THE KINETICS 0F OXYGEN UPTAKE BY 6,7-DEMETWLESTER PROTOF’ORPHYRTN EX ERONUB) EN SEVERAL 8033358 Thesis for the Degree of M: S. MICHIGAN STATE UNIVERSITY JAMES R. BAKER ' 1972 J LIBKAF; 1 ”T Michigan Sate 1; U nivetsity A = é'HBING BY .. HUM} & SUNS' 800K amnmv mc. LIBRARY amosns ”mtmcmm ABSTRACT THE KINETICS OF OXYGEN UPTAKE BY 6,7-DIMETHYLESTER PROTOPORPHYRIN IX IRON(II) IN SEVERAL SOLVENTS BY James R. Baker The kinetics of the reaction of the complex 6,7- dimethylester protoporphyrin IX iron(II) with oxygen has been investigated in benzene/pyridine and pyridine/imidazole solutions. Though the results of the reaction in the benzene/pyridine solution have been reported in a previous workl, substantial reason was found to justify a reexamination of the kinetics results. Also obtained in this work were the kinetics of this same reaction in a pyridine solution saturated with imidazole. It was previously reported2 that the latter system yielded a product that was reversible in its reaction with oxygen. In order to describe the reaction more fully, kinetics data have been obtained. In both systems a prominent visible absorption spectrum was obtained for the non-oxygenated 6,7-dimethylester protoporphyrin IX iron(II). As the complex was allowed to react with oxygen, the visible spectra showed marked alterations in their visible absorption and the intensities James R. Baker of the absorbances decreased as the reaction proceeded. By using absorbance as a measure of the concentration of the non-oxygenated complex, the decrease of the absorbance with time measures the rate of this reaction. In reexamining the reaction of the iron(II) complex with oxygen in benzene/pyridine solution a new rate equation of a mixed first- and second-order character was found, while the previous results showed mainly first-order character for the reaction. These results suggested a new mechanism should be proposed. The kinetics of this same reaction in a pyridine solution saturated with imidazole also fit a mixed-order rate equation. Unfortunately, not enough data were obtained and not enough is known about the reaction to propose a reaction mechanism. It was, however, verified by a visible spectral study that this was, indeed, a reversible oxygen carrying system. 1. I. A. Cohen and W. S. Caughey, Biochem., 1, 636 (1968). 2. A. H. Corwin and S. D. Bruck, g. Amer. Chem. Soc., g9, 4736 (1958). THE KINETICS OF OXYGEN UPTAKE BY 6,7-DIMETHYLESTER PROTOPORPHYRIN IX IRON(II) IN SEVERAL SOLVENTS BY James R. Baker A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1972 . I" V .t :1 ' 7' ‘i d 4‘ I a - J .- .' To my Mother and Father ACKNOWLEDGMENTS The author wishes to express his sincere appre- ciation for the advice and assistance of Dr. Carl H. Brubaker, Jr. under whose direction this investigation was made. The guidance of Dr. Theodore J. Williams in the initial stages of my graduate research was invaluable. The financial support of this project by the National Science Foundation is also gratefully acknowl- edged. iii, LIST OF TABLES LIST OF FIGURES INTRODUCTION EXPERIMENTAL RESULTS DISCUSSION BIBLIOGRAPHY APPENDIX A APPENDIX B TABLE OF CONTENTS iv 17 40 47 49 52 Table LI ST OF TABLES Results for Kinetics of Oxygen Uptake by Fe(II)heme Diester in Benzene/Pyridine Solutions Results for Kinetics of Oxygen Uptake by Fe(II)heme Diester in Pyridine Solutions Saturated with Imidazole 33 37 Figure 10 LIST OF FIGURES The Nitrogen Purification Train Reaction Flask for Kinetics Measurements Infrared Spectra of (A) Hemin, (B) Product of Process One, (C) Product of Process Two Oxidation of 02 of (A) Product of Process One and (B) Product of Process Two Kinetic Spectra of Fe(II)heme Diester in Benzene/Pyridine Solution Fit of Rate Data to a First-Order Equation Fit of Rate Data to a Second-Order Equation Fit of Rate Data to a Mixed-Order Equation Spectra of Fe(II)heme Diester in a Pyridine Solution Saturated with Imidazole as a Function of Time Spectra of Initial, Oxygenated, and Recycled Fe(II)heme Diester vi 19 22 25 28 3O 31 35 39 INTRODUCTION In recent years a considerable amount of interest has been expressed in the study of the reversible oxygen carrying ability of hemoglobin.1 Of course, hemoglobin was long ago identified as a component of blood and as the prime oxygen carrier in most animal systems. Heme (protoporphyrin Ix iron(II)) is the actual oxygen carrying unit of hemoglobin and can combine reversibly with oxygen and has been the subject of considerable study for many yearszv3t4r5. Throughout these studies, however, the actual oxygen to iron interaction and the detailed mechanism of how oxygen can enter through the globin for reaction with the iron heme has not yet been established. Hemoglobin6r7r8 consists of four protein subunits each containing a heme group surrounded by the polypeptide globin that consists of four subunits (a1, 02, 81, and 82). Perutz9 provides an ingenious description of the conform- ational changes which occur in the a and 8 subunits of hemoglobin when a ligand attacks each of the four heme groups. The kinetics of oxygen uptake by hemoglobin under suitably restricted conditions have been examined by Gibson and co-workerslovllrl2 by using a stopped-flow technique developed for this purpose. 2 The actual study of the heme iron to oxygen interaction is complicated by the size (over 64,000 mw unit.) and complexity of the hemoglobin molecule and suggests that a model of a reversible oxygen carrying system less complex than hemoglobin might be useful. Since it is the iron porphyrin segment of the hemoglobin molecule which is the actual oxygen carrier, isolation of this unit might well serve as the model sought and, indeed, a number of workers have isolated13r14 synthesizedls, analyzedls, and studied the kineticsl7118r19 of reactions of the iron porphyrin unit. One basic problem inherent in this approach is the need to maintain the iron in the iron(II) state throughout the reaction. In most cases iron has successfully complexed with oxygen but oxidation to the iron(III) state has also occurred and rendered the reaction irreversible. In 1956 the first reversible oxygen reaction with protoheme was described. Corwin and Reyes20 reported a crystalline diimidazole protohemochrome which combined reversibly with oxygen in the absence of water. In 1958 Corwin and Bruck21 reported first, that this same protohemochrome reversibly took up oxygen when placed in pyridine solution saturated with imidazole and, second, that a spectral change also accompanied oxygenation. However, no subsequent research has been reported with this system. This work has been an attempt to describe the protohemo- chrome system more fully by studying the kinetics of the oxygen reaction. For the sake of conVEniefice one small 3 alteration has been made, the protoheme used by Corwin and Bruck is protoporphyrin IX iron(II), which takes up oxygen almost instantaneously in most systemsl7, thus making it difficult to study. By esterifying the two carboxylic acid groups on the porphyrin ring the rate of reaction with oxygen by 6,7-dimethylester protoporphyrin IX iron(II) becomes slow enough to study by conventional techniques. By using the spectral change in the conversion of the de-oxy to oxyhemochrome, time versus absorbance data are collected and rate constants calculated. The kinetics reported by Cohen and Caughey19 were reexamined. They also used 6,7-dimethylester protoporphyrin IX iron(II) (hereafter to be identified as heme diester) in their study, but used a different solvent system. Their results show the conventional iron(II) to iron(III) oxidation upon oxygenation of the complex. However, upon repeating their work certain differences in the rate of reaction were noted which greatly affected the rate constants. The experiments have been repeated in this work, new rate constants calculated, and the reasons for the differences are discussed. EXPERIMENTAL General Procedure A11 visible spectra were recorded by use of a Unicam SP800 Ultraviolet Spectrophotometer. Infrared spectra were recorded by a Perkin-Elmer 457 Grating Infrared Spectro- photometer. All solvents were reagent grade and were purified as specified below or were used as received. Solid chemicals were used as received. Glassware was cleaned by immersion for several hours in a potassium hydroxide/ethanol solution, then by several hours immersion in a dilute hydrochloric acid bath, followed by thoroughly rinsing in water, and, finally, by drying in an oven at about 130°. Purification‘gf Solvents Methanol:22 Clean, dry Mg turnings (10 g) and 12 (l g) were placed in a 3 i flask fitted with a reflux condenser. Methanol (100 ml) was added through the condenser and the mixture was warmed on a water bath until all the Mg was converted to magnesium methoxide. Methanol (2 2) was next added into the flask and the mixture refluxed for 30 minutes. The methanol was then distilled, moisture was excluded, and the 65-66° fraction was collected. 5 Chloroform: This solvent was washed three times with water in a seperatory funnel, dried over CaC12 and filtered before use. Benzene: The solvent was washed twice with concen- trated sulfuric acid, three times with water, and refluxed over CaO before use. Pyridine: The pyridine was passed through an alumina column and refluxed over BaO before use. Acetone: Acetone was dried over CaClZ and filtered before use. Trifluoroacetic acid: The acid was used as received. Purification of Gases Oxygen: Oxygen (received from Liquid Carbonic, industrial grade) was passed through a column (4 cm by 45 cm) of potassium hydroxide before use. Nitrogen: It was vitally important to use oxygen-free nitrogen during the preparation and reaction of many of these compounds in order to prevent oxidation. This nitrogen (Central Welding Supply, Lansing, MI, prepurified) was further purified in the following manner; the nitrogen was passed through a purification train23 which consisted of the following components, connected in series, and schematically illustrated in Figure 1; (A) a column (4 cm by 40 cm) of BTS catalyst (from BASF Corp.) heated to about 140°, Figure l The Nitrogen Purification Train ms 1““? \/J catalyst (A) ’CrSOu 4. VA. wzn(Hg) 9“ ‘T‘ heating ? _,,+ .._4 . coil ‘7" 0.15 g . ‘ H SO ’1 , 2 x { ; (BL; ~ J 3 ;‘ ,J" g 2 . a: “”1 .1_.L __1; ‘ to reaction l flask ; vi aquasorb (D) \_/1 cold trap (C) 8 (B) two identical gas scrubbing towers each containing chromium(II) sulfate solution in 0.15 N sulfuric acid and zinc amalgam, (C) a dry ice/acetone cold trap, (D) a column (70 cm by 3 cm) of aquasorb (Mallenkrodt Chemical). Component A is the high capacity oxygen remover, while component B should remove any trace oxygen remaining. Components C and D insure that no water is allowed to enter the reaction areas. The BTS catalyst was prepared and used as directed by instructions supplied by BASF Corp. The chromium(II) sulfate solution used for oxygen removal was prepared as follows:27 a solution 0.15 N in sulfuric acid is initially prepared and added to the gas scrubbing towers. The zinc amalgam was prepared by adding 20 g of Zn, 0.6 g HgClz, and 40 m1 of 1.0 M HCl in a beaker and stirring for several minutes. The resulting solid was washed several times with distilled water and added to the gas scrubbing towers. Nitrogen was then bubbled through the solution while chromium(II) was prepared. Several pieces of chromium metal (about 10 g) are placed in enough concentrated HCl to cover all the pieces and remain in the HCl until vigorous reaction proceeds. The HCl was then decanted, the metal washed several times with water, and then was added to the gas scrubbing towers. To facilitate the dissolving of the chromium in the acid solution, the solution was heated to 9 about 40°. The solution was ready for use when a blue color appeared. Esterification of Hemin, Process One24 Hemin (from Eastman-Kodak, 97+% pure), 0.1 g, was dissolved in 25 m1 of methanol containing 0.2 g KOH. Then 10 ml of trifluoroacetic acid was added to this solution and the solution was refluxed for 10-15 hours. This solution was evaporated to dryness and 50 m1 of acetone was added. After dissolving as much residue as possible in the acetone, the solution was filtered, and again evaporated to dryness. The residue was then treated with three successive 20 m1 portions of chloroform, each was filtered through the same fine-fritted filter. The filtrate was dried overnight at 50° in a desiccator. Esterification of Hemin, Process Tw025 A solution containing 50 m1 of cold, absolute methanol and 2.6 g of concentrated sulfuric acid was prepared. In this solution 0.2 g of hemin (Eastman-Kodak, 97+% pure) was dissolved and allowed to stand in the refrigerator (ca. 5°) for 48 hours. The solution was transferred to a 500 m1 seperatory funnel containing 150 ml dry chloroform and the contents were mixed well. It was washed twice with 100 ml portions of water, twice with 100 ml portions of 2 N NH4OH, twice again with 100 ml portions of water, twice with 100 m1 of l N HCl, and, finally, three times with 100 ml portions of 10 water. Then the chloroform layer was separated, placed in a 1000 ml round bottom flask, and taken to dryness in a rotary evaporator. By use of acetone (ca. 50 ml), the solid residue was transferred to a small flask (100 ml) to facilitate product collection and the solution was again taken to dryness. Finally, the product was scraped off the sides of the flask, powdered as much as possible, and dried overnight under vacuum. Crystallization of Esterified Product22 The hemin ester was allowed to dissolve in a small volume of freshly distilled and dried chloroform (1 m1 CHC13 for 20-30 mg of ester). To the boiling solution an equal volume of boiling, absolute methanol was added. As boiling continues, chloroform distilled off and the methanol concentration increased. This process is allowed to continue until crystals form on cooling. Crystals were then collected, washed with H20, and dried overnight in a desiccator. This procedure for crystallization of the ester was used to prepare crystals from both processes one and two for preparation of the ester. Fe(II)hemin Diester Stock Solutions24 To a 60 m1 seperatory funnel 10 ml water, 20 ml pyridine, and 15 ml benzene were added. Then approximately 15 mg purified chlorohemin diester was added and the mixture was shaken well to mix the contents thoroughly. The space 11 in the funnel above the solution was flushed briefly with nitrogen and about 0.15 g Na28204 was added. Then the seperatory funnel was quickly capped, shaken well for about one minute, and a bright orange color appeared. The benzene layer was allowed to separate and the lower aqueous layer was discarded. The benzene phase was delivered into a capped 50 ml flask containing 10 m1 pyridine and 10 g anhydrous NaZSO4, which had been previously cooled in an ice bath. The mixture was stirred for about 5 minutes and filtered through a fine-fritted funnel into a receiver cooled in an ice bath. This solution was stored in the refrigerator at about 5°. Even under these low temperature storage conditions, oxidation of the Fe(II)heme diester became significant after three to four days, after which time the stock solution was discarded and a new stock solution prepared. Procedure for Measuring the Rate of Oxygen Uptake by Fe(II)heme Diester in Benzene/Pyridine SolutiOns The novel reaction flask for this procedure is shown in Figure 2. With this apparatus the reaction solution can be prepared under vacuum, the required nitrogen and oxygen pressures applied, and transfer to a spectrophotometer cell can all be accomplished. The procedures for preparations of the solution to be used in the kinetics experiments and the rate measurements are described below. Into Flask A was placed 20.0 ml of a solution containing pyridine in benzene (concentrations are varied by 12 Figure 2 Reaction Flask for Kinetics Measurements 75 mm- JL. 14/20 Joints 100 mm 2 mm stopcock to .————~... N2 ' 1 02 10/30 inner joint 3.. 13 mm standard stopcocks W. :1 joints tygon tubing 1 cm quartz Spectral cell *” l4 diluting a stock solution of pyridine in benzene to obtain 2 to 10—3 pyridine concentrations of 10- M). The solution was frozen, the vessel was evacuated, the solid was thawed and re-frozen, the vessel evacuated, then filled with nitrogen, and the solution was thawed again. Next, the solution was shaken vigorously, again it was frozen and the flask was evacuated. This procedure deoxygenated the solution very effectively. Atmospheric pressure was noted on the vacuum gauge attached to the system, a partial pressure of oxygen was applied (oxygen pressures of 150, 300, and 500 torr were used to evaluate the dependence of rates on p02) and a partial pressure of nitrogen was added to attain a total pressure in the flask of one atmosphere. The solution was next thawed and shaken for at least 5 minutes. To flask B was next added 2.0 m1 of stock Fe(II)heme diester solution. The Fe(II)heme diester solution was evaporated to dryness by application of a vacuum. Flasks A and B were then connected and the space between them was evacuated. Flask B was next filled with partial pressures of nitrogen and oxygen exactly the same as in flask A above. The entire apparatus (flasks A and B connected to a spectrophotometer cell) was immersed in a temperature bath and allowed to reach thermal equilibrium (about 30 minutes). After this time, the solution and solid were allowed to mix by opening the stopcocks between flasks and swirling the solution for a few seconds. The solution was immediately 15 transferred to the spectrophotometer cell, placed into the spectrophotometer, and a series of visible spectra taken at about 5 minute intervals (intervals varied depending on the rate of the reaction). Spectra were measured by means of a Unicam SP800 Ultraviolet Spectrophotometer fitted with a time delay and a temperature controlled cell holder. The time delay allows one to measure a spectrum at any specified time interval and at each specified time interval the spectrum will be recorded indefinitely. The cell holders are jacketed so that water can be circulated through them from a temperature regulated water bath and gives temperature stability during the course of the measurements. Procedure for Measuring the Oxygen Uptake b Fe(II)heme DIester 12.3 Pyridine Solution Saturated w1th ImidazoIE The procedure used for these measurements was identical with the procedures described above except that the solvent system used was a pyridine solution saturated in imidazole. It is crucial that saturation of pyridine with imidazole be obtained because the concentration greatly affects the rate constants. Particular care was taken in preparing these solutions, so that not only was saturation obtained, but also all the solutions were of uniform composition. Imidazole (recrystallized, from Research Organic/ Inorganic Chemical Corp.) was recrystallized three times 16 from pyridine to insure a pure product. The product from the final crystallization was dried overnight under vacuum. To prepare a saturated solution of imidazole in pyridine enough imidazole was placed into 200 m1 of pyridine at about 30° so that the imidazole no longer dissolved in the solution. The solution was cooled to about 20° and allowed to remain at this temperature for several hours. The temperature of this solution was slowly increased and the solution was stirred until 25° was reached. At this point, the stirring was stopped, the solid was allowed to settle while 25° was maintained, and, finally, 10 m1 portions were removed and placed in separate vials to be used when needed. RESULTS Esterification Procedure Two methods of preparation of the heme diester from hemin are described in this thesis. The first, refluxing of hemin in trifluoroacetic acid/methanol solution was the 17 in their initial studies of method used by Kao and wang the oxidation of this complex by molecular oxygen. The second, placing hemin in a sulfuric acid/methanol solution and allowing the reaction to proceed for about 48 hours, is the method suggested by Falkzz. In order to decide which product and method would best suit the needs of the following experiments, a comparison of the properties of these compounds was made. First, the infrared spectrum of each of these products was obtained and compared with that of the initial hemin spectrum (see Figure 3). Since the infrared spectra of both hemin and hemin 22'26 simple comparisons diester have been reported elsewhere could be made with these spectra. The most definitive absorbance, in terms of the ester side group is a shift of 1 to about 1740 cmTl. the hemin ester absorption at 1685 cm- This shift is found in spectra of both esterified products. However, in the spectrum of product one an unassigned peak at about 1770 cm.1 is found. This band is found in no other 17 18 Figure 3 Infrared Spectra of Initial Commercial Hemin the Product of the Trifluoroacetic Acid/Methanol Method of Preparation of the Esterified Product the Product of the Sulfuric Acid/Methanol Method of Preparation of the Esterified Product 19 A. product of process C. (coo 20 reported spectrum of the ester. It has become obvious that the two ester products are not identical and that further 1) of observations in the "fingerprint" region (750-1250 cm- these spectra substantiate the supposition that these are not identical products. Comparison with the previously reported spectra of the heme diester showed product two was essentially the same and suggests it is the one to be used. To test the two products further, the reaction of the reduced form of each with oxygen when placed in pyridine was observed. The results of previous studiesl7'19'21 suggest that the Fe(II)heme diester is impassive to reaction with oxygen when pyridine is the solvent used. However, in Figure 4 one can see the results of dissolving the reduced forms of products one and two in a pyridine solution in an oxygen atmosphere. Figure 4a shows that over a period of about 35 minutes the visible spectrum of product one showed a drastic change at its 555 nm absorption, indicative of reaction with oxygen. Figure 4b shows that product two showed no such change in its visible spectrum, suggesting product two was impassive to oxygen while in pyridine. These results implied that product two provided the heme diester which should and was used in all subsequent experiments. Kinetics of the Reaction of Fe(II)heme Diester with Oxygen i2 BenzenE7nyidine Solutions The experimental procedure described above for kinetics measurements in the benzene/pyridine solution results in a 21 Figure 4 Oxidation of 02 of A. the Product of Method One in Pyridine B. the Product of Method Two in Pyridine 0’4 ; 0.2 :2: . 0'0 ()2 22 the product of process one in pyridine F about 35 min. L; . f :i for spectral change : ;_J ,TT . . i ' .. ....... .1... _ . . _ . --.; if _ ”1'- ---._- ..-.g' t - .'r 45°" 50° 550 600 650 23 series of spectra (a typical example is found in Figure 5). Over a period of time the absorbance at 555 nm decreases and is a measure of the rate of reaction of the Fe(II)heme diester with oxygen. In treating the data (time vs. absorbance) one can safely assume the order does not depend on the changing pyridine or oxygen concentrations. By making pyridine > 10.3 M and oxygen > 10.3 M with the Fe(II)heme diester 2 10-5 M there is about a one hundredfold excess of pyridine and oxygen over the heme diester concentration and, hence, the pyridine and oxygen concentrations can be considered constant. With this assumption, the data are fitted to pseudo first-order, pseudo second—order, and a mixed pseudo first- and second-order rate equations of the forms -dA/dt = k1(A - Am), -dA/dt = k2(A — Am)2, and -dA/dt = ki(A - Am) + (A - Am)2 respectively, where k's are rate constants, A is the observed absorbance, and A00 is the absorbance when the complex is completely oxygenated. The correction factor, Am, is necessary because both the oxygenated and the non-oxygenated complexes absorb at 555 nm28 (see Appendix B for a more detailed calculation of this correction factor). By using the above first-, second-, and mixed-order forms of the rate equation, rate constants for each were calculated by a general purpose curve fitting program of Dye and Nicely29. The computer program adjusts the rate constants and initial absorbance until the calculated constants give results most nearly 24 Figure 5 Kinetic Spectra of Fe(II)heme Diester in Benzene/Pyridine Solution mZOMUHZHAAHZ mBuzmAm> chguc >1 r (haw: utau wt. :_ 1:4 _:_;1 .tpa.:;.4C :7: .c—Trs.3n1uu ,c .owpchn v. o_mwo It!) :uv: v_ u >ch. tr_:a :rpantt.at : u.qi: t ...acc .a.:r:.:ucxu (c mcnmcoc. u pzuxmauz_._cou3c_. u sch—c: ur» hc u24<>._c.uxr_. u 3.. a.» at . be; .1_le it ...x.. r. econ—cc. u bzuauzu7_.~cou.cs. a pic—3 at» we w:.«>._cowcs.. u bin; ... .u ;:_:a ....oc.x.. it ._._.u vs V313! u vzcwa r .ccupiuz mucxuq 29 Figure 7 Fit of Rate Data to a Second-Order Rate Equation 30 0&5 —u—-—-mu——n——-mo-nu—u-Vu—u-uunra——u——-mu—u-uu-mu—un—u-U‘uu—nnuqrun—u-u-rI-hn——-¢-— ~cuuneo. cc.u_oc. U I P u! U) U T IoHfiRDOI IIHGOIQZOQIH d Oh ducn and. no Bah w H U m w m u‘ u‘ w u “R U ‘ u pan?» no Dru ioooooooooooo :9: C “*P‘U’F‘H—I—Iu‘h—t—u—c—wu—u—u—o—u‘u—u—u—p—ru—u—u—u—m—c—u—n—‘r—u—J—n—w—n—I-c—m—u-I—I-mh 1‘ ch - r n. I ---f--"w----w----v----r----v----m nl. .U P'l'-f----f'-'lU'Iv'IV---I-h---'— u “ Id u > «how: >¢ a «pow: uzcv wt» z. 214 ~7.ca :u»<.:;.¢c era .cmppcon ¢_ c_ruc rut: cum: v. u >42:. »7_ca auhqo:t_¢: a chmz : u pzuxmauz_._coua—_. a pzuauaoz_.~cou.c~. .<»ru1_1uox. zc u2¢u1 u ._;_ru .ahrua.:uo.a (a v2._oowrm.. u 1:. ur— —< i:.¢> .c_Vua we I etc—w C1» hc u:.4>._cowcsu. u put. rt» pa urdc> o.».1cty.w ._o\.‘ w— .(Uub1u> s7 .—o_.n V— VVCICC 31 Figure 8 Fit of Rate Data to a Mixed-Order Rate Equation —— ”...—n‘ _‘ I32 _m----u--u-m m V! —I--I—I—u‘—-—-—qr-—0--WI-—I—I-U‘—u—-—¢r—-nun-nmc-un—u—U‘ho—hnblfI-hnhu-U'~-—I~ m Nceuocs. coon—9c. ZnHBtadu ¢H§MOIQflHH8 < 09 «Eta cat: no Duh u‘ I I u “R uh U') I hzwwaUI—o~oouO——o I blwawmur_o~eok—eho 5.1.. . ‘.. ...... tfifivaocxofit LIB u“ nu---lT-'--’-"-r---'r----J'I-"I-"'If'--'.I.I'I-r ----.aU'..u ---- "U'. I. \ LI | \l RI \ I I u u u H u u u u u Ids-n—o—IIb—b. ' ~~—m1—~—bu ~—~.I—4U"—-h'b'--J ——u—-—r—-..—- I —--—_J A > (name >8 - «bawo wtcv wt» 2. mac p7-Cd :upq.:cch c:< .4»:t:_1icuu (c v24“: u .owppoaa m— c—vwc rut: cum: v— u >42c. »z_ca sun‘s:oaco 4 wreuz : .»._Ci at...._;uo-t a: vzcux u I teppoc wt» pc w:4<>._c.uom_. a cop nth .4 12.4) .a_uui .1 I rte—c UZb >1 U34¢>ogeow:s_o u puma wt» be u:.«> ....3<22.. .— 5a.. .«.. r. ..u_»au> .....x v. wvczc. 33 Table 1 Results for Kinetics of Oxygen Uptake By Fe(II)heme Diester in Benzene/Pyridine Solutionsa [py] mM [02]b mM k1(s-1)X103 k2(A-ls-1)x103 2.424 1.79 0.17 (110%) 0.083 (112%) 2.424 3.57 0.88 (i 9%) 0.26 (110%) 2.424 5.95 2.40 (i15%) 2.60 (110%) 4.857 1.79 0.22 (1 5%) 0.042 (110%) 4.857 3.57 0.97 (i12%) 0.128 (f 8%) 4.857 5.95 1.28 (i 5%) 1.96 (i 6%) 9.714 3.57 0.30 (i 9%) 0.096 (f 8%) 9.714 5.95 0.61 (i 7%) 1.75 (i 7%) a. Temperature = 24.8°C in all cases, total pressure = 760 torr. b. [02] = (11.9x10'6 from the O2 solubility in benzene. M/torr) p02 with this constant derived 30 34 Figure 9 Spectra of Fe(II)heme Diester in a Pyridine Solution Saturated with Imidazole as a Function of Time -—— mic/mu EH AA HZ mhczmw. mat... one coo 0mm oom one P _ — A 4 _ a . 36 Under these suitably restrictive conditions one has a system with the ability to react reversibly with oxygen and verifies the work of Corwin and Bruck21. However, it was necessary to try to repeat this work because it has generally been ignored. Figure 10a is the spectrum of the initial non-oxygenated form of the complex Fe(II)heme diester in pyridine saturated with imidazole, Figure 10b is the completely oxygenated Fe(II)heme diester, and Figure 10c is the spectrum of the oxygenated product which has been under vacuum about 4 hours at 40° and was then redissolved in pyridine/imidazole with the exclusion of oxygen. Each of the spectra correlate well with those of Corwin and Bruck (note especially the shift to higher wavelength and the change in relative intensities of the absorbances in the oxygenated form, and note that in recycling to the original product a spectrum not as intense, but with the same wavelength maxima is obtained). These agreements show there is no need for further substantion of the reversibility of this complex with oxygen. 37 Table 2 Results for Kinetics of Oxygen Uptake by Fe(II)heme Diester in Pyridine Solutions Saturated with Imidazolea p02 (torr) k1(s’1)x103 k2(A'1s'1)xio3 150 0.49 (t 8%) 1.16 (i8%) 300 2.61 (110%) 2.42 (16%) 500 6.92 (i 8%) 4.56 (19%) 24.8°C in all cases, pressure = 760 torr. a. Temperature 38 Figure 10 Spectra of Initial, Oxygenated, and Recycled Fe(II)heme Diester (a) ----- spectrum of initial Fe(II) complex (b) -'-'- spectrum of completely oxygenated complex spectrum of recycled complex (c) 39 6mm unohoflstAHs nowuoaobaz . 8n 4 l N.o O FIT .s eouaqzosqv DISCUSSION Kinetics of the Reaction of Fe(II)heme Diester with Oxygen w—————- u T— 1'— £2 BenzenE7Pyr1d1ne Solutions It was found that the trifluoroacetic acid/methanol method (product one in the experimental section) of esterifying the hemin provided a poor product for use in the kinetics experiments. This method was used by Kao and Wang17 to provide them with the heme diester to be used in their kinetics experiments. These results indicate that the work of Kao and Wang and the rate constants are suspect. Cohen and Caugheylg also repeated some of the work of Kao and Wang and used yet another method to prepare the esterified product25. Their results also gave rate constants differing from those of Kao and Wang but, again, there was a discrepancy in their work. The mechanism they described has as its final step; py-N4Fe-02-FeN4-py + 2H20«——9 2N4Fe(III)OH + H202+py They report water was always in amounts sufficient to be available for reaction. However, they claim that water was rigorously excluded from solvents and the reaction flask throughout their experimental work, and in no place during their experimental procedure was water added to the reaction solution. How, then, can water enter into this mechanism? 40 41 If the amount of water is not a controlled variable, then certainly a variable amount of water, particularly if its concentration is near that of the heme diester concentration, could alter the rate of reaction. One finds up to 50% error in their data, far greater error than what should be considered acceptable. It was, therefore, necessary to repeat the work of Kao and Wang and Cohen and Caughey for the system Fe(II)heme diester in benzene/pyridine solution. It should be noted that very rigorous procedures for excluding water from the system have been used and that the results differ from either of the previous reports. Upon comparing the results of this study with those of Cohen and Caughey three notable differences are observed; (1) the mixed-order rate equation provides the best fit to the present data throughout the range of pyridine and oxygen concentration, whereas Cohen and Caughey report a first- order rate at oxygen concentrations > 3.6 mM, (2) the error within the present results are 15% or less - Cohen and Caughey have error as high as 50%,and (3) the numerical values of the rate constants calculated from this study tend to be about a factor of ten lower than those of Cohen and Caughey. Eliminating water from the system can very possibly account for these differences, and better reproducibility has been obtained. Also, eliminating water prevents the final step in the mechanism of Cohen and Caughey's and suggests a new mechanism must be found. 42 Since a mixed-order rate equation seems to be followed over the range of concentrations used in this experiment, a mechanism more complex than that proposed by either Kao and wang or Cohen and Caughey had to be found. This becomes more complicated because no simple relationship between the concentrations of pyridine and oxygen and the rate constants could be found, but as [pyridine] increases the rate Fl constants (k1 and k2) decrease and as p02 increases the rate constants increase. A possible mechanism consistent with the observed results is suggested as follows: &j K , Fe(py)2 ————E——A Fepy + py fast (1) K0 Fe(py)2 + 02 -——————A>Fepy02 + py fast (2) k1 _ Fe(py)02 ——-;———A pyFe(III)o2 slow (3) -1 k2 2- FepyO2 + Fepy ———————a>pyFe(III)02Fe(III)py slow (4) k pyFe(III)O;+Fepy ———§—4)pyFe(III)O§Fe(III)py fast (5) The mechanism corresponds to a mixed-order rate equation of the form; -d[Fe(II)] /dt = kiIFe(II)] + ké[Fe(II)]2 where ( k KMIO ] ) k3K [Fe(II)]-k_1([PY]+K +K [02 1) k' (1+ 72“""TT’ 1 =T§y1 + KP + KO [027 k 3Kp [FéTif)]+k _1([py]+Kp +K [o 43 kZKpKo [02] k'= .2 ([py] + Kp + KO [021) (see Appendix A for details of the derivation). Thus, ki and ké are pseudo first- and second-order rate constants depending on the experimental [py] and p02 and k3KplFe(II)], k_1[py] and k_lKo[02]. This mechanism provides for a first-order dissociative process (equation 3) followed by second-order dimer formations (equations 4 and 5). Both of these features have also been observed in autoxidations in aqueous systems. George31 found that the reactions of oxygen with ferrous perchlorate solutions proceeded through an intermediate Fe(II)O2 complex and eventually to the Fe(III)O; product that is analogous to equations 2 and 3 in the above 32 found that the oxidation of mechanism. Haim and Wilmarth pentacyanocobaltate ion by molecular oxygen yielded the bridged peroxo complex (NC)5-Co(III)-O§--Co(III)-(CN)5 and Huffman and Davidson 33 proposed in the reaction of ferrous ammonium sulfate in 1 M sulfuric acid with molecular oxygen that a peroxo diiron(III) bridged complex, (H20)5-Fe(III)- O§--Fe(III)-(H20)5 probably formed. Both are very similar to the bridged complexes in equations 4 and 5. The present mechanism seems to be the best to conform to the data. IfEl'l‘lui-‘lllllll 44 Kinetics of the Reaction 9: Fe(II)heme Diester with Oxygen i3 Pyridine Solutions Saturated with Imidazole Certainly the interest in this system can be traced to the ability of the Fe(II)heme diester to take up molecular oxygen reversibly and only to do so when in a pyridine solution saturated with imidazole. It can be compared with the benzene/pyridine system previously discussed that does not have the reversibility, but rather leads to oxidation of the iron from (II) to (III). It was hoped kinetics experiments with this system would lead to a greater understanding of the mechanism by which reversible oxygenation takes place. But the experiments undertaken here were not without their, as of yet, unresolved problem and sources of concern. The molecular oxygen concentration in the pyridine/imidazole has not yet been determined (unlike the benzene/pyridine solution in which oxygen solubility was already known). This problem eliminates the possibility of determining the exact dependence of the rate constant on the oxygen concentration but does not prevent one from varying the oxygen concentration, since the partial pressure of oxygen in the reaction flask could be controlled. Also the concentration of imidazole in saturated pyridine has not been accurately determined. It is vitally important that saturation with imidazole is attained, for as the concentration of imidazole falls below saturation the rate becomes slower. However, care was taken to insure a 45 saturated solution even though its concentration was not known. One may question the method used to prepare the diimidazole heme diester complex and whether, in fact, this is the actual complex used. Although the elaborate procedure of Corwin and Bruck21 is not employed to isolate this diimidazole complex, the initial visible spectrum is identical with that they reported. Thus it is a good indication that the correct complex was used. Since pyridine is the only other ligand in the system available to complex to the iron heme, it is comforting to note that 34'35 show that the affinity of imidazole for other reports the iron in a heme group is much greater than the affinity of pyridine for the same iron heme and implies that imidazole is the coordinated complex on the heme diester. The analysis of the results obtained in the reaction of oxygen with Fe(II)heme diester in a pyridine/imidazole solution becomes very difficult due to their small number obtained in this study. However, some general observations can be made: the mixed first- and second-order character of the reaction rate is again found and suggests a complicated mechanism may well be found in this system,too. The rate of reaction is about the same in both of these systems, i.e., the values of the rate constants, k1 and k2, are in the same range as the rate constants of the benzene/pyridine system. Though it seems that there are some comparisons to be made between this system and the 46 benzene/pyridine system, it must be admitted that much less is known about the way the reversible oxygen system behaves or why the change from benzene/pyridine to pyridine/imidazole solvent causes the reversibility. Unfortunately, the lack of complete data at this time prevents the presentation of a detailed reaction mechanism. It is hoped, however, that the research will be continued until a complete understanding of this system is obtained. BIBLIOGRAPHY 11. 12. l3. 14. 15. 16. BIBLIOGRAPHY E. Antonini and M. Brunori, Ann. Rev. Biochem., 39, 977 (1970). F. Haurowitz, "The Chemistry and Function of Proteins", 2nd ed., New York, Academic Press, 1963. W. S. Caughey, Ann. Rev. Biochem., 36, 653 (1957). R. Lemberg and J. W. Legge, "Hematin Compounds and the Bile Pigments", Interscience Publishers, Inc., New York, 1949. R. E. Dickerson, Ann. Rev. Biochem., 4;, 815 (1972). M. F. Perutz, g. Mol. Biol., 13, 646 (1965). M. F. Perutz, M. Muithead, M. Cox, L. C. G. Goaman, F. S. Mathews, L. L. McGandy and L. E. Webb, Nature, 219 (1968). M. F. Perutz, M. Muirhead, M. Cox, L. C. G. Goaman, Nature, 219, 131 (1968). M. F. Perutz, Nature, 228, 726 (1970). O. H. Gibson and L. J. Parkhurst, g. Biol. Chem., 243, 5521 (1968). O. H. Gibson, L. J. Parkhurst and G. Geraci, J. Biol. Chem., 244, 4668 (1969). O. H. Gibson, g. Biol. Chem., 245, 3285 (1970). M. Schalfejeff, Chem. Ber., 18, 232 (1885). T. C. Chu and E. J. Chu, g. Biol. Chem., 212, 1 (1955). R. L. N. Harris, A. W. Johnson, and I. T. Kay, Quart. * Rev.,’ 39, 211 (1966). R. J. P. Williams, Chem. Rev., 26, 299 (1956). 47 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 48 O. H. Kao and J. H. wang, Biochem., J, 342 (1965). I. A. Cohen and W. S. Caughey, in "Hemes and Hemoproteins“, Chance, Estabrook, and Yonetani, eds., Academic Press, New York, 1966, pp. 577. I. A. Cohen and W. S. Caughey, Biochem., J, 636 (1968). A. H. Corwin and Z. Reyes, J. Amer. Chem. Soc., 18, 2437 (1956). A. H. Corwin and S. D. Bruck, J. Amer. Chem. Soc., 89, 4736 (1958). J. E. Falk, "Porphyrins and Metalloporphyrins", Elsevier, Pub. Co., Amsterdam, 1964. James D. Hoeschelle, Ph. D. thesis, 1969, Michigan State University, pp. 38-40. J. H. Wang, A. Nakahara and E. B. Fleischer, J. Amer. Chem. Soc., 82, 1109 (1958). J. E. Falk, E. I. B. Dresel, A. Benson and B. C. Knight, Biochem. J., JJ, 87 (1956). J. O. Alben, W. H. Fuchsman, C. A. Beaudreau and W. S. Caughey, Biochem., J, 624 (1968). R. E. Dodd and P. C. Robinson, "Experimental Inorganic Chemistry", Elsevier, Amsterdam, 1957, pp. 167-168. R. J. Porra and O. T. G. Jones, Biochem. J., 81, 181 (1963). Landolt—Bornstein, II. Bund, 2. Teil, Bundteil b (1962): PP. 1-74. J. L. Dye and V. A. Nicely, J. Chem. Educ., 28, 443 (1971). P. George, J. Chem. Soc., (1954), 4349. A. Haim and W. K. Wilmarth, J. Amer. Chem. Soc., JJ, 509 (1961). R. E. Huffman and N. Danidson, J. Amer. Chem. Soc., 78, 4836 (1956). —_ R. W. Cowgill and W. M. Clark, J. Biol. Chem., 198, 33 (1952). ‘— J. N. Philips, Rev. Pure App. Chem., lg, 35 (1960). APPENDICES APPENDIX A Assuming that the mechanism of the reaction of Fe(II)heme diester with oxygen in benzene/pyridine solvent is that given below, this appendix will show the derivation of the rate equations for the mixed-order equation. Mechanism K Fe(py)2 ————E——A Fepy + py K Fe (py)2 + 02 ___°__. Fepy02 + py k 1 - FepyOZ'—-----]-<-———-‘pyFe(III)O2 -1 k FepyO2 + Fepy -———3——e)pyFe(III)O§_Fe(III)py k pyFe(III)O; + Fepy ————2——9 pyFe(III)O§ Fe(III)pY Derivations of Rate Constants K = [Fer][E¥] P Fe PY 2 [Fe(py)02][py] Ko = fia(py)21[ozT [Fe(II)] = [Fe(py)2] + [Fepy] + [Fepyozl applying the_steady state approximation 49 50 d[Fe(lII)py0;] - dt = 0 = kIIFepyOZ] — k_l[Fe(III)py02] - k3[Fe(III)py02][Fepy] [Fe(III)py0;] = k1[Fepy02]/(k_1 + k3lFepyl) -d[Fe(II)]/dt k1[Fepy02] - k_1[pyFe(IlI)o;] + k2[Fepy02][Fepy] + k3[pyFe(III)O;][Fepy] = (k1 + kZTFepy])[Fepy02] + [pyFe(III)O;] (kBIFepyl - k_1) klIFePYOZ] = [FePY02](k1+k2[FePY])+ E:I:E;TFEPYT (k3[F€PY] - k-l) from equilibrium conditions: [Fepy] = KPIFe(py)2]/[pyl [Fepyozl = KOTFe(py)2][02]/lpy] [Fe(II)] = [Fe(py)2](1 + Kp/[py] + KOTOZJ/[py]) [Fe(py)21 = [Fe(II)1/(1 + Kp/[py] + KOTOZJ/[Py}) [Fe(pY)2] = ([Fe(II)][py])/([py] + KP + Kolozl) [Fepy] = KplFe(II)]/([py] + KP + KOIOZI) let Kd [py] + KP + Ko[02] 2 [o ] k K K [o ][Fe(II)] ”1 g 2 [Fe(II)] + 2 P 0 2 2 d d -d[Fe(II)]/dt = K +k 10K KIO2 ] k 3K [Fe(II)]/Kd - k _[Fe‘11’1‘E—KETretii)1/K3+r: 51 _d Fe(II) = leOIOZ][Fe(II)] (1 + k3KPIFe(II)] - k_le) dt Kd k3Kp[Fe(II)] + k_1Kd k K K (o 1 + 299 g 2 [Fe(II)]2 Kd Assuming kBKPIFe(II)] is much greater than k-le we now have a rate equation both first- and second-order in [Fe(II)]. APPENDIX B The rate constants found in tables 1 and 2 are calculated by fitting the data to a mixed-order rate equation of the form -dA/dt = k1(A - Am) + k2(A - Am)2 where the A00 correction is necessary because both the non-oxygenated and oxygenated Fe(II)heme diester absorb at the wavelength at which the absorbances are measured. It is the purpose of this appendix to show how this Am term arises and that the absorbance is actually a measure of the concentration. The relationship of the absorbance to the concentration can be found from the Beer's Law equation: A =£3c1 where; A = absorbance, e = molar absorptivity, c = concentration of the absorbing species, 1 = path length in cm of the cell. Having used a 1 cm path length spectrophotometer cell we find; A = EC = c + 8II II 8IIICIII where gII and C11 refer to the non-oxygenated Fe(II)heme diester and EIII and cIII refer to the oxygenated Fe(II)heme diester. Since the total concentration (cT) is equal to c + + ). But when II 6111’ A = EIICII complete oxygenation has occurred only the oxygenated EIII(°T ' C11 52 53 Fe(II)heme diester is present, so that EIIICT = A00 and )cII + EIIICT = ASCII + Am. iRearranglng Unfortunately,Ae has not been calculated A = (€11 ' 8III (A - Am)/A€ = CII' for the complexes but it is a constant, so A - A00 is directly proportional to C11. The rate equation is more _ 2 . properly stated as, -chI/dt — leII + kZCII , thus if we replace the cII terms with A - A00 and incorporate the As term intolthe rate constant one obtains a perfectly valid rate equation of the form; -dA/dt = k1(A - A-) + k2(A - A...)2