Enzymes for carbon sequestration: neutron crystallographic studies of carbonic anhydrase.

Carbonic anhydrase (CA) is a ubiquitous metalloenzyme that catalyzes the reversible hydration of CO(2) to form HCO(3)(-) and H(+) using a Zn-hydroxide mechanism. The first part of catalysis involves CO(2) hydration, while the second part deals with removing the excess proton that is formed during the first step. Proton transfer (PT) is thought to occur through a well ordered hydrogen-bonded network of waters that stretches from the metal center of CA to an internal proton shuttle, His64. These waters are oriented and ordered through a series of hydrogen-bonding interactions to hydrophilic residues that line the active site of CA. Neutron studies were conducted on wild-type human CA isoform II (HCA II) in order to better understand the nature and the orientation of the Zn-bound solvent (ZS), the charged state and conformation of His64, the hydrogen-bonding patterns and orientations of the water molecules that mediate PT and the ionization of hydrophilic residues in the active site that interact with the water network. Several interesting and unexpected features in the active site were observed which have implications for how PT proceeds in CA.

Preliminary joint neutron and X-ray crystallographic study of human carbonic anhydrase II.

Carbonic anhydrases catalyze the interconversion of CO(2) to HCO(3)(-), with a subsequent proton-transfer (PT) step. PT proceeds via a proposed hydrogen-bonded water network in the active-site cavity that is stabilized by several hydrophilic residues. A joint X-ray and neutron crystallographic study has been initiated to determine the specific water network and the protonation states of the hydrophilic residues that coordinate it in human carbonic anhydrase II. Time-of-flight neutron crystallographic data have been collected from a large ( approximately 1.2 mm(3)) hydrogen/deuterium-exchanged crystal to 2.4 A resolution and X-ray crystallographic data have been collected from a similar but smaller crystal to 1.5 A resolution. Obtaining good-quality neutron data will contribute to the understanding of the catalytic mechanisms that utilize water networks for PT in protein environments.

Kinetic analysis of product inhibition in human manganese superoxide dismutase.

Manganese superoxide dismutase (MnSOD) cycles between the Mn(II) and Mn(III) states during the catalyzed disproportionation of O(2)(*-), a catalysis that is limited at micromolar levels of superoxide by a peroxide-inhibited complex with the metal. We have investigated the role in catalysis and inhibition of the conserved residue Trp161 which forms a hydrophobic side of the active site cavity of MnSOD. Crystal structures of mutants of human MnSOD in which Trp161 was replaced with Ala or Phe showed significant conformational changes on adjacent residues near the active site, particularly Gln143 and Tyr34 which in wild-type MnSOD participate in a hydrogen bond network believed to support proton transfer during catalysis. Using pulse radiolysis and observing the UV absorbance of superoxide, we have determined rate constants for the catalytic dismutation of superoxide. In addition, the rates of formation and dissociation of the product-inhibited complex of these mutants were determined by direct observation of the characteristic visible absorption of the oxidized and inhibited states. Catalysis by W161A and W161F MnSOD was associated with a decrease of at least 100-fold in the catalytic rate of reduction of superoxide, which then promotes a competing pathway leading to product inhibition. The structural changes caused by the mutations at position 161 led to small changes, at most a 6-fold decrease, in the rate constant for formation of the inhibited complex. Solvent hydrogen isotope effects support a mechanism in which formation of this complex, presumably the peroxide dianion bound to the manganese, involves no rate-contributing proton transfer; however, the dissociation of the complex requires proton transfer to generate HO(2)(-) or H2O2.

The catalytic properties of human carbonic anhydrase IX.

Human carbonic anhydrase IX (CA IX) is an integral membrane protein and a member of the alpha class of carbonic anhydrases that includes the human and animal enzymes. We have prepared a truncated, recombinant form of human CA IX of 255 residues consistent with full-length human CA II, among the most efficient of the carbonic anhydrases. Catalysis by and inhibition of this form of human CA IX has been investigated using stopped-flow spectrophotometry and 18O exchange measured by mass spectrometry. In kinetic constants for the hydration of CO2, CA IX closely resembled CA II with maximal proton transfer-dependent 18O exchange near 1 micros(-1) and kcat/Km near 55 microM(-1) x s(-1). Human CA IX was very strongly inhibited by three classic sulfonamides and cyanate, with inhibition constants that are close to those for CA II.

Redox properties of human manganese superoxide dismutase and active-site mutants.

The redox potential of human manganese superoxide dismutase (MnSOD) has been difficult to determine because of the problem of finding suitable electron mediators. We have found that ferricyanide and pentacyanoaminoferrate can be used as electron mediators, although equilibration is very slow with a half-time near 6 h. Values of the midpoint potential were determined both by allowing enzyme and mediators to equilibrate up to 38 h and by reductive titration adding dithionite to enzyme and mediator. An overall value of the midpoint potential was found to be 393 +/- 29 mV. To elucidate the role of His30 and Tyr34 in the active site of human MnSOD, we have also measured the redox properties of the site-specific mutants His30Asn (H30N) and Tyr34Phe (Y34F) and compared them with the wild-type enzyme. Crystal structures have shown that each mutation interrupts a hydrogen bond network in the active site, and each causes a 10-fold decrease in the maximal velocity of catalysis of superoxide dismutation as compared with wild type. The present study shows that H30N and Y34F human MnSOD have very little effect, within experimental uncertainty, on the redox potential of the active-site metal. The redox potentials determined electrochemically were 365 +/- 28 mV for H30N and 435 +/- 30 mV for Y34F MnSOD. These results suggest that the role of His30 and Tyr34 is more in support of catalysis, probably proton transport, and not in the tuning of the redox potential.

Bicarbonate as a proton donor in catalysis by Zn(II)- and Co(II)-containing carbonic anhydrases.

Catalysis of (18)O exchange between CO(2) and water catalyzed by a Co(II)-substituted mutant of human carbonic anhydrase II is analyzed to show the rate of release of H(2)(18)O from the active site. This rate, measured by mass spectrometry, is dependent on proton transfer to the metal-bound (18)O-labeled hydroxide, and was observed in a site-specific mutant of carbonic anhydrase II in which a prominent proton shuttle residue His64 was replaced by alanine, which does not support proton transport. Upon increasing the concentration of bicarbonate, the rate of release of H(2)(18)O increased in a saturable manner to a maximum of 4 x 10(5) s(-)(1), consistent with proton transfer from bicarbonate to the Co(II)-bound hydroxide. The same mutant of carbonic anhydrase containing Zn(II) had the rate of release of H(2)(18)O smaller by 10-fold, but rate of interconversion of CO(2) and HCO(3)(-) about the same as the Co(II)-containing enzyme. These data as well as solvent hydrogen isotope effects suggest that the bicarbonate transferring the proton is bound to the cobalt in the enzyme. The enhancement of (18)O exchange caused by increasing bicarbonate concentration during catalysis by the Zn(II)-containing carbonic anhydrase from the archaeon Methanosarcina thermophila suggests that a very similar mechanism for proton donation by bicarbonate occurs with this wild-type enzyme.

Structural and kinetic analysis of the chemical rescue of the proton transfer function of carbonic anhydrase II.

Histidine 64 in human carbonic anhydrase II (HCA II) functions in the catalytic pathway of CO(2) hydration as a shuttle to transfer protons between the zinc-bound water and bulk water. Catalysis of the exchange of (18)O between CO(2) and water, measured by mass spectrometry, is dependent on this proton transfer and was decreased more than 10-fold for H64A HCA II compared with wild-type HCA II. The loss of catalytic activity of H64A HCA II could be rescued by 4-methylimidazole (4-MI), an exogenous proton donor, in a saturable process with a maximum activity of 40% of wild-type HCA II. The crystal structure of the rescued complex at 1.6 A resolution shows 4-MI bound in the active-site cavity of H64A HCA II, through pi stacking interactions with Trp 5 and H-bonding interactions with water molecules. In this location, 4-MI is about 12 A from the zinc and approximates the observed "out" position of His 64 in the structure of the wild-type enzyme. 4-MI appears to compensate for the absence of His 64 and rescues the catalytic activity of the H64A HCA II mutant. This result strongly suggests that the out conformation of His 64 is effective in the transfer of protons between the zinc-bound solvent molecule and solution.

Purification and kinetic analysis of recombinant CA XII, a membrane carbonic anhydrase overexpressed in certain cancers.

Carbonic anhydrase XII (CA XII) is a transmembrane glycoprotein with an active extracellular CA domain that is overexpressed on cell surfaces of certain cancers. Its expression has been linked to tumor invasiveness. To characterize its catalytic properties, we purified recombinant secretory forms of wild-type and mutant CA XIIs. The catalytic properties of these enzymes in the hydration of CO(2) were measured at steady state by stopped-flow spectrophotometry and at chemical equilibrium by the exchange of (18)O between CO(2) and water determined by mass spectrometry. The catalysis of CO(2) hydration by soluble CA XII has a maximal value of k(cat)/K(m) at 34 microM(-1) small middle dots(-1), which is similar to those of the membrane-associated CA IV and to soluble CA I. The pH profiles of this catalysis and the catalyzed hydrolysis of 4-nitrophenylacetate indicate that the pK(a) of the zinc-bound water in CA XII is 7.1. His64 in CA XII acts as a proton shuttle residue, as evidenced by the reduced rate constant for proton transfer in the mutants containing the replacements His64 --> Ala and His64 --> Arg, as well as by the selective inhibition of the proton transfer step by cupric ions in wild-type CA XII. The catalytic rate of CO(2) hydration by the soluble form of CA XII is identical with that of the membrane-bound enzyme. These observations suggest a role for CA XII in CO(2)/HCO(3)(-) homeostasis in cells in which it is normally expressed. They are also compatible with a role for CA XII in acidifying the microenvironment of cancer cells in which CA XII is overexpressed, providing a mechanism for CA XII to augment tumor invasiveness and suggesting CA XII as a potential target for chemotherapeutic agents.

Enhancement of catalytic efficiency by the combination of site-specific mutations in a carbonic anhydrase-related protein.

A single mutation, involving the replacement of an arginine residue with histidine to reconstruct a zinc-binding site, suffices to change a catalytically inactive murine carbonic anhydrase-related protein (CARP) to an active carbonic anhydrase with a CO2-hydration turnover number of 1.2 x 104 s-1. Further mutations, leading to a more 'carbonic anhydrase-like' active-site cavity, results in increased activity. A quintuple mutant having His94, Gln92, Val121, Val143, and Thr200 (human carbonic anhydrase I numbering system) shows kcat = 4 x 104 s-1 and kcat/Km = 2 x 107 M-1.s-1, greatly exceeding the corresponding values for carbonic anhydrase isozyme III and approaching those characterizing carbonic anhydrase I. In addition, a buffer change from 50 mM Taps/NaOH to 50 mM 1, 2-dimethylimidazole/H2SO4 at pH 9 results in a 14-fold increase in kcat for this quintuple mutant. The CO2-hydrating activity of a double mutant with His94 and Gln92 shows complex pH-dependence, but the other mutants investigated behave as if the activity (kcat/Km) is controlled by the basic form of a single group with pKa near 7.7. In a similar way to human carbonic anhydrase II, the buffer behaves formally as a second substrate in a ping-pong pattern, suggesting that proton transfer between a zinc-bound water molecule and buffer limits the maximal rate of catalysis in both systems at low buffer concentrations. However, the results of isotope-exchange kinetic studies suggest that proton shuttling via His64 is insignificant in the CARP mutant in contrast with carbonic anhydrase II. The replacement of Ile residues with Val in positions 121 or 143 results in measurable 4-nitrophenyl acetate hydrolase activity. The pH-rate profile for this activity has a similar shape to those of carbonic anhydrase I and II. CD spectra of the double mutant with His94 and Gln92 are variable, indicating an equilibrium between a compact form of the protein and a 'molten globule'-like form. The introduction of Thr200 seems to stabilize the protein.

Multiple replacements of glutamine 143 in human manganese superoxide dismutase: effects on structure, stability, and catalysis.

Glutamine 143 in human manganese superoxide dismutase (MnSOD) forms a hydrogen bond with the manganese-bound solvent molecule and is investigated by replacement using site-specific mutagenesis. Crystal structures showed that the replacement of Gln 143 with Ala made no significant change in the overall structure of the mutant enzyme. Two new water molecules in Q143A MnSOD were situated in positions nearly identical with the Oepsilon1 and Nepsilon2 of the replaced Gln 143 side chain and maintained a hydrogen-bonded network connecting the manganese-bound solvent molecule to other residues in the active site. However, their presence could not sustain the stability and activity of the enzyme; the main unfolding transition of Q143A was decreased 16 degrees C and its catalysis decreased 250-fold to k(cat)/K(m) = 3 x 10(6) M(-)(1) s(-)(1), as determined by stopped-flow spectrophotometry and pulse radiolysis. The mutant Q143A MnSOD and other mutants at position 143 showed very low levels of product inhibition and favored Mn(II)SOD in the resting state, whereas the wild type showed strong product inhibition and favored Mn(III)SOD. However, these differences did not affect the rate constant for dissociation of the product-inhibited complex in Q143A MnSOD which was determined from a characteristic absorbance at 420 nm and was comparable in magnitude ( approximately 100 s(-)(1)) to that of the wild-type enzyme. Hence, Gln 143, which is necessary for maximal activity in superoxide dismutation, appears to have no role in stabilization and dissociation of the product-inhibited complex.

Marcus rate theory applied to enzymatic proton transfer.

The hydration of CO(2) and the dehydration of HCO(3)(-) catalyzed by the carbonic anhydrases is accompanied by the transfer of protons between solution and the zinc-bound water molecule in the active site. This transfer is facilitated by amino acid residues of the enzyme which act as intramolecular proton shuttles; variants of carbonic anhydrase lacking such shuttle residues are enhanced or rescued in catalysis by intermolecular proton transfer from donors such as imidazole in solution. The resulting rate constants for proton transfer when compared with the values of the pK(a) of the donor and acceptor give Bronsted plots of high curvature. These data are described by Marcus theory which shows an intrinsic barrier for proton transfer from 1 to 2 kcal/mol and work terms or thermodynamic contributions to the free energy of reaction from 4 to10 kcal/mol. The interpretation of these Marcus parameters is discussed in terms of the well-studied pathway of the catalysis and structure of the enzymes.

Proton transfer to residues of basic pK(a) during catalysis by carbonic anhydrase.

The maximal velocity in the hydration of CO(2) catalyzed by the carbonic anhydrases in well-buffered solutions is limited by an intramolecular proton transfer from zinc-bound water to acceptor groups of the enzyme and hence to buffer in solution. Stopped-flow spectrophotometry was used to accumulate evidence that this maximal velocity is affected by residues of basic pK(a), near 8 to above 9, in catalysis of the hydration of CO(2) by carbonic anhydrases III, IV, V, and VII. A mutant of carbonic anhydrase II containing the replacement His-64-->Ala, which removes the prominent histidine proton shuttle (with pK(a) near 7), allows better observation of these basic groups. We suggest this feature of catalysis is general for the human and animal carbonic anhydrases and is due to residues of basic pK(a), predominantly lysines and tyrosines more distant from the zinc than His-64, that act as proton acceptors. These groups supplement the well-studied proton transfer from zinc-bound water to His-64 in the most efficient of the carbonic anhydrases, isozymes II, IV, and VII.

Kinetic and spectroscopic characterization of the gamma-carbonic anhydrase from the methanoarchaeon Methanosarcina thermophila.

The zinc and cobalt forms of the prototypic gamma-carbonic anhydrase from Methanosarcina thermophila were characterized by extended X-ray absorption fine structure (EXAFS) and the kinetics were investigated using steady-state spectrophotometric and (18)O exchange equilibrium assays. EXAFS results indicate that cobalt isomorphously replaces zinc and that the metals coordinate three histidines and two or three water molecules. The efficiency of either Zn-Cam or Co-Cam for CO(2) hydration (k(cat)/K(m)) was severalfold greater than HCO(3-) dehydration at physiological pH values, a result consistent with the proposed physiological function for Cam during growth on acetate. For both Zn- and Co-Cam, the steady-state parameter k(cat) for CO(2) hydration was pH-dependent with a pK(a) of 6.5-6.8, whereas k(cat)/K(m) was dependent on two ionizations with pK(a) values of 6.7-6.9 and 8.2-8.4. The (18)O exchange assay also identified two ionizable groups in the pH profile of k(cat)/K(m) with apparent pK(a) values of 6.0 and 8.1. The steady-state parameter k(cat) (CO(2) hydration) is buffer-dependent in a saturable manner at pH 8. 2, and the kinetic analysis suggested a ping-pong mechanism in which buffer is the second substrate. The calculated rate constant for intermolecular proton transfer is 3 x 10(7) M(-1) s(-1). At saturating buffer concentrations and pH 8.5, k(cat) is 2.6-fold higher in H(2)O than in D(2)O, suggesting that an intramolecular proton transfer step is at least partially rate-determining. At high pH (pH > 8), k(cat)/K(m) is not dependent on buffer and no solvent hydrogen isotope effect was observed, consistent with a zinc hydroxide mechanism. Therefore, at high pH the catalytic mechanism of Cam appears to resemble that of human CAII, despite significant structural differences in the active sites of these two unrelated enzymes.

Role of tryptophan 161 in catalysis by human manganese superoxide dismutase.

Tryptophan 161 is a highly conserved residue that forms a hydrophobic side of the active site cavity of manganese superoxide dismutase (MnSOD), with its indole ring adjacent to and about 5 A from the manganese. We have made a mutant containing the conservative replacement Trp 161 --> Phe in human MnSOD (W161F MnSOD), determined its crystal structure, and measured the catalysis of the resulting mutant using pulse radiolysis to produce O(2)(*)(-). In the structure of W161F MnSOD the phenyl side chain of Phe 161 superimposes on the indole ring of Trp 161 in the wild type. However, in the mutant, the hydroxyl side chain of Tyr 34 is 3.9 A from the manganese, closer by 1.2 A than in the wild type. The tryptophan in MnSOD is not essential for the half-cycle of catalytic activity involving reduction of the manganese; the mutant W161F MnSOD had k(cat)/K(m) at 2.5 x 10(8) M(-)(1) s(-)(1), reduced only 3-fold compared with wild type. However, this mutant exhibited a strong product inhibition with a zero-order region of superoxide decay slower by 10-fold compared with wild type. The visible absorption spectrum of W161F MnSOD in the inhibited state was very similar to that observed for the inhibited wild-type enzyme. The appearance of the inhibited form required reaction of 2 molar equiv of O(2)(*)(-) with W161F Mn(III)SOD, one to form the reduced state of the metal and the second to form the inhibited complex, confirming that the inhibited complex requires reaction of O(2)(*)(-) with the reduced form of the enzyme. This work suggests that a significant role of Trp 161 in the active site is to promote the dissociation of product peroxide, perhaps in part through its effect on the orientation of Tyr 34.

Interrupting the hydrogen bond network at the active site of human manganese superoxide dismutase.

Histidine 30 in human manganese superoxide dismutase (MnSOD) is located at a site partially exposed to solvent with its side chain participating in a hydrogen-bonded network that includes the active-site residues Tyr(166) and Tyr(34) and extends to the manganese-bound solvent molecule. We have replaced His(30) with a series of amino acids and Tyr(166) with Phe in human MnSOD. The crystal structure of the mutant of MnSOD containing Asn(30) superimposed closely with the wild type, but the side chain of Asn(30) did not participate in the hydrogen-bonded network in the active site. The catalytic activity of a number of mutants with replacements at position 30 and for the mutant containing Phe(166) showed a 10-40-fold decrease in k(cat). This is the same magnitude of decrease in k(cat) obtained with the replacement of Tyr(34) by Phe, suggesting that interrupting the hydrogen-bonded active-site network at any of the sites of these three participants (His(30), Tyr(34), and Tyr(166)) leads to an equivalent decrease in k(cat) and probably less efficient proton transfer to product peroxide. The specific geometry of His(30) on the hydrogen bond network is essential for stability since the disparate mutations H30S, H30A, and H30Q reduce T(m) by similar amounts (10-16 degrees C) compared with wild type.

Characterization of the product-inhibited complex in catalysis by human manganese superoxide dismutase.

The reduction with excess H(2)O(2) of human Mn(III) superoxide dismutase (SOD) and the active-site mutant Y34F Mn(III)SOD was measured by scanning stopped-flow spectrophotometry and revealed the presence of an intermediate in the reduction of the manganese. The visible absorption spectrum of this intermediate closely resembled that of the enzyme in the inhibited, zero-order phase of the catalyzed disproportionation of superoxide. The decay of the visible spectrum of this intermediate was 2-fold faster for the wild-type compared with the mutant Y34F Mn-SOD. This correlates with the enhanced product inhibition of Y34F during the catalysis of O-(2) dismutation. The visible spectrum of the product-inhibited complex resembles that of the azide-Mn-SOD complex, suggesting that the inhibited complex has expanded geometry about the metal to octahedral. This study shows that the inhibited complex responsible for the zero-order phase in the catalysis by Mn-SOD of superoxide dismutation can be reached through both the forward (O-(2)) and reverse (H(2)O(2)) reactions, supporting a mechanism in which the zero-order phase results from product inhibition.

Introduction of histidine analogs leads to enhanced proton transfer in carbonic anhydrase V.

The rate-limiting step in the catalysis of the hydration of CO2 by carbonic anhydrase involves transfer of protons between zinc-bound water and solution. This proton transfer can be enhanced by proton shuttle residues within the active-site cavity of the enzyme. We have used chemical modulation to provide novel internal proton transfer groups that enhance catalysis by murine carbonic anhydrase V (mCA V). This approach involves the site-directed mutation of a targeted residue to a cysteine which is then subsequently reacted with an imidazole analog containing an appropriately positioned leaving group. Compounds examined include 4-bromoethylimidazole (4-BEI), 2-chloromethylimidazole (2-CMI), 4-chloromethylimidazole (4-CMI), and a triazole analog. Two sites in mCA V, Lys 91 and Tyr 131, located on the rim of the active-site cavity have been targeted for the introduction of these imidazole analogs. Modification of the introduced Cys 131 with 4-BEI and 4-CMI resulted in enhancements of up to threefold in catalytic activity. The pH profiles indicate the presence of a new proton shuttle residue of pKa near 5.8, consistent with the introduction of a functional proton transfer group into the active site. This is the first example of incorporation by chemical modification of an unnatural amino acid analog of histidine that can act as a proton shuttle in an enzyme.

Carbon dioxide transport by proteic and facilitated transport membranes.

Membrane separation of gases is governed by the permeability of each species across the membrane. The ratio of permeabilities yields the selectivity. Use of certain organic carriers in facilitated transport membranes and the CO2 converting enzyme carbonic anhydrase (CA) in proteic and facilitated transport membranes allows a dramatic increase in CO2 selectivity over other gases. CA has a low Km (9 mM), which we predicted would allow it to scavenge CO2 to very low partial pressures. Our goal was to determine if CA could remove CO2 from an environment at levels of 0.1% or less. Prior measurements of CO2 transport across thin supported liquid membranes showed that addition of CA enhanced CO2 flux by 3- to 100-fold. Proteic films use bifunctional reagents (e.g., glutaraldehyde) to cross-link the enzyme forming a gel. Bovine serum albumin (BSA) is often added for structural stability. Using such a preparation we examined the ability of proteic films to improve CO2 selectivity and to scavenge CO2 from a mixed gas stream. Proof-of-concept results, measured by mass spectrometry, showed a fivefold improvement in CO2 capture rate with maximal improvement at CO2 values of 1% partial pressure difference in the presence of 0 atm absolute difference. At 0.1% CO2 the membrane exhibited a 76% improvement over controls. At 0.3% CO2 the improvement is about threefold. CA proteic membranes exhibit selectivity for CO2 over oxygen and nitrogen in excess of three orders of magnitude. A CA-based proteic or facilitated transport membrane should readily achieve CO2 partial pressures of 0.05% under CELSS conditions. In addition to proteic membranes we are exploring direct immobilization of engineered CA to ultra-high-permeability teflon membranes. Site-directed mutagenesis was used to add functional groups while retaining full enzymatic activity. These results provide a basis for development of far more efficient CO2 capture proteic and facilitated transport membranes with increased selectivity to values closer to 100-fold at 1% CO2. The result will be CO2 selectivity at 0.1% on the order of 400-fold. These results exceed those obtained with other technologies.