Carbonic anhydrase activators: L-Adrenaline plugs the active site entrance of isozyme II, activating better isoforms I, IV, VA, VII, and XIV

The activation of the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1) with L-adrenaline and histamine has been investigated by kinetic and X-ray crystallographic studies. L-Adrenaline behaves as a potent activator of isozyme CA I (activation constant of 90 nM), being a much weaker activator of isozyme CA II (activation constant of 96 microM). Isoforms CA IV, VA, VII, and XIV were activated by L-adrenaline with K(A)s in the range of 36-63 microM. The X-ray crystal structure of the CA II-L-adrenaline adduct revealed that the activator plugs the entrance of the active site cavity, obstructing it almost completely.     

Carbonic anhydrase inhibitors: the X-ray crystal structure of ethoxzolamide complexed to human isoform II reveals the importance of thr200 and gln92 for obtaining tight-binding inhibitors

Ethoxzolamide, an almost forgotten inhibitor of the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1), is the only classical inhibitor whose structure in adduct with any isoform was not reported yet. We report here the inhibition data of this molecule with the 12 catalytically active mammalian isozymes (CA I-CA XIV) and the X-ray crystal structure with the cytosolic, ubiquitous isoform CA II. These data are presumably useful for the design of novel CA inhibitors, targeting various CA isozymes, considering that ethoxzolamide was already the lead molecule to obtain the second generation inhibitors, dorzolamide and brinzolamide, clinically used antiglaucoma agents with topical action, as well as various other investigational agents.     

Carbonic anhydrase inhibitors: the very weak inhibitors dithiothreitol, beta-mercaptoethanol, tris(carboxyethyl)phosphine and threitol interfere with the binding of sulfonamides to isozymes II and IX

The inhibition of the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1) with dithiothreitol, 2-mercaptoethanol, tris(carboxyethyl)phosphine (reducing agent frequently added to enzyme assay buffers) and threitol has been investigated. The agents were very weak inhibitors of isozymes CA II and CA IX, but unexpectedly, strongly influenced the binding of the low nanomolar sulfonamide inhibitor acetazolamide (5-acetamido-1,3,4-thiadiazole-2-sulfonamide). Acetazolamide affinity for all investigated CAs diminished orders of magnitude with increasing concentrations of these agents in the assay system. DTT and similar derivatives should not be added to the assay buffers used in monitoring CA activity/inhibition, as they lead to under-estimation of the binding constants, by a mechanism probably involving the formation of ternary complexes.     

Reconsidering anion inhibitors in the general context of drug design studies of modulators of activity of the classical enzyme carbonic anhydrase

Inorganic anions inhibit the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1) generally by coordinating to the active site metal ion. Cyanate was reported as a non-coordinating CA inhibitor but those erroneous results were subsequently corrected by another group. We review the anion CA inhibitors (CAIs) in the more general context of drug design studies and the discovery of a large number of inhibitor classes and inhibition mechanisms, including zinc binders (sulphonamides and isosteres, dithiocabamates and isosteres, thiols, selenols, benzoxaboroles, ninhydrins, etc.); inhibitors anchoring to the zinc-coordinated water molecule (phenols, polyamines, sulfocoumarins, thioxocoumarins, catechols); CAIs occluding the entrance to the active site (coumarins and derivatives, lacosamide), as well as compounds that bind outside the active site. All these new chemotypes integrated with a general procedure for obtaining isoform-selective compounds (the tail approach) has resulted, through the guidance of rigorous X-ray crystallography experiments, in the development of highly selective CAIs for all human CA isoforms with many pharmacological applications.     

Carbonic anhydrase inhibitors: Benzenesulfonamides incorporating cyanoacrylamide moieties are low nanomolar/subnanomolar inhibitors of the tumor-associated isoforms IX and XII

A series of benzenesulfonamides incorporating cyanoacrylamide moieties (tyrphostine analogues) have been obtained by reaction of sulfanilamide with ethylcyanoacetate followed by condensation with aromatic/heterocyclic aldehydes, isothiocyanates or diazonium salts. The new compounds have been investigated as inhibitors of the metalloenzyme carbonic anhydrase (CA, EC 4. 2.1.1), and more specifically against the cytosolic human (h) isoforms hCA I and II, as well as the transmembrane, tumor-associated ones CA IX and XII, which are validated antitumor targets. Most of the new benzenesulfonamides were low nanomolar or subnanomolar CA IX/XII inhibitors whereas they were less effective as inhibitors of CA I and II. The structure–activity relationship for this class of effective CA inhibitors is also discussed. Generally, electron donating groups in the starting aldehyde reagent favored CA IX and XII inhibition, whereas halogeno, methoxy and dimethylamino moieties led to very potent CA XII inhibitors.     

Carbonic anhydrase inhibitors: Inhibition of the new membrane-associated isoform XV with phenols

Inhibition of the newest isoform of the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1), CA XV, with a series of phenols was investigated. Murine CA XV showed an inhibition profile by phenols distinct of those of the cytosolic human isoforms CA I and II. Phenol and some of its 2-, 3-, and 4-substituted derivatives incorporating hydroxy, fluoro, carboxy, and acetamido moieties were effective CA XV inhibitors, with inhibition constants in the range of 7.20-11.30 microM, whereas compounds incorporating 4-amino-, 4-cyano, or 3-hydroxy groups were less effective (K(I)s of 335-434 microM). The best phenol inhibitor was clioquinol (K(I) of 2.33 microM). Phenols show a different inhibition mechanism as compared to sulfonamides and their isosteres, and may lead to the design of compounds with selectivity for inhibiting different CA isozymes with medicinal chemistry applications.     

Carbonic anhydrase inhibitors: interactions of phenols with the 12 catalytically active mammalian isoforms (CA I-XIV)

The inhibition of the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1) with three phenols was investigated. Phenol was an effective CA I-IV, IX, XII and XIV inhibitor (K(I)s of 2.7-11.5 microM) and a less effective one against the other isoforms, CA VA, VB, VI, VII, and XIII (K(I)s of 208-710 micraoM). 3,5-Difluorophenol was an effective inhibitor of CA III, IV, IX, and XIV (K(I)s of 0.71-10.7 microM) being a weaker one for CA I, II, VA, VB, VI, VII, XII, and XIII (K(I)s of 33.9-163 microM). Clioquinol (5-chloro-7-iodo-8-quinolinol) was the best phenol inhibitor against all isozymes, with inhibition constants in the range of 3.3-16.0 microM. These data prove that the phenol OH moiety can be considered as a new 'zinc-water binding group' for the design of CA inhibitors possessing a different inhibition mechanism as compared to the classical sulfonamide inhibitors that bind the metal ion within the active site cavity.     

Acipimox inhibits human carbonic anhydrases

Acipimox, a nicotinic acid derivative in clinical use for the treatment of hyperlipidaemia, incorporates a free carboxylic acid and an N-oxide moiety, functionalities known to interact with the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1) and inhibit its activity. Herein we report that acipimox acts as a low micromolar CA inhibitor (CAI) against most human (h) isoforms possessing catalytic activity, hCA I – XIV. By using computational techniques (docking and molecular dynamics simulations), we propose that acipimox coordinates through its carboxylate group to the zinc ion from the enzyme active site cavity, whereas the N-oxide group is hydrogen-bonded to the proton shuttle His residue in some isoforms (hCA I) or to active site Thr or Gln residues in other isoforms (hCA II, III, IV, VII, etc). As some CA isoforms are involved in lipogenesis, these data may be useful for the design of more effective CAIs with antiobesity activity.     

An inhibitor-like binding mode of a carbonic anhydrase activator within the active site of isoform II

The 2,4,6-trimethylpyridinium derivative of histamine is an effective activator of the zinc enzyme carbonic anhydrase (CA, EC 4.2.1.1). However, unlike other CA activators, which bind at the entrance of the active site cavity, an X-ray crystal structure of hCA II in complex with the 1-[2-(1H-imidazol-4-yl)-ethyl]-2,4,6-trimethylpyridinium salt evidenced a binding mode never observed before either for activators or inhibitors of this enzyme, with the 2,4,6-trimethylpyridinium ring pointing towards the metal ion deep within the enzyme cavity, and several strong hydrophobic interactions stabilizing the adduct. Indeed, incubation of the activator with the enzyme for several days leads to potent inhibitory effects. This is the first example of a CA activator which after a longer contact with the enzyme behaves as an inhibitor.     

Interaction of the unique competitive inhibitor imidazole and related compounds with the active site metal of carbonic anhydrase: linkage between pH effects on the inhibitor binding affinity and pH effects on the visible spectra of inhibitor complexes with the cobalt-substituted enzyme

Previous studies on the interaction of carbonic anhydrase (CA) with the unique CO2 competitive inhibitor imidazole and related compounds were all interpreted as showing that an ionizable water ligand on the metal of this zinc metalloenzyme is not displaced by inhibitor binding. Internal inconsistencies in the pH dependence of binding and the pH dependence of the visible spectra of complexes with cobalt-substituted enzyme prompted us to reinvestigate this binding. Visible spectroscopy was used to measure the binding of imidazole and 1,2,4-triazole to Co(II)-substituted human CA I and active site carboxymethylated human CA I (CmCA I) and the binding of 1,2,4-triazole to bovine CoIICA II. The limiting visible spectra for these enzyme-inhibitor adducts were also computed and examined for pH dependence. It was shown that the pKa of visible spectral changes can be independently predicted from studies on the pH dependence of binding. After consideration of possible contributions from effects of His-200 ionization in CA I and CmCA I, and His-64 in CA II, the pH effects on binding affinity and spectra were found to be of the correct magnitude to establish linkage between binding and an ionization. It was also shown, however, that pH effects on binding and spectra cannot distinguish whether neutral imidazole binds to both ionization forms of the enzyme (Zn-OH2 and Zn-OH) or whether neutral imidazole and its anion both bind to only the acid form of the enzyme, presumably after displacing the water. These findings have implications to the crystallographic interpretations on the imidazole-enzyme complex and to the catalytic mechanism of CO2 hydration.     

Carbonic anhydrase inhibitors. Interaction of isozymes I, II, IV, V, and IX with phosphates, carbamoyl phosphate, and the phosphonate antiviral drug foscarnet

A detailed inhibition study of five carbonic anhydrase (CA, EC 4.2.1.1) isozymes with inorganic phosphates, carbamoyl phosphate, the antiviral phosphonate foscarnet as well as formate is reported. The cytosolic isozyme hCA I was weakly inhibited by neutral phosphate, strongly inhibited by carbamoyl phosphate (K(I) of 9.4 microM), and activated by hydrogen- and dihydrogenphosphate, foscarnet and formate (best activator foscarnet, K(A)=12 microM). The cytosolic isozyme hCA II was weakly inhibited by all the investigated anions, with carbamoyl phosphate showing a K(I) of 0.31 mM. The membrane-associated isozyme hCA IV was the most sensitive to inhibition by phosphates/phosphonates, showing a K(I) of 84 nM for PO(4)(3-), of 9.8 microM for HPO(4)(2-), and of 9.9 microM for carbamoyl phosphate. Foscarnet was the best inhibitor of this isozyme (K(I) of 0.82 mM) highly abundant in the kidneys, which may explain some of the renal side effects of the drug. The mitochondrial isozyme hCA V was weakly inhibited by all phosphates/phosphonates, except carbamoyl phosphate, which showed a K(I) of 8.5 microM. Thus, CA V cannot be the isozyme involved in the carbamoyl phosphate synthetase I biosynthetic reaction, as hypothesized earlier. Furthermore, the relative resistance of CA V to inhibition by inorganic phosphates suggests an evolutionary adaptation of this mitochondrial isozyme to the presence of high concentrations of such anions in these energy-converting organelles, where high amounts of ATP are produced by ATP synthetase, from ADP and inorganic phosphates. The transmembrane, tumor-associated isozyme hCA IX was on the other hand slightly inhibited by all these anions.     

Inhibition profiling of human carbonic anhydrase II by high-throughput screening of structurally diverse, biologically active compounds

Human carbonic anhydrase II (CA II), a zinc metalloenzyme, was screened against 960 structurally diverse, biologically active small molecules. The assay monitored CA II esterase activity against the substrate 4-nitrophenyl acetate in a format allowing high-throughput screening. The assay proved to be robust and reproducible with a hit rate of approximately 2%. Potential hits were further characterized by determining their IC(50) and K(d) values and tested for nonspecific, promiscuous inhibition. Three known sulfonamide CA inhibitors were identified: acetazolamide, methazolamide, and celecoxib. Other hits were also found, including diuretics and antibiotics not previously identified as CA inhibitors, for example, furosemide and halazone. These results confirm that many sulfonamide drugs have CA inhibitory properties but also that not all sulfonamides are CA inhibitors. Thus many, but not all, sulfonamide drugs appear to interact with CA II and may target other CA isozymes. The screen also yielded several novel classes of nonsulfonamide inhibitors, including merbromin, thioxolone, and tannic acid. Although these compounds may function by some nonspecific mechanism (merbromin and tannic acid), at least 1 (thioxolone) appears to represent a genuine CA inhibitor. Thus, this study yielded a number of potentially new classes of CA inhibitors and preliminary experiments to characterize their mechanism of action.     

Inhibition of carbonic anhydrases I and II by N-unsubstituted carbamate esters.

We previously showed that the zinc metalloenzyme carbonic anhydrases (CA I and II isozymes) bind "neutral" amides and related compounds as anions through coordination of their deprotonated amide nitrogen to the active site zinc (Rogers, J. I., Mukherjee, J., and Khalifah, R. G. (1987) Biochemistry 26, 5672-5679). Urethan, the ethyl carbamate ester, was among such compounds. The present study was designed to test whether other N-unsubstituted carbamate esters of pharmacological interest (as sedatives, hypnotics, anxiolytics, and skeletal muscle relaxants) were capable of binding to CA in the same manner. We studied the interaction of human CA I and II with urethan, phenyl carbamate, ethinamate, meprobamate, and methocarbamol. Phenyl carbamate studies were greatly complicated by its uncatalyzed hydrolysis via an elimination mechanism to form cyanate, a powerful CA inhibitor. In general, the compounds display: 1) slow on-off inhibition binding kinetics in the seconds range, 2) maximal inhibitor affinity at alkaline pH, and 3) characteristic three-band visible spectra of their complexes with cobalt-substituted CA I. These properties are shared with the previously studied amide inhibitors and are taken as evidence that the deprotonated carbamate nitrogen coordinates to the active site metal ion. CA I appeared to bind carbamate esters more tightly than CA II, an unusual 1000-fold selectivity being seen in the case of methocarbamol. The inhibition by these drugs is not sufficiently strong to implicate CA I and II in their mechanism of action. However, it does suggest the possible existence of previously unsuspected similarities between binding to CA and to their physiological receptors or targets, particularly the involvement of zinc.     

Carbonic anhydrase activators: kinetic and X-ray crystallographic study for the interaction of D- and L-tryptophan with the mammalian isoforms I-XIV

An activation study of mammalian carbonic anhydrase (CA, EC 4.2.1.1) isoforms I-XIV with D- and L-tryptophan has been performed both by means of kinetic and X-ray crystallographic techniques. These compounds show a time dependent activity against isozyme CA II, with activation constants of 1.13 microM for L-Trp and 0.37 microM for D-Trp, respectively, after 24 h of incubation between enzyme and activator. The high resolution X-ray crystal structure of the hCA II-D-Trp adduct revealed the activator to bind in a totally unprecedented way to the enzyme active site as compared to histamine, L-/D-Phe, L-/D-His or L-adrenaline. D-Trp is anchored at the edge of the CA II active site entrance, strongly interacting with amino acid residues Asp130, Phe131 and Gly132 as well as with a loop of a second symmetry related protein molecule from the asymmetric unit, by means of hydrogen bonds and several weak van der Waals interactions involving Glu234, Gly235, Glu236 and Glu238. Thus, a second activator binding site (B) within the CA II cavity has been detected, where only D-Trp was shown so far to bind, in addition to the activator binding site A, in which histamine, L-/D-Phe, and L-/D-His are bound. These findings explain the strong affinity of D-Trp for CA II and may be useful for designing novel classes of CA activators by using this compound as lead molecule.     

Immobilized carbonic anhydrase: preparation,characteristics and biotechnological applications

Carbonic anhydrase (CA) is an essential metalloenzyme in living systems for accelerating the hydration and dehydration of carbon dioxide. CA-catalyzed reactions can be applied in vitro for capturing industrially emitted gaseous carbon dioxide in aqueous solutions. To facilitate this type of practical application, the immobilization of CA on or inside solid or soft support materials is of great importance because the immobilization of enzymes in general offers the opportunity for enzyme recycling or long-term use in bioreactors. Moreover, the thermal/storage stability and reactivity of immobilized CA can be modulated through the physicochemical nature and structural characteristics of the support material used. This review focuses on (i) immobilization methods which have been applied so far, (ii) some of the characteristic features of immobilized forms of CA, and (iii) biotechnological applications of immobilized CA. The applications described not only include the CA-assisted capturing and sequestration of carbon dioxide, but also the CA-supported bioelectrochemical conversion of CO₂ into organic molecules, and the detection of clinically important CA inhibitors. Furthermore, immobilized CA can be used in biomimetic materials synthesis involving cascade reactions, e.g. for bone regeneration based on calcium carbonate formation from urea with two consecutive reactions catalyzed by urease and CA.     

Combining the tail and the ring approaches for obtaining potent and isoform-selective carbonic anhydrase inhibitors: Solution and X-ray crystallographic studies

5-(3-Tosylureido)pyridine-2-sulfonamide and 4-tosylureido-benzenesulfonamide (ts-SA) only differ by the substitution of a CH by a nitrogen atom, but they have very different inhibitory properties against the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1). By means of X-ray crystallography on the human CA II adducts of the two compounds these differences have been rationalized. As all sulfonamides, the two compounds bind in deprotonated form to the Zn(II) ion from the enzyme active site and their organic scaffolds extend throughout the cavity, participating in many interactions with amino acid residues and water molecules. However the pyridine derivative undergoes a tilt of the heterocyclic ring compared to the benzene analog, which leads to a very different orientation of the two scaffolds when bound to the enzyme. This tilt also leads to a clash between a carbon atom from the pyridine ring of the first inhibitor and the OH moiety of Thr200, leading to less effective inhibitory properties of the pyridine versus the benzene sulfonamide derivative. Indeed, ts-SA is a promiscuous, low nanomolar inhibitor of 7 out of 10 human (h) CA isoforms, whereas the pyridine sulfonamide is a low nanomolar inhibitor only of the tumor-associated hCA IX and XII, being less effective against other 9 isoforms. Thus, a difference of one atom (N vs CH) in two isostructural sulfonamides leads to drastic differences of activity, phenomenon understood at the atomic level through the high resolution crystallographic structure and kinetic measurements reported in the paper. Combining the tail and the ring approaches in the same chemotype leads to isoform-selective, highly effective sulfonamide CA inhibitors.     

The active and the inactive plasminogen activator inhibitor from human endothelial cell conditioned medium are immunologically and functionally related to each other

In human endothelial cell conditioned medium a fast-acting inhibitor of tissue-type plasminogen activator and urokinase has been detected. Moreover, an inactive inhibitor of these plasminogen activators is present, that can be activated by denaturing agents such as sodium dodecyl sulphate (SDS). The mutual relationship between these inhibitors was studied. The fast-acting plasminogen activator inhibitor from human endothelial cell conditioned medium was purified in a complex with tissue-type plasminogen activator by immune adsorption, using an immobilized anti-tissue-type plasminogen activator antibody. With the complex as an antigen, specific antibodies were raised against this inhibitor in rabbits. The antiserum immunoreacted with both the inactive and the fast-acting plasminogen activator inhibitor. Endothelial cell conditioned medium (containing the inactive plasminogen activator inhibitor) was treated with SDS and the inhibitory activity that emerged was purified. The SDS-generated product formed complexes with tissue-type plasminogen activator with the same molecular mass as those formed with the fast-acting inhibitor. Moreover, the inhibitory activity generated by SDS treatment showed the same kinetic behaviour with tissue-type plasminogen activator as did the fast-acting inhibitor. These data show that the fast-acting and the inactive plasminogen activator inhibitor are immunologically and functionally related to each other, and probably represent different molecular forms of the same protein.     

Properties and intracellular localization of calpain activator protein

In this paper, we have further analyzed the properties of calpain activator (CA) in order to better define its physiological function. The activator shows a pH optimum approximately 7.8-8.0, independently of the nature of the buffer used. Although the maximal activity is observed with human acid-denatured globin, the effect of CA is detectable with other protein substrates, such as casein and insulin. A comparable activating effect is observed also with the synthetic substrate Succ-Leu-Tyr-AMC. The activatory effect has been evaluated in a reconstructed system, using plasma membrane Ca(2+)-ATPase as substrate. CA is localized in erythrocyte precursor cells on the inner surface of the plasma membrane in very high amount and its level profoundly decreases up to 10% of the original value when cells reach the terminal differentiated state.     

Autosomal recessive hyponatremia due to isolated salt wasting in sweat associated with a mutation in the active site of Carbonic Anhydrase 12

Genetic disorders of excessive salt loss from sweat glands have been observed in pseudohypoaldosteronism type I (PHA) and cystic fibrosis that result from mutations in genes encoding epithelial Na+ channel (ENaC) subunits and the transmembrane conductance regulator (CFTR), respectively. We identified a novel autosomal recessive form of isolated salt wasting in sweat, which leads to severe infantile hyponatremic dehydration. Three affected individuals from a small Bedouin clan presented with failure to thrive, hyponatremic dehydration and hyperkalemia with isolated sweat salt wasting. Using positional cloning, we identified the association of a Glu143Lys mutation in carbonic anhydrase 12 (CA12) with the disease. Carbonic anhydrase is a zinc metalloenzyme that catalyzes the reversible hydration of carbon dioxide to form a bicarbonate anion and a proton. Glu143 in CA12 is essential for zinc coordination in this metalloenzyme and lowering of the protein-metal affinity reduces its catalytic activity. This is the first presentation of an isolated loss of salt from sweat gland mimicking PHA, associated with a mutation in the CA12 gene not previously implicated in human disorders. Our data demonstrate the importance of bicarbonate anion and proton production on salt concentration in sweat and its significance for sodium homeostasis.     

Metal and metal-ATP interactions with brain and cardiac adenylate cyclases.

Metal (Me) and MeATP interactions with adenylate cyclases associated with rabbit ventricular particles and with a detergent-dispersed preparation from rat cerebellum have been studied. data were simulated to fit kinetic models in which an inhibitor (HATP or ATP) is added in constant proportion to the variable substrate (MeATP). The specific models considered were that the enzyme binds (a) MeATP as the substrate; (b) MeATP as the substrate and HATP or ATP as an inhibitor; (c) MeATP as the substrate and free Me as an activator; and (d) MeATP as the substrate, free Me as an activator, and HATP or ATP as an inhibitor. Both equilibrium-ordered and random (rapid equilibrium assumption) types of sequential kinetic models were considered. The various models were tested using cardiac particulate adenylate cyclase in the presence of either a phosphoenolpyruvate-pyruvate kinase or a creatine phosphate-creatine kinase ATP-regeneration system. Although the enzyme with either system appeared to bind Mg2+ as an activator, one or both ATP-regeneration systems also seemed to interact directly with adenylate cyclase, making clear interpretations difficult. With the phosphoenolpyruvate-pyruvate kinase system, kinetic patterns on double reciprocal plots were linear as a function of MgATP, but with creatine phosphate-creatine kinase, kinetic patterns were concave downward. The kinetic models were further tested using the detergent-dispersed cerebellar enzyme, a preparation with low adenosine triphosphatase activity and not requiring the addition of an ATP-regeneration system. Reciprocal plots were linear and intersecting as a function of either MeATP or Me (Me = Mg2+ or Mn2+), and secondary replots of slopes and intersecting as function of either MeATP or Me (Me = Mg2+ or Mn2+), and secondary replots of slopes and intercepts also were linear. These data indicate that the brain detergent-dispersed enzyme conforms to a bireactant, sequential mechanism where free cation is a required activator and free ATP is not a potent inhibitor.