Catalytic Properties of Murine Carbonic Anhydrase IV*

(Received for publication, November 18, 1996, and in revised form, March 18, 1997)

Jonathan D. Hurt Dagger , Chingkuang Tu §, Philip J. Laipis Dagger and David N. Silverman §

From the § Department of Pharmacology and Therapeutics and the Dagger  Department of Biochemistry and Molecular Biology, University of Florida College of Medicine, Gainesville, Florida 32610-0267

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES


ABSTRACT

A cDNA encoding the murine carbonic anhydrase IV (mCA IV) gene, modified to resemble a form of mature human carbonic anhydrase IV (Okuyama, T., Waheed, A., Kusumoto, W., Zhu, X. L., and Sly, W. S. (1995) Arch. Biochem. Biophys. 320, 315-322), was expressed in Escherichia coli. Inactive inclusion bodies were collected and refolded, and active enzyme was purified; the resulting mCA IV was used to characterize the catalysis of CO2 hydration using stopped flow spectrophotometry and 18O exchange between CO2 and water. Unlike previously studied isozymes in this class of carbonic anhydrase, the pH profile for kcat for hydration of CO2 catalyzed by mCA IV could not be described by a single ionization, suggesting multiple proton transfer pathways between the zinc-bound water molecule and solution. A role for His64 in transferring protons between the zinc-bound water and solution was confirmed by the 100-fold lower activity of the mutant of mCA IV containing the replacement His64 right-arrow Ala. The remaining activity in this mutant at pH levels near 9 suggested a second proton shuttle mechanism. The maximal turnover number kcat for hydration of CO2 catalyzed by mCA IV was 1.1 × 106 s-1 at pH > 9. A pKa of 6.6 was estimated for the zinc-bound water molecule in mCA IV.


INTRODUCTION

The mammalian carbonic anhydrases (CAs1) constitute a gene family of at least seven distinct isozymes (1) that catalyze the hydration of CO2 to form bicarbonate and a proton: CO2 + H2O left-right-arrows  HCO3- + H+. Although these isozymes are characterized by a high degree of amino acid identity (28-59%) (2), they are quite diverse in their cellular distribution, catalytic activity, and physiological function (reviewed in Ref. 3). Among these isozymes, CA IV is the only known membrane-associated form. It was first identified and purified from bovine lung (4), although the presence of a membrane-bound carbonic anhydrase activity had been observed earlier (reviewed in Refs. 5 and 6). Subsequent purifications from human kidney (7, 8), human lung (8), and lung microsomal membranes from a variety of mammals (9) identified CA IV as a 35-52-kDa protein anchored to the membrane by a glycosyl phosphatidylinositol linkage to its C terminus. The distribution of membrane-associated carbonic anhydrases is widespread; they have been found in many secretory tissues, where they play a prominent role in, for example, the formation of ocular fluid, cerebrospinal fluid, and other secretions (reviewed in Ref. 10). Moreover, CA IV is the luminal CA in the proximal tubule of the kidney and is estimated to mediate 85% of renal bicarbonate reabsorption (10). Membrane-associated carbonic anhydrases have been found in many other tissues including the capillaries of skeletal and cardiac muscle, the colon, and the reproductive tract.

The crystal structure of human CA IV shows considerable backbone similarity to that of CA II, especially in the region of the active site (11). Two disulfide bridges appear in the structure of human CA IV (11) that are not present in CA II; these are likely responsible for the enhanced stability of CA IV against heat and SDS. Whitney and Briggle (4) reported that CA IV, unlike the other isozymes, was stable for several hours in 1 to 5% SDS, an observation that has been used to facilitate its purification.

Measurements near physiological pH have identified human (7) and bovine CA IV (12) as a fast isozyme, with a catalytic turnover kcat of approximately 2 × 105 s-1 at 0 to 1 °C, close in magnitude to that of CA II, the most efficient of the carbonic anhydrase isozymes. Early studies showed that the membrane-bound carbonic anhydrase from the brush border of the dog kidney had catalytic constants nearly identical to those of carbonic anhydrase II, suggesting that adherence to the membrane did not significantly diminish the activity to below that of CA II (13). Sulfonamides inhibit CA IV, but the average inhibition constant is 17-fold less than for CA II (12).

cDNA clones for human (14), rat (15), and mouse (16) CA IV have been isolated. Until recently, obtaining sufficient amounts of CA IV for kinetic and other studies was limited to the enzyme produced from tissue isolation or overexpression in a mammalian cell line (17). Refolding of CA IV inclusion bodies obtained from expression in Escherichia coli2 have produced sufficient quantities of active, SDS-resistant, murine carbonic anhydrase IV (mCA IV) to characterize its kinetic constants in the hydration of CO2 over a range of 4-5 pH units, allowing an examination of the mechanism of catalysis including the pKa of the zinc-bound water molecule. In this study we present the full pH profiles for catalysis of the hydration of CO2 by mCA IV using both initial velocities measured by stopped flow and 18O exchange between CO2 and water measured by mass spectrometry. The data show a very efficient enzyme but with features that are unique among the studied isozymes of the human and animal carbonic anhydrases; the maximal turnover appears to be dependent on at least two ionizations, suggesting that more than one proton shuttle group is functioning in the catalytic pathway. One of these is a shuttle group with a pKa near 7, which is identified as His64 by site-specific mutagenesis. These data are discussed in the context of a number of recent cases suggesting the role of multiple proton transfers in catalysis and proton-translocating proteins.


MATERIALS AND METHODS

Expression and Isolation of Murine CA IV

A murine Balb/c lung carbonic anhydrase IV cDNA was isolated,2 and a portion of the coding sequence (see legend to Fig. 1) was expressed in E. coli strain BL21(DE3)pLysS using the pET31 T7 expression vector described by Tanhauser et al. (18). The mutant H64A mCA IV was constructed using a mutating oligonucleotide (18, 19) and verified by DNA sequencing. The bacterial cells were lysed, and mCA IV-enriched inclusion bodies were isolated, denatured in guanidine hydrochloride, and refolded into an active form.2 Further purification was carried out using a combination of gel filtration (Ultrogel AcA 44, LKB) and ion exchange (DEAE-Sephacel, Sigma) chromatography (20). The purity of the isolated enzyme was estimated at greater than 95% by electrophoresis on 10% polyacrylamide gels. The concentration of active carbonic anhydrase was determined by titration with the carbonic anhydrase inhibitor ethoxzolamide (Ki = 16 nM) while observing the catalyzed 18O exchange between CO2 and water; more than 96% of the enzyme was active, indicating that nearly all of the purified mCA IV had been refolded into the active conformation.


Fig. 1. An alignment of the full-length murine CA IV protein sequence with that of human CA II as determined by IALIGN (40). An essentially identical alignment was obtained using BESTFIT (Version 8, 1994, Genetics Computer Group, Madison, WI). The two arrows indicate the initial and terminal amino acids of the murine CA IV utilized in this study. The first arrow marks the E right-arrow M deletion/mutation that removes the N-terminal signal peptide from CA IV, and the second arrow marks the P right-arrow TAA (termination) mutation that removes the hydrophobic C terminus, resulting in the mature or "CA II-like" form. The = in the CA II sequence marks a counted gap necessary to align its sequence with that of human CA I, creating the canonical CA amino acid numbering. None of the other gaps are counted.
[View Larger Version of this Image (41K GIF file)]

18O Exchange

The catalyzed and uncatalyzed rates of 18O exchange from species of CO2 into water and the rates of exchange of 18O between 12C-containing and 13C-containing species of CO2 were measured at chemical equilibrium using a mass spectrometer. Equations 1 and 2 demonstrate the catalytic pathway for the exchange of 18O from bicarbonate to water. In Equation 2, B- is a buffer in solution and/or an amino acid side chain in the enzyme.
<UP>HOCO</UP><SUP>18</SUP><UP>O</UP><SUP><UP>−</UP></SUP>+E<UP>ZnH</UP><SUB>2</SUB><UP>O</UP> ⇆ E<UP>Zn</UP><SUP>18</SUP><UP>OH</UP><SUP><UP>−</UP></SUP>+<UP>CO</UP><SUB>2</SUB>+<UP>H</UP><SUB>2</SUB><UP>O</UP> (Eq. 1)
E<UP>Zn</UP><SUP>18</SUP><UP>OH</UP><SUP><UP>−</UP></SUP>+<UP>BH</UP> ⇆ E<UP>Zn</UP><SUP>18</SUP><UP>OH</UP><SUB>2</SUB>+<UP>B</UP><SUP><UP>−</UP></SUP> <LIM><OP><ARROW>⇆</ARROW></OP><UL><UP>H</UP><SUB>2</SUB><UP>O</UP></UL></LIM>  (Eq. 2)
E<UP>ZnH</UP><SUB>2</SUB><UP>O</UP>+<UP>H</UP><SUB>2</SUB><SUP>18</SUP><UP>O</UP>+<UP>B</UP><SUP><UP>−</UP></SUP>
This method has been described by Silverman (21) and is capable of determining two rates in the catalytic pathway, as described below. The first is R1, the rate of interconversion of CO2 and HCO3- at chemical equilibrium. Equation 3 expresses the substrate dependence of R1.
R<SUB><UP>l</UP></SUB>/[E]=k<SUB><UP>cat</UP></SUB><SUP><UP>ex</UP></SUP>[S]/(K<SUB><UP>eff</UP></SUB><SUP><UP>S</UP></SUP>+[S]) (Eq. 3)
Here [E] is the total enzyme concentration, kcatex is a rate constant for maximal HCO3- to CO2 interconversion, [S] is the substrate concentration of HCO3- and/or CO2, and KeffS is an apparent substrate binding constant (22). This equation, when applied to the data for varying substrate concentration or to measurement of R1 when [S] <<  KeffS, can determine the values of kcatex/KeffS.

The second rate determined by this method is RH2O, the rate of release from the enzyme of water labeled with 18O (Equation 2). A proton donated from a donor group BH converts the zinc-bound hydroxide to zinc-bound water, which readily exchanges with unlabeled water. The 18O label is effectively lost by dilution into the solvent water. The value of RH2O can be interpreted in terms of the rate constant from a predominant proton donor group to the zinc-bound hydroxide according to Equation 4 (23), in which kB is the rate constant for proton transfer to the zinc-bound hydroxide, KB is the ionization constant for the donor group, and KE is the ionization constant of the zinc-bound water molecule.
R<SUB><UP>H</UP>2<UP>O</UP></SUB>/[E]=k<SUB><UP>B</UP></SUB>/{(1+K<SUB><UP>B</UP></SUB>/[<UP>H</UP><SUP><UP>+</UP></SUP>])(1+[<UP>H</UP><SUP><UP>+</UP></SUP>]/K<SUB><UP>E</UP></SUB>)} (Eq. 4)
Measurements of the rate of distribution of 18O were determined using an Extrel EMX-200 mass spectrometer and a membrane inlet permeable to dissolved gases (21). Solutions contained 5 µM EDTA (except for measurement of Cu2+ inhibition), and the total ionic strength of solution was maintained at 0.2 M by the addition of Na2SO4. Unless indicated otherwise, experiments were carried out in the absence of buffers, which were not needed to maintain pH since these experiments were carried out at chemical equilibrium.

Steady-state Constants

A stopped flow spectrophotometer (Applied Photophysics Model SF.17MV) was used to measure initial velocities of the hydration of CO2. Since this catalysis produces protons as well as HCO3-, we measured the initial rate of hydration by recording the absorbance change of a pH indicator (24). Saturated CO2 solutions were made by bubbling CO2 into water at 25 °C. Syringes with gas-tight seals were used to make CO2 dilutions from 17 to 0.24 mM. The pKa of the buffer indicator pairs, and the observed wavelengths, were as follows: Mes (pKa 6.1) with chlorophenol red (pKa 6.3, 574 nm); Mops (pKa 7.2) with p-nitrophenol (pKa 7.1, 400 nm); Hepes (pKa 7.5) with phenol red (pKa 7.5, 557 nm); Taps (pKa 8.4) with m-cresol purple (pKa 8.3, 578 nm); and Ches (pKa 9.3) with thymol blue (pKa 8.9, 590 nm). The buffer concentration was 25 mM, unless indicated otherwise, and the total ionic strength for each buffer-indicator pair system was maintained at 0.1 M by the addition of the appropriate amount of Na2SO4. Solutions contained 4 µM EDTA. The mean of four to eight reaction traces of the first 5 to 10% of the reaction was used to determine initial rates. The uncatalyzed rates were subtracted, and the rate constants kcat and kcat/Km were determined by nonlinear least squares methods (Enzfitter, Elsevier-Biosoft). The values of Km for hydration of CO2 were below 10 mM at pH < 8.5 and reached a maximal value near 20 mM at pH approaching 10; hence, we were able to achieve concentrations of CO2 approaching saturation over most of the pH range studied. The S.E. values in kcat and kcat/Km were ±8% at most.

Rate of Ester Hydrolysis

The catalysis by mCA IV of the hydrolysis of 4-nitrophenyl acetate was measured by following the increase in absorbance at 348 nm, corresponding to the isosbestic point of nitrophenol and the nitrophenolate ion using the molar absorptivity 5.0 × 103 M-1·cm-1 (25). Because the catalysis of this hydrolysis by mCA IV was slow, we determined the overall rate and subtracted from it the rate in the presence of 20 µM ethoxzolamide to inhibit mCA IV. The difference was the component of the overall hydrolysis due to the hydrolysis at the active site. Initial velocities were determined under the conditions given in the legend to Fig. 4. The value of Km for catalysis was too large to measure because it exceeded the solubility of substrate. As a result, we were limited to observing catalytic rates that were first order in the substrate from which we obtained kcat/Km.


Fig. 4. The kinetic constant kcat/Km for the hydrolysis of 4-nitrophenyl acetate catalyzed by murine CA IV at 25 °C in solutions containing 5 µM EDTA and 25 mM of one of the following buffers: pH 5.2-6.2, Mes; pH 6.8-7.4, Mops; pH 7.8-8.0, Hepes. The data represent that component of the catalytic velocity that was abolished by the addition of 20 µM of the carbonic anhydrase inhibitor ethoxzolamide. The solid line is a least squares fit of the data to a single ionization with pKa of 6.5 ± 0.2 and a maximal value of kcat/Km of 20 ± 1 M-1·s-1.
[View Larger Version of this Image (12K GIF file)]


RESULTS

Murine CA IV

A murine CA IV cDNA coding sequence closely corresponding in structure to human CA II was inserted into a pET31 vector and protein expressed in E. coli BL21(DE3)pLysS (18). An alignment of the full-length murine CA IV protein sequence with that of human CA II is shown in Fig. 1. The two arrows in Fig. 1 show the initial and terminal amino acids of the expressed mCA IV used in this study. Mature human and rat CA IV isolated from lung have been shown to be N terminally truncated by 18 and 17 amino acids, respectively, to remove the putative plasma membrane targeting sequence (9). The murine CA IV coding sequence was truncated to mimic that of the endogenous rat protein, removing the first 17 amino acids and converting the next residue into a start methionine. The aspartate in the mouse sequence that immediately follows this methionine corresponds to the second residue in the mature rat protein (9, 15). Human CA IV is C terminally cleaved during maturation immediately after serine 266 and attached to a glycosylphosphatidylinositol anchor by this residue. This membrane anchor can be enzymatically removed without altering enzyme activity, and a fully active truncated form of human CA IV can be expressed without the residues C terminal to serine 266 (corresponding to position 258 in Fig. 1) and lacking attachment to a glycosylphosphatidylinositol anchor (17). The form of murine CA IV examined here is C terminally truncated two amino acids beyond the corresponding serine in the human sequence (17). One additional change occurred during construction of the murine CA IV expression clone; this converted a lysine at position 12 to a glutamate (Fig. 1). Residue 12 is a glutamate in human CA IV (14). Occurring in a loop region of human CA IV, Glu12 is a surface residue that extends into solvent and has no interactions with other residues in the crystal form of the enzyme (11).3 We anticipate that the presence of glutamate at position 12 has no effect on the catalysis by murine CA IV.

Catalytic Activity

The rate constant R1/[E] for the interconversion of CO2 and HCO3- at chemical equilibrium catalyzed by mCA IV was measured as a function of total concentration of all species of CO2, with data at pH 6.7 given in Fig. 2. The dependence of R1/[E] on the sum of the concentrations [CO2] + [HCO3-] was nearly linear with very little evidence of saturation. For example, the very slight curvature in the data of Fig. 2 is consistent with a value of KeffS of 270 ± 30 mM, where S represents all CO2 species, both CO2 and HCO3-. Because this large value of KeffS greatly exceeded the solubility of CO2 and HCO3-, we were not able to obtain a value of kcatex in Equation 3. However, we were able to obtain kcatex/KeffS from the slope of plots such as the one in Fig. 2 or from studies with ([CO2] + [HCO3-]) <<  KeffS.


Fig. 2. The rate constant for interconversion of CO2 and HCO3-, R1/[E], catalyzed by murine CA IV as a function of the concentration of all species of CO2, [CO2] + [HCO3-], at pH 6.7 and 25 °C. Solutions contained no buffers, and total ionic strength of solution was held at 0.2 M by addition of Na2SO4.
[View Larger Version of this Image (15K GIF file)]

The ratio kcat/Km for the hydration of CO2 was determined by two methods, measurement by mass spectrometry of the exchange of 18O between CO2 and water and measurement by stopped flow of the initial velocity of CO2 hydration. When [S] <<  Km there is an equilibrium distribution of enzyme forms also in steady state; hence, the ratio kcatex/KeffCO2 (for S = CO2) obtained by 18O exchange is in theory and in practice equivalent to kcat/Km for hydration of CO2 obtained in steady-state measurements (22). The 18O method was carried out at a total concentration of 25 mM of all CO2 species under conditions (given in the legend to Fig. 3) for which ([CO2] + [HCO3-]) <<  KeffS. This approach for catalysis by mCA IV yielded a maximal value of kcat/Km of 3.2 ± 0.1 × 107 M-1·s-1 with an apparent pKa for the catalysis of 6.6 ± 0.1 (Fig. 3). The presence of up to 200 mM imidazole in solution had no effect on kcat/Km measured by 18O exchange (shown for 100 mM in Fig. 3). The measurement of kcat/Km for hydration of CO2 by stopped flow gave data with less precision (data not shown); the analysis of these results yielded a maximal value of kcat/Km of 5.0 ± 0.2 × 107 M-1·s-1 with an apparent pKa of 7.3 ± 0.2. The solvent hydrogen isotope effect on kcat/Km determined for mCA IV by 18O exchange was D(kcat/Km) = 0.83 ± 0.11, measured at pH 6.8 and 25 °C in solutions containing no buffers. The 18O exchange studies extended to catalysis by the mutant H64A mCA IV gave a maximal value of kcat/Km of 6.3 ± 0.2 × 107 M-1·s-1 with an apparent pKa of 7.3 ± 0.1 (Fig. 3).


Fig. 3. The kinetic constant kcat/Km for hydration of CO2 catalyzed by murine CA IV in the absence of buffers (bullet ), murine CA IV in the presence of 100 mM imidazole (square ), and murine H64A CA IV in the absence of buffers (triangle ). Data were obtained at 25 °C using 18O exchange. The total concentration of all CO2 species ([CO2] + [HCO3-]) was 25 mM, and the total ionic strength of solution was maintained at 0.2 M by addition of Na2SO4. Solutions also contained 5 µM EDTA. The solid line is a least squares fit to a single ionization for murine CA IV in the absence of buffers with pKa of 6.6 ± 0.1 and a maximal value of kcat/Km of 3.2 ± 0.1 × 107 M-1·s-1; the dashed line is a fit to H64A CA IV and has pKa of 7.3 ± 0.1 and a maximal value of kcat/Km of 6.3 ± 0.2 × 107 M-1·s-1.
[View Larger Version of this Image (18K GIF file)]

The value of the apparent pKa for catalysis by mCA IV was confirmed by measurement of the catalytic hydrolysis of 4-nitrophenyl acetate. The inherent efficiency of mCA IV in this catalysis was very low. To ensure that we were measuring catalysis at the same active site as is involved in hydration of CO2, we determined that component of the overall rate of catalytic hydrolysis of 4-nitrophenyl acetate that was inhibited in the presence of 20 µM ethoxzolamide, a specific and potent (Ki = 16 nM) inhibitor of mCA IV. The maximal value of kcat/Km for this hydrolysis was 20 ± 1 M-1·s-1 with an apparent pKa of 6.5 ± 0.2 (Fig. 4).

The rate constant RH2O/[E] for the proton transfer-dependent release of 18O-labeled water from the active site (Equation 2) was measured by mass spectrometry in the absence of buffers; the values of RH2O/[E] as a function of pH are shown in Fig. 5. RH2O/[E] was evaluated by Equation 4 to determine a rate constant for proton transfer to the zinc-bound hydroxide as well as values of pKa for the predominant proton donor group and for the zinc-bound water molecule. Equation 4 was fit by least squares analysis to the data for catalysis by mCA IV (Fig. 5) to yield the rate constant for proton transfer (kB = 1.4 ± 0.2 × 106 s-1) with pKa = 6.9 ± 0.1 for the proton donor group and pKa = 6.6 ± 0.1 for the zinc-bound water molecule. This latter value is in agreement with the estimate for the zinc-bound water molecule obtained by measurement of kcat/Km for hydration of CO2 and for ester hydrolysis. The solvent hydrogen isotope effect on kB was D(kB) = 1.9 ± 0.4. RH2O/[E] catalyzed by H64A mCA IV was less than that for the unmodified enzyme by close to 100-fold (Fig. 5). The value of RH2O/[E] catalyzed by H64A mCA IV was enhanced more than 10-fold and in a saturable manner by the addition of imidazole with an apparent Kmbuffer of 80 ± 20 mM and a maximal value of 1.9 ± 0.2 × 105 s-1 at pH 7.2 (data not shown). In contrast, the value of RH2O/[E] catalyzed by wild-type mCA IV was enhanced 50% by addition of imidazole at the level of saturation (50-100 mM imidazole) under the same conditions.


Fig. 5. Variation with pH of RH2O/[E] catalyzed by murine CA IV (bullet ) and by murine H64A CA IV (triangle ), both in the absence of buffers. RH2O is the rate of release from the enzyme of 18O-labeled water, and [E] is the total concentration of enzyme. Experimental conditions were as given in Fig. 2. The solid line for murine CA IV is a least squares fit of Equation 4 to the data yielding (pKa)ZnH2O = 6.6 ± 0.1, (pKa)donor = 6.9 ± 0.1, and kB = 1.4 ± 0.2 × 106 s-1.
[View Larger Version of this Image (17K GIF file)]

The catalytic turnover number for the hydration of CO2 catalyzed by mCA IV and measured at steady state had a maximal value at high pH of kcat = 1.1 ± 0.1 × 106 s-1 (Fig. 6). This maximal value of kcat is very similar to that of kB determined from 18O exchange. The pH dependence of kcat could not be fit to a single ionization and shows the influence of another ionizable group or an inhibition from other sources in the region of pH between 7 and 9 (Fig. 6). Data collected in the presence of 25 mM imidazole partially overcame this effect on kcat (Fig. 6), suggesting that it is in part due to impeded proton transfer in the active site. The turnover number kcat for CO2 hydration catalyzed by H64A mCA IV was considerably less than kcat for the unmodified mCA IV (Fig. 6).


Fig. 6. The pH dependence of the steady-state turnover number kcat for the hydration of CO2 catalyzed by murine CA IV (bullet ) and by murine H64A CA IV (triangle ) measured at 25 °C in the presence of 4 µM EDTA and 25 mM of one of the following buffers: pH 5.6-6.9, Mes; pH 6.6-7.3, Mops; pH 7.0-7.9, Hepes; pH 8.0-8.7, Taps; pH 8.7-10, Ches. The experiments identified by the open square were identical to those for murine CA IV with the exception that they also contained 25 mM imidazole in addition to one each of the buffers listed above. The solid line for wild-type murine CA IV was obtained by least squares fit and yielded two values of pKa: 6.3 ± 0.4 and 9.1 ± 0.1.
[View Larger Version of this Image (20K GIF file)]

Inhibition of mCA IV

The Ki values, describing the inhibition of 18O exchange between CO2 and water catalyzed by mCA IV, were determined from data collected at 25 °C. For ethoxzolamide, Ki = 16 ± 1 nM; for cyanate, Ki = 97 ± 13 µM. Both of these inhibitors blocked kcat/Km and RH2O equally. However, cupric ion Cu2+ blocked only RH2O with a value of IC50 = 0.6 ± 0.1 µM (Fig. 7); the value R1/[E] and hence of kcat/Km determined by 18O exchange was not affected by cupric ion concentrations as great as 40 µM (Fig. 7).


Fig. 7. R1/[E] (square ) and RH2O/[E] (bullet ) plotted against the concentration of CuSO4. The concentration of murine carbonic anhydrase IV was 12 nM in solution with 100 mM Hepes at pH 7.5 and 25 °C. The total ionic strength of solution was 0.2 M achieved by addition of Na2SO4.
[View Larger Version of this Image (15K GIF file)]


DISCUSSION

In this study we describe the catalytic properties of murine CA IV lacking an N-terminal signal peptide and C-terminal extension, an enzyme designed to resemble both the mature form of CA IV and cytosolic CA II, a well studied and efficient isozyme of carbonic anhydrase. An amino acid alignment of this murine CA IV with that of human CA II is shown in Fig. 1. The overall sequence identity of 37% includes residues involved in coordination to the zinc and in hydrogen bonding to the zinc-bound water molecule (Thr199, Glu106) that are conserved in all of the animal carbonic anhydrases. Moreover, the crystal structures of human CA IV and human CA II in the region of their active sites are very similar (11). Through these similarities, the general catalytic pathway of murine CA IV can be rationalized by analogies to isozyme II as described in Equations 5 and 6.
<UP>CO</UP><SUB>2</SUB>+E<UP>ZnOH</UP><SUP><UP>−</UP></SUP>+<UP>H</UP><SUB>2</SUB><UP>O</UP> ⇄ <UP>HCO</UP><SUB>3</SUB><SUP><UP>−</UP></SUP>+E<UP>ZnH</UP><SUB>2</SUB><UP>O</UP> (Eq. 5)
<UP>His</UP><SUP>64</SUP>-E<UP>ZnH</UP><SUB>2</SUB><UP>O</UP>+<UP>B</UP><SUP><UP>−</UP></SUP> ⇄ <UP>H<SUP>+</SUP>His</UP><SUP>64</SUP>-E<UP>ZnOH</UP><SUP><UP>−</UP></SUP>+<UP>B</UP><SUP><UP>−</UP></SUP> ⇄  (Eq. 6)
<UP>His</UP><SUB>64</SUB>-E<UP>ZnOH</UP><SUP><UP>−</UP></SUP>+<UP>BH</UP>
Here BH represents buffer in solution and/or a possible proton shuttle of the enzyme. This scheme of Equations 5 and 6 has been supported by considerable evidence for the mechanism of CA II (26), and its use for mCA IV is consistent with a number of features of this work: a solvent hydrogen isotope effect near unity, D(kcat/Km) = 0.83 ± 0.11, indicating the lack of proton transfer in Equation 5; a solvent hydrogen isotope effect on RH2O/[E] of 1.9 ± 0.4 consistent with primary proton transfer in Equation 6; and an effect of buffer imidazole on kcat but not on kcat/Km.

However, despite these similarities between CA II and CA IV, catalysis by murine CA IV appears unique among the isozymes of carbonic anhydrase based on the evidence of Fig. 6 that its catalysis involves multiple proton transfer pathways. This evidence as well as the role of His64 in the proton shuttle is discussed below.

Role of His64

The maximal turnover for CO2 hydration catalyzed by CA II has been carefully considered in previous work to be determined in rate almost entirely by the intramolecular proton transfer from the zinc-bound water molecule to His64 (27, 28). The similar values of kcat for isozymes II and IV as well as the presence and similar conformation of His64 in both of these isozymes (11) strongly suggest that the same intramolecular proton transfer contributes to the maximal velocity of CA IV. The maximal turnover number kcat for the hydration of CO2 catalyzed by mCA IV at 25 °C was observed to be 1.1 × 106 s-1 (Fig. 6), quite close in magnitude to the kcat value of 1.4 × 106 s-1 measured under comparable conditions for human CA II (24).

Further support for the role of His64 in CA IV comes from 18O-exchange studies. The rate constant for the release of 18O-labeled water from mCA IV, RH2O/[E] shown in Fig. 5, reveals a bell-shaped dependence on pH very similar to that of CA II (29). This curve is fit well by the superposition of two ionizations (Equation 4) describing the presence of the zinc-bound hydroxide and the presence of a second group that donates a proton to the zinc-bound hydroxide, such as the imidazolium side chain of His64. These 18O-exchange data confirm the pKa of 6.6 for the ionization of the zinc-bound water and also provide the pKa of 6.9 ± 0.1 for the second ionization (Fig. 5). Again this value is close to that determined by the titration of the proton magnetic resonance of the C-2 proton of His64 in human CA II, which yields a value of pKa = 7.1 (30). These same 18O-exchange data of Fig. 5 give a maximal rate constant for intramolecular proton transfer from His64 to the zinc-bound hydroxide molecule of kB = 1.4 ± 0.2 × 106 s-1, consistent with the maximal value of kcat, which is also determined by this intramolecular proton transfer. That these values should be nearly identical for proton transfer-dependent processes in the hydration and dehydration directions is the result of the values of the pKa for donor and acceptor being nearly identical at pKa = 7.

The role of His64 as proton donor in CA IV is further supported by the inhibition of catalysis by cupric ions and by the much reduced values of kcat and RH2O for the mutant H64A mCA IV. The data for the unmodified mCA IV showed inhibition of RH2O by cupric ions (IC50 = 6 × 10-7 M), with no effect on the rate of interconversion of CO2 and HCO3- (R1/[E]) up to a concentration of 40 µM (Fig. 7). These data are consistent with an inhibition of the proton-transfer pathway with no effect on the function of the zinc-bound hydroxide in the first stage of catalysis that produces HCO3-. An analogous result was obtained with CA II, in which RH2O was inhibited by Cu2+ with an IC50 of 1.0 × 10-7 M and by Hg2+ with an IC50 of 1.6 × 10-7 M with no effect on R1 (31). This was interpreted as an inhibition of the function of His64 as a proton shuttle and was confirmed by the crystal structure of the Hg2+·CA II complex showing a superposition of structures, about half with mercuric ion bound to Ndelta and about half bound to Nepsilon of the imidazole side chain of His64 (32). The similar inhibition of CA IV and CA II by cupric ions is strong evidence for the role of His64 in proton transfer by CA IV. Moreover, the inhibited value of RH2O/[E] near 2 × 104 s-1 for CA IV in the presence of excess Cu2+ (Fig. 7) is close to the value of RH2O/[E] for the mutant H64A CA IV in which His64 is replaced by alanine (Fig. 5) and is close to the value of RH2O/[E] for H64A CA II (29), suggesting that this binding of Cu2+ has nearly completely inhibited the proton transfer role of a shuttle group. The remaining proton-transfer activity is due to other less efficient proton-transfer residues of the protein or to water in the active site.

These inhibition data suggesting a proton shuttle role for His64 were supported by the much lower values of kcat for hydration and RH2O/[E] catalyzed by the mutant H64A mCA IV (Figs. 5 and 6). Both of these rate constants are determined largely by proton transfer from or to the zinc-bound water or hydroxide (27, 29) in CA II. This is further confirmed by the significant enhancement of RH2O and kcat for H64A mCA IV by addition of imidazole, achieving a chemical rescue by providing the proton-transfer group from solution (Fig. 6).

Multiple Proton-transfer Pathways

Although the maximal values of kcat for hydration are similar for mCA IV and CA II, there is a significant difference: the pH dependence of CA II can be described by a single ionization (24), that of the proton acceptor His64, but the pH dependence of kcat for mCA IV is more complex, indicating the influence of at least two ionizable groups (Fig. 6).4 One of these groups is His64 acting as a proton shuttle, as described above, and the second is at least one other proton shuttle of pKa near 9 that has not yet been identified but that is similar to that observed in CA V (33). The activation of kcat through chemical rescue caused by the presence of 25 mM imidazole (Fig. 6) suggests that the depression in the pH profile of kcat for mCA IV near pH 7 is due to reduced capacity of the proton shuttle. That is, mCA IV has the capacity to attain higher values of kcat with an external proton acceptor, a result that suggests that the function of the internal proton shuttle His64 in mCA IV is partially blocked. One possible explanation for the more complex pH profile of mCA IV may be the presence of threonine at position 65, whereas human CA II has an alanine at this position (Fig. 1). This Thr65 may block the proton shuttle function of His64 in a manner similar to that of Phe65 in CA V (33). However, in CA II Jackman et al. (34) found no effect of the replacement Ala65 right-arrow Thr on the function of the proton-shuttle mechanism involving His64. It is noteworthy that murine CA IV has glutamine at residue 63, whereas human CA II has glycine at this position; hence, in this position also murine CA IV has a bulkier side chain that could hinder the role of His64 as a proton shuttle. Tamai et al. (41) have reported evidence consistent with the influence of residue 63 in the overall catalysis by CA IV; the replacement Gly63 right-arrow Gln in bovine CA IV decreases catalytic activity measured in a colorimetric assay near neutral pH, and the replacement Gln63 right-arrow Gly in murine CA IV increases activity. This role of residue 63 can also explain the differences in the pH profile for kcat obtained for murine CA IV in this work and that of Baird et al. (42) for human CA IV. Baird et al. observed a pH profile for kcat catalyzed by human CA IV that is consistent with a single ionization; at physiological pH this value of kcat is greater by about 40-fold than that of murine CA IV. These observations represent one of the largest kinetic differences both qualitatively and quantitatively between a single isozyme of carbonic anhydrase from different animal species.

The data for kcat describing CO2 hydration catalyzed by H64A mCA IV, although lacking the proton shuttle His64, clearly show the emergence of a second proton shuttle in the enhanced activity at the higher pH range of Fig. 6. Hence, murine CA IV demonstrates at least two pathways for intramolecular proton transfer. Its further study may clarify the properties of multiple proton transfers and clarify the role of such processes in more complex cases of multiple proton pathways recently reported in enzymes (glycinamide ribonucleotide transformylase (35)) and in proton-translocation proteins (cytochrome oxidase (36) and photosynthetic reaction center (37)). Each of these cases has a similarity to our results in that mutation of a single proton-transfer residue leaves considerable activity, indicating that these residues are nonessential when mutated individually and suggesting multiple proton-transfer pathways.

Interconversion of CO2 and HCO3-

The ratio kcat/Km contains rate constants for the catalysis up to the first irreversible step in the pathway, which in the hydration direction is the release of HCO3- in Equation 5. Thus, kcat/Km contains information on the rate of interconversion of CO2 and HCO3- but not on the proton-transfer steps by which the zinc-bound hydroxide is regenerated (Equation 6). The maximal value of kcat/Km for hydration of CO2 catalyzed by mCA IV was similar when measured at 25 °C by 18O exchange (3.2 × 107 M-1·s-1; Fig. 3) and by stopped flow spectrophotometry (5.0 × 107 M-1·s-1; data not shown). The apparent pKa for this activity was 6.6 (Fig. 3). Hence, when Wistrand and Knuuttila (7) and Maren et al. (12) made measurements of kcat/Km for hydration at pH near 7.0-7.8 they were close to maximal values; their measurements were made at 0 to 1 °C and yielded kcat/Km near 2 × 107 M-1·s-1 for human and bovine CA IV. These values are to be compared with the maximal value of kcat/Km for human CA II of 1.5 × 108 M-1·s-1 at 25 °C (24).

The pH dependence of kcat/Km for hydration of CO2 catalyzed by mCA IV gave an apparent pKa of 6.6 ± 0.1 measured by 18O exchange at chemical equilibrium (Fig. 3); this is a direct indication of the pKa of the zinc-bound water molecule (38). A consistent result is the pKa of 6.5 ± 0.2 for the hydrolysis of 4-nitrophenyl acetate catalyzed by mCA IV (Fig. 4). However, for reasons that are unclear, the apparent pKa measured from kcat/Km at steady state gave a value somewhat larger, 7.3 ± 0.2. The value of the corresponding pKa for CA II is 6.9 but depends significantly on ionic strength and sulfate concentration (38). In view of these similarities, it is interesting that the maximal value of kcat/Km for the hydrolysis of 4-nitrophenyl acetate catalyzed by mCA IV at 20 ± 1 M-1·s-1 is less by a factor of at least 100 than that measured for CA II, which is as great as 3 × 103 M-1·s-1 (27, 39). The replacement His64 right-arrow Ala in mCA IV had an effect on kcat/Km for hydration of CO2 measured by 18O exchange; the maximal value is enhanced somewhat and the apparent pKa of 7.3 ± 0.1 is also increased compared with that of the unmodified mCA IV (Fig. 3). These are small effects relative to those observed for kcat and RH2O but reflect the influence of the side chain of residue 64 on the ionization of the zinc-bound water and the related effect on the rate of interconversion of CO2 and HCO3-.


FOOTNOTES

*   This work was supported by Grant GM25154 from the National Institutes of Health.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
   To whom correspondence should be addressed: Box 100267 Health Center, University of Florida College of Medicine, Gainesville, FL 32610-0267. Tel.: 352-392-3556; Fax: 352-392-9696.
1   The abbreviations used are: CA, carbonic anhydrase; mCA, murine carbonic anhydrase; H64A mCA IV, the mutant of murine carbonic anhydrase IV containing the replacement His64 right-arrow Ala; Ches, 2-(N-cyclohexylamino)ethanesulfonic acid; Mes, 2-(N-morpholino)ethanesulfonic acid; Mops, 3-(N-morpholino)propanesulfonic acid; Taps, 3[tris(hydroxymethyl)methyl]aminopropanesulfonic acid; D(k), the solvent hydrogen isotope effect on k, (k)H2O/(k)D2O.
2   J. D. Hurt, C. K. Tu, and P. J. Laipis, in preparation.
3   D. W. Christianson, personal communication.
4   It is interesting that the depression in the pH profile for kcat (Fig. 6) was not detected in the 18O-exchange experiment (Fig. 5). This observation is most likely related to experimental conditions; the 18O-exchange experiment was performed at chemical equilibrium in the absence of buffers, did not involve a significant proton transfer to solution, and would not be as greatly affected by limited mobility of His64 as kcat in the steady-state experiment, which requires proton transfer to buffer in solution. Another interesting observation is that there appears to be an effective proton acceptor with pKa >=  9 that contributes to an enhancement of kcat for H64A mCA IV and possibly for the wild-type mCA IV as well (Fig. 6). A similar observation was made for murine CA V (33). Again, this feature was not observed in the 18O-exchange experiment (Fig. 5); the transfer of protons from a group with pKa >=  9 to the zinc-bound 18O-labeled hydroxide the conjugate acid of which has pKa near 6.6 (Equation 2) involves a thermodynamic barrier that renders the rate slow relative to proton transfer from His64.

ACKNOWLEDGEMENTS

We acknowledge the technical assistance of Yanping Zhang and Yang Wang.


REFERENCES

  1. Hewett-Emmett, D., and Tashian, R. E. (1996) Mol. Phylogenet. Evol. 5, 50-77 [CrossRef][Medline] [Order article via Infotrieve]
  2. Tashian, R. E. (1989) BioEssays 10, 186-192 [Medline] [Order article via Infotrieve]
  3. Dodgson, S. J. (1991) in The Carbonic Anhydrases (Dodgson, S. J., Tashian, R. E., Gros, G., and Carter, N. D., eds), pp. 297-306, Plenum Publishing Corp., New York
  4. Whitney, P. L., and Briggle, T. V. (1982) J. Biol. Chem. 257, 12056-12059 [Abstract/Free Full Text]
  5. Maren, T. H. (1980) Ann. N. Y. Acad. Sci. 341, 246-258 [Abstract]
  6. Wistrand, P. J. (1984) Ann. N. Y. Acad. Sci. 429, 195-206 [Medline] [Order article via Infotrieve]
  7. Wistrand, P. J., and Knuuttila, K.-G. (1989) Kidney Int. 35, 851-859 [Medline] [Order article via Infotrieve]
  8. Zhu, X. L., and Sly, W. S. (1990) J. Biol. Chem. 265, 8795-8801 [Abstract/Free Full Text]
  9. Waheed, A., Zhu, X. L., and Sly, W. S. (1992) J. Biol. Chem. 267, 3308-3311 [Abstract/Free Full Text]
  10. Sly, W. S., and Hu, P. Y. (1995) Annu. Rev. Biochem. 64, 375-401 [CrossRef][Medline] [Order article via Infotrieve]
  11. Stams, T., Nair, S. K., Okuyama, T., Waheed, A., Sly, W. S., and Christianson, D. W. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 13589-13594 [Abstract/Free Full Text]
  12. Maren, T. H., Wynns, G. C., and Wistrand, P. J. (1993) Mol. Pharmacol. 44, 901-905 [Abstract]
  13. Vincent, S. H., and Silverman, D. N. (1980) Arch. Biochem. Biophys. 205, 51-56 [Medline] [Order article via Infotrieve]
  14. Okuyama, T., Sato, S., Zhu, X. L., Waheed, A., and Sly, W. S. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1315-1319 [Abstract]
  15. Fleming, R. E., Crouch, E. C., Ruzicka, C. A., and Sly, W. S. (1993) Am. J. Physiol. 265, L627-L635 [Abstract/Free Full Text]
  16. Tamai, S., Cody, L. B., and Sly, W. S. (1996) Biochem. Genet. 34, 31-43 [CrossRef][Medline] [Order article via Infotrieve]
  17. Okuyama, T., Waheed, A., Kusumoto, W., Zhu, X. L., and Sly, W. S. (1995) Arch. Biochem. Biophys. 320, 315-322 [CrossRef][Medline] [Order article via Infotrieve]
  18. Tanhauser, S. M., Jewell, D. A., Tu, C. K., Silverman, D. N., and Laipis, P. J. (1992) Gene 117, 113-117 [Medline] [Order article via Infotrieve]
  19. Kunkel, T. A., Bebenek, K., and McClary, J. (1991) Methods Enzymol. 204, 125-139 [Medline] [Order article via Infotrieve]
  20. Tu, C. K., Thomas, H. G., Wynns, G. C., and Silverman, D. N. (1986) J. Biol. Chem. 261, 10100-10103 [Abstract/Free Full Text]
  21. Silverman, D. N. (1982) Methods Enzymol. 87, 732-752 [Medline] [Order article via Infotrieve]
  22. Simonsson, I., Jonsson, B.-H., and Lindskog, S. (1979) Eur. J. Biochem. 93, 409-417 [Abstract]
  23. Silverman, D. N., Tu, C. K., Chen, X., Tanhauser, S. M., Kresge, A. J., and Laipis, P. J. (1993) Biochemistry 32, 10757-10762 [Medline] [Order article via Infotrieve]
  24. Khalifah, R. G. (1971) J. Biol. Chem. 246, 2561-2573 [Abstract/Free Full Text]
  25. Verpoorte, J. A., Mehta, S., and Edsall, J. T. (1967) J. Biol. Chem. 242, 4221-4229 [Abstract/Free Full Text]
  26. Silverman, D. N., and Lindskog, S. (1988) Acc. Chem. Res. 21, 30-36
  27. Steiner, H., Jonsson, B.-H., and Lindskog, S. (1975) Eur. J. Biochem. 59, 253-259 [Abstract]
  28. Rowlett, R. S. (1984) J. Protein Chem. 3, 369-393
  29. Tu, C. K., Silverman, D. N., Forsman, C., Jonsson, B. H., and Lindskog, S. (1989) Biochemistry 28, 7913-7918 [Medline] [Order article via Infotrieve]
  30. Campbell, I. D., Lindskog, S., and White, A. I. (1975) J. Mol. Biol. 98, 597-614 [Medline] [Order article via Infotrieve]
  31. Tu, C. K., Wynns, G. C., and Silverman, D. N. (1981) J. Biol. Chem. 256, 9466-9470 [Free Full Text]
  32. Eriksson, A. E., Jones, T. A., and Liljas, A. (1986) in Zinc Enzymes (Bertini, I., Luchinat, C., Maret, W., and Zeppezauer, M., eds), pp. 317-328, Birkhauser, Boston
  33. Heck, R. W., Boriack-Sjodin, P. A., Qian, M. Z., Tu, C. K., Christianson, D. W., Laipis, P. J., and Silverman, D. N. (1996) Biochemistry 35, 11605-11611 [CrossRef][Medline] [Order article via Infotrieve]
  34. Jackman, J. E., Merz, K. M., and Fierke, C. A. (1996) Biochemistry 35, 16421-16428 [CrossRef][Medline] [Order article via Infotrieve]
  35. Warren, M. S., Marolewski, A. E., and Benkovic, S. J. (1996) Biochemistry 35, 8855-8862 [CrossRef][Medline] [Order article via Infotrieve]
  36. Mitchell, D. M., Fetter, J. R., Mills, D. A., Ädelroth, P., Pressler, M. A., Kim, Y., Aasa, R., Brzezinski, P., Malmström, B. G., Alben, J. O., Babcock, G. T., Ferguson-Miller, S., and Gennis, R. B. (1996) Biochemistry 35, 13089-13093 [CrossRef][Medline] [Order article via Infotrieve]
  37. Baciou, L., and Michel, H. (1995) Biochemistry 34, 7967-7972 [Medline] [Order article via Infotrieve]
  38. Simonsson, I., and Lindskog, S. (1982) Eur. J. Biochem. 123, 29-36 [Medline] [Order article via Infotrieve]
  39. Pocker, Y., and Stone, T. (1967) Biochemistry 6, 668-678 [Medline] [Order article via Infotrieve]
  40. Dayhoff, M. O., Barber, W. C., and Hunt, L. T. (1983) Methods Enzymol. 91, 524-545 [Medline] [Order article via Infotrieve]
  41. Tamai, S., Waheed, A., Cody, L. B., and Sly, W. S. (1997) Proc. Natl. Acad. Sci. U. S. A. 93, 13647-13652 [Abstract/Free Full Text]
  42. Baird, T. T., Waheed, A., Okuyama, T., Sly, W. S., and Fierke, C. A. (1997) Biochemistry 36, 2669-2678 [CrossRef][Medline] [Order article via Infotrieve]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.