Structure of PhnP, a phosphodiesterase of the carbon-phosphorus lyase pathway for phosphonate degradation

Abstract

Carbon-phosphorus lyase is a multienzyme system encoded by the phn operon that enables bacteria to metabolize organophosphonates when the preferred nutrient, inorganic phosphate, is scarce. One of the enzymes encoded by this operon, PhnP, is predicted by sequence homology to be a metal-dependent hydrolase of the beta-lactamase superfamily. Screening with a wide array of hydrolytically sensitive substrates indicated that PhnP is an enzyme with phosphodiesterase activity, having the greatest specificity toward bis(p-nitrophenyl)phosphate and 2',3'-cyclic nucleotides. No activity was observed toward RNA. The metal ion dependence of PhnP with bis(p-nitrophenyl)phosphate as substrate revealed a distinct preference for Mn(2+) and Ni(2+) for catalysis, whereas Zn(2+) afforded poor activity. The three-dimensional structure of PhnP was solved by x-ray crystallography to 1.4 resolution. The overall fold of PhnP is very similar to that of the tRNase Z endonucleases but lacks the long exosite module used by these enzymes to bind their tRNA substrates. The active site of PhnP contains what are probably two Mn(2+) ions surrounded by an array of active site residues that are identical to those observed in the tRNase Z enzymes. A second, remote Zn(2+) binding site is also observed, composed of a set of cysteine and histidine residues that are strictly conserved in the PhnP family. This second metal ion site appears to stabilize a structural motif.

FIGURE 1.

Multiple sequence alignment of PhnP homologues. Amino acid sequences are for E. coli K-12 PhnP (gi: 536936), Pseudomonas stutzeri PhnP (gi: 40804950), Pyrococcus horikoshii OT3 PhnP (gi: 14591382), Marinobacter aquaeolei VT8 PhnP (gi: 120555210), Pseudomonas putida PqqB (gi:56967240; Protein Data Bank code 1xto), E. coli ZipD (gi: 90109091; Protein Data Bank code 2cbn), and B. subtilis tRNaseZ (gi: 60594108; Protein Data Bank code 1y44). The percentage of sequence identity to E. coli PhnP is shown in parentheses. α-Helices and β-strands observed in the E. coli PhnP structure are indicated as cylinders and block arrows, respectively. Residues involved in binding the two active site metal ions are highlighted in red. The putative general acid catalyst (GAC) is highlighted in turquoise. Residues involved in binding the structural zinc ion are highlighted in yellow. The exosite of the tRNase Z enzymes involved in binding tRNA is underlined. Strictly, highly and moderately conserved residues are indicated by asterisks, colons, and periods, respectively. The sequence alignment was performed using ClustalW (53) and then edited manually.

FIGURE 2.

Screening PhnP hydrolytic activity against a series of phosphate diesters and phosphoanhydrides. PhnP (2 μg) was incubated with each substrate (0.25 mm) in the presence of a mixture of Mg²⁺ (5 mm), Mn²⁺, Ni²⁺, and Co²⁺ (0.5 mm each) in pH 8.5 buffer for 20 min at 37 °C. After treatment with alkaline phosphatase to hydrolyze any monoester product generated by PhnP, the liberated orthophosphate was quantified by absorbance at 630 nm with a malachite green assay. Assay conditions are described under “Experimental Procedures.”

FIGURE 3.

TLC (polyethyleneimine-cellulose) analysis of the reaction product of PhnP with 2′,3′-cAMP. Lanes 1 and 2, standards of 2′-AMP and 3′-AMP, respectively (5 μl each from 10 mm solutions in reaction buffer). Lane 3, an aliquot (5 μl) of the PhnP reaction with 2′,3′-cAMP (26 μm PhnP, 10 mm 2′,3′-cAMP in reaction buffer, incubated at 21 °C for 7 h). Lane 4, an aliquot (5 μl) of 2′,3′-cAMP (10 mm in reaction buffer) incubated under the same conditions as the PhnP reaction. Reaction conditions and TLC development are described under “Experimental Procedures.”

FIGURE 4.

Size exclusion chromatogram of PhnP (monomer molecular mass = 28.67 kDa). The void volume and elution volumes of selected protein standards (alcohol dehydrogenase, 150 kDa; bovine serum albumin, 66 kDa; carbonic anhydrase, 29 kDa; cytochrome c, 12.4 kDa) are indicated with black triangles. Inset, plot of log molecular weight for protein standards versus the ratio of elution volume (Ve) to void volume (Vo). The linear fit yields a slope of −0.90 ± 0.06 and a y intercept of 3.7 ± 0.1 (correlation coefficient = −0.994). The data point marked as an open square corresponds to dimeric PhnP (57.3 kDa) based on the observed Ve/Vo = 2.28.

FIGURE 5.

The crystal structure of dimeric PhnP. A, PhnP dimer. Subunit A is in magenta, and subunit B is in blue. Malate is in yellow, Mn²⁺ ions are in orange, and Zn²⁺ ion is in red. B, GRASP representation of the PhnP dimer with (S)-malate shown in both active sites. C, the active site with (S)-malate. Mn²⁺ ions are shown in orange.

FIGURE 6.

A comparison of PhnP with close structural homologues. A, PhnP compared with PqqB (Protein Data Bank code 1xto) and B. subtilis tRNase Z (Protein Data Bank code 1y44). B, alignment of B. subtilis tRNase Z (magenta) and PhnP (cyan) active site residues. Zinc ions observed in the tRNase active site are shown in red, Mn²⁺ ions observed in the PhnP active site are in orange. PhnP residues are indicated in boldface type, and tRNase Z residues are given in brackets. C, alignment of P. putida PqqB (slate) and PhnP (cyan) residues comprising the structural Zn²⁺ ion site. PhnP Zn²⁺ ion is in red, and PqqB Zn²⁺ ion is in raspberry. PhnP residues are indicated in boldface type, and PqqB residues are given in brackets.

FIGURE 7.

Active site of PhnP. A, interaction of (S)-malate with Mn²⁺ ions and Asp⁸⁰. A bound water molecule is shown in blue. The difference density for Mn²⁺ ions at the 5 σ level is shown as red mesh. B, interactions of (S)-malate with His²⁰⁰ and adjacent monomer (residues are shown in gray). The difference density is contoured at the 5 σ level. Distances between interacting groups are given in Table 5.