Some Characteristics of Human, Bovine, and Horse Carbonic Anhydrases As Revealed by Inactivation Studies*

Abstract Two representatives of the low (human Enzyme B and horse Enzyme B) and two representatives of the high (bovine Enzyme B and human Enzyme C) catalytic activity forms of erythrocyte carbonic anhydrase were reacted with bromoacetate and bromoacetazolamide with the aim of elucidating structural differences at their active sites. The dissociation constants of the reversible inhibition of bromoacetate and bromoacetazolamide have been determined. It was found that the constants for bromoacetate showed 25- to 30-fold differences between the low and high activity forms of the enzyme, but only 1.1- to 6-fold differences within one activity type. The dissociation constants of the enzyme-bromoacetazolamide complex were almost the same for the two groups. Several lines of evidence suggest that bromoacetazolamide inactivates both types of enzymes with simultaneous alkylation of a histidine (or histidines) at or near the active site at the 3-nitrogen position. The rate of alkylation is, however, markedly different for the two types. Stoichiometric amounts of this inhibitor partially alkylate a histidine of the high catalytic activity forms, but do not react with the low catalytic activity forms. At higher inhibitor concentrations both types react, but the high activity forms react significantly faster. By contrast, bromoacetate at low concentrations inactivates exclusively the low catalytic activity forms with the carboxymethylation of a histidine at the 3-nitrogen position. Evidence is presented that this reaction for the horse enzyme occurs at or near the active site. At higher inhibitor concentrations the high catalytic activity forms also became inactivated, but this reaction is nonspecific. The half-times of the inactivation by bromoacetate were also determined. These values for the high catalytic activity enzymes were found to be 36 times greater than could be explained on a basis of enzyme-inhibitor complex concentration. It is suggested that a difference in the conformation of the active site, or a steric hindrance brought about by an amino acid side chain sufficiently close to the reactive histidine, would account for the observed differences in the rates of alkylation and inactivation.

It was found that the constants for bromoacetate showed 25-to 30-fold differences between the low and high activity forms of the enzyme, but only l.l-to 6-fold differences within one activity type.
The dissociation constants of the enzyme-bromoacetazolamide complex were almost the same for the two groups.
Several lines of evidence suggest that bromoacetazolamide inactivates both types of enzymes with simultaneous alkylation of a histidine (or histidines) at or near the active site at the j-nitrogen position.
The rate of alkylation is, however, markedly different for the two types. Stoichiometric amounts of this inhibitor partially alkylate a histidine of the high catalytic activity forms, but do not react with the low catalytic activity forms.
At higher inhibitor concentrations both types react, but the high activity forms react significantly faster.
By contrast, bromoacetate at low concentrations inactivates exclusively the low catalytic activity forms with the carboxymethylation of a histidine at the j-nitrogen position. Evidence is presented that this reaction for the horse enzyme occurs at or near the active site. At higher inhibitor concentrations the high catalytic activity forms also became inactivated, but this reaction is nonspecific.
The half-times of the inactivation by bromoacetate were also determined. These values for the high catalytic activity enzymes were found to be 36 times greater than could be explained on a basis of enzyme-inhibitor complex concentration.
It is suggested that a difference in the conformation of the active site, or a steric hindrance brought about by an amino acid side chain sufficiently close to the reactive histidine, would account for the observed differences in the rates of alkylation and inactivation. *  by haloacetic acids and bovine Enzyme B by bromoacet'azolamide has recently been shown (l-3).
The inactivation was found to occur in bot,h cases by a modification of 1 eq of histidine residues at the 3-nitrogen position (l-3).
It has also been shown that the innctivation is preceded by the formation of an enzyme-inhibitor complex, indicating that the inactivation occurs via the active site (l-3).
These investigations have revealed that bromoacetate at a concentration that almost' completely inactivated human B does not inactivate human C (2) or bovine B carbonic anhydrases (3). The latter two enzymes differ from the previous one in that they catalyze several times faster the reversible hydration of COZ as well as the hydrolysis of esters (4).
It is the purpose of the present paper to describe further in vestigations  of the covalent interaction  of bromoacetate  and  bromoacetazolamide with human, bovine, and horse carbonic anhydrases and to consider the implications of the resulting dat,a on the relationship between the chemical reactivity of the active site histidine and the catalytic activity of carbonic anhydrase isoenzymes.

Preparation of Carbonic
Ilnhydrases-Crude human carbonic anhydrases were prepared from red blood cells.' The conditions for hemolysis, as well as the preparation of crude carbonic anhydrase by chloroform-ethanol treatment, were similar to those described by Armstrong et al. (5) with the exception that the dialyzed supernatant was freeze-dried. A column measuring 4 x 40 cm was packed with DEAE-cellulose (previously washed according to Peterson and Sober (6)) and equilibrated with 0.01 M Tris chloride buffer at pH 8. To the column 800 mg of crude carbonic anhydrase were applied and &ted at 5" with a concave gradient that was developed by two metering pumps.
The flow rate from the reservoir (which contained 0.1 M 'I'ris chloride buffer, pH 8) to the mixing chamber (containing 2500 mI of 0.01 M Tris chloride buffer, pH 8) \vas 80 ml l)er hour. The flow rate from the mixing clhambcr to the column XV:LS set at 230 ml per hour. The effluent was continuously monitored with the aid of an 1x0 ultraviolet analyzer (model UA 2).
Two well separated fractions wcw obtained. The faster moving fraction (400 to 55s ml, Fraction I) gave two bands on starch gel corresponding to Enzymes 1~ and C (50 to 60 mg). The slower moving fraction (690 lo 996 ml, Fraction 11) was pure Enzyme B (350 to 400 mg). Fraction 1 was dialyzed against dist,illcd water and freeze-dried; a total of 200 mg of this fraction was applied to a DEAE-Sephadex J-50 column (2.5 X 25 cm) which ~vas equilibrated with 0.05 M Tris chloride buffer at, 1111 8 and 5". The column was eluted with the same buffer at 5" with a flow rate of 26 ml per hour. Two fractions were obtained.
Horse blood was purchased from Woodlyn Farms, Guelph, Ontario.
Crude horse carbonic auhydrascs were prepared as described by Furth ('7); 500 mg of crude enzyme mixture, which still contained considerable :unount,s of hemoglobin, were chromntogral)hed on a DEAF-cellulose column, 4 x 40 cm, at pH 8 as described above for the scpwat~ion of crude human carbonic anhydrases.
Two major fractions were obtained. The starch gel electrophorogram indicated that the fastest moving fraction (450 to 640 ml, Fractioll 1, 35 mg) contained horse Enzyme C and two unidentified proteins.
The slower moving fraction (660 to 1040 ml, Fraction II, 80 mg) was impure horse Enzyme B (on st,arch gel, in addition to Knzymc 11, a faint unidentified band was seen). After dialysis against, distilled water and freeze-drying, 50 mg of Fractioll I I \verc rcchromatographed on a DEAEcellulose column, 1 x 60 cm, with the same gradient as was used for the separation of crude horse carbonic anhydrases.
The mixing chamber cont,aincd 204 ml of 0.01 III Tris chloride buffer at pII 8. The flop rale t,o the mixing chamber was set at, 8 ml per hour and to the column at 25 ml per hour.
The major peak (146 to 175 ml) was starch gel electrophoretically pure horse Enzyme I< (31 mg).
For bovine carbonic anhydrascs, 800 mg of crude commercial enzyme (Worthington) was separated into three fractions on a DEAE-cellulose column (4 x 40 cm) previously equilibrated wit,h 0.01 M Tris chloride buffer at l)H 8. Elution was achieved by a linear gradient from 0.01 lo 0.1 nr at pII 8 with a flow rate of' 260 ml per hour. The volume of the buffer in the reservoir was I liter. The first, major fract;ion (1010 to 1300 ml, Fraction I, 350 mg) was starch gel electrophorctically pure bovine Enzyme B; the second fraction (1387 to 1,586 ml, 70 mg) was impure Enzyme A. At t,his point the gradient was replaced by 0.1 M Tris chloride at pH 8 to elute a third fraction (1820 to 2028 ml, 61 mg) which was still euzymatically active. On starch gel, however, this gave three bands: two corresponded to Enzymes B and A and t,he third one migrated between the two.
Zinc-jree Humun Carbonic Anhydrase C-This was prepared according to Lindskog (8) at pH 5.5. *\i"ter 12 days of dialysis against o-phenanthroline followed by dialysis against distilled water, t,he supernatant was separated from the denatured protein by centrifugation and freeze-dried. Zinc analyses were carried out by atomic absorption spectroscopy (Perkin Elmer, model 303).
Alkylated Human B and Bovine B Carbonic Anhydrases-Carboxymethylated human I':nzymc 13 was prepared by a slight modification of the method of Wlrit,ney, Nyman, and Malstrijm (2). The enzyme was reacted with a 60.fold molar excess (I .8 X lop2 11) of bromoacelnte in 0.1 M Tris sulfate buffer at pH 7.6 for 96 hours at 24". Alkylated bovine Enzyme B was prepared and purified by DEAE-cellulose chrolll:ltogr:11)113T as reportctl previously (3).
Inhibifors and Reagents-Bromoacetazolamide was prepared as reported (3). Commercial bromoacct,ic acid (Eastman) was redistilled under reduced pressure before use. l-140-l~romoacetic acid was purchased from New England Nuclear and diluted with freshly distilled bromoacetic acid to obtain a specific activity of 68 PC1 per mmole. Urea was recrystallized from ethanol, p-nitrophenyl acetate (Eastman) twice from acetonewater.
Dimethylformamide was redistilled under reduced prcssure. DEAE-cellulose (high capacity) was purchased from Bio-Itad and DEAE-Sephadex A-50 frorn Pharmacia (Canada). All other chemicals were reagent grade and were used without further purification.
Enzyme Assays--CO2 hydr&on activit>y was measured by the Wilbur-Anderson method at> 4". Specific :&ivit,y units were defined by the equation given earlier (9). Est,erase activity was determined with p-nitrophenyl acetate as substrate (lo), following the optical density change at' 348 rnp (Zeiss PM& II or Gilford recording spectrophotometer) (5). The reaction cuvette contained 0.6 X 1O-6 M to 1 X lop6 M enzyme plus 4 X lo-" M p-nitrophenyl acetate in 0.025 M Tris sulfate buffer plus 0.8% acetone at pH 7.6 and 24".
Determination of Inhibition Constants-The dissociation COW stants (K,) for reversible inhibition of bromoacelate and bromoacetazolamide were determined by the method of Dixon (II) by esterase activity, assuming that the inhibition is noncompetitive.
Starch Gel ElectrophoresisThis was done by reported procedures. For the human and bovine enzymes, l)H 8.9 (9) and, for the horse enzyme, pH 8.6 (7) were used. Amino 11 cid flnalysis-Protein samples were cvacuatcd, t,reated with 6 N HCl, sealed, and hydrolyzed for 22 hours at 110". Since cystine overlaps His(3-Cm)2 the modified human and horse enzymes were oxidized prior to acidic hydrolysis (12) with performic acid 30.fold in excess of that required to oxidize the cysteine aud methionine in the modified proteins.
Amino acid analysis of the hydrolysates was performed by the method of Spackman, Stein, and Moore (13). A synthetic mixture of carboxymethyl histidines (14) was used to identify the newly formed peaks in the hydrolysates.
The integration constant, given by Crestfield, Stein, and Moore (14) was used for the quantitative determination of His(B-Cm) and His(l-Cm). Molar ratios of amino acids were calculated on the basis of glycine equal to 20 for bovine B, 16 for human B, 22 for human c', and 23 for horse B carbonic anhydrases.
Determination of Radioactivity-All of the counting was carried out in a Packard Tri-Carb liquid scintillat)ion counter (model 3375). Protein samples were finely suspended in 0.1 ml of dimethylformamide and then dissolved in 1 ml of Hyamine. Absolute colmt rates were determined by either the internal or external standard method.
The preparations of pure bovine K, human B, human C, and horse B carbonic anhydrases described in this paper were obtained by Jnodificat,ions of kJlowil procedures. This was necessary because some of' the published methods did not give in our hands electrophoretically pure enzyme samples. The modified procedures have been repeated in two laboratories several times and have provctl highly reproducible.
The specific catalytic activities expressed in \\'ilbur-Anderson unit's were 40,000 to 50,000 for bovine II, 8,000 to 9,000 for human 13, 35,000 to 45,000 for human (:, and 2,000 for horse 1< enzymes. Each enzyme migrated as a single band OJL starch gel, and its amino acid content agreed well with rrlxxtetl data (5,7,18).
Sine-Iree Human Carbonic ;I nhydrase C-Our zinc-free human c preparation contained less than 0.05 g atom of zinc per mole alld was almost, completely inactive.
Upon the addition of 1 g atom of zinc the enzyme regained its original activity.
l<ovine Enzyme 11 alkylatcd 1% ith bromoacetazolamide was isolated as a chromatographically pure sample which on starch gel at $1 X.9 migrated as a single band toward the anode ahead of the native enzyme.
The COZ hydration property of the two slkylal cd enzymes could not be determined with sufficient accuracy by i he '\Vitbur-&demon method to tell whether they possessed any residual activity.
Reversible Inhibition-h Table 1 are S~OWJJ the inhibitiorr const,auts of bromoacetate and bromoacet,azolamide at pII 7.6 for the four enzymes. For bromoacetate this constant, varies between 9 X 1OP M for human Enzyme 1s (which possesses low catalytic activity) and 2.7 X 10-l M for bovine Enzyme I I (which possesses high catalytic activity).
This variat,ion corresponds to a 30.fold difference in the binding of bromoacetate by these two enzymes.
It is interesting to note that the binding of this reagent to bovine 13 aud hiJlaJ1 (' enzymes, bot,h having high catalytic activity, is almost the same, whereas t,hc difference in binding between the two low activity forms (human 13 and horse I%) is approximat,ely B-fold. The constants for bromoacctazolamide arc of the same order of magnitude.
The variation between the two extreme values is 3-fold. Under identical conditions ncithcr human 13 nor horse B enzymes (low activity) reacted at all.
In Table III it can be seen that bromoacetate had an opposite effect. Although a 20.fold molar excess of this reagent' at pH 7.6 brings about 50 t'o 60yC inactivation with human 13 and horse IS enzymes, with the simultaneous formation of 3-cnrbox~~methyl histidine, neither bovine I{ 11or human C enzymes showed any significant inactivation or carboxTmethylation.
If the concentration of bromoacet,azolamide relative to Ihc enzyme was increased to 20.fold (Table II), all four carbonic anhydrases showed inhibition and alkylation, although the quantitative aspects of the reaction were significantly diffcrcnt. The high catalytic activity forms yielded an almost inact,ive enzyme and approximately 0.8 eq of alkylated histidine; the low activity forms revealed only 0.30 to 0.46 eq of alkylated histidine and a corresponding inactivation.
Sinlilarly, >I 60.fold   Enzyme, 9 mg, was dissolved in 1 ml of 0.1 M Tris sulfate buffer at pH 7.6 and 2.5 mg of bromoacetate (1.8 X 1O-2 M, 60-fold molar excess) were added. At different time intervals an aliquot was withdrawn and diluted 50 times and its e&erase activit.y was determined.
When 50% inactivat,ion was attained the reaction mixture was dialyzed against distilled water, freeze-dried, and analyzed.
The calculated half-t'imes were derived by assuming an inverse relationship of half-time to enzyme-inhibitor complex concentration.
The observed value for human Enzyme B was taken as a reference point for these calculations. molar excess of bromoaeebate almost complet,ely inactivated human 13 and horse 1% enzymes with the formation of 0.8 to 0.9 eq of His(3-Cm) ; under identical conditions bovine 13 and human C enzymes showed 10 to 11 y0 inactivation with a parallel formation of less than 0.1 eq of His(3-Cm) (Table III).
In selected cases, the react'ions with bromoacetazolamide and bromoacetate were carried out at pH X.7. These results, which are also summarized in Tables II and III,  where (EI) is the fraction of total enzyme that formed a reversible complex, Et is the total enzyme concentration excluding the portion that has become inactivated, RI is the inhibition constant, and (I) is the amount of inhibitor.
The results of this calculation are listed also in Table I.
Irreversible Inactivation at Near Equal Concentrations of Enzyme-Inhibitor Complex-Since t,he concentrations of the enzymebromoacetate complexes differ markedly for the two groups of enzymes (Table I), it was important to compare their inactivations at near equal concentrations of the enzyme-inhibitor compies. The results in Table IV indicate that under t'hese conditions human B and horse 13 enzymes formed 0.22 and 0.16 eq of His(3-Cm), respectively, while bovine 73 and human C enzymes yielded only trace amounts.
If the difference in the Kr values would account for the observed difference in the rate of alkylat,ion of the two groups of enzymes, rat,her than the react.ivity of the active site histidine, equal amounts of His@Cm) would have been formed from all four enzymes.
The observed values of alkylation tabulated in Table II  A kinetic study with bromoacetazolamide to determine the half-time of the reaction was unsuccessful.
A recently revealed pseudoirreversible binding of acetazolamide and bromoacetazolamide made it impossible to separate the reversible and irreversible binding of these reagents by the usual dilution techniques. Further details of this phenomenon will be given elsewhere.
Speci$city of Inactivation- Table  VI indicates that human C carbonic anhydrase had lost approximately 81% of its activity when it reacted with a 20.fold molar excess of bromoacetazolamide at pH 7.6. As a result of this inactivation, 0.7 eq of histidine became alkylated at the a-nitrogen position.
Under identical conditions the metal-free apoenzyme yielded only 0.1 eq of alkylated histidine.
When 1 eq of Zn++ was added to this metalfree enzyme preparat'ion 92% of its usual activity was regained. It is well documented that Zn++ . IS an essential part of all of the known mammalian erythrocyte carbonic anhydrases and is necessary for the binding of sulfonamide inhibitors.
The lack of any significant reaction of bromoacetazolamide with the zincfree enzyme is a good indication that it is first bound to the active to 7 ml and the e&erase activity was determinrd. site and then forms a covalent bond with a hhtidine at or close. to the active site. Furthermore, if the native enzyme is reacted under identical conditions, but in the presence of 6 M urea, the formation of alkylated histidine drops to 0.2 eq, confirming the specificity of the reaction. When 1.2.fold molar excess of the reagent was used (Table II) a good correlation wab found between irreversible inactivation and His(3-Cm) formation. The' mere fact that a 20.fold molar excess of bromoaceta.te did not bring about any significant inactivation and alkylation in 24 hours (Table III) can also be considered as evidence for t,he, specificity of the reaction of human C enzyme with bromoacctazolamide.
In order to determine the active site-directed nature of the irreversible inactivation of horse carbonic anhydrase 13 with bromoacetat,e a different, course was followed.
The rate ol' the, reaction was determined at 9 1llM and at 45 my concentrations of the reagent (Fig. 1) and a comparison of the estimated halftimes indicated a "rate sat,uratioll effect" (I 9j. The rate of the reaction increased by a factor of only 3.3 and not by a factorof 5, which would be expected if the covalent bond formed outside the active site by a bimolecular mechanism. From Equation 1 the calculated value for the rate sat,uration effect, is 3.2, which agrees well with the observed value. The good xtoichiometry of carboxymethylation, inhibition, and 14C-bromoacet:L1 r illcorporation substantiates the view t,hat the rraction occ2rurec-l via the active site (Table III).
The specificity of the reactions of human carbonic anhydrase B with bromoacetazolamide and of bovine Enzyme I( w4h brdmoacetate was also studied. Carboxymethylated human Enzyme B was reacted with 20-fold molar excess of bromoacetazolamide and it was found that the initial value of His(Y-Cm) (1 eq) did not increase significantly (1.07 eq), indicating that, the His(S-Cm) formed from human Enzyme 1% with bromoacetazolamide occurs at the active site. By contrast, bromoacetate at 1.8 X 1 OP M concentration (60-fold molar excess) did react with monoalkylated bovine Enzyme R. The initial value of IIis(3-Cm) (0.9 eq) increased to 1.2 eq, indicating that' at t,his high concentration bromoacetate reacts with a histidine or histidines outside the active site. The nonspecific nature of the reaction is also indicated by the significant nonequivalence of IIis(3-Cm) formation and inactivation, which is shown in Table V. Rorine Enzyme 13 when it was 5001, inactivaled with bromoacetate contained 0.83 eq of His(3-Cm) and 0.15 cq of His(l-Cm).
The 0.33 eq excess of His(3-Cm) relative to the inactivation almost completely accounts for the His  formed from the rnonoalkylat,ed enzyme with bromoacet,at,e.
A similar nonspecificity of the reaction of human Enzyme C with bromoacetate is seen. in Table V. DISCUSSION Carbonic anhydrase is present in mammalian erythrocgles in multiple molecular forms. Most of the species studied contain at least two major isoenzymes, one possessing specific catalytic activity 4 to 10 times higher than the other (7,9,20). From bovine erythrocytes two major forms have been isolated; both possess the same high specific cat,alytic activity and it is believed that they are conformers (18). It seemed reasonable to assume that the differences in catalytic activity of the various enzymes reside in the chemical structure of t,heir active sites. Tn t,he last 2 years three laboratories have reported that, t,here is a chemically react,ive histidine at or close to the active site of human Enzyme 13 and bovine Enzyme R (1 -3). Although at present there are serious doubts about the significance of this histidine in catalysis (2,21), we reasoned that, a l~obe of its chemical reactivity might give some clue to the differences in the act,ive sites of carbonic anhydrase isoenzymes.
This reasoning found some support from the foIlowing observations. Whitney et al. (2) as well as Bradbury (21) reported that haloacetates rapidly inactivate the low catalytic activity form of human isoenzymes (human U) with the labeling of 1 eq of histidine at the 3-nitrogen position.
Whitney et al. (2) also reported briefly that iu one experiment human Enzyme C, which possesses high catalytic activity, did not react with bromoacetate.
Concurrent with this investigation we (3) found that bovine Enzyme B, which like human C possesses high catalytic activity, does not react with haloacetates even when 20.fold molar excess of t'his reagent is used, but it does react with bromoacetazolamide.
We demonstrated that this reaction occurs at or close to t,he active site of this enzyme with the alkylation of 1 eq of histidine at the 3-nit,rogen position.
In order to study further the chemical reactivity of the active site histidine of the two types of carbonic anhydrase isoenzyrnes, bromoacetate and bromoacetazolamide were reacted with bovine B and human C (high activity forms) and human I< and horse B (low activity forms).
We selected these enzymes as typical representatives of the two types of carbonic anhydrase isoenzymes since they are the most readily available and the best characterized. The data presented in t,his paper show conclusively that a difference exists between the chemical reactivity of a histidine of the high and the low catalytic activity forms of mammalian carbonic anhydrases when tested with bromoacetate and bromoacetazolamide, respectively.
In order to make these differences meaningful it was necessary to prove that the histidines that reacted with the two reagents are located at or close to the active site. This requirement seems to have been fulfilled in the case of the reaction of haloacetate and human carbonic anhydrase B (I, 2, 21), as well as in the case of the reaction of bromoacetazolamide and bovine Enzyme 15 (3). Based on the evidence given in this paper we believe that the reactions of bromoacetat,e with horse Enzyme B and of bromoacetazolamide with human B and human C enzymes also occur via the active site.3 By contrast, t,he reaction of bovine B and human C enzymes with bromoacetate was nonspecific at the concentration used, while the low activit'y human B enzyme reacted specifically with this reagent.
Our results provide a basis for the following deduction.
In the case of high catalytic activity enzymes, although bromoacetate reacts with a histidine that is at or close to the active site, it is also capable of reacting with a histidine (or histidines) outside the active sit,e. The Iat,ter reaction, however, does not affect enzyme activity.
The results also indicate that bromoacetazolamide is a better reagent for labeling the active site of carbonic anhydrases possessing high catalytic activity.
Although horse Enzyme 12 relatively rapidly, the inactivation is not conplete. The enzyme reached 96% inactivation in about 5 hours, and this value did not increase further even after 24 hours. Moreover, a chromatographically purified alkylated bovine Enzyme B, migrating as a single band on starch gel electrophoresis, retains 4 to 5% of its esterase activity.
A smail residual esterase activity was found previously by Whitney et al. (2) and I{radbury (21) for carboxymethylated human Enzyme B. Our results suggest that a residual esterase activity is a general phenorncnon of mammalian erythrocyte carbonic anhydrases, after alkylat iou at the active site histidine.
In the reactions described here a combination must occur initially between the enzyme active site and the inhibitor, which then reacts with a properly oriented histidine.
The rate of inactivat'ion would be controlled by the conccnt&ion of the cnzyme-inhibitor complex, which in turn depends on both the concentration and the inhibition constant of the inhibitor. A considerable difference was found in the rate of inactivation and alkylation of one particular isoenzyme with bromoacetate and bromoacetazolamide.
Human carbonic anhydrase 1% with bromoacetate at 1.8 X 1OP' M concentration, when 66.6% of the enzyme was reversibly complexed, yielded 1 eq of His(S-Cm) ( Table V), while with bromoacetazolamide at 6.6 x IOP M concentration, when 96% of the enzyme was present in combined forms, it yielded only 0.31 eq of His(3-Cm) (Table II). This indicates that bromoacetate reacts with human Enzyme 1% at least 3 times faster than bromoacetazolamide, although the concentration of enzyme-bromoacetate complex is only two-thirds the concentration of the enzyme-bromoacetazolamide complex. A direct comparison of the rate of alkylation can, however, be misleading, for the relative electrophilicity of the two reagents is not known at this time.
A comparison of the rates of alkylation of the high catalytic activity isoenzymcs by the two reagents would appear to be difficult since at a concentration of bromoacetate that yields measurable amount's of His&Cm) the reaction is nonspecific for the high activity enzymes.
A striking difference was also found between the two types of isoenzymes in the rate of alkylation by either reagent when compared at equal molar concentrations of the reagent,s (Tables II  and III) and at near equal equilibrium concentratjions of enzymeinhibitor complex (Table IV). This is a clear indication that a difference in chemical reactivity, and not in binding, causes t,he difference in rates of alkylation.
As Tables II, III, and V indicate, the remarkable isoenzyme specificity of the inhibitor holds only if the inhibitor concentration is kept low. At higher inhibitor concent,rations both reagents react with the four enzymes, although the rates of the reactions are significantly different.
This observation indicates that an almost complete lack of the reaction at lower concentrations cannot be due to the lack of a sterically available histidine at or around the active site.
Bradbury (21) found a pK of 5.8 for the reactive histidine of human carbonic anhydrase B. The pK of the reactive hi&dine of bovine Enzyme B is not known at the present time. We have found (3) that the inactivation and alkylation of the act,ive site histidine of this enzyme reach a maximum around $1 8. Although these results might indicate a difference in the pK of the reactive histidines of the two enzymes, it is conceivable that at, pI1 8.7, at which pH the differences in the reaction are equally significant, the histidine of either enzyme will be present largely in unprot,onated form and can react as a nucleophile.