HAD hydrolase function unveiled by substrate screening: enzymatic characterization of Arabidopsis thaliana subclass I phosphosugar phosphatase AtSgpp

This work presents the isolation and the biochemical characterization of the Arabidopsis thaliana gene AtSgpp. This gene shows homology with the Arabidopsis low molecular weight phosphatases AtGpp1 and AtGpp2 and the yeast counterpart GPP1 and GPP2, which have a high specificity for dl-glycerol-3-phosphate. In addition, it exhibits homology with DOG1 and DOG2 that dephosphorylate 2-deoxy-d-glucose-6-phosphate. Using a comparative genomic approach, we identified the AtSgpp gene as a conceptual translated haloacid dehalogenase-like hydrolase HAD protein. AtSgpp (locus tag At2g38740), encodes a protein with a predicted Mw of 26.7 kDa and a pI of 4.6. Its sequence motifs and expected structure revealed that AtSgpp belongs to the HAD hydrolases subfamily I, with the C1-type cap domain. In the presence of Mg2+ ions, the enzyme has a phosphatase activity over a wide range of phosphosugars substrates (pH optima at 7.0 and Km in the range of 3.6–7.7 mM). AtSgpp promiscuity is preferentially detectable on d-ribose-5-phosphate, 2-deoxy-d-ribose-5-phosphate, 2-deoxy-d-glucose-6-phosphate, d-mannose-6-phosphate, d-fructose-1-phosphate, d-glucose-6-phosphate, dl-glycerol-3-phosphate, and d-fructose-6-phosphate, as substrates. AtSgpp is ubiquitously expressed throughout development in most plant organs, mainly in sepal and guard cell. Interestingly, expression is affected by abiotic and biotic stresses, being the greatest under Pi starvation and cyclopentenone oxylipins induction. Based on both, substrate lax specificity and gene expression, the physiological function of AtSgpp in housekeeping detoxification, modulation of sugar-phosphate balance and Pi homeostasis, is provisionally assigned.


Introduction
The haloacid dehalogenase-like hydrolase (HAD) superfamily is a large group of proteins with diverse substrate specificity whose members, despite the family name, are involved not only in the enzymatic cleavage by nucleophilic substitution of carbon-halogen bonds (C-halogen), but also in a variety of hydrolytic enzyme activities including phosphatase (CO-P), phosphonatase (C-P) and phosphoglucomutase (CO-P hydrolysis and intramolecular phosphoryl transfer) reactions (Koonin and Tatusov 1994;Allen and Dunaway-Mariano 2004).
Within the HAD superfamily, the wide family of magnesium-dependent acid phosphatases and phosphomutases is characterized by an amino-terminal conserved Asp as the nucleophile (Collet et al. 1998;Selengut 2001). All members share the a/b core domain catalytic scaffold, with the active site formed by four loops containing highly conserved sequence motifs (loops 1, DxD; 2, T/S; 3, K/R; 4, E/DD, GDxxxD, or GDxxxxD), spatially arranged around a single binding cleft, that position substrate-cofactor binding and catalytic residues that are involved in the core chemistry (Burroughs et al. 2006). Many of them also possess a smaller cap domain, linked to the core by two hinge-like solvated peptide linkers, which acts as a dynamic rigid lid over the core active site. Although its primary function might be active-site desolvation, the cap domain contains the helix-loop-helix (loop 5) with a stringently conserved Gly flanked by residues whose side chains contribute to the catalytic site formed at the domain-domain interface. These residues are responsible for the chemistry diversification within the family and provide substrate specificity Lahiri et al. 2004). Thus, the elements involved in substrate-recognition and chemical-catalysis are located separately; substrate recognition is delegated to cap residues, whereas phosphoryl transfer is mediated by residues located deep inside the core cleft (Wang et al. 2002;Lu et al. 2005Lu et al. , 2008. Based on the presence, topology and location of the cap domain, the HAD superfamily is divided into three subfamilies (I, IIA-IIB and III) (Morais et al. 2000;Zhang et al. 2002;Tremblay et al. 2006). In subfamily I, the small a-helical-bundle cap is inserted between loops 1 and 2 in the core domain, whereas in subfamily II the larger b-sandwich is placed between loops 2 and 3; the third group contains no insertion (Selengut and Levine 2000;Shin et al. 2003). Unlike the core catalytic domain, the cap has undergone extensive evolutionary diversity in substrate exploration during HADs evolution (Burroughs et al. 2006). While the a/b Rossmann core active sites are superimposable, the architecture of the cap domain differs even between the closest structural homologs (Rao et al. 2006).
In HAD phosphatases, after the binding of the substrate the active site closes and Mg 2? cofactor interacts with the negatively charged phosphate, preparing it for nucleophilic attack by the first conserved Asp, forming an intermediate acyl-phosphate with the carboxyl group. Then, the enzyme opens again allowing the escape of the leaving group and the entry of solvent; water is deprotonated by the second Asp, hydrolyzing the acyl-phosphate intermediate and returning the enzyme to the native status (Burroughs et al. 2006). In phosphohydrolases, the Asp nucleophile is located on loop 1, loop 2 positions a conserved Ser/Thr that binds the substrate phosphoryl group, whereas loop 3 positions a conserved Arg/Lys that orients and shields charge in the Asp nucleophile and the phosphoryl group and, finally, loop 4 positions two or three Asp residues that bind the Mg 2? cofactor (Lu et al. 2005).
Due to their sequence divergence, only the highly conserved catalytic motifs and the similar folds and active-site structures allow identification between members of the HADs (Morais et al. 2000). However, the catalyzed reaction and substrate specificity are difficult to predict and have to be determined empirically (Kuznetsova et al. 2006).
In an earlier work, we cloned and characterized two Arabidopsis thaliana isoforms AtGpp1 (At4g25840) and AtGpp2 (At5g57440) of the DL-glycerol-3-phosphatase, involved in plant glycerol metabolism (Caparrós-Martín et al. 2007). The analysis of the sequence indicates that AtGpp1 and AtGpp2 phosphatase are members of the HAD haloacid dehalogenase hydrolase superfamily. AtGpp1 and AtGpp2 show high homology with the yeast phosphatases GPP1 and GPP2 (Norbeck et al. 1996), which have a high specificity for DL-glycerol-3-phosphate, as well as with DOG1 and DOG2 that dephosphorylate 2-deoxy-D-glucose-6-phosphate (Rández-Gil et al. 1995).
The Arabidopsis genome contains homologous loci, other than AtGpp1 and AtGpp2, with similar scores and general function predicted as phosphatase/phosphohexomutase (unknown At2g38740, putative At4g39970, catalytic/ hydrolase At3g48420 and catalytic/hydrolase/phosphoglycolate phosphatase At2g33255). Attention was focused on these HAD superfamily-encoded loci, sharing significant similarity, presuming a related sequence-based assignment of activity on targeting substrates with similar structural characteristics, in Arabidopsis.
Following a screening approach, the four purified proteins were tested for phosphatase activity against a set of sugar phosphoesters. Pi release was only detected in assays with the product of locus At2g38740 so-called AtSgpp. Structural prediction and the chemistry analysis reveal AtSgpp as a typical phosphomonoesterase of subclass I (C1-type cap). At odds with other subfamily I representatives such as phosphonatase, phosphoserine phosphatase, 2,3-diketo-Lphospho-5-thiomethylpentane phosphatase, 2-deoxy-D-glucose-6-phosphate phosphatase, and glycerolphosphate phosphatase, rather than being substrate specific, AtSgpp shows a broad-range sugar phosphate phosphatase activity. Curiously enough, similar specificity has been ascribed to the Bacteroides thetaiotaomicron BT4131 gene (Lu et al. 2005). BT4131 enzyme is member of the haloalkanoate dehalogenase superfamily (subfamily IIB, C2B-type cap), with a tentatively assigned physiological function in chitin metabolism. The expected structure, specificity and kinetics of plant AtSgpp chemistry were analyzed and compared with those of bacterial BT4131.

Plant material, growth conditions and stress treatments
Arabidopsis thaliana ecotype Columbia (Lehle Seeds, Round Rock, TX, USA) was grown in the greenhouse at 25°C for 8 h in the dark and 16 h in light. For seedling stress assays, wild-type (WT) and transgenic surface-sterilized Arabidopsis seeds were sown in Petri dishes containing 3 ml MSS medium [MS (Murashige and Skoog 1962 Genome, sequence and structural analysis Comparative genomics were performed using programs such as BLAST (Altschul et al. 1990) and data bank resources from the NCBI. Protein domain families were generated with the ProDom program from the Swiss-Pro and TrEMBL sequence databases (Corpet et al. 2000). CLUSTAL W was used for the progressive multiple sequence alignment (Higgins et al. 1994). GENEVESTI-GATOR Arabidopsis microarray database was utilized for the expression analysis (Zimmermann et al. 2004). 3D models of the protein have been built using the ESyPred3D web server (Lambert et al. 2002).
cDNA cloning Using peptide sequence motifs shared between the yeast DL-glycerol-3-phosphatases (Norbeck et al. 1996), virtual clones were isolated by BLAST (Altschul et al. 1990) search screening from the predicted conceptual translated proteins of the A. thaliana genomic library. The corresponding homologous genes were cloned by reverse PCR using leaf total RNA as template for the first strand cDNA synthesis together with an oligo(dT) primer and avian myeloblastosis virus reverse transcriptase. The first strand cDNA template was PCR amplified using REDTaq DNA polymerase and the 5 0 -forward and 3 0 -reverse gene-specific adapted primers 5 0 -CGAGGAATTCATGAATGGCTTCT CTGATCTTAATCC-3 0 /5 0 -CCGGGTCGACTTAAGACT TGTTATCAAGTTCTTCC-3 0 ; 5 0 -CGAGGAATTCATGG CGGTTTCTTGCAACCACTCTGC-3 0 /5 0 CCGGGTCGAC TTAAGCTGCAGTGACTATTGTTTGAAGC-3 0 ; 5 0 -CGA GGAATTCATGGCCACTGTGAAAATCTCTCTTTCC-3 0 /5 0 -CCGGGTCGACTTAACTAACGAACTGTTTCCGG AG-3 0 ; and 5 0 -CGAGGAATTCATGGCGAATTTAACG ACGAACGC-3 0 /5 0 -CCGGGTCGACCTACGGGTTCAGG TCGAAGTTCG-3 0 for At2g38740, At4g39970, At3g484 20 and At2g33255, respectively. The PCR products were cloned as 735, 956, 960 or 675 bp EcoRI/SalI fragment into the pBluescript SK ? vector. Sequencing of the pBluescript SK ? clones revealed that the sequence of the proteins is the same as that published in the Gen-Bank TM . The cDNAs, containing the entire gene coding region, were subcloned as EcoRI/SalI fragments into the pMAL-c2X expression vector and transformed into the expression strain E. coli TB1 for recombinant protein production. The cloning site used in the pMAL-c2X polylinker (located downstream of the malE gene), adding vector-encoded residues Ile-Ser-Glu-Phe fused between the factor Xa cleavage site and the NH 2 -terminal methionine residue of the cloned proteins. To improve the expression of the eukaryotic genes in the E. coli system, E. coli TB1 cells were co-transformed with pMAL-c2X At2g38740, pMAL-c2X At4g39970, pMAL-c2X At3g48420 or pMAL-c2X At2g33255 and, in each case, the helper plasmid pSBETa. The positive co-transformed colonies were selected on 200 lg/ml ampicillin (Sigma) and 100 lg/ml kanamycin (Sigma).

Northern blots and hybridization
Northern analyses were performed using approximately 20 mg of total RNA per track. Isolated DNA fragments were nick-translated in the presence of a-[ 32 P]dCTP to be used as probes (Maniatis et al. 1982). Probe AtSgpp cDNA was a 735-bp EcoRI/SalI fragment, which contains the complete AtSgpp coding region. Hybridization was performed in 39 SSC (saline sodium citrate), 0.05 % PVP, 0.05 Ficoll, 1 % SDS and 50 lg ml -1 ssDNA (salmon sperm DNA) (Sigma) at 55°C. Filters were washed at high stringency (0.19 SSC and 0.5 % SDS) at 55°C. The experiments were performed more than once and the data shown are representative.

Isolation of A. thaliana AtGpp-homologous genes
In previous work, the complete sequence of the Arabidopsis genome (The Arabidopsis Genome Initiative 2000) was used for the identification of the two uncharacterised A. thalianaDL-glycerol-3-phosphatase genes. With the known budding yeast GPP1 and GPP2 protein sequences and using the BLAST program (Altschul et al. 1990) two putative A. thaliana, named AtGpp1 and AtGpp2, were identified, cloned and characterized (Caparrós-Martín et al. 2007). Using a comparative genome approach, four additional homologous genes were detected, with similar scores and general function predicted as phosphatase/phosphohexomutase (unknown At2g38740, putative At4g39970, catalytic/hydrolase At3g48420 and catalytic/hydrolase/ phosphoglycolate phosphatase At2g33255). The virtual isolated genes were cloned by RT-PCR and sequenced.

Sequence comparisons
The four plant proteins show significant similarity (28-39 %, sequence identity of 18-23 %) to A. thalianaDLglycerol-3-phosphatases AtGpp1 and AtGpp2 (Fig. 1). This homology is ubiquitously distributed although, as it would be expected among HAD members, the greatest similarity occurs at the four loops forming the catalytic scaffold of the active site platform framed by the core domain Lu et al. 2005), particularly in the highly conserved sequence motifs by which family members are recognized, contained in loops 1-4 (1, FDxDG; 2, xT/Sx; 3, KPxP; 4, ED, or GDxxxDD), that positions substrate-cofactor binding and catalytic residues that are involved in the core chemistry (Burroughs et al. 2006). Fewer sequence homologies are shared at the predicted cap domain substrate recognition loop 5, responsible for the chemical diversification within the family, apart from the stringently conserved loop marker Gly (G), flanked by Lys/Arg (K/R) and residues, whose side chains contribute to the catalytic site, probably operating in domain-domain binding, active-site desolvation and/or catalysis (Lahiri et al. 2004), that must provide the signature-based substrate specificity.

Protein expression, purification and substrate screening
To examine the phosphatase activity, the AtGpp homologous At2g38740, At4g39970, At3g48420 and At2g33255 (26.7, 34.7, 34.3 and 27.5 kDa, respectively) were expressed as MBP (42.7 kDa) fusion proteins, purified from the corresponding E. coli clones, by amylose affinity chromatography, and their phosphatase activity determined in the presence of Mg 2? ions (Fig. 2).

AtSgpp gene expression
To gain insight into the physiological importance of the AtSgpp hydrolytic activity in plants, the expression of the A. thalianaAtSgpp gene was analyzed, by Northern hybridization, in different tissues and under stress conditions (Fig. 5). In all blot experiments a band was detected corresponding in size to the AtSgpp transcripts. The AtSgpp pattern of expression was examined in roots, shoots, leaves, flowers and developing siliqua (Fig. 5a). Transcripts of the gene were detectable on northern blots from all of these organs. High expression was found in flowers, pointing to a prominent role of this gene during floral development.
AtSgpp mRNA accumulation was also observed in shoots, siliqua, roots and the lowest in leaves. AtSgpp expression was also examined in seedling subjected to salt (100 mM NaCl and 15 mM LiCl), osmotic (200 mM sorbitol) or oxidative (5 mM H 2 O 2 and 1 lM methyl viologen) stresses (Fig. 5b). AtSgpp expression seems to be affected in seedlings that undergo all these abiotic stresses.
Signature sequence substrate specificity loop 5 While the catalytic core is superimposable, the cap domain has evolved through the evolution, remaining responsible of targeting diversification among members of the subfamily. The cap contains the so-called substrate specificity loop 5 that provides the signature sequence motif 5 which, in the closed conformation of the enzyme, defines the active site and its specific chemistry. Core sequence motifs 1 and 2 determine the boundaries of the cap sequence segment in subfamily I, and the conserved Gly identifies the loop 5 motif within it, allowing detection of cap domain-derived active site residues, in the absence of a three-dimensional structure (Lahiri et al. 2004). Using this effective tool for function assignment together with their predicted structure, the substrate specificity loop 5 from known HAD phosphohydrolases GPP (Norbeck et al. 1996), DOG (Rández-Gil et al. 1995), AtGpp (Caparrós-Martín et al. 2007), AtSgpp and unknown homologous representatives, were compared (Fig. 6). In baker's yeast subclass I DL-glycerol-3-phosphatases GPP1 and GPP2 and 2-deoxy-D-glucose-6-phosphatases DOG1 and DOG2, their putative shared motif 5 (SHGW/ AR) presents a single shift of Trp (W56 in GPP1 and GPP2) per Ala (A49 in DOG1 and DOG2) which could impart substrate specificity, while between yeast GPP1 and GPP2 and its plant counterparts AtGpp1 and AtGpp2, just the central Gly (G55 in GPP1 and GPP2 and G57 in AtGpp2), essential in determining the loop 5 conformation and flexibility (Lahiri et al. 2004), and surrounding Lys (R57 in GPP1 and GPP2 and R58 in AtGpp2) are conserved (Fig. 6a). By contrast, the motif 5 is well preserved between genuses within the kingdom itself (Fig. 6b, c). Lack of shared homology is also appreciated among phosphosugar phosphatases, subfamily IIB bacterial BT4131 and subclass I plant AtSgpp, despite their analogous catalysis and substrate specificity (not shown).

Discussion
The objective of this study was to search for the sequencefunction relationship, which drives the substrate discrimination between HAD phosphohydrolases subfamily members. Application of general enzymatic screenings and substrate profiling is a useful tool to discover new enzymes (Kuznetsova et al. 2005); likewise, the substrate prediction of unknown HAD hydrolases and their confirmed function, by applying the predicted substrates to the purified protein, has been an elegant strategy that has been successfully a c b d Fig. 5 Expression of AtSgpp mRNA during development and under stress treatment. a, b Northern blot of Arabidopsis RNA probed with radiolabelled AtSgpp cDNA. c, d AtSgpp pattern of expression from the microarrays data bank Genevestigator. a From different tissues: 1 root, 2 shoot, 3 leaves, 4 flowers, 5 siliqua. b From 12-day-old seedling cultivated on MSS medium: 1 control, 2 MSS ? 100 mM NaCl, 3 MSS ? 200 mM sorbitol, 4 MSS ? 15 mM LiCl, 5 MSS ? 5 mM H 2 O 2 and 6 MSS ? 1 lM methyl viologen. In Northern blots, ethidium bromide-stained RNA was used as loading control. c Expression profile over time throughout the live cycle of the Arabidopsis. d Partial hierarchical clustering displaying the level of the anatomical expression across of tissue types carried out (Lu et al. 2005). Following sequence-function analysis, as those used in the characterization of Arabid-opsisDL-glycerol-3-phosphatases AtGpp1 and AtGpp2 (Caparrós-Martín et al. 2007), the physiological substrate specificity of the closest AtGpp homologues At2g38740, At4g39970, At3g48420, and At2g33255 was analyzed. Rather than for their genomic context, loci were chosen for the remarkable shared similarity of the encoding genes. In the spirit of the initial strategy it was to contribute to the hypothesis that might be possible to infer guidelines for function assignment within the HAD family, based on the sequence data alone (Lahiri et al. 2004); abounding into the syllogism that, if the cap domain defines substrate specificity and the physiological substrate determines the biochemical function, logically the relationship between cap domain and substrate structure is the essential clue to the sequence-based assignment of its function (Tremblay et al. 2006).
The AtSgpp catalytic hydrolysis of cyclic sugars occurs with inconspicuous specificity and efficiency, k cat /K m in the range of 2.5-10.7 9 10 3 M -1 s -1 . Such a modest catalysis and substrate promiscuity has been reported by the hexose phosphate phosphatase HPP (BT4131 from Bacteroides thetaiotaomicron VPI-5482), whose catalytic efficiency is low: k cat /K m *1 9 10 3 M -1 s -1 compared to the 1 9 10 6 to 1 9 10 8 M -1 s -1 ''gold-standard'' for single-substrate enzymes involved in primary metabolism (Lu et al. 2005). Bacterial BT4131 is a member of the haloalkanoate dehalogenase superfamily (subfamily IIB) with a C2B-type cap domain, whilst the expected structure of the plant enzyme yet enrolls AtSgpp as a member of the subfamily I with the C1-type cap domain. Besides the lax specificity for the ring size and stereochemistry of the sugar phosphate, AtSgpp shares with BT4131 the ability to discriminate between isomers that do not accommodate into the active site solvent cage, such as D-glucose-1-phosphate or a-D-mannose-1-phosphate and, to a lesser extent, D-fructose-6-phosphate. As suggested for other HAD monophosphatases (Kuznetsova et al. 2006;Tremblay et al. 2006), the kinetic parameters of AtSgpp demonstrated an overlapping catabolic activity, with undefined boundaries on physiologically related substrates that, in this case, point to a role in the metabolic regulation of phosphosugars intermediates of the glycolysis and the pentose phosphate pathway. The activity of AtSgpp on 2-deoxy-D-glucose-6phosphate also arouses interest. 2-deoxy-D-glucose is a a b c Fig. 6 Comparison of the signature sequence motif 5. a The amino acid sequence of A. thalianaDL-glycerol-3-phosphatases AtGpp1 and AtGpp2 was compared with homologous representatives from S. cerevisiaeDL-glycerol-3-phosphatases GPP1 and GPP2 and 2-deoxy-D-glucose-6-phosphatases DOG1 and DOG2. b, cArabidop-sisDL-glycerol-3-phosphatase AtGpp2 and phosphosugar phosphatase AtSgpp were also compared with orthologous candidates from other genus by NCBI BLink: Vitis (CBI17840), Oryza (BAD05444), Sorghum (EES14772), Zea (ACF85466), Chlamydomonas (EDP00037), and Homo (AAH12494) (b) and Vitis (CBI33932), Oryza (BAD36300), Sorghum (EER89058), Zea (ACF83536), Chlamydomonas (EDP04776), and Pseudomonas (EGH95925) (c), respectively. Conserved residues are labeled with asterisks or dots and the putative shared motifs in bold. In the absence of a three-dimensional structure, cap domain-derived active site residues were identified by the conjoined information from predicted 3D models, the boundaries of the cap sequence segment defined by motifs 1 and 2, and the conserved Gly (G) that identify the loop 5 motif within this segment non-metabolizable analog of glucose that becomes toxic after its phosphorylation to 2-deoxy-D-glucose-6-phosphate. So far, no plant analogous to the yeast DOG1 and DOG2 genes (Rández-Gil et al. 1995), which dephosphorylate 2-deoxyglucose-6-phosphate, has been found. However, DOG2 transgenic over-expression in tobacco improves the tolerance to the deleterious effects of 2-deoxy-D-glucose on growth, chlorophyll content and the expression of genes related to photosynthesis (Cutanda 2003). In vitro phosphatase activity against 2-deoxyglucose-6-phosphate has also been observed in E. coli YniC (HAD1) protein; E. coli YniC-overproducing strain grew well in the presence of 2-deoxy-D-glucose, demonstrating that the 2-deoxy-D-glucose-6-phosphatase activity of YniC (HAD1), plays an important role for the resistance of E. coli cells to 2-deoxyglucose (Kuznetsova et al. 2006). Substrate binding regulates, by the hinge motion of the solvated domain linkers, the conformational cap closed/ open interconversion. In the cap-closed conformation, cap and core domain interfaced and residues from the cap substrate specificity loop enter the active site, participating in substrate binding and catalysis, whereas the cap-open conformation allows the access to the active site of the solvent, facilitating the release of the product . Similarly, as previously reported for subclass I phosphonatase catalysis (Zhang et al. 2002;Morais et al. 2004;Lahiri et al. 2006), one could also suggest that the substrate specificity loop of AtSgpp contributes the catalytic Lys (K71) residue to form a Schiff base with the substrate. Proton dissociation from the K71 in the cap domain versus protonation of His (H72) would be required for the AtSgpp cap-core domains closure. Together with the ionization of several enzyme and substrate groups, the loss of activity at low pH could be due to the protonation of H72 and at high pH to deprotonation of K71. Thus, extreme pH would modify the intramolecular protontransfer, introducing a change in the electron density of the system and on the stability of the phosphoaspartate intermediate compound. This being the case, the pH dependency of the AtSgpp chemistry on D-ribose-5-phosphate/Mg 2? would reflect a narrower chemoselectivity in the active site than those conformed by the remaining ligands tested.
Similar pattern of AtSgpp gene expression, in all Arabidopsis organs, have been shown for DL-glycerol-3-phosphatase AtGpp (Caparrós-Martín et al. 2007), with transcripts particularly abundant in developing siliqua. Particularly interesting is the induced expression under nitrogen and iron deficiency and the abscisic acid-induced and reactive oxygen species-dependent expression in guard cell (Böhmer and Schroeder 2011). Foremost, the greatest expression is observed under Pi starvation and cyclopentenone oxylipins induction. Oxylipins induce the expression of genes related to detoxification, stress responses, and secondary metabolism, when accumulated in response to stress stimuli such as wounding and pathogen infection, what has been attributed to an increase in ROS (Mueller et al. 2008). Coincidentally, a housekeeping function has also been assigned to the hydrolyzing activity of AtCoase, a pyrophosphatase from Arabidopsis thaliana which cleaves coenzyme-A to 4 0 -phosphopantetheine and 3 0 ,5 0 -adenosinediphosphate; the CoA cleaving enzyme, whose ubiquitous expression improves plant development, is a member of the Nudix hydrolases, pyrophosphatases that hydrolyze nucleoside diphosphates (Kupke et al. 2009).
Since the referred signature sequence motif 5 drives the substrate targeting, and consequently may be used as protein tagging, from the sequence motif comparisons, it would be worth considering that, within the same species a unique point mutation Trp/Ala (W/A) may be sufficient to discriminate between structural related substrates, pivoting the stereospecificity for DL-glycerol-3-phosphate or 2-deoxy-D-glucose-6-phosphate. In contrast, orthologous enzymes as DL-glycerol-3-phosphatases, from different kingdoms, differ in their patterns. Most intriguing, in the evolutionary divergence within the HAD family, is the ability of analogous Bacteria-BT4131 and Plantae-AtSgpp to perform similar function with dissimilar topology and location of the cap domain. Considering that, although both the I and II subfamilies act predominantly on small encapsulated substrates , in subfamily I the small a-helical-bundle C1-type cap is inserted between loops 1 and 2 of the core domain, whereas in subfamily IIB the larger b-sandwich C2B-type is between loops 2 and 3 (Selengut and Levine 2000;Shin et al. 2003); moreover, BT4131 uses two substrate specificity loops in substrate recognition (Lu et al. 2005).
In conclusion, this paper reports the enzymatic characterization of A. thaliana locus At2g38740. Rescued from the HAD unknown members and hereafter named as phosphosugar phosphatase AtSgpp, its expected structure lists the enzyme as being a member of the HAD subfamily I C1-type cap domain. Extensive substrate screening reveals that AtSgpp presents substrate promiscuity, with broad-range sugar phosphate phosphatase activity, preferentially detectable on D-ribose-5-phosphate and also 2-deoxy-D-ribose-5-phosphate, 2-deoxy-D-glucose-6-phosphate, D-mannose-6-phosphate, D-fructose-1-phosphate, D-glucose-6-phosphate, DL-glycerol-3-phosphate, and D-fructose-6-phosphate. The kinetic parameters show a humble specificity and efficiency that resemble those of the bacterial BT4131, which is a member of the HAD subfamily IIB with a C2B-type cap domain. BT4131 biochemical function was identified by the conjoined information from residues stationed on substrate specificity loops, the active site solvent cage and the genome context of the encoding gene (Lu et al. 2005) while for AtSgpp, instead of the genomic context, was the close homology with ArabidopsisDL-glycerol-3-phosphatase AtGpp who designed the sugar phosphoesters screening. AtSgpp phosphatase activity is optimal at pH 7.0, albeit fairly pH independent on D-ribose-5-phosphate cleaving. At present, it is not known if the ribose ester is the main natural hydrolyzing substrate of AtSgpp, or if it could be another non-screened candidate. Also, if using different assay conditions, the Arabidopsis homologous At4g39970, At3g48420 and At2g33255 will cleave sugar phosphomonoesters, or if they hydrolyze different substrates. AtSgpp is ubiquitously expressed throughout development and its expression is affected by abiotic and biotic stresses, being the greatest under Pi starvation and cyclopentenone oxylipins induction. Therefore, taking into account both the activity and the expression data, the AtSgpp physiological function appears to be somehow related to housekeeping detoxification, sugar-phosphate modulation and in maintaining the homeostatic balance of Pi in the cell. Another aim of this work was to provide further information as to the relationship that binds the function to the substrate specificity loop, to contribute to answer the formerly enunciated question about how then do HAD phosphohydrolases distinguish their substrates? It could be claimed that functional peers can be thereby identified by their cap domain signature-sequence motif, though this function assignment would remain constrained to a single kingdom.