THE JOURNAL OF BIOLOGICAL CHEMISTRY
`© 1984 by The American Society of Biological Chemists, Inc.
`
`Vol. 259, No. 6, Issue of March 25, pp. 3620-3624, 1984
`Printed in U.S.A,
`
`Iron-catalyzed Hydroxy] Radical Formation
`STRINGENT REQUIREMENT FOR FREE IRON COORDINATIONSITE*
`
`Ernst Graft§, John R. Mahoney, Robert G. Bryant|| **, and John W. Eatoni +} §§
`From the {Departments of Laboratory Medicine and Pathology, ||Chemistry, and t{Medicine, University of Minnesota,
`Minneapolis, Minnesota 55455
`
`(Received for publication, August 8, 1983)
`
`The rapid and nonspecific reactivity of OH renders this
`free radical particularly dangerous. It may abstract hydrogen
`from, or hydroxylate, most biomolecules, causing cell injury
`or death. Hydroxyl radical is believed to be the etiological
`agent for several diseases and mayalso be involvedin natural
`aging.
`Underphysiological conditions,ironis only slightly soluble,
`tending to form precipitates with anions such as OH™ and
`inorganic phosphate. However, a variety of chelating agents
`greatly increase the solubility of iron. For example, the addi-
`tion of EDTA to an 03-generating system in the presence of
`Fe** markedly potentiates cytotoxicity (4), OH formation (4,
`7), and lipid peroxidation (8), presumably by increasing the
`effective concentration of Fe** at neutral pH. Similarly, ADP
`and ATP form iron chelates of high catalytic activity (9, 10).
`However, the powerful iron chelators DTPA and Desferal
`(desferrioxamine B methanesulfonate) totally inactivate Fe**
`for previously unexplained reasons(4, 8).
`We have investigated the chemical basis for the different
`behavior of iron chelates with respect to radical formation.
`Recognizing the importance of aquosites in transition metal
`catalysis (11) and the crucial role of heme-associated water
`in spontaneous hemoglobin oxidation (12), we postulated that
`ÿ 
`for catalytic activity, iron requires at least one coordination
`ÿ 
`site that is open or occupied by a readily dissociable ligand
`ÿ
`such as water or azide. We tested this hypothesis by deter-
` ÿ
`
`mining coordinated water in a number of disparate iron
`
`chelates spectroscopically and by measuring their catalytic
`ÿ
`activity in the formation of OH using an enzymatic O3-
`

ÿ
`generating system.
`
`
`
`8107‘ZZJsnBnyUojsonsAq/S10°9q['MMM//:dy)yWOYpopeojuMog
`
`EXPERIMENTAL PROCEDURES
`
`Materials
`
`Materials were purchased from the following sources: acetylace-
`tone, ADP, ATP, dimethyl sulfoxide, DTPA, EDTA, EGTA, EHPG,
`hypoxanthine, sodium ascorbate, sodium azide, sodium phytate, and
`xanthine oxidase from Sigma; 2,3-dihydroxybenzoic acid and nitrilo-
`triacetic acid from Aldrich; FeSO, and FeCl,-6H,0 from J. T. Baker
`Chemical Co.; H,O, from Mallinckrodt Chemical Works; Desferal
`from Ciba Pharmaceuticals, Inc.; and bleomycin from Bristol Labo-
`ratories. All other chemicals used were of analytical grade.
`
`Hydroxyl Radical Formation
`
`Hypoxanthine-xanthine Oxidase System—O} was generated by hy-
`poxanthine and xanthine oxidase. The formation of OH was moni-
`tored by reaction with dimethyl sulfoxide which yields formaldehyde
`(13, 14). Triplicate 1.0-ml samples of 50 mM Tris, pH 7.4, 50 um Fe™,
`500 uM chelator, 50 mm dimethyl sulfoxide, 300 uM hypoxanthine,
`and 18 milliunits of xanthine oxidase were incubated at 37 °C for 30
`min. The reaction was initiated by the addition of the enzyme and
`terminated with 50 yl of 100% trichloroacetic acid. The blanks were
`incubated in the absence of enzyme and stopped with trichloroacetic
`acid, followed by the addition of xanthine oxidase. Formaldehyde was
`
`3620
`
`MAIA Exhibit 1045
`
`IPR PETITION
`
`MAIA V. BRACCO
`
`  
` ÿÿ
`
`
`
`The catalysis by iron of the formation of reactive
`oxygen species in biological systems has been well
`documented. In this present study, we have investi-
`gated the hypothesis that iron-catalyzed formation of
`hydroxyl radical (OH) from superoxide anion radical
`(03) and H2Q, requires the availability of at least one
`iron coordination site that is open or occupied by a
`readily dissociable ligand such as water. This hypoth-
`esis was tested by measuring the catalytic activity of
`12 different iron chelates using hypoxanthine and xan-
`thine oxidase to generate 03. In these same chelates,
`we also determined the presence or absence of coordi-
`nated water by UV-visible spectroscopy and 1H NMR
`relaxation measurements. Of all chelates tested, only
`Fe** coordinated to diethylenetriamine pentaacetic
`acid; ethylenediamine di(o-hydroxyphenylacetic acid),
`phytate, and Desferal lacked coordination water; and
`only these four complexes failed to produce hydroxyl
`radical. Separate determinations of the two redoxhalf-
`reactions involved(i.e. Fe** + 0; — Fe** + Oz and Fe**
`+ H.02 — Fe** + ‘OH + OH’)indicate that an available
`coordination site is necessary for the latter (Fenton)
`reaction. This principle governing iron reactivity may
`help advance our understanding of the mechanism of
`oxidative damage in biological systems and may also
`permit the design of more effective chelators for the
`control of iron in biological systems.
`
`Stimulated phagocytes produce large quantities of O7'
`which promote the destruction of ingested microorganisms
`(1) and of erythrocyte targets in a model system (2). OH has
`been implicated as a derivative toxic species (3, 4), and the
`mechanism of iron-catalyzed OH formation has been well
`established (5, 6).
`
`O3 + Fe** — Fe** + 0;
`
`H;02 + Fe** + Fe** + ‘OH + OH-
`
`* The costs of publication of this article were defrayed in part by
`the payment of page charges. This article must therefore be hereby
`marked “advertisement” in accordance with 18 U.S.C. Section 1734
`solely to indicate this fact.
`§ Supported by National Institutes of Health Grants HL-16833
`and AI-18823.
`1] Supported by National Institutes of Health Grant HL-16833.
`** Supported by National Institutes of Health Grant GM-29428
`and National Science Foundation Grant PCM-8106054.
`§§ Supported by National Institutes of Health Grant HL-16833.
`Recipient of a National Institutes of Health Research Career Devel-
`opment Award.
`1The abbreviations used are: 03, superoxide anion radical; OH,
`hydroxy] radical; DTPA, diethylenetriamine pentaacetic acid; EGTA,
`ethylene glycol bis(S-aminoethy! ether)-N,N,N’ ,N’-tetraacetic acid;
`CDTA,
`trans-1,2-diaminocyclohexane-N,N,N',N’-tetraacetic acid;
`EHPG, ethylenediamine di(o-hydroxyphenylacetic acid).
`
`
`
`
`MAIA Exhibit 1045
`MAIA V. BRACCO
`IPR PETITION
`
`

`

`Free Coordination Sites and Iron Reactivity
`
`cat1
`
`3621
`
`Os
`
`3 o4
`
`; 03o2
`
`0.
`0
`
`350
`
`500
`
`450
`400
`Wavelength (nm)
`Fic. 1. Effect of azide on visible absorption of Fe**-EDTA.
`Solutions of 0.3 mm Fe**, 5.0 mm EDTA, 50 mm Tris, pH 7.4,
`containing increasing amounts of NaN; (50, 100, 200, 500, 1000, 1500
`mM)were scanned versus Fe**-EDTA in Tris, pH 7.4.
`
`ge
`T —— +
`r
`
`
`
` ÿÿ
`
`
`
`
`
`
`
`8107‘ZZIsNBNYUOyson3Aq/S10'0qfMMM//:d}}YWoOpopeojuMog
`
`ÿ ÿ
`ÿ ÿ
`
` ÿ

ÿ 
`
`  
`
`(NzJ(m)
`Fic. 2. Azide dependence of absorption intensity of N3-
`Fe*+. EDTA. Experimental conditions were the same as in Fig. 1.
`All absorption readings were corrected for the minor contribution of
`azide alone. The curve shown is a computer-generated binding iso-
`therm (Equation 1) using the parameters obtained from the double
`reciprocal plot shown in the inset.
`
`ment between experimentaland theoretical binding isotherms
`supports the assumption that the only equilibrium in consid-
`eration is that between Nj and Fe**-EDTA.Theaffinity of
`N; for Fe** is too low to displace any EDTA based on both
`theoretical considerations and on the observations that 1) no
`N;- Fe** precipitate could be detected even after several days,
`2) all of the above spectral properties were independent of
`EDTAconcentration, and 3) no spectral change was observed
`upon the addition of azide to other iron chelates of similar
`Fe**-chelator association constants containing no aquo coor-
`dinationsite.
`This spectroscopic method permits ready determination of
`the presence offirst coordination sphere water in a number
`of different Fe**-chelates. Table I demonstrates thatall but
`four chelates contained at least one coordinated water mole-
`cule. The maximum wavelength, the extinction coefficient of
`the complex, and the association constant between Nz and
`Fe**-chelates are highly variable and depend on the type of
`bonding between Fe** and chelator, the stereochemistry of
`the complex, and the numberof coordination positions. A
`complete interpretation of these spectral data, however, is
`beyond the scope of the present study. A spectroscopictitra-
`tion of Nz with Fe** - nitrilotriacetic acid similar to that shown
`in Fig. 2 revealed an extinction coefficient of 4257 cm™' m7!
`and an association constant of 2.49 M~!. No biphasic character
`of the binding isotherm was observed, suggesting the presence
`of only one watersite.
`Theability of the optical spectroscopic method to determine
`the presence of coordinated water was confirmed by nuclear
`
`determined colorimetrically by the Hantzsch reaction (15). Equal
`volumes of the above trichloroacetic acid extracts and of a solution
`containing 50 mM acetic acid, 20 mM acetylacetone, and 2 M ammo-
`nium acetate were incubated at 37 °C for 40 min in the dark, and the
`absorption of the resultant diacetyldihydrolutidine was measured at
`412 nm. Exact timing and dimming of lights during spectrophoto-
`metric assay were very important due to the instability and light
`sensitivity of diacetyldihydrolutidine.
`Fe**-Ascorbate System—Triplicate samples of 2 mM ascorbate, 167
`uM Fe**, 50 mM dimethyl sulfoxide, 2.0 mM chelator, 50 mM Tris, pH
`7.4, were incubated at 37°C for 30 min, and formaldehyde was
`determined by the Hantzsch reaction as described above.
`
`Decomposition of HsOz
`Triplicate 1.0-ml samples of 0.3 mM Fe**, 0.5 mM EDTA, 0.1 mm
`H.02, 100 mM NaN, and 50 mM Tris, pH 7.4, were incubated at
`37 °C for 10 min. Hydrogen peroxide was determined colorimetrically
`by the iodine-starch method (16). Hydroxy] radical generation could
`not be measured dueto the interference by Nz, a good OH scavenger
`(17).
`
`'H NMRRelaxation
`
`1H NMR relaxation measurements were conducted using a spe-
`cially constructed field cycling nuclear magnetic relaxation spectrom-
`eter described previously (18). In the present determinations, the
`measure field was 7.25 MHz, and the probe was of the Helmholtz
`single coil design. Each relaxation rate is the result of at least 30
`measurements on the decay curve. Samples were placed in 10-mm
`Pyrex culture tubes and thermostated at 286.8 K with liquid Freon
`via a Neslab RTE-8 refrigerated heat exchanger. 'The rf source was a
`crystal oscillator gated by Vari-L SS-30 gates, amplified by an ENI
`10-watt broad band amplifier. The receiver consisted of a Mitek
`preamplifier, a cascade of Avantek UTO-500 series amplifiers oper-
`ating into a Merrimac doubly balanced mixer. The 250-kHz inter-
`mediate frequency was detected using an absolute valve detector. A
`two-pulse sequence was used at the measure field to generate a spin
`echo, the amplitude of which was captured using a sample and hold
`circuit read by an IBM 7406 device coupler. The experiment was
`controlled by an IBM 5120 computer.
`
`UV-visible Spectroscopy
`All spectroscopic scans were recorded on a Beckman DU-8 spec-
`trophotometer at a scan speed of 100 nm/min. Therate of oxidation
`of Fe?* and Fe**-chelates was followed by recording the increase in
`absorption at 310 nm (Fe**), 328 nm (Fe?*.-EDTA), and 355 nm
`(Fe**. DTPA) every 12 s following the addition of Fe’* to 50 mm Tris,
`pH7.4, or Tris, 0.6 mM chelator.
`
`RESULTS
`
`As mentioned earlier, chelation of Fe** by EDTA greatly
`enhances iron-driven ‘OH generation. This may appear puz-
`zling because EDTA formsa hexadentate chelate with Fe**.
`However, it is known that the structure of the Fe**-EDTA
`complex includes a seventh coordination site occupied by
`water (19). This coordination water may be competitively
`displaced by a ligand such as N3. As shown in Fig. 1,
`Nz - Fe**- EDTAdisplays an absorption peak at approximately
`409 nm which is absent in H,O-Fe**-EDTA. This spectral
`property may be employed to quantitate the interaction be-
`tween Nz and Fe**.EDTA.Fig. 2 shows a binding isotherm,
`wherein the open circles represent the actual data points, and
`the solid line was calculated using Equation 1.
`
`AXE! x (A x E7'- 0.3 x 107% - kK")
`AxE™"-0.3 x 107%
`
`[Ns] =
`
`(1)
`
`where [N;] = total azide concentration (molar), A = absorp-
`tion at 409 nm, E = extinction coefficient of N3z-Fe**.
`EDTA at 409 nm, and K = association constant for Nz and
`Fe**.EDTA.
`The values for the parameters K and E were determined to
`be 0.61 m7! and 3171 cm7! mM“, respectively, from the double
`reciprocal plot shown by the inset in Fig. 2. The good agree-
`
`
`
`
`

`

`3622
`
`Free Coordination Sites and Iron Reactivity
`
`TABLE |
`
`Catalytic and spectroscopic correlation of various Fe**-chelates
`oa
`log K, of
`Adide-induced spectral
`iron chela-
`_....,.
`.
`shift
`tor
`Catalyticactivity
`Chelator
`
`Maximum Absorption
`Fe**
`Fe**
`
`nm
`
`TABLE II
`'H nuclear relaxation rates at 0.01 MHzfor several iron(II)
`
`complexes
`
`ee
`
`Complex*
`
`YT,
`=
`7.2
`Fe(III)-EDTA
`5.7
`Fe(III): EDTA + Nz
`4.0
`Fe(III): DTPA
`4.0
`Fe(III). DTPA + Nz
`10.5
`Fe(III)-ATP
`5.5
`Fe(III)- ATP + Ng
`6.9
`Fe(III) - Desferal
`
`71
`Fe(III) -Desferal + Ny
`* All complexes contained 2.0 mm Fe** and 5.0 mM chelating agent
`in 50 mM Tris, pH 7.4. The azide ion concentration was 1000 mm.
`
`4 4
`
`4
`
`ee
`
`wi
`
`|
`26-
`
`t
`
`None
`EDTA
`EGTA
`NTA?
`CDTA
`Phytate
`EHPG
`DTPA
`Desferal
`ADP
`
`ATP
`
`Bleomycin
`DHBA
`
`nmols HCHO/
`30 min
`81.1414
`16.1 + 0.3
`10.4 + 0.5
`23.8 + 0.3
`10.4+0.1
`0.0
`0.0
`0.0
`0.0
`63.2 + 1.0
`
`409
`456
`404
`371
`Noshift
`Noshift
`No shift
`Noshift
`347
`426
`340
`443
`356
`22.1 + 0.7
`400
`69.5+ 1.9
`
`490
`
`11.9 + 0.8
`
`0.391
`0.619
`0.945
`0.453
`
`14.3° 25.0°
`11.8° 20,5°
`8.3°
`15.9°
`18.9° 30.0°
`
`14.3° 33.9°
`16.4° 28.0°
`30.7°
`
`21.4
`
`0.940
`0.842
`0.855
`0.789
`0.471
`0.260
`0.314
`
`* The formation of ‘OH in the presence of 50 uM Fe** and 500 um
`chelator was determined by the xanthine oxidase method as described
`under “Experimental Procedures.”
`> The peak positions and their absorption intensity were obtained
`from differential spectroscopic scansof 0.3 mM Fe**, 5.0 mM chelator,
`1.0 M NaNs, 50 mM Tris, pH 7.4, versus the same solution minus
`azide.
`‘From Martell and Smith (23).
`4 NTA,nitrilotriacetic acid; DHBA,2,3-dihydroxybenzoic acid.
`* From Crumblisset al. (24).
`‘From Martell and Smith (25).
`
`
`
`
`(s)mB
`
`
`
`
` oRELAXATIONRATE,st
`
`
`
`
`ooa“T_T.TT11
`
`
`8107‘TZISNBNYUOyson3Aq/S10'0qfMMM//:dyYWoOpopeopuUMog
`
` ÿÿ
`
`1.1_
`
`
`Lo
`
`  
`
`ÿ 
`ÿ
`ÿ
`
`3 8
`ÿ
`
`
` ÿ

ÿ 
`
`2
`
`i
`2
`
`tt
`
`|
`
`2
`
`_t@TOTOT
`
`HCHOProduction{nmoles) ——o__
`
`
`
`°h
`
`22
`
`[chelator] / [Fe°*]
`Fic. 4, Effect of chelator/iron ratio on hydroxyl radical
`formation. The assay was performed using 300 uM hypoxanthine,
`18 milliunits of xanthine oxidase, and 50 um Fe** as described under
`“Experimental Procedures.” Each point represents the mean and
`standard deviation of three determinations. Solid symbols are EDTA/
`Fe**-chelate and open symbols represent phytate/Fe**-chelate.
`
`
` id 10
`
`FREQUENCY, MHz
`Fic. 3. The ’H nuclear magnetic relaxation rates for 20 mM
`Fe**.5.0 mm EDTA complex at pH 7.4 in the presence (@) and
`absence (A) of 1000 mm N; measured at 286.8 K as a function
`of magnetic field strength plotted as the ‘H Larmorfrequency.
`
`magnetic relaxation which is also sensitive to water partici-
`pation in the first coordination sphere of paramagnetic ions
`such as Fe(III). It is clear both from the frequency dependence
`(Fig. 3) and the data in Table II that the addition of azide ion
`decreases the 'H relaxation rate only for those complexes
`where dramatic changesin the optical spectrum are observed
`on the addition of azide ion.
`Analysis of the capability of all these Fe*t complexes to
`catalyze the formation of ‘OH from O03 and H,O, demon-
`strates a direct relationship between the presenceofa first
`coordination sphere water molecule on Fe** and the catalytic
`activity of the complex (Table I). Virtually identical results
`
`were obtained with the Fe**-ascorbate system. Thus, spectro-
`scopic determination of azide binding is a simple andreliable
`method for predicting the oxidative reactivity of a particular
`Fe**-chelate. In contrast, the correlation between the associ-
`ation constantfor chelating agent with iron(IID) and catalytic
`reactivity of the chelate is very poor (Table I).
`The effect of the EDTA/Fe** ratio on ‘OH formation
`demonstrates the relative reactivities of Fe(H.O)3* and H.0-
`Fe**.EDTA (Fig. 4). Freshly prepared Fe(H.O)%* is very
`soluble and, as expected, a superior redox catalyst. Upon
`titration with EDTA, the activity decreases down to the
`equivalence point and,
`thereafter,
`is independent of the
`EDTA/Fe™*ratio. At the sametime, the extinction coefficient
`of Nz-Fe**-EDTA at 409 nm remains constant between
`EDTA/Fe*ratios of 1 to 20 (data not shown). Unlike H.O-
`Fe**.EDTA, mono-, di-, tri-, and tetraferric phytate lack
`available aquo coordination sites as determined spectroscop-
`ically and, indeed, produce no ‘OHatanyratio (Fig. 4).
`To investigate further the role of aquo coordination sites
`
`
`
`
`

`

`Free Coordination Sites and Iron Reactivity
`
`3623
`
`in iron catalysis of OH production, we measured the spon-
`taneous oxidation rates of Fe** and Fe**-chelates (Fig. 5).
`Although DTPA blocked iron-driven ‘OH formation, this
`chelator only slowed, but did not prevent, oxidation of Fe**
`to Fe’. The oxidation of Fe?t.DTPA followed first order
`kinetics with a t,, of 525 s, which wasonly slightly lower than
`that of Fe** alone (t, = 151s). The addition of EDTA to Fe**
`accelerated Fe** formation (t, = 16 s), and the reaction
`approached second order kinetics due to the rapid oxygen
`consumption. As expected, the oxidation rates of H.0-Fe*.
`EDTAand N;-Fe?*.EDTA were found to be identical. Also,
`100 mm Nzhad noeffect on the decomposition of H,O2 by
`Fe*t. EDTA, since both water and azide should be readily
`displaceable by HO, from the iron coordinationsite.
`DISCUSSION
`
` 05
`
`REFERENCES
`
`One
`
`* E. Graf, J. R. Mahoney, and J. W. Eaton (1984) Science (Wash.
`D. C.), submitted for publication.
`
`!
`1000
`
`oa
`
`Fe
`eEDTA
`
`03
`
`==
`
`2 02
`Ol
`
`0
`
`_
`200
`
`0
`
`.
`600
`Time (s)
`Fic. 5. Oxidation rate of Fe** and Fe?*-chelates. The oxida-
`tion of Fe** to Fe** was followed spectrophotometrically at 25 °C as
`described under “Experimenta] Procedures.” Thesolutions contained
`0.5 mM Fe**, 0.6 mM chelator, 50 mM Tris, pH 7.4. The reactions
`were initiated by the addition of Fe.
`
`the insolubility of iron in phosphate buffers and suggests that
`the latter should not be used to assess iron-mediated OH
`generation.
`important
`The observations reported here have several
`implications. For example, in view of the facilitation of iron-
`mediated ‘OH generation by iron-EDTAchelates,it might be
`wise to reconsider the use of EDTA as a preservative of
`enzymes, resins, and food items. EDTA might be replaced
`with phytate, a stable and abundant polyphosphate which
`comprises 1-6% by weightofall plant seeds. Its main function
`in seeds is believed to be protection against oxidative damage
`during storage.? Indeed, as shown herein, the iron-phytate
`chelate is totally inert in the Fenton reaction. We have
`successfully applied this property to the preservation of fresh
`fruits and vegetables.” A multitude of other applications based
`on this antioxidant property of phytic acid have been sum-
`marized in a recent review (21).
`The above results demonstrate that in free iron and iron
`Of equal importance are the medical applications of our
`chelates, the availability of at least one coordination site is
`results. A numberof drugs are strong iron-chelating agents,
`required for catalysis of ‘OH generation. Iron catalysis of
`and their potential for catalyzing ‘OH production mayeasily
`‘OH formation from O3 and HQ, consists of two half-reac-
`be predicted by the spectroscopic methods reported herein.
`tions: the reduction of Fe** by 03, and the oxidation of Fe**
`One example is the natural antibiotic bleomycin whose chem-
`by H,0,. The availability of a coordination site thatis free or
`istry, biochemistry, and clinical effectiveness have been re-
`occupied by an easily displaceable ligand such as water facil-
`viewed extensively (22). Fe*t-bleomycin possesses an iron-
`itates the first half-reaction, yet it is not a stringent require-
`coordinated water molecule and is a good catalyst for ‘OH
`ment, as demonstrated by the differential oxidation rates of
`formation (Table I). This property of the iron-bleomycin
`Fe**.EDTA and Fe**-DTPA by O2. However, occupation of
`chelate may help explain the well known pulmonary toxicity
`all iron coordination sites by a chelator with displacement of
`of bleomycin. In contrast, the clinically useful iron chelator
`water in the first coordination sphere precludes the binding
`Desferal does not accelerate iron-driven oxidation and may
`of HO, to Fe’*-chelates and the subsequent Fenton reaction.
`
`actually suppress such reactions in vivo.
`Thus, chelation of iron by DTPA, phytate, EHPG, or Desferal
`
`In summary, our results support the idea that chelates in
`excludes iron-associated water and uncouples the oxidation
`which a coordination site on iron remains open (or loosely
`of Fe?* from the formation of ‘OH.
`associated with ligands such as water) are efficient catalysts
`In the absence of chelating agents, Fe(H.0)%* rapidly cat-
`of ‘OH generation. Hindranceofall coordination sites on iron
`alyzes the formation of ‘OH. However, Fe(H2O)%* aggregates
`effectively blocks OH production, although other redox prop-
`slowly as oxygen-bridged polynuclear complexes that precip-
`erties of the chelated iron are evidently not greatly modified.
`itate. At physiological pH, the concentration of free iron at
`Should this general principle hold true for all iron chelates
`equilibrium is estimated to be 10~** M (20). Thus, under most
`(and, perhaps, chelates of other transition metals),
`it will
`circumstances, EDTA appearsto raise iron reactivity by in-
`provide a valuable basis for prediction of the behavior of iron
`creasing the available iron concentration, even though catal-
`complexes and for the design of novel chelators.
`ysis by soluble hexaaquoiron(III) ions is more efficient than
`
`ÿ
`that by H,0-Fe**.EDTA. These results contradict earlier
`reports of very low lipid peroxidation and ‘OH formation by
`

ÿ
`iron in the absence of EDTA,followed by an increase concom-
`. Rosen, H. & Klebanoff, S. J. (1979) J. Exp. Med. 149, 27-39
` 
`. Weiss, S. J. (1980) #. Biol. Chem. 255, 9912-9917
`itant with the addition of EDTA (8). However, those earlier
`
`. Repine, J. E., Eaton, J. W., Anders, M. W., Hoidal, J. R. & Fox,
`assays were carried out in phosphate buffer, which immedi-
`R. B. (1979) J. Clin. Invest. 64, 1642-1651
`ately precipitates Fe*t. We have noticed that the release of
`4. Rosen, H. & Klebanoff, S. J. (1981) Arch. Biochem. Biophys.
`50% of Fe** from 167 uM Fe**, 80 mm phosphate, pH 7.4, to
`208, 512-519
`0.4 mM DTPArequired 177 min at 25 °C, Thisclearly reflects
`5. Walling, C. (1975) Accts. Chem. Res. 8, 125-131
`cooTtT
`6. Koppenol, W. H., Butler, J. & VanLeewen, J. W. (1978) Photo-
`chem, Photobiol. 28, 655-660
`7. McCord, J. M. & Day, E. D., Jr. (1978) FEBS Lett. 86, 139-142
`8. Gutteridge, J. M. C., Richmond, R. & Halliwell, B.
`(1979)
`Biochem. J. 184, 469-472
`9. Floyd, R. A. & Lewis, C. A. (1983) Biochemistry 22, 2645-2649
`10. Svingen, B. A., Buege, J. A., O’Neal, F. O. & Aust, S. D. (1979)
`J. Biol. Chem. 254, 5892-5899
`11. Martell, A. E., Gustafson, R. & Chaberek, S. (1957) in Advances
`in Catalysis IX (Farkas, A., ed) pp. 319-329, Academic Press,
`New York
`12. Perutz, M. F. & Lehmann, H. (1968) Nature (Lond.) 219, 902-
`908
`13. Klein, S. M., Cohen, G. & Cederbaum,A. I. (1980) FEBS Lett.
`116, 220-222
`14. Klein, 8S. M., Cohen, G. & Cederbaum,A.I. (1981) Biochemistry
`20, 6006-6012
`
`15. Nash, T. (1953) Biochem. J. 55, 416-421
`
`
`
`
`
` ÿÿ
`
`  
`
`ÿ ÿ ÿ ÿ
`
`
`
`
`8107‘TZISNBNYUOyson3Aq/S10‘0qgfMMM//:dyYWOTpopeopuUMog
`
`
`
`
`

`

`3624
`
`Free Coordination Sites and Iron Reactivity
`
`16. Graf, E. & Penniston, J. T. (1980) Clin. Chem. 26, 658-660
`17. Dorfman, L. M. & Adams, G. E.
`(1973) in Reactivity of the
`Hydroxyl Radical in Aqueous Solutions (Dorfman L. M., and
`Adams, G. E., eds) National Bureau of Standards, Washington,
`D.C.
`18. Brown, R. D., III, Brewer, C. F. & Koenig, S. H. (1977) Biochem-
`istry 16, 3883-3896
`19. Lind, M. D., Hamor, M. J., Hamor, T. A. & Hoard, J. L. (1964)
`Inorg. Chem. 3, 34-44
`20. Spiro, T. G. & Saltman, R. (1974) in Iron in Biochemistry and
`Medicine (Jacobs, A. & Worwood, M., eds) Vol. 1, pp. 1-28,
`Academic Press, New York
`21. Graf, E. (1983) J. Am. Oil Chem. Soc. 60, 1861-1867
`
`in Advances in Inorganic Chemistry
`(1982)
`22. Dabrowiak, J. C.
`(Eichhorn, G. L. & Marzilli, L. G., eds) Vol. 4, pp. 70-113,
`Elsevier Scientific Publishing Co., New York
`23. Martell, A. E. & Smith, R. M. (1974) Critical Stability Constants,
`Vol. 1, Plenum Press, New York
`24. Crumbliss, A. L., Palmer, R. A., Sprinkle, K. A. & Whitcomb,D.
`R. (1975) in Symposium on the Development of Iron Chelators
`for Clinical Use (Anderson, W. F. & Hiller, M. C., eds) pp. 175-
`212, (Departmentof Health, Education, and Welfare, Bethesda,
`Publication NIH 76-994
`25. Martell, A. E. & Smith, R. M. (1982) Critical Stability Constants,
`Vol. 5, Supple. 1, Plenum Press, New York
`
`
`
`
`
`8107‘ZZJsnBnyUoysonsAq/S10°9q'MMM//:dy1yWOYpopeojuMog
`
`  
` ÿÿ
`
`ÿ 
`ÿ 
`ÿ
` ÿ
`
`
`ÿ
`

ÿ
`
`
`
`
`
`
`

`

` !"#ÿ$"%"!ÿ!ÿ&'ÿ()ÿ*+'ÿ&ÿ&ÿ
`ÿ
`ÿ ÿÿÿ 
`

ÿÿÿ 

ÿ

ÿÿÿ

`
`,-./0ÿ23-456278562/9:9ÿ;<=>9ÿ?@AB9ÿ
`YMMRZZÿEDRÿ[NZEÿ\F]^ER]ÿ_ROZ`NQÿNaÿED`Zÿ^OE`MbRÿ^Eÿ ÿDEEFGHHIIIJKLMJNOPHMNQERQEHSTUHVHWVSX
`
`
`ÿÿ
`
`ÿYbROEZGÿÿÿ
`
`
`
`gb`MjÿDROR ÿcDRQÿED`Zÿ^OE`MbRÿ`ZÿM`ER]dÿÿÿENÿMDNNZRÿaON[ÿ^bbÿNaÿefghZÿRi[^`bÿ^bROEZ
`
`
`dÿÿ ÿcDRQÿ^ÿMNOORME`NQÿaNOÿED`Zÿ^OE`MbRÿ`ZÿFNZER]
`mD`Zÿ^OE`MbRÿM`ERZÿXÿORaRORQMRZnÿXÿNaÿID`MDÿM^QÿLRÿ^MMRZZR]ÿaORRÿ^E
`ÿDEEFGHHIIIJKLMJNOPHMNQERQEHSTUHVHWVSXJa\bbJDE[bkORaib`ZEil
`
`ÿÿ
`
`{

€ ÿ ÿ
`
`
`
`
`xyzz{
`
`{{|}o~| pz
`
`ÿo
`ÿp
`q rÿ

`
`ÿs
`qp
`
`qrÿttÿtuvw
`
`
`
`
`