|Home | About | Journals | Submit | Contact Us | Français|
Soluble inorganic pyrophosphatases (PPAs) that hydrolyze inorganic pyrophosphate (PPi) to orthophosphate (Pi) are commonly used to accelerate and detect biosynthetic reactions that generate PPi as a by-product. Current PPAs are inactivated by high salt concentrations and organic solvents, which limits the extent of their use. Here we report a class A type PPA of the haloarchaeon Haloferax volcanii (HvPPA) that is thermostable and displays robust PPi-hydrolyzing activity under conditions of 25% (vol/vol) organic solvent and salt concentrations from 25 mM to 3 M. HvPPA was purified to homogeneity as a homohexamer by a rapid two-step method and was found to display non-Michaelis-Menten kinetics with a Vmax of 465 U · mg−1 for PPi hydrolysis (optimal at 42°C and pH 8.5) and Hill coefficients that indicated cooperative binding to PPi and Mg2+. Similarly to other class A type PPAs, HvPPA was inhibited by sodium fluoride; however, hierarchical clustering and three-dimensional (3D) homology modeling revealed HvPPA to be distinct in structure from characterized PPAs. In particular, HvPPA was highly negative in surface charge, which explained its extreme resistance to organic solvents. To demonstrate that HvPPA could drive thermodynamically unfavorable reactions to completion under conditions of reduced water activity, a novel coupled assay was developed; HvPPA hydrolyzed the PPi by-product generated in 2 M NaCl by UbaA (a “salt-loving” noncanonical E1 enzyme that adenylates ubiquitin-like proteins in the presence of ATP). Overall, we demonstrate HvPPA to be useful for hydrolyzing PPi under conditions of reduced water activity that are a hurdle to current PPA-based technologies.
Inorganic pyrophosphatases (PPAs) (EC 18.104.22.168) catalyze the hydrolysis of the phosphoanhydride bond of inorganic pyrophosphate (PPi) (P2O74−) to form 2 mol of orthophosphate (Pi) (PO43−) (1). PPi is a common by-product of metabolism, including the biosynthesis of DNA, RNA, protein, peptidoglycan, lipids (e.g., cholesterol), cellulose, starch, and other biopolymers (2). PPi is also formed during the posttranslational modification of proteins, including adenylation, uridylation, and ubiquitylation (2).
The hydrolysis of PPi by PPA releases a considerable amount of energy (ΔG′° = −19.2 kJ/mol) that can drive unfavorable biochemical transformations to completion. One example is in the synthesis of DNA by DNA polymerase. In this endergonic (ΔG′° = +2.1 kJ/mol) reaction, the 3′-hydroxyl group of the nucleotide that resides at the 3′ end of the growing DNA strand serves as a nucleophile in the attack of the α phosphorus of the incoming deoxynucleoside 5′-triphosphate (dNTP), thus releasing PPi (2). The polymerization of DNA is highly dependent on PPA to hydrolyze the energy-rich PPi to 2Pi and to drive the synthesis reaction forward (2). Under standard conditions, DNA polymerase alone converts DNA to dNTPs.
PPAs are used in a wide variety of biotechnology applications based on the ability of these enzymes to drive reactions forward and generate an easily assayed product. PPAs prevent the accumulation of PPi during DNA sequencing reactions (3, 4), PCR (5), and single-base extension (SBE) reactions (6). PPAs are also used to increase the yields of RNA by in vitro transcription (7) and of GDP-modified sugars by enzymatic synthesis (8, 9). PPAs are used routinely to quantify rates of reactions that release PPi as a by-product, such as single nucleotide polymorphism (SNP) genotyping reactions (10,–12), RNA synthesis by viral RNA-dependent RNA polymerases (13), and aminoacyl-tRNA synthetase activity (14, 15). The advantage of PPA-coupled assays is that PPA hydrolyzes PPi to a product (2Pi) which is readily detected by colorimetric assay (16, 17).
PPAs that operate in a wide variety of organic solvents and salt concentrations are desirable in bioindustry to increase the solubility of hydrophobic substrates, allow for novel synthetic chemistry, alter substrate specificity, ease product recovery, and reduce microbial contamination (18,–24). Biotechnological applications of solvent-tolerant PPAs can be envisioned in the biosynthesis of hydrophobic compounds derived from carbon skeletons such as cholesterol and rubber. In cholesterol biosynthesis, PPi is released in several of the early and foundational steps leading to the production of steroid hormones and vitamin D (25). Likewise, natural rubber is synthesized from allylic diphosphate and polymers of isopentenyl diphosphate (IPP), releasing one PPi per IPP incorporated by this rate-limiting reaction (26, 27). Solvent-tolerant PPAs could be used to drive forward the initiation and elongation reactions of rubber and cholesterol-derived compounds by performing the reactions in extracting reagents that enhance reactant and product solubility. In addition to increasing natural product yield, the hydrophobic products generated by this type of approach could be retrieved by precipitation under aqueous conditions while allowing for reuse of any enzymes that are solvent tolerant.
Here we propose and demonstrate that new archaeal PPAs of the class A type (IPR008162 family) hold the answer to the solvent and salt tolerance dilemma. PPAs of the class A type are not associated with the membrane and are widespread in all domains of life, including Archaea. Class A type PPAs of hyperthermophilic archaea are already used in PCR and DNA sequencing reactions (4, 5). In contrast, PPAs with DHHA2 domains (IPR004097) and PPAs of the PPi-energized H+ pump (IPR004131), Mn2+-dependent (IPR022934), and PpaX-type (IPR023733) families are not as widely distributed in archaea and/or can be difficult to purify due to their association with membranes.
Biochemicals were from Sigma-Aldrich (St. Louis, MO). Other organic and inorganic analytical-grade chemicals were from Fisher Scientific (Atlanta, GA). Phusion and Taq DNA polymerases, restriction endonucleases, and T4 DNA ligase were from New England BioLabs (Ipswich, MA). Desalted oligonucleotides were from Integrated DNA Technologies (Coralville, IA). Agarose used for routine analysis of DNA was from Bio-Rad Laboratories (Hercules, CA).
The strains used in this study are listed in Table 1. Escherichia coli TOP10 was used for routine recombinant DNA analysis. E. coli GM2163 was used for preparation of plasmid DNA prior to transformation of Haloferax volcanii H26 by standard methods (28). E. coli strains were grown at 37°C in Luria-Bertani (LB) medium supplemented with ampicillin (Amp) (0.1 mg · ml−1) as needed. H. volcanii strains were grown at 42°C in ATCC 974 medium supplemented with novobiocin (Nv) (0.1 μg · ml−1) as needed. Cells were grown in liquid cultures with rotary shaking at 200 rpm and on solid medium (1.5% [wt/vol] agar plates). Growth was monitored by measuring the optical density at 600 nm (OD600), where 1 OD600 unit equals approximately 1 × 109 CFU · ml−1.
The plasmids and primers used in this study are listed in Table 1. Primers 1 and 2 were used for PCR-based amplification of the gene encoding PPA of H. volcanii (HvPPA) (HVO_0729; UniProt: D4GT97) with strain H26 genomic DNA as the template. The 0.55-kb PCR product was ligated into the NdeI to BlpI sites of pJAM503 to generate plasmid pJAM2920 for expression of HvPPA with an N-terminal polyhistidine tag (His6-HvPPA) linked with a thrombin cleavage site. Plasmid DNA was isolated with a QIAprep spin miniprep kit (Qiagen, Valencia, CA). PCRs were according to standard methods using an iCycler (Bio-Rad Laboratories). Genomic DNA was extracted from H. volcanii cells by boiling colonies resuspended in double-distilled water (ddH2O) (29) or by DNA spooling (28). Phusion DNA polymerase was used for high-fidelity PCR-based cloning. Taq DNA polymerase was used for colony screening. DNA fragments were separated by 0.8 to 2% (wt/vol) agarose gel electrophoresis (90 V, 30 to 45 min) in TAE buffer (40 mM Tris, 20 mM acetic acid, 1 mM EDTA, pH 8.0). Gels were stained with ethidium bromide at 0.25 μg · ml−1 and visualized with a Mini visionary imaging system (Fotodyne, Hartland, WI). Hi-Lo DNA molecular weight markers (Minnesota Molecular, Minneapolis, MN) were used for comparison. DNA fragments were isolated directly from PCR by MinElute PCR purification (Qiagen) or from 0.8% (wt/vol) SeaKem GTG agarose (FMC Bioproducts, Rockland, ME) gels in TAE buffer at pH 8.0 using the QIAquick gel extraction kit (Qiagen) as needed. The fidelity of DNA plasmid constructs was verified by Sanger DNA sequencing (UF ICBR DNA sequencing core).
HvPPA was expressed with an N-terminal polyhistidine (His6) tag in H. volcanii H26-pJAM2920 (Table 1). H. volcanii cells were grown to stationary phase (OD600 of 3 to 3.5) (4 1-liter cultures in 2.8-liter Fernbach flasks) and harvested by centrifugation (10 to 15 min at 9,200 × g and 25°C). Cell pellets were resuspended at 5 ml per 1 g (wet weight) cells in Tris-salt buffer (20 mM Tris-HCl [pH 7.5], 2 M NaCl, and 2.5 mM MgCl2) supplemented with 40 mM imidazole and 1 minitablet protease inhibitor cocktail (Roche product no. 05892791001) per 10 ml buffer. Cells were lysed by passage through a French press (three times at 20,000 lb/in2). Whole-cell lysate was clarified by centrifugation (twice for 30 min at 9,200 × g and 4°C) and sequential filtration using 0.8-μm and 0.2-μm cellulose acetate filters (Thermo Scientific Nalgene). Clarified cell lysate was applied to a HisTrap HP column (5 ml) (17-5248-01; GE Healthcare) preequilibrated and washed in 100 ml of Tris-salt buffer with 40 mM imidazole. Fractions containing HvPPA were eluted in Tris-salt buffer with a 25-ml gradient from 40 mM to 500 mM imidazole. Fractions were tested for activity, dialyzed overnight with a buffer change after 4 h against Tris-salt buffer containing 2.5 mM MgCl2 and 1 mM dithiothreitol (DTT), and concentrated by centrifugal filtration using an Amicon Ultra-x ml 10K device (EMD Millipore). HvPPA was further purified by size exclusion chromatography (SEC) in which protein (500 μl at 14.5 mg · ml−1) was applied at a flow rate of 0.3 ml · min−1 to a Superdex 200 10/300 GL column (GE Healthcare) equilibrated in Tris-salt buffer supplemented with fresh 1 mM DTT. HvPPA fractions eluting at 14.8 ml (hexamer) and 15.9 ml (trimer) were further purified using a similar SEC strategy. The purity of HvPPA was assessed with Coomassie blue-stained SDS-polyacrylamide gels and PPi activity assay. Molecular mass standards used for analytic SEC were blue dextran (for void volume), β-amylase, cytochrome c, bovine serum albumin (BSA), and alcohol dehydrogenase (Sigma-Aldrich, no. MWGF200-1KT). HvPPA fractions were pooled and stored at 4°C.
HvPPA-mediated hydrolysis of inorganic pyrophosphate (PPi) to orthophosphate was determined spectrophotometrically. Reagents were in nanopure water (Barnstead/Thermolyne Nanopure lab water system). Sodium pyrophosphate tetrabasic decahydrate (Sigma-Aldrich) was used as the substrate. For kinetic measurements, reaction mixtures (500 μl total) contained 0.5 to 1 μg HvPPA and 1 mM PPi in high-salt buffer (3 M NaCl and 20 mM Tris-HCl, pH 8.5). Reaction mixtures were incubated at 42°C for 1 to 3 min. Orthophosphate levels were determined by malachite green assay as previously described (16) with modification. Briefly, 2.5 ml of 14% (wt/vol) (NH4)2MoO4 and 0.2 ml of 11% (vol/vol) Tween 20 were added to 10-ml of color reagent (containing 1.67 ml concentrated sulfuric acid, 8.33 ml nanopure H2O, and 12.22 mg malachite green). In triplicate, 50 μl of the color reagent was mixed with 200 μl of the reaction mixture and allowed to react at room temperature for 10 min. The formation of (MG+)(H2PMo12O40) (where MG+ represents ionized malachite green) was monitored by A630. A less sensitive malachite green assay was also used as previously described (17), in which reaction aliquots (50 μl) were mixed with 250 μl Itaya color reagent [1 volume of 4.2% (wt/vol) (NH4)2MoO4 in 5 N HCl was mixed with 3 volumes of 0.2% (wt/vol) malachite green, and after 30 min the solution was filtered with a 0.45-μm filter and stored at room temperature] and 10 μl 1.5% (vol/vol) Tween 20 in triplicate. The assay was immediately monitored by A650. Freeze-dried KH2PO4 was used as the standard. Assay mixtures with PPi minus HvPPA were used for individual background subtraction. All proteins used in this assay were buffer exchanged with Tris–high-salt buffer in nanopure H2O prior to use. All experiments were performed in triplicate, and the mean ± standard deviation (SD) was calculated. Hill coefficients were calculated for Mg2+ and PPi using the Enzyme Kinetics Module of SigmaPlot, version 13.0.
UbaA-mediated hydrolysis of ATP to AMP and PPi in the presence of the ubiquitin-like SAMP1 was monitored by coupled assay with HvPPA. Proteins were buffer exchanged with HEPES-salt buffer in nanopure H2O prior to use. Reaction mixtures (500 μl total) containing 20 μM UbaA, 20 μM SAMP1, 0.5 μM HvPPA, 2.5 mM nucleotide, 2.5 mM MgCl2, and 50 μM ZnCl2 in high-salt buffer (2 M NaCl and 50 mM HEPES in nanopure H2O, pH 7.5) were incubated at 42°C for 1 h. Nucleotides were ATP, AMP, ADP, AMP-PNP, CTP, GTP, TTP, and UTP. Proteins were removed by Ultracel-3 centrifugal filtration prior to determining orthophosphate levels by malachite green assay (16) modified as described above. Assay mixtures with nucleotide alone were used for individual background subtraction.
The molar protein concentration was calculated using absorption at 280 nm and an extinction coefficient of 26,025 M−1 · cm−1 (with the assumption that all cysteines were cystines). These values were comparable to protein concentrations calculated by the Bradford method (30) using bovine serum albumin (BSA) (Thermo Scientific) as the standard.
HvPPA protein (40 μl at 1.5 mg/ml in buffer made with nanopure water containing 20 mM Tris-HCl [pH 7.5], 2 M NaCl, 2.5 mM MgCl2, and 1 mM DTT) was frozen at −80°C for 20 min in a 1.5-ml Eppendorf tube and lyophilized over a 4-h period using a VirTis 2K BenchTop XL freeze dryer. The sample was stored at room temperature and rehydrated with 40 μl nanopure H2O prior to assay.
Proteins were separated by reducing SDS-PAGE according to the Laemmli system (31). His-tagged HvPPA was analyzed by immunoblotting using a monoclonal unconjugated anti-His IgG2 antibody from mouse (27-4710-01; GE Healthcare) and alkaline phosphatase-linked goat anti-mouse IgG antibody (A5153; Sigma-Aldrich). Immunoreactive antigens were detected by chemiluminescence using CDP-Star (Applied Biosystems) as the alkaline phosphatase substrate and X-ray film (Research Products Intl. Corp.).
Evolutionary analyses were conducted in MEGA6 (32). The evolutionary history of archaeal PPAs was inferred using the neighbor-joining method (33). The evolutionary distances were computed using the p-distance method (34) and were in units of number of amino acid differences per site. The analysis involved 225 amino acid sequences. All ambiguous positions were removed for each sequence pair. A total of 263 positions were in the final data set.
To identify new PPAs with novel biochemical properties for use in bioindustry, we first analyzed the PPA sequences available in the public databases. Our focus was on archaeal PPAs of the class A type (IPR008162), since members of this family are widespread in archaea compared to PPAs of other protein families (i.e., IPR004131, IPR022934, IPR004097, and IPR023733). Based on hierarchical clustering, the class A type PPA homologs of the halophilic archaea were found to share a close evolutionary relationship that was distinct from the PPAs that have been biochemically characterized as represented by dendrogram plot (Fig. 1).
The PPA homolog of the halophilic archaeon Haloferax volcanii (HvPPA) was further analyzed by multiple amino acid sequence alignment and Phyre2-based homology modeling. HvPPA was found to be 42 to 55% identical and 60 to 71% similar in amino acid sequence to biochemically characterized PPAs of the class A type, including those from the archaea Thermoplasma acidophilum (4, 35, 36), Pyrococcus horikoshii (5, 37,–39), Sulfolobus sp. (40,–44), Methanothermobacter thermautotrophicus (Methanobacterium thermoautotrophicum) (45), and Thermococcus thioreducens (46) (Fig. 2A). Three-dimensional (3D) homology modeling suggested HvPPA to have an OB fold consisting of a central β-barrel structure and α-helices associated in a β1–8-α1-β9-α2 topology and to homo-oligomerize into a trimer and/or dimer of trimers (Fig. 2A and andB).B). Active-site residues of PPi hydrolysis were found conserved in HvPPA, including Asp69, which is predicted to provide the carboxylate functional group that performs the nucleophilic attack on the PPi substrate when Mg2+ ions are present (47, 48). Interestingly, the two cysteine residues (Cys24 and Cys85) of HvPPA were found in a Cys-X63-Cys configuration that was highly conserved among haloarchaeal PPAs and distinct from other class A type PPAs (Fig. 2A and andB).B). These cysteine residues were at a significant distance from the putative active site of HvPPA, with Cys85 predicted to reside at an intrasubunit interface, suggesting that redox status could modulate the quaternary structure of this enzyme. HvPPA was also found to have an unusually high abundance of acidic residues on its surface (Fig. 2C), which we argue could enhance the solubility and flexibility of this PPA under high-salt conditions (49, 50). In contrast, proteins with less surface charge have a tendency to aggregate and become rigid under conditions of reduced water activity. Thus, the PPAs of the halophilic archaea were pursued as a resource for identifying new enzymes with novel biochemical properties for use in biotechnology.
HvPPA was purified over 300-fold from H. volcanii H26-pJAM2920 (a strain that ectopically expressed the protein with an N-terminal His6 tag from an rRNA P2 promoter) (Table 2). HvPPA was purified by a two-step method that relied upon Ni2+-based immobilized metal ion affinity chromatography (IMAC) and size exclusion chromatography (SEC) (Fig. 3A). Based on SEC, HvPPA was associated in trimeric and hexameric configurations (Fig. 3B). The trimeric form was 23% less active than the hexamer and was not further characterized. The HvPPA hexamer was composed primarily of the His6-tagged form but included genome-encoded (untagged) forms (Fig. 3B). The molecular masses observed (25.6 and 36.6 kDa) for the HvPPA subunits by SDS-PAGE were 5 to 8 kDa larger than the theoretical molecular masses, likely due to altered SDS coating of the acidic polypeptide, which would retard its migration by SDS-PAGE (51, 52). Our finding that HvPPA purified as homotrimers and -hexamers was consistent with our 3D model and subunit arrangement of class A type PPAs from thermophilic and hyperthermophilic archaea (Table 3). We suggest that HvPPA is a dimer of trimers, as has been observed in X-ray crystal structures of related PPA enzymes (e.g., PDB 1QEZ and 3I98).
HvPPA hexamers were found to readily hydrolyze PPi to Pi, with optimal activity detected at 42°C and basic pH (pH 8 to 9) (Fig. 4A and andB).B). Supplementation of reaction mixtures with NaCl (in a 120-fold concentration range of 25 mM to 3 M) had little, if any, effect on the catalytic activity (Fig. 4C). HvPPA was inactivated when divalent cations were removed from the reaction by dialysis against the metal chelator EDTA. The PPi-hydrolyzing activity of the EDTA-treated HvPPA could be partially restored by supplementation with Mn2+ or Mg2+ ions but not by addition of Zn2+, Ca2+, Co2+, or Ni2+ (Fig. 4D). We note that the EDTA-treated enzyme was restored to only half the activity of the untreated control when assayed with 2.5 mM MgCl2. Furthermore, the untreated HvPPA was significantly stimulated by addition of high concentrations of Mg2+ to the reaction buffer, with optimal PPi-hydrolyzing activity at 20 to 40 mM MgCl2 (Fig. 4E). Addition of other divalent cations such as Mn2+ did not stimulate the activity of the HvPPA when it was not pretreated with EDTA. Together, these results suggest that HvPPA is most likely coordinated to Mg2+ (and not Mn2+) ions upon purification from H. volcanii and requires relatively high concentrations of Mg2+ for full activity. Mg2+ ions are present at high concentrations in the cytosol of haloarchaea, with reported concentrations at ~120 mM for Halobacterium salinarum (53).
Similarly to other class A type PPAs, HvPPA was inhibited by sodium fluoride (NaF), with Ki values of 1.8 mM NaF at pH 8.5 and 0.2 mM NaF at pH 7.5 (Fig. 5A). Increased sensitivity to fluoride inhibition at pH 6 to 7 compared to basic pH is commonly observed for class A type PPAs (54) and other hydrolyzing enzymes such as catalases (55) and peroxidases (56). F− ions inhibit activity of class A type PPAs by substituting the attacking nucleophile in the PPi hydrolysis reaction (47). Consistent with the NaF inhibition of HvPPA, amino acid residues interacting with the F−-, PPi-, H2O-, and Mg2+-bound molecules in the X-ray crystal structure of E. coli PPA (PDB 2AUU) were found conserved in the haloarchaeal enzyme (Fig. 5B).
HvPPA displayed non-Michaelis-Menten kinetics for PPi hydrolysis. When assayed at 42°C, HvPPA was found to have a Vmax of 465 U · mg−1 and a Km of 0.55 mM for the PPi substrate. In contrast, HvPPA had a reduced Vmax of 53 U · mg−1 and a Km of 0.26 mM for PPi at 25°C. Sigmoidal kinetic profiles indicative of positive cooperative binding were detected for Mg2+, with the degree of cooperativity represented by a Hill coefficient of 2.62 at 25°C (and a Km of 13 mM for Mg2+ determined under these conditions). HvPPA hydrolysis of nucleoside triphosphates (ATP, TTP, GTP, or CTP) or nucleoside diphosphate (ADP) was not detectable, making HvPPA useful for coupling PPAs with nucleotide-dependent enzymes in assays. Based on these results, HvPPA was found to catalyze the hydrolysis of PPi with kinetic properties that were most closely related to those of the PPA of M. thermautotrophicus among the archaeal class A type PPAs (Table 3). The low affinity of HvPPA for Mg2+ and PPi based on Km values is consistent with the unusually high levels of these types of ions in the cytosol of haloarchaea (57).
Haloarchaeal proteins are notable for their stability at 40 to 65°C and for their high tolerance of organic solvents (58,–62). Here we found HvPPA to be moderately thermostable, with a half-life of thermal inactivation of 2 h at 65°C, and to retain 82% activity after incubation for 2 h at 42°C (Fig. 6A). In contrast, class A type PPAs of the hyperthermophilic archaea are highly resistant to thermal inactivation (e.g., Pyrococcus horikoshii PPA displays a half-life of 1 h at 105°C  [Table 3]). The unique feature of HvPPA was its high tolerance of organic solvents, with little if any enzyme inactivation after 2 h of incubation in buffers supplemented with 50% (vol/vol) dimethyl sulfoxide (DMSO), dimethylformamide (DMF), ethanol, or methanol (Fig. 6B). The catalytic activity of HvPPA was also found to be robust when assayed in organic solvents. HvPPA displayed 110 to 150% activity in buffers supplemented with 25% (vol/vol) methanol and ethanol and 63 to 94% activity in buffers with 25% (vol/vol) DMSO and DMF, compared to the no-solvent controls (Fig. 6C). HvPPA was found to be more stable when stored in buffers supplemented with 2 to 3 M NaCl than in those with 1.5 M NaCl or less (Fig. 6D). However, the enzyme was fully active over a wide range of salts (as noted in Fig. 4C) and was active after storage at −20°C in 20% (vol/vol) glycerol and after lyophilization (with 100% and 85% of the activity of the untreated control, respectively).
PPAs are not yet available for use under conditions of high salt concentrations or organic solvents to drive the activity of enzymes that generate PPi as a by-product. Here we demonstrate the use of HvPPA with a PPi-generating enzyme that functions at reduced water activity and high temperature by a coupled assay. In particular, HvPPA was used to monitor the PPi by-product of the “salt-loving” enzyme UbaA of H. volcanii at 42°C in a buffer system with 2 M NaCl. UbaA has a NAD/FAD-binding fold domain common to ubiquitin-activating E1 family enzymes and is required for the formation of ubiquitin-like bonds in archaea (63). UbaA is presumed to adenylate the C-terminal α-carboxylate groups of ubiquitin-like proteins (named SAMPs in archaea) and to release PPi as a by-product (Fig. 7A). To monitor this activity, HvPPA was used in a coupled assay to drive UbaA-mediated adenylation of SAMP1 and hydrolyze the PPi by-product to 2Pi for detection by colorimetric assay. Significant levels of Pi were detected when UbaA and HvPPA were coupled with ATP and SAMP1 in the reaction (Fig. 7B). Pi was not detected when ATP, UbaA, HvPPA, or SAMP1 was omitted from the adenylation assay (Fig. 7B). Deletion of the C-terminal diglycine residues of SAMP1 (ΔGG) was found to significantly reduce the level of Pi detected. Likewise, the reaction was found to be highly specific for ATP, with little if any Pi generated enzymatically when ATP was replaced by other nucleotides (AMP, ADP, AMP-PNP, CTP, GTP, TTP, and UTP) (Fig. 7B). Based on these results, HvPPA was found to be useful for hydrolysis of PPi in coupled assays that require conditions of reduced water activity and high temperature.
Here we report a class A type PPA of the haloarchaeon H. volcanii (HvPPA) that is evolutionarily, structurally, and biochemically distinct from PPAs that have been characterized. HvPPA displays thermostable and solvent-tolerant properties and has catalytic activities that are useful for biotechnology applications. HvPPA was found to be useful for coupled assay with enzymes that generate PPi as a by-product, and it can perform this activity under conditions of high temperature and reduced water activity and is functional over a wide range of salt concentrations. We demonstrate the use of HvPPA in a novel coupled assay to detect the PPi by-product released at reduced water activity (2 M NaCl) by ATP-dependent adenylation of the ubiquitin-like SAMP1 by the salt-loving E1-like enzyme UbaA. In contrast, current PPAs are inactivated in dose-dependent manner by salt and organic solvents (22,–24). Our discovery of HvPPA and its function opens new possibilities for the hydrolysis of PPi and related compounds in systems which benefit from the use of high-ionic-strength compounds and/or organic solvents.
We thank S. Shanker at the UF ICBR Genomics Core for Sanger DNA sequencing.
Funds awarded to J.A.M.-F. through the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences and Biosciences, Physical Biosciences Program (grant DE-FG02-05ER15650), were used for optimizing enzymatic conversion of high-energy molecules. Funds to J.M.-F. through the National Institutes of Health (grant R01 GM57498) were for in vitro analysis of ubiquitin-like proteins and noncanonical E1 enzymes of archaea.
We do not have a conflict of interest to declare.
U.S. Department of Energy (DOE) funding was from the Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences and Biosciences, Physical Biosciences Program. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.