ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 7, No. 2 Copyright © 1977, Institute for Clinical Science

The Role of Metals in Enzyme Activity* JAMES F. RIORDAN, Ph.D. Biophysics Research Laboratory, Department of Biological Chemistry, Harvard Medical School, and Division of Medical Biology, Veter Bent Brigham Hospital, Boston, MA 02115

ABSTRACT Metal ions play important roles in the biological function of many en­ zymes. The various modes of metal-protein interaction include metal-, ligand-, and enzyme-bridge complexes. Metals can serve as electron donors or acceptors, Lewis acids or structural regulators. Those that participate directly in the catalytic mechanism usually exhibit anomalous physicochemical characteristics reflecting their entatic state. Carboxypeptidase A, liver alcohol dehydrogenase, aspartate transcarbamoylase and al­ kaline phosphatase exemplify the different roles of metals in metalloenzymes while the nucleotide polymerases point to the essential role of zinc in maintaining normal growth and development. Introduction Certain metals have long been recog­ nized to have important biological func­ tions primarily as a consequence of nu­ tritional investigations.14,15,22 Thus, the absence of a specific, essential metal from the diet of an organism invariably leads to a deficiency state characterized by metabolic abnormalities with altered or retarded growth. Because such metals are usually present in tissues in very small amounts it was reasonable to sus­ pect that they might play a catalytic role, perhaps participating in enzymatic reac­ tions. The actual discovery of metalloenzymes, however, required the availability of accurate, sensitive, analytical method­ ology. As a consequence, the unequivo­ cal demonstration of a role for metals in enzyme action is of relatively recent vin­ tage. 119

At present, reliable measurements of small concentrations of metals present in tissues, cells, subcellular particles, body fluids and biomacromolecules can be performed by colorimetry, fluorimetry, polarography, emission spectrometry with spark, flame or plasma excitation sources, x-ray and atomic fluorescence, atomic absorption and neutron activation analysis, among other methods. Metals that have been detected by such techniques and currently known to be components of metalloenzymes include cobalt, copper, iron, manganese, molyb­ denum, nickel, selenium and zinc (table I). Aside from its role in vitamin B 12, cobalt, to date, has been found to be a * Supported by Grant-in-Aid GM-15003 from the National Institutes of Health of the Department of Health, Education and Welfare.

RIORDAN

12 0

TABLE I Metals Present in Naturally Occurring Metalloenzymes

Enzyme Function

Metal

Transcarboxylation Oxidoreduction Oxidoreduction Various Oxidoreduction Urease Peroxidase Various

Cobalt Copper Iron Manganese Molybdenum Nickel Selenium Zinc

component of but one enzyme, the biotin-dependent, zinc-containing oxaloacetate transcarboxylase of Pro­ pionibacterium shermanii.19 Copper is present in a large number of enzymes that catalyze oxidoreduction reactions such as tyrosinase, lysyl oxidase and cytochrome oxidase.20 Iron is also found primarily in enzymes that participate in oxidoreduction reactions; in addition, it plays a major role in oxygen transport.18 Manganese has been identified as a com­ ponent of pyruvate carboxylase from chicken liver and is present in Es­ cherichia coli superoxide dismutase.15 It also serves as an activator for many metal-activated enzymes; however, in most of these cases, magnesium and other divalent cations can fulfill the same func­ tion. Molybdenum is found most frequently in flavin-dependent enzymes, usually in conjunction with non-heme iron and acid-labile sulfur. A typical example is xanthine oxidase. A molybdoheme pro­ tein, sulfite oxidase, has been described TABLE

II

Currently Known Zinc Metalloenzymes

I n t e r n a t io n a l Union o f B io ch em istry System

E.C.l E.C.2 E.C.3 E.C.4 E.C.5 E.C.6

Oxidoreductases Transferases Hydrolases Lyases Isomerases Ligases

Number

7 8 23 19 1 1

Example

Alcohol dehydrogenases DN A polymerase Carboxypeptidase 6-ALA dehydratase Mannose-P isomerase Pyruvate carboxylase

as well as a molybdoferrodoxin, a com­ ponent of the nitrogenase system of nitrogen-fixing bacteria Azotobacter vinelandii and Clostridium pasteurianum-15 Nickel has been found to be present in urease 50 years after the enzyme was first crystallized.8 Selenium, which has been recognized as an essential nutrient for more than a dozen years, has recently been shown to be a component of an en­ zyme, glutathione peroxidase from eryth­ rocytes, the first example of a selenoenzyme.10 Zinc enzymes are among the most common of the metalloenzymes number­ ing over 70 and representing each of the six categories of enzymes designated by the International Union on Biochemistry (IUB) commission on enzymes (table II). Zinc metalloenzymes exhibit perhaps the greatest diversity both of catalytic func­ tion and of the role played by the metal atom.15,21,22’23,27'28,30 The metal is present in several dehydrogenases, aldolases, peptidases and phosphatases. Zinc en­ zymes participate in carbohydrate, lipid, protein and nucleic acid synthesis or de­ gradation. Several examples of zinc en­ zymes will be cited to illustrate the role metals in metalloenzymes and the gen­ eral importance of zinc to metabolism. Other metals such as sodium, potas­ sium, calcium and magnesium can also assist in the action of enzymes. With th­ ese, the mode of metal-enzyme interac­ tion is complex and often difficult to es­ tablish. Still other metals, such as chromium, vanadium and tin, have been shown to be either essential for growth in certain species or components of biologi­ cal macromolecules. However, their rela­ tionship to enzyme mechanisms has not been established. Enzymes affected by metal ions have been operationally defined as either metalloenzymes or metal-enzyme com­ plexes.28 A metalloenzyme contains a firmly bound, stoichiometric quantity of a

ROLE OF METALS IN ENZYME ACTIVITY metal as an integral part of the protein molecule. Removal of the metal by treat­ ment with chelating agents, for example, abolishes catalytic activity. In instances where the resultant apoenzyme is structually stable, restoration of the metal can regenerate full biological function. In contrast, metal-enzyme complexes are more loosely associated, the criterion for association being metal activation of catalysis. The metal ion is frequently not an integral part of the molecule when iso­ lated, and the enzyme may exhibit partial activity in the absence of the metal ion. Obviously, the difference between these two classes of metal-enzyme systems de­ pends on the magnitude of the metalprotein stability constant which can be a function of the metal atom as well as en­ vironmental conditions such as pH, buf­ fer and ionic strength. The metalloenzymes are better suited for elucidation of the metal protein interaction and for ex­ trapolating such information to the un­ derstanding of enzymic mechanisms. Moreover, they lend themselves more readily to a definite assessment of the physiological role of the metal. Metalenzyme complexes, however, have been of great theoretical importance in the un­ derstanding of catalytic phenomena and general mechanisms of catalysis by metalloenzymes. At present some 2,000 or more different enzymes have been isolated and charac­ terized and it has been estimated that at least one-third of these require or contain metal ions.14 In fact, the actual number of metal-dependent enzymes may be even greater for it has been pointed out that “There probably does not exist a single enzyme-catalyzed reaction in which either enzyme, substrate, product, or a combination of these is not influenced in a very direct and highly specific manner by the precise nature of the inorganic ions which surround and modify it” .17 This paper will be concerned primarily

121

with the role of metals in metalloen­ zymes and will not attempt to cover the interesting but voluminous literature de­ aling with metal-enzyme complexes. The Interaction of Metal Ions With Enzymes A number of schemes have been pres­ ented24 to describe the types of interac­ tions that can occur between metals, en­ zyme proteins and substrate (or inhibitor) (figure 1). The first of these represents an interaction between the substrate and the metal ion to form a complex that acts as the true substrate. Substrate-metal complexation can occur prior or subsequent to the formation of the enzyme-substrate complex. This type of behavior is typi­ cally observed with metal-activated en­ zymes. The second scheme indicates that the metal first binds to the protein and then serves as a site of interaction with substrate. In this instance, the metal can function either as a binding site, as a component of the catalytic apparatus of the enzyme or both. An example of both such possibilities is given by the role of zinc in carboxypeptidase A. Here the zinc atom is believed to interact with a peptide substrate via the carbonyl oxygen atom of its terminal

M + S = MS E + MS = EMS 2) E + M = EM } EM + S = EMS

3, E } ME

+ M = ME + S = MES

F i g u r e 1. Interactions between metal (M), en­ zyme (E) and substrate (S).

122

RIORDAN

peptide bond, i.e., the one that is suscep­ tible to hydrolysis. However, even though some kind of metal-substrate bond may be formed, the metal does not appear to be essential for peptide sub­ strate binding. Peptides bind to the metal-free apoenzyme as well as they do to the metalloenzyme, even though they are not hydrolyzed.2 Thus, for peptide substrates the metal presumably serves as a catalytic site. On the other hand, ester substrates of carboxypeptidase do not bind to the apoenzyme. It has been proposed that differences in the mode of interaction between substrate and metal account for the numerous kinetic differ­ ences that have been observed for car­ boxypeptidase acting on ester and pep­ tide substrates, respectively.2 A third scheme would have the metal acting at a site on the enzyme remote from the active site. In such instances, the metal could either serve to maintain protein structure and only influence catalytic activity indirectly or else it could regulate activity by stabilizing more or less active conformations of the protein. The latter situation would more likely pertain for metal-activated en­ zymes where the metal-protein interac­ tion is more readily controlled by manip­ ulation of the ambient metal ion concen­ tration. It should be emphasized that these schemes are not all mutually exclu­ sive and that some metalloenzymes are known to contain functionally different classes of metal ions. The Role of Metals in the Mechanism of Catalysis Iron, copper and molybdenum are most commonly encountered in enzymes catalyzing oxidoreduction reactions. In the majority of cases, the metal ion par­ ticipates directly in the electron transfer process and undergoes a cyclic change in oxidation state. Oftentimes the free metal is capable of catalysis by itself as with the

iron-promoted decomposition of hydro­ gen peroxide although in this case catalase is at least a million times more effective than iron alone. Thus, the pro­ tein component of a metalloenzyme con­ tributes many of the critical aspects of the catalytic mechanism. Zinc, on the other hand, does not undergo a change in oxidation state dur­ ing enzymatic catalysis even though it participates in oxidoreduction reactions, e.g., as a component of alcohol dehydro­ genase. The zinc cation has a stable, d 10 electronic configuration and has little tendency to accept or to donate single electrons. Instead, it serves as a Lewis acid interacting with electronegative donors to increase the polarity of chemi­ cal bonds and thus promote the transfer of atoms or groups. Substitution reactions of simple metal chelates generally pro­ ceed via intermediates with an open coordination position or a distorted coor­ dination sphere. Zinc (and also cobalt) can readily accept a distorted geometry and, hence, would appear to be well suited to participate in substitution reac­ tions as, for instance, in carbonic anhydrase, carboxypeptidase and alkaline phosphatase. Entasis and Metalloenzyme Active Sites What then is the role of the protein in the mechanism of action of a metalloen­ zyme? As indicated for catalase, it makes a major contribution to the enhancement of reaction rate. It creates a proper bind­ ing locus to ensure substrate specificity. It juxtaposes catalytic residues in the precise orientation with respect to the susceptible reaction centers of the sub­ strate. It provides a suitably balanced hydrophobic-hydrophilic environment and serves to collect all the participating species in reactions between several molecules. It also provides liganding groups for binding the metal. The number, nature, orientation and im­

ROLE OF METALS IN ENZYME ACTIVITY mediate chemical environmental of these groups will dictate, in large part, the chemical characteristics of the bound metal ion. It is this total combination that manifests as the catalytic activity of the enzyme. In other words, the protein con­ tributes a constellation of ligands at the metal binding site that prepares the metal for its catalytic role. Prior to the interaction with substrate the protein has already poised the metal for catalysis. In the case of iron or copper, for example, the metal may be held in a compromised geometry between those normally assumed by its two oxidation states and approximating that of the plausible transition state for the reaction in which it is involved. Vallee and Wil­ liams have called this an entatic state im­ plying a state of tension at the active site of the enzyme.32,33 It has been defined by them as “the existence in the enzyme of an area with energy, closer to that of a unimolecular transition state than to that of a conventional stable molecule, thereby constituting an energetically poised domain.” Spectral Properties of Metalloenzymes On the basis of entasis, the physicochemical properties of metals in metalloenzymes might be expected to di­ ffer from those observed for the same metals present in well-defined model coordination complexes. Both the absorp­ tion and EPR* spectra of copper and nonheme iron enzymes are unusual when compared to those of simple copper and iron complexes.32,32 In particular the so-called copper blue enzymes exhibit absorption bands that differ strikingly in intensity and fine structure from those of non-catalytic copper proteins. The zinc ion has neither intrinsic color nor unpaired electrons hence its

* Electron spin resonance.

123

FIGURE 2. Visible absorption spectra of E. coli alkaline phosphatase containing 4 g atoms of cobalt per mole of protein compared with spectra of cobalt model complexes.

physicochemical properties are difficult to assess. However, in virtually every zinc metalloenzyme where it has been at­ tempted, the zinc ion can be replaced by cobalt to form an enzymically active de­ rivative with a visible absorption spec­ trum.29 The resulting spectra differ signif­ icantly from the spectra of model cobalt (II) complexes as shown for alkaline phosphatase in figure 2. Moreover, altera­ tion of the coordination sphere by the ad­ dition of another ligand such as an inTABLE

III

Activities of Metallocarboxy-Peptidases

Activity Peptidase* Metal

Apo Zinc Cobalt Nickel Manganese Cadmium Mercury Rhodium Lead Copper

(v/vzinc X 100)

(v' vzinc x 100>

0 100 200 47 27 0 0 0 0 0

Esteraset

0 100 114 43 156 143 86 71 57 0

*0.02 M benzyloxycarbonylglycyl-L-phenylalanine, pH 7.5, 0° C. tO . 01 M benzoylglycyl-DL-phenyllactate, p H 7.5, 25° C.

riordan

124

F i g u r e 3. Schematic representation of the mechanism of peptide hydrolysis catalyzed by carboxypeptidase A.

hibitor anion converts the irregular spec­ trum to one closely resembling a regular tetrahedral cobalt ion. It is important to note that these properties can be ob­ served in the absence of substrate. While it has not been possible to interpret these unusual properties of metals in metalloenzymes in terms of precise geometries and, ultimately, mechanisms of enzyme action, nevertheless they are quite con­ sistent with views on the entatic nature of active sites. The Role of Zinc in Metalloenzymes The Ro le

of

Z in c

in

C a r b o x y p e p t id a s e

Carboxypeptidase A is a classic zinc metalloenzyme.31 It contains one g atom of zinc per molecular weight (34,500). Removal of the metal atom either by dialysis at low pH or by treatment with chelating agents gives a totally inactive apoenzyme.22 23 Activity can be restored by readdition of zinc or one of a number of other divalent metal ions (table III).

The cobalt enzyme, for example, has twice the peptidase activity of the zinc enzyme while the nickel and manganese enzymes are much less active. The pep­ tidase activity of cadmium carboxypep­ tidase is a function of the particular pep­ tide substrate examined. In most cases, it is usually less than a few percent of that of the native zinc enzyme. Mercury, rhodium, lead and copper carboxypeptidases are essentially inactive as pep­ tidase. A comparison of the kinetic parameters for the zinc, cobalt, man­ ganese and cadmium enzyme-catalyzed hydrolysis of benzoyl-glycyl-glycyl-Lphenylalanine (table IV) reveal a range of kcat values from 6000 min ”1 for the cobalt enzyme to 43 min -1 for the cadmium en­ zyme.2 The Kmvalues, on the other hand, are almost totally independent of the par­ ticular metal present. Thus, it would ap­ pear that the primary role of the metal is to function in the catalytic process and that it has little to do with substrate bind­ ing. This is consistent with previous studies showing that peptide substrates bind to apocarboxypeptidase and prevent the reassociation of the metal-free protein with zinc. X-ray analysis of carboxypeptidase crystals together with amino acid se­ quence information has identified three protein ligands to the zinc.416 They are glutamic acid-72, histidine-69 and histidine-196. Both histidyl residues are held in position by hydrogen bonding to carboxyl side chains. In the crystalline state, a fourth coordination site is oc­ cupied by water. Diffusion of the very slowly hydrolyzed dipeptide, glycyl-Ltyrosine, into the crystals gives an enzyme-dipeptide complex. The car­ bonyl oxygen atom of the substrate’s pep­ tide bond is thought to displace the coor­ dinated water atom and interact directly with the zinc. This polarizes the carbonyl bond and promotes an attack by the car­ bonyl group of glutamic acid-270 at the

ROLE OF METALS IN ENZYME ACTIVITY carbonyl carbon atom either directly or through a water molecule (figure 3). A key feature of the catalytic mechanism porposed by the x-ray crystallographers is a 12A movement of tyrosine-248, a resi­ due thought to serve as a proton donor to the susceptible peptide nitrogen atom.16 Several conclusions drawn from studies carried out with carboxypeptidase in solution differ significantly from those derived from x-ray analysis. Using a chemically modified derivative of carboxypeptidase in which tyrosine-248 was coupled with diazotized arsanilic acid, Johansen and Vallee demonstrated that the azotyrosyl residue interacts directly with the zinc ion thus constituting a fourth and a fifth protein ligand.12,13 Such a tyrosyl-zinc interaction would preclude the large conformational change pro­ posed on the basis of crystal structure analysis. In addition, the water molecule on the zinc atom has been shown by 35C l NMR investigations to interact with glutamic acid-270 though the metal may interact with this residue directly.26 It should also be noted that the interaction of the en­ zyme with ester substrates is quite differ­ ent from that with peptides indicating that there may be at least two alternative catalytic mechanisms.2 Such studies em­ phasize the need for caution in ex­ trapolating from crystal structures to solu­ tion mechanisms. The significant feature of the zinc atom is carboxypeptidase emerging from these solution studies is its unusual coordina­ tion state. Direct complexation with two residues implicated in the catalytic mechanism suggests that this interaction poises not only the metal but simultane­ ously the organic components of the ac­ tive site. Moreover, the grouping of amino acid side chains around the zinc would effectively exclude water mole­ cules from the substrate binding pocket and perhaps, as a consequence, enhance

125 TABLE

IV

Metallocarboxypeptidase - Catalyzed Hydrolysis of Benzy1-glycylglycyl-L-phenylalanine k

M etal

Cobalt Zinc Manganese Cadmium

c a t (mm

-1

)

6,000 1,200 230 41

Km

(mM 1)

1.5 1.0 2.8 1.3

the catalytic properties of both glutamic acid-270 and tyrosine-248. The most im­ portant thing, however, is that zinc ion is central to the overall process of hyd­ rolysis. Its major function is to be able to undergo ligand exchange when triggered by the entrance of substrate into the ac­ tive site. The Role Al c o h

ol

of

D

Z in c

in

L iv e r

ehydrogenase

Liver alcohol dehydrogenase is a di­ meric enzyme with two identical sub­ units of molecular weight 40,000.5 Each subunit contains two g atoms of zinc, only one of which is involved in catalytic ac­ tivity; the other is thought to stabilize structure.7 X-ray crystallographic studies reveal two important differences be­ tween these zinc ions. First, the active site zinc is liganded in a distorted tet­ rahedral geometry to two cysteinyl sulfurs and the imidazole group of a his­ tidine. The fourth coordination position contains a water molecule. All four lig­ ands of the second zinc atom are cys­ teines. Second, the active site zinc is lo­ cated at the bottom of a hydrophobic poc­ ket about 25 Â from the protein surface and can be approached from one direc­ tion by substrate and from a second by the coenzyme nicotinamide-adenine di­ nucleotide (NAD). The structural zinc is located much closer to the enzyme sur­ face, and some 20 Â away from the active site. Its lack of a readily exchangeable ligand precludes its interaction with chelating agents, coenzyme or substrate.

RIORDAN

126

E - Zn - H20

NAD+

NAD+

NADH

NADH

pulsory binding mechanism, addition of coenzyme induces a conformational •H NAD+ change in the protein that, among other things, results in the displacement of a - E - Zn - OH" + H+ proton from the zinc-bound water mole­ cule. Substrate now binds to the zinc, +4- RCH2 - OH presumably as the negatively charged al­ cohólate ion and displaces the hydroxyl E-Zn - "OCH2R + H20 ion. Hydride transfer to the coenzyme now occurs, and the resultant aldehyde ++ dissociates from the enzyme to be re­ placed by water. This mechanism is E-Zn - OCHR analogous to that of the MeerweinPondorf-Oppenauer reaction. Both in­ ++ volve hydride transfer to carboxyl com­ pounds and require participation of a E-Zn - H20 + RCHO strong Lewis acid. Role

E - Zn - H20 +

NADH

F i g u r e 4.

Schematic reaction mechanism for alcohol dehydrogenase catalysis (adapted from reference 5).

What are the roles of the two distinct classes of zinc in alcohol dehydrogenase? All attempts to remove either or both classes of zinc by treatment with chelators or dialysis at low pH have led to the irreversible interaction of the en­ zyme. This implies a marked lability of the metal-free protein and suggests that one pair of metal ions, presumably the non-active site pair, serves to stabilize structure. Since there is no evidence to support any other role for these metal ions and since it is unlikely that they do nothing, a stabilizing function is the most obvious assignment at this time. However, x-ray structural analysis provides no indication that this part of the molecule is necessary for structure stabilization and it has been suggested that the second zinc might have evolutionary implications (5). The role of the active site zinc is be­ lieved to be the mediation of electrophilic catalysis (figure 4). Following the well-known Theorell-Chance com­

of

Z in c

in

A spartate

Tr a n sc a r b a m o y la se

Aspartate transcarbamoylase from E. coli has been studied extensively be­ cause of interest in the mechanism of its allosteric feedback regulation.11 The en­ zyme can be dissociated into two types of subunits, one which retains catalytic ac­ tivity and one which binds the regulator molecule, CTP, but is inactive. It should be noted that the regulatory subunits con­ tain zinc, one g atom per 17,000 pro­ tomeric weight. The role of zinc in the regulation of aspartate transcarbamoylase is not entirely understood. Zinc seems to stabilize the tertiary structure of the reg­ ulatory protomeric unit, promote its dimerization and is important for recon­ stitution of the native enzyme from its separated subunits. Substitution of Hg2+ or Cd2+ for zinc gives a derivative with properties nearly identical to those of the native enzyme. Zinc does not appear to be involved in binding the allosteric lig­ and, CTP, to the regulatory subunit. The Role

of

M

e t a l s in

A l k a l in e P h o s p h a t a s e

Escherichia coli alkaline phosphatase is a zinc metalloenzyme containing four g atoms zinc per molecular weight of

ROLE OF METALS IN ENZYME ACTIVITY 89,0003,25 As with alcohol dehyrogenase, each of the two identical subunits con­ tains two zinc atoms, one at the active site and one at another site. In addition the enzyme, when isolated at neutral pH, contains 1.3 g atoms of magnesium per mole.1,3 Magnesium alone does not acti­ vate the apoenzyme but increases the ac­ tivity of the enzyme containing two g atoms of zinc by about four-fold and that of the four zinc enzyme by 20 percent. Hence, magnesium regulates the activ­ ity of alkaline phosphatase while zinc serves, on the one hand, to stabilize struc­ ture and, on the other, to participate in the catalytic process. Magnesium inter­ acts directly with the enzyme and does not seem to exert its regulatory role by means of substrate binding. Studies with phosphatase containing cobalt instead of zinc indicate that magnesium binding in­ duces a change in the coordination geometry of the active site cobalt ions and alters the relative affinities of cobalt or zinc for the catalytic, structural or reg­ ulatory sites.1 While it is not yet possible to define more precisely the role of the three dif­ ferent classes of metal ions in alkaline phosphatase, this example illustrates quite well the emerging general princi­ ple. Metals in metalloenzymes can have any one of three different roles,— catalytic, structural or regulatory. The same metal, e.g., zinc, can have any one, two or all three of these roles in the same enzyme. Alternatively, different metals can fulfill these functions in a given en­ zyme. Hence, the analytical demonstra­ tion of the presence of a particular metal species in an enzyme is not sufficient to establish the specific role of that metal in biological function. The Role and

of

Z

P r o t e in M

in c in

N u c l e ic A c id

e t a b o l is m

Zinc has long been known to be essen­ tial for the normal growth and develop-

127

DNA

TRANSCRIPTION

REVERSE.

^TRANSCRIPTASE

___* ♦

RNA POLYMERASES

(r-R N A TRANSLATION

^

*DNA*

POLYM ERASE^

m -R N A )

t-R N A

|,------- A A -t-R N A ELONGATION FACTOR

OTHER Z n -E N Z Y M E S AND Zn-PROTEINS

I

PROTEIN

AMINO ACIDS PROTEASES PEPTIDASES

FIGURE 5. Zinc enzymes in nucleic acid and protein metabolism.

ment of microorganisms, plants, animals and, more recently, man .8,9,30 As evi­ denced by the few examples cited, the primary role of zinc would be to function in zinc metalloenzymes. However, it seems unlikely that disrupting the activ­ ity of carboxypeptidase or alcohol dehy­ drogenase would have profound effects on growth. Moreover studies on the con­ sequences of zinc deficiency, particularly in Euglena gracilis, indicated defects in nucleic acid, protein synthesis and cellu­ lar division.8 Peptides, amino acids, nucleotides and polyphosphate all accumulate under these conditions and the rate of incorpo­ ration of [3H]-uridine into ribonucleic acid (RNA) is markedly decreased. Cytofluorometric analysis of the metabolism of deoxyribonucleic acid (DNA) during the cell cycle of E. gracilis has revealed that all of the biochemical processes es­ sential for cells to pass from G! into S, from S into G 2 and from G 2 to mitosis re­ quire zinc.9 It is now clear that zinc defi­ ciency disrupts these critical steps in the normal growth process because many of the important enzymes are zinc enzymes (figure 5). Thus, DNA polymerase, the various RNA polymerases, certain elon­ gation factors and perhaps some amino acyl t-RNA syntheses all require zinc. Moreover, the RNA-dependent DNA polymerases from avian, simian, feline and RD-114 tumor viruses have all been found to be zinc metalloenzymes.30 Such data extend the role of zinc in enzymes essen­ tial to normal nucleic acid metabolism to

128

RIORDAN

others presumed to play a role in leukemic processes.

Conclusions Metalloenzymes are now well estab­ lished entities in biochemistry, and their catalytic activities reflect the nutritional importance of the corresponding essential minerals. At present it is recognized that copper and iron are especially important in enzymes catalyzing oxidoreduction proc­ esses while zinc, which can be a compo­ nent of enzymes involved in a wide variety of reaction types, is critically associated with the fundamental steps of transcrip­ tion and translation. The identification of individual metalloenzymes is a relatively recent occurrence, particularly for zinc en­ zymes which are usually colorless. Two decades ago only three or four zinc enzymes were known while 20 times that many are known today, and the number increases steadily. Several dozen iron and copper enzymes have been investigated and, despite their more obvious charac­ teristics, new ones continue to be found. F ewer examples of cobalt, manganese and selenium enzymes are presently known; however, based on experience with other metals, this situation is expected to change with time as well. As the number of known metalloen­ zymes increases, the metabolic and physiologic consequences of metal defi­ ciency begin to be understood. The growth retardation and teratogenic ef­ fects of zinc deficiency or the weakening of elastic tissue owing to copper defi­ ciency, can be traced, at least in part, to specific enzymes or groups of enzymes. However, many aspects of trace metal metabolism remain unknown. The rela­ tionships between trace metals, such as the reciprocal alterations in zinc and copper concentrations in blood serum, the effects of manganese and iron on

copper and zinc concentrations as well as the hormonal influences on all of these, are yet to be defined. A complex inter­ play between storage proteins, carriers and functional macromolecules would seem to underlie many of the biological responses to changes in trace metal nutri­ tion. Moreover, it should be noted that nucleic acids as well as proteins are known to bind many of the metals men­ tioned. Hence, some effects will not be due to changes in an enzyme activity but rather to the inability of these molecules to exercise their assigned biological func­ tions. On the basis of current knowledge, it would seem that a good deal of progress has been made in deciphering the role of metals in enzymes in elucidating their overall mode of action in biology.

References 1. A n d e r s o n , R. A., K e n n e d y , F . S., and V a l LEE, B. L.: The effect of Mg(II) on the spectral properties o f C o (II) alkaline phosphatase. Biochemistry 25:3710-3716, 1976. 2. A u l d , D. S. and H o l m q u is t , B.: Carboxypeptidase A. Differences in the mechanisms of ester and peptide hydrolysis. Biochemistry 13:4355-4361, 1974. 3. B osron , W. F., Ke n n e d y , F. S., and V a l l e e , B. L.: Zinc and magnesium content of alkaline phosphatase from Escherichia coli. Biochemis­ try 24:2275-2282, 1975. 4. B r a d s h a w , R. A., E r i c c s o n , L. H., W a l s h , K. A., and N e u r a t h , H.: The amino acid se­ quence of bovine carboxypeptidase A. Proc. Nat. Acad. Sei. 63:1389-1394, 1969. 5. B r ä n d e n , C. I., J Ö r n v a l l , H., E k l u n d , H., and F u r u g r e n , B.: Alcohol dehydrogenases. The Enzymes, vol. XI. Boyer, P. D., ed. New York, Academic Press, pp. 103-190, 1975. 6. D ix o n , N. E., G a z z o l a , C., Bl a k e l e y , R. L., and ZERNER, B.: Jack bean urease (EC3.5.1.0). A metalloenzyme. A simple biological role for nickel? J. Amer. Chem. Soc. 97:4131-4133, 1975. 7. D r u m , D . E. and V a l l e e , B. L.: Differential chemical reactivities of zinc in horse liver al­ cohol dehydrogenase. Biochemistry 9:40784086, 1970. 8. F al c h u k , K. F., F a w c et t , D. W., and V al ­ l e e , B. L.: Role of zinc in cell division of Euglena gracilis. J. Cell. Sei. 17:57-78, 1975. 9. F a l c h u k , K. F ., K r i s h a n , A., and V a l l e e , B. L.: DNA distribution in the cell cycle of Eu­ glena gracilis. Cytofluorometry of zinc defi­ cient cells. Biochemistry 24:3439-3444, 1975.

ROLE OF METALS IN ENZYME ACTIVITY 10. F l o h e , L., G u n z l e r , W. A., an d Sc h o c k , N. H.: G lu ta th io n e peroxidase: A s e le n o e n zy m e . F ed. E u ro p e a n B io c h e m . Soc. L etters 32:132-136, 1973. 11. J a c o b s o n , G. B. and S t a r k , G. B.: Aspartate transcarbamylases. The Enzymes, vol. IX. Boyer, P. D., ed. New York, Academic Press, pp. 225-308, 1973. 12. J o h a n s e n , J. T. and V a l l e e , B. L.: Differ­ ences betw een the conform ation of arsanilazotyrosine 248 of carboxypeptidase in the crystalline state and in solution. Proc. Nat. Acad. Sci. 68:2532-2535, 1971. 13. J o h a n s e n , J. T. and V a l l e e , B. L.: Conforma­ tions of arsanilazotyrosine 248 carboxypepti­ dase Aa,p,y. Comparison of crystals and so­ lutions. Proc. Nat. Acad. Sci. 70:2006-2010, 1973. 14. LEHNINGER, A. L.: Bole of metal ions in en­ zyme systems. Physiol. Bev. 30:393-429, 1950. 15. Li, T. K. and V a l l e e , B. L.: Trace elements, Section B. The biochemical and nutritional role of trace elements. Modem Nutrition in Health and Disease. Goodhart, R. S. and Shils, M. E., eds. Philadelphia, Lea and Febiger, pp. 372-399, 1973. 16. L i p s c o m b , W . N ., H a r t s u c k , J. A., R e e k e , G . N., J r ., Q u io c h o , F. A., B e t h g e , P. H ., L u d ­ w i g , M . L ., S t e i t z , T. A., M u i r h e a d , H ., and C o p p o l a , J. C .: The structure of carboxy­ peptidase A. V II. The 2.0 A resolution studies of the enzyme and of its complex with glycyl tyrosine, and m echanistic deductions. Brookhaven Symp. Biol. 21:24-90, 1968. 17. M a h l e r , H. R.: Interrelationships with en­ zymes. Mineral Metabolism, vol. 1, Part B. Comas, C. L. and Bronner, F., eds., New York, Academic Press, pp. 743-879, 1961. 18. M OORE, C. V.: Major Minerals. Section C, Iron. Modern Nutrition in H ealth and Disease. Goodhart, R. S. and Shils, M . E., eds. Philadel­ phia, Lea and Febiger, pp. 297-323, 1973. 19. N o r t h r o p , D. B. and W o o d , H. G.: Transcar­ boxylase V. The presence of bound zinc and cobalt. J. Biol. Chem. 244:5801-5807, 1969.

129

20. O ’D e l l , B. L.: Biochemistry and physiology of copper in vertebrates. Trace Elem ents in Human Health and Disease, Zinc and Copper. Prasad, A. S., ed., New York, Academic Press, pp. 391-413, 1976. 21. R lO R D A N , J. F.: Biochemistry of zinc. The Med­ ical Clinics of North America. 60:661-674,1976. 22. B i o r d a n , J. F. and V a l l e e , B. L.: The func­ tional role o f metals in m etalloenzym es. Protein-Metal Interactions. Friedman, M., ed. Advances in Exp. Med. Biol. 48:33-58, 1974. 23. R i o r d a n , J. F. and VALLEE, B. L.: Structure and function of zinc metalloenzymes. Trace Elements in Human Health and Disease, Vol. 1, Zinc and Copper. Prasad, A. S., ed., New York, Academic Press, pp. 227-256, 1976. 24. SCRUTTON, M. C.: Metal enzymes. Inorganic Biochemistry, vol. 1. Eichorn, G. L., ed. Amsterdam, Elsevier, pp. 381-437, 1973. 25. S im p s o n , R. T. and V a l l e e , B. L.: Two differ­ entiable classes of metal atoms in alkaline phosphatase of Escherichia coli. Biochemistry 7:4343-4350, 1968. 26. S t e p h e n s , R. S., J e n t o f t , J. E., and B r y a n t , B. G.: 35C1 nuclear magnetic resonance inves­ tigation of carboxypeptidase A. J. Amer. Chem. Soc. 96:8041-8045, 1974. 27. K n a p p e i s , G. G. and C A R L S O N , F.: Ultrastruc­ ture of Z-disc in skeletal muscle. J. Cell. Biol. 23:323-335, 1962. 28. V a l l e e , B. L.: Zinc and metalloenzymes. Adv. Protein Chem. 10:317-334, 1955. 29. V A L L E E , B. L.: C obalt substituted zinc metalloenzymes. Metal Ions in Biological Sys­ tems. Dhar, S. K., ed. New York, Plenum, pp. 1-12, 1973. 30. V a l l e e , B. L.: Zinc biochemistry: a perspec­ tive. Trends in Biochem. Sci. 2:88-91, 1976. 31. V a l l e e , B. L. and N e u r a t h , H.: Carboxypep­ tidase: a zinc metalloprotein. J. Amer. Chem. Soc. 76:5006-5007, 1954. 32. V a l l e e , B. L. and W i l l i a m s , R. J. P.: Enzyme action: views derived from metalloenzyme studies. Chem. in Britain 4:397-402, 1968.