Mn2+ Is a Native Metal Ion Activator for Bacteriophage λ Protein Phosphatase†

Bacteriophage λ protein phosphatase (λPP) is a member of a large family of metal-containing phosphoesterases, including purple acid phosphatase, protein serine/threonine phosphatases, 5′-nucleotidase, and DNA repair enzymes such as Mre11. λPP can be activated several-fold by various divalent metal ions, with Mn2+ and Ni2+ providing the most significant activation. Despite the extensive characterization of purified λPP in vitro, little is known about the identity and stoichiometry of metal ions used by λPP in vivo. In this report, we describe the use of metal analysis, activity measurements, and whole cell EPR spectroscopy to investigate in vivo metal binding and activation of λPP. Escherichia coli cells overexpressing λPP show a 22.5-fold increase in intracellular Mn concentration and less dramatic changes in the intracellular concentration of other biologically relevant metal ions compared to control cells that do not express λPP. Phosphatase activity assessed using para-nitrophenylphosphate as substrate is increased 850-fold in cells overexpressing λPP, indicating the presence of metal-activated enzyme in cell lysate. EPR spectra of intact cells overexpressing λPP exhibit resonances previously attributed to mononuclear Mn2+ and dinuclear [(Mn)2] species bound to λPP. Spin quantitation of EPR spectra of intact E. coli cells overexpressing λPP indicates the presence of approximately 40 μM mononuclear Mn2+-λPP and 60 μM [(Mn)2]-λPP. The data suggest that overexpression of λPP results in a mixture of apo-, mononuclearMn2+, and dinuclear-[(Mn)2] metalloisoforms and that Mn2+ is a physiologically relevant activating metal ion in E. coli. The bacteriophage λ protein phosphatase (λPP)1 was originally identified and characterized as a phosphatase by Cohen et al. on the basis of significant amino acid sequence homology with mammalian protein phosphatases (PP) 1 and 2A (9, 10). One hundred fifteen residues of the N-terminus of λPP have 35% sequence identity to the N-terminal sequences of protein phosphatases 1 and 2A (PP1 and PP2A, respectively) (9, 10). PP1, PP2A, and λPP belong to a large family of metallophosphoesterases, which includes bacterial/ cyanobacterial (11), archaeal (12), fungal (13-15), protist (16), plant (17, 18), and animal (19) protein phosphatases, Mre11 nuclease (7), 5′-nucleotidase (6), and purple acid phosphatase (20-24). The enzymes in this family share a common phosphoesterase motif, DXH(X)nGDXXD(X)mGNHD/E (25-27). The amino acids highlighted in bold are situated in loops within a common secondary structural motif, the âRâRâ-fold, and are metal ligands to an active site dinuclear metal center2 (1-8). Each phosphatase in the family appears to have different metal ion requirements. Various metalloisoforms of purple acid phosphatase have been isolated, including Fe-Fe, FeZn, and Fe-Mn forms (20-24). Both PP1 and PP2A are activated in vitro by Mn2+, Co2+, and Fe2+/ascorbate, but the identity of the native metal ions is not yet resolved (2832). Calcineurin (PP2B), an Fe-Zn enzyme activated by Ca2+/calmodulin (19, 33, 34), can also be activated in vitro by Mn2+ and Ni2+, but there is no evidence that Mn2+ is an intrinsic metal activator (33, 35-41). λPP and other bacterial phosphatases can be stimulated severalfold by divalent metals, with Mn2+ and Ni2+ providing the most significant activation (27, 42). The ability of purified λPP to bind a dinuclear [(Mn)2] cofactor has been confirmed by EPR † This work was supported by a grant from the National Institutes of Health, GM46865. * Corresponding author. Address: 665 Huntington Ave. 1-512, Boston, MA 02445. Tel: (617) 432-2501. Fax: (617) 432-0377. E-mail: [email protected]. ‡ Deceased. 1 Abbreviations: BSA, bovine serum albumin; EPR, electron paramagnetic resonance; λPP, bacteriophage lambda protein phosphatase; λPPpT77, bacteriophage lambda protein phosphatase T7promoter based protein expression vector; ICP-ES, inductively coupled plasma emission spectroscopy; IPTG, isopropyl â-D-1-thiogalactopyranoside; PMSF, phenylmethylsulfonyl fluoride; pNPP, para-nitrophenyl phosphate; PP, protein phosphatase; PP1, protein phosphatase 1; PP2A, protein phosphatase 2A; PP2B, protein phosphatase 2B (calcineurin); PrPA, protein phosphatase A. 2 Four variations of the metallophosphatase active site have been classified by the Structural Classification of Proteins (SCOP) database (http://scop.mrc-lmb.cam.ac.uk/scop/). The hallmark is a dinuclear metal center with separation between metal ions from 3.1 to 3.4 Å (1-8). Of the two metal ions, designated M1 and M2, the M2 site is most highly conserved. In all cases, the ligands to the M2 site are a carboxamide ligand from an asparagine group, two histidine imidazole ligands, a carboxylate group from a conserved aspartate that bridges M1 and M2 ions, and an additional bridging oxygen ligand usually from solvent. The coordination environment about M1 differs for the four classes: In the protein serine/threonine phosphatases (λPP, PP1, calcineurin A), besides the bridging ligand atoms, the M1 ion is coordinated by a histidine imidazole group and an aspartate carboxylate (1-5). These, plus an additional glutamine ligand, are observed coordinated to the M1 ligand in E. coli 5′-nucleotidase (6), whereas an additional histidine is found in the DNA repair enzyme Mre11 from P. furiosus (7). In the purple acid phosphatases, a tyrosine residue substitutes for the histidine, and a histidine replaces a solvent molecule, for a net tyrosine for solvent substitution in the coordination sphere of the M1 site (8). 15404 Biochemistry 2002, 41, 15404-15409 10.1021/bi026317o CCC: $22.00 © 2002 American Chemical Society Published on Web 11/27/2002 spectroscopy (43, 44) and X-ray crystallography (1). Nevertheless, several other divalent metal ions are suitable activators. Extensive characterization of purified λPP has been performed, but little is known about the identity and stoichiometry of metal ions utilized in vivo. In this report, whole cell EPR spectroscopy, metal analysis, and activity measurements are utilized to investigate metal binding and activation of λPP in vivo. The data indicate that λPP preferentially binds and is activated by Mn2+ when overexpressed in E. coli. EXPERIMENTAL PROCEDURES Cell Culture and Protein Expression. Competent cells of the E. coli strain BL-21 Star (DE3)pLysS (Invitrogen) were transfected with one of two plasmids: λPPpT77, which contains the λPP gene upstream of the T7 polymerase promoter and is used to overexpress λPP, or pT7-7, the parent vector without the gene for λPP (45). Cells were grown in LB media or M9 minimal media with ampicillin (0.1 mg/mL). For LB media, cultures were grown with or without metal supplementation consisting of 504 μM FeCl2, 25.6 μM CaCl2, 20.8 μM H3BO3, 4.0 μM MnCl2, 1.52 μM CoCl2, 49.8 μM ZnCl2, 50.0 μM Na2MoO4, and 4.0 μm NiCl2 (46). For cells grown in M9 minimal medium (per liter: 6 g Na2HPO4, 3 g KH2PO4, 1 g NH4Cl, 0.5 g NaCl, 0.12 g MgSO4, 11.1 mg CaCl2) culture were grown containing 0.4% glucose, 5 mL of 200× vitamin solution (per liter: 200 mg biotin, 200 mg choline chloride, 200 mg folic acid, 200 mg of nicotinic acid, 200 mg of pantothenate, 200 mg of pyridoxal, 200 mg of riboflavin, pH adjusted to 7), 0.5 mL of 0.1% thiamine, and a 50 μM MnCl2, FeCl2, or ZnCl2 metal supplement. Cultures in LB (350 mL) were grown at 37 °C until the absorbance at 600 nm was ≈1.00. Cultures were then induced with 1 mM isopropyl â-D-1-thiogalactopyranoside (IPTG), and the temperature shifted to 27 °C. After 21 h, the cells were harvested and washed three times with 100 mL of Chelex-treated 50 mM Tris-Cl, pH 7.00, by resuspension followed by centrifugation at 11 900g for 20 min at 4 °C. Crude extract was prepared by resuspending 0.5 g wet cell paste in 5 mL of 0.1 M Tris-Cl, pH 8.00, containing 4 mM phenylmethylsulfonyl fluoride (PMSF) and 0.4 mg/mL lysozyme (Sigma), incubated for 3 h at 4 °C; homogenized (Ultra-Turrax T8, IKA Labortechnik) three times for 30 s, and centrifuged at 23 700g (4 °C) for 45 min. The supernatant was used directly for phosphatase assays and protein concentration determination (described below). Cultures in minimal media were grown and cell lysate prepared as described above, with the exception that induction was performed for 14 h. Metal Analysis. Concentrated nitric acid (11 M) was added to an aliquot of the supernatant prepared as above to a final concentration of 1 M. The suspension was incubated overnight at 4 °C and then centrifuged at 16 000g, 4 °C, for 15 min. Each supernatant was diluted in 0.1 M Tris-Cl, pH 8.0 for metal analysis. Mn and Ni analyses were performed using atomic absorption spectroscopy (Perkin-Elmer 3100, Shelton, CT), and Fe, Zn, Cu, Ca, and Mg analyses were performed using Inductively Coupled Plasma Emission Spectroscopy (ICP-ES) in the Mayo Metals Laboratory. Metal concentrations reported represent the concentration in crude cell extract. Metal Stoichoimetry. The metal stoichiometry of λPP in cell lysate was determined from the metal concentrations, determined as described above, and an estimation of the concentration of λPP, determined by comparing the Coomassie blue-stained intensity of an aliquot of cell lysate against known quantities of purified λPP by use of SDSPAGE. For each gel, the intensity of bands corresponding to λPP were analyzed using ImageQuant software (Molecular Dynamics), and a standard curve of intensity versus micrograms of purified λPP was used to estimate the concentration of λPP in cell lysate. Phosphatase ActiVity Assays. Cell lysate prior to nitric acid treatment was assayed for phosphatase activity in 0.1 M TrisCl, pH 7.80, using 20 mM p-nitrophenylphosphate (pNPP) as substrate. Product formation was measured spectrophotometrically at 410 nm using ∆ 410 ) 14 400 M-1 cm-1 at pH 7.8. It was necessary to dilute cell lysate from overexpressing cells 20-fold in 0.1 M Tris-Cl, pH 7.80, prior to use in the phosphatase assays but there was no need to dilute the lysate from nonexpressing control cells. Protein concentrations were determined using the Coomassie Plus Protein Assay Reagent (Pierce, Rockford, IL) using BSA as a standard. EPR Spectroscopy. Cultures (75 mL) of E. coli BL-21 Star (DE3) cells transfected with λPPpT77 or control cells without this plasmid were grown with metal supplementation in LB with ampicillin (0.1 mg/mL) or LB, respectively, and induced with IPTG as described above. After 21 h, 40 mL of each culture was centrifuged at 23 700g, 4 °C, for 5 min to pellet the cells. The supernatant was discarded and each cell pellet was resuspended in 800-825 μL of 0.1 M Tris-Cl, pH 8.00 (Chelex-treated), to normalize samples to equivalent optical densities at 600 nm. An aliquot (250 μL) of each cell suspension was transferred to a quartz EPR tube and frozen by immersion in liquid nitrogen. EPR spectra were recorded using a Bruker ESP 300E spectrometer operating at X-band microwave frequency equipped with an Oxford Instruments ESR 900 continuousflow cryostat for cryogenic temperature regulation. Signal averaging to improve signal-to-noise was performed by averaging 20 scans for each EPR sample of whole cells. Samples of mononuclear Mn2+and dinuclear [(Mn)2]λPP were prepared as described (43). Following desalting, the mononuclear Mn2+-λPP sample had an Mn/protein ratio of 0.63. The dinuclear [(Mn)2]-λPP sample was prepared by addition of two equivalents of Mn2+ to the enzyme sample. Estimation of the concentration of mononuclear Mn2+and dinuclear [(Mn)2]-λPP in whole cell EPR samples was performed by comparing the intensity of specific EPR signals in the spectra from intact E. coli cells to the corresponding signals observed in spectra of mononuclear-Mn2+ and dinuclear-[(Mn)2] forms of λPP, prepared as described previously (43, 44). EPR spectra were measured at equivalent microwave power and temperature and corrected for gain and number of scans.