Metal Clusters in Proteins - American Chemical Society

Chapter 9

Fe(III) Clusters on Ferritin Protein Coats and Other Aspects of Iron Core Formation

Downloaded by CORNELL UNIV on October 23, 2016 | http://pubs.acs.org Publication Date: June 21, 1988 | doi: 10.1021/bk-1988-0372.ch009

Elizabeth C . Theil Department of Biochemistry, North Carolina State University, Raleigh, N C 27695-7622 F e r r i t i n , found i n plants, animals, and some bacteria, serves as a reserve of iron for iron-proteins, such as those of respiration, photosynthesis, and DNA synthe s i s , as well as providing a safe s i t e for d e t o x i f i c a tion of excess iron. The structure of f e r r i t i n , unique among proteins, is a protein coat of multiple, highly conserved polypeptides around a core of hydrous f e r r i c oxide with variable amounts of phosphate. Variations in ferritin protein coats coincide with variations i n iron metabolism and gene expression, suggesting an interdependence. Iron core formation from protein coats requires Fe(II), at least experi mentally, which follows a complex path of oxidation and hydrolytic polymerization; the roles of the protein and the electron acceptor are only p a r t l y understood. It is known that mononuclear and small polynuclear Fe clusters bind to the protein early i n core formation. However, v a r i a b i l i t y in the s t o i c h i ometry of Fe/oxidant and the apparent sequestration and s t a b i l i z a t i o n of Fe(II) i n the protein for long periods of time indicate a complex microenvironment maintained by the protein coats. F u l l understanding of the r e l a t i o n of the protein to core formation, p a r t i c u l a r l y at intermediate stages, requires a systematic analysis using defined or engineered p r o t e i n coats. The Structure and Function of F e r r i t i n Iron is required by e s s e n t i a l l y a l l l i v i n g organisms, p a r t i c i pating i n respiration, photosynthesis, and DNA synthesis. An apparent exception is certain members of the b a c t e r i a l genus Lactobacillus (1). In general, the reactions mediated by iron proteins are electron transfer or activation of dioxygen or n i t r o gen. However, the very properties of iron which render i t so 0097-6156/88/0372-0179$06.00/0 ° 1988 American Chemical Society

Que; Metal Clusters in Proteins ACS Symposium Series; American Chemical Society: Washington, DC, 1988.

180

METAL CLUSTERS IN PROTEINS

useful are hazardous when uncontrolled. For example, the transfer of a single electron from Fe(II) to dioxygen or to organic com pounds produces dangerous radicals which can derange the delicate structures required for normal l i f e . Moreover, the oxidation product, hydrated Fe(III), i s extraordinarily insoluble [K 10" M (2) at the pH of b i o l o g i c a l materials], forming r u s t - l i k e precipitates. S t a b i l i z i n g Fe(III) as a complex with, e.g., c i t r a t e creates additional problems. For example, the amount of c i t r a t e required to s t a b i l i z e the iron released from senescent red c e l l s of a human i n a single day could only be supplied by drink ing 5 l i t e r s of orange juice (3), an amount l i k e l y to disrupt the normal acid/base balance i n the body and to derange the metabolism of other metal ions. The problem of using iron i s solved i n b i o l o g i c a l systems by the use of f e r r i t i n (3), a protein which sequesters the iron i n a safe, available form for plants, animals, and bacteria. Such a widespread d i s t r i b u t i o n of an iron storage protein and the high conservation of the protein structure suggests that f e r r i t i n i s an old protein. Certainly the need for f e r r i t i n is old and may even have preceded the appearance of dioxygen in the atmosphere, since even the most p r i m i t i v e organisms appear to have depended upon iron. g


18

F e r r i t i n Structure. F e r r i t i n i s a large complex protein composed of a protein coat which surrounds a core of polynuclear hydrous f e r r i c oxide (FeO'OH or Fe203«nH 0). Points of contact which have been observed (4) between the inner surface of the protein coat and the iron core may r e f l e c t sites of cluster formation and core nucleation. 2

The Protein Coat. Twenty-four polypeptides assemble into a hollow sphere, of ca. 100-120 X i n outer diameter, to form the protein coat of f e r r i t i n . The diameter of the i n t e r i o r , which becomes f i l l e d with hydrous f e r r i c oxide, i s ca. 70-80 X. Subunit assembly appears to be spontaneous; the coat remains assembled even without the iron core. Subunit biosynthesis i s actually controlled by the amount of iron to be stored by a c e l l ; the subunit templates (mRNAs) are stored i n the cytoplasm of a c e l l i n a repressed form and are recruited for biosynthesis when the concentration of iron increases (3). Many of the current ideas about the shape of the f e r r i t i n molecule are derived from the high resolution x-ray c r y s t a l l o graphic studies of Harrison and coworkers (5,6) (Figure 1) on the protein coat of f e r r i t i n from the spleen of horses, i n which e s s e n t i a l l y a l l (> 90%) of the polypeptide subunits are i d e n t i c a l . However, protein coats of f e r r i t i n from other animals, and indeed from different c e l l s and tissues i n the same animal, can be composed of assemblages of s i m i l a r , but d i s t i n c t , subunits (3). The maximum number of different f e r r i t i n subunits possible i n a given c e l l , tissue, or organism is not known, but three d i s t i n c t subunits have been characterized so far (7). Although as many as 16 different DNA fragments encoding f e r r i t i n - l i k e sequences have been i d e n t i f i e d (8-11) i n the genome of a number of animals, some of the sequences may not be functional genes; the exact number of r e a l and pseudoferritin genes i s the subject of intense



9. THEIL

Fe(III) Clusters on Ferritin Protein Coats

Figure 1. The protein coat of f e r r i t i n from horse spleen. N refers to the N-terminus and E refers to the location of the E h e l i x , a short helix with an axis perpendicular to the long axis of the subunit and which lines the channels formed at the four fold axes. (Reproduced with permission from Ref. 5. Copyright 1983 E l s e v i e r . )


181


182


investigation. Overall, the morphology of the assembled protein coat appears to be conserved not only within an organism but also among plants, animals, and bacteria (3,12), in spite of variations i n the ratio of f e r r i t i n subunit types, the number of f e r r i t i n subunit types, and, in the case of f e r r i t i n from bacteria, the presence of heme (13-15). Dimeric, t r i m e r i c , and tetrameric interactions of f e r r i t i n subunits each have d i s t i n c t i v e features which may relate to the function of the protein coat. Among vertebrates, at least, the amino acid sequence in such regions is highly conserved, and forms structures which may have functional roles. The structure of the dimer interface i n the crystals from the protein coats of horse spleen f e r r i t i n (5,6), for example, is a densely packed mass of amino acid sidechains on the outer surface, where the protein would be i n contact with the cytoplasm, but forms a f l e x i b l e groove with potential ligands for iron on the inner surface facing the iron core. Speculation that the dimer interface participates i n nucleation of the iron core has been frequent, not only because of the structure (6) but also because of the effect of amino acid modification (16) and of c r o s s - l i n k i n g subunit pairs (17). Trimeric and tetrameric subunit interactions form channels through the protein coat large enough to permit the movement of iron and other small molecules and leading to the hypothesis that the channels function i n iron uptake and release from f e r r i t i n ; preliminary data support such ideas (18,19) but the p o s s i b i l i t y that the protein coat i s s u f f i c i e n t l y f l e x i b l e to permit the entrance or exit of iron at other sites cannot be discounted. The Iron/Protein Interface. Interactions of iron with the protein coat of f e r r i t i n are most easily characterized i n the early stages of core formation when most, i f not a l l , of the iron present is i n contact with the protein coat. In the complete core, bulk iron i s inorganic. To date, the protein coat has been l i t t l e examined early i n iron core formation except i n terms of effects on the iron environment. Studies of the iron early i n core formation w i l l be discussed l a t e r . The Iron Core. A l l f e r r i t i n molecules store iron inside the pro tein coat as a polynuclear oxo-bridged complex. Experimentally, f e r r i t i n can only be formed from protein coats using Fe(II). Two reactions are involved: oxidation of Fe(II) and hydrolysis of Fe(III) to form oxo-bridged complexes. [Some Fe(II) atoms might form oxo or hydroxo bridges before oxidation, depending upon l o c a l microenvironments i n the growing core.] The reverse reaction, i.e., release of iron from f e r r i t i n as Fe(II), requires addition of electrons and hydration of the oxo bridges (or ligand exchange). Whether or not the two reactions are temporally coupled is not known, although i t is sometimes assumed. However, Watt and Frankel observed the apparent uncoupling of the two reactions when coulometric reduction of iron i n f e r r i t i n cores produced a f e r r i t i n with 50% Fe(ll) but with e s s e n t i a l l y a l l the iron retained inside the protein coat (20).



9. THEIL

Fe(III) Ousters on Ferritin Protein Coats

183

Variations i n f e r r i t i n iron cores include the number of iron atoms, composition, and the degree of order (3,6,21-23). Size variations of the iron core range from 1-4500 Fe atoms and appear to be under b i o l o g i c a l control (e.g. Ref. 15). The d i s t r i b u t i o n of iron core sizes in a p a r t i c u l a r f e r r i t i n preparation can be easily observed after sedimentation of f e r r i t i n through a gradient of sucrose. Known compositional variations of f e r r i t i n iron cores only involve phosphate, which can range from as much as 80% (21) to as l i t t l e as 5% of the iron (21); in normal mammalian l i v e r or spleen, the amount of phosphate i n the f e r r i t i n iron core i s ca. 12% of the iron (24). When the phosphate content is high, the d i s t r i b u t i o n of phosphate is c l e a r l y throughout the core rather than on the surface. However, i n t e r i o r locations for phosphate are also suggested when the phosphate content is lower, by data on an Fe(III)ATP model complex (P:Fe = 1:4) (25) or by phosphate a c c e s s i b i l i t y studies i n horse spleen f e r r i t i n (P:Fe = 1:8) (24). Based on model studies, other possible variations i n core composi tion could include H2O or sulfate (26). The degree of order i n f e r r i t i n iron cores has been shown to vary depending upon the organism and the amount of phosphate. For example, high resolution electron microscopy shows that the iron cores of many mammalian f e r r i t i n s are microcrystalline and are composed of several small or one large c r y s t a l (27,28). Such crystals are s i m i l a r to f e r r i h y d r i t e , a mineral (29) which can also be formed experimentally by heating solutions of f e r r i c nitrate (30); f e r r i h y d r i t e has no phosphate. In contrast, the cores of f e r r i t i n from bacteria (21) and some invertebrates (22) are highly disordered. While the disorder i s related to the amount of phosphate i n some cases, i n others the disorder appears to depend on the environment i n which the core forms (the protein coat? other anions?). In mammalian f e r r i t i n s , i t is not known i f phosphate might be responsible for regions of disorder that were not detected by the analysis of the d i f f r a c t i o n pattern. In contrast to phosphate, sulfate might increase the order of f e r r i t i n iron cores since, i n model systems, the amount of order i n soluble hydrous f e r r i c oxides appears to be increased by sulfate (26). However, the effect of sulfate on f e r r i t i n iron cores has not yet been examined. F e r r i t i n iron cores, or polynuclear iron complexes i n l i p i d vesicles or i n matrices of protein and complex carbohydrates, appear to be the precursors of minerals such as hematite and magnetite that form i n certain bacteria (31), marine invertebrates (22), insects, and birds. The conversion from f e r r i t i n - l i k e iron cores requires p a r t i a l changes i n the oxidation state of and/or ordering of the iron atoms, and may depend on some of the natural variations i n f e r r i t i n core structure. Clearly, the f i n a l structure of the iron core of f e r r i t i n is the resultant of effects of the protein coat and the environment on the formation of the polynuclear oxo-bridged iron complex. Although the important interplay among iron, various anions, and different f e r r i t i n protein coats (or other organic surfaces) has only begun to be examined i n a systematic way, the early results promise new insights to understanding both the significance of


184


variations i n f e r r i t i n protein coats and core structures as well as the mechanisms of biomineralization.


F e r r i t i n Function The function of a l l f e r r i t i n molecules is to store iron. However, the mechanisms by which iron enters the core or is released from the core in vivo i s poorly understood. Experimentally, Fe(II), but not Fe(III), mixed with f e r r i t i n protein coats forms normal iron cores. Moreover, reductants such as thioglycollate or reduced flavins can reverse the process of core formation and release Fe(ll) from the core. Since such reductants occur in vivo, reduction of f e r r i t i n cores may also occur in vivo. In contrast to the dearth of information about physiological uptake and release mechanisms for iron stored i n f e r r i t i n , there i s a great deal of information about the physiological roles of iron stored i n f e r r i t i n . There are at least three purposes for which the stored iron can be used. F i r s t , f e r r i t i n can store iron for ordinary c e l l u l a r metabolism such as the synthesis of iron proteins, e.g., the cytochromes or ribonucleotide reductase; a l l c e l l s store iron for such housekeeping purposes, but the amount of iron (and of f e r r i t i n ) w i l l vary depending upon the quantity of iron proteins needed, cf. hemoglobin i n red blood c e l l s to cyto chromes i n an adipocyte. Second, f e r r i t i n can store excess iron for detoxification. Although i n normal c e l l s the incorporation of iron by a c e l l i s t i g h t l y coupled to c e l l u l a r needs, the safe guards can be breached i n abnormal conditions that produce iron overload. For example, excess iron accumulates i n the genetic disease hemochromatosis or i n copper poisoning, where excess iron i n the spleen i s stored i n a f e r r i t i n with a modified protein coat that allows more iron to be sequestered/molecule (17). Finally, some c e l l s have the specialized role of storing extra iron (beyond the need for housekeeping or detoxification) for other c e l l s . There are several types of specialized iron storage c e l l s and each appears to release iron under different circumstances depending on the need of the animal, e.g., d a i l y recycling of iron, long term iron storage for emergencies such as hemorrhaging, or iron storage for large and rapid demands during development and growth (3). It is tempting to think that the different types of f e r r i t i n protein coats i n a s p e c i f i c c e l l type or in a p a r t i c u l a r physiological condition reflect the different iron storage roles of f e r r i t i n and different modes of gene regulation. While some observations support such an idea (7,17), i t is far from being proven. Iron Clusters and the Early Stages of F e r r i t i n Iron Core Formation The sequence of steps in the biosynthesis of the f e r r i t i n iron core has been studied by analyzing the incorporation of F e into f e r r i t i n during synthesis of the protein in vivo. Ferritin, collected at various intervals after the induction of synthesis, was fractionated according to iron core size by sedimentation through gradients of sucrose (32). F e appeared f i r s t in f e r r i t i n with small amounts of Fe, and l a t e r , the F e appeared i n fractions further down the gradient as the core size and the r a t i o 5 9

5 9

5 9



9. THEIL


185

of Fe to protein increased. The data indicate the assembly of the polypeptide subunits into protein coats with no or small amounts of iron, followed by accretion of the iron core. Iron core forma tion i n f e r r i t i n has been studied experimentally either by adding small amounts C< 10-12 Fe atoms) of Fe(II) or other metal ions to f e r r i t i n protein coats or by adding enough Fe(II) to form r e l a t i v e l y complete cores (400-2000 Fe atoms/protein coat). Note that cores do not form experimentally with Fe(III) and protein coats, although small amounts of Fe(III) can be added to existing f e r r i t i n iron cores (33). Each experimental approach has p a r t i c u l a r virtues. For example, at low Fe/protein ratios the contributions of Fe bound to the protein are emphasized. At high Fe(II)/protein ratios [and Fe(II) concentrations], measurement of oxygen uptake or Fe(II) oxidation i s f a c i l i t a t e d . Each set of experimental conditions has a b i o l o g i c a l counterpart. However, experimental conditions which correspond to the intermediate states of core formation, i.e. the repeated addition of small amounts of Fe(II), have been l i t t l e used as yet. Low Fe/Protein (Fe 8 because the variance is so small. The variance is 3.05 x 10~ for Fe-O and Fe-C interactions, 1.60 x 10~ for the Fe-Fe and distant Fe-O interactions, and 8.56 x 10~ for the distant Fe-O interaction. (Reproduced from ref. 39. Copyright 1987 American Chemical Society.) Continued on next page. _1

3

5

5


10


188


-12 - 8 - 4

0

A

8

12

VELOCITY (mm/s) 57

Figure 2. Continued (C) Mossbauer spectra of Fe(III)-apoferritin complex in zero field: (1) 1.5, (2) 4.2, and (3) 10 K. (Reproduced from ref. 39. Copyright 1987 American Chemical Society.)



9. THEIL

189


Where i n the protein coat the Fe(III) clusters form is not yet known. Nor is i t known how many Fe(III) atoms can be accom modated by the protein coat; eventually, during iron core formation, Fe atoms added to the cluster w i l l be i n the environ ment of bulk core iron. Each conundrum has as a possible solution a testable hypothesis. In the case of the location of the s i t e s of Fe(III) cluster formation, inspection of conserved carboxylate residues i n the sequences of f e r r i t i n s from mammals and frogs (3) shows a cluster of three at the trimer interface and a line of three pairs on the inner surface, the dimer interface; no patches of carboxylate ligands appear to exist on the outer surface of the protein coat of f e r r i t i n , making i t unlikely that the Fe(III) cluster is on the surface. The trimer interface is not a l i k e l y s i t e for core nucleation, as indicated above, because the growing core would block the channels and because polymerization of iron could extend to the outside of the protein coat. Moreover, the trimer interaction appears to be quite i n f l e x i b l e i n the crystals and unlikely to expand readily to admit an Fe(III) cluster. In contrast, the dimer interface has the following features: 1. more potential Fe ligands on the inner surface than the three-fold channel (3); 2. a structure which can orient core growth toward the hollow center of the protein coat, i.e. a f l e x i b l e groove on the inner surface toward the hollow center of the protein, and a boundary for the groove at the outer surface of the protein coat, which is a seemingly impenetrable mass of amino acid side chains (6); and 3. altered iron uptake when the dimer interaction i s altered by covalent cross-links (17). Analysis of iron core formation i n a recently prepared form of f e r r i t i n i n which carboxylate residues at the dimer interface have been replaced using s i t e - d i r e c t e d mutagenesis (S. Sreedharan and E. T h e i l , unpublished results) w i l l test the hypothesis. In the case of the number of Fe(III) atoms that can be accommodated by the protein coat of f e r r i t i n , the number of binding sites (8-12) x the number of Fe(III) atoms/cluster (3-5) defines the lower l i m i t of the number of clustered Fe(III) bound (24-60). Analysis of x-ray absorption spectra i n complexes with a variety of numbers of Fe atoms added can provide l i m i t s to the numbers of iron atoms the protein can accommodate i n the clusters. High Fe/Protein (Fe = 400-2000/Molecule). The reconstitution of large f e r r i t i n iron cores from Fe(II) and protein coats involves the oxidation and polymerization of Fe. Usually, dioxygen is the electron acceptor i n the experiments. A number of investigators have examined the stoichiometry of Fe(II) oxidation (Table I) (41-43), with results varying from 1.5-4 F e / 0 ; a stoichiometry of 2 Fe/02 was the value most frequently obtained. The amounts of iron and the buffers used were s i m i l a r , although the method of measuring dioxygen consumption varied from manometry (41), to incorporation of 0 from 0 (42), to the use of an oxygen electrode (43). No evidence for oxygen radical intermediates was obtained. The lack of a t r u l y satisfying explanation for the stoichiometry of oxygen consumption and Fe oxidation emphasizes the complexity of the path of f e r r i t i n iron core formation and the amount of knowledge s t i l l to be acquired. Recent evidence 2

1 8

1 8

2


190


Table I.

Fe(II) oxidized O2 consumed

n

Stoichiometry Fe/02 * F e r r i t i n Formation [Horse Spleen Apof e r r i t i n , Fe( II)NHi*SOiJ Reference

Fe/Molecule

Buffer

Analysis

3.8-4.1

2300-1500

Imidazole, pH 6.4

Manometric

(41)

1.78-2.00

240-1900

Mass Spec Hepes-Mestrometry Imidazole, H 0 , pH 7.0

(42)

1 8

2


2.5-3.2*

At 22 Fe/molecule,

222-2220

Hepes, T r i s Imidazole, pH 7.4

Oxygen Electrode

(43)

the r a t i o : Fe/02 consumed = 1.5.

emphasizes how much s t i l l needs to be learned (Figure 3; 44). For example, when changes i n the oxidation state of bulk Fe are measured d i r e c t l y , e.g. by x-ray absorption spectroscopy, the changes continue for up to 24 hours. Such observations contrast with the results obtained when the changes i n the Fe environment are measured i n d i r e c t l y . For example, when colorless solutions of Fe(II) and a p o f e r r i t i n are mixed, the solution quickly becomes the amber color of solutions of f e r r i t i n . The absorption spectrum of both f e r r i t i n and Fe(II) mixed with a p o f e r r i t i n is broad, with a series of transitions i n the v i s i b l e range. Measurements of color formation can be made at wavelengths between 310 and 500 nm; extinction coefficients vary with the amount of iron/molecule (34). The color change of the solutions, which previously has been used as a measure of Fe(II) oxidation during f e r r i t i n iron core formation, is complete long before the x-ray absorption spectrum changes from that of Fe(II) to that of Fe(III) (Figure 3). Note that the x-ray absorption spectrum of mixtures of solu tions of Fe(ll) and Fe(III) s a l t s i n 0.1 N HN0 can be reproduced very precisely by mathematically varying the weighting of averages of the spectra of solutions of pure Fe(II) or Fe(III) s a l t s equivalent to those i n the actual mixtures. Such results indicate that x-ray absorption spectra can be used to estimate r e l i a b l y the r e l a t i v e amounts of Fe(II) and Fe(III) i n a mixture. Other analyses of the rate of oxidation of Fe(II) i n f e r r i t i n iron core formation have used the decline i n r e a c t i v i t y of Fe(II) with o-phenanthroline or b i p y r i d y l as an index. Almost immediately, upon mixing Fe(II) with solutions of f e r r i t i n protein coats, the a v a i l a b i l i t y to o-phenanthroline diminishes [the aliquots were mixed with the chelator i n 0.01 N HC1 (Figure 3) to allow complex formation before adjusting the pH]; a l l of the Fe(II) i n solutions of Fe(ll) without protein reacted with o-phenanthroline. By 2 hours after Fe(ll) solutions were mixed with solutions of the protein, l i t t l e of the Fe(II) i n the protein solution was a v a i l able to complex with o-phenanthroline, although the x-ray absorp tion spectrum was e s s e n t i a l l y unchanged and was indistinguishable 3



9. THEIL


TIME

191

(hours)

Figure 3. Comparison of the rate of oxidation of Fe(II) when mixed with a p o f e r r i t i n coats (480 Fe/molecule) i n 0.15 M Hepes*Na, pH 7.0, using absorbance at 420 nm ( • — • — • ), a v a i l a b i l i t y to react with o-phenanthroline ( o—o—o ), change in the x-ray absorption near edge structure (XANES) ( • — • — • ). A l l three types of measurements were made under the same experimental condi tions, including the sample holder. (Data are taken from Ref. 44.) Fe(ll) is released to react with o-phenanthroline after b o i l i n g the protein i n 1 N HC1.


192



from that for solutions of Fe(II). Boiling the protein solution in acid to denature the protein coats rendered Fe(Il) available to complex with o-phenanthroline (J. Rohrer and E. T h e i l , unpublished results). Apparently, the protein rapidly incorporated Fe(II) inside the coats i n an environment unavailable to the complexing agent, where the oxidation and/or the t r a n s i t i o n to polynuclear hydrous f e r r i c oxide occurred slowly. What causes the apparent s t a b i l i z a t i o n of Fe(II) inside the protein coats is unknown, but a microenvironment r i c h in protons or deficient in electron acceptors are possible explanations. Models for Nucleation of F e r r i t i n Cores. A number of small polynuclear Fe(lII) complexes of defined size and s t a b i l i z e d by organic coats have been c r y s t a l l i z e d and characterized (e.g. 4548). A complete description of their properties is the subject of Chapter 10 i n t h i s volume by S. M. Gorun. Summary and Conclusions F e r r i t i n is a complex composed of a protein coat and a hydrous f e r r i c oxide core with variable amounts of phosphate; points of contact which occur between the inner surface of the protein and the iron core may represent the sites of Fe(III) cluster formation and core nucleation. Animals, plants, and bacteria use f e r r i t i n to maintain a safe, available source of iron that overcomes both the t o x i c i t y of Fe(II) and the i n s o l u b i l i t y of Fe(III). The protein coat of f e r r i t i n is assembled from 24 polypeptide chains (subunits) to form a hollow sphere with channels at sites of the t r i m e r i c and tetrameric subunit interactions and a f l e x i b l e groove on the inner surface of the dimeric subunit contacts. Variations in the types of subunits (n >3) and the numbers of each type of subunit occur i n the f e r r i t i n s of different c e l l types of an organism, possibly to accommodate variations in the role of the stored iron such as normal i n t r a c e l l u l a r metabolism (house keeping), detoxification, long-term storage in specialized c e l l s , or rapid recycling of iron in specialized c e l l s which process iron from old c e l l s . Variations i n subunits of the protein coat are superimposed on a highly conserved sequence, which is found not only i n a l l the f e r r i t i n of different c e l l s in an animal but also i n different animals. Variations also occur in the structure of the iron core of f e r r i t i n and include differences in size, composition (e.g. phosphate and, possibly, other anions), and degree of order. The functional significance of differences i n core structure and the r e l a t i o n to difference in protein structure are only beginning to be explored. F e r r i t i n forms by addition of iron to assembled protein coats, which are stable even with no iron. The protein coats of f e r r i t i n can bind a v a r i e t y of metal i o n s , e.g. Fe, Cd, Mn, V, Tb, and Zn, but to date, only Fe has been observed to form a core. In v i t r o , Fe(II) is required to i n i t i a t e core formation. The steps include binding to the protein coat, oxidation and migration to form an Fe(lII) cluster on the protein (or migration and oxida tion on a cluster already formed on the protein), followed by the addition and oxidation of hundreds to thousands of Fe atoms; a



9. THEIL


193

subfraction of the i r o n , added to protein coats i n small amounts to r e s t r i c t core s i z e , has been detected as a putative Fe(II)-0Fe(III) binuclear species i n d i c a t i v e of a possible intermediate step. The l o c a t i o n of the Fe(III) cluster i n the protein coat i s not yet known, but consideration of the structure of the groove on the inner surface of the subunit dimer interface, the effect of covalent l i n k i n g of subunits into dimers, and the a b i l i t y of the dimer interface to orient i r o n core growth toward the hollow center of the protein coat suggest i t i s a reasonable s i t e to consider for the cluster and core nucleation. Complete conversion of bulk Fe(II) to polynuclear Fe(III) appears to be r e l a t i v e l y slow (16-24 hours at room temperature and pH = 7.0), although sequestration of the Fe(II) i s r e l a t i v e l y r a p i d . The detection both of Fe(III) clusters on the protein and the apparent control by the protein of Fe(II) sequestration and oxidation emphasize the important contribution of the f e r r i t i n protein coat to iron core formation, as well as providing indices for assessing the effect of variations i n the structure of the protein coats. In addition, the study of the i n t e r r e l a t i o n s h i p between the f e r r i t i n protein coat and the structure and formation of the iron core should not only c l a r i f y the significance of the structure of f e r r i t i n related to function, but should also provide lessons for understanding the formation of iron biominerals and mineralization on organic surfaces i n general. Acknowledgments The author has been fortunate to collaborate on much of this work with Dale E . Sayers, N. Dennis Chasteen, Boi Hanh Huynh, and A l a i n Fontaine, and to have received support from the North Carolina A g r i c u l t u r a l Research Service and the National Institutes of Health (grants DK20251 and GM34675). Literature Cited

1. 2. 3. 4. 5.

6. 7. 8.

Neilands, J.B. Microbial Iron Metabolism; Academic Press: New York, 1974; p 25. Biedermann, G.; Schindler, P. Acta Chem. Scand. 1957, 11, 731740. Theil, E.C. Ann. Rev. Biochem. 1987, 56, 289-315. Massover, W.H. J. Mol. Biol. 1978, 123, 721-726. Rice, D.W.; Ford, G.C.; White, J . L . ; Smith, J.M.A.; Harrison, P.M. In Advances in Inorganic Biochemistry; Theil, E.C.; Eichhorn, G.L.; Marzilli, L.G., Eds.; Elsevier: New York, 1983; Vol. 5, pp 39-50. Ford, G.C.; Harrison, P.M.; Rice, D.W.; Smith, J.M.A.; Treffry, A. Philos. Trans. R. Soc. Lond. B Biol. Sci. 1984, 304, 561-565. Dickey, L.F.; Sreedharan, S.; Theil, E.C.; Didsbury, J.R.; Wang, Y.-H.; Kaufman, R.E. J. Biol. Chem. 1987, 262, 79017907. Didsbury, J.R.; Theil, E.C.; Kaufman, R.E.; Dickey, L.F. J. Biol. Chem. 1986, 261, 949-955.


194 9. 10. 11. 12.


13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

METAL CLUSTERS IN PROTEINS Leibold, E.A.; Munro, H.N. J. Biol. Chem. 1987, 262, 73357341. Santoro, C.; Marone, M.; Ferrone, M.; Constanzo, F.; Colombo, M.; Minganti, C.; Cortese, R.; Silengo, L. Nucleic Acids Res. 1986, 14, 2863-2876. Constanzo, F.; Colombo, M.; Staempfli, S.; Santoro, C.; Marone, M.; Rainer, F.; Hajo, D.; Cortese, R. Nucleic Acids Res. 1986, 14, 721-736. Clegg, G.A.; Fitton, J . E . ; Harrison, P.M.; Treffry, A. Prog. Biophys. Mol. Biol. 1981, 36, 53-86. Stiefel, E.F.; Watt, G.D. Nature 1979, 279, 81-83. Yariv, J.; Kalb, J.; Sperling, R.; Banninger, E.R.; Cohen, S. G.; Ofer, S. Biochem. J. 1981, 197, 171-175. Moore, G.R.; Mann, S.; Bannister, J.V. J. Inorg. Biochem. 1987, 28, 329-336. Wetz, K.; Crichton, R.R. Eur. J. Biochem. 1976, 61, 545-550. Mertz, J.R.; Theil, E.C. J. Biol. Chem. 1983, 258, 1171911726. Treffry, A.; Harrison, P.M. Abstracts, Proc. 8th Int. Conf. on Proteins of Iron Transport and Storage, 1987, p 8. Levi, S.; Luzzago, A.; Ruggeri, G.; Cozzi, A.; Campanini, S.; Arioso, P.; Cesarini, G. Abstracts, Proc. 8th Int. Conf. on Proteins of Iron Transport and Storage, 1987, p 53. Watt, G.D.; Frankel, R.B.; Papaefthymiou, G.C. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 3640-3643. Mann, S.; Bannister, J.V.; Williams; R.J.P. J. Mol. Biol. 1986, 188, 225-232. St. Pierre, T.G.; Bell, S.H.; Dickson, D.P.K.; Mann, S.; Webb, J.; Moore, G.R.; Williams, R.J.P. Biochim. Biophys. Acta 1986, 870, 127-134. Bell, S.; Weir, M.P.; Dickson, D.P.K., Gibson, J.F.; Sharp, G.A.; Peters, T.J. Biochim. Biophys. Acta 1984, 787, 227-236. Granick, S.; Hahn, P.F. J. Biol. Chem. 1944, 155, 661-669. Mansour, A.N.; Thompson, C.; Theil, E.C.; Chasteen, N.D.; Sayers, D.E. J. Biol. Chem. 1985, 260, 7975-7979. Yang, C.-Y.; Bryan, A.M.; Theil, E.C.; Sayers, D.E.; Bowen, L.H. J. Inorg. Biochem. 1986, 28, 393-405. Harrison, P.M.; Fischbach, F.A.; Hoy, T.G.; Haggis, G.A. Nature 1967, 216, 1188-1190. Massover, W.H.; Crowley, J.M. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 3847-3851. Towe, K.M.; Bradley, W.F. J. Colloid. Interface Sci. 1967, 24, 384-392. Schwertmen, V.; Schulze, D.G.; Murad, E. Soil Sci. Soc. Amer. J. 1982, 46, 869-875. Mann, S.; Frankel, R.B.; Blakemore, R.P. Nature 1984, 310, 405-407. Hoy, T.G.; Harrison, P.M. Brit. J. Haem. 1976, 33, 497-504. Treffry, A.; Harrison, P.M. Biochem. J. 1979, 181, 709-716. Treffry, A.; Harrison, P.M. J. Inorg. Biochem. 1984, 21, 920. Chasteen, N.D.; Theil, E.C. J. Biol. Chem. 1982, 257, 76727677.


9. THEIL



36.

195

Wardeska, J.G.; Viglione, B.; Chasteen, N.D. J. Biol. Chem. 1986, 261, 6677-6683. 37. Theil, E.C. In Advances in Inorganic Biochemistry; Theil, E.C.; Eichhorn, G.L.; Marzilli, L.G., Eds.; Elsevier: New York, 1983; Vol. 5, pp 1-38. 38. Rosenberg, L.P.; Chasteen, N.D. In The Biochemistry and Physiology of Iron; Saltman, P.; Hegenauer, J., Eds.; Elsevier Biomedical: New York, 1982; pp 405-407. 39. Yang, C.-Y.; Meagher, A.; Huynh, B.H.; Sayers, D.E.; Theil, E.C. Biochemistry 1987, 26, 497-503. 40. Chasteen, N.D.; Aisen, P.; Antanaitis, B.C. J. Biol. Chem. 1985, 260, 2926-2929. 41. Melino, G.; Stefanini, S.; Chiancone, E.; Antonini, E. FEBS Lett. 1978, 86, 136-138. 42. Mayer, D.E.; Rohrer, J.S.; Schoeller, D.A.; Harris, D.C. Biochemistry 1983, 22, 876-880. 43. Treffry, A.; Sowerby, J.M.; Harrison, P.M. FEBS Lett. 1978, 95, 221-224. 44. Rohrer, J.S.; Joo, M.-S.; Dartyge, E . ; Sayers, D.E.; Fontaine, A.; Theil, E.C. J. Biol. Chem. 1987, 262, 1338513387. 45. Murch, B.P.; Boyle, P.D.; Que, L., Jr. J. Am. Chem. Soc. 1985, 107, 6728-6729. 46. Gorun, S.M.; Lippard, S.J. J. Am. Chem. Soc. 1985, 107, 4568-4570. 47. Gorun, S.M.; Lippard, S.J. Nature 1986, 319, 666-668. 48. Wieghardt, K.; Pohl, K.; Jebril, I.; Huttner, G. Angew. Chem., Int. ed., 1984, 230, 77-78. RECEIVED

December 18, 1987


Metal Clusters in Proteins - American Chemical Society

Recommend Documents