Methods for Study of Microbe – Mineral Interactions, 2006
ANALYSIS OF (BIO)GEOCHEMICAL KINETICS OF FE(III) OXIDES SUSAN L. BRANTLEY, SHANE RUEBUSH, JE-HUN ... more ANALYSIS OF (BIO)GEOCHEMICAL KINETICS OF FE(III) OXIDES SUSAN L. BRANTLEY, SHANE RUEBUSH, JE-HUN JANG, AND MING TIEN ABSTRACT 80 INTRODUCTION 81 ABIOTIC NONREDUCTIVE MINERAL DISSOLUTION SYSTEMS 81 Reaction Kinetics 81 Abiotic ...
Shewanella oneidensis demonstrates respiratory flexibility by the transfer of electrons to Fe (II... more Shewanella oneidensis demonstrates respiratory flexibility by the transfer of electrons to Fe (III) and Mn (IV) oxides under anaerobic conditions. Researchers postulate that the bacterium utilizes surface proteins to facilitate the respiratory mechanism for dissimilatory iron(III) reduction. Previous genetic and biochemical studies has shown that iron reduction is associated with the outer membrane of the cell. The identity of the terminal reductase is not yet known. S. oneidensis has been shown to use soluble extra-cellular compounds to facilitate iron(III) reduction as well as expression of novel proteins on the cell surface when interacting with iron(III) oxides. Our results show that the outer membrane fraction possess enzymatic activity for converting Fe(III) to Fe(II) as measured by ferrozine complexation. AQDS, extra-cellular organic extracts, and iron(III) both soluble and solid have been assayed for activity with outer membrane fractions. Zymograms of the membrane fractions separated by isoelectric focusing and native PAGE electrophoresis stained using ferrozine have implicated proteins that are directly involved in the Fe(III) reduction process. A proteomics analysis of outer membrane proteins has also been implemented to identify different expression patterns under Fe(III) reducing conditions. Proteins that are unique to Fe(III) reduction have been isolated and identified using N-terminal sequence analysis. We will also attempt to examine the effect of enzymatic iron(III) reduction on isotopic partitioning from in vitro assays.
The isotopic fractionation of iron during the biological reduction of iron by microbes has receiv... more The isotopic fractionation of iron during the biological reduction of iron by microbes has received much attention due to the possible use of iron isotopes as an indicator of biological activity in ancient and extraterrestrial environments. However the mechanisms of dissimilatory iron reduction have not been fully characterized. We are investigating the mechanisms by which Shewanella putrefaciens strain CN32 reduces ferric iron in the form of goethite, as well as, the resulting iron isotopic fractionation. In the experiments a PIPES buffered minimal media was used in an effort to eliminate or control the formation of secondary ferrous-iron solids. S. putrefaciens is thought to also produce an electron shuttle, which carries electrons from the cell to the iron solid. In one set of experiments, S. putrefaciens was cultured in minimal media containing goethite both with and without anthraquinone-2,6-disulfonate (AQDS, an artificial electron shuttle). Preliminary data indicates that the fractionation of iron in solution in the AQDS amended cultures is -1.57 per mil lighter than the starting goethite. This fractionation corresponds well with previously reported fractionations in similar systems. However, other researchers have shown that, in these systems, much of the reduced Fe(II) sorbs to the goethite. An acid extraction is often used to remove this sorbed Fe(II) and determine the total amount of reduced iron. This extraction was used to extract sorbed Fe(II) for isotopic analysis. Although the extraction itself may cause a fractionation effect, less than 1% of the total iron in the extraction can be attributed to this effect. Therefore, the observed fractionation should be primarily a function of the microbially reduced iron and not an artifact of the extraction. The isotope fractionation in the extraction, which includes both soluble and sorbed Fe(II), is -2.42 per mil relative to the starting goethite. We are currently combining parts of the cell involved in iron reduction (cell wall components) with an electron shuttle and goethite to accomplish in vitro Fe reduction. We will compare the in vitro iron isotope fractionations that occur without live cells to those with live cultures in an effort to elucidate iron reducing mechanisms and pathways.
This study documents the first example of in vitro solid-phase mineral oxide reduction by enzyme-... more This study documents the first example of in vitro solid-phase mineral oxide reduction by enzyme-containing membrane fractions. Previous in vitro studies have only reported the reduction of aqueous ions. Total membrane (TM) fractions from iron-grown cultures of Shewanella oneidensis MR-1 were isolated and shown to catalyze the reduction of goethite, hematite, birnessite, and ramsdellite/pyrolusite using formate. In contrast, nicotinamide adenine dinucleotide (NADH) and succinate cannot function as electron donors. The significant implications of observations related to this cell-free system are: (i) both iron and manganese mineral oxides are reduced by the TM fraction, but aqueous U(VI) is not; (ii) TM fractions from anaerobically grown, but not aerobically grown, cells can reduce the mineral oxides; (iii) electron shuttles and iron chelators are not needed for this in vitro reduction, documenting conclusively that reduction can occur by direct contact with the mineral oxide; (iv) electron shuttles and EDTA stimulate the in vitro Fe(III) reduction, documenting that exogenous molecules can enhance rates of enzymatic mineral reduction; and (v) multiple membrane components are involved in solid-phase oxide reduction. The membrane fractions, consisting of liposomes of cytoplasmic and outer membrane segments, contain at least 100 proteins including the enzyme that oxidizes formate, formate dehydrogenase. Mineral oxide reduction was inhibited by the addition of detergent Triton X-100, which solubilizes membranes and their associated proteins, consistent with the involvement of multiple electron carriers that are disrupted by detergent addition. In contrast, formate dehydrogenase activity was not inhibited by Triton X-100. The addition of anthraquinone-2,6-disulfonate (AQDS) and menaquinone-4 was unable to restore activity; however, menadione (MD) restored 33% of the activity. The addition of AQDS and MD to reactions without added detergent increased the rate of goethite reduction. The Michaelis-Menten Km values of 71 ± 22 m 2/L for hematite and 50 ± 16 m 2/L for goethite were calculated as a function of surface area of the two insoluble minerals. Vmax was determined to be 123 ± 14 and 156 ± 13 nmol Fe(II)/min/mg of TM protein for hematite and goethite, respectively. These values are consistent with in vivo rates of reduction reported in the literature. These observations are consistent with our conclusion that the enzymatic reduction of mineral oxides is an effective probe that will allow elucidation of molecular chemistry of the membrane-mineral interface where electron transfer occurs.
We measured the Fe isotope fractionation during the reactions of Fe(II) with goethite in the pres... more We measured the Fe isotope fractionation during the reactions of Fe(II) with goethite in the presence and absence of a strong Fe(III) chelator (desferrioxamine mesylate, DFAM). All experiments were completed in an O2-free glove box. The concentrations of aqueous Fe(II) ([Fe(II)aq]) decreased below the initial total dissolved Fe concentrations ([Fe(II)total], 2.15 mM) due to fast adsorption within 0.2 day. The concentration of adsorbed Fe(II) ([Fe(II)ads]) was determined as the difference between [Fe(II)aq] and the concentration of extracted Fe(II) in 0.5 M HCl ([Fe(II)extr]) (i.e., [Fe(II)ads]=[Fe(II)extr]−[Fe(II)aq]). [Fe(II)ads] also decreased with time in experiments with and without DFAM, documenting that fast adsorption was accompanied by a second, slower reaction. Interestingly, [Fe(II)extr] was always smaller than [Fe(II)total], indicating that some Fe(II) was sequestered into a pool that is not HCl-extractable. The difference was attributed to Fe(II) incorporated into goethite structure (i.e., [Fe(II)inc]=[Fe(II)total]−[Fe(II)extr]). More Fe(II) was incorporated in the presence of DFAM than in its absence at all time steps. Regardless of the presence of DFAM, both aqueous and extracted Fe(II) (δ56/54Fe(II)aq and δ56/54Fe(II)extr) became isotopically lighter than or similar to goethite (−0.27‰) at day 7, implying that the isotope exchange occurred between bulk goethite and aqueous Fe. Consistently, the mass balance indicated that the incorporated Fe is isotopically heavier than extracted Fe. These observations suggested that (i) co-adsorption of Fe(II) with DFAM resulted in more pervasive electron transfer, (ii) the electron transfer from heavy Fe(II) in the adsorbed Fe(II) to light Fe(III) in goethite results in the fixation of heavy adsorbed Fe(III) on the surface and accumulation of Fe(II) within the goethite, and (iii) desorption of the reduced, light Fe from goethite does not necessarily occur at the same surface sites where adsorption occurred.
Methionine aminopeptidases (MetAPs) are ubiquitous metallohydrolases that remove the N-terminal m... more Methionine aminopeptidases (MetAPs) are ubiquitous metallohydrolases that remove the N-terminal methionine from nascent polypeptide chains. Although various crystal structures of MetAP in the presence of inhibitors have been solved, the structural aspects of the product-bound step has received little attention. Both perpendicular- and parallel-mode electron paramagnetic resonance (EPR) spectra were recorded for the Mn(II)-loaded forms of the type-I (Escherichia coli) and type-II (Pyrococcus furiosus) MetAPs in the presence of the reaction product l-methionine (L-Met). In general, similar EPR features were observed for both [MnMn(EcMetAP-I)]-L-Met and [MnMn(PfMetAP-II)]-L-Met. The observed perpendicular-mode EPR spectra consisted of a six-line hyperfine pattern at g = 2.03 (A = 8.8 mT) with less intense signals with eleven-line splitting at g = 2.4 and 1.7 (A = 4.4 mT). The former feature results from mononuclear, magnetically isolated Mn(II) ions and this signal are 3-fold more intense in the [MnMn(PfMetAP-II)]-L-Met EPR spectrum than in the [MnMn(EcMetAP-I)]-L-Met spectrum. Inspection of the EPR spectra of both [MnMn(EcMetAP-I)]-L-Met and [MnMn(PfMetAP-II)]-L-Met at 40 K in the parallel mode reveals that the [Mn(EcMetAP-I)]-L-Met spectrum exhibits a well-resolved hyperfine split pattern at g = 7.6 with a hyperfine splitting constant of A = 4.4 mT. These data suggest the presence of a magnetically coupled dinuclear Mn(II) center. On the other hand, a similar feature was not observed for the [MnMn(PfMetAP-II)]-L-Met complex. Therefore, the EPR data suggest that L-Met binds to [MnMn(EcMetAP-I)] differently than [MnMn(PfMetAP-II)]. To confirm these data, the X-ray crystal structure of [MnMn(PfMetAP-II)]-L-Met was solved to 2.3 A resolution. Both Mn1 and Mn2 reside in a distorted trigonal bipyramidal geometry, but the bridging water molecule, observed in the [CoCo(PfMetAP-II)] structure, is absent. Therefore, L-Met binding displaces this water molecule, but the carboxylate oxygen atom of L-Met does not bridge between the two Mn(II) ions. Instead, a single carboxylate oxygen atom of L-Met interacts with only Mn1, while the N-terminal amine nitrogen atom binds to M2. This L-Met binding mode is different from that observed for L-Met binding [CoCo(EcMetAP-I)]. Therefore, the catalytic mechanisms of type-I MetAPs may differ somewhat from type-II enzymes when a dinuclear metalloactive site is present.
The gene encoding for the methionyl aminopeptidase from the hyperthermophilic archaeon Pyrococcus... more The gene encoding for the methionyl aminopeptidase from the hyperthermophilic archaeon Pyrococcus furiosus (PfMetAP-II; EC 3.4.11.18) has been inserted into a pET 27b(+) vector and overexpressed in Escherichia coli. The new expression system resulted in a 5-fold increase in purified enzyme obtained from a 5 L fermentor growth. The as-purified PfMetAP-II enzyme, to which no exogenous metal ions or EDTA was added, was found to have 1.2 equiv of zinc and 0.1 equiv of iron present by ICP-AES analysis. This enzyme had a specific activity of 5 units/mg, a 60-fold decrease from the fully loaded Fe(II) enzymes. When an additional 2 equiv of Zn(II) was added to the as-purified PfMetAP-II, no activity could be detected. The combination of these data with previously reported whole cell studies on EcMetAP-I further supports the suggestion that the in vivo metal ion for all MetAP's is Fe(II). Both Co(II)- and Fe(II)-loaded PfMetAP-II showed similar substrate specificities to EcMetAP-I. Substrate binding was largely affected by the amino acid in the P1 position and the length of the polypeptide. The substrates MSSHRWDW and MP-p-NA showed the smallest K(m) values while the substrates MGMM and MP-p-NA provided the highest turnover. The catalytic efficiency (k(cat)/K(m)) of PfMetAP-II for MP-p-NA at 30 degrees C was 799 500 and 340 930 M(-1) s(-1) for Co(II)- and Fe(II)-loaded PfMetAP-II, respectively. Maximum catalytic activity was obtained with 1 equiv of Co(II) or Fe(II), and the dissociation constants (K(d)) for the first metal binding site were found to be 50 +/- 15 and 20 +/- 15 nM for Co(II)- and Fe(II)-substituted PfMetAP-II, respectively. Electronic absorption spectral titration of a 1 mM sample of apo-PfMetAP-II with Co(II) provided a dissociation constant of 0.35 +/- 0.02 mM for the second metal binding site, a 17500-fold increase compared to the first metal binding site. The electronic absorption data also indicated that both Co(II) ions reside in a pentacoordinate geometry. PfMetAP-II shows unique thermostability and the optimal temperature for substrate turnover was found to be approximately 85 degrees C at pH 7.5 in 25 mM Hepes and 150 mM KCl buffer. The hydrolysis of MGMM was measured in triplicate between 25 and 85 degrees C at eight substrate concentrations ranging from 2 to 20 mM. Both specific activity and K(m) values increased with increasing temperature. An Arrhenius plot was constructed from the k(cat) values and was found to be linear over the temperature range 25-85 degrees C, indicating that the rate-limiting step in PfMetAP-II peptide hydrolysis does not change as a function of temperature. Co(II)- and Fe(II)-loaded PfMetAP-II have similar activation energies (13.3 and 19.4 kJ/mol, respectively). The thermodynamic parameters calculated at 25 degrees C are as follows: DeltaG++ = 46.23 kJ/mol, DeltaH++ = 10.79 kJ/mol, and DeltaS++ = -119.72 J.mol(-1).K(-1) for Co(II)-loaded PfMetAP; DeltaG++ = 46.44 kJ/mol, DeltaH++ = 16.94 kJ/mol, and DeltaS++ = -99.67 J.mol(-1).K(-1) for Fe(II)-loaded PfMetAP. Interestingly, at higher temperatures (> 50 degrees C), Fe(II)-loaded PfMetAP-II is more active (1.4-fold at 85 degrees C) than Co(II)-loaded PfMetAP-II.
Methods for Study of Microbe – Mineral Interactions, 2006
ANALYSIS OF (BIO)GEOCHEMICAL KINETICS OF FE(III) OXIDES SUSAN L. BRANTLEY, SHANE RUEBUSH, JE-HUN ... more ANALYSIS OF (BIO)GEOCHEMICAL KINETICS OF FE(III) OXIDES SUSAN L. BRANTLEY, SHANE RUEBUSH, JE-HUN JANG, AND MING TIEN ABSTRACT 80 INTRODUCTION 81 ABIOTIC NONREDUCTIVE MINERAL DISSOLUTION SYSTEMS 81 Reaction Kinetics 81 Abiotic ...
Shewanella oneidensis demonstrates respiratory flexibility by the transfer of electrons to Fe (II... more Shewanella oneidensis demonstrates respiratory flexibility by the transfer of electrons to Fe (III) and Mn (IV) oxides under anaerobic conditions. Researchers postulate that the bacterium utilizes surface proteins to facilitate the respiratory mechanism for dissimilatory iron(III) reduction. Previous genetic and biochemical studies has shown that iron reduction is associated with the outer membrane of the cell. The identity of the terminal reductase is not yet known. S. oneidensis has been shown to use soluble extra-cellular compounds to facilitate iron(III) reduction as well as expression of novel proteins on the cell surface when interacting with iron(III) oxides. Our results show that the outer membrane fraction possess enzymatic activity for converting Fe(III) to Fe(II) as measured by ferrozine complexation. AQDS, extra-cellular organic extracts, and iron(III) both soluble and solid have been assayed for activity with outer membrane fractions. Zymograms of the membrane fractions separated by isoelectric focusing and native PAGE electrophoresis stained using ferrozine have implicated proteins that are directly involved in the Fe(III) reduction process. A proteomics analysis of outer membrane proteins has also been implemented to identify different expression patterns under Fe(III) reducing conditions. Proteins that are unique to Fe(III) reduction have been isolated and identified using N-terminal sequence analysis. We will also attempt to examine the effect of enzymatic iron(III) reduction on isotopic partitioning from in vitro assays.
The isotopic fractionation of iron during the biological reduction of iron by microbes has receiv... more The isotopic fractionation of iron during the biological reduction of iron by microbes has received much attention due to the possible use of iron isotopes as an indicator of biological activity in ancient and extraterrestrial environments. However the mechanisms of dissimilatory iron reduction have not been fully characterized. We are investigating the mechanisms by which Shewanella putrefaciens strain CN32 reduces ferric iron in the form of goethite, as well as, the resulting iron isotopic fractionation. In the experiments a PIPES buffered minimal media was used in an effort to eliminate or control the formation of secondary ferrous-iron solids. S. putrefaciens is thought to also produce an electron shuttle, which carries electrons from the cell to the iron solid. In one set of experiments, S. putrefaciens was cultured in minimal media containing goethite both with and without anthraquinone-2,6-disulfonate (AQDS, an artificial electron shuttle). Preliminary data indicates that the fractionation of iron in solution in the AQDS amended cultures is -1.57 per mil lighter than the starting goethite. This fractionation corresponds well with previously reported fractionations in similar systems. However, other researchers have shown that, in these systems, much of the reduced Fe(II) sorbs to the goethite. An acid extraction is often used to remove this sorbed Fe(II) and determine the total amount of reduced iron. This extraction was used to extract sorbed Fe(II) for isotopic analysis. Although the extraction itself may cause a fractionation effect, less than 1% of the total iron in the extraction can be attributed to this effect. Therefore, the observed fractionation should be primarily a function of the microbially reduced iron and not an artifact of the extraction. The isotope fractionation in the extraction, which includes both soluble and sorbed Fe(II), is -2.42 per mil relative to the starting goethite. We are currently combining parts of the cell involved in iron reduction (cell wall components) with an electron shuttle and goethite to accomplish in vitro Fe reduction. We will compare the in vitro iron isotope fractionations that occur without live cells to those with live cultures in an effort to elucidate iron reducing mechanisms and pathways.
This study documents the first example of in vitro solid-phase mineral oxide reduction by enzyme-... more This study documents the first example of in vitro solid-phase mineral oxide reduction by enzyme-containing membrane fractions. Previous in vitro studies have only reported the reduction of aqueous ions. Total membrane (TM) fractions from iron-grown cultures of Shewanella oneidensis MR-1 were isolated and shown to catalyze the reduction of goethite, hematite, birnessite, and ramsdellite/pyrolusite using formate. In contrast, nicotinamide adenine dinucleotide (NADH) and succinate cannot function as electron donors. The significant implications of observations related to this cell-free system are: (i) both iron and manganese mineral oxides are reduced by the TM fraction, but aqueous U(VI) is not; (ii) TM fractions from anaerobically grown, but not aerobically grown, cells can reduce the mineral oxides; (iii) electron shuttles and iron chelators are not needed for this in vitro reduction, documenting conclusively that reduction can occur by direct contact with the mineral oxide; (iv) electron shuttles and EDTA stimulate the in vitro Fe(III) reduction, documenting that exogenous molecules can enhance rates of enzymatic mineral reduction; and (v) multiple membrane components are involved in solid-phase oxide reduction. The membrane fractions, consisting of liposomes of cytoplasmic and outer membrane segments, contain at least 100 proteins including the enzyme that oxidizes formate, formate dehydrogenase. Mineral oxide reduction was inhibited by the addition of detergent Triton X-100, which solubilizes membranes and their associated proteins, consistent with the involvement of multiple electron carriers that are disrupted by detergent addition. In contrast, formate dehydrogenase activity was not inhibited by Triton X-100. The addition of anthraquinone-2,6-disulfonate (AQDS) and menaquinone-4 was unable to restore activity; however, menadione (MD) restored 33% of the activity. The addition of AQDS and MD to reactions without added detergent increased the rate of goethite reduction. The Michaelis-Menten Km values of 71 ± 22 m 2/L for hematite and 50 ± 16 m 2/L for goethite were calculated as a function of surface area of the two insoluble minerals. Vmax was determined to be 123 ± 14 and 156 ± 13 nmol Fe(II)/min/mg of TM protein for hematite and goethite, respectively. These values are consistent with in vivo rates of reduction reported in the literature. These observations are consistent with our conclusion that the enzymatic reduction of mineral oxides is an effective probe that will allow elucidation of molecular chemistry of the membrane-mineral interface where electron transfer occurs.
We measured the Fe isotope fractionation during the reactions of Fe(II) with goethite in the pres... more We measured the Fe isotope fractionation during the reactions of Fe(II) with goethite in the presence and absence of a strong Fe(III) chelator (desferrioxamine mesylate, DFAM). All experiments were completed in an O2-free glove box. The concentrations of aqueous Fe(II) ([Fe(II)aq]) decreased below the initial total dissolved Fe concentrations ([Fe(II)total], 2.15 mM) due to fast adsorption within 0.2 day. The concentration of adsorbed Fe(II) ([Fe(II)ads]) was determined as the difference between [Fe(II)aq] and the concentration of extracted Fe(II) in 0.5 M HCl ([Fe(II)extr]) (i.e., [Fe(II)ads]=[Fe(II)extr]−[Fe(II)aq]). [Fe(II)ads] also decreased with time in experiments with and without DFAM, documenting that fast adsorption was accompanied by a second, slower reaction. Interestingly, [Fe(II)extr] was always smaller than [Fe(II)total], indicating that some Fe(II) was sequestered into a pool that is not HCl-extractable. The difference was attributed to Fe(II) incorporated into goethite structure (i.e., [Fe(II)inc]=[Fe(II)total]−[Fe(II)extr]). More Fe(II) was incorporated in the presence of DFAM than in its absence at all time steps. Regardless of the presence of DFAM, both aqueous and extracted Fe(II) (δ56/54Fe(II)aq and δ56/54Fe(II)extr) became isotopically lighter than or similar to goethite (−0.27‰) at day 7, implying that the isotope exchange occurred between bulk goethite and aqueous Fe. Consistently, the mass balance indicated that the incorporated Fe is isotopically heavier than extracted Fe. These observations suggested that (i) co-adsorption of Fe(II) with DFAM resulted in more pervasive electron transfer, (ii) the electron transfer from heavy Fe(II) in the adsorbed Fe(II) to light Fe(III) in goethite results in the fixation of heavy adsorbed Fe(III) on the surface and accumulation of Fe(II) within the goethite, and (iii) desorption of the reduced, light Fe from goethite does not necessarily occur at the same surface sites where adsorption occurred.
Methionine aminopeptidases (MetAPs) are ubiquitous metallohydrolases that remove the N-terminal m... more Methionine aminopeptidases (MetAPs) are ubiquitous metallohydrolases that remove the N-terminal methionine from nascent polypeptide chains. Although various crystal structures of MetAP in the presence of inhibitors have been solved, the structural aspects of the product-bound step has received little attention. Both perpendicular- and parallel-mode electron paramagnetic resonance (EPR) spectra were recorded for the Mn(II)-loaded forms of the type-I (Escherichia coli) and type-II (Pyrococcus furiosus) MetAPs in the presence of the reaction product l-methionine (L-Met). In general, similar EPR features were observed for both [MnMn(EcMetAP-I)]-L-Met and [MnMn(PfMetAP-II)]-L-Met. The observed perpendicular-mode EPR spectra consisted of a six-line hyperfine pattern at g = 2.03 (A = 8.8 mT) with less intense signals with eleven-line splitting at g = 2.4 and 1.7 (A = 4.4 mT). The former feature results from mononuclear, magnetically isolated Mn(II) ions and this signal are 3-fold more intense in the [MnMn(PfMetAP-II)]-L-Met EPR spectrum than in the [MnMn(EcMetAP-I)]-L-Met spectrum. Inspection of the EPR spectra of both [MnMn(EcMetAP-I)]-L-Met and [MnMn(PfMetAP-II)]-L-Met at 40 K in the parallel mode reveals that the [Mn(EcMetAP-I)]-L-Met spectrum exhibits a well-resolved hyperfine split pattern at g = 7.6 with a hyperfine splitting constant of A = 4.4 mT. These data suggest the presence of a magnetically coupled dinuclear Mn(II) center. On the other hand, a similar feature was not observed for the [MnMn(PfMetAP-II)]-L-Met complex. Therefore, the EPR data suggest that L-Met binds to [MnMn(EcMetAP-I)] differently than [MnMn(PfMetAP-II)]. To confirm these data, the X-ray crystal structure of [MnMn(PfMetAP-II)]-L-Met was solved to 2.3 A resolution. Both Mn1 and Mn2 reside in a distorted trigonal bipyramidal geometry, but the bridging water molecule, observed in the [CoCo(PfMetAP-II)] structure, is absent. Therefore, L-Met binding displaces this water molecule, but the carboxylate oxygen atom of L-Met does not bridge between the two Mn(II) ions. Instead, a single carboxylate oxygen atom of L-Met interacts with only Mn1, while the N-terminal amine nitrogen atom binds to M2. This L-Met binding mode is different from that observed for L-Met binding [CoCo(EcMetAP-I)]. Therefore, the catalytic mechanisms of type-I MetAPs may differ somewhat from type-II enzymes when a dinuclear metalloactive site is present.
The gene encoding for the methionyl aminopeptidase from the hyperthermophilic archaeon Pyrococcus... more The gene encoding for the methionyl aminopeptidase from the hyperthermophilic archaeon Pyrococcus furiosus (PfMetAP-II; EC 3.4.11.18) has been inserted into a pET 27b(+) vector and overexpressed in Escherichia coli. The new expression system resulted in a 5-fold increase in purified enzyme obtained from a 5 L fermentor growth. The as-purified PfMetAP-II enzyme, to which no exogenous metal ions or EDTA was added, was found to have 1.2 equiv of zinc and 0.1 equiv of iron present by ICP-AES analysis. This enzyme had a specific activity of 5 units/mg, a 60-fold decrease from the fully loaded Fe(II) enzymes. When an additional 2 equiv of Zn(II) was added to the as-purified PfMetAP-II, no activity could be detected. The combination of these data with previously reported whole cell studies on EcMetAP-I further supports the suggestion that the in vivo metal ion for all MetAP's is Fe(II). Both Co(II)- and Fe(II)-loaded PfMetAP-II showed similar substrate specificities to EcMetAP-I. Substrate binding was largely affected by the amino acid in the P1 position and the length of the polypeptide. The substrates MSSHRWDW and MP-p-NA showed the smallest K(m) values while the substrates MGMM and MP-p-NA provided the highest turnover. The catalytic efficiency (k(cat)/K(m)) of PfMetAP-II for MP-p-NA at 30 degrees C was 799 500 and 340 930 M(-1) s(-1) for Co(II)- and Fe(II)-loaded PfMetAP-II, respectively. Maximum catalytic activity was obtained with 1 equiv of Co(II) or Fe(II), and the dissociation constants (K(d)) for the first metal binding site were found to be 50 +/- 15 and 20 +/- 15 nM for Co(II)- and Fe(II)-substituted PfMetAP-II, respectively. Electronic absorption spectral titration of a 1 mM sample of apo-PfMetAP-II with Co(II) provided a dissociation constant of 0.35 +/- 0.02 mM for the second metal binding site, a 17500-fold increase compared to the first metal binding site. The electronic absorption data also indicated that both Co(II) ions reside in a pentacoordinate geometry. PfMetAP-II shows unique thermostability and the optimal temperature for substrate turnover was found to be approximately 85 degrees C at pH 7.5 in 25 mM Hepes and 150 mM KCl buffer. The hydrolysis of MGMM was measured in triplicate between 25 and 85 degrees C at eight substrate concentrations ranging from 2 to 20 mM. Both specific activity and K(m) values increased with increasing temperature. An Arrhenius plot was constructed from the k(cat) values and was found to be linear over the temperature range 25-85 degrees C, indicating that the rate-limiting step in PfMetAP-II peptide hydrolysis does not change as a function of temperature. Co(II)- and Fe(II)-loaded PfMetAP-II have similar activation energies (13.3 and 19.4 kJ/mol, respectively). The thermodynamic parameters calculated at 25 degrees C are as follows: DeltaG++ = 46.23 kJ/mol, DeltaH++ = 10.79 kJ/mol, and DeltaS++ = -119.72 J.mol(-1).K(-1) for Co(II)-loaded PfMetAP; DeltaG++ = 46.44 kJ/mol, DeltaH++ = 16.94 kJ/mol, and DeltaS++ = -99.67 J.mol(-1).K(-1) for Fe(II)-loaded PfMetAP. Interestingly, at higher temperatures (> 50 degrees C), Fe(II)-loaded PfMetAP-II is more active (1.4-fold at 85 degrees C) than Co(II)-loaded PfMetAP-II.
Uploads