Isolation, cloning, and expression of an acid phosphatase containing phosphotyrosyl phosphatase activity from Prevotella intermedia

Abstract

A novel acid phosphatase containing phosphotyrosyl phosphatase (PTPase) activity, designated PiACP, from Prevotella intermedia ATCC 25611, an anaerobe implicated in progressive periodontal disease, has been purified and characterized. PiACP, a monomer with an apparent molecular mass of 30 kDa, did not require divalent metal cations for activity and was sensitive to orthovanadate but highly resistant to okadaic acid. The enzyme exhibited substantial activity against tyrosine phosphate-containing peptides derived from the epidermal growth factor receptor. On the basis of N-terminal and internal amino acid sequences of purified PiACP, the gene coding for PiACP was isolated and sequenced. The PiACP gene consisted of 792 bp and coded for a basic protein with an M(r) of 29,164. The deduced amino acid sequence exhibited striking similarity (25 to 64%) to those of members of class A bacterial acid phosphatases, including PhoC of Morganella morganii, and involved a conserved phosphatase sequence motif that is shared among several lipid phosphatases and the mammalian glucose-6-phosphatases. The highly conservative motif HCXAGXXR in the active domain of PTPase was not found in PiACP. Mutagenesis of recombinant PiACP showed that His-170 and His-209 were essential for activity. Thus, the class A bacterial acid phosphatases including PiACP may function as atypical PTPases, the biological functions of which remain to be determined.

FIG. 1

SDS-PAGE (16% polyacrylamide) analysis of the purified PiACP. Lane A, molecular mass markers; lane B, purified PiACP (4 μg). The positions and molecular masses of standard proteins are indicated at the left.

FIG. 2

Substrate dependence curves for PiACP. Purified PiACP was assayed at 37°C in 50 mM sodium acetate (pH 4.9) buffer with a synthetic peptide corresponding to human EGFR, either A-E-N-A-E-Y(P)-L-R-V [0.04 to 0.4 mM; Y(P), phosphotyrosine] or pNPP (0.25 to 10 mM). The data were plotted by the Lineweaver-Burk method and subjected to linear regression. The K_m and V_max values for EGFR were determined to be 0.83 ± 0.035 mM and 8.44 ± 0.24 μmol/min · mg, respectively. K_m and V_max values for pNPP were determined to be 0.24 ± 0.009 mM and 32.46 ± 0.53 μmol/min · mg, respectively.

FIG. 3

Dephosphorylation of tyrosine-phosphorylated proteins in A431 lysate by PiACP. A 2-μl aliquot of purified PiACP was incubated with 10 μl of lysate of the human epidermoid carcinoma cell line A431 for 1 h (lane b), 3 h (lane c), or 15 h (lane d). Lane a, control. Proteins were resolved by SDS-PAGE and transferred to Immobilon-P membranes for immunoanalysis as described in Materials and Methods. The molecular size corresponding to EGFR (170 kDa) is indicated. Molecular mass markers are at the left.

FIG. 4

Nucleotide and predicted amino acid sequences of PiACP. The underlined areas indicate areas of the protein that were previously analyzed by amino acid sequencing. The predicted site of proteolytic cleavage of the putative signal sequence is indicated by an arrowhead. The two convergent arrows indicate repeat sequences (details in the text). RBS?, potential ribosome-binding site.

FIG. 5

Comparison of the deduced amino acid sequence of PiACP with those of the class A bacterial ACPases. ACP-Sfl, ACPase from S. flexneri; ACP-Pst, ACPase from P. stuartii; ACP-Mm, ACPase from M. morganii; Apy-Sfl, apyrase from S. flexneri; ACP-Sty, ACPase from S. typhimurium; ACP-Zm, ACPase from Z. mobilis. Identical residues are indicated by an asterisk; conservative amino acid substitutions are indicated by a colon or dot.

FIG. 6

Overexpression of recombinant PiACP in E. coli. The ORFs for PiACP and its mutants were cloned and expressed in pET3a, as described in Materials and Methods. BL21(DE3)/pLysS cells containing the following recombinant PiACP clones were induced with (lane +) or without (lane −; control) IPTG (0.5 mM), and total extracts (lanes − and +), the denaturant-dissolved fraction (lane 1), or the purified fraction (lane 2; gel filtration fraction) were analyzed by SDS-PAGE followed by staining with Coomassie blue. Lanes −, +, 1, and 2, wild-type PiACP; lane 3, ΔGSYPSGHT mutant; lane 4, H170Q mutant; lane 5, H209T mutant; lane M, protein standards as in Fig. 1.

FIG. 7

Chemical modification of recombinant PiACP in E. coli. The residual activity was measured at 37°C (pH 4.9) with pNPP as the substrate after the following treatments: PAO (10 mM [□] and 20 mM [■]), DEPC (1 mM [○] and 5 mM [●]), and phenylglyoxal (1 mM [▵] and 5 mM [▴]).

FIG. 8

Phylogenetic tree of the class A bacterial ACPase family and other phosphatases. The unrooted tree was constructed with NjPlot of the CLUSTAL W software package. All of these phosphatases contain the motif KXXXXXXRP-(X_12–54)-PSGH-(X_31–54)-SRXXXXXHXXXD as described in the text. Abbreviations for phosphatases are as follows: E. coli phosphatidylglycerol phosphate phosphatase, PGP-Ec; Haemophilus influenzae phosphatidylglycerol phosphate phosphatase, PGP-Hi; T. denticola neutral phosphatase, NP-Td; rat G6Pase, G6P-rat; and human G6Pase, G6P-human. The other abbreviations are as defined in the legend for Fig. 5.