatic activities of the proteins shown in the Table: M, monoesterase (m, low activity); D, diesterase; T, triesterase; C, cyclic nucleotide phosphodiesterase. RCSB PDB codes: MPPED2, 3RL3; Vps29, 2R17; MJ0936, 1S3N; Rv0805, 3IB7; GpdQ, 3D03; Mre11, 1II7; YfcE, 1SU1; DR1281, 1T70; CthPnkP, 4J6O; λ-PP, 1G5B; YmdB, 4B2O; Dbr1, 4PEG; Lmo2624, 2XMO; pig PAP, 1UTE. Residues coloured red show deviations from the conserved MPE signature motif.

 
 

Phosphoprotein phosphatases

Phosphoprotein phosphatases, enzymes that dephosphorylate phosphoserine/threonine residues in proteins, cluster together in the cladogram shown in Figure 2. These include mammalian phosphoprotein phosphatases (PPP5 and calcineurin/PP2B), lambda phage phosphatase (λ-PP), and some putative bacterial phosphatases (PrpE from Bacillus subtilis). During our analysis of these sequences, we observed that all of these enzymes have a non-polar amino acid (valine/isoleucine/leucine/alanine/proline) in sequence block V instead of the canonical histidine found in most other MPEs (Figure 1 and Table 1). Mutation of the equivalent histidine in the cyclic nucleotide phosphodiesterase Rv0805 [11] and Dbr1 nuclease [25] severely compromises the phosphoesterase activity of these enzymes. However, protein phosphatases seem to function with this substitution, indicating that compensatory changes may have occurred in their active site. Furthermore, the fact that this substitution was found in bacterial as well as eukaryotic enzymes is suggestive of either convergent evolution or the presence of a common ancestor for these bacterial and eukaryotic phosphatases. The first of these two possibilities would imply a functional role for this substitution, perhaps to allow hydrophobic interactions with substrate proteins. This has not yet been tested, but would no doubt shed light on the architecture of MPE active sites. The second possibility has been alluded to by Uhrig et al. [27]. They used sequence data to show that some protein phosphatases in plants may have been derived from mitochondrial or archaeal genomes, or by horizontal gene transfer from bacteria [27]. Additionally, similarity between some eukaryotic protein phosphatases and bacterial diadenosine tetraphosphatases (or ApaHs) is also known [28] and this is clearly seen from the relatively close branching of ApaH enzymes and protein phosphatases (Figure 2). Importantly, we observed that several ApaH enzymes also harbour a similar active-site substitution. Moreover, recent studies on PrpE, a protein phosphatase from Myxococcus xanthus, demonstrated that this enzyme also possesses diadenosine tetraphosphatase activity in vitro [17], strengthening the case for a bacterial origin for protein phosphatases.

Cyclic nucleotide phosphodiesterases

Cyclic nucleotide phosphodiesterases from the MPE superfamily form two distinct branches containing CpdA-like and CpdB/2′3′-cyclic nucleotide phosphodiesterase-like proteins (Figure 2). CpdA was originally identified in Escherichia coli as a regulator of 3′5′-cAMP levels, an important second messenger molecule that mediates catabolite repression [29]. Subsequently, homologues of CpdA have been identified from several bacteria such as Pseudomonas aeruginosa [30], Haemophilus influenzae [31] and Mycobacterium tuberculosis (referred to as Rv0805) [32]. These enzymes show no sequence similarity with well-characterized mammalian cyclic nucleotide phosphodiesterases, and hence are referred to as Class III cyclic nucleotide phosphodiesterases in the literature [33]. For instance, whereas mammalian cyclic nucleotide phosphodiesterases show relatively low Km values for specific (i.e. cAMP or cGMP) substrates (10–1000 nM [34]) Class III cyclic nucleotide phosphodiesterases have a high Km (∼10 μM–5 mM [11,29,30,32,35]) and can hydrolyse both cAMP and cGMP [30,32]. It is important to note, however, that whereas all CpdA-like enzymes do hydrolyse 3′5′-cAMP in vitro, their roles in vivo may not be restricted to regulating levels of this nucleotide. For instance, our studies with the Rv0805 protein have revealed that the major role of this enzyme in modulating the cell wall properties of mycobacteria is independent of its activity as a cAMP phosphodiesterase [11,36,37].

CpdB-like enzymes are capable of hydrolysing 2′3′-cAMP, a positional isomer of 3′5′-cAMP and, unlike CpdA orthologues, may also possess 3′-nucleotidase activity [38,39]. It may therefore not be surprising that CpdB and its orthologues branch with 5′-nucleotidase from E. coli in the MPE tree (Figure 2), possibly representing convergent evolution of their MPE domains to accommodate similar substrates. It is noteworthy here that 5′-nucleotidase from E. coli has a large (188 amino acids) C-terminal regulatory domain [40,41] that is absent from CpdB-like enzymes. In Gram-negative bacteria such as Yersinia enterolytica, CpdB-like enzymes may be involved in utilization of extracellular 2′3′-cAMP for growth [42]. CpdB-like 2′3′-cAMP phosphodiesterases have been identified and characterized from other bacteria too, for instance SpdA from Sinorhizobium meliloti [43], but their biological roles are unclear. It must be noted that several other MPEs from different branches of the MPE cladogram can also hydrolyse 2′3′-cAMP effectively in vitro. For instance, the Rv0805 phosphodiesterase utilizes 2′3′-cAMP more efficiently that 3′5′-cAMP, even though it is more similar to CpdA than CpdB [11,35]. MPEs thought to be involved in RNA processing such as PnkP from Clostridium thermocellum [35], YmdB from B. subtilis [44] and as well as protein phosphatase from the λ-phage [35] also display robust activity against 2′3′-cAMP in vitro. This raises some important questions regarding the biological relevance of the 2′3′-cAMP hydrolytic activity of these enzymes, given that the presence of naturally occurring free 2′3′-cAMP has not been reported in most systems.

Pyrophosphatases and nucleotidases

Enzymes from three different branches of the MPE cladogram possess pyrophosphatase activity (Figure 2). The first two of these are ApaH-like [45,46] and 5′-nucleotidase-like enzymes [47,48] that act on pyrophosphate linkages in nucleotides and activated sugars respectively. The third branch contains LpxH-like enzymes that act on UDP-2,3-diacylglucosamine (UDP-DAGn) to produce Lipid X [13]. Lipid X in turn is required for the production of Lipid A, an important component of Gram-negative bacterial cell walls. Thus deletion of the lpxH gene from E. coli results in accumulation of UDP-DAGn in the bacterium [49]. Interestingly, MPE domains associated with the small subunits of DNA polymerases from archaeal and eukaryotic genomes are part of the same clade as LpxH (Figure 2). These were initially identified using computational methods by Aravind and Koonin [50] and predicted to have pyrophosphatase activity on the basis of the association between pyrophosphatase domains (not MPEs) with DNA polymerases from bacteria. Eukaryotic polymerase-associated MPE domains, however, lack the necessary motifs for catalytic activity (discussed in greater detail below) and hence may possess non-catalytic regulatory roles [50].

Purple acid phosphatases

Catalytically competent PAPs form two distinct clades in the MPE cladogram (Figure 2). Whereas enzymes from plant and invertebrate genomes are present in both of these clades, mammalian PAPs are restricted to a single branch. Mammalian-like PAPs are typically smaller in size and lack a large N-terminal regulatory domain found in plant enzymes [51,52]. Despite their separation into two clades and differences in domain organization, enzymes from both groups have a histidine-to-glycine substitution in sequence block I (Figure 1). The influence of this change on catalysis remains to be tested, since this substitution does lead to a significant alteration in the conformation of the active site (see Figure 5B and Table 1).

The physiological substrates of most PAPs are unknown. However, they are known to perform important functions such as regulating bone remodelling in mammals [53] and organophosphate degradation in plants [54]. Recently, the PAP PPD1 from yellow lupin seeds was reported to demonstrate pyrophosphatase activity along with phosphodiesterase activity, a feature not common in plant PAPs [14]. A third clade containing PAPs is also present, but no information is available on the biochemistry of the enzymes in this clade (Figure 2).

‘Small’ MPEs

Finally, all MPEs that do not possess any additional domains and contain only the core MPE fold cluster together (Figure 2). These include the MJ0936 protein from Methanococcus janaschii [55], YfcE from E. coli [56] and Vps29 from eukaryotes [8]. Different enzymes from this clade show different substitutions in active-site residues and hence possess different biochemical properties. For example, YfcE and its orthologues contain a histidine-to-cysteine substitution in sequence block III that renders them inactive against cyclic phosphodiester bonds [35,56] (Figure 1 and Table 1). At the same position, MJ0936 and its orthologues possess an asparagine residue. Vps29 proteins, on the other hand, have more than one active-site residue substituted for non-catalytic alternatives and have evolved to be scaffolds for other proteins rather than enzymes [8,57]. Despite this difference, the clustering of these proteins in the MPE cladogram (Figure 2) suggests that MPE domains lacking associated domains have evolved convergently, perhaps due to structural constraints that may be relevant in the absence of stabilizing interactions with accessory structural elements.

STRUCTURE OF THE MPE DOMAIN: VARIATIONS ON A CONSERVED THEME

The crystal structures of several members of the MPE superfamily are available. A dominant theme in all these structures is the conservation of the core MPE fold and architecture of the active site. Different enzymes have evolved several additional structural elements, often entire domains, which may modify and/or regulate their functions.

The core MPE domain and active site

The MPE domain has a compact globular shape, with the three-dimensional fold of the polypeptide chain composed of a characteristic alternation of secondary-structure elements, described as a βαβαβ architecture (Figures 3 and 4). For historical reasons, this fold is called the calcineurin-like fold. The heart of the structure harbours a β-sandwich, built of two parallel β-sheets. Within each sheet, the organization of β-strands is of a mixed parallel/antiparallel nature. The β-sandwich is decorated by α-helices. This organization of secondary-structural elements in best appreciated for the MJ0936 protein from M. janaschii which is the smallest member of the MPE superfamily characterized to date (Figures 3 and 4A).

Typical three-dimensional fold of the MPE domain, represented by one of the smallest members of the MPE family, MJ0936 (PDB code 1S3N)

Figure 3
Typical three-dimensional fold of the MPE domain, represented by one of the smallest members of the MPE family, MJ0936 (PDB code 1S3N)

The structure is shown as ribbons, with magenta α-helices and blue β-strands. The active-site residues are depicted as sticks, with carbon in yellow (except for the non-conserved residue Asn60, where carbon is green), oxygen in red and nitrogen in blue, and active-site metals (Mn2+) as magenta spheres, marked by their position as metals α and β. N- and C-termini are marked and secondary-structure elements are numbered.

Figure 3
Typical three-dimensional fold of the MPE domain, represented by one of the smallest members of the MPE family, MJ0936 (PDB code 1S3N)

The structure is shown as ribbons, with magenta α-helices and blue β-strands. The active-site residues are depicted as sticks, with carbon in yellow (except for the non-conserved residue Asn60, where carbon is green), oxygen in red and nitrogen in blue, and active-site metals (Mn2+) as magenta spheres, marked by their position as metals α and β. N- and C-termini are marked and secondary-structure elements are numbered.

Organization of secondary-structure elements

Figure 4
Organization of secondary-structure elements

Organization of secondary structure elements in (A) MJ0936 (PDB code 1S3N), (B) Rv0805 (PDB code 3IB7) and (C) Dbr1 (PDB code 4PEG) MPEs. α-Helices of MPE domain are shown as magenta cylinders and β-strands are shown as blue arrows. α-Helices of accessory C-terminal domains of Rv0805 (B) and Dbr1 (C) are in orange. Active-site residues from the MPE motif are shown in yellow boxes. Residues deviating from the consensus are shown in green boxes. N- and C-termini are marked.

Figure 4
Organization of secondary-structure elements

Organization of secondary structure elements in (A) MJ0936 (PDB code 1S3N), (B) Rv0805 (PDB code 3IB7) and (C) Dbr1 (PDB code 4PEG) MPEs. α-Helices of MPE domain are shown as magenta cylinders and β-strands are shown as blue arrows. α-Helices of accessory C-terminal domains of Rv0805 (B) and Dbr1 (C) are in orange. Active-site residues from the MPE motif are shown in yellow boxes. Residues deviating from the consensus are shown in green boxes. N- and C-termini are marked.

It is noteworthy that the numbers of secondary-structure elements can vary significantly between MPEs. For instance, six β-strands in the N-terminal sheet and five in the other (6+5) of the β-sandwich are observed in the case of MJ0936 [55], with four flanking α-helices (Figures 3 and 4A). The Vps29 protein maintains a (6+5) pattern of the β-sandwich, but has three flanking α-helices [8]. In the Rv0805 phosphodiesterase from M. tuberculosis, the β-sandwich is altered to a (5+6) pattern and is associated with six α-helices [11] (Figure 4B), whereas in Dbr1, the β-strand pattern is (6+6) with nine α-helical elements [58] (Figure 4C). This increased complexity in the core MPE domain may, in some cases, have important regulatory consequences. For instance, in the Dbr1 enzyme, a unique loop between β7 and α5, called the lariat recognition loop (LRL), is necessary for substrate recognition [58] (Figure 4C).

The active site of the MPE enzymes resides on the top of the central β-sandwich (Figure 3). Two metal ions (MI and MII or α and β) in the active-site cleft are usually octahedrally co-ordinated by seven conserved residues of the MPE motif as well as by a metal-bridging water molecule (Figures 1 and 5) and either a phosphate/sulfate/acetate moiety found in the crystallization buffer or substrate/product/inhibitor analogues added during protein preparation and crystallization, or simply by water molecules in the absence of these compounds. Metal ion α is co-ordinated by the side chains of two histidine and two aspartate residues via N and O atoms respectively, and metal ion β via N atoms of imidazole side chains of two histidine residues and O-atoms of one aspartate and one asparagine side chain (Figure 5A). It is important to point out that these residues are not universally invariant, and some substitutions in active-site residues and their repercussions on catalysis have already been alluded to in the previous section (Table 1 and Figure 1).

Active site of MPEs

Figure 5
Active site of MPEs

(A) Close-up view of the active site of the Rv0805 MPE (PDB code 3IB8) as a representative protein with all of the canonical residues of MPEs present in the sequence, and these residues are shown as sticks with carbon in yellow, oxygen in red and nitrogen in blue. Other parts of the structure are shown as a blue ribbon. Active-site metals are shown as spheres, with cyan for Fe3+ and magenta for Mn2+, and the metal-bridging water molecule shown as a red sphere. AMP bound to the active site of Rv0805 is shown in sticks with carbon in green, oxygen in red, nitrogen in blue and phosphorus in orange. Octahedral co-ordination of the metals is depicted by black broken lines. (B) Overlay of the active site of Rv0805 with the active site of porcine PAP (PDB code 1UTE). Rv0805 is shown as a grey ribbon and its active-side residues are shown as sticks, with carbon in orange, oxygen in red and nitrogen in blue. PAP is shown as a blue ribbon and its active-site residues are shown as sticks, with carbon in yellow, oxygen in red and nitrogen in blue. Metal ions Fe3+ and Mn2+ are shown as cyan and magenta spheres respectively, and metal-bridging water molecules are shown as red spheres. Spheres belonging to PAP are marked α′, β′ and O′. PAPs contain a unique tyrosine residue (Tyr55 in pig PAP) that takes over the function of histidine (His23 in Rv0805) in co-ordinating metal at the α-site.

Figure 5
Active site of MPEs

(A) Close-up view of the active site of the Rv0805 MPE (PDB code 3IB8) as a representative protein with all of the canonical residues of MPEs present in the sequence, and these residues are shown as sticks with carbon in yellow, oxygen in red and nitrogen in blue. Other parts of the structure are shown as a blue ribbon. Active-site metals are shown as spheres, with cyan for Fe3+ and magenta for Mn2+, and the metal-bridging water molecule shown as a red sphere. AMP bound to the active site of Rv0805 is shown in sticks with carbon in green, oxygen in red, nitrogen in blue and phosphorus in orange. Octahedral co-ordination of the metals is depicted by black broken lines. (B) Overlay of the active site of Rv0805 with the active site of porcine PAP (PDB code 1UTE). Rv0805 is shown as a grey ribbon and its active-side residues are shown as sticks, with carbon in orange, oxygen in red and nitrogen in blue. PAP is shown as a blue ribbon and its active-site residues are shown as sticks, with carbon in yellow, oxygen in red and nitrogen in blue. Metal ions Fe3+ and Mn2+ are shown as cyan and magenta spheres respectively, and metal-bridging water molecules are shown as red spheres. Spheres belonging to PAP are marked α′, β′ and O′. PAPs contain a unique tyrosine residue (Tyr55 in pig PAP) that takes over the function of histidine (His23 in Rv0805) in co-ordinating metal at the α-site.

PAPs, in addition to the canonical MPE motif residues, show the presence of an additional conserved tyrosine residue in sequence block II (Figure 1). This residue in its deprotonated form co-ordinates the metal ion α instead of a canonical histidine in sequence block I, which is lacking in PAPs (Figure 5B). This interaction between the conserved tyrosine and ferric ion results in a charge transfer transition and is thus responsible for the characteristic colour of PAPs [59,60]. These deviations in the amino acid sequences of the active site as well as the shape and hydrophobicity of the active-site cleft and the substrate-binding site account for the differences in substrate and metal ion specificity and affinity in diverse MPEs.

Variable tertiary structures of MPEs

MPEs exist in a variety of different homo-oligomeric states ranging from monomers to hexamers (Figure 6). MPEs reported to be monomers include Vps29 [8], Dbr1 [58], mammalian PAPs [61] and the 2′3′-cyclic nucleotide phosphodiesterase SpdA [43]. Some examples of dimeric MPEs are Rv0805 [11,32,62] and Mre11 [63,64], the former being a swapped dimer [11]. Interestingly, dimerization of Rv0805 is regulated by metal occupancy at the active site [11,62]. Large plant PAPs (∼55 kDa) also form dimers and exhibit two types of structural organization. One comprises enzymes with subunits bound by a disulfide bridge formed by cysteine residues located in the C-terminal region (Figure 6). The second type lacks cysteine residues in this position and therefore does not form disulfide bridges between subunits [65].

Oligomerization of MPEs

Figure 6
Oligomerization of MPEs

Structures of oligomeric MPEs are shown as ribbon representations. In each case, MPE domains of different protomers are coloured either blue or grey, whereas accessory domains are shown in orange/light orange (C-terminal domain) or green/light green (N-terminal domain). Active-site metals are shown as spheres, with cyan for Fe3+, magenta for Mn2+, green for Zn2+ and red for Co2+, as well as the cysteine residues (yellow) forming the disulfide bridge in red kidney bean PAP. PDB codes are indicated within parentheses.

Figure 6
Oligomerization of MPEs

Structures of oligomeric MPEs are shown as ribbon representations. In each case, MPE domains of different protomers are coloured either blue or grey, whereas accessory domains are shown in orange/light orange (C-terminal domain) or green/light green (N-terminal domain). Active-site metals are shown as spheres, with cyan for Fe3+, magenta for Mn2+, green for Zn2+ and red for Co2+, as well as the cysteine residues (yellow) forming the disulfide bridge in red kidney bean PAP. PDB codes are indicated within parentheses.

Larger homo-oligomers of MPEs have also been reported. For instance, some small MPEs such as YfcE from E. coli form functional homo-tetramers, with the substrate-binding sites located in the conserved clefts formed between the two sets of dimers [56]. A plant PAP belonging to a novel group of high-molecular-mass (75 kDa) PAPs was shown to form a non-covalent homohexamer, a dimer of trimers [14]. Hexameric complexes have also been reported for GpdQ from E. aerogenes, which is a trimer of three symmetric dimers [9]. Interestingly, dimers of GpdQ are very similar in structure to those seen in Rv0805 (Figure 6). As seen in Rv0805, GpdQ is a swapped dimer, where C-terminal ‘cap’ elements are exchanged between two protomers. In GpdQ, these cap elements are linked to the core MPE domain by disulfide bonds [9]. It is important to note that the interfaces between the protomers can be very different for different enzymes [9,11,56,64,66]. A notable exception is the high similarity of the homodimer interface in Rv0805 and GpdQ [9,11]. Thus the ability of MPEs to oligomerize may not be ancestral, but may have evolved multiple times later in evolution.

Curiously, each protomer of oligomeric MPEs is self-sufficient in forming the active site and co-ordinating catalytic metals. Why then do some MPEs form oligomers? In some cases, we do have answers to this question. For instance, different structural organization (i.e. with and without disulfide bonding) is reflected in substrate preferences of plant PAPs [65]. Similarly, the oligomeric state is linked to catalytic efficiency in the case of GpdQ. Mutants in the GpdQ protein selected for greater catalytic efficiency tend to form smaller oligomers, but at the expense of thermodynamic stability [67]. In other cases, such as Rv0805 and Mre11, oligomerization may be crucial for the non-catalytic functions of these proteins. The dimeric structure of Rv0805 is predicted to be a crucial factor behind its ability to act as phosphodiesterase and at the same time moonlight as a non-catalytic cell-envelope-binding protein [11,37]. Although dimerization is not critical for nuclease activity of Mre11, it may enable architectural DNA-binding functions of the Mre11–Rad50–Nbs1 (MRN) complex by promoting high-affinity DNA interactions [68].

MECHANISM OF PHOSPHOESTER HYDROLYSIS BY MPEs

The mechanism by which MPEs catalyse phosphoester hydrolysis has been investigated using structural and biophysical techniques for some members of this family. There have been several focused reviews on the catalytic mechanism of protein phosphatases [69,70], PAPs [71] and the GpdQ glycerophosphodiesterase [72,73]. In this section, we merely outline the general scheme for catalysis, thus facilitating a greater understanding of the subsequent sections, and also identify points of difference between enzymes of the MPE superfamily.

MPEs rely on the two metal ions co-ordinated in their active site for their activity. Similar to other structurally distinct binuclear metallohydrolases, such as arginases and β-lactamases, a metal-co-ordinated water molecule acts as the attacking nucleophile initiating hydrolysis of the substrate phosphoester and formation of a penta-co-ordinated transition state. Following nucleophilic attack, the leaving group is protonated by an active-site amino acid and released from the protein, leaving behind the phosphate. The active site is then regenerated by releasing the bound phosphate upon addition of at least two water molecules [40,41,69,71,7476]. It has been suggested that, for some MPEs such as GpdQ, phosphate release may be the rate-limiting step of the catalytic cycle [74].

Although the above general mechanism is conserved across MPEs, some variations on this theme do exist. For instance, it is clear that the attacking nucleophile is a metal-co-ordinated water molecule, but the identity of the attacking nucleophile is debatable. For human PPP5 and PP1 [76,77] and pig PAP, it has been argued that a metal-bridging μ-hydroxide (abstracted from water) acts as the attacking nucleophile, whereas, in the cases of Mre11 from Pyrococcus furiosus [64], GpdQ from E. aerogenes [74] and red kidney bean PAP [78], a hydroxide bound by one of the two metals is thought to play this role. In other enzymes, such as 5′-nucleotidase (5′-NTdase) from E. coli, crystal structures suggest that both of these scenarios are operational, although the latter may be preferred [79]. Variations on the mechanism of catalysis are translated to differences in some of the biochemical properties of individual MPEs. For example, whereas in most MPEs, the protonation of the leaving group is thought to be performed by an active-site histidine residue, in some PAPs, such as sweet potato PAP [80] and human PAP [81], this role is thought to be performed by a glutamate or aspartate residue respectively. This may represent an important modification in the functioning of PAPs, as, unlike other MPEs that work best at slightly alkaline pH, PAPs function at slightly acidic pH [80]. Most strikingly, the modes of substrate (phosphate) binding at the active sites of different MPEs, as inferred from structural data, are also not identical. In the case of 5′-NTdase, binding of phosphate in the active site is monodentate (only to β) [79]. However, phosphate binding in the active sites of some PAPs [61,78] and human protein phosphatases [76] is predicted to be bidentate, engaging both metal ions. As a consequence of this difference, PAPs are subject to inhibition by oxoanions such as phosphate (product inhibition), vanadate and tungstate [82,83], whereas 5′-NTdase is reported to be relatively resistant to these [79].

METAL IONS IN CATALYSIS AND REGULATION

The importance of metal ions in phosphoester hydrolysis by MPEs is apparent from the above section. Abstraction of metals from these enzymes by EDTA treatment, or reducing metal occupancy in the active site of MPEs by mutation, severely compromises their activity [12,32,8487]. It follows then that the metal-binding residues of MPEs are likely under strong evolutionary purifying selection, and therefore contribute significantly to the sequence signature of MPE domain-containing proteins (Figure 1).

Despite the high degree of conservation in metal-binding residues, the repertoire of metals that different MPEs utilize is very diverse. Table 2 summarizes the metals that support catalysis of some well-characterized MPEs. It is immediately apparent that Mn2+ can be almost universally utilized as the catalytic cation. In several cases, such as P. furiosus Mre11 [64,88] and yeast Dbr1 [25], Mn2+ supports the highest phosphoesterase activity among the divalent cations tested. Indeed, for both of these enzymes, replacing Mn2+ with Mg2+ led to compromised catalytic activity [25,64,88]. For Mre11, this was explained by the inability of Mg2+ to occupy both metal-binding sites in the active site [64]. However, Mn2+ is not the optimal choice for catalysis for some MPEs. For instance, Ni2+ is preferred over Mn2+ by MJ0936, a phosphodiesterase from M. jannaschii [55]. It must also be noted that some MPEs harbour a homonuclear metal centre, i.e. the same metal ion occupying both sites, exemplified by Mre11 (Mn2+–Mn2+) [63,64], but others utilize a heteronuclear metal ion centre, for instance Rv0805 (Fe3+–Mn2+) [11,62] and red kidney bean PAP (Fe3+–Zn2+) [75,78]. A peculiar example of a heteronuclear metal ion centre is the YmdB phosphoesterase from B. subtilis that uses iron in two different oxidation states (i.e. Fe3+ and Fe2+) at its two binding sites [44]. Mammalian PAPs may also use a similar heteronuclear metal ion centre, containing iron in two different oxidation states [61,71].

Table 2
Biochemical properties and roles of accessory domains in some MPEs

Substrate utilization, metal preference and modulatory roles of accessory domains of some well-characterized MPEs. Metals in bold are those seen at the α- and β-sites in crystal structures. Other metals support catalytic activity, but no structural information is available to determine whether one or both sites are occupied by metal ions. Note that the structure of Dbr1 shows only a single Mn2+ ion at the β-site, but it is not clear whether binding of a single metal ion is sufficient for catalytic activity, or whether a second metal ion binds along with the substrate during the enzymatic reaction. Bis-pNPP, bis(p-nitrophenyl) phosphate; pNPP, p-nitrophenyl phosphate; pNPPP, p-nitrophenyl-(phenyl) phosphonate; Tm-pNPP, thymidine 5′-monophosphate-p-nitrophenyl ester.

Enzyme (source organism) Substrates Catalytic metals (α-β) Roles for accessory domains Reference(s) 
Rv0805 (Mycobacterium tuberculosisBis-pNPP, pNPPP, 3′5′-cGMP, 3′5′-cGMP, 2′3′-cAMP Fe3+–Mn2+ C-terminal domain mediates localization to the cell envelope, regulates expression levels, enhances activity against linear phosphoesters and stabilizes the catalytic domain [11,32,37,62
Mre11 (Homo sapiens, Saccharomyces cerevisiae, Pyrococcus furiosusssDNA, dsDNA, DNA hairpins Mn2+–Mn2+ C-terminal accessory domain keeps the catalytic domain autoinhibited, mediates interactions with Rad50 and regulates substrate preference [64,88,94,98,102,108
5′-Nucleotidase (Escherichia coliATP, ADP, AMP, pNPP, UDP-glucose Mn2+–Mn2+, Co2+ C-terminal domain regulates substrate binding and utilization [40,41,47,48,79,86,112,113
GpdQ (Enterobacter aerogenespNPP, bis-pNPP, organophosphate triesters Zn2+–Zn2+, Fe2+–Fe2+, Fe2+–Co2+, Fe2+–Mn2+, Mn2+–Mn2+, Fe2+–Cd2+, Cd2+–Cd2+ C-terminal cap may help in oligomerization [9,67,74,89
Dbr1 (Entamoeba histolytica, Saccharomyces cerevisiae2′5′-Phosphodiester ends in RNA (lariat intermediates) X–Mn2+, Mg2+, Ni2+ C-terminal domain required for catalysis possibly due to functions in substrate binding [25,58
MPPED2 (Rattus rattusBis-pNPP, pNPPP, Tm-pNPP Mn2+–Mn2+, Ni2+ N-terminal domain required for catalysis possibly by maintaining protein's stability [12,84
Enzyme (source organism) Substrates Catalytic metals (α-β) Roles for accessory domains Reference(s) 
Rv0805 (Mycobacterium tuberculosisBis-pNPP, pNPPP, 3′5′-cGMP, 3′5′-cGMP, 2′3′-cAMP Fe3+–Mn2+ C-terminal domain mediates localization to the cell envelope, regulates expression levels, enhances activity against linear phosphoesters and stabilizes the catalytic domain [11,32,37,62
Mre11 (Homo sapiens, Saccharomyces cerevisiae, Pyrococcus furiosusssDNA, dsDNA, DNA hairpins Mn2+–Mn2+ C-terminal accessory domain keeps the catalytic domain autoinhibited, mediates interactions with Rad50 and regulates substrate preference [64,88,94,98,102,108
5′-Nucleotidase (Escherichia coliATP, ADP, AMP, pNPP, UDP-glucose Mn2+–Mn2+, Co2+ C-terminal domain regulates substrate binding and utilization [40,41,47,48,79,86,112,113
GpdQ (Enterobacter aerogenespNPP, bis-pNPP, organophosphate triesters Zn2+–Zn2+, Fe2+–Fe2+, Fe2+–Co2+, Fe2+–Mn2+, Mn2+–Mn2+, Fe2+–Cd2+, Cd2+–Cd2+ C-terminal cap may help in oligomerization [9,67,74,89
Dbr1 (Entamoeba histolytica, Saccharomyces cerevisiae2′5′-Phosphodiester ends in RNA (lariat intermediates) X–Mn2+, Mg2+, Ni2+ C-terminal domain required for catalysis possibly due to functions in substrate binding [25,58
MPPED2 (Rattus rattusBis-pNPP, pNPPP, Tm-pNPP Mn2+–Mn2+, Ni2+ N-terminal domain required for catalysis possibly by maintaining protein's stability [12,84

Whether homonuclear or heteronuclear, the two metal-ion-binding sites in MPEs do not have the same affinity for cations. In GpdQ, for example, Zn2+ occupancy was found to be higher at the α-site than at the β-site [9,74], as also observed in most PAPs [71]. For the Dbr1 phosphodiesterase, it was observed that only the β-pocket was occupied by Mn2+ in the recombinantly expressed and purified enzyme [58], and a higher affinity of the β-site was also observed in the case of λ-PP [87]. In an extreme case, the affinity for metals can be very low, as seen in Vps29 [57]. Furthermore, although different metals may occupy the active sites of most MPEs, not all combinations of metals support phosphoesterase activity to equal extents. Daumann et al. [89] demonstrated that the GpdQ phosphodiesterase can form complexes with a number of different metals (Fe2+, Co2+, Zn2+ and Mn2+) and, even though the purified enzyme had a Fe2+–Zn2+ ion centre (due to the relatively high concentrations of these metals in bacteria), a homonuclear Mn2+–Mn2+ centre supported the highest substrate turnover (kcat). Similarly, for 5′-NTdase from E. coli (also referred to as UshA), it was demonstrated that replacing Zn2+ with Co2+ at the low-affinity site stimulated activity close to 40-fold [86]. Finally, it was possible to inhibit the activity of the PnkP phosphoesterase from C. thermocellum by replacing active-site-bound Ni2+ with Co2+, Cu2+ or Zn2+ ions [90].

Interestingly, the effect of metal ion replacement is also influenced by the substrate used. This phenomenon is perhaps best characterized for Cth-PnkP that has both monoesterase and diesterase activities. Keppetipola and Shuman [85] elegantly demonstrated that the effect of mutations in the active site of this enzyme on monoesterase and diesterase activities was strongly influenced by the metal cofactor used. Most dramatically, mutation of the metal-co-ordinating residue His189 to aspartate resulted in severely compromised phosphodiesterase activity in the presence of Ni2+, but stimulation was seen in the presence of Mn2+ as a consequence of a 24-fold increase in substrate affinity (i.e. Km) [85]. This curious observation identifies tremendous plasticity in the active sites of MPEs and indicates that the mode of interaction of active-site residues with different metals is not identical. Furthermore, it may also be argued that the metal occupancy of MPE active sites may play important regulatory roles in vivo by determining substrate specificity, particularly since some of these enzymes can hydrolyse multiple substrates in vitro.

Observations such as those described above raise some intriguing questions. For instance, which metals do these enzymes co-ordinate in vivo? How is this regulated by the cell? In an interesting review related to these questions, Valdez et al. [91] argue that cells may regulate local concentrations of metals in order to ensure that appropriate metals are chelated by metalloenzymes. Therefore, in the case of eukaryotic cells, compartmentalization may be the key to deciding which metal is bound in the active site of MPEs. Although there is no intracellular compartmentalization reported in prokaryotes, perhaps subcellular localization is the trick used by MPEs to regulate metal occupancy. Some bacterial MPEs, exemplified by YkuE (B. subtilis), are known to be secreted [92], whereas others, such as CpdA (E. coli) [29] are predicted to be intracellular. Our studies with Rv0805 (M. tuberculosis) have shown that the protein is partitioned between the cell envelope and cytosolic compartments [11,37]. It is likely that the repertoire of metals chelated by cytosolic and envelope-localized MPEs are different, providing an additional level of regulation of their activity and possibly also influencing the substrates that they utilize in these different compartments. Moreover, the concentrations of different metal ions in a cell may govern whether or not an MPE will be active. If such a scenario were true, it can be envisaged that some MPEs may play the role of ‘metal sensors’ and themselves contribute to regulating the concentrations of various metal ions in cells. Interestingly, the Cdc1 MPE from Saccharomyces cerevisiae has been implicated in regulating the levels of cytosolic Mn2+ ions through an as yet unknown mechanism [93]. Given the ability of most MPEs to bind Mn2+, this observation may support the above hypothesis.

REGULATORY ROLES OF ACCESSORY DOMAINS

Structural characterization of several MPEs has revealed that, along with the core MPE domain, these enzymes often possess additional structural elements (Figures 4 and 7, and Table 2). For example, MPE domains can be fused to N-terminal domains of various sizes (e.g. MPPED2 [84] or large plant PAPs [78]), or to C-terminal domains, as found in Rv0805 [11], Dbr1 [58], Mre11 [63,94], GpdQ [9] and Lmo2642 [95]. Finally, the MPE domain may also be sandwiched between N- and C-terminal domains, as found in PPP5 [76,96] and Cth-Pnkp [97]. In several cases, these elements are proximate to the active site, possibly acting as ‘gates’ or ‘caps’, and hence have been referred to as cap domains [9,37,64]. The size, sequence and structure of these elements are extremely variable [9,11,25,44,64,79,84], indicating that different MPEs acquired these additional structural elements independently. It therefore follows that these elements may fine-tune the functions of MPEs in different ways and with different effects. We refer to these domains as accessory domains in the following discussion, although it may be noted that the term ‘domain’ is being used rather loosely in this context.

Unambiguous functional assignments for these domains are available for a few enzymes and range from modulating catalysis and substrate utilization to mediating functionally relevant interactions (Table 2). A few illustrative examples are discussed below.

Regulation of catalysis

Given the catalytic promiscuity of MPEs, it is clear that cells must regulate their diverse activities carefully in order to prevent wasteful catalytic cycles and prevent toxicity. One of the ways in which this is achieved is through the regulatory influence of cap domains. For instance, part of the C-terminus domain of the Mre11 nuclease from the T4 bacteriophage is autoinhibitory, keeping the N-terminal MPE domain inactive [98]. Gao and Nelson [98] demonstrated that the negative charge of the C-terminus domain was crucial for its inhibitory function, suggesting that the C-terminus may competitively inhibit DNA binding to the active site of the nuclease domain. Furthermore, this inhibition was relieved upon binding of Mre11 to the Rad50 protein, possibly because of restructuring of the C-terminus [98].

In a strikingly similar case, human PP2B (also known as calcineurin), is also subject to autoinhibition by its C-terminus. Interaction with its regulators, such as calmodulin, allow this autoinhibition to be relieved [99]. A variation on this theme is observed in PPP5 and PAP from humans. In the case of PPP5, the regulatory N-terminal domain, which is composed of tetratricopeptide repeats (TPRs), acts to inhibit the catalytic domain by blocking the catalytic channel [96]. A similar autoinhibitory role is played by the ‘repression loop’ of mammalian PAPs [100,101]. Although the repression loop is not a canonical ‘cap domain’, it does point to a general need for cells to regulate the activity of MPEs through diverse mechanisms.

It is necessary to point out that not all accessory domains are inhibitory. Some, exemplified by the N-terminal cap of rat MPPED2 [84] or the C-terminal domain of Dbr1 from yeast [25], are indispensable for the activity of the protein and their deletion can result in complete loss of catalytic activity. In the case of MPPED2, this may be due to a role for the N-terminal cap in maintaining protein stability, since its deletion drives much of the protein into inclusion bodies upon heterologous expression (Janardan, V. and Visweswariah, S.S., unpublished work). A similar role in maintaining stability was also proposed for the cap domain of the GpdQ phosphodiesterase [67]. For Dbr1, the role of the C-terminal cap domain in catalysis is proposed to be promoting interaction with RNA via stabilizing the LRL of this enzyme [25].

Substrate specificity/utilization

The accessory domains of some MPEs serve to regulate the utilization of substrates. This phenomenon is perhaps most clearly illustrated for 5′-NTdase from E. coli. 5′-NTdase possesses two domains: the N-terminal catalytic domain and the C-terminal regulatory domain. Whereas the catalytic centre of the enzyme is entirely in the N-terminal domain, the role of substrate binding is performed by the C-terminal domain. These two domains are linked together by an α-helical linker [79]. Catalysis by the enzyme depends on a large relative rotation between the two domains that ‘delivers’ the bound substrate to the active site of the enzyme [40]. Interestingly, the N-terminal domain alone shows phosphoesterase activity in vitro, with substrates such as ATP, ADP and pNPP, albeit with a marked increase in Km (i.e. a decrease in substrate affinity) [41]. Most strikingly, the N-terminal domain alone was incapable of hydrolysing AMP, whereas the full-length enzyme efficiently utilized it as a substrate [41]. Similar effects, although not as dramatic, have been observed for the Rv0805 phosphodiesterase, which has a C-terminal cap domain. When the cap domain of Rv0805 was deleted, the utilization of linear phosphodiester substrates was compromised, whereas cyclic substrates such as 3′5′-cAMP were hydrolysed with comparable efficiency [11].

The ability of cap domains to modulate substrate specificity and utilization has been elegantly exploited by Nature to direct the activity of the Mre11 nuclease towards specific substrates. The Mre11 nuclease can act as an endonuclease as well as an exonuclease in vitro [88]. Which of these two activities predominates is determined by its interacting partner, Rad50. Mre11 forms multiple interactions with Rad50, one of which is through the former's ‘capping’ domain [64,94]. Cycles of ATP binding and hydrolysis by Rad50 lead to formation of ‘closed’ or ‘open’ complexes. The closed complex acts predominantly as an endonuclease, whereas the open complex is predominantly an exonuclease [102]. This phenomenon has also been investigated structurally and is now known to be crucially dependent on the conformation of the capping domain of Mre11, that in turn serves to ‘open’ or ‘close’ the substrate-binding pocket in Mre11 [94].

Functional interactions

In addition to directly modulating catalysis, the accessory domains of some MPEs may also modulate their functions by allowing interactions with other cellular components. This has already been alluded to for the interaction between Mre11 and Rad50 proteins [64,94], and for the catalytic subunit of calcineurin/PP2B with its regulators calcineurin B and calmodulin [99]. Apart from interactions with other proteins, the cap domain of MPEs may also mediate interaction with other components of the cell. We have shown recently that the C-terminus domain of the Rv0805 phosphodiesterase mediated the interaction of this protein with the cell wall of mycobacteria [37]. This interaction was crucial for retention of the enzyme in the cell envelope fraction of mycobacteria [37]. Interestingly, mutations in the active site of Rv0805 could mimic C-terminal deletion by preventing interaction with the cell wall [37], arguing for intimate communication between the capping and catalytic domains of MPEs.

From the above examples, it is clear that the regulatory domains of MPEs are functionally relevant, and a complete understanding of MPE function cannot be achieved without clear knowledge of the roles of these domains. In most cases, however, their roles remain elusive, representing a conspicuous lacuna in our understanding of these enigmatic proteins. Notably, members of the Vps29/YfcE branch [8,56] of the MPE tree (Figure 2) and proteins similar to the MJ0936 phosphoesterase from M. janaschii [55] lack any additional domains. These proteins, which possess the MPE domain alone, are argued to be similar to the ancestral MPE domain that then may have evolved additional structural elements to fulfil regulatory roles [56]. This does not mean, however, that these proteins have not evolved or specialized in function. As discussed above, both Vps29 and YfcE have critical differences in their active-site residues that render them either inactive (Vps29) or unable to hydrolyse 2′3′-cyclic nucleotides (YfcE) [8]. Additionally, it has been proposed that substrate binding by YfcE, which forms a tetramer in vitro, occurs in a cleft between two protomers [56]. Thus, whereas these proteins may not contain regulatory domains, they seem to achieve specialization for their cellular roles by modifying the active site directly.

THE MPE FOLD AS A PROTEIN-INTERACTION SCAFFOLD: FUNCTIONS BEYOND CATALYSIS

In addition to recruiting proteins through their accessory domains, some MPEs, particularly eukaryotic enzymes, also utilize the MPE domain itself to interact with other proteins of the cell. For example, eukaryotic protein phosphatases such as PP1 and PP2A may interact with a number of regulatory proteins that govern the substrate specificity of these otherwise promiscuous phosphatases. It is crucial to note that most eukaryotes possess a relatively small number to genes coding for the catalytic subunits of protein phosphatases (1–4 for PP1), but hundreds of regulatory partners that allow the same catalytic subunit to be used in different signalling pathways and against several different substrates. This method of regulating the activity of protein phosphatases appears to be an important adaptation to harness their catalytic activity economically, in a tightly regulated manner. Comprehensive and excellent reviews by Shi [70] and by Virshup and Shenolikar [103] provide greater detail.

Yet another example of functionally important interactions with the MPE domain is the interaction of the Mre11 nuclease with the Nbs1 protein (Xsr2 from yeast), forming the MRN complex with Rad50. Whereas interaction with Nbs1 is not required for the repair of double-stranded breaks by the MRN complex, its importance in vivo is highlighted by its actions as a signalling protein that activates other checkpoint proteins such as the ATM (ataxia telangiectasia mutated) kinase in response to double-stranded breaks [104,105]. Critical for this function is the ability of Nbs1 to bind to Mre11 through the former's C-terminus. We now know that, unlike Rad50 which interacts primarily with the accessory domains of Mre11, the C-terminus of Nbs1 forms at least two independent contacts with the catalytic domain of Mre11 directly, each consisting of multiple polar, hydrophobic and stacking interactions [63]. The crucial importance of these interactions is demonstrated by the fact that several hypomorphic mutations in Mre11 associated with ATLD are mapped to the regions of interaction between the two proteins [23,63].

Remarkably, the ability of the Vps29 protein, a component of the retromer complex, to recruit other proteins has now become its primary function, and it appears to have lost its catalytic ability. Indeed, Vps29 orthologues from most eukaryotes have lost more than one catalytically important residue at the active site, demonstrating a lack of selection for these residues during the course of evolution. The crystal structure of mouse Vps29 demonstrated that, despite these changes, the overall structure and fold of the protein was similar to that of other MPEs [8]. Mouse Vps29 also bound Mn2+ in its active site when crystals of the protein were soaked in MnSO4, but, unlike several other MPEs, was unable to bind sulfate in its active site [8]. Subsequent solution studies showed that Vps29 can also bind Zn2+, albeit with significantly lower affinity than other MPEs, but neither Mn2+ nor Zn2+ supported catalysis [57]. Importantly, interaction of Vps29 with its interacting partner Vps35 was not significantly affected by metal occupancy [57]. This is in contrast with the Mre11 protein, whose interaction with Nbs1 is significantly potentiated by metals [63].

Whereas Vps29 represents an extreme example of catalysis-independent functionality, other catalytically competent MPEs may also have catalysis-independent roles. For instance, the nuclease activity of Mre11 is dispensable for some (but not all) of its functions in S. cerevisiae, namely mating-type switching and telomere maintenance [106,107]. Furthermore, the human Mre11-3 variant, which is rendered inactive due to a mutation in its active site, but retains interaction with Rad50 and Nbs1, could assemble into a partially functional complex with its interacting partners in cells, once again demonstrating that some of the roles of this protein are catalysis-independent [108]. We showed that overexpression of the Rv0805 phosphodiesterase in mycobacteria could result in cell wall perturbation that was independent of its catalytic activity [11,36,37]. Another prime candidate for catalysis-independent functions is the eukaryotic MPPED2 protein that has very poor catalytic activity in vitro [12]. The low catalytic efficiency of MPPED2 was attributed to the presence of a glycine residue instead of histidine in sequence block V (Figure 1 and Table 1). This substitution also allowed the active site of MPPED2 to bind 5′-AMP and 5′-GMP with high affinity (∼50 nM), further inhibiting enzymatic activity in vitro [84]. These observations suggest that, under physiological conditions, the activity of MPPED2 is kept inhibited by the 5′-nucleotides and hence its functions in the cell may not involve phosphoester hydrolysis. Additionally, structural elements present in this protein, such as a proline-rich N-terminus and a solvent-exposed hydrophobic loop strongly suggest that this protein may serve as a protein interaction scaffold in vivo [84]. These avenues are yet to be explored and will no doubt add to our appreciation of the non-enzymatic functions of members of the MPE superfamily.

SUMMARY AND PERSPECTIVE

As we continue to uncover the functions of MPE domain-containing proteins, it has become apparent that, despite marked structural conservation, these proteins cannot be assigned a single functional role. Diversity in the various functions performed by MPEs is generated primarily through embellishments on the conserved MPE domain. These may include substitutions in active-site residues, the influence of interacting proteins or unique structural elements. Thus the MPE domain may be viewed as an example of a protein domain that has been elaborated on several times for novel functions. During the course of evolution, there appears to have been a steady expansion in the number of MPE domain-containing proteins. The fact that bacterial genomes such as those of E. coli and B. subtilis encode <10 MPE domain-containing proteins, yeasts S. cerevisiae and S. pombe encode close to 20 MPE domain-containing proteins, Drosophila melanogaster encodes 79 MPE domain-containing proteins, and human and mouse genomes code for close to 100 MPE domain-containing proteins is illustrative of this [7] (Figure 8). This increase in the number of MPE domain-containing proteins in the genome is also accompanied by re-appropriation of their functions. For instance, we see the evolution of intron-processing enzymes from an ancestral MPE in eukaryotes, and the use of MPEs as protein-interaction scaffolds, a function altogether independent of catalysis (Figure 8).

Domain organization of MPEs

Figure 7
Domain organization of MPEs

Left: schematic depictions of the organization of various domains in MPEs. Structures of protomers/monomers of illustrative proteins are also provided in comparable orientation of the MPE domain. In the schematic diagrams as well as in structures, the MPE domains are in blue, whereas the accessory domains at the N- or C-terminus are in green and orange respectively. Active-site residues are shown as sticks, with carbon in yellow, oxygen in red and nitrogen in blue, whereas active-site metals are shown as spheres with cyan for Fe3+/Fe2+, magenta for Mn2+, green for Zn2+ and red for Co2+. The first row shows single-domain proteins with the smallest MPE fold and the second row shows the single MPE domain proteins with larger MPE domains due to additional insertions within the MPE domain. PDB codes are indicated within parentheses.

Figure 7
Domain organization of MPEs

Left: schematic depictions of the organization of various domains in MPEs. Structures of protomers/monomers of illustrative proteins are also provided in comparable orientation of the MPE domain. In the schematic diagrams as well as in structures, the MPE domains are in blue, whereas the accessory domains at the N- or C-terminus are in green and orange respectively. Active-site residues are shown as sticks, with carbon in yellow, oxygen in red and nitrogen in blue, whereas active-site metals are shown as spheres with cyan for Fe3+/Fe2+, magenta for Mn2+, green for Zn2+ and red for Co2+. The first row shows single-domain proteins with the smallest MPE fold and the second row shows the single MPE domain proteins with larger MPE domains due to additional insertions within the MPE domain. PDB codes are indicated within parentheses.

Expansion of MPE domain-containing proteins during evolution

Figure 8
Expansion of MPE domain-containing proteins during evolution

The average numbers of MPE domain-containing proteins per genome from genera representing viruses, bacteria, archaea and eukarya were obtained from Pfam. For a few species, the number of MPE domain-containing proteins are indicated in parentheses and corresponding bars are coloured red. Identified functions for MPEs are overlaid on the histogram. The number of MPE domain-containing proteins and the kinds of functions they perform show a steady increase from simpler to more complex organisms. DNA-repair enzymes and phosphoprotein phosphatases are found across all domains of life, whereas increased organism complexity is accompanied by the appearance of MPEs with more specialized functions such as RNA splicing or non-catalytic protein scaffolding.

Figure 8
Expansion of MPE domain-containing proteins during evolution

The average numbers of MPE domain-containing proteins per genome from genera representing viruses, bacteria, archaea and eukarya were obtained from Pfam. For a few species, the number of MPE domain-containing proteins are indicated in parentheses and corresponding bars are coloured red. Identified functions for MPEs are overlaid on the histogram. The number of MPE domain-containing proteins and the kinds of functions they perform show a steady increase from simpler to more complex organisms. DNA-repair enzymes and phosphoprotein phosphatases are found across all domains of life, whereas increased organism complexity is accompanied by the appearance of MPEs with more specialized functions such as RNA splicing or non-catalytic protein scaffolding.

To what may we attribute the apparent success of MPE domains in evolution? Versatility in substrate utilization is a property that is seen almost universally across members of this enzyme family and may be a possible answer. In different contexts, the esterase activity is controlled stringently though diverse mechanisms such as the presence of accessory domains, metal ion occupancy and interacting partners. Curiously, specificity is by and large not built into the active site of these enzymes, which may imply a certain advantage to maintaining promiscuous substrate utilization in MPEs. This may seem counterintuitive, as it has been argued that catalytic promiscuity in MPEs such as GpdQ comes at the cost of catalytic efficiency [89]. However, the promiscuity of GpdQ in substrate and metal utilization make it remarkably adaptable and hence able to function under a wide range of conditions [89]. Directed evolution experiments showed that the reactivity and specificity of GpdQ could be vastly altered with just a few rounds of mutagenesis [67], suggesting that these enzymes may be ready substrates for evolution to combat changing environments. This promiscuity and adaptability of MPEs has triggered interest in creating mimetics of their active site for the purpose of bioremediation and detoxification of organophosphates [72].

Another possible reason for the ubiquitous presence of MPE domain-containing proteins could be the structural stability of the MPE domain. Whereas the MPE fold may have been initially optimized for metal co-ordination and catalysis, it is now clear that several MPE proteins that show low metal occupancy at the active site can still adopt the MPE fold. For instance, the structures of the apo Vps29 protein [8], the N97A mutant of Rv0805 (a mutant with reduced active-site metal occupancy) [62] and the apo form of MJ0936 [55] all closely resemble metal-bound forms. It is possible to envisage that this property may have led to the selection of these domains as scaffolds for protein–protein interactions, best exemplified by Vps29. In the future, it will be interesting to identify whether or not other MPEs also possess similar ‘scaffolding’ functions. Is this an ancestral property of MPE domains or a relatively new function restricted to complex organisms? Furthermore, from available information on the Mre11 nuclease and eukaryotic protein phosphatases, we know already that, for some enzymes, catalytic activity and protein interaction are interrelated. It is likely that this property is also true for bacterial and archaeal enzymes, several of which have as yet unknown substrates in vivo. It may therefore be fruitful to analyse the interactomes of these enzymes to be able to infer their functionally relevant substrates.

A large majority of the enzymes belonging to the MPE superfamily remain uncharacterized and, for several that have been structurally and biochemically studied, we do not know their precise functions in vivo. In the present review, we have outlined the enormous diversity in substrate utilization and cellular functions of these enzymes, and this fact must be kept in mind when characterizing novel members of this superfamily. As more and more enzymes are described, it may be possible in the future to identify sequence signatures in MPEs that confer specificities to various substrates and hence allow reliable prediction of which substrates these enzymes may utilize. Until then, however, we shall have to rely on structural, biochemical and genetic approaches to unravel the roles of these enzymes. We hope that the present review will serve as a framework for a better appreciation of this family of proteins and provide directions for further research into functions of as yet uncharacterized MPEs.

Abbreviations

     
  • ATLD

    ataxia-telangiectasia-like disease

  •  
  • LRL

    lariat recognition loop

  •  
  • MPE

    metallophosphoesterase

  •  
  • MPPED2

    metallophosphoesterase domain-containing 2

  •  
  • MRN

    Mre11–Rad50–Nbs1

  •  
  • 5′-NTdase

    5′-nucleotidase

  •  
  • PAP

    purple acid phosphatase

  •  
  • pNPP

    p-nitrophenyl phosphate

  •  
  • λ-PP

    lambda phage phosphatase

  •  
  • PP

    protein phosphatase

  •  
  • PPD1

    diphosphonucleotide phosphatase/phosphodiesterase

  •  
  • PPP

    phosphoprotein phosphatase

  •  
  • UDP-DAGn

    UDP-2,3-diacylglucosamine

  •  
  • Vps29

    vacuolar protein sorting 29

FUNDING

N.M. is a recipient of a Research Associate fellowship from the Indian Institute of Science. S.S.V. acknowledges the Department of Science and Technology, Government of India for the J.C. Bose Fellowship, and Department of Biotechnology for funding. M.P. acknowledges the Slovenian Research Agency [programme grant number P1-0104]. M.P. and S.S.V. were recipients of an Indo-Slovene Research Grant from the Department of Science and Technology (India) and the Slovenian Research Agency.

References

References
1
Hunter
T.
Why nature chose phosphate to modify proteins
Philos. Trans. R. Soc. Lond. B Biol. Sci.
2012
, vol. 
367
 (pg. 
2513
-
2516
)
[PubMed]
2
Schroeder
G.K.
Lad
C.
Wyman
P.
Williams
N.H.
Wolfenden
R.
The time required for water attack at the phosphorus atom of simple phosphodiesters and of DNA
Proc. Natl. Acad. Sci. U.S.A.
2006
, vol. 
103
 (pg. 
4052
-
4055
)
[PubMed]
3
Lad
C.
Williams
N.H.
Wolfenden
R.
The rate of hydrolysis of phosphomonoester dianions and the exceptional catalytic proficiencies of protein and inositol phosphatases
Proc. Natl. Acad. Sci. U.S.A.
2003
, vol. 
100
 (pg. 
5607
-
5610
)
[PubMed]
4
Moorhead
G.B.G.
De Wever
V.
Templeton
G.
Kerk
D.
Evolution of protein phosphatases in plants and animals
Biochem. J.
2009
, vol. 
417
 (pg. 
401
-
409
)
[PubMed]
5
Koonin
E.V.
Bacterial and bacteriophage protein phosphatases
Mol. Microbiol.
1993
, vol. 
8
 (pg. 
785
-
786
)
[PubMed]
6
Koonin
E.V.
Conserved sequence pattern in a wide variety of phosphoesterases
Protein Sci.
1994
, vol. 
3
 (pg. 
356
-
358
)
[PubMed]
7
Finn
R.D.
Bateman
A.
Clements
J.
Coggill
P.
Eberhardt
R.Y.
Eddy
S.R.
Heger
A.
Hetherington
K.
Holm
L.
Mistry
J.
, et al. 
Pfam: the protein families database
Nucleic Acids Res.
2014
, vol. 
42
 (pg. 
D222
-
D230
)
[PubMed]
8
Collins
B.M.
Skinner
C.F.
Watson
P.J.
Seaman
M.N.J.
Owen
D.J.
Vps29 has a phosphoesterase fold that acts as a protein interaction scaffold for retromer assembly
Nat. Struct. Mol. Biol.
2005
, vol. 
12
 (pg. 
594
-
602
)
[PubMed]
9
Jackson
C.J.
Carr
P.D.
Liu
J.W.
Watt
S.J.
Beck
J.L.
Ollis
D.L.
The structure and function of a novel glycerophosphodiesterase from Enterobacter aerogenes
J. Mol. Biol.
2007
, vol. 
367
 (pg. 
1047
-
1062
)
[PubMed]
10
McLoughlin
S.Y.
Jackson
C.
Liu
J.W.
Ollis
D.L.
Growth of Escherichia coli coexpressing phosphotriesterase and glycerophosphodiester phosphodiesterase, using paraoxon as the sole phosphorus source
Appl. Environ. Microbiol.
2004
, vol. 
70
 (pg. 
404
-
412
)
[PubMed]
11
Podobnik
M.
Tyagi
R.
Matange
N.
Dermol
U.
Gupta
A.K.
Mattoo
R.
Seshadri
K.
Visweswariah
S.S.
A mycobacterial cyclic AMP phosphodiesterase that moonlights as a modifier of cell wall permeability
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
32846
-
32857
)
[PubMed]
12
Tyagi
R.
Shenoy
A.R.
Visweswariah
S.S.
Characterization of an evolutionarily conserved metallophosphoesterase that is expressed in the fetal brain and associated with the WAGR syndrome
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
5217
-
5228
)
[PubMed]
13
Young
H.E.
Donohue
M.P.
Smirnova
T.I.
Smirnov
A.I.
Zhou
P.
The UDP-diacylglucosamine pyrophosphohydrolase LpxH in lipid A biosynthesis utilizes Mn2+ cluster for catalysis
J. Biol. Chem.
2013
, vol. 
288
 (pg. 
26987
-
27001
)
[PubMed]
14
Antonyuk
S.V.
Olczak
M.
Olczak
T.
Ciuraszkiewicz
J.
Strange
R.W.
The structure of a purple acid phosphatase involved in plant growth and pathogen defence exhibits a novel immunoglobulin-like fold
IUCrJ
2014
, vol. 
1
 (pg. 
101
-
109
)
[PubMed]
15
Monds
R.D.
Newell
P.D.
Wagner
J.C.
Schwartzman
J.A.
Lu
W.
Rabinowitz
J.D.
O'Toole
G.A.
Di-adenosine tetraphosphate (Ap4A) metabolism impacts biofilm formation by Pseudomonas fluorescens via modulation of c-di-GMP-dependent pathways
J. Bacteriol.
2010
, vol. 
192
 (pg. 
3011
-
3023
)
[PubMed]
16
Nishimura
A.
Moriya
S.
Ukai
H.
Nagai
K.
Wachi
M.
Yamada
Y.
Diadenosine 5′,5″′-P1,P4-tetraphosphate (Ap4A) controls the timing of cell division in Escherichia coli
Genes Cells
1997
, vol. 
2
 (pg. 
401
-
413
)
[PubMed]
17
Sasaki
M.
Takegawa
K.
Kimura
Y.
Enzymatic characteristics of an ApaH-like phosphatase, PrpA, and a diadenosine tetraphosphate hydrolase, ApaH, from Myxococcus xanthus
FEBS Lett.
2014
, vol. 
588
 (pg. 
3395
-
3402
)
[PubMed]
18
Connelly
J.C.
Leach
D.R.
Tethering on the brink: the evolutionarily conserved Mre11–Rad50 complex
Trends Biochem. Sci.
2002
, vol. 
27
 (pg. 
410
-
418
)
[PubMed]
19
Buis
J.
Wu
Y.P.
Deng
Y.B.
Leddon
J.
Westfield
G.
Eckersdorff
M.
Sekiguchi
J.M.
Chang
S.
Ferguson
D.O.
Mre11 nuclease activity has essential roles in DNA repair and genomic stability distinct from ATM activation
Cell
2008
, vol. 
135
 (pg. 
85
-
96
)
[PubMed]
20
Xiao
Y.H.
Weaver
D.T.
Conditional gene targeted deletion by Cre recombinase demonstrates the requirement for the double-strand break repair Mre11 protein in murine embryonic stem cells
Nucleic Acids Res.
1997
, vol. 
25
 (pg. 
2985
-
2991
)
[PubMed]
21
Fernet
M.
Gribaa
M.
Salih
M.A.M.
Seidahmed
M.Z.
Hall
J.
Koenig
M.
Identification and functional consequences of a novel MRE11 mutation affecting 10 Saudi Arabian patients with the ataxia telangiectasia-like disorder
Hum. Mol. Genet.
2005
, vol. 
14
 (pg. 
307
-
318
)
[PubMed]
22
Lee
J.H.
Ghirlando
R.
Bhaskara
V.
Hoffmeyer
M.R.
Gu
J.
Paull
T.T.
Regulation of Mre11/Rad50 by Nbs1: effects on nucleotide-dependent DNA binding and association with ataxia-telangiectasia-like disorder mutant complexes
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
45171
-
45181
)
[PubMed]
23
Stewart
G.S.
Maser
R.S.
Stankovic
T.
Bressan
D.A.
Kaplan
M.I.
Jaspers
N.G.J.
Raams
A.
Byrd
P.J.
Petrini
J.H.J.
Taylor
A.M.R.
The DNA double-strand break repair gene hMRE11 is mutated in individuals with an ataxia-telangiectasia-like disorder
Cell
1999
, vol. 
99
 (pg. 
577
-
587
)
[PubMed]
24
Chapman
K.B.
Boeke
J.D.
Isolation and characterization of the gene encoding yeast debranching enzyme
Cell
1991
, vol. 
65
 (pg. 
483
-
492
)
[PubMed]
25
Khalid
M.F.
Damha
M.J.
Shuman
S.
Schwer
B.
Structure–function analysis of yeast RNA debranching enzyme (Dbr1), a manganese-dependent phosphodiesterase
Nucleic Acids Res.
2005
, vol. 
33
 (pg. 
6349
-
6360
)
[PubMed]
26
Nam
K.
Lee
G.
Trambley
J.
Devine
S.E.
Boeke
J.D.
Severe growth defect in a Schizosaccharomyces pombe mutant defective in intron lariat degradation
Mol. Cell. Biol.
1997
, vol. 
17
 (pg. 
809
-
818
)
[PubMed]
27
Uhrig
R.G.
Kerk
D.
Moorhead
G.B.
Evolution of bacterial-like phosphoprotein phosphatases in photosynthetic eukaryotes features ancestral mitochondrial or archaeal origin and possible lateral gene transfer
Plant Physiol.
2013
, vol. 
163
 (pg. 
1829
-
1843
)
[PubMed]
28
Andreeva
A.V.
Kutuzov
M.A.
Widespread presence of “bacterial-like” PPP phosphatases in eukaryotes
BMC Evol. Biol.
2004
, vol. 
4
 pg. 
47
 
[PubMed]
29
Imamura
R.
Yamanaka
K.
Ogura
T.
Hiraga
S.
Fujita
N.
Ishihama
A.
Niki
H.
Identification of the cpdA gene encoding cyclic 3′,5′-adenosine monophosphate phosphodiesterase in Escherichia coli
J. Biol. Chem.
1996
, vol. 
271
 (pg. 
25423
-
25429
)
[PubMed]
30
Fuchs
E.L.
Brutinel
E.D.
Klem
E.R.
Fehr
A.R.
Yahr
T.L.
Wolfgang
M.C.
In vitro and in vivo characterization of the Pseudomonas aeruginosa cyclic AMP (cAMP) phosphodiesterase cpdA, required for cAMP homeostasis and virulence factor regulation
J. Bacteriol.
2010
, vol. 
192
 (pg. 
2779
-
2790
)
[PubMed]
31
Macfadyen
L.P.
Ma
C.X.
Redfield
R.J.
A 3′,5′ cyclic AMP (cAMP) phosphodiesterase modulates cAMP levels and optimizes competence in Haemophilus influenzae Rd
J. Bacteriol.
1998
, vol. 
180
 (pg. 
4401
-
4405
)
[PubMed]
32
Shenoy
A.R.
Sreenath
N.
Podobnik
M.
Kovacevic
M.
Visweswariah
S.S.
The Rv0805 gene from Mycobacterium tuberculosis encodes a 3′,5′-cyclic nucleotide phosphodiesterase: biochemical and mutational analysis
Biochemistry
2005
, vol. 
44
 (pg. 
15695
-
15704
)
[PubMed]
33
Richter
W.
3′,5′-cyclic nucleotide phosphodiesterases class III: members, structure, and catalytic mechanism
Proteins
2002
, vol. 
46
 (pg. 
278
-
286
)
[PubMed]
34
Rousseau
E.
Gagnon
J.
Lugnier
C.
Biochemical and pharmacological characterization of cyclic nucleotide phosphodiesterase in airway epithelium
Mol. Cell. Biochem.
1994
, vol. 
140
 (pg. 
171
-
175
)
[PubMed]
35
Keppetipola
N.
Shuman
S.
A phosphate-binding histidine of binuclear metallophosphodiesterase enzymes is a determinant of 2′,3′-cyclic nucleotide phosphodiesterase activity
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
30942
-
30949
)
[PubMed]
36
Matange
N.
Hunt
D.M.
Buxton
R.S.
Visweswariah
S.S.
Overexpression of the Rv0805 phosphodiesterase elicits a cAMP-independent transcriptional response
Tuberculosis (Edinb.)
2013
, vol. 
93
 (pg. 
492
-
500
)
[PubMed]
37
Matange
N.
Podobnik
M.
Visweswariah
S.S.
The non-catalytic “cap domain” of a mycobacterial metallophosphoesterase regulates its expression and localization in the cell
J. Biol. Chem.
2014
, vol. 
289
 (pg. 
22470
-
22481
)
[PubMed]
38
Anraku
Y.
A new cyclic phosphodiesterase having a 3′-nucleotidase activity from Escherichia coli B. I. Purification and some properties of the enzyme
J. Biol. Chem.
1964
, vol. 
239
 (pg. 
3412
-
3419
)
[PubMed]
39
Anraku
Y.
A new cyclic phosphodiesterase having a 3′-nucleotidase activity from Escherichia coli B. II. Further studies on substrate specificity and mode of action of the enzyme
J. Biol. Chem.
1964
, vol. 
239
 (pg. 
3420
-
3424
)
[PubMed]
40
Schultz-Heienbrok
R.
Maier
T.
Strater
N.
A large hinge bending domain rotation is necessary for the catalytic function of Escherichia coli 5′-nucleotidase
Biochemistry
2005
, vol. 
44
 (pg. 
2244
-
2252
)
[PubMed]
41
Krug
U.
Patzschke
R.
Zebisch
M.
Balbach
J.
Strater
N.
Contribution of the two domains of E. coli 5′-nucleotidase to substrate specificity and catalysis
FEBS Lett.
2013
, vol. 
587
 (pg. 
460
-
466
)
[PubMed]
42
Trulzsch
K.
Roggenkamp
A.
Pelludat
C.
Rakin
A.
Jacobi
C.A.
Heesemann
J.
Cloning and characterization of the gene encoding periplasmic 2′,3′-cyclic phosphodiesterase of Yersinia enterocolitica O:8
Microbiology
2001
, vol. 
147
 (pg. 
203
-
213
)
[PubMed]
43
Mathieu-Demaziere
C.
Poinsot
V.
Masson-Boivin
C.
Garnerone
A.M.
Batut
J.
Biochemical and functional characterization of SpdA, a 2′,3′ cyclic nucleotide phosphodiesterase from Sinorhizobium meliloti
BMC Microbiol.
2013
, vol. 
13
 
[PubMed]
44
Diethmaier
C.
Newman
J.A.
Kovacs
A.T.
Kaever
V.
Herzberg
C.
Rodrigues
C.
Boonstra
M.
Kuipers
O.P.
Lewis
R.J.
Stulke
J.
The YmdB phosphodiesterase is a global regulator of late adaptive responses in Bacillus subtilis
J. Bacteriol.
2014
, vol. 
196
 (pg. 
265
-
275
)
[PubMed]
45
Leveque
F.
Blanchinroland
S.
Fayat
G.
Plateau
P.
Blanquet
S.
Design and characterization of Escherichia coli mutants devoid of Ap4n-hydrolase activity
J. Mol. Biol.
1990
, vol. 
212
 (pg. 
319
-
329
)
[PubMed]
46
Mechulam
Y.
Fromant
M.
Mellot
P.
Plateau
P.
Blanchinroland
S.
Fayat
G.
Blanquet
S.
Molecular cloning of the Escherichia coli gene for diadenosine 5′,5″′-P1,P4-tetraphosphate pyrophosphohydrolase
J. Bacteriol.
1985
, vol. 
164
 (pg. 
63
-
69
)
[PubMed]
47
Ruiz
A.
Hurtado
C.
Meireles Ribeiro
J.
Sillero
A.
Gunther Sillero
M.A.
Hydrolysis of bis(5′-nucleosidyl) polyphosphates by Escherichia coli 5′-nucleotidase
J. Bacteriol.
1989
, vol. 
171
 (pg. 
6703
-
6709
)
[PubMed]
48
Wang
L.
Zhou
Y.J.
Ji
D.B.
Lin
X.P.
Liu
Y.X.
Zhang
Y.X.
Liu
W.J.
Zhao
Z.K.
Identification of UshA as a major enzyme for NAD degradation in Escherichia coli
Enzyme Microb. Technol.
2014
, vol. 
58–59
 (pg. 
75
-
79
)
49
Babinski
K.J.
Kanjilal
S.J.
Raetz
C.R.H.
Accumulation of the lipid A precursor UDP-2,3-diacylglucosamine in an Escherichia coli mutant lacking the lpxH gene
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
25947
-
25956
)
[PubMed]
50
Aravind
L.
Koonin
E.V.
Phosphoesterase domains associated with DNA polymerases of diverse origins
Nucleic Acids Res.
1998
, vol. 
26
 (pg. 
3746
-
3752
)
[PubMed]
51
Schenk
G.
Korsinczky
M.L.
Hume
D.A.
Hamilton
S.
DeJersey
J.
Purple acid phosphatases from bacteria: similarities to mammalian and plant enzymes
Gene
2000
, vol. 
255
 (pg. 
419
-
424
)
[PubMed]
52
Schenk
G.
Guddat
L.W.
Ge
Y.
Carrington
L.E.
Hume
D.A.
Hamilton
S.
de Jersey
J.
Identification of mammalian-like purple acid phosphatases in a wide range of plants
Gene
2000
, vol. 
250
 (pg. 
117
-
125
)
[PubMed]
53
Ketcham
C.M.
Baumbach
G.A.
Bazer
F.W.
Roberts
R.M.
The Type 5, acid phosphatase from spleen of humans with hairy cell leukemia: purification, properties, immunological characterization, and comparison with porcine uteroferrin
J. Biol. Chem.
1985
, vol. 
260
 (pg. 
5768
-
5776
)
[PubMed]
54
Bozzo
G.G.
Raghothama
K.G.
Plaxton
W.C.
Purification and characterization of two secreted purple acid phosphatase isozymes from phosphate-starved tomato (Lycopersicon esculentum) cell cultures
Eur. J. Biochem.
2002
, vol. 
269
 (pg. 
6278
-
6286
)
[PubMed]
55
Chen
S.
Yakunin
A.F.
Kuznetsova
E.
Busso
D.
Pufan
R.
Proudfoot
M.
Kim
R.
Kim
S.H.
Structural and functional characterization of a novel phosphodiesterase from Methanococcus jannaschii
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
31854
-
31862
)
[PubMed]
56
Miller
D.J.
Shuvalova
L.
Evdokimova
E.
Savchenko
A.
Yakunin
A.F.
Anderson
W.F.
Structural and biochemical characterization of a novel Mn2+-dependent phosphodiesterase encoded by the yfcE gene
Protein Sci.
2007
, vol. 
16
 (pg. 
1338
-
1348
)
[PubMed]
57
Swarbrick
J.D.
Shaw
D.J.
Chhabra
S.
Ghai
R.
Valkov
E.
Norwood
S.J.
Seaman
M.N.
Collins
B.M.
VPS29 is not an active metallo-phosphatase but is a rigid scaffold required for retromer interaction with accessory proteins
PLoS ONE
2011
, vol. 
6
 pg. 
e20420
 
[PubMed]
58
Montemayor
E.J.
Katolik
A.
Clark
N.E.
Taylor
A.B.
Schuermann
J.P.
Combs
D.J.
Johnsson
R.
Holloway
S.P.
Stevens
S.W.
Damha
M.J.
Hart
P.J.
Structural basis of lariat RNA recognition by the intron debranching enzyme Dbr1
Nucleic Acids Res.
2014
, vol. 
42
 (pg. 
10845
-
10855
)
[PubMed]
59
Yang
Y.S.
McCormick
J.M.
Solomon
E.I.
Circular dichroism and magnetic circular dichroism studies of the mixed-valence binuclear non-heme iron active site in uteroferrin and its anion complexes
J. Am. Chem. Soc.
1997
, vol. 
119
 (pg. 
11832
-
11842
)
60
Averill
B.A.
Davis
J.C.
Burman
S.
Zirino
T.
Sandersloehr
J.
Loehr
T.M.
Sage
J.T.
Debrunner
P.G.
Spectroscopic and magnetic studies of the purple acid-phosphatase from bovine spleen
J. Am. Chem. Soc.
1987
, vol. 
109
 (pg. 
3760
-
3767
)
61
Guddat
L.W.
McAlpine
A.S.
Hume
D.
Hamilton
S.
de Jersey
J.
Martin
J.L.
Crystal structure of mammalian purple acid phosphatase
Structure
1999
, vol. 
7
 (pg. 
757
-
767
)
[PubMed]
62
Shenoy
A.R.
Capuder
M.
Draskovic
P.
Lamba
D.
Visweswariah
S.S.
Podobnik
M.
Structural and biochemical analysis of the Rv0805 cyclic nucleotide phosphodiesterase from Mycobacterium tuberculosis
J. Mol. Biol.
2007
, vol. 
365
 (pg. 
211
-
225
)
[PubMed]
63
Schiller
C.B.
Lammens
K.
Guerini
I.
Coordes
B.
Feldmann
H.
Schlauderer
F.
Mockel
C.
Schele
A.
Strasser
K.
Jackson
S.P.
Hopfner
K.P.
Structure of Mre11–Nbs1 complex yields insights into ataxia-telangiectasia-like disease mutations and DNA damage signaling
Nat. Struct. Mol. Biol.
2012
, vol. 
19
 (pg. 
693
-
700
)
[PubMed]
64
Hopfner
K.P.
Karcher
A.
Craig
L.
Woo
T.T.
Carney
J.P.
Tainer
J.A.
Structural biochemistry and interaction architecture of the DNA double-strand break repair Mre11 nuclease and Rad50-ATPase
Cell
2001
, vol. 
105
 (pg. 
473
-
485
)
[PubMed]
65
Olczak
M.
Morawiecka
B.
Watorek
W.
Plant purple acid phosphatases: genes, structures and biological function
Acta Biochim. Pol.
2003
, vol. 
50
 (pg. 
1245
-
1256
)
[PubMed]
66
Alves-Pereira
I.
Canales
J.
Cabezas
A.
Cordero
P.M.
Costas
M.J.
Cameselle
J.C.
CDP-alcohol hydrolase, a very efficient activity of the 5′-nucleotidase/UDP-sugar hydrolase encoded by the ushA gene of Yersinia intermedia and Escherichia coli
J. Bacteriol.
2008
, vol. 
190
 (pg. 
6153
-
6161
)
[PubMed]
67
Yip
S.H.
Foo
J.L.
Schenk
G.
Gahan
L.R.
Carr
P.D.
Ollis
D.L.
Directed evolution combined with rational design increases activity of GpdQ toward a non-physiological substrate and alters the oligomeric structure of the enzyme
Protein Eng. Des. Sel.
2011
, vol. 
24
 (pg. 
861
-
872
)
[PubMed]
68
Williams
R.S.
Moncalian
G.
Williams
J.S.
Yamada
Y.
Limbo
O.
Shin
D.S.
Groocock
L.M.
Cahill
D.
Hitomi
C.
Guenther
G.
, et al. 
Mre11 dimers coordinate DNA end bridging and nuclease processing in double-strand-break repair
Cell
2008
, vol. 
135
 (pg. 
97
-
109
)
[PubMed]
69
Barford
D.
Das
A.K.
Egloff
M.P.
The structure and mechanism of protein phosphatases: insights into catalysis and regulation
Annu. Rev. Biophys. Biomol. Struct.
1998
, vol. 
27
 (pg. 
133
-
164
)
[PubMed]
70
Shi
Y.G.
Serine/threonine phosphatases: mechanism through structure
Cell
2009
, vol. 
139
 (pg. 
468
-
484
)
[PubMed]
71
Mitic
N.
Smith
S.J.
Neves
A.
Guddat
L.W.
Gahan
L.R.
Schenk
G.
The catalytic mechanisms of binuclear metallohydrolases
Chem. Rev.
2006
, vol. 
106
 (pg. 
3338
-
3363
)
[PubMed]
72
Daumann
L.J.
Gahan
L.
Schenk
G.
Larrabee
J.A.
Comba
P.
Spectroscopic and mechanistic studies of dinuclear metallohydrolases and their biomimetic complexes
J. Biol. Inorg. Chem.
2014
, vol. 
19
 pg. 
S236
 
73
Schenk
G.
Mitic
N.
Gahan
L.R.
Ollis
D.L.
McGeary
R.P.
Guddat
L.W.
Binuclear metallohydrolases: complex mechanistic strategies for a simple chemical reaction
Acc. Chem. Res.
2012
, vol. 
45
 (pg. 
1593
-
1603
)
[PubMed]
74
Hadler
K.S.
Tanifum
E.A.
Yip
S.H.
Mitic
N.
Guddat
L.W.
Jackson
C.J.
Gahan
L.R.
Nguyen
K.
Carr
P.D.
Ollis
D.L.
, et al. 
Substrate-promoted formation of a catalytically competent binuclear center and regulation of reactivity in a glycerophosphodiesterase from Enterobacter aerogenes
J. Am. Chem. Soc.
2008
, vol. 
130
 (pg. 
14129
-
14138
)
[PubMed]
75
Schenk
G.
Elliott
T.W.
Leung
E.
Carrington
L.E.
Mitic
N.
Gahan
L.R.
Guddat
L.W.
Crystal structures of a purple acid phosphatase, representing different steps of this enzyme's catalytic cycle
BMC Struct. Biol.
2008
, vol. 
8
 pg. 
6
 
[PubMed]
76
Swingle
M.R.
Honkanen
R.E.
Ciszak
E.M.
Structural basis for the catalytic activity of human serine/threonine protein phosphatase-5
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
33992
-
33999
)
[PubMed]
77
Egloff
M.P.
Cohen
P.T.
Reinemer
P.
Barford
D.
Crystal structure of the catalytic subunit of human protein phosphatase 1 and its complex with tungstate
J. Mol. Biol.
1995
, vol. 
254
 (pg. 
942
-
959
)
[PubMed]
78
Klabunde
T.
Strater
N.
Frohlich
R.
Witzel
H.
Krebs
B.
Mechanism of Fe(III)–Zn(II) purple acid phosphatase based on crystal structures
J. Mol. Biol.
1996
, vol. 
259
 (pg. 
737
-
748
)
[PubMed]
79
Knofel
T.
Strater
N.
Mechanism of hydrolysis of phosphate esters by the dimetal center of 5′-nucleotidase based on crystal structures
J. Mol. Biol.
2001
, vol. 
309
 (pg. 
239
-
254
)
[PubMed]
80
Schenk
G.
Gahan
L.R.
Carrington
L.E.
Mitic
N.
Valizadeh
M.
Hamilton
S.E.
de Jersey
J.
Guddat
L.W.
Phosphate forms an unusual tripodal complex with the Fe–Mn center of sweet potato purple acid phosphatase
Proc. Natl. Acad. Sci. U.S.A.
2005
, vol. 
102
 (pg. 
273
-
278
)
[PubMed]
81
Mitic
N.
Valizadeh
M.
Leung
E.W.
de Jersey
J.
Hamilton
S.
Hume
D.A.
Cassady
A.I.
Schenk
G.
Human tartrate-resistant acid phosphatase becomes an effective ATPase upon proteolytic activation
Arch. Biochem. Biophys.
2005
, vol. 
439
 (pg. 
154
-
164
)
[PubMed]
82
Crans
D.C.
Simone
C.M.
Holz
R.C.
Que
L.
Jr
Interaction of porcine uterine fluid purple acid phosphatase with vanadate and vanadyl cation
Biochemistry
1992
, vol. 
31
 (pg. 
11731
-
11739
)
[PubMed]
83
Vincent
J.B.
Crowder
M.W.
Averill
B.A.
Spectroscopic and kinetics studies of a high-salt-stabilized form of the purple acid phosphatase from bovine spleen
Biochemistry
1991
, vol. 
30
 (pg. 
3025
-
3034
)
[PubMed]
84
Dermol
U.
Janardan
V.
Tyagi
R.
Visweswariah
S.S.
Podobnik
M.
Unique utilization of a phosphoprotein phosphatase fold by a mammalian phosphodiesterase associated with WAGR syndrome
J. Mol. Biol.
2011
, vol. 
412
 (pg. 
481
-
494
)
[PubMed]
85
Keppetipola
N.
Shuman
S.
Distinct enzymic functional groups are required for the phosphomonoesterase and phosphodiesterase activities of Clostridium thermocellum polynucleotide kinase/phosphatase
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
19251
-
19259
)
[PubMed]
86
McMillen
L.
Beacham
I.R.
Burns
D.M.
Cobalt activation of Escherichia coli 5′-nucleotidase is due to zinc ion displacement at only one of two metal-ion-binding sites
Biochem. J.
2003
, vol. 
372
 (pg. 
625
-
630
)
[PubMed]
87
White
D.J.
Reiter
N.J.
Sikkink
R.A.
Yu
L.
Rusnak
F.
Identification of the high affinity Mn2+ binding site of bacteriophage lambda phosphoprotein phosphatase: effects of metal ligand mutations on electron paramagnetic resonance spectra and phosphatase activities
Biochemistry
2001
, vol. 
40
 (pg. 
8918
-
8929
)
[PubMed]
88
Hopfner
K.P.
Karcher
A.
Shin
D.
Fairley
C.
Tainer
J.A.
Carney
J.P.
Mre11 and Rad50 from Pyrococcus furiosus: cloning and biochemical characterization reveal an evolutionarily conserved multiprotein machine
J. Bacteriol.
2000
, vol. 
182
 (pg. 
6036
-
6041
)
[PubMed]
89
Daumann
L.J.
McCarthy
B.Y.
Hadler
K.S.
Murray
T.P.
Gahan
L.R.
Larrabee
J.A.
Ollis
D.L.
Schenk
G.
Promiscuity comes at a price: catalytic versatility vs efficiency in different metal ion derivatives of the potential bioremediator GpdQ
Biochim. Biophys. Acta
2013
, vol. 
1834
 (pg. 
425
-
432
)
[PubMed]
90
Keppetipola
N.
Shuman
S.
Mechanism of the phosphatase component of Clostridium thermocellum polynucleotide kinase-phosphatase
RNA
2006
, vol. 
12
 (pg. 
73
-
82
)
[PubMed]
91
Valdez
C.E.
Smith
Q.A.
Nechay
M.R.
Alexandrova
A.N.
Mysteries of metals in metalloenzymes
Acc. Chem. Res.
2014
, vol. 
47
 (pg. 
3110
-
3117
)
[PubMed]
92
Monteferrante
C.G.
Miethke
M.
van der Ploeg
R.
Glasner
C.
van Dijl
J.M.
Specific targeting of the metallophosphoesterase YkuE to the Bacillus cell wall requires the twin-arginine translocation system
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
29789
-
29800
)
[PubMed]
93
Paidhungat
M.
Garrett
S.
Cdc1 and the vacuole coordinately regulate Mn2+ homeostasis in the yeast Saccharomyces cerevisiae
Genetics
1998
, vol. 
148
 (pg. 
1787
-
1798
)
[PubMed]
94
Mockel
C.
Lammens
K.
Schele
A.
Hopfner
K.P.
ATP driven structural changes of the bacterial Mre11:Rad50 catalytic head complex
Nucleic Acids Res.
2012
, vol. 
40
 (pg. 
914
-
927
)
[PubMed]
95
Kim
Y.G.
Jeong
J.H.
Ha
N.C.
Kim
K.J.
Structural and functional analysis of the Lmo2642 cyclic nucleotide phosphodiesterase from Listeria monocytogenes
Proteins
2011
, vol. 
79
 (pg. 
1205
-
1214
)
[PubMed]
96
Yang
J.
Roe
S.M.
Cliff
M.J.
Williams
M.A.
Ladbury
J.E.
Cohen
P.T.
Barford
D.
Molecular basis for TPR domain-mediated regulation of protein phosphatase 5
EMBO J.
2005
, vol. 
24
 (pg. 
1
-
10
)
[PubMed]
97
Wang
L.K.
Smith
P.
Shuman
S.
Structure and mechanism of the 2′,3′ phosphatase component of the bacterial Pnkp–Hen1 RNA repair system
Nucleic Acids Res.
2013
, vol. 
41
 (pg. 
5864
-
5873
)
[PubMed]
98
Gao
Y.
Nelson
S.W.
Autoinhibition of bacteriophage T4 mre11 by its C-terminal domain
J. Biol. Chem.
2014
, vol. 
289
 (pg. 
26505
-
26513
)
[PubMed]
99
Kissinger
C.R.
Parge
H.E.
Knighton
D.R.
Lewis
C.T.
Pelletier
L.A.
Tempczyk
A.
Kalish
V.J.
Tucker
K.D.
Showalter
R.E.
Moomaw
E.W.
, et al. 
Crystal structures of human calcineurin and the human FKBP12–FK506–calcineurin complex
Nature
1995
, vol. 
378
 (pg. 
641
-
644
)
[PubMed]
100
Funhoff
E.G.
Ljusberg
J.
Wang
Y.
Andersson
G.
Averill
B.A.
Mutational analysis of the interaction between active site residues and the loop region in mammalian purple acid phosphatases
Biochemistry
2001
, vol. 
40
 (pg. 
11614
-
11622
)
[PubMed]
101
Strater
N.
Jasper
B.
Scholte
M.
Krebs
B.
Duff
A.P.
Langley
D.B.
Han
R.
Averill
B.A.
Freeman
H.C.
Guss
J.M.
Crystal structures of recombinant human purple acid phosphatase with and without an inhibitory conformation of the repression loop
J. Mol. Biol.
2005
, vol. 
351
 (pg. 
233
-
246
)
[PubMed]
102
Majka
J.
Alford
B.
Ausio
J.
Finn
R.M.
McMurray
C.T.
ATP hydrolysis by RAD50 protein switches MRE11 enzyme from endonuclease to exonuclease
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
2328
-
2341
)
[PubMed]
103
Virshup
D.M.
Shenolikar
S.
From promiscuity to precision: protein phosphatases get a makeover
Mol. Cell
2009
, vol. 
33
 (pg. 
537
-
545
)
[PubMed]
104
Lee
J.H.
Paull
T.T.
Direct activation of the ATM protein kinase by the Mre11/Rad50/Nbs1 complex
Science
2004
, vol. 
304
 (pg. 
93
-
96
)
[PubMed]
105
Lee
J.H.
Paull
T.T.
ATM activation by DNA double-strand breaks through the Mre11–Rad50–Nbs1 complex
Science
2005
, vol. 
308
 (pg. 
551
-
554
)
[PubMed]
106
Moreau
S.
Ferguson
J.R.
Symington
L.S.
The nuclease activity of Mre11 is required for meiosis but not for mating type switching, end joining, or telomere maintenance
Mol. Cell. Biol.
1999
, vol. 
19
 (pg. 
556
-
566
)
[PubMed]
107
Ritchie
K.B.
Petes
T.D.
The Mre11p/Rad50p/Xrs2p complex and the Tel1p function in a single pathway for telomere maintenance in yeast
Genetics
2000
, vol. 
155
 (pg. 
475
-
479
)
[PubMed]
108
Arthur
L.M.
Gustausson
K.
Hopfner
K.P.
Carson
C.T.
Stracker
T.H.
Karcher
A.
Felton
D.
Weitzman
M.D.
Tainer
J.
Carney
J.P.
Structural and functional analysis of Mre11-3
Nucleic Acids Res.
2004
, vol. 
32
 (pg. 
1886
-
1893
)
[PubMed]
109
Bressan
D.A.
Baxter
B.K.
Petrini
J.H.
The Mre11–Rad50–Xrs2 protein complex facilitates homologous recombination-based double-strand break repair in Saccharomyces cerevisiae
Mol. Cell. Biol.
1999
, vol. 
19
 (pg. 
7681
-
7687
)
[PubMed]
110
Kim
J.W.
Kim
H.C.
Kim
G.M.
Yang
J.M.
Boeke
J.D.
Nam
K.
Human RNA lariat debranching enzyme cDNA complements the phenotypes of Saccharomyces cerevisiae dbr1 and Schizosaccharomyces pombe dbr1 mutants
Nucleic Acids Res.
2000
, vol. 
28
 (pg. 
3666
-
3673
)
[PubMed]
111
Shin
D.H.
Proudfoot
M.
Lim
H.J.
Choi
I.K.
Yokota
H.
Yakunin
A.F.
Kim
R.
Kim
S.H.
Structural and enzymatic characterization of DR1281: a calcineurin-like phosphoesterase from Deinococcus radiodurans
Proteins
2008
, vol. 
70
 (pg. 
1000
-
1009
)
[PubMed]
112
Neu
H.C.
The 5′-nucleotidase of Escherichia coli. II. Surface localization and purification of the Escherichia coli 5′-nucleotidase inhibitor
J. Biol. Chem.
1967
, vol. 
242
 (pg. 
3905
-
3911
)
[PubMed]
113
Neu
H.C.
The 5′-nucleotidase of Escherichia coli. I. Purification and properties
J. Biol. Chem.
1967
, vol. 
242
 (pg. 
3896
-
3904
)
[PubMed]