'-'.‘o‘u\'o'- Hut. .120 o -'.;\ .m. . . . “pv'nul‘f .-|o«-'_~':,- u I; 33-. “.4 - " .. l.-,‘, . a dig. r-T'J ‘ ‘. I f, 12‘“ '91'.,..r'r 1‘9‘ .‘r'. 12,313} at" 9f '13:“? ’l" 1‘ f D} .2 w - . n f’ . . ,' ~ 735.22»? xvi-‘99» ::~t.‘.’..ff $35.25» :‘p' '*' N.- ' ‘ ’ ~f t ‘n‘ e .rrr'r.. ’ . ?‘- 41‘- '2 . . ' v“ ’ ’1'." w; , ,4... ”v: {‘0 2‘ Jr" I.“ '3 .,. J AI. :; ..‘ wuura- 1'4“»; ' at ..."",.,’_',v:’,’"" 19¢” . .f' . 'frfiffifr“ . X- {3"}; :T‘z J v .’-., ". , _ U» ‘3' 5'17- E? :9i»« " ' IL: 4,5“ . :1 t ' 7:". 1“". f" , r or 3}" '4 $7.? - ‘ pm- Dr '9“- _‘ - ‘ "1 ‘ ' - , #0:.“ an!“ :3 ', 2 H; . 2 , c. . _—:;,,.4".wr. Jufv - .. ,.. ,., I. u" ... . .. ... .- A. um . . " ' a2 ', M 212:? 1“ .aa¢.n-U,.- ’2! ~.;o{'nu ~~m ‘ 2 3‘1}; gig: "Egg; ' -,‘ ”If ’v ‘f’r‘ A. ,1: ~'."' («1" . ”“72 4’25. if}??? Apmf ‘é‘fif‘x . am“; Gil“. who! Sfiflv m— , . .' ~.- i-J”fi.;{ 1;... $157351 ,~....5..'.‘. ‘4. 4—,!!231“, Wm: :"V: mtg/”.M.I . :r'fiw . ‘ j,“ 530’” 2 :* «dirt: 2: .. yfi! f firs.— “5??!“ 31?; “ flog— I“ rota- Awh' .';”‘J."‘ I. it?” In 35:34?" ‘9’ neg? ~ 4&2: 313.7 ~ I4 553’? If V”? ' T"',é.:¢ - ’/ 2i “ . ‘1‘ ‘ 5' M 42;...» €+' w 1"i-Vffi.‘ )rfi'k' ‘ , fin“..- . Maw-lung 3 1293 00561 4858 LEBRARY Michigan State University This is to certify that the dissertation entitled Study on Heme g1 Of Cytochrome Qéi Nitrite Reductase presented by Weishih Wu has been accepted towards fulfillment of the requirements for ph o D 0 degree in Chemi Stry Major professor ca .2. 4471 MM MS U i: an Affirmative Action/ Equal Opportunity Institution 0-12771 )V1£Sl_} RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from .—:—. your record. FINES will be charged if book is returned after the date stamped below. 3“" .‘m- “:9 vi: “1995' ‘ STUDY ON HEME $11 or CYTOCHROME gal NITRITE REDUCTASE By Weishih Wu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1988 ABSTRACT STUDY ON HEME g1 or CYTOCHROME £511 NITRITE REDUCTASE By Weishih Wu Cytochrome Ed] nitrite reductase found in a large number of microbial denitrifiers, which play an important role in the biological nitrogen cycle, contains an extractable green heme prosthetic group. This so-called d1 heme has been assumed to be a chlorin since the late 1950's. Recently, a structure determination of isolated heme _d_] by another laboratory also favored a chlorin structure. However, critical analysis of the available spectral data has led us to propose that heme _d_l is not a chlorin but a novel porphyrindione compound. The spectra of the model compounds derived from octaethylporphyrin and mesoporphyrin IX dimethyl ester provided solid evidence of the proposed structure. The methodologies developed in the synthesis of the LIB-porphyrindione core structure and the acrylate side chain ultimately led to the total synthesis of heme 911. By comparison with its stereo- and regioisomers, heme d has been shown to possess a 6-acrylic side chain and a cis arrangement of its two angular acetic side chains. Physicochemical studies have demonstrated that the most distinctive features of porphyrindione, as compared with porphyrin and isobacteriochlorin, are its electronegativity due to the two conjugated keto groups on the macrocycle and its enlarged core size which is brought about by the saturation of the pyrrole rings. Reconstituted cytochrome g1 nitrite reductase from Pseudomonas stutzeri with both native and synthetic heme d1 exhibit identical spectral properties and NO/NZO producing activities. This proves that the porphyrindione structure is the true form of d1 and that the synthetic compound is completely bioactive. To explain the biosynthetic origin of this unprecedented d1 heme, two possible pathways, via protoporphyrin or sirohydrochlorin, have been proposed. To all friends of my generation, those who lost their lives during the Cultural Revolution (1966--1976, China); those who lost their opportunities to recieve education during Mao's era; those who have never lost their faith in building up a democratic, prosperous new China and those who are working hard for a world filled with freedom, equality and fraternity. iv ACKNOWLEDGEMENTS I would like to express my special thanks to Professor C. K. Chang, for his advice, friendship and guidance throughout the course of this work. I would like to thank Professor E. LeGoff, for serving as the second reader and for always being ready to help. I would also like to thank Professor C. H. Brubaker and Professor A. Tulinsky for serving as members on my guidance committee. Gratitude must be expressed to the National Institutes of Health for financial support in the form of research assistantship. In addition, I thank the Department of Chemistry at Michigan State University for providing support in the form of teaching assistantship and for rewarding of a one year SOHIO fellowship and a summer BASF fellowship. Great thanks must be extended to the past and present members of Professor Chang's group - Dr. M. Kondylis, Dr. I. Abdalmuhdi, Dr. M. Koo, Dr. A. Salehi, Ms. C. Aviles and Mr. W. Lee for their encouragement and friendship. Particularly, I wish to thank Dr. C. Sotiriou, not only for her inseparable contribution to the "green heme" project, but also for being a good friend. It has been my good fortune to be associated with a number of other research groups at MSU, especially professor I. Tiedje's and professor I. Babcock's groups. It was fun to do the enzyme purification and protein reconstitution experiment with Dr. E. Weeg-Aerssens in a 4 °C cold room and especially interesting to take Raman spectra with Dr. W. A. Otertling in the dark but colorful laser lab. I am also grateful to the NMR group of the Chemistry department for their prompt help, particularly to Dr. L. D. Le for his demonstration of the NOE technique. Thanks also go to my friends and colleagues at Chengdu Institute of Organic Chemistry Chinese Academy of Sciences. Especially, I wish to thank Professor G. N. Li, who was the director of the institute and my master thesis advisor, for his introducing me into the colorful porphyrin chemistry and his caring throughout my graduate career. I also extend my thanks to Mr. S. L. Pan for his helping me go through the Chinese red tapes otherwise my oversea study would not be so easy. The deepest appreciation is due to my parents for their love, encouragement and faith in their youngest son, and above all, for always taking the education of the children as their first priority even during the most diffith time of their lives. Thanks also go to my brother, sister and my wife's family for their consistent support over the years. Finally, I would like to thank my wife Xiaoming. I appreciate sincerely the support and love she has given me and looking forward a new life together with our two lovely children. vi TABLE OF CONTENTS Page LIST OF TABLES ........................................................... x LIST OF FIGURES ......................................................... xii CHAPTER 1 INTRODUCTION: THE GREEN HEIWE PROSTHETIC GROUP OF CYTOCHROME m1 NITRITE REDUCTASE .......................... 1 I. SIGNIFICANCE AND BACKGROUND .............................. I A. NITROGEN CYCLE AND DENITRIFICATION ...................... 3 B. CYTOCHROME _C_d1 NITRITE REDUCT ASE ........................ 5 C. ON THE MECHANISM OF NITRITE REDUCTION .................. 9 D. HEMP. Q1 ................................................ 10 II. OBJECTIVES OF THE PRESENT WORK ............................ I 1 RESULTS AND PRESENTATION ................................. 13 IV. NOMENCLATURE OF HEME Q1 AND RELATED COMPOUNDS ........ 14 CHAPTER 2 FURTHER CHARACTERIZATION OF THE HENIE d1 STRUCTURE ....... 16 I. EVIDENCE OF THE PORPHYRINDIONE STRUCTURE OF HEME _d_] ...... 16 II. STUDY WITH MODEL COMPOUNDS .............................. 21 A. FORMATION OF THE 1,3-PORPHYRINDIONE CORE STRUCTURE . . . . 22 1. On the 0504 Oxidation of Porphyrin Tetraacetate ................. 25 2. On the HZOZ-HZSO4 Oxidation of Mesoporphyrin IX .............. 27 3. On the Oxidation of Zn(II) Porphyrinone ....................... 29 B. FORMATION OF THE ACRYLIC SIDE CHAIN .................... 34 C. SPECTROSCOPIC STUDIES OF MODEL COMPOUNDS ............. 38 III. EXPERIMENTAL ............................................ 48 CHAPTER 3 TOTAL SYNTHESIS or HEME 511 ................................ 60 I. RATIONAL OF THE SYNTHETIC STRATEGY ....................... 60 vii .< CHAPTER4 FROM 1,4-PORPHYRIN DIACET ATE - A TEST ...................... 61 FROM PORPHYRINS WITH MASKED ACETTC SIDE CHAINS - A BYPASS ................................................ 63 A. 1,4-BIS(2CHLOROETHYL)-PORPHYRm ........................ 63 B. 1,3~BIS-(2-CI-ILOROETI-IYL)-PORPHYRII\I ........................ 67 C. OXIDATION OF 1 ,3-I’ORPI-IYRINDIONE SIDE CHAINS .............. 70 FROM 2,4-PORPI—IYRIN DIACETATE - REACHING THE GOAL .......... 72 g. ANALOGUES FROM COPROPORPHYRIN IV .................... 81 EXPERIMENTAL ............................................. 85 ON THE STRUCTURE or HEME 311 ............................ 121 THE STEREO AND REGIOISOMERS ............................. 121 A. Cis- AND trans-£11 -— STEREOCHEMISTRY DEDUCED FROM NMR SHIFT REAGENT ..................................... 121 B. Cis- AND tran-iso-Ql - THE LOCATION OF ACRYLIC SIDE CHAIN . . . . 124 THE NATIVE FORM OF HEME d1 ............................... 124 THE BIOSYNTHETIC ORIGINAL OF $11 .......................... 128 PHYSICOCHEMICAL PROPERTIES OF HEIWE d1 AND MODEL SYSTEMS .................................................. 130 GENERAL CONSIDERATION .................................. 130 ABSOPTION SPECTRA ....................................... 131 A. SPECTRAL FEATURES OF PORPHYRINONES .................... 131 B. SPECTRAL FEATURES OF PORPHYRINDIONES .................. 136 NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY .............. 142 A. 1H-NMR SPECTRA ........................................ 145 B. 13C-NMR SPECTRA ........................................ 147 VIBRATIONAL SPECTROSCOPY ............................... 152 A. INFRAREDSPECTRA ...................................... 152 viii B. RESONANCE RAMAN SPECTRA .............................. 161 v. X-RAY DIFFRACTTON ANALYSIS ............................... 165 VI. REDOX POTENTIAL ......................................... 168 VII. BASICITY OF CENTRAL NITROGEN ............................ 170 VIII. EXPERIMENTAL ............................................ 171 CHAPTER 6 RECONSTTTUTION 0F CYTOCHROMEgd NITRTTEREDUCTASE wrrH NATIVE AND SYNTHETIC HEM 1 ...................... 173 L PREVIOUS WORK ........................................... 173 II. EXPERIMENTAL ............................................ 174 A. PREPARATION OF APOPROTETN ...... , ..................... 174 B. PREPARATION OF HEME g1 SOLUTION ....................... 176 C. RECONSTITUIION OF NTTRTTE REDUCI‘ASE ................... 176 D. ESTIMATION OF PROTEIN ................................. 177 E. ACTIVITY ASSAY ......................................... 177 III. SPECTRAL CHARACTERIZATION .............................. 178 Tv. RECOVERY OF ACTIVITY ..................................... 183 IV. DISCUSSION .............................................. 188 CHAPTER 7 SUMMERY AND PROSPECTS ................................. 191 1. EVALUATION OFTHE PRESENT WORK ......................... 191 IT. FUTURE WORK ............................................ 192 REFERENCES .......................................................... 194 ix Table 10 11 12 13 14 LIST OF TABLES Page 13C-NMR chemical shifts of LIB-porphyrindione 6, model compound 10 in comparison with that of natural heme g1 ........................... 47 1H-NMR chemical shifts of Q. free base in comparison with its stereo- and regioisomers ........................................... 126 Meso proton chemical shifts differences between hemes and free base .......... 127 UV-vis spectra of varies porphyrinones in comparison with chlorins ........ 133 UV-vis spectra of Cu-metallated and protonated porphyrinones in comparison with chlorins .......................................... 135 UV-vis Spectra of Cu-metallated and protonated porphyrindiones ........... 139 UV-vis spectra of 1,3-porphyrindiones in comparison with isobacteriochlorins .............................................. 143 1H—NMR chemical shifts of heme d related porphyrin, porphyrinones and porphyrindiones ............................................. 146 1H-NMR chemical shifts of the meso protons of porphyrindiones in comparison with their analogue bacteriochlorins and isobacteriochlorins ...... 149 13C-NMR chemical shifts of heme d1 related porphyrin, porphyrinone and porphyrindione .............................................. 150 Infrared absorption bands of _d_l and porphyrindiones in comparison with sirohydrochlorin and isobacteriochlorin ............................... 154 Infrared absorption bands of porphyrinones in comparison with chlorin ........ 159 Redox potentials of porphyrin, porphyrinones and porphyrindiones ........... 169 The calculated energies for the HOMOS and LUMOS of porphyrin, porphyrinone and porphyrindione zinc complexes ........................ 170 15 Recovery of nitrite reductase activity after reconstitution of the apoprotein with native and synthetic heme g1 ........................... 189 FIGURE 10 11 12 13 14 15 LIST OF FIGURES page Examples of metalloporphyrinoids and their parent ring structures ............ 2 The shape, dimension and symmetry of cytochrome Cd] ...................... 7 UV-vis absorption spectra of heme d1 versus 1,3-OEPdione 8 in CI-IZCIZ ........ 17 UV-vis absorption spectrum changes of heme d1 during hydrogenation in formic acid ................................................... 19 Natural abundance, broad-band proton-decoupled 13C-NMR of the free base methyl ester of g1 in CDC13 at 32 °C ........................... 20 UV—vis absorption spectra of 1,3-dione 8, 53 and 10 in CHZCIZ ............... 40 UV-vis absorption spectra of model compound 10 (A) in the form of Cu(II) chelate in Cl-ICl3 and (B) in the protonated form in formic acid ......... 42 UV-vis absorption spectra of 10, (A) ferriheme chloride in acetone with a trace amount of HCl; (B) alkaline ferriheme in CHZCIZ/ acetone containing tetrabutylammonium hydroxide; (C) pyridine hemochrome in CH2C12/ pyridine ................................................ 43 250 MHz 1H-NMR spectrum of model compound 10 in CDC]3 ............... 44 Natural abundance, broad-band proton-decoupled 13C-NMR of model compound 10 .................................................... 46 250 MHz 1H-NMR spectra of cis-(_l1 versus trans-_d_1 in CD03 ............... 122 The Eu(fod)3 induced chemical shift (in ppm) of meso protons of cisrdione 59a and trans-dione 59b ................................... 123 UV-vis absorption spectra of cis-_cl1 versus cis-iso-g1 in CH2C12 .............. 125 UV-vis absorption spectrum of porphyrinone 135 in CHZCIZ ................ 132 UV-vis absorption spectra of 2,3-dione 32, 1,3.dione 8 and 1,4-dione 33 in CHZClz ........................................... 137 xii 16 17 18 19 20 21 22 23 24 25 26 27 28 29 UV-vis absorption spectra of 1,5—dione 123 and 1,6dione 124 in CH2C12 ....... 138 UV-vis absorption spectra of Cu (11) 2,3-dione 32 and 1,3-dione 8 in CHZCIZ ....................................................... 140 UV-vis absorption spectra of 1,4-dione 33 and 1,3-dione 8 in CHsz with CF3C02H (1%) ............................................. 141 250 MHz 1I--I-NMR spectrum of porpnyrinone 111 in CDC13 .................. 148 FT-IR spectra of g1 and Cu (II) 91—1 , samples are prepared as a thin films on NaCl pellets ................................................. 153 FT-IR spectra of porphyrin 61 and acryloporphyrin 146, samples are prepared as a thin films on NaCl pellets .............................. 157 Resonance Raman spectra of (A) natural —dl Cu (II) complex; (B) 1,3-dione 6 Cu (11) complex and (C) acrylo-lS-dione 10 Cu (II) complex at ~2 °C, sample A and B in CH2C12, and sample C as ~1 mg/ 100 mg KBr .............. 162 Resonance Raman spectra of Cu (11) complexes of synthetic Q] and 1,3-dione 59 in CHZCIZ ............................................ 163 (A) Molecular structure and atom names of Cu (II) 1,3-dione 8, (B) bond distance, (C) deviation from the plane of the four nitrogens. (D) Free base of 1,3-dione 8, (E) Another View of D ....................... 166 Geometric model of the ligand periphery of Ni (11) isobacteriochlorin ......... 167 UV-vis spectra of Fe (111) £11 in acetone containing 0.024 N HCl and about 10% of water, (A) synthetic heme g]; (B) native heme Q] from P. stutzeri ..................................................... 179 UV-vis spectra of synthetic d1 in 0.25 M of phosphate buffer at PH 7.3 ........ 181 UV-vis spectra of pure preparation of nitrite reductase from P. stutzeri in 0.25 M phosphate buffer at PH 7.0 .................................. 182 UV-vis spectra of apoprotein of nitrite reductase from P. stutzeri after removal of heme d1, the apoprotein was dissolved in 0.25 M of xiii 30 31 32 phosphate buffer at PH 7.0 ......................................... 184 _ UV-vis spectra of reconstituted nitrite reductase in 0.25 M of phosphate buffer at PH 7.0, (A) native d1 reconstituted and (B) synthetic d1 reconstituted .......................................... 185 Progress curves of nitric oxide and nitrous oxide production from 1 mM of nitrite by intact nitrite reductase from P. stutzeri ......................... 186 Progress curves of nitric oxide and nitrous oxide production from 1 mM of nitrite by the reconstituted nitrite reductase, (A) reconstituted with synthetic heme d1 and (B) reconstituted with native heme d1 ............... 187 xiv CHAPTER 1 INTRODUCTION: THE GREEN HEME PROSTHETIC GROUP OF CYTOCHROME 9511 NITRITE REDUCTASE I. SIGNIFICANCE AND BACKGROUND Chemists interested in the bioorganic and bioinorganic fields are always fascinated by the rich and colorful chemistry of porphyrinoids. Varying a single structure theme, that of uroporphyrinogen, Nature has selected a rich variety of magnificent structures to take part in a diversity of fundamental biological functions in all kinds of organism ranging from bacteria to man.1 Examples shown in Figufl are the better known porphyrin family compounds; hemin (iron porphyrin), chlorophyll _a_ (magnesium chlorin), bacteriochlorophyll a (magnesium bacteriochlorin), siroheme (iron isobacteriochlorin), vitamin Bu (cobalt corrin) and coenzyme £430 (nickel corphin). In fact, metalloporphyrinoids are of so much importance that they deserve the label as "the pigments of life".2 This thesis concerns a new member of metalloporphyrinoids: heme 11 (1)3 This iron heme has an unprecedented LIB-porphyrindione core structure with an arcylic side chain present, and has been found as the cofactor of cytochrome ggl-type nitrite reductase in a large number of denitrifying bacteria. The present work addresses various synthetic aspects of the unique structure of d1 heme, the physicochemical properties of its novel macrocyclic ligand and the biological significance of this prosthetic group in the I" .‘ a" CO CH: CO C "03¢ co'H colphy‘yl’ CO'WS,“ "I hemin chlorophyll a bacteriochlorophyll a siroheme vitamin B12 Figure 1 Examples of metalloporphyrinoids and their parent ring structures. denitrification process. Ho,c ° 902“ O \ H02C COZH 1. Heme £11 A. Nitrogen Cycle and Denitrification4 Though the early history of the earth involved massive geochemical and geophysical changes, living things are today responsible for the major chemical transformations which take place on our planet. It is convenient to express these changes, on a gross scale, in the form of hypothetical cycles of the biologically active elements, such as nitrogen, sulfur and phosphorus. A simplified nitrogen cycle is shown in Scheme 1 to indicate the major biological processes involved. The immediate product of biological dinitrogen fixation is ammonia, which is taken by plants and microbes to make their protein and other organic nitrogen matters, which then serves as the nutrients for human and animals. The degradation process brings organic nitrogen back to ammonia which is oxidized to nitrate. The reduction of nitrate and nitrite undergoes two different routes: the assimilatory nitrite-reducing process to generate ammonia; and the dissimilatory route producing nitrous oxide and dinitrogen gas. The balance between fixation-nitrification and denitrification is fundamental to the persistence of life on this planet. N20 DENITRIFICATION Dissimilatory Nitrous nitrite reductase reductase Nitrate reductase NO3‘ = r NOz' N2 Assimilafory nitrite reductase N ITROFIC ATION FIX ATION NH 3 DEGRADATION < ) ASSIMILATION ORGANIC NITROGEN (plsnts, microbes) ( uman, animals) Scheme 1. The biological nitrogen cycle. The important but often contradictory consequences of the denitrification process include that: (1) it generated in the past almost all of the gaseous nitrogen in the earth's atmosphere and now maintains the standing stock by continual production in an annual turnover of some 2 x 108 tons of nitrogen; (2) it causes up to 30% loss of fertilizer-fixed nitrogen from agricultural soil, thus limiting plant productivity; (3) it emits free nitrous oxide which has been found to contribute to ozone destruction in the stratosphere and to the increased planet temperature through the so called "green house effect"; (4) it removes nitrate or nitrite from waste-water and finds important industrial uses in waste-water treatment plants; (5) finally, it results in the temporary accumulation of toxic nitrite which is known to react with secondary amine to form carcinogenic nitrosamines in food, water or digestive system. Man has dreamed of controlling and curtailing denitrification for decades. But without proper knowledge of the chemical mechanisms and its ecological impacts, there are enormous risks involved in tempering with a process of such global significance. The delineation of the general pathway of this process only occurred recently. B. Cytochrome cdl Nitrite Reductase It has been found that NO3' is first reduced by molybdenum-containing nitrate reductase to nitrite.5' 6 and the dissimilatory nitrite reductase reduce N 02' to nitrous oxide (N20) which is then converted to dinitrogen through a separate nitrous oxide reductase (a copper enzyme).7r 8 Clearly, it is nitrite reduction that defines denitrification. There are two types of denitrifying nitrite reductases reported;4 one is a multiheme protein named cytochrome £21 and another is a Cu—containing enzyme. The cytochrome g1 enzyme appears to be more prevalent in nature and so far has been isolated from a large number of chemoautotrophic denitrifying bacteria, including Pseudomonas aeruginosa.9 Thiobacillgs denitrificans 10 Alcaligenes faecalis (formerly Pseudomonas denitrificans),11 Paracoccus denitrificans (formerly Micrococcus denitrificans),12 Paracoccus hilodenitrificans13 and Pseudomonas stutzeri.14 In 1958, Horio extracted and partially purified four different kinds of soluble respiratory components from Pseudomonas aeruginosa.15r 16 Among the purified components, there was a greenish-brown fraction which possessed a complex spectrum containing both a g-type cytochrome and a so-called 12 cytochrome. This enzyme component exhibited general properties of a cytochrome oxidase: aerobically it oxidizes ascorbate, hydroquinone and reduced cytochrome 9551. These reactions were found strongly inhibited in the presence of cyanide and carbon monoxide. During the early purification, this preparation was called gtochrome Q because of its greenish-brown color; later it was renamed pseudomonas cytochrome oxidase.17 The property of the same entity as a nitrite reductase was recognized by Yamanaka and Okunuki in the subsequent years.18r 19 It was uncertain at that time whether this preparation is one enzyme, which itself has two activities, or the preparation contains two distinct enzymes. Yamanaka and Okunuki20 in 1962 isolated the greenish-brown component in a pure crystalline state and proved it as a single enzyme. They further confirmed that the enzyme contains two different hemochromes and possess a dual enzymatic property, that is, it acts as both a nitrite reductase and a cytochrome oxidase. Thus the name "pseudomonas cytochrome £551 : nitrite, Oz oxidoreductase" was once used by these workers to indicate the double function of the enzyme when cytochrome £551 was involved in the redox reaction.21 Although the function initially attributed to this enzyme was that of cytochrome oxidase (or oxygen-reductase) and has been until recently the best known function, it is now considered as a secondary and nonphysiological function. The 92 heme has been renamed as 91, and later to -d—1 after 70's, thus this enzyme is most properly named as cytochrome £21 nitrite reductase. As shown in Figure 2. cytochrome $11 consists of two identical subunits and has a twofold axis. Both subunits have a oblong Shape with a length of 90-100 A. The dimension of the enzyme is less than 73 A and the volume is around 145 nm3.22' 23 Each subunit of molecular weight of 63,000 dalton contains a red g-type heme prosthetic group and a green colored heme d1 <73A com Ho,c 60,14 Figure 2. The shape, dimension and symmetry of cytochrome g1. moiety.24r 25 In both oxidized and reduced forms of the enzyme, heme g and heme d1 groups in each subunit are oriented perpendicularly to each other with a closest distance of 13-15 A29 27 All four heme prosthetic groups are found at one end of the protein molecule.22 Heme g is covalently bonded to the polypeptide backbone by two thioether linkages to cystine. The iron atom of heme g is six-coordinate and low spin in both Fe(II) and Fe(III) oxidation states and the axial ligands are thought to be histidine and methionine. Heme d is noncovalently associated in the protein pocket. In the absence of exogenous ligands, the Fe(II) heme Q] is high spin and is thought to be five-coordinate with a nearby axial N donor, possibly, an imidazole residue.28 The Fe(III) d1 has been suggested as low spin and presumably six-coordinated with an additional endogenous protein residue or a water molecule providing the other axial ligand.29 In vivo this enzyme, while water soluble after isolation, is normally associated with cell membrane and its physiological electron donor is cytochrome £551 and/ or a blue copper protein azurin.10' 42 The kinetics work to date supports the idea that the heme 9 sites are associated with electron uptake, while the heme g1 sites are responsible for electron donation to exogenous ligands, i.e. the substrates of the enzyme, such as oxygen and nitrite. This enzyme is only produced anaerobically in the presence of nitrite and serves its intended function as nitrite reductase. Once formed, it can also donate electrons to molecular oxygen, reducing oxygen to water, thus functions also as cytochrome oxidase. The turnover number of cytochrome 9% as oxygen reductase under the physiological condition corresponds to 600 moles of oxygen consumption per mole of the enzyme per minute at 37 °C . Under anaerobic condition, one mole of this enzyme reduces 4,000 moles of nitrite to nitrogen oxides per minute at the same temperature.30 Thus its nitrite reductase activity is about sixfold more than its oxidase activity. The constitutive cytochrome-oxidases are found always more active than cytochrome 9d] as the oxygen reductase. Most recently, another nonphysiological property, the carbon monoxide oxygenase activity, of cytochrome c_d_1 was studied by Timkovich and Thrasher.31 C. On the Mechanism of Nitrite Reduction Although nitrite reduction is the alleged function of cytochrome g1, controversy has arisen about the nature of the physiological intermediates or products in the course of reduction. Mainly, there are three proposed pathways concerning how N02” is converted to N 20, Scheme 2. The first, +2e' +2e', 4H+ I | Flea-N203 ———>Ee—NZO32' _— lie-N20 : N02. I ’21'120 Scheme of Averill and Tiedje Fe + N20 +2H+, -H20 F'e-NO+ +12 [FoNO-l -——> Fa-NO' ——->Fe + NO’ , Scheme of Hollocher ZNO' ' N20 +1e’ \ E-NO' ———’ E + NO' NO° E-NO' Scheme 2. The mechanism of nitrite reduction. also the oldest, proposes two enzymes: a nitrite reductase and a nitric oxide reductase with NO as the obvious intermediate.4 This pathway offers no explanation how the N -N bond is formed. The second pathway, proposed by Averill and Tiedje}2 requires that the conversion be carried out by one enzyme and that N-N bond formation occurs before reduction. The third pathway, proposed by Hollocher,33' 3’4 can also be carried out by a single enzyme, and suggests that reduction of nitrite to nitroxyl occurs, followed by dimerization of nitroxyl (NO') to form N20. This pathway could 10 accommodate a separate NO reductase. These three schemes have coexisted since 1982 but it has not yet been resolved as to which one is actually responsible for denitrification, since this determination is not trivial. Evidence of the pros and cons for each proposal has been presented in the literature.32'39 It can be simply pointed out that the key reaction of N-N bond formation must be intimately associated with the heme dl-nitrosyl [Fe-NO+] complex, which in turn, must depend on the nature of the d1 heme prosthetic group. D. Heme (11 The green heme cytochrome cytochrome Q] from Pseudomonas aeruginosa was first observed by Horio and coworkers in 1958. It was first classified as an "az-type" heme for having a typical absorption maximum around 630 nm according to the designation Keilin assigned for cytochromes absorbing in the red region.“ This heme was first isolated by Yamanaka and Okuniki in 1961,41 and its visible spectra in both oxidized and reduced forms and of pyridine, CN’, NO and CO derivatives were carefully documented at the same time.42 It was frequently compared with and related to another green heme, Barrett's green heme from Aerobacter aerogenes and Escherichia @1443 which was also designated as "_a_2". Since Barrett had previously determined his heme _a_z to be an iron chlorin it was assumed that "heme _a_2" from Pseudomona_s aeruginosa and other denitrifying bacteria must also be a chlorin. To avoid confusion with 1:13 hemes of mammalian cytochrome oxidase heme g2 was renamed "d" after 1970. Lemberg and Barrett noticed the spectral and solubility differences that exist between the "classical heme d (a2)", as obtained from sources such as L g91_i, and the extractable heme of pseudomonas nitrite reductase and therefore suggested the name heme 511 to 11 distinguish the latter.44 However, the idea of heme d1 being a chlorin with a structure like 2 was unchallenged in more than two decades. Timkovich in 198445r 46 managed to isolate and purify enough material to allow a careful structure determination with the aid of modern instruments. H . OH Ho.c / no.6 0.1-1 com Ho,c Com / Haze co!" Hole COIH HOIC COIH 2 3 uroporphyrinogen III The Q] structure concluded by Timkovich has also a chlorin core as shown by 3 with the most unusual arrangement of substituents. This structure is considered surprising Since it defies all known biosynthetic pathways by which the other porphyrinoids come into being. To date there are no known exceptions to the fact that all the naturally occurring tetrapyrrolic macrocycles derive their substituent pattern from uroporphyrinogin III and structure 3 is obviously not one of them. After a careful examination on the spectra data published, C. K. Chang3 in 1985 proposed the revolutionary structure 1 for heme _d_ 1, which possesses a porphyrindione (dioxo- isobacteriochlorin) core structure hitherto unknown in the biological world. This structure not only fits all the spectroscopic and analytical data better but also is compatible with the common porphinoid biosynthetic pathway. II. OBJECTIVES OF THE PRESENT WORK 12 The principal objectives of our study are to understand the structure and function of £11 heme prosthetic group in cytochrome 9d] nitrite reductases. Specifically we intended to study the following: 1. Further characterization of heme (11 structure. To verify the proposed porphyrindione (dioxo-isobacteriochlorin) structure of the _cll moiety by chemical derivatization of the natural pigment and by comparing its spectral properties side-by-side with those of well-characterized synthetic model compounds. 2. Synthesis of d1 prosthetiggroup and its structural and functional analogues. To provide definitive proof of structure and produce a copious supply of heme $11 and its analogues for other experiments. 3. Physicochemical properties of heme d1_and model systems. To build a knowledge base of the intrinsic properties and reaction profile of heme £11 and to identify unique properties not present in the other systems. Using synthetic heme d1 and its analogues, we plan to examine the spectral properties of these compounds by UV-Vis, NMR, IR, and RR spectroscopies, and determine their redox chemistry by electrochemical means. Throughout these studies, whenever necessary, comparative studies on chlorin and isobacteriochlorin compounds will be carried out such that the attributes of the porphyrindione system can be evaluated. 4. Reconstitution of gytochrome cdl and other protein with smthetic (11 m its analogues. To replace protoheme of myoglobin with dl-type hemes to allow studies of heme (_11 in a well defined protein environment. To replace the natural heme £11 in cytochrome Q] with synthetic heme d1 and other related hemes so that the functional role of heme d1 moiety may be revealed. Cytochrome g_d_1 will be grown and isolated from P. aeruginosa or P. stutzeri. 13 Standard measurements and reactivity assays will be performed with reconstituted enzymes to document any structure-function relationships. In particularly, the ability to produce N 20 will be monitored in the reconstituted systems. 111. RESULTS AND PRESENTATION Significant progress has been made during the last three years toward the above objectives. Our proposal that the green-colored heme gl_1 is not a chlorin but a porphyrindione (dioxo-isobacteriochlorin) has been fully substantiated; the suggested structure has been proven entirely correct; total synthesis of this new chromophore is now achieved; redox and coordination chemistry has been studied; native heme d1 and cytochrome g1 have been isolated from Pseudomonas stutzeri: apoprotein of cytochrome 9d] has been reconstituted successfully with both native and synthetic heme $11 as well as a number of d1 analogues and related hemes. In the following chapters, Chapter 2 provides further evidence on the heme 4. structure as we proposed, describing the synthesis and properties of several model compounds. The methodologies employed for the formation of 1,3-dione core structure and acrylic side chain are described in detail. Chapter 3 describes the total synthesis of heme Q. from different approaches and the convenient synthesis of a £11 analogue from coproporphyrin IV. In Chapter 4, the steric and the structural isomers of Q] are compared, the questions concerning the native form as well as biosynthetic origin of this novel heme are also discussed. Chapter 5 is devoted to the physicochemical property of $11 and its model systems in comparison with those of the other porphyrinoids, especially chlorin and isobacteriochlorin systems. Chapter 6 14 deals with the reconstitution of cytochrome _c_c_ll nitrite reductase with both native and synthetic heme d1 in an effort to understand the mechanism of nitrite reduction. Finally, some thoughts on the future work are presented in Chapter 7. IV. NOMENCLATURE OF HEME g1 AND RELATED PORPHYRINS The nomenclature of porphyrinoids with keto group(s) on the ring has not been standardized. Neither the names "geminiporphyrine-diketone and geminiporphyrine—monoketone" as Inhoffen47' 43 used nor "oxochlorin and dioxo-isobacteriochlorin" as Johnson49r 50 named are convenient and specific. The prefix "oxo-", in fact, could be confused with "oxyporphyrin" or "oxophlorin", which denotes a porphyrin with an oxygen attached to the methine bridge. Furthermore, we now know that these ketone derivatives have very little chemical properties in common with those of the corresponding chlorins, isobacteriochlorins, or bacteriochlorins.51 It is probably more appropriate to consider them "quinones" of porphyrins, hence we propose the use of "porphyrinone" and "porphyrindione". In addition, we suggest the trivial names "dioneheme" for heme _d_l (pronounced either like dye-own-heme or d-I-heme) and the "6-acrylo-1,3-porphyrindione" for the metal-free _d_l, a modification of Timkovich's "acrylochlorin".45' 46 The prefix numbers "6." and "1,3-" are used to specify the positions of the acrylic side chain and the keto groups on the ring. The advantage of our nomenclature is that all common names of precursor porphyrins can be retained and put to work. For example, the monoketone derivative of mesoporphyrin (5) is named as "mesoporphyrinone" and the diketone derivative (6) is named as "1,3-mesoporphyrindione", and the diketone 7 15 from coproporphyrin IV is "2,3-coproporphyrin(IV)-dione". Moo.c COIM- Moozc co,m Moo,c comic 5 6 7 As illustrated in the structure 4 of metal free heme d], we use the conventional Fischer 1, 2 ------ , 8 system for the substituents, and the positions of the two keto groups are assigned as "1,3-". The four pyrrole rings are numbered A, B, C and D clockwise starting from the up-left ring bearing the first keto group and the four meso positions are indicated as OT, p, Y and 5 accordingly. For convenience, the position assignment of the substituents of heme g1 (structure 4) is taken as a standard, thus the acrylic side chain is at position 6 on ring C and the propionic chain is at position 7 on ring D. All the substituents, including the ring keto group, of other porphyrins, porphyrinones and porphyrindiones are numbered accordingly throughout this work. CHAPTER 2 FURTHER CHARACTERIZATION OF HEME g1 STRUCTURE I. EVIDENCE OF THE 1,3-PORPHYRINDION E STRUCTURE OF HEME g. Absorption Spectra The most persuasive evidence in favor of a non-chlorin structure of $11 is from the visible spectrum of this natural pigment in the form of its free base methyl ester. Examination of more than twenty of the naturally occurring or synthetic chlorins revealed that the chlorin visible spectra are remarkably homogeneous in that the overall pattern is relatively unperturbed by electronic effects of the peripheral substituents. However the overall shape of the d spectrum does not resemble that of any chlorin compound, but that of a typical isobacteriochlorin, such as sirohydrochlorin,52 except that all peaks are red shifted. This leads to a less known isobacteriochlorin-type compound: a 1,3-porphyrindione 8 (dioxo-isobacteriochlorin)53 whose absorption peaks are significantly red shifted from ordinary isobacteriochlorins. Metal complexes of the natural heme g1 and compound 0 8 8 are made and their visible spectra are measured quantitatively. AS shown in Figure 3a and 3b, the neutral, free base spectra of the two are dissimilar in 16 17 Com - done-Ole I 88¢ 1 ‘ 1 ‘ ' “:- 3” m 500 m 700 O” 400 900 m m m Figure 3 UV-vis absorption spectra of heme 511 versus 1,3-0El’dione 8 in CHZCIZ. 18 both Soret and visible regions. However, the multiplicity of bands seen in the free bases due to the broken of symmetry by the central protons and vibronic overtone diminishes when the rings are protonated and metallated.54 The correspondence between d1 and 8 become more apparent when these forms are compared as shown in Figure 3c-3f. The low extinction coefficient of the Soret absorption bands and particularly the low ratio of A 50,“ versus Avis band seen in both compounds certainly argues forcibly for the unlikelihood of chlorins or porphyrins. A consistent feature of the spectrum comparisons is the ca. 20 nm red shift of the highest wavelength of Q]. The proposed structure 4 does contain the extra conjugation of the acrylate which could account for the red shift. As shown in mm, when this olefinic substituent of heme d1 is hydrogenated by treatment with HZ-PtOz, the band shifts to a wavelength coincident with that of dione 8. Lligh Resolution Fast Atom Bombardment Mass Spectra ' One of the key questions of Timkovich's original assignment 345 is the inability to obtain the parent MS ion of 714; an "(M-2)“ of 712 was observed instead. This was attributed to the loss of inner pyrrolic protons, an event observed in the other porphyrins.52 The 6-acrylo-1,3-porphyrindione structure 4, has two less hydrogen than 3 and therefore fits perfectly with the observation. New fast atom bombardment mass spectra were obtained in a matrix that consistently gives (M+1)+ ions for porphyrins.55 Mass units for the protic methyl ester, (M+1) = 713, the 2H3 methyl ester (725, 726, and 727 corresponding to two 1H, one 1H and one 2H, or two 2H at the inner pyrrolic NH's), the copper chelate of the protic methyl ester (773), and the copper chelate of the ethyl ester (829) are only consistent with structure 4. 19 - hydrogenation of d, T o +- ‘ o c. o e A o a a o q B A nm 400 500 600 700 800 Figure 4 UV-vis absorption spectrum changes of heme 511 during hydrogenatipn in formic acid/Pt. (A) 111 before hydrogenation; (B) after 30 s; (C) after 90 s; (D) model compound 8 in formic acid. The d1 ring was also hydrogenated during the time, leading to decreased absorption. 20 13C-NMR spectra The natural abundance 13C-NMR of about 2 mg of natural a. methyl ester, after 708,625 transients (11 days), yielded a spectrum shown in Figure 5. |50 IOO 50 0 PW“ Figure 5. Natural abundance, broad-band proton-decoupled 13C-NMR of the free base methyl ester of 511 in CDC13 at 32 °C. The chemical shifts are given in Table l. The chlorin structure 3 has two hydroxymethyl groups on the p-pyrrole carbons while structure 4 proposes acetate esters. The expected 13C chemical shift is very different for the methylene carbons of these two moieties. Methyl groups on the saturated B-pyrrolic carbons in model porphyrins are usually very close to 23 ppm.56r 57 The incremental shift for replacing an H with -C02R is about 20 ppm, predicating 43 ppm for -_C;H2COZR. The incremental shift for replacement with -OH is 48 ppm, predicating 71 ppm for 21 -_C_HZOH. The spectrum of d1 showed no evidence of any peaks near 71 ppm while two distinct resonances are evident at 43 ppm giving compelling support for the structure 4. With the above data we were in a position to dismiss the chlorin structure for heme d1. However, further work was still required to ascertain that Q] does have a 1,3-porphyrindione core structure and the conjugating acrylic side chain is attached at the position 6 of ring C as we proposed. Therefore, we turned to synthesizing well-characterized model compounds. 11. STUDY WITH MODEL COMPOUNDS M.°2c 0 C01". 0 o O 0 0 Moore cot". \ / "Cage COQM. M003: C02Mo 9 10 1 1 Initially, model compounds 9 and 10 were designed to duplicate the structure character of 4. Model 9, with a 1, 3-dione nucleus and with one acetate function attaching to each of its two tertiary carbons, has the elements of the northern half of 4, its NMR spectra would be invaluable for verification purpose. Compound 10, with an acrylic side chain setting at the desired position on ring C, besides providing useful NMR data, should give a visible spectrum virtually identical to what has been observed with $11. To verify the position of the acrylic side chain in relation to the two keto groups, a regioisomeric compound 11 of 10 was also considered necessary. 22 To obtain these model compounds we were facing two challenging problems, that is, developing an elegant pathway to form the 1,3-porphyrindione core structure and establishing an effective method to introduce selectively the acrylic side chain to the ring C of the macrocycle. A. Fomjjon of the 1, 3-Porphyrindione Skeleton We suggested originally that the keto groups in structure 1 of d1 are possibly derived from a pinacolic rearrangement of vicinal diol or epoxide.3 Precedence of such reactions in porphyrin so far has been limited to only a few systems. Hans Fischer and his Students58 reported in 1930 a novel method to obtain green-colored chlorin from porphyrin. They reacted mesoporphyrin IX in concentrated sulfuric acid with hydrogen peroxide and yielded a product which was designated as "dioxymesoporpyrin" even though the elementary analysis results were ambivalent in showing whether one or two oxygen atoms had been added to the porphyrin. Indeed they later concluded that the green product resulted from this acidic medium contains only one extra oxygen atom and called it "anhydrochlorin".59 The structure they proposed, formulated as a epoxide ring cross a pyrrole double bond, was again incorrect. Johnson”! 50 and Inhoffen,47' 48 about 20 years ago, independently reinvestigated such oxidation reaction using symmetrical etioporphyrin and octaethylporphyrin (OEP) and established that the true identity of the major product from this hydrogen peroxide-sulfuric acid oxidation is a keto chlorin (porphyrinone) formed by a pinacol rearrangement of the intermediate diol or epoxide. By our inference, Fischer's "dioxymesoporphyrin" must also be some sort of keto derivative but the exact products of the mesoporphyrin oxidation was far from clear. For being an unsymmetrically substituted 23 porphyrin, mesoporphyrin could produce up to 8 isomeric porphyrinones. Furthermore, as reported originally by Inhoffen and Nolte,47' 43 and more recently by Chang60 using OEP, the oxidation reaction does not stop at the monoketone level; diketones, and even triketones, arise almost simultaneously under the reaction condition optimized for porpyhrinone. With OEP there are five diketones and four triketones identified; for mesoporphyrin, there could be 14 diketone regioisomers alone statistically without counting the diastereomers! The sheer number of anticipated isomeric products from mesoporphyrin must have dissuaded attempts to reexamine this reaction for after more than a half century, Fischer's pioneering yet unsolved work stands unsettled. A central question concerning the pinacol rearrangement is the migratory aptitude when the substituents involved are not equivalent,61 such as in the mesoporphyrin's case, therefore OEP is inapplicable to address this question. Hans Fischer”! 62 demonstrated that hydroxylation of type-IX porphyrin can be achieved with osmium tetroxide although the resultant isomeric dihydroxyl compounds were not individually identified. Recently, our group devised a convenient method63 with 0804 to effective dihydroxylation of porphyrins to obtain chlorin diols. These diols indeed rearranged in acidic medium to give excellent yields of corresponding porphyrinones. Although this synthetic study is more useful in preparing for heme _c1 and its model compounds,64 it offers a potential way to prepare porphyrindione. Employing a variety of specifically synthesized porphyrins, C. Sotiriou of our group65 completed a series of experiments aiming at elucidating the reactivity as well as the migratory aptitude of the biologically important porphyrin side chains. The results are cited here: 1). Relative reactivity of the pyrrole double bond towards dihydroxylation is 24 proportional to, barring electronic effect, the size of the substituents, the larger the substituent, the slower the rate, thus: H = Methyl (Me) > Ethyl (Et) > Acetic (A) > propionic (P). 2). Migratory aptitude of the substituents is mainly related to their electronic effect; hydrogen, ethyl, alkyl including propionate side chain will migrate over methyl group and acetate side chain has a lower mobility than methyl. These general rules hold true in most porphyrins. Although the two-step OsO4 oxidation-acid catalyzed pinacol rearrangement may offer some control over the unwanted porphyrinone isomers, it is not useful for producing the 1,3-dione of the isobacteriochlorin-type derivatives. In the presence of excess amounts of osmium tetraoxide only tetrahydroxybacteriochlorin was observed. Even deuteroporphyrin, which has built-in steric advantages, was found to react with an excess of 0504 to yield only the tetrahydroxybacteriochlorin without any trace of isobacteriochlorin. This reaction pattern may be due to the preferred diagonal n-electron delocalization pathway presents in all porphyrins, which prompts the saturation of the two isolated, diagonal pyrrole p,p'-doub1e bonds with minimum lost of fl-energy. A similar argument has been advanced to account for the exclusively diagonal reduction of the tetraphenylchlorin by diimide to yield bacteriochlorin.66 In a medium of sulfuric acid, however, porphyrin could become doubly protonated, the influence of the valence tautomerism would become insignificant, and isobacteriochlorin may be formed. Indeed, in the reaction of OEP with hydrogen peroxide-sulfuric acid, the combined yield of three dioxoisobacteriochlorins (1,3-porphyrindiones) is better than that of two dioxobacteriochlorins (1,5- and 1,6-porphyrindiones).60 Therefore the HZOZ-HZSO4 oxidation was applied in an attempt to prepare model 25 compounds 9 from porphyrin tetraacetate 12 and the core structure of 10 from mesoporphyrin IX respectively. 1. On the H292fl2594 oxidation of porphyrin tetrJaacetitg The porphyrin tetraacetate 12 was synthesized through a 2+2 dipyrrylmethene condensation pathway, Scheme 3, following Fischer's classical method.67r 68 The dipyrrylmethene 14 was easily obtained in crystalline form by brominating pyrrole 13 in acetic acid. Heating 14 in fused methylsuccinic acid at 130° for 6 hours, gave porphyrin 12 in a yield of 12%. However, the four symmetrically substituted acetate groups render this compound not only poorly soluble in most organic solvents, but also unexpectedly inert toward HZOZ-HZSO4 oxidation and subsequent pinacolic rearrangement. Only a very low yield of porphyrinone 15 was obtained together with some y-lactone compound 16 after a prolonged reaction of 12 in the hydrogen peroxide-sulfuric acid medium. No trace of the desired porphyrindione 9 was detected. An alternative OsO4 oxidation of this porphyrin gave only a poor yield of dihydroxychlorin 17, which refused to undergo pinacolic rearrangement to 15 in concentrated sulfuric acid but cyclized instead almost quantitatively to lactone 16. These results indicated clearly that the four electron-withdrawing acetate side chains have rendered the pyrrole double bond inactive toward H202 oxidation as well as osmic attack. The pinacolone formation was found sluggish due to the electron-deficiency on the ring, which might destabilize the cation necessary for rearrangement, thus the reaction tends to form the thermodynamically stable S-membered lactone 16. Similar results were also observed in our later experiments. 26 C02Mo 13 .....,.Z;£ :— H CH,OH/ H’ 1) com. . 0 co.m OH H 0 com- 16 a ' ’ 12 MoO,C MoO,C M.°3c M.Ozc 0,0. COZMQ OH OH COzMQ . 17 MOOIC MeO,C Scheme 3 27 2. On the HZQZflZSOQxidation of mesoporphyrin We were optimistic on the oxidation of mesoporphyrin with hydrogen peroxide-sulfuric acid based on the consideration that no strong electron-withdrawing group like acetate are present and that the pinacolic rearrangement of the diols would be dictated by the specific migratory aptitudes of the substituents. As mentioned earlier, both ethyl and propionate side chains in a Vic-dihydroxychlorin have a higher migratory aptitude as compared with the methyl group. Because that the diol formation is highly sensitive to the size of the side chain, the "northern diols" should be preferred versus the "southern" diols and the desired 1,3-porphyrindione 6 would be the favored product. Thus, mesoporphyrin dimethyl ester dissolved in concentrated sulfuric acid was reacted with H202, and after about 30 minutes the solution was neutralized by sodium acetate. The solid product, collected by filtration, contained most of the ketone products with intact propionic esters. Chromatography of this material on silica gel went surprisingly well, and nine different compounds, excluding the unreacted mesoporphyrin, were obtained (Scheme 4); the total yield was about 30% (reproducible in three separated runs).68 Structure identification in most cases was straightforward, aided by UV-vis absorption and 1H-NMR spectra. The differentiation of monoketones 5 and 18 and also 20 and 21 was accomplished by nuclear Overhauser enhancements (NOE),63r 69 which was also of great value in confirming the assignment of 3,7-dione 22. The overall products distribution shows that indeed our anticipated reactivity and migratory aptitudes hold remarkably well with only two exceptions, the 2,3-dione 19 and 1,8-dione 24, but even in these two, one of the keto group is in the "correct" place. We suspect that 19 and 24 may arise from the respective precursors 5 and 18; the 28 mgc CO¢MO l t I t o o F x o Moo,c com. MoO,C com. MoO,C com. 5 18 6 o o «1.0,: 0 E 0 com. mo,c cow. com. more 19 21 20 0 Home o co, M. CO, M. mate C0,". 22 23 24 29 subsequent pinacol rearrangement went to the "wrong" direction because the formation of adjacent diketones is energetically favorable, which offsets the regular migratory trend. In all diketones, the presence of diastereomers can be detected by NMR (bifurcation of the pyrroline substituent signals), but attempts to separate them have not been successful. The clarification of the old literature problem has ample rewards. First, the demonstration that the oxidation of mesoporphyrin behaves in a predictable and reproducible manner and the separation and identification of individual regioisomers can be accomplished by using routine laboratory facilities and techniques immediately provide access to a rich source of all types of porphyrinone and porphyrindione derivatives. Secondly, the availability of these compounds suggests expeditious synthetic strategies for reduced porphyrin macrocycles.63r 69 The monoketones and diketones are convenient precursors to alkylated chlorins and isobacteriochlorins such as bonellin70 and sirohydrochlorin“! 72 Above all, the 1,3-porphyrindione 6 with the core structure of model compound 10, has been successfully prepared, albeit at an unimpressive yield (4.5%). 3. 0504 oxidation of Zn(II) porphyrinone The disadvantage of the above HZOz-HZSO4 oxidation of the p-substituted porphyrins, such as mesoporphyrin, is that it produces a complex mixture of isomeric products containing one, two, and three keto groups on the ring with uniformly low yields. Porphyrinone, such as 26 from octaethylporphyrin (OEP), can be prepared with significantly higher yield by an alternative 2-step reaction via OsO4 oxidation and acid catalyzed pinacolic rearrangement. However, further oxidation of 26 by 0804 invariably leads to bacteriochlorin-type compound 27, which upon 3O rearrangement gives two isomeric products, 1,5-porphyrindione 28 and 1,6-porphyrindione 29. We found that the osmium tetroxide addition preference can be altered dramatically in favor of isobacteriochlorin-type 1,3-porphyrindione 8 formation simply by metallation of the ring (Scheme 5). The zinc complex of 26 was found to react with 0504 (1.5 equivalent) in CH2C12 containing 1% pyridine to give predominantly dihydroxyporphyrinone 30 (>60% yield), which can be treated with sulfuric acid to give 1,3—dione 8. A small amount of the ring D diol 31 was also obtained which rearranged to yield about equally 8 and 32. If the synthetic goal is 8, the crude dihydroxylation products can be used directly in the pinacol rearrangement as the ratio of dione 8 to 32 is usually greater than thirty-fold. That the osmate addition mainly occurred at ring B of 26 is possibly a consequence of the electron-withdrawing effect of the carbonyl group rendering the adjacent ring D double bond less reactive. It is also noteworthy that during the pinacol rearrangement of 30 or its free base, none of the possible 1,4-dione 33 was observed. The sterically unfavorable arrangement of the four geminal ethyl groups in 33 might be the reason for its absence. Insertion of other metal ions such as Cu(II) and Ni(II) has the same effect on switching the osmate addition pattern but the yields of the osmate esters were less satisfactory. The remarkable alteration of the site of attack by metallation in the chlorin-type system appears to be a general phenomenon. Previously it has been observed that the diimide reduction of free base tetraphenylchlorin (TPC) produces only tetraphenylbacteriochlorin whereas Zn(II) TPC gives exclusively Zn(II) tetraphenylisobacteriochlorin.73 Similarly, reduction of the Ni(II) pheophorbide family of chlorin by Raney nickel promotes the formation of isobacteriochlorin.74 Whitlock and Oster 31 OEP 0304 H0 2.9 32 I 31 33 Scheme 5 32 suggested that the saturation of a diametrical pyrrole double bond in the free base chlorin may be prompted by the diagonal TI-electron delocalization pathway that bypasses the outer p-p' double bonds of the pyrroline ring and its opposite partner, leading to the bacteriochlorin formation with minimum loss of n-energy.65r 73 The preference of this valence tautomer would be diminished when the system become metallated, or protonated as in our HZOZ-HZSO4 system. However this hypothesis did not explain why the double bond saturation occurs exclusively at the adjacent ring in the metal complex since one would expect that the absence of a preferred TI-delocalizing pattern only favors a more random attack. Neither did the previous MO calculations of Zn(II)TPC show a significant difference in fl-electron density between the opposite and adjacent p-p' double bonds.65 The cause of the selectivity remains unclear. The selective saturation of porphyrinone double bond has made possible the synthesis of a variety of 1,3-porphyrindiones bearing peripheral substituents at the specific positions. As we have shown that dione 6, with the core structure of model compound 10, could be prepared by the HZOZ-HZSO4 oxidation of mesoporphyrin, with a <5% yield after repeatedly separation from a mixture of no less than nine ketone products. With the zinc method Scheme 6, dione 6 was prepared from mesoporphyrin cleanly with a much higher overall yield and the unreacted mesoporphyrin and porphyrinone 5 could always be recovered for recycling. The intermediacy of porphyrinone 5 seems to be necessary. Attempts to react zinc mesoporphyrin directly with an excess of 0504 have only resulted in intractable pigments. The two-stage oxidation via isolated porpyrinone has also imparted a high degree of regioselectivity for the isobacteriochlorin-type 1,3-dione formation. In the present case, if the isomeric porphyrinone 38 is used, the major 36 39 M00,C MOO,C Mame “_— CO,M¢ 60.". 60,". 33 3% B‘E Scheme 6 Home ".0 ,C MoO,C l C0,". C0,". cow. cot". 35 37 4O 41 34 product is 41, despite the steric advantage for osmic attack at ring A. In comparison with mesoporphyrin , the osmate selectivity of northern ring A and B versus southern ring C and D is about 4 to 1. The pinacolic rearrangement product from 40 is exclusively 3,5-porphyrindione 41, apparently reversing the migratory aptitude of methyl < propionate observed in the case of simple Vic-dihydroxychlorin but fully agreeing with the above observation with porphyrindione 8, that the formation of the 1,3-dione is preferred to its 1,4-dione regioisomer. This observation has a strong influence on our later synthetic strategies to the Q] structure 4, since we recognize that the best way to form 1,3-porphyrindione is to take advantage of both tendencies: the migratory aptitude of the substituents involved in the pinacolic rearrangement and the preferential formation of 1,3-porphyrindione from isobacteriochlorin-type Vic-dihydroxyporphyrinone, as illustrated in the formation of 6 here. However if the migratory aptitudes of the substituents involved are not in agreement with the rearrangement direction, the 1,3-dione formation is still highly possible, as in the case of dione 41, since formation of the 1,4-dione is not detected. This strategy is best utilized later in the synthesis of the heme d1 analogue 130. B. Formation of the Acrylic Side Chain There has been no precedence example of converting a porphyrin propionate side chain directly to an acrylic functionality. Acrylic porphyrins are usually synthesized by applying Kenovenagel-type condensation or Wittig reaction to corresponding formylporphyrins.75 The necessary formylporphyrins can be obtained by electronic substitution of peripherally unsubstituted hemes with dichloromethyl ether,76 or by degradation of vinylporphyrins with oxidants.77 Owing to the inaccessibility of peripherally 35 unsubstituted porphyrins, especially those with the position of formyl group specified, a multistep total synthesis has often become an unavoidable choice.78 Originally, a total synthesis pathway was contemplated to approach the model compound 10 (Scheme 7). The starting porphyrin 45 was to be assembled by a stepwise method from pyrrole 43, 44 and dipyrrylmethene 42. The dichloromethyl ether reaction would bring 45 to its formyl derivative 46, which would then be converted to a 1,3-porphyrindione 47, further to model compound 10. Since the electron-withdrawing -CHO was expected to retard the hydroxylation and lead to side reactions, we planned, as a backup, to replace the -CH0 with a CHZCHCICOZMe group. After the formation of the 1,3-dione core structure, a strong base, such as DBU [1,8-diazabicyclo(5,4,0) undec-7ene] might generate the acrylic acid. This scheme is obviously tedious and requires a substantial amount of time and effort to accomplish. After several attempts aiming at the synthesis of porphyrin 45 without yielding satisfactory results, we turned to another direction seeking a way to form acrylic side chain directly from the propionic side chain on a 1,3-porphyrindione. As shown in Scheme 8, a reaction of the free base of 1,3-dione 8 with 1.5 equivalent of 0504 gave surprisingly only one single compound whose structure was determined by 1H- NMR and NOE measurements to be diol 48. Presumably for the same reason as mentioned for the diol 30 formation, the ring D is less favored toward osmic attack due to the electron withdrawing effect of the keto group adjacent to it. This result is contradictory to the structure presumed by Inhoffen and Nolte who thought ring D diol 49 is the product. When diol 48 was treated with sulfuric acid, a triketone 50 was obtained. However, if heated in dilute hydrochloric acid, another chain of 36 \ \ \ 42 \ H~- Br’ COt-Bu "N CHO \ \ 43 H H M00,C 45 II o o CHO CHO 47 max: I Moo,c 46 o O 10 Home com. Scheme 7 37 50 O O OH 52 51 O O \ Scheme 8 38 events took place; the diol 48 underwent elimination to generate the pyrrole double bond with the concomitant formation of an alcohol at the OL-position of one of its ethyl groups on the ring. Possibly, the reaction went through an intermediate stage of exocyclic alkenes prior to dehydration (Scheme 2), a pathway perhaps similar to the suggested mechanism in Inhoffen's chlorOphyll p synthesis.79 Two alcohols were isolated and identified as 51 and 52 with similar yields. Compound 52 was found to be quite inert toward dehydration and remained unchanged even after a lengthy reaction time, but to form readily its methoxy derivative 54 in the presence of methanol and catalytic amount of acid or on TLC plates. In contrast, compound 51 underwent smoothly dehydration to give a vinyl porphyrindione 53. When 1,3-mesoporphyrindione 6 was treated with 0504, the dihydroxylation occurred overwhelmingly at the ring C to give diol 55. Heating this compound in diluted hydrochloric acid resulted in almost quantitative formation of model compound 10 with the acrylic double bond formed at the ring C propionate side chain (Scheme 10). Even there were two alcohol intermediates possible, the (B-hydroxylpropionate derivative 56 predominated. The formation of an acrylic side chain conjugating with the aromatic macrocycle seems to be thermodynamically favored so that it drives the reaction to 10 without stopping at 56 and this must have depressed the formation of hydroxymethyl isomer 57. C. Spectroscopic Studies of Model Compounds The UV-visible spectra of model compound 8, 53, and 10 are compared in Fi ure . It is obvious that the absorbance maxima are red shifted, especially in the visible region. There are about 10 nm shift toward red for 39 s? *6 €01". €03". C0251} N I —-> COgMe COgMO C03“. Scheme 9 O O o 0.. ———> OH OH M O O 2C 6 :02". MOOIC 55 CO2". MOO: C \COQMO . . o O OH HO MOOzc cat”. ".01: C03". 56 57 Scheme 10 4O I I I I I I I l I 440 _ ” 421 d , o _ O c a _ .n I. o _ 0 a d < _ o.oo Ann 400 soo 600 700 800 ' Figure 6 UV-vis absorption spectra of 1,3-dione 8, 53 and 10 in CHZClz. 41 all the dione absorption bands with the addition of a vinyl group and more than 20 nm shift with an additional carbonyl group. Model compound 10 has a visible spectrum virtually indistinguishable from that of Q] (Figure 3a) arguing that these two compounds possess the same fl-system. As we expected that the influence of the electronegative acrylic acid is both prominent and specific, i. e., the correct spectrum can only be obtained by an unique arrangement of the acrylic auxochrome in relation to the two keto groups on the ring, we were lucky to have the correct regioisomer 10 with the acrylic acid on ring C at the first try. A regio-isomer with the acrylic group on ring D was not obtained until we had achieved the total synthesis of d1. The spectra of 10 in the forms of its copper complex and acid salt are given in Figpre 2, they are nearly identical to those of the natural d1 (Figpre 3g and fig). To scrutinize further, iron complex was prepared for 10 and three representative spectra (hemin chloride, alkaline ferriheme and pyridine hemochrome) were taken as shown in Figure 8. These spectra were compared with literature spectra of Q] (Figure 1, 2 and 3 of Yamanaka and Okunuki114 and Figure 3 and 5c of Walsh et a1115). The similarity between $11 and model compound 10, again, borders on the identical. It is noticed that the triply split Soret band of the pyridine hemochrome has not been observed before with any porphyrin and chlorin hemes. A comparison of the 1H-NMR spectra of 10 and those of heme d1 also shows a similarity that is striking for molecules of such complexity (flgyr_e 9). One of the main reasons of Timkovich's original interpretation of the Q] data went astray was the observation of g1 meso resonances as grouped into two distinct pairs. This is almost universally observed for chlorin model compounds with the upfield pair representing the two meso protons adjacent to the sole saturated pyrrole. In most isobacteriochlorin, such as 42 T F I r l I I I I 3 . A . 80" 437 " 60l- .. 409 b d 644 40)- 4 20" 590 ‘ b q enj- m 445 B . 646 1 500 600 700 nm Figure 7 UV-vis absorption spectra of model compound 10 (A) in the form of Cu(II) chelate in CHC13 and (B) as the protonated form in formic acid. 43 I I ‘soT I I f I I I A Ferriheme chloride .. 0.6 l' " i- -I 004 "' d o 2 P 530 577 BIO ‘ O O 1 L L L 1 L L I 1 c . g r B "' Alkaline ferriheme . '6 a I- .n a .. 0.4 L- 0.2 '- C 426 0'8 ' ‘0‘ Pyridine hemochrome 006 " .l Oe4 " " 0.2 " d l- u l l g l g l l l L 400 500 600 700 mu 800 Figure 8 UV-vis absorption spectra of 10, (A) ferriheme chloride in acetone with a trace amount of HCl; (B) alkaline ferriheme in CHZClzlacetone containing tetrabutylammonium hydroxide; (C) pyridine hemochrome in CH2C12/pyridine. ’.79/ 1.90 1.92 0.68 , . 0.50 o .3.41. 2.57 0.6] ' 0.65 2.53 , 1.31 0 (.075) 183 (8.26) 8.26 9.13 (3.22) (3.31)3.32 3.26(3.24) (5.30) 6.91 (5.37) 8.96 / 6.07 (4.37) 3.07 (3.35) MOOzc COZMO 6.01 3.65 (8.00) {3.52) LIL TlefirITfilIfijTYlTT 3.9 8.. 7.32 XI UU “LU fijlfiTT—TIWWTjTjjTfi erTT’T ‘LB 3.IZ.I1.D Figure 9 250 MHz 1H-NMR spectrum of model compound 10 in CDCI3 (The bifuration of the peaks of the pyrroline ethyl and methyl groups is an indication of the presence of diastereomers). The numbers in parenthesis are the chemical shifts of natural d1. 45 sirohydrochlorin,71 the usual pattern is one meso proton down-field (the one between the unsaturated pyrroles), a pair at the intermediate shift (the two between a saturated and an unsaturated pyrrole), and one relatively upfield (the one between the two saturated pyrroles). In 1,3-porphyrindiones, the deshielding effects of the carbonyl oxygens have distorted the usual isobacteriochlorin pattern into an apparent chlorin pattern. The 3-carbonyl oxygen has deshielded OL—proton from its furthest upfield normal position to a shift comparable to p-proton, while the 1-carbonyl has deshielded 6-proton to a range comparable to that of y. The 1H-NMR of model compound 10 has also clarified the question of chemical shift for the NH resonances of g1. In 10, they appear at 0.97 ppm, which is unusually more downfield for the inner NH protons of porphyrin, but presumably reflects the decreased ring current and electron-withdrawing effect of the carbonyl groups on the ring. The corresponding chemical shift of d] around 0.9 ppm had been dismissed as residue impurity. The 13C-NMR of model compound 10 is shown in Figure 19. The resonances are assigned based on analogies to known models64r 63 in Table; and contrasted with the spectra of its precursor compound 6 and that of natural Q]. It has been pointed out that the meso carbons are sensitive to the level of the saturation of the porphyrin core as well as precise substituents.64 The highly similar pattern of the individual meso carbon resonances between model compound 10 and $11 is the evidence for the common core structure. Furthermore, the resonances at 207.2 and 208.7 ppm of 10, and 209.1 and 209.3 of 6 are assigned to the carbonyls which were not observed in the original d1 spectrum. The similarity between the Resonance Raman spectra of model compound 10 in the form of its Cu(II) complex and those of Cu(II) d1 was also 46 .H ozah 5 02» v.8 angmmmmn xoom .Ue an an 3 “£3928 336 we 522-02 pea—38%.:28m 653685 .8595? 3.532 3 853.4 a: am 3... Sr am 3; n: 3: am an: a: PD» — >.vab._bb.bb—rr>r_lPubLbhurPLubl—rPLVIP—Ih u p u _ r oiPr—.rP>iLv_>buP-.—FP-U.V.—D PP. Fri? Db-h.P.—~.Pb u — ..PF_— —D [b— rrrL.rrt—P*F—rtnrr—E :13 is}: is} EN. SN- 8.._ Sm~ 8m— SPF Em_ Efi_ SEN G—N SNN .r .LELI: . _ EIFLctfz .E _-E.-E_rt-._ Etc: _E ._E_ EerELFLIFLrtr if; Table 1. 13C-NMR chemical shiftsa of 1,3-mesoporp compound 10 in comparison with that of natural heme d1 hyrindione 6, model tetramethyl ester. compound 5(ppm) carbon dione 6 model 10 heme 511 a and b 131.55 13222 134.7, 137.1 (pyrrole) (131.98, 13203)b (133.87, 133.93) (132.24, 132.32) 134.90 135.69, 136.74 (136.07, 136.19) 136.92, 137.10 136.84, 137.46 139.52 139.72 (144.16, 14424) (14219, 14234) (147.45, 147.59) (151.66, 151.78) (161.72, 161.87) (155.07, 155.19) 237.33% °°°°°°°°°°°°°° 53.11, 55.73 53.03, 55.42 gizgizgogm 36.19 36.24 36.66 21.18, 21.33 21.16 21.5 173.12 17283 51.62 51.73 8 """"""" -_ 173.12 17283 QH=§H 0913 12248 1229 143.83 51.91 51.9 01280013 41.7, 42.5 521 61330-13 ......................... 8286 ................................. 5 73, 8.85 ................... CH3(pymlme) ............ 2251,2366 .............. 2229., 23 01 - 23.4, 24.1 EH3 (pyaol.‘>"““‘“ """16I71',‘11.66 10.72, 1281 11.2, 13.2 360 (ring) """" " 209115, 209.29 ' 2072520871 meso-H """ 91.31, 96.86," 90.03, 91.67“ 902, 91.9 99.01, 99.11 98.33 98.5, 102.9 (10288, 1029) aReferenced against the center of CHC13 (77.0 ppm). 5Fairs in parenthsises indicate bifurcation due to diastereomers. 48 observed and will be discussed in Chapter 5. III. EXPERIMENTAL General 1H— and 13C-NMR were obtained at 250 MHz on a Bruker WM-250 instrument. Spectra were mostly recorded in CDC13; the residue CHC13 was used as the internal standard set at 7.24 ppm. Nuclear Overhauser enhancement (NOE) was measured by difference between a spectrum with preirradiation on a target peak minus a spectrum with equivalent preirradiation at a dummy position. Magnitudes of NOEs were calculated as the area of the enhanced resonance in difference spectra divided by the area in the control spectrum with no enhancement. Mass spectra were obtained using a Finnigan 4021 GC-MS (direct insertion probe, 70ev, 200-300 °C), or a IEOL HX 110-HF spectrometer equipped with a fast atom bombardment (FAB) gun. A matrix of thioglycerol-dithioerythreitol-dithiothreitol, 2:1:1, containing 0.1% trifloroacetic acid was used for the FAB-MS. Visible absorption spectra (in CHzClz or CHC13) were measured with a Cary 219 or a Shimadzu 160 spectrophotometer. IR spectra were obtained from KBr pellets or NaCl films on a Nicolet IR/42 spectrometer. Melting points were obtained on an electrothermal melting point apparatus and are uncorrected. Preparative TLC plates were purchased from Analtech (silica gel G, 1000 or 1500 pm). A. Synthesis and oxidation of porphyrin tetramethyl acetate 5-Bromo-2,4'-di-(2-methoxycIabonylmethyl)-2'.4,5'-trimethyl-2,2'-dipyrryl- 49 methenium bromide (14) To a solution of 26.7 g (0.1 mol) of pyrrole 13 in 160 ml of AcOH, 15 ml (2.9 mol) of bromine in 50 ml of AcOH were added in 15 min with stirring. The mixture was stirred for another 1 h after the addition. Then the large part of the solvent was removed by vacuum and the concentrated solution was allowed to stand at room temperature until dipyrrylmethenium bromide (13) was crystallized. The product was collected by filtration and vacuum dried. yield, 15.0 g, 46%; mp 173-175 (dec.); lH-NMR 6 2.07, 2.37, 2.70 (3 H each, 8, Me), 3.48, 3.82 (2 H each, s, CH2COZ), 3.69, 3.71 (3 H each, s, COZCH3), 7.31 (1 H, s, methine), 7.95, 8.06 (1 H each, br s, NH); MS (direct probe 70 eV) m/ e 410 (M+). 2,4,6,8-tetramethyl-pogphyrin-1 ,3,5,7-tetramethyl acetate (12) The above dipyrrylmethenium 14, 13.0 g (0.02 mol), and 250 g of methylsuccinic acid were ground together into fine powder and dried under vacuum for 8 h. The mixture was then fused in an oil bath for 6 h at 130 ° C protecting from moisture. To the cooled black melt 150 ml of MeOH and 100 ml of CHC13 were added, dry HCl gas was then bubbled into the solution for 10 min. After standing for 4 h, the solution was diluted by addition of anther 200 ml of CHC13, washed first with saturated N aOAc aqueous solution (2 x 150 ml), then with water (2 x 100 ml), and dried over NaZSO4. The solvent was removed by vacuum and the residue was chromatographed on a silica gel column (60-250 mesh). The methylsuccinic acid was eluted out with MeOH / CH2C12 (1/100) and the porphyrin 12 was rinsed out with 5%MeOH/CH2C12. To achieve a better purification result the crude porphyrin was column chromatographed once more under the similar condition then crystalized from MeOH-CHZClZ. Yield, 0.75 g (11%); 1H-N MR 50 5 -3.7 (2H, 5, NH), 3.7 (12 H, s, CH3), 3.8 (12H, s, CHZCOZQH3), 5.1(8H, s, QILIZCOZCH3), 10.2 (4H, s, meso); UV-vis hmax (rel. int.) 401 nm (1.00), 499 (0.12), 533 (0.08), 569 (0.07), 623 (0.05); MS, found m/ e 655.7233 for (M+H)+, C36I-138N406 requires 655.7342. Tetramethyl-ZJB,5,7-tetgmethylporphyrinone-2,4,6,8-tetracetlte (15) and trimethyl-2-hydroxyl-1,3,5,7-tetramethyl-chlorin—1,2-(Y-lactone)-4,6,8- triacetate (16) HzOz-HZSO4 oxidation of porphyrin 12 was carried out in the same way as will be described later in the mesoporphyrin's reaction. Only less than 5% of 15 and small amount of y-lactone 16 were obtained and no other oxidation products were detected from the reaction system. (15) 1H-NMR 5 1.95 (3 H, s, Me saturated), 2.97 (3 H, s, CHzco;,_C_H_1 saturated) 3.50, 3.59, 3.62 (3 H each, ring Me), 3.75, 3.78, 3.82 (3 H each, CHZCOZC_H3), 3.89, 4.00 (1 H each, dd, J=17.5 Hz ngOz saturated), 4.98, 5.04 (2 H each, s, CHZCOZ), 5.01, 5.07 (1 H each, dd, I=17.5 Hz C_H2COZ), 9.10, 9.85, 9.90, 9.98 (1 H each, s, meso Q, 5, [3, y), -2.88, -2.83 (1 H each, br 5, NH); UV-vis hmax (rel. int.) 405 nm (1.00), 505 (0.08), 543 (0.07), 587 (0.05), 642 (0.18). (16)1H-NMR 5 2.37 (3 H, s, Me saturated), 3.41, 3.46, 3.57 (3 H each, s, ring Me), 3.70, 3.75, 3.81 (3 H each, s, COZQHB), 4.83, 4.86 (2 H each, s, CHZCOZ), 4.89, 4.99 (1 H each, dd, ]=17.6 Hz, gizCOz), 9.15, 9.17, 9.74, 9.87 (1 H each, s, meso 6,01,Y,p ), -2.85 (2 H, br 5, NH); UV-vis hmax (rel. int.) 392 nm (1.00), 495 (0.11), 540 (0.04), 585 (0.05), 640 (0.26). B. HzOz-H2804 oxidation of mesoporphyrin \ 51 Mesoporphyrin IX dimethyl ester (350 mg) was dissolved in concentrated sulfuric acid (30 ml d=1.84) in an ice bath. To this solution under stirring was added 6% H202 (2 m1) dropwise such that the temperature of the reaction mixture was kept below 10 °C. After the addition was complete (15 min), the dark red solution was stirred an additional 10 min in an ice bath and then at room temperature for 25 min or until the solution became dark green. The reaction was quenched by pouring the mixture into a large beaker containing N aOAc (20 g) and crushed ice. After standing at room temperature for 2 h, the solids were collected by filtration, washed with water, and dried (ca. 200 mg). The filtrate was concentrated in vacuo and mixed with acidified methanol (200 ml of MeOH+2 ml of H2504) and chloroform (100 ml) to effect esterification. This mixture, after washing with water, afforded about 150 mg of solid material containing small amounts of unreacted mesoporphyrin dimethyl ester, together with other intractable oxidation products. Therefore, in later reaction runs, the acid filtrate was discarded. The solid oxidation product was chromatographed on a 2 x 10 in. silica gel column eluted with methanol / methylene chloride (2/ 98). A fairly pure porphyrinone 18 was obtained from first 20-30 m1 eluent, and the rest of pigments were collected into three fractions. Each fraction was concentrated and chromatographed again on preparative TLC plates (CHZCI2 with 1% MeOH) to give eight additional compounds. Alternatively, as a group, the diones can be cleanly separated from monoketones on a silica gel column by using methylene chloride containing 5% of formic acid as eluent. Subsequent separations can be carried out on preparative TLC plates. The nine keto products are tabulated roughly according to their Rf values are given in Scheme 4. 3-Mesopogphyrinone dimethyl ester (18) Yield, 29.5 mg (8.2%); 1H-NMR 52 5 0.41 (3 H, t, CHzgfl3 saturated), 1.82 (3 H, t, CH2C_I-I_3), 2.07 (3 H, s, Me saturated), 2.77 (2 H, q, CfizCH3 saturated), 3.26 (4 H, m, CH2C_H2CO?_), 3.48, 3.55, 3.58 (3 H each, s, Me), 3.68 (6 H, s, COZCH3), 4.02 (2H, q, C_HZCH3), 4. 25, 4. 40 (2 H each, q, C__HZCHZCOZ), 9.14, 9.86, 9.88, 9.90 (1 H each, s, meso 0, Q, 5 and y), -2.96 (2 H, br 5, NH); UV-vis hmax (EM) 642 nm (34600), 585 (5800), 547 (12400), 508 (8800), 407 (165200); MS found m/ e 611.3242 for (M+H)+, C36H43N4OS requires m / e 611.3236. l-Mesoporphyrinone dimethyl ester (5) Yield, 35.6 mg (9.9%); 1H-N MR 5 0.41 (3 H, t, CH2_C_H3 saturated), 1.80 (3 H, s, CHZQIgl_3), 2.06 (3 H, s, Me saturated), 2. 75 (2 H, q, C_H2CH3) saturated), 3. 21 (4 H, m, CH2C__H2COZ), 3. 45, 3.56, 3.59 (3 H each, s, Me), 3.62, 3.65 (3 H each, s, COZCH3), 4.01 (2 H, q, _CHZCH3), 4.22, 4.38 (2 H each, q, _C_H,_CH2C02), 9.10, 9.80, 9.82, 9.92 (1 H each, s, meso on, 5, [3, and y), -2.97, -2.81 (1 H each, br 5, NH); UV-vis hmax (EM) 642 11111 (33300), 585 (6000), 547 (12000), 508 (10000), 407 (175000); MS (direct probe, 70 eV), m/ e 610 (M+). 1,3-Mesoporphyrindione dimethyl ester (6) Yield, 16.8 mg (4.5%); 1H-NMR 5 0.50, 0.70 (3 H each, s, Me saturated), 2.62 (4 H, m, CHZCH3saturated), 3.11 (4 H, m, CHZQ-IZCOZ), 3.27, 3.32 (3H each, ring Me), 3.60, 3.63 (3 H each, s, COZCH3), 4.16 (4H, m C_HZCHZCOZ), 8.42, 8.63, 9.28, 9.51 (1 H each, s, meso p, 01, 5, Y), -0.04 (2H, br 5, NH); UV-vis Amax (EM) 638 nm (16800), 592 (15300), 584 (15700), 544 (9600), 438 (97000), 417 (94000), 402 (74800); MS found m/ e 627.3178 for (M+H)+, C36H43N 406 requires m/ e 627.3185. 2,3-Mesoporphyrindione dimethyl ester (19) Yield, 7.5 mg (2%); 1H-NMR 5 0.55 (6H, t, CHZQH3 saturated), 1.95, 1.97 (3 H each, s, Me 53 saturated), 2.68 (4 H, q, CHZCH3 saturated), 3.22 (4 H, t, C_HZCHZCOZ), 3.46 (6 H, s, Me), 3.62 (3 H, s, COZQLI3), 4.37 (4H, t, CHZCHZCOZ), 8.90 (2H, s, meso 0 and 5), 9.74, 9.90 (1 H each, s, meso CL and y), -1.63 (2H, br 5, NH); UV-vis hmax (EM) 622 (18000), 592 (9400), 435 (102000), 417, (133000); MS found m/ e 627.3194 for (M+H)+, C36H43N406 requires m / e 627.3185. 6-Mesoporph3ginone dimethyl ester (20) Yield, 4.5 mg (1.2%); 1H-N MR 5 1.53, 2.10 (1 H each, m, .C_H2CH2C02 saturated), 1.79, 1.80 (3 H each, t, CH2_C_H3), 2.10 (3H, 8, Me), 3.09 (2H, t, CHZQIiZCOz saturated), 3.22 (2 H, t, CH2C_}12C02), 3.24, 3.44, 3.60 (3 H each, 5, ring Me), 3.62, 3.64 (3 H each, s, COZCH3), 3.90, 4.04 (2 H each, q, C_flzCHg), 4.34 (2H, t, C_HZCHZCOZ),9.17, 9.82, 9.86, 9.90 (1 H each, s, meso p, y, 5 and 01), -2.92 (2 H, br s, NH); UV-vis hmax (EM) 642 nm (33300), 585 (6000), 547 (12000), 508 (10000), 407 (175000); MS found m/ e 611.3245 for (M+H)"'. C36H43N4OS requires m/ e 611.3236. 7-Mesoporphyrinone dimethyl ester (21) Yield, 4.5 mg (1.2%); 1H-NMR 5 1.50, 2.08 (1 H each, m, C_HZCHZCOZ saturated), 1.78, 1.80 (3 H each, t, CHZQH3), 2.10 (3 H, 5, Me saturated), 3.08 (2 H, t, CH2C_H2C02 saturated), 3.23 (2 H, t, CHZQHZCOZ), 3.25, 3.43, 3.59 (3 H each, 5, ring Me), 3.60, 3.70 (3 H each, s, cozglia), 3.87, 4.02 (2 H each, t, _C_H2CH3) 4.34 (2 H, t, QHZCHZCOZ), 9.12, 9.78, 9.80, 9.90 (1 H each, s, meso 5, y, 01 and p), -2.95, -2.84 (1 H each, br 8, NH); UV-vis hmax (EM) 642 nm (33000), 585 (5900), 547 (12000), 508 (9900), 407 (176000); MS (direct probe, 70 eV) 610 (M+). 3,7-Mesoporphyrindione dimethyl ester (22) Yield, 9.5 mg (2.6%); 1H-NMR 5 0.44 (3 H, t, CHZQ-j3 saturated),1.56, 2.15 (1 H each, m, C_H_2c:H2co2 saturated), 1.75 (3 H, t, CH2C_H_3), 1.99, 2.02 (3 H each, 8, Me saturated), 2.71, (2 54 H, q, C_I'I_2CH3), 3.02 (2 H, t, CPIZQHQCOZ samrated), 3.29 (2 H, t, CHzgizCOZ), 3.49, 3.50 (3 H, 8, ring Me), 3.51, 3.75 (3 H, s, COZCH3), 3.93 (2 H, t, C__I-IZCH3), 4.29 (2 H, tgflzCHzCOz), 9.05, 9.06, 9.66, 9.74 (1 H each, s, meso 5, 0, y, and 01), ~2.78, -2.74 (1 H each, 5, br 5, NH); UV-vis Kmax (EM) 685 nm (95000), 652 (7300), 622 (6800), 556 (10700), 514 (8100), 486 (5700), 411 (187000), 401 (164000); MS, found m/ e 627.3198 for (M+H)+, C36H43N406 requires m/ e 627.3185. L7-Mesoporphyrindione dimethyl ester (23) Yield, 7.0 mg (1.9%); 1H-NMR 5 0.61 (3 H, t, CH2C_H_3 saturated), 1.70 (3 H, t CH2Q13), 1.76, 2.15 (1 H each, m, C_H_2CH2C02 saturated), 1.95, 1.96 (3 H each, 5, Me saturated), 2.63 (2 H, q, C_HZCH3 saturated), 2.94 (2 H, t, CH2C_H_2COZ saturated), 3.13 (2 H, t CHZQIJZCOZ), 3.43, 3.44 (3 H each, 8, ring Me), 3.46, 3.72 (3 h each, s, conga), 3.75 (2 H, q, C_H_2CH3), 4.14 (2 H, t, QHZCHZCOZ), 8.57, 8.78, 9.32, 9.52 (1 H each, s, meso 01, 5, y, and p), -0.61 (2 H, br s, NH); UV-vis hmax (EM) 637 nm (15800), 592 (14400), 583 (15000), 544 (9100), 437 (92000), 417 (90000), 402 (72000); MS (direct probe 70 eV), m/ e 626 (M+). 1,8-Mesoporphyrindione dimethyl ester (24) Yield, 2.0 mg (0.5%); 1H-NMR 5 0.52, (3H, m, CHzg:_i_i3 saturated), 1.77 (3 H, t, CHZCHB), 1.82, 2.20 (1 H each, m, C_HZCHZCOZ saturated), 1.98, 2.01 (3 H each, s, Me saturated), 2.68 (2H, q, _C_H2CH3 saturated), 3.00 (2 H, t, CH2C_H2COZ saturated), 3.17 (2 H, t, CH2_C_H2COZ)3.34, 3.37 (3 H each, 5, ring Me), 3.57, 3.65 (3 H, s, COZQ-I_3), 3.91 (2 H, t, _CfiZCH3), 4.25 (2 H, t, C_HZCHzCOz), 8.93, 8.97, 9.66, 9.80 (1 H each, s, meso y, 0t, 5, and p); UV-vis hmax (EM) 623 nm (19000), 592 (5900), 436 (100000), 417 (135000); MS, found m/ e 627.3910 for (M+H)+, C36H43N 406 requires m/ e 627.3185. 55 C. Preparation of 1,3-porphyrindione from zinc porphyrinone” Typically, 1 mmol of porphyrinone was dissolved in 100 m1 of CHC13 and 50 ml of MeOH, and to this solution 2 ml of saturated Zn(OAc)2 methanol solution and a pinch of N aOAc were added. The solution was brought to refluxing and causing the brown-color to turn gradually into green. The completion of zinc insertion can be monitored by TLC or by UV-vis spectrum (the bands of the free base at 406, 546, and 508 nm should disappear). The excess of zinc acetate was then washed away with water (3 x 100 ml) and the solvent was evaporated. The residue was vacuum dried and redissolved in 100 ml of dry methylene chloride. To this solution 1 ml of dry pyridine and 380 mg (1.5 mmol) of osmium tetroxide in 0.38 ml of anhydrous ether was added, and the mixture was allowed to stir at room temperature, under nitrogen, in the dark for 20 h. It was then quenched with methanol (50 ml) and bubbled with H25 for 10 min to decompose the osmic ester. The precipitated osmium sulfide was removed by filtration, and the crude product in the filtrate was chromatographed on a silica gel column. Unreacted Zn(II) porphyrinone was eluted first with 1% MeOH/CHzClz, and the slower moving dihydroxy compounds were then rinsed down quickly by increasing the amount of methanol in the eluent. To effect pinacolic rearrangement, the diol containing fractions were brought to dryness and treated with ~10 m1 of sulfuric acid directly. The central zinc ion was replaced by protons during this time. After stirring for a few minutes at room temperature, the acid solution was carefully diluted with 100 ml of methanol in an ice bath with continual stirring. The solution was further diluted with 150 ml of CH2C12 and washed with saturated 56 N aOAc solution (3 x 100 ml), water (2 x 100 ml), and brought to dryness. The residue was chromatographed once again on preparative TLC (1~2% MeOH /CH2C12) to separate the major 1,3-dione from a small amount of 2,3-dione. In both OEP and mesoporphyrin systems the yield of dihydroxy compounds were usually higher than 60%, and the yield of 1,3-dione from diols through pinacolic rearrangement was above 80% while that of the 2,3-dione was less than 5%. *The above procedure can only be applied to porphyrinones with alkyl side chains or other substituents with the similar electronic property. When electron-withdrawing substituents are involved stronger acidic condition has to be used to effect the rearrangement and more isomeric products are formed (to be discussed in chapter 3). D. Acrylate side chain formation 2,2,4,4,5,7,8-heptaethyl-6-vinyl-1,3-porphyrindione (53) Osmium tetroxide (384 mg, 1.5 mmol) in 3.84 ml of ether and 1.5 ml of pyridine were added to a solution of 566 mg (1 mmol) of dione 7 in 150 ml of CH2C12, and the reaction was allowed to proceed, under argon, in the dark, at room temperature for 24 h. The solution was then treated with 50 ml of methanol and bubbled with H25 for 10 min. The precipitated osmium sulfide was removed by filtration on celite and the crude product was dried by vacuum before chromatographed on a silica gel column. Unreacted green dione 8 (55%) was eluted first with 1% MeOH/CHZCIZ and the dark-grey dihydroxy compound 48 was eluted off with 5% MeOH/CH2C12; yield, 222 57 mg, 37%; 1H-NMR 5 0.41, 0.44, 0.56, 0.64, 0.71, (3 H each, t, CHZCH3 sat.), 1.41 (3 H, s, CHzg-ls sat.), 1.47, 1.48 (3 H each, t, CH2_C_I_-I3), 2.21, 2.45 (6 H each, m, _C_H2CH3 sat.), 3.35 (4 H, m, _C_H2CH3), 3.81, 4.09 (1 H each, br s. OH), 7.12, 7.57, 7.68, 8.48 (1 H each, s, (3, 01, y, 5); UV-vis hmax (EM) 669 run (8800), 611 (6700), 547 (8200), 516 (6400), 417 (38500), 385 (48800), 368 (39700). Diol 48 (222 mg, 0.37 mmol) was dissolved in 50 ml of dioxane with 3 ml of diluted hydrochloric acid (5%). The mixture was heated on a steam bath until the dark-grey solution turned into a bright-green color. The reaction was allowed to go for another 10 min before cooled down and diluted with 100 ml of CH2C12. The organic solution was washed twice with water and evaporated to dryness. Separation on TLC plate (1% MeOH/CHzClz) gave the fast moving band of vinyl dione 53 followed by the two Q-hydroxyethyl dione 51 and 52. An OL-methoxyethyl dione 54 was also separated, which was found derived from 52 on the plate in the presence of methanol and can be easily made by treating 52 with acidified methanol. The structures of above products were confirmed by NOE. (51) Yield, 34 mg, 16%; 1H-NMR 5 0.35 (3 H, t , CH2C_H3 sat.), 0.97 (9 H, m, CHZQB sat.), 1.39 (9 H, m CH2C_H3) 2.29 (3 H, d, CHOHC_H_3), 2.84 (8 H, m, _CflzCH3), 3.40 (6 H, m, CLIZCH3), 6.34 (1 H, q, QHOHCH3), 8.25, 8.50, 9.02, 9.92 (1 H, each, s, meso, p, 01, 5, Y); UV-vis hmax (rel. int.) 401 (0.82), 418 (1.00), 441 (0.93), 549 (0.17), 591 (0.28), 640 (0.25). (52) Yield, 67 mg, 31%; lH-NMR 5 0.38, 0.43 (3 H each, t, CH2c_H_3 sat.), 0.55 (6 H, t, CH2c_H_3 saturated), 1.67 (9 H, m, CHZCH3), 2.11 (3 H, d, CHOHQfla), 2.55 (8 H, m, CHZCH3), 3.73 (6 H, m, CHZCH3), 6.23 (1 H, q, QIOHCH3), 8.30, 8.47, 9.18, 9.96 (1 H each, s, meso (3, Q, 5, Y); UV-vis hmax (rel. int.) 401 (0.83), 419 (1.00), 439 (0.94), 548 (0.18), 587 (0.30), 640 (0.27). 58 (53) Yield, 79 mg, 38%; 1H-NMR 5 0.42, 0.55 (6 H each, t, CH29H3 sat.), 1.67 (9 H, m, CHZCH3), 2.54 (8 H, m Q_H_2CH3 sat.), 3.71 (6H, m, C_HZCH3), 6.09 (2 H, dd, CH=Q_Hz), 7.85 (1 H, dd, CI_-I_=CH2), 8.34, 8.46, 9.17, 9.49 (1 H each, s, meso [5, On, 5, y), 1.50 (2 H, br s, NH); UV-vis Kmax (rel. int.) 422 nm (1.00), 444 (0.82), 554 (0.17), 596 (0.28), 646 (0.22). (54) Yield, 24 mg, 11%; 1H-NMR 5 0.41, 0.56 (6 H each, t, CH2C_H3 sat.), 1.67 (9 H, m, CHZCH3), 2.09 (3 H, d, CH(OCH3)§_H3), 2.57 (8 H, m QHZCH3), 3.50 (3 H, s. og_l3), 3.73, (6 H, m, C_HZCH3), 5.61 (1 H, d, C_H_OHCH3), 8.33, 8.49, 9.21, 10.02 ( 1 H each, s, meso p, on, 5, y); UV-vis Kmax (rel. int.) 4.02 (0.80), 4.19 (1.00), 4.39 (0.87), 5.46 (0.17), 5.88 (0.25), 6.39 (0.21). Dimethyl-l ,3-mesoporphy_1;indione~6—acrylate-7-propiona_t§ (10) The 1,3-mesoporphyrindione (6) was treated with OsO4 in the same way as described above to effect the dihydoxylation. Typically, 100 mg (147 mmol) of 5 gave 42 mg (59 mmol) of 5,6—dihydroxyl-1,3—mesoporphyrindione 55. The dehydration was accomplished smoothly by heating 55 in 10 ml of dioxane with 2.5 m1 of 5% HQ (or alternatively, refluxing 55 in 50 ml of benzene with a few drops of concentrated hydrochloric acid). After the distinctive color change from grey to bright-green color change had taken place, 1 ml of concentrated sulfuric acid in 10 ml of MeOH was added to the reaction solution. The mixture was heated for another 5 min on steam bath and then set aside for 4 h to ensure esterification of propionic side chain. The solution containing crude product was combined with 50 ml of CH2C12 and washed with water and vacuum dried. The crude product was further purified on a TLC plate (1%MeOH/CH2C12) to give the green pigment 10, 32 mg. 59 (55) Yield, 40%; 1H-NMR 5 0.45, 0.64 (3 H each, t, CH2C_H3 saturated), 1.52, 1.57, 1.68, (3 H each, s, Me saturated), 2.22 (4 H, m, C_HZCH3), 2.40 (2 H, m, QIQCHZCO m saturated), 2.79 (2 H, t, QHZCHZCOZ), 2.85 (3 H, 5, ring Me), 2.92 (2 H, m, CH2C_H2C02 sat.), 3.52 (2 H, t, CHZCHZCOZ), 3.62, 3,74 (3 H, s, C02C_H3), 3.98 (2 H, t, QHZCHZCOZ), 7.10, 7.56, 7.69, 8.44 (1 H each, s, meso p, on, Y, 5), 3.55, 3.82 (1 H each, br s, OH); UV-vis Kmax (rel. int.) 368 nm (0.82), 383 (1.00), 412 (0.87), 507 (0.14), 540 (0.18), 604 (0.16), 659 (0.21). (10) Yield, 80%; 1H-NMR 5 0.48, 0.50 (3 H, t, CHZQI-_‘I3 sat.), 0.63, 0.64 (3 H, t, CH2C_H3 sat.), 1.81, 1.83 (3 H, s Me sat.), 1.90, 1.12 (3 H, 8, Me sat.), 2.55 (4 H, m, _C_H_2CH3), 3.07 (2 H, t, CHng—gcoz)r 3.24, 3.32 (3 H each, s, ring Me), 3.65 (3 H, s, CHZCH2C02_C_H3), 4.01 (3 H, s, CH=CHCOCH3), 4.07 (2 H, t, C_HZCH2C02), 6.91, 8.96 (1 H each, d, gl-I_=C_H_C02), 8.26 (1 H, s, meso (3), 8.39, 8.40 (1 H, s, meso Cl), 9.13 (1 H, s, meso 5), 9.40 (1 H, s, meso v), 0.97 (2 H, br s, NH); UV-vis Kmax (rel. int.) 423 nm (1.00), 446 (0.71), 568 (0.34), 611 (0.32), 661 (0.21); MS found m/ e 625.2946 for (M+H)+, C46H41N406 requires 625.3030. CHAPTER 3 TOTAL SYNTHESIS OF HEME g1 I. RATION ALE OF THE SYNTHETIC STRATEGY At first, it might seem easy to achieve the total synthesis of d1 as the major synthetic hurdles, e.g., the formation of 1,3-dione core structure and the introduction of acrylic double bond, have already been overcome. All that left was to replace the ethyl side chains in model compound 10 by acetic acid groups. In principle, a double OsO4 oxidation-pinacolic rearrangement of a porphyrin diacetate like 58 would furnish the core structure 59 of heme M.°2c C01". M0020 0 C01”. 0 CO M CO M MIO,C 58 2 Q M.°2c 59 2 9 Cl Moo,c CO¢MO C Cl Cl Moo,c 60 C0151. MOO,C 6‘1 CO¢MQ Moo,c 62 com. d1. However, in view of the difficulties experienced in the synthesis of model compound 9, where electron-withdrawing acetate side chains were 60 61 involved, we expected that the £14 synthesis may not be so simple. Nevertheless, we chose to make a symmetric porphyrin such as 60 first to study the property of porphyrin bearing two acetic side chains and to see if the desired 1,3-dione structure can be formed. To circumvent the difficulties associated with acetic side chains, porphyrins such as 61 and 62 with the masked acetic functions (in the form of chloroethyl side chains here) were also considered. If the pinacolic rearrangement pathway should fail, we also planned alternatively to examine the direct oxygenation of the unsubstituted p-positions on an isobacteriochlorin. We thought that an isobacteriochlorin like 63 would undergo oxygenation to yield the £11 core structure 59. Several syntheses of unsubstituted isobacteriochlorin had been reported by Battersby's group.30r 31 It is possible to follow their procedures to prepare a simple model compound such as 64 to see if this idea would work. If so, we would then incorporate the other substituents and prepare 63 by this approach. MoO,C com. II. FROM 1,4-PORPHYRIN DIACETATE -- A TEST The 1,4-porphyrin diacetate 60 was first synthesized in 36% yield by the 62 condensation of symmetric 5,5'-dimethyl-dipyrrylmethenium bromide 65 with 5,5'-dibromo-dipyrromethenium perbromide 66 in formic acid with one equivalent of bromine (Scheme 11). Compound 65 was easily prepared by the self-condensation of pyrrole 67 upon refluxing in a mixture of formic acid and hydrobromic acid for 2 h. Dipyrrylmethene 66 was best prepared via its benzyl ester precursor by catalytic hydrogenolysis and subsequent bromination.82 MQO’C COIM. / / \ \ MOO,C COM: H HN‘ 65 Br ‘ ".0 ,C 6015‘ H 67 66 M0036 60 CO, M9 Scheme 11 Reaction of porphyrin 60 with 1.2 equivalent of 0504 in CH2C12, followed with H25 effected dihydroxylation at two possible positions to give northern diol 67 (13%) and southern diol 68 (48%). The electron -withdrawing effect of the acetate side chains may have rendered the osmic attack at the northern pyrroles unfavorable. The two diols were separated on preparative TLC plate and 67 was then treated with concentrated sulfuric acid. While porphyrinone 69 was isolated as a minor product, a green colored major fraction was identified as the y-lactone 70. This compound was found resistant to pinacolic rearrangement in acid and to hydrolysis in basic medium. Another minor product 71 was found with a hydroxymethyl substituent which might be formed by a similar dehydration-hydration 63 mechanism as described in Chapter 2. Further reaction of the zinc complex of 69 with 0504 and followed with sulfuric acid yielded predominately the 3,5-porphyrindione 73 derived obviously from ring-C diol 72 with the migration of methyl group. No trace amount of 1,3-dione 59 was detected. Some starting porphyrinone 69 as well as a hydroxymethyl porphyrinone 74 were also isolated from the reaction mixture W- The result of the reaction of porphyrin 60 in HZOZ-HZSO4 was messy and gave no indication of the formation of any porphyrindiones and a prolonged reaction only resulted in decomposition. Apparently the strong electron-withdrawing effect of the two acetic groups, plus the "wrong arrangement" of the substituents at the northern pyrroles have prevented the formation of the desired 1,3-dione 59 from this porphyrin. III. FROM PORPHYRIN WITH MASKED ACETIC SIDE CHAINS-A BYPASS A. 1 ,4-Bis-( 2-chloroethyl z-porphyrin Porphyrin 61, chosen for reasons of convenient preparation and separation of its isomeric products in the later reaction steps, was synthesized by a pathway similar to that used in preparing 60. This porphyrin has been made previously by Smith83 and Battersby's group84 using different approaches. The 2—ethyl acetate side chains of pyrrole 67 which was synthesized as described in experimental part, were selectively reduced with diborane to hydroxyethyl groups which were then reacted with thionyl chloride to form the 2-chloroethyl derivative 75. The bis-(2-chloroethy1)-dipyrrylmethenium bromide 76 was obtained in the same way as described above, and condensed similarly in a "2+2" pattern with 66 to generate 61 in a yield of 32% (Scheme Moo,c 130,1». M0026 co,m H 60 mote €02". MCO’C cot“. 68 Meme 0" 0 H C0,". 67 Home emu. Moo, OH cog» 69 71 Mg: :03“. MQO,C C0,". MeO,C €0,310 CO No n.0,: «1.0,: 0 ' + 72 74 Home mo,c com. Scheme 12 65 13). Cl Cl ‘3' cu I 66 / / \ _ wc015t H H L v H Br' 75 76 61 M.°2c COIMQ Scheme 13 The reaction of porphyrin 61 with osmium tetroxide, Scheme 14 produced the northern diol 77 (11%) and the southern diol 78 (31%) respectively. The acid catalyzed pinacol rearrangement of 77 was accomplished smoothly in sulfuric acid to give exclusively porphyrinone 79 with the migration of chloroethyl group which was verified by NOE measurement. OsO4 oxidation of the zinc complex of 79 gave mainly two vic-diols at its ring B (80) and ring D (81) in a ratio about 2 to 1 with a total yield of 25%. The mixture of this two diols were treated with H2504 directly without separation since they had been found not very stable in the ordinary working-up process. About 48% of the porphyrinone 79 was found regenerated from the rearrangement reaction mixture, and the other major products included 1,3-dione 82 (8%), 1,7-dione 83 (14%) and 1,8-dione 84(7%). Replacing the of acetate side chain by chloroethyl group did not seem to improved the electron-deficient problem significantly, as the yield of the northern diol 77 was still lower than that of the south. The surprising result was vic-diol 92 going back to porphyrinone 79 rather than giving the "normal" pinacolic product. Also the reaction between the zinc complex of porphyrinone 79 and OsO4 did not follow the empirical rules observed in the 66 6_1 H 0 H0 I C‘ CI HO H0 73 90.0,: 602'“ n.0,: com. CI ll c: M = 2H cu cu M = Zn F F — 85 0 co,» 01.0.1: 79 com. Home 81 mote C0,”. Home com. max: cot... Scheme 14 67 model compound study: not only ring B, but also ring D diol has been formed. However, the differential formation of isobacteriochlorin-type compound by zinc insertion still held true, there was no bacteriochlorin-type compound such as 1,6-dione 85 has been detected. The low yield of 1,3-porphyrindione 82 was probably resulted from the reluctant migration of the ring B methyl group of 80 since the methyl group has a lower mobility than that of the chloroethyl group. To achieve a higher yield of 1,3-dione 82 the positions of these two groups on ring B should be exchanged to suit their intrinsic migratory tendencies, i. e., a porphyrin like 62 is needed. B. 1,3-Bis-(2-chloroethyl)-porphyrin Porphyrin 62 was synthesized by derivatization of protoporphyrin IX according to literature method35 with modifications (Scheme 15). The two 0M0 / mo 86 87 Mco,c com. 1.1.0.0 COsM- Meme COtM- . H protoporphyrin IX ‘3' OH cu 88 62 Moo,c com. mo,c 602". Scheme 15 68 reactive vinyl groups of protoporphyrin were converted to the chloroethyl side chains by oxidation with T1(NO3)3 to the diacetal 86 and to dialdehyde 87, followed by reduction to 88 with NaBH4, and then by chlorination with PhCOCl-DMF to 62, each step giving essentially a quantitative yield. Osmium tetroxide oxidation of porphyrin 62 (Scheme m offered all four possible vic-diols; ring A 89 (6.8%), ring B 90 (6.9%), ring C 91 ( 22% ) and ring D 92 (26%). The two northern diols were separated from the two south and subjected to acid catalyzed pinacolic rearrangement directly without further separation. The two porphyrinone, 93 and 94, were obtained in very high yield (~90%). We learned from the 1,4-bis-(2-chloroethyl)-porphyrin pathway that for porphyrinone with such substituents, the osmic attack would occur on both pyrroles flanking to the keto group, so both zinc complexes 93 and 94 could offer 1,3-dione 82 even though the former would be the major producer. Thus OsO4 oxidation of the mixture 93 and 94 and rearrangement produced six porphyrindiones in addition to the regenerated starting porphyrinones 93 (16%) and 94 (17%). Among them, 1,3-porphyrindione 82 was obtained in highest yield (22%), others including 3,5—dione 95 (17%), 3,7—dione 96 (5%), 1,8-dione 97 (4%), 2,3-dione (3%) 98 and 1,7-dione 99 (2%). The structures of these compounds were identified by their visible spectra and the NOE measurements. According to the position of the second keto group introduced, we could deduce that dione 109 and 99 must be derived from porphyrinone 93, whereas 95, 96 and 98 from 94. The 3,7-porphyrindione 96 was the first bacteriochlorin-type product observed in the zinc porphyrinone oxidation. Although the percentage yield of 96 was not so significant, the appearance of this compound implied that less electron-rich substituents, even like the chloroethyl group, could render the regioselectivity of osmic attack toward zinc porphyrinone less specific. This 69 91 62 90 MOO,C CO,Me mote 60,900 3 Cl O, Cl 0 Cl 94 1000.6 1000.6 emu. CI 0 82 97 MOO,C C02". ‘ CO,Me M0013 C0,". c. 0 Cl c o 3' (:1 o 0 CI 95 mo,c o 96 98 "'0': C0,". cot". M.°:c C0,". Scheme 16 70 phenomenon has been seen at least in one other case (vide infra). C. Oxidation of 1,3-Porphyrindione Side Chains In order to transform the chloroethyl group to acetate side chain, porphyrinone 79 was first tested (Scheme 12). Heating of 79 in pyridine with KOH resulted in the formation of compound 100 with a hydroxyethyl group derived from the saturated ring A chloroethyl side chain by a nucleophilic substitution, and a vinyl group through the elimination of a HCl from the aromatic ring B chloroethyl side chain. The latter transformation was successfully used in the heme g1 prosthetic group synthesis to regenerate the vinyl groups from the protected chloroethyl forms.64 The alcohol side chain of 100 was converted to an aldehyde 101 by Swern oxidation86 in a good yield (>70%). To oxidize 101 further to acid, in its ester form 102, several reagents, such as argentic oxide (AgO)87 and pyridinium dichromate (PDC)88 have been tried. PDC oxidation gave 102 in a yield of 55% and the argentic method gave a slightly lower yield. PDC was also found to oxidize alcohol 100 in DMF directly to 102 after a lengthy reaction time. The vinyl group survived throughout these reactions. Unfortunately, this mild procedure turned out to be impractical when 1,3-dione 82 was applied. As shown in Scheme 18, reaction of this compound with KOH in pyridine produced the 2,4-bis-hydroxyethyl-1,3-porphyrindione 103 as a mixture of cis and trans isomers with poor yield (<45%). Attempt to oxidize 103 directly to the diacid of 59 with PDC ended up with a bleached solution containing unidentified species. Even the result of Swern oxidation was messy, only a poor yield of dialdehyde compound 104 could be detected. Oxidation by using AgO led also to miserable decomposition. 71 CI CI :l I I/ /, °:p| I I/ /, M'Ozc C0270! MCO: C Gag". M002C cot". 79 100 101 MQO,C . / 3 Scheme 17 MOO}: €02". 102 CI HO 0 CI 0 CH0 0 0::104; MeO,C 82 COzMe MeO,C com. M00, c CO,Me MIC O €0,019 0 X : Scheme 18 M0036 60,?“ 159 72 IV. FROM 1,3-PORPHYRIN DIACETATE -— REACHING THE GOAL Although the foregoing experiments, either from 1,4-porphyrin diacetate or from porphyrins with masked acetic functions, did not provide a practical synthesis of heme 311, they were educational. Having failed to bypass the "acetate problem", we had to press forward in the face of difficulties -- starting again from porphyrins with acetic side chains. Since we had already made porphyrinone diacetate 69 from 1,4-porphyrin diacetate 60 it should not be impossible to make 1,3-porphyrindione 59 if the migratory aptitude of all the substituents involved in pinacolic rearrangement are in tune with the migratory direction favoring the 1,3-dione formation. Porphyrin 58, which fits the above requirement was therefore constructed. The two acetate substituents of this porphyrin at the right positions are expected to direct the subsequent double migration of the adjacent methyl groups to generate the desired 1,3-dione nucleus. Owing to the reluctance of the pinacolic rearrangement caused by the two electron-withdrawing acetate groups, stronger acidic medium was considered to force the reaction. As shown in Scheme 12, the dialdehyde porphyrin 87, obtained previously, was oxidized with the Jones reagent and esterified in acidified methanol to provide porphyrin 58 in 92% yield. OsO4 attacked porphyrin 58 at all four pyrrole rings. After the separation of the unreacted 58 on a silica gel column, the two northern diols (105 and 106, 20%) were further separated from the two southern diols (107 and 108, 51%) on preparative TLC plates. The two southern diols were found unstable on TLC plates and could only be isolated as the spiro y-lactone derivatives 109 and 110. The southerndiols and their lactone derivatives could be "00,6 com. 87 one 1m 0 1 €0.94. "00,6 cot". "00,6 cot". "00.6 C0,". ' mo e H * com. .H 107 “ / on 108 ”00.6 60.". mtg co’”. com. me e H' ’ 0H , com. a" mo,e 1 com. CO."- 58 105 106 “00.6 CW“- M.o,e com. mop emu. «1.0,: 0 com. ‘ 111 . 112 3400.6 + '30,". “1.01;; C0,". CO,MQ 113 X MOOQC CO.MO Scheme 19 74 recycled, however, by reducing with a mixture of AcOH-HI—H3POZ to recover porphyrin 5863' 89 In order to achieve a higher yield for the pinacolic rearrangement of the northern diols, several strong acidic media were tested. We found that by using sulfuric acid alone, only less than 15% of porphyrinones 111 and 112 could be obtained with the major products being the inert y-lactone 113 and 114; using Nafion or Magic acid, diol decomposition was observed. The reaction condition was finally optimized by treating the diols with FSO3H-HZSO4-fuming H2504 (10:10z1), and porphyrinones 111 and 112 were obtained as the major products (45%) with the amount of lactones 113 and 114 reducing to about 10%. The separation of 111 and 112 was accomplished on TLC plates by "over-developing", and the ratio of these two porphyrinones was about one to one. It is necessary to purify porphyrinone 111 at this stage or the multiple isomeric products derived from 112 would surely add more difficulties to the already tedious separation in the next step. Porphyrinone 111 was metalated with zinc(II) acetate in CH3Cl-MeOH and the zinc complex was oxidized with 1.5 equivalent of OsO4. The oxidation products were quickly separated from the unreacted zinc porphyrinone on a silica gel column by flash chromatography under argon. Following the zinc removal by washing with aqueous HCl solution (10%), the purple colored porphyrinone diol 115 (18%), which was unstable on TLC plate, was separated from its ring C regioisomer 116 (15%) and ring D 117 (12%) on a chromatotron protecting from air and light. Rearrangement of 127 in concentrated sulfuric acid containing 10% of fluorosulfonic acid generated the desired 1,3-porphyrindione 59 (as the mixture of cis and trans diastereomers) in a modest yield of 12%. Other compounds isolated from the reaction mixture included the regenerated porphyrinone 111 (20%), a 75 y-lactone 118 (8%), a porphyrinone OL-hydroxyacetate 119 (15%) and an Q-hydroxymethyl porphyrinone 120 ( 4%) (Scheme 29). Formation of 119 and 120 was probably from a process analogous to that we encountered in the acrylate side chain formation (Chapter 2, B). As we mentioned earlier for zinc complexes bearing less electron-rich substituents, such as chloroethyl groups in compound 79, the osmic attack on the porphyrinone is less selective. In the present case, with the strong electron-withdrawing acetate side chains, the reaction is even less selective and the dihydroxylation occurred at all the three pyrrole rings without remarkable preferences. These results have therefore overridden the empirical rules observed in our model compound study where only simple alkyl substituents were involved. However, the advantage of using the zinc method is still obvious: without zinc insertion, the 0504 oxidation would overwhelmingly lead to the bacteriochlorin-type ring C diol 117, and the desired ring B diol 115 would not have been formed at all. The presence of acetate substituents further hindered the second pinacolic rearrangement on ring B of 115 as demonstrated by the low yield of 1,3-dione 59 even under the best condition tested. When the above FSO3H-H2SO4-fuming H2504 (10:10:1) mixture, which worked best in the first pinacol rearrangement, was applied here, the predominate product was the regenerated porphyrinone 111. By using sulfuric acid with only 10% of FSO3H, this "reduction" was largely prevented. The similar reduction has been observed in the rearrangement of porphyrinone diols bearing chloroethyl substituents (section 3). The precise mechanism for the reduction, like most alcohol reduction, is obscure. We believe that the diol may be reduced via the intermediate b-j (Scheme 21) which are favored by the electron-withdrawing acetate group as 76 MOOQC CO’M. . MQO’C cor“. HO OH HO OH MeO,C 302M. MeO,C CO,“ 116 117 mo,e com. M = 2H M = Zn 111 M.o,e com. MeO,C 0 C0,". MOO,C OH OH o 0... C0,". M.o,e 59 302". mo,e 1 15 60,“. n.0,: 1 18 com. n.0,e 0" h" com. MOO C C0,”. MOO C CO MO ' 120 ' 119 ’ Scheme 20 N~ OH CO, ”C Scheme 21 78 well as by a very electron-negative porphyrinone core structure. In strong acid, benzylic alcohols are known to undergo reduction by nucleophilic substitution.90 Porphyrinone diol 115 was unstable and difficult to purify. To avoid the unnecessary loss, the mixture of diols obtained from the 0504 oxidation of Zn(II) porphyrinone 111 was treated with HZSO4-FSO3H(10%) directly to effect the rearrangement. More than eight compounds were isolated in this manner and the structures and yields are given in Scheme 22. In addition to the three compounds derived from 115 as described above, 1,7-dione 121, 1,8-dione 122 and porphyrinone 6-acrylate 125 were derived from porphyrinone diol 116, and 1,5-dione 123 and 1,6-dione 124 were produced from diol 117. All these porphyrindiones were obtained as diastereomeric mixtures. The cis—1,3-dione 59a and its trans isomer 59b were separable by TLC plates, approximately in a ratio of 1.5 to 1. Their structure assignments will be described in next chapter. These compounds were found rather stable, even a prolonged developing on TLC plate in air did not result in noticeable decomposition or isomerization. The formation of acrylic double bond was accomplished by separately treating 59a and 59b with 1.5 equivalent of 0504 in methylene chloride to effect the dihydroxylation followed by reacting with a catalytical amount of HCl in boiling benzene (Scheme 23). In both cases, two regioisomers, with the acrylate substituent either on ring C or ring D were isolated in a ratio about 5 to 1. The formation of diols 127a and 127b, once again, demonstrated the reduced selectivity of the osmic attack duo to the acetate substituents. We use the prefix "iso-" in naming the ring-D acrylic compounds, thus there are cis-d1 (128a=4a) and trans-_d_1 (128b=4b), cis-iso-d1 (129a) and trans-iso-gl (129b). 79 "“323 com. M = 2H 111 M = Zn MeO,C CO2!“ MOOzc col”. 0 (17%) mo; 0 121 C03". n.0,e com- mo,e com. 0 CO¢MO \ mo,e cow. M.o,e M.o,e com. 123 (4%) 124 (9%) 125 (3%) 59 (6%) 118 (4%) 119 (7%) 120(2%) Scheme 22 MeO,C "Cote ".01: o €0,950 0 H o H o 126 MOO.C cot". M004: 0 60,!“ O M.°2c cot”. 128a (cis) 128b (trans) o C02Me 59a (cis) 59b (trans) com. \ MOOzc 0 cat". a on " 127 M0026 com. M.o,c 0 com. 0 6100.0 cot". 129a (cis) 129b (trans) Scheme 23 81 Among above four isomers only cis-d1 (128a) with the structure 4 as we proposed is truly identical with the natural d1 tetramethyl ester in a careful comparison of 1H and 13C NMR spectra, UV-Visible absorption, HPLC and TLC. V. A g1 ANALOGUE FROM COPROPORPHYRIN IV "001‘: O €0,940 0 M0036 CO¢MO 130 The most successful example of applying all the methodologies developed in the 511 synthesis was the preparation of homo-d1 130 in which two pyrroline propionate replaced the acetate side chains. Compound 130 may be used as substitutes and spectral probes for heme £11 in a variety of experiment conditions including protein reconstitution. The total yield of 130 from the same reaction pathway is significantly higher than that of d1 owing to the replacement of troublesome acetic side chains by the less electron-withdrawing propionic side chains. As shown in Scheme A, the symmetric coproporphyrin IV tetramethyl ester/‘starting compound, was conveniently synthesized through a condensation of dipyrrylmethenes 132 and 66 in a yield of 47%. Compound 132 was made by standard method from corresponding pyrrole 131in formic acid with hydrobromic acid. 82 Me? M'. PM. colt-Bu \ H H T H Br‘ MeP PM. 131 132 Scheme 24 The OsO4 oxidation generated two diol compound 133 and 134, with the northern isomer 133 being the favored product, which upon pinacolic rearrangement in sulfuric acid produced predominately porphyrinone 135. Further reaction of 135 with osmium tetraoxide after zinc insertion gave three diols with the ring B isomer 136 as the major product. Treating the mixture of diols with HZSO4-FSO3H (9:1) led to the formation of 1,3-dione 139 with the migration of methyl group (32%, yield), together with the regenerated porphyrinone 135 (20%) and other keto regioisomers (Scheme 25). The formation of acrylic side chain was accomplished as usual by further reaction with 0504, followed by dehydration in boiling benzene with a small amount of hydrochloric acid. The homo-_d_1130 was obtained almost quantitatively without any indication of the formation of its ring D regioisomer 144, however, a minor hydroxymethyl compound 145 was observed (Scheme 26). 83 PM. co.» Moo,e Scheme 25 139 C0194. ..... i 0 com. o 0."! \ / mme 145 email. mmc 130 emu. name 144 emu. Scheme 26 In conclusion, the successful synthesis of heme £11 and its stereo and regioisomers as well as its structural isomers not only established the Q] structure we proposed, but also provided us a reasonable supply of materials to study physical properties and functions of this green heme in biological systems. 85 VI. EXPERINIENTAL A. Porphyrin-lA-diacetate system Tetramethyl-2,3,5,8—tetramethyl-porphyrin-1,4-diacetate-5,6-dipropionate (60) 4,4'-di—ethoxycarbonylmethyl-3,3',5,5'-tetramethyl-2,2',-dipyrryl- methenium bromide 65, 12.9 g (0.051 mol), and 5,5'-dibromo-3,3'- di-methoxycarbonylmethyl-4,4'-dimethyl-2,2'-dipyrrylmethenium bromide 66, 29.1 g (0.05 mol), were suspended in formic acid (250 ml, 98-100%) and treated with 2.6 ml of bromine. The mixture was refluxed in an oil bath for 3 h and the solvent was then allowed to boil off over a period of 1 h with a stream of air. To the dried reaction residue were added 500 ml of methanol and 10 ml of concentrated sulfuric acid, followed by 40 ml of trimethyl orthoformate. After standing overnight, protected from moisture, the reaction mixture was diluted with 600 ml of CHzClz first and then 400 ml of concentrated aqueous N aOAc solution. The organic layer was separated, washed once again with 300 ml of N aOAc solution and then with water (3 x 300 ml). After evaporation of the solvent, the crude product was chromatographed on a silica gel column (50 to 250 mesh) using 1% MeOH/ CHzClz as eluent first then gradually increasing methanol to 5%. A dark non-fluorescent forerun was discarded and the moving of porphyrin band on column can be monitored by using an UV-lamp to ensure a complete collection. The porphyrin containing fractions were combined and brought to dryness in vacuo and then crystallized from CHZClz-MeOH. yield, 12.3 g, 36%; lH-NMR 5 3.25 (4 H, t, CH2C_I-12COZ), 3.44, 3.57 (6 H each, s, Me), 3.66 (6 H, s, CHZCHZCOZQH3), 3.73 (6 H, s, CH2C02_C_H3), 9.79 (1 H, s, meso Q), , 9.91 (2 H, s, meso p , 5), 9.96 (1 H, s, meso v), -4.06 (2 H, br s, NH); UV-vis Kmax (rel. int.) 399 (1.00), 498 (0.10), 531 (0.07), 568 (0.06), 622 (0.05); MS found ' 86 m/ e 683.7840 for (M+H)+, C38H43N408 requires 683.7884. Dihydroxychlorin 67 and 68 Osmium tetroxide (600 mg, 2.4 mmol) in anhydrous ether (6 ml) was added to a methylene chloride (200 ml) solution of porphyrin 60 (1.36 g, 2.0 mmol). Dry pyridine (1 ml) was added subsequently, and the mixture was allowed to stir at room temperature, under argon, in the dark for 20 h. The reaction was then quenched with 100 m1 of methanol and bubbled with H25 for 10 min. The precipitated black osmium sulfide was removed by filtration through celite, and the crude product in the filtrate was brought to dryness and chromatographed on a silica gel column. Unreacted porphyrin was eluted first with 1% MeOH/CHZCIZ. The faster moving isomer which turned out to be northern diol 67 was eluted then with 2% MeOH/CHZCIZ and the slower moving southern diol 68 was rinsed down with 4% MeOH/ CH2C12. Diol 67 was further purified on preparative TLC plates, developed with 5% EtOAc/CHzClz. Yield: unreacted porphyrin 60, 476 mg (35%), northern diol 67 186 mg (13%), southern diol 68, 687 mg (48%). Tetramethyl-3,4-dihydroxyl-2,3,5,8-tetramethylchlorin-I,4-diacetate-6,7- dipropionate (67) 1H-NMR 5 1.91 (3 H, s, Me sat.), 3.15 (4 H, t, QHZCHZCOZ), 3.20, 3.38, 3.41 (3 H each, s, ring Me), 3.62, 3.67 (3 H each, s, CHZCHZCOZQH3), 3.94 (3 H, s, C_H_2COZCH3), 4.18 (4 H, m, C_IizCOz sat., _Ci-IZCHZCOZ) 4.56, 4.68 (1 H each, d, I=16 Hz, C_HZCOZ), 9.04, 9.10, 9.57, 9.73 (1 H each, s, meso 0t, 0, y, 5), 12.88 (2 H, br 5, NH); UV-vis Kmax (rel. int.) 392 nm (1.00), 494 (0.11), 521 (0.04), 589 (0.05), 642 (0.29); MS found m/ e 717.8043 for (M+H)+, C38H45N40lo requires 717.8031. Tetramethyl-5,6-dihydroxy-2,3,5,8-tetramethyl-1,4-diacetate-6,7- 87 dipropionate (68) 1H-NMR 5 2.14 (3 H, s, Me sat.), 2.54, 2.75 (2 H each, m, C_H2C_H2COZ), 3.23, 3.33, 3.36 (3 H each, 5, ring Me), 3.48 (3 H, s, CHZCHzcoz_c_H3 sat.,) 3. 65 (3 H, s, CHZCH2C02C_H3), 3. 70, 3. 71 (3 H each, s, CHZCOZCH3), 4. 05 (2 H, m, .C__H2CH2COz)r 4.58, 4. 67 (1 H each, d, I: 16 Hz, QHZCOZ), 4.71 (2 H, s, CHZCOZ), 8.91 (2 H, s, meso p, y), 9.53, 9.62 (1 H each, s, meso Q, 5), -2.29, -2.36 (1 H each, br 5, NH); UV-vis Kmax (rel. int.) 392 run I (1.00), 497 (0.10), 522 (0.03), 588 (0.04), 640 (0.26); MS found m/ e 717.8025 for (M+H)+, C38H45N40lo requires 717.8013. Porphflinone 69, lactone 70 and methoxyr_nethyl compound 71 To the north diol 67 (186 mg, 0.26 mmol) was added 20 ml of concentrated sulfuric acid, and the mixture was stirred for 2 h at room temperature before being quenched with 100 ml of methanol in an acetone/dry-ice bath. The solution was then allowed to stand for 5 h, protected from moisture, to ensure the re-esterification. The solution was further diluted with 200 ml of CHzClz and washed with aqueous NaOAc ( 25%, 3 x 150 ml) and water (3 x 150 ml). The solvent was evaporated and the residue was chromatographed on preparative TLC plates, developed with 2% MeOH/ CH2C12. The fastest moving brown band was porphyrinone 69 (43 mg, 24%), followed by red colored hydroxymethyl porphyrin 71 (16 mg, 9%) and lactone 70 (83 mg, 47%). Tetramethyl-2,4,5,8-tetramethyl-porphyrinone-1,4—diacetate-6,6- dipropionate (69) 1H-NMR 5 1.94 (3 H, 5, Me sat.), 2.94 (3 H, s, COZC_I-I_3 sat.), 3.18, 3.25 (2 H each, t, CH2C_H_2COZ), 3.46, 3.54, 3.59 (3 H each, 5, ring Me), 3.65, 3. 66 (3 H each, s, CHZCHZCOZC 3H), 39,8 3. 89 (1 H each, d, ]=16 Hz, CHZCOZ sat.), 4.22, 4.37 (2 H each, t, CflzCHzCOZ), 5.02 (2 H, s, _C_I-_I_2COZ), 9.07, 9.86, 9.92, 9.94 (1 H each, s, meso (3, ol, y, 5), -2.82, -2.97 (2 H, br s, NH); UV-vis ‘Amax 88 (rel. int.) 404 nm (1.00), 506 (0.09), 543 (0.09), 587 (0.06), 614 (0.04), 643 (0.24); MS m / e found 699.7854, C38H43N409 requires 699.7878. Trimethyl-1-hydroxy-2,3,5,8-tetramethyl-1,2-(y-lactone)chlorin-.4- acetate-6,7-dipropionate (70) lH-NMR d 2.35 (3 H, s, Me sat.), 2.87 (4 H, m, C_HZCHZCOZ), 3.08, 3.26, 3.44 (3 H each, s, ring Me), 3.55 (2 H, m Q—IZCOZ sat.), 3.53, 3.61 (3 H each, s, CHZCHZCOZCHg), 3.74 (3 H, s, CH2C02C_H3), 3.80 (4 H, m, C_H_2CH2COZ), 4.83, 4.84 (1 H, each, s, CHZCOZ), 9.11, 9.13, 9.15, 9.60 (1 H each, s, meso 01, p, y, 5), -3.10 (2 H, br 8, NH); UV-vis hmax (rel. int.) 390 (1.00), 493 (0.10), 541 (0.03), 587 (0.04), 640 (0.24); MS found m/ e 685.7633 for (M+H)+, C37H41N409 requires 685.7607. Tetramethyl-3-hydroxymethyl-2,5,8-trimethylporphyrin-1,4-diacetate-6,7— dipropionate (71) 1H-NMR 5 3.26, 3.27 (2 H each, t, C_HZCH3C02), 3.59 (3 H, s, ring Me), 3.64 (6 H, s, ring Me), 3.65 (6 H, s, CHZCHZCOZQH3), 3.74, 3.75 (3 H, s, _C_H2COZCH3), 4.36, 4.41 (2 H each, t, _C_I;I_2CH2COZ), 5.04, 5.15 (2 H each, s, QIQCOZ), 5.94 (2 H, s, _Cflon), 10.04, 10.06, 10.16, 10.29 (1 H each, s, meso); UV-vis kmax (rel, int.) 401 nm (1.00), 499 (0.10), 534 (0.07), 568 (0.06), 622 (0.05), 642 (0.043); MS found m/e 699.7847 for (M+H)+, C38H43N409 requires 699.7878. Tetramethyl-3,5-porphflindione-1,4-diacetate-6,7-dipropionate (73) Starting with 69, the zinc insertion, OsO4 oxidation and sulfuric acid catalyzed rearrangement were accomplished as described previously. Chromatographic separation on TLC plates, developed with 2% MeOH/ CH2C12, gave 3,5-dione 73 as the major product in pair of cis and trans isomers. Yield, 44%; 1H-NMR 1.61, 2.15 (1 H each, m, _C_H_2CH2COZ), 181, 193 89 (3 H each, 5, Me sat.), 2.89 (2 H, t, CHZQLIZCOZ sat.), 3.09 (2 H, t, CH2C_H_2COZ), 3.17 (3 H, s, COZC_I-l3 sat.), 3.35, (2 H, t, CH2C_H2COZ), 3.37, 3.41 (3 H each, s, ring Me), 3.62 (3 H, s, CHZCHZCOZCH3), 3.73 (3 H, s, CH2C02C_H3), 3.90 (2 H, m, C_H2C02 sat.), 4.09 (2 H, t, CH2CH2COZ), 481 (2 H, s, C_HZCOZ), 858, 865, 943, 955 (1 H each, s, meso y, p, on, 5), -0.50 (2 H, br s, NH); UV-vis Kmax (rel. int.) 416 nm (1.00), 437 (0.83), 546 (0.13), 587 (0.19), 639 (0.21); MS found m/ e 715.7836 for (M+H)+, C38H43N40lo requires 715.7872. B. 1,4-Bis-(2-chloroethyl)-porphyrin system Ethyl-diacetoacetag In a 2 L flask were placed 138 g (1 mol ) of KZCO3, 30 g (0.18 mol) mol (crushed up finely), 100 g (1 mol) of 2, 4—pentanedione, 167 g (1 mol) of ethyl bromoacetate and 350 ml of 2-butanone. The mixture was heated slowly to reflux. After about 30 min, the refluxing became vigorous but subsided gradually. The reaction was continued for 4 h and then the mixture was cooled and diluted with 1 L of acetone. The inorganic salts were filtered off and washed further with acetone. The solvent was evaporated with a rotavapor and the orange colored crude product was distilled under vacuum. Yield, 153 g (82%). 1H-NMR (as ketone -- enol tautomers) 5 1.27 (3 H, t, COZCH2QH3), 2.15, 2.18, 2.27 (6 H, s, COC_1-I_3), 2.86, 2.89, 3.24, 3.26 (3 H, d, s, C_I_-I_C_H2), 4.14 (2 H, q, _CflzCH3). Ethleacetoacetate oxime 136 g of ethyl acetoacetate was dissolved in 300 ml of acetic acid and the solution was cooled in an ice bath. The saturated aqueous solution of sodium nitrite , 73 g (1.06 mol), was added dropwise with stirring and the temperature was controlled below 20 °C. The reaction was continued for 90 another 2 h after the addition. The orange-colored oxime solution was kept at low temperature or used immediately. Ethyl-4-ethoxycarbonylmethyl-3,S-dimethylpyrrole-2-carboxylate To the well stirring solution of ethyl-diacetoacetate (186 g, 1 mol) in 350 ml of AcOH was added the above oxime solution (1.05 mol) dropwise while adding zinc powder (200 g, 3.06 mol) such that to maintain the reaction temperature between 90 to 95 °C. After the addition, the reaction mixture was heated for 1 h before being poured onto 1000 g of ice. The crude product was precipitated as a yellow solid which was collected by filtration and washed with water. The solid was then redissolved in 500 ml of CHZCIZ, filtered again to remove zinc powder and dried over N a2504. After evaporation of solvent the pyrrole was crystallized from ethanol. Yield, 103.7 g (41%); MP, 85-87 0C; lH-NMR 5, 1.21 (3 H, t, CH2C02CH2_C_H3), 1.32 (3 H, t, COZCH2_C_H3), 2.21, 2.26 (3 H each, s, Me), 3.34 (2 H, s, C_HZCOZ), 4.09 (2 H, q, CHZCOZ_CI_12CH3), 4.27 (2 H, q, COZCflzCH3), 8.97 (1 H, br 5, NH). Eth 1-3- 2-h dro eth 1-3 5-dimeth l ole-2-carbo late Boron trifloride-ether complex (130 ml) was added dropwise to a solution of sodium borohydride (25 g) in diglyme (50 ml), and the diborane generated was flushed with a slow steam of nitrogen into a solution of foregoing pyrrole (25.3 g, 0.1 mol) in 100 ml of THF. The reaction was continued for 45 min, methanol was then carefully added to quench the excess 82H6 until effervescence ceased. The solvent was evaporated and the crude product pyrrole was crystallized form methanol. Yield, 18.7 g (89%); MP, 109-111 0C; 1H-NMR 5 1.26 (3 H, t, COZCH2C_H3), 2.20, 2.23 (3 H each, s, Me), 2.57, 3.51 (2 H each, t, _C_H_2C_HZOH), 4.19 (2 H, q, COsz-IZCH3), 10.11 (1 H, 91 br 5, NH). Eth 1-3- 2-chloroeth 1-2 4-dimeth l ole-5-carbox late 7 The above hydroxyethylpyrrole pyrrole (10.05 g 0.05 mol) was dissolved in 200 ml of benzene, 20 g of anhydrous KZCO3, and 6 ml of thionyl chloride were added subsequently. The mixture was heated under refluxing for 3 h. After cooling, the solution was filtered to remove the inorganic salts and the filtrate was brought to dryness. The crude product was crystallized from methanol to give 10.3 g of pyrrole 75, 90%. MP, 110-112 °C;1H-NMR 5 1.22 (3 H, t, COZCH2_C_Ii3), 2.11, 2.15 (3 H each, 5, Me), 2.72, 3.38 (2 H each, t, C_HZC_H_2C1), 4.17 (2 H, q, COZQHZCH3), 8.60 (1 H, br 5, NH). -3 5 5'-tetrameth 1-2 2'-di rr lmethenium bromide Pyrrole 75, 11.5 g (0.05 mol), in 30 ml of formic acid (98-100%) was heated on a steam bath. To this solution 7 ml of concentrated hydrobromic acid (48%) was added in one portion and the heating was continued for about 2 h until gas evolution ceased. On standing at room temperature overnight the dipyrrylmethenium salts were crystallized as a dense chunky solid. It was collected by filtration, washed with cold methanol to yield 5.1 g of 76, (50%). MP, 217219 0c; 1H-NMR 5 2.28, 2.66 (6 H each, s, Me), 2.96, 3.38 (2 H each, t, Cflzngl), 7.08 (1 H, s, methine), 8.01 (2 H, 5, NH). Dimethyl—1,4-bis-( -chloroeth11)-2,3,5,8-tetramethylporphyrin-6,7-_c_l_i; propionate (61) The condensation of 4.06 g (0.01 mol) of dipyrrylmethenium bromide 76 and 5.83 g (0.01 mol) of dipyrrylmethenium bromide 66 was carried out in the 92 same way as described in the synthesis of porphyrin 60. The porphyrin 61 obtained from column chromatography was crystallized from CHzClz-MeOH as sparkling purple blades, 2.12g (32%). MP, sintered at 202 oC and melted at 208 0c (lit. MP 201-203 0C); 1H-NMR 3.26 (4 H, t, CHZQHZCOZ), 3.49, 3.57 (6 H each, 5, ring Me), 3.66 (6 H, s, CHZCH2C02QH3), 4.06 (4 H, t, C_HZCHZCOZ), 4.36 (8 H, m, C_HZQHZCI), 9.81, 9.83, 9.89, 10.01 (1 H each, s, meso), -3.99 (2 H, br 5, NH); UV-vis kmax (rel. int.) 400 nm (1.00), 498 (0.13), 532 (0.10), 568 (0.08), 620 ( 0.06); MS (direct probe) m/ e 663 (M+). Dihydroxylchlorin 77,78, porphyrinone 79, 1,3-porphyrindione 82, 1,7-porplLWindione 83 and 1,8-porphyrindione &4_. To a solution of 1,4-bis-(2-chloroethyl)-porphyrin 61, 6.63 g (10 mmol) in 200 ml of CH2C12, pyridine (5 ml) was added, followed by 3.10 g (12 mmol) of 0504. The reaction was stirred in the dark, under nitrogen, for 24 h before being diluted with 100 m1 of methanol and quenched with H25. The oxidation products were chromatographed on a silica gel column. Porphyrin 61 (1.52 g, 23% recovered) was eluted first with methylene chloride while the mixture of the two dihydroxylchlorins were washed out with 2% MeOH/CHzClz. This mixture was then chromatographed on TLC plates, developed with 8% EtOAc/ CH2C12, to separate the fast moving northern diol 77 (0.767 g, 11%) from the southern diol 78 (2.16 g, 31%). The distinction of 77 and 78 was based on derivatization: the southern isomer underwent dehydration in dioxane with diluted acid to form an acrylate side chain, which gave the specific proton NMR peak at 4.10 ppm (CH=CH2COZC_Ii3) together with two pairs of doublet at 6.92 and 9.09 ppm (CH=CHCOZ). The above north diol 77 (1.1 mmol) was stirred with 10 ml of concentrated sulfuric acid for 30 min to effect the pinacolic rearrangement 93 followed by re-esterification with methanol. The porphyrinone 79 was obtained in a yield of 680 mg (91%) after being chromatographed on TLC plates. The zinc insertion was carried out by reacting 79 with zinc acetate in 100 ml of CHC13 and 20 ml of methanol. After 30 min refluxing, the mixture was concentrated, the zinc complex was precipitated by the addition of methanol and filtered in virtually quantitative yield. To a solution of the above zinc complex in 150 ml of CHzClz, 2 ml of pyridine and 380 mg (1.5 mmol) of osmium tetroxide were added and the reaction was allowed to proceed in the dark, under argon, for 24 h at room temperature. The reaction mixture was worked up by treating with methanol and H25 as described before and chromatographed on silica gel column (1% MeOH/ CH2C12, then 4% MeOH/CHzClz). Yield, recovered Zn(II) 79, 379 mg (51 %), and a mixture of two diols (80 and 81), 350 mg (45%). The mixture of zinc diols (0.45 mmol) was dissolved 10 ml of concentrated H2504 and allowed to stir at room temperature for 30 min. After being cooled in a dry ice—acetone bath, 100 ml of MeOH (100 ml) was added carefully to the solution, which was then partitioned between CHzClz (200 m1) and water (100 ml). The organic layer was separated, washed with aqueous sodium acetate (25%, 2 x 100 ml), water (2 x 100 ml), dried over N a2504 and evaporated to dryness, to give, after separation on TLC plates (1% MeOH/ CHzClz), 146 mg (48%) of regenerated porphyrinone 79, 25 mg (8%) of 1,3-dione 82, 44 mg (14%) of 1,7-dione 83 and 22 mg (7%) of 1,8-dione 84. Dimethyl-l,4-bis-(2-chloroethyl)-1,2-dihydroxyl-2,3,5,8~tetramethyl- chlorin-6,7-dipropionate (77) 1H-NMR 5 2.03 (3 H, s, Me sat.), 2.65, 3.20 (2 H each, m, QHZCfiZCl sat.), 2.77, 2.79 (2 H each, t, CHZQHZCOZ), 3.15, 3.16, 3.24 (3 94 H each, 5, ring Me), 3.62, 3.68 (3 H each, s, COZC_H_3), 3.90 (8 H, m, QflzC_H_2C1 and C_I-1_2CH2COZ), 8.65, 8.66, 9.29, 9.31 (1 H each, s, meso on, 5, y, (3); UV-vis kmax (rel. int.) 393 nm (1.00), 494 0.11), 521 (0.05), 589 (0.06), 642 (0.29). Dirnethyl-1,4-bis-(2-chloroethyl)-7,8-dihydroxyl—2,3,5,8-tetramethyl- chlorin-6,7-propionate (78) lH-NMR 5 2.01 (3 H, s, Me sat.), 2.50 (2 H, m, QflzCHZCOz sat.), 3.01, 3.29 (2 H each, t, CH2_C_H_2COZ), 3.03, 3.18, 3.25 (3 H each, 5, ring Me), 3.57, 3.71 (3 H, each, s, COZQH3), 3.94 (2 H, t, Q_I_-I_2CH2COZ), 3.78 -4.22 (8 H, m, CflzgflzCl), 8.72, 8.77, 9.30, 9.35 (1 H each, s, meso y, 5, 01, p); UV-vis Kmax (rel. int.) 392 nm (1.00), 495 (0.12), 522 (0.04), 588 (0.05), 641 (0.27). Dimethyl-2,4-bis-(2-chloroethyl)-2,3,5,8-tetramethyl-porphyrinone- 6,7-dipropionate (79) 1H-NMR 5 2.16 (3 H, s, Me sat.), 2.40, 3.28 (1 H each, m, _CflzCHZCl sat.), 3.22 (4 H, t, CHZQLIQCOZ), 3.32 (2 H, m, CH2C_H2C1), 3.40, 3.53, 3.55 (3 H each, s, ring Me), 3.66, 3.67 (3 H each, s, COZC_H3), 4.18 (4 H, m, QflzCHZCOZ), 4.26, 4.42 (2 H each, t, QHZQL-IZCI), 9.16, 9.66, 9.85, 9.86 (1 H each, s, on, (3, 5,Y), -3.01 (2 H, br s, NH); UV-vis Kmax (rel. int.) 409 nm (1.00), 510 (0.10), 548 (0.11), 589 (0.08), 615 (0.06), 644 (0.25); MS (direct probe 70 eV), m/ e 679 (M+). Dimethyl-2,4-bis-(2-chloroethyl)-2,4,5,8-tetramethyl-1,3-porphyrindione- 6,7-dipropionate (82) lH-NMR 5 1.95, 1.94 (3 H each, s, Me sat.), 3.25 (12 H, m, C_IizgfizCl sat., CHzgflzCOz), 3.34, 3.36 (3 H each, s, ring Me), 3.58, 3.60 (3 H each, s, CHZCHZCOZQH3), 4.23 (4 H, m, C_HZCHZCOZ), 8.61, 8.79, 9.41, 9.45 (1 H each, s, meso p, Q, 5, v), -0.67 (2 H, br s, NH); UV-vis kmax (rel. int.) 418 nm (1.00), 439 (0.94), 545 (0.13), 592 (0.18), 637 (0.22); MS found m/ e 696.6521 for (M+H)+, C36H41N406C12 requires 696.6574. 95 Dimethyl-1,4-bis-(2—chloroethyl)-2,3,5,8-tetramethyl—1,7-porphyrindione- 6,7—dipropionate (83) 1H-NMR 5 1.76, 2.10 (1 H each, m, _C_I-_12CH2COZ sat.), 1.95, 1.97 (3 H each, s, Me sat.), 2.89 (2 H, m, CHZQIZCOzsatJ, 3.01, 3.32 (6 H, m, C_H2C_H2Cl sat. and CHZQHZCOZ), 3.35, 3.42 (3 H each, 5, ring Me), 3.45 (3 H, s, COz_C_Ii3 sat.), 3.70 (3 H, s, COz_C_H3), 4.00-4.50 (6 H, m, QfizglizCl and QHZCHZCOZ), 8.59, 8.78, 8.33, 9.51 (1 H each, s, meso 0L, 5, y, p), -0.60 ( 2 H, br 5, NH); UV-vis Xmax (rel. int.) 403 nm (0.75), 419 (1.00), 440 (0.94), 545 (0.14), 593 (0.20), 638 (0.24); MS found m/ e 696.6492 for (M+H)+, C36H41N406C12 requires 696.6574. Dimethyl-2,4-bis-(2-chloroethyl)-2,3,5,7-tetramethyl-1,8-porphyrindione- 6,7-dipropionate (84) 1H-NMR 5 1.67, 2.15 (1 H each, m, gizCHzCOz sat.), 1.98, 1.99 (3 H each, s, Me sat.), 3.10-3.30 (4 H, m, _C_H2C_I:I_2C1), 2.97 (2 H, t, CH2C_H_2C02 sat.), 3.14 (2 H, t, CHngizCOz), 3.33 (3 H, s, COZ_C_H_3 sat), 3.49, 3.57 (3 H each, 5, ring Me), 3.62 (3 H, s, COZC_H3), 4.11 (2 H, t, C_HZCHZCOZ), 4.22, 4.55 (2 H each, m, Qfl2C_H2Cl), 8.93, 8.98, 9.63, 9.77 (1 H each, 5, 0L, y, 5, (3); UV-vis Xmax (rel. int.) 417 (1.00), 436 (0.76), 539 (0.06), 592 (0.11), 623 (0.17), 694 (0.04); MS found m / e 696.6543 for (M+H)+, C36H41N406C12 requires 696.6574. C. 1,3-Bis-(2-chloroethyl)-porphyrin system Dimethyl-l,3-bis-(2,2-dimethoxyethyl)-2,4,5,8-tetramethylporphflin-6,7- digropionate (86) To a solution of protoporphyrin IX dimethyl ester (5.91 g, 0.01 mol ) in 1.5 L of methylene chloride and 250 ml of methanol was added 15.5 g (0.035 mol) of Tl(NO3)3.3HZO dissolved in 500 ml of Methanol. The solution was stirred at 40 °C for 10 min and the white TlNO3 was precipitated out. 96 Hydrogen sulfide was then bubbled into the solution for 10 min followed by addition of 25 ml of concentrated HCl to destroy the T1(III)-porphyrin complex. The mixture was stirred for 5 min and then filtered to remove the solid inorganic salt. The organic filtrate was washed with aqueous sodium acetate (25%, 2 x 200 ml), water (3 x 300 m1) and then brought to dryness. Porphyrin 86 was recrystallized from CHzClz-MeOH with a yield of 6.65 g (95%). MP, 230-232 °C; lH-NMR 5 3.28, 3.32 (2 H each, t, C_HZCHZCOZ), 3.45, 3.47 (6 H each, s, CH(OC_H3)2), 3.57, 3.59. 3.60, 3.62 (3 H each, s, ring Me), 3.66, 3.68 (3 H each, s, COZQH3), 4.09, 4.20 (2 H each, d, C_HZCH(OCH3)2), 4.34, 4.39 (2 H each, t, C_I-lzC_H2COZ), 5.08, 5.14 (1 H each, t, CH2C_H(OCH3)2), 10.00, 10.03, 10.06, 10.09 (1 H each, s, meso), -4.02 (2 H, br 5, NH); UV-vis Kmax (EM) 400 nm (169800), 498 (12300), 533 (7900), 567 (5500), 621 (3300); Dimethyl-I,3-bis-(hydrogencarbonylmethyl)-2,4,5,8-tetramethy1-6,7- dipropionate (8]) Porphyrin 86, 3.6 g (0.05 mol) was dissolved in 1000 ml of tetrahydrofuran and the solution was heated to reflux, followed by addition of 36 ml of diluted hydrochloric acid (4.5 ml of concentrated HCl in 31.5 ml of water) in one portion. The mixture was refluxed for 5 min and then cooled immediately in an ice bath. The solution was diluted with 1000 ml of chloroform, and washed with aqueous sodium acetate (25%, 500 ml) and water (2 x 800 ml). The solvent was evaporated under vacuum and the residue was crystallized from CHzClz-MeOH (or used directly to undergo either reduction by N aBH4 to 88 or oxidation by Jones reagent to porphyrin 58). Isolated yield, 2.56 g (81.3%); lH-NMR 5 3.23 (4 H, t, CHngzcoz), 3.37 (6 H, s, Me), 3.49, 3.50 (3 H each, s, Me), 3.66 (6 H, s, COZ_C_H_3), 4.29, 4.30 (2 H each, t, _C_H2CH2COZ), 4.75, 4.80 (2 H each, d, _C_fl2CHO), 9.55, 9.62, 9.80, 9.24 ( 1 H 97 each, s, meso), 10.10, 10.13 (1 H each, t, CH0), -4.35 (2 H, br 5, NH). Dimethyl-1,3-bis-(2-dihydroxyethyl)—2,4,5,8-tetramethylporphylin-6, - dipropionate (88) The foregoing porphyrin dialdehyde diester (87) (containing a small amount of monoester), without further purification, was stirred in a mixture of THF and MeOH (4:1) in an ice bath and treated with 10.5 g of NaBH4 in cold methanol. The reaction was allowed to proceed for 10 min before 35 m1 of acetic acid was carefully added to quench the excess borohydride. The solvent was evaporated under vacuum and the residue was treated with acidified methanol (5 ml H2504 in 200 ml MeOH). After 8 h, the acidic solution was diluted with 500 ml of CHzClz, washed with 500 ml of N aOAc solution (25%), then with water (3 x 500 ml). After dried by vacuum, the crude product was chromatographed on a neutral alumina column (Brockman Grade 111) with CHC13 to separate a small amount of unreacted acetal porphyrin, and the porphyrin dialcohol 88 was eluted with 3% MeOH/CHCl3 to give a yield of 2.28 g (73%). 1H-NMR 5 3.25 (4 H, t, CHZCfiZCOz), 3.57 (12 H, s, ring Me), 3.65 ( 6 H, s, COZQH3), 4.23 (4 H, t, _CfizCHZCOZ), 4.37 (8 H, m Q—I_2C_H_ZOH), 9.98, 10.01, (2 H each, s, meso), -3.82 (2 H, br s, NH). UV-vis Xmax (rel. int.) 399 nm (1.00), 499 (0.11), 533 (0.08), 566 (0.06), 620 (0.05). Dimethyl-1,3-bis-( -chloroethyl)-2,4,5,8-tetramethylporphflin-6,7- dipropionate (62) To a solution of above porphyrin dialcohol 88 (2.5 g, 4 mmol) in 500 ml of dry DMF was added 30 ml of benzoyl chloride. The mixture was heated at 98 °C for 1 h under nitrogen and then allowed to be cooled. Water (600 ml) 98 and triethylamine (40 ml) were added. The precipitated porphyrin 62 was collected by filtration and washed with water. The crude porphyrin was passed through a short silica gel column and then crystallized from CHzClz-MeOH. Yield, 2.38 g (90%); MP 216-218 °C (lit. 216-217 0C); 1H-N MR 5 3.24, 3.26 (2 H each t, CH2C_H2COZ), 3.46, 3.50, 3.53, 3.56 (3 H each, 5, ring Me), 3.66, 3.67 (3 H each, s, COZQi-I_3), 4.19-4.38 (12 H, m QflzgfizCl and QHZCHZCOZ), 9.75, 9.82, 9.87, 9.96 (1 H each, s, meso); UV-vis Xmax (rel. int.) 399 nm (1.00), 498 (0.11), 533 (0.08), 567 (0.06), 621 ( 0.05); MS (direct probe 70 eV), m/ e 663 (M+). Porphyrinone 93 and L4 Porphyrin 62, 6.63 g (10 mmol), in 400 ml on CHzClz and 5 ml of pyridine was treated with 3 g (12 mmol) of osmium tetroxide under the same condition as described before. After working up, the mixture of the oxidation products was chromatographed on a silica gel column, eluted with 1% MeOH/CH2C12 to give 1.59 g (24%) of unreacted porphyrin 62. The faster moving northern diols were separated partially from the slower moving southern diols. Further purification on TLC plates with 8% EtOH/CH2C12 gave a combined yield of 1.25 g (18%) for the northern diols and 3.34 g (48%) for the southern diols. The mixture of the two northern diols (1.79 mmol) was dissolved in 15 ml of concentrated sulfuric acid and stirred at room temperature for 30 min. Porphyrinone 93 and 94 were obtained, after working up and separation by chromatography, in a total yield of 91%, 559 mg for 93 and 547 mg for 94. To achieve a higher yield of 1,3-dione the mixture of these two porphyrinones can be used directly without separation for the next step. Dimethyl-l,3-bis-(2-chloroethyl)-2,4,5,8-tetramethyl-1-porphyrinone- 99 6,7-dipropionate (93) 2.09 (3 H, 5, Me sat.), 2.88, 3.28 (1 H each, m, QfizCHzCl sat.), 3.20 (6 H, m, CHzg-IZCI sat. and 2CH2C_H2COZ), 3.45, 3.57, 3.62 (3 H each, s, ring Me), 3.63, 3.65 (3 H each, s, CH2C02C_H3), 4.21, 4.24, 4.36, 4.45 (2 H each, t, CH2_C_H_2C1 and 2_Cfl2CH2COZ), 9.11, 9.84, 9.87, 9.95 ( 1 H each, s, meso on, 5, p, y), -2.92, -2.98 (1 H each, br s, NH); NOE test: selective irradiation of 8-Me at 3.45 ppm caused an increase of 9.7% in intensity for 5 -proton, and irradiation of 4- and 5-Me at 3.57 and 3.62 ppm gave p-meso proton an increase of 9.8% and 10.8%, respectively; UV-vis Xmax (rel. int.) 407 nm (1.00), 508 (0.09), 547 (0.10), 588 (0.07), 645 (0.23). Dimethyl-1,4-bis-(2-chloroethy1)-2,4,5,8-tetramethyl-3-porphyrinone-6,7- dipropionate (94) lH-NMR 5 2.09 (3 H, s, Me sat.), 2.86, 3.29 (1 H each, m C_HZCHZCI sat.), 3.23, (6 H, m, CHZCflZCl sat. and 2CH2QHZCOZ), 3.42, 3.51, 3.57 (3 H each, s, ring Me), 3.67 (6 H, s, CHZCHZCOZCH3), 4.21, 4.55 (4 H each, m, QflzgflzCl, and 2CfizCH2COZ), 9.12, 9.77, 9.82, 9.87 (1 H each, s, meso p, (X, 5, Y), -2.92, -3.01 (1 H each, br 5, NH); UV-vis Xmax (rel. int.) 407 nm (1.00), 508 (0.10), 547 (0.11), 587 (0.08), 643 (0.24). 1,7-dione 22 Zinc insertion to the forgoing porphyrinones 105 and 106 was accomplished as usual in CHCl3-MeOH with Zn(OAc)2. The mixture of these two complexes (744 mg, 1 mmol), was treated with osmium tetroxide (380 mg, 1.5 mmol) to effect dihydroxylation as described previously in a total yield of 404 mg ( 52%). 305 mg (41%) of the unreacted starting material was recovered. The rearrangement of diols were carried out in the concentrated sulfuric acid (15 ml), yielding mainly six porphyrinones, in addition to the regenerated porphyrinones (133 mg, 33%), with the 1,3-dione 82 as the major 100 product after separation. Dimethyl-2,4-bis-(2-chloroethyl)-2,4,5,8-tetramethyl-1,3-porphyrindione- 6,7-dipropionate (82) It was found the same as that obtained form 1,4-bis- (2-chloroethyl-porphyrin system. Yield, 79 mg (22%). Dimethyl-l,4-bis-(2-chloroethy1)-2,4,6,8-tetramethyl-3,5-porphyrindione— 6,7-dipropionate (95) Yield, 61 mg (17%); 1H-NMR 5 1.74, 2.20 ( 1 H each, m, _CflZCHzCOZ sat.), 1.94, 1.97 ( 3 H each, 5, Me sat.), 2.98 (2 H, m, CH2C_H2C02 sat.), 3.03 to 3.35 (4 H, m _CfizgflzCl sat.), 3.21 (2 H, m, CH2_C_H2COZ), 3.39 (3 H, s, CHZCH2C02_C_I_-I_3 sat), 3.37, 3.44 ( 3 H each, 5, ring Me), 3.64 (3 H, s, CHZCH2C02C_H3), 4.08, 4.18, 4.31 (2 H each, t, C_HzgfizCl, and C_HZCHZCOZ), 8.64, 8.74, 9.40, 9.50 (1 H each, s, meso y, (3, ol, 5), -0.42 (2 H, br 5, NH); UV-vis kmax (rel. int.) 403 nm (0.75), 419 (0.93(, 439 (1.00), 540 (0.13), 592 (0.18), 636 (0.21), 684 (0.05). Dimethyl-1,4-bis-(2-chloroethyl)-2,4,5,8-tetramethyl-3,7-porphyrindioneo 6,7-dipropionate (96) Yield, 18 mg (5%); 1H-NMR 5 1.54, 2.12 (1 H each, m, C_HZCHZCOZ), 2.05 (6 H, 5, Me sat.), 2.95, 3.30 (4H, m, C_H_2_C_H2Cl sat.), 3.28 (3 H, s, COZQHB sat.) 3.04 (2 H, t, CHZQHZCO2 sat.), 3.26 (2 H, t, CH2Q12COZ), 3.54, 3.56 (3 H each, 5, ring Me), 3.76 (3 H, s, COZ_C_I_i3, 4.27, 4.31, 4.44 (2 H each, t, C_Hzgljol and C_HzCHZCOz), 9.08, 9.09, 9.67, 9.77 (1 H each, s, meso 5, p, y, CL), -2.70, -2,75 (1 H each, br s, NH); UV-vis kmax (rel. int.) 402 nm (0.86), 411 (1.00), 483 (0.05), 511 (0.06), 553 (0.07), 623 (0.05), 652 (0.06), 686 (0.52). Dimethyl-2,3-bis-(2-chloroethy1)-2,4,5,8-tetramethyl-1,8—porphyrindione (97) Yield, 14 mg (4%); 1H-NMR 5 1.67, 2.18 (1 H each, m, C_I-_12CH2C02 sat.), 2.00, 2.02 (3 H each, 5, Me sat.), 3.00 (2 H, t, CHZQHZCOZ sat.), 3.09-3.35 ( 4 H, m, 101 (2129?le sat.), 3.16 (2 H, t, CH2C_H2COZ), 3.35, (3 H, s, COZC_I-l3 sat.), 3.57, 3.61 (3, H each, s, ring Me), 3.63 (3 H, s, COZC_H3), 4.11, 4.21, 4.56 (2 H each, t, C_HZQHZCI and QIjZCHzCOz), 8.91, 8.89, 9.63, 9.84 (1 H each, s, meso y, Q, 5, (3), -1.60 (2 H, br 5, NH); UV-vis Xmax (rel. int.) 416 nm (1.00), 436 (0.75), 541 (0.06), 591 (0.10), 623 (0.16), 690 (0.04). Dimethyl-1,4-bis-(2-chloroethyl)-1,4,5,8-tetramethyl-2,3-porphyrindione- 6,7-dipropionate (98) Yield, 11 mg (3%); 1H-NMR 5 1.98 2.00 (3 H each, s, Me sat) 2.90-3.35 (8 H, m, C_HZCHZCI sat.), 3.21 (4 H, t, CH2C__H2COZ), 3. 46 (6 H, s, ring Me), 3.60 (6 H, s, COZC_H3), 8.92 (2 H, s, meso p, 5), 9.66, 10.00 (1 H each, s, meso Q, y), -1.60 ( 2 H, br 5, NH); UV-vis Xmax (rel. int.) 416 nm (1.00), 435 (0.76), 591 (0.10), 625 (0.17), 686 (0.06). Dimethyl-3,3-bis-(2-chloroethyl)-2,4,5,8-tetraethyl-1,7-porphyrindione- 5,7-dipropionate (99) Yield, 7 mg (2%); 1H-NMR 5 1.75, 2.11 (1 H each, m, C_HZCHZCOz), 1.97, 1.99 ( 3 H each, 5, Me sat.), 2.93, (2 H, t, CH2§_H_2COz sat.), 3.02-3.83 (4 H, m, _CflzgflzCl), 3.11 (2 H, t, CH2C__H2CO,_), 3. 44 (3 H, s, COzfia sat), 3.,45 3.71 (3Heach, 5, ring Me), 3.71 (3H, s, COZC_ H3), 41,2 4.21 (6H, m, QHZQLIZCI and C_flzCHZCOz), 8.58, 8.79, 9.33, 9.58 (1 H each, s,meso Q, 5, Y, (B), -0.53 (2 H, br 5, NH); UV-vis Xmax (rel. int.) 418 nm (1.00), 439 (0.93), 544 (0.12), 592 (0.18), 636 (0.21). D. Oxidation of porphyrinone and 1,3-porphyrindione side chains Dimethyl-2-hydroxyr_nethyl-2,3,5,8-tetramethyl-2-vinylporphginone-6,7-di- propionate (100) To a solution of 2,4-bis-(2-chloroethyl)-porphyrinone 79 (68 mg, 0.1 mmol) in 40 ml of pyridine, under argon, 0.8 g of KOH in 3 m1 of water was 102 added and the solution was refluxed for 5 h. The mixture was then cooled and evaporated to dryness under vacuum. The residue was dissolved in 20 ml of methanol with 0.5 ml of sulfuric acid and allowed to stand at room temperature for 6 h. The crude product was worked up by washing the solution with aqueous sodium acetated and then water, dried over NaZSO4, evaporated under vacuum, chromatographed on a TLC plate to give 56 mg (90%) of 112. 1H-NMR 5 2.10, (3 H, s, Me sat.), 2.90-3.20 (4 H, m, Qfl2C_HZOH), 3.17, 3.18 (2 H each, t, CHZQHZCOZ), 3.42, 3.51, 3.62 (3 H each, 5, ring Me), 3.61, 3.64 (3 H each, s, COZC_H3), 4.18, 4.29 (2 H each, t, QflZCHZCOZ), 6.30 (2 H, dd, CH=Q§2), 8.19 (1 H, dd, _C_I_H_=CH2), 9.19, 9.84, 9.90, 9.97 ( 1 H each, s, meso Q, 0, 5, Y), -2.92 (2 H , br s, NH); UV-vis Xmax (rel. int.) 412 (1.00), 511 (0.11), 550 (0.09), 596 (0.07), 652 (0.24); MS (direct probe, 70 eV), m/ e 625 (M+). Dimethyl-g-(hydrogencarbonvlmethfivl)-2,3,5,8-tetramethyl-4-vinyl-pogphyrin- one-6,7-diprogionate (101) . A solution of 10 m1 of CHzClz and 0.4 ml 0.44 mmol of oxalyl chloride was stirred in an dry-ice/ acetone bath under nitrogen. 0.68 ml (8.8 mmol) of dimethyl sulfoxide in 2 ml of CH2C12 was added. The mixture was allowed to stir for 5 min before porphyrinone 112 (31 mg, 0.05 mmol) in 5 ml of CHzClz was added. After stirring for another 15 min, 0.28 ml 20 mmol) of triethyl amine was added and the reaction mixture was gradually warmed up to room temperature. The solution was diluted with 20 ml of CH2C12, washed with saturated NaCl solution (20 ml), aqueous HCl (5%, 2 x 20 ml), NazCO3 (5%, 20 ml), water, and then brought to dryness. Separation on TLC plate, developed with 1% MeOH/ CH2C12, gave 101 in a yield of 24 mg (78%). lH-NMR 5 2.00 (3 H, s, Me sat.), 3.19, 3.22 (2 H each, t, CH2_C_H2C02), 3.43, 3.58, 3.61 (3 H each, s, ring Me), 3.62, 3.64, (3 H each, s, COZQH3), 3.98 (2 H, dd, I=17 103 Hz, gHzcoz), 4.22, 4.38 (2 H each, t, C__HZCHZCOZ), 6. 32 (2 H, dd, CH=CH2), 8. 22 (1 H, dd, QH=CH2), 9.09, 9.90, 9.95, 9.99 (1 H each, s, meso Q, 0, 5, Y), -2.95, -2.87 (1 H each, d, NH); UV-vis Xmax (rel. int.) 412 (1.00), 512 (0.11), 548 (0.09), 594 (0.07), 651 (0.23); MS (direct probe, 70 eV), m/ e 623 (M+). Trimethyl-2,3,5,8-tetramethyl-4-vinylporphjginone-Z-acetate-6,7- dipropionate (192) To a solution of porphyrinone aldehyde 101 (20 mg, 0.03 mmol) in 15 ml of THF with 1 ml of water, 15 mg (0.12 mmol) of argentic oxide was added and the mixture was stirred at room temperature for 30 h. The solid was then filtered and the green filtrate was bought to dryness under vacuum. The residue was dissolved in 20 ml of methanol with 0.5 ml of sulfuric acid and allowed to stand for 5 h before being diluted with 30 ml of CH2C12. The solution was washed with aqueous sodium acetate (25%, 2 x 20 ml), water (3 x 20 ml) and then evaporated to dryness. The crude product was purified on TLC plate (1%MeOH/CH2C12) to give 15 mg (76%) of 102. 1H-NMR 5 2.04 (3 H, s, Me sat.), 2.87 (3 H, s, CHZCOZ_C_H3), 3.23 (4 H, m, CHzgflzCOz), 3.57, 3.63 (3 Heach, s,ring Me), 3. 65 (6H, s, CHZCHZCOZ__ 11,3) 367 (2H dd, C_HZCOZ), 426, 4.40 (2 H each, t, CHZCHZCOZ), 6. 33 (2 H, dd, CH2=CH), 8. 22 (1 H, CH2=_H_), 9.25, 9.91, 10.04, 10.08 (1 H each, s, meso Q, [3, 5, y), -2.91 (2 H, br 5, NH); UV-vis Kmax (rel. int.) 413 nm (1.00), 514 (0.12), 548 (0.08), 594 (0.07), 652 (0.25); MS (direct probe 70 eV), 653 for (M4) Dimethyl-gA-bis-(Z-hydroxyethyl )-2,4,5,8-tetramethyl-1 ,3-porphyrin-dione- 6,7-dipropionate (103) 2,4-Bis-(2-chloroethyl)-1,3-porphyrindione 82 (69 mg, 0.1 mmol) was treated with KOH as described above led to 103 in a total yield of 44% as a 104 mixture of cis and trans isomers. The two compound were separated on TLC plates after a prolonged developing time to give the slower moving band assigned as cis-dione 103a (13 mg) and faster moving component as the-trans isomer 103b (16 mg). (103a) 1H-NMR 1.94, 1.98 ( 3 H each, s, Me sat.), 2.80-3.14, 3.55, (8 H, m, QZQHZOH), 3.10 (4 H, m, CHZQHZCOZ), 3.29, 3.34 ( 3 H each, 5, ring Me), 3.60, 3.62 (3 H each, COZC_H3), 4.17, 4.18 (2 H each, t, QflzCHzCOz), 8.52, 8.72, 9.35, 9.62 (1 H each, s, meso, p, Q, 5, Y); UV-vis Xmax (rel. int.) 413 nm (1.00), 436 (0.73), 487 (0.10), 544 ( 0.13), 588 (0.16), 636 (0.17); MS (direct probe, 70 eV), 658 (M+). (1035) 1H-NMR 1.92, 1.95 (3 H each, s, Me sat.), 2.75-3.05, 3.54 ( 8 H, m, CHZSLHQOH), 3.08, 3.11 (3 H each, s, ring Me), 3.60, 3.62 (3 H each, s, COzC_Ha), 4.17, 4.18 (2 H each, t, CHzCHzCOz), 8.46, 8.66, 9.30, 9.55 (1 H each, s, meso, p, Q, 5, Y); UV-vis hmax (rel. int.) 415 nm (1.00), 437 (0.87), 545 ( 0.19), 588 (0.26), 637 (0.26); MS (direct probe, 70 eV), 658 (M+). E. Heme d1 synthesis from porphyrin-2,4-diacetate Tetramethyl-I ,3,5,8-tetramethylporphyrin-2,4-diacetate-6,7-dipropionate (58) Porphyrin 87 (3.12 g, 5 mmol) was dissolved in 400 ml of acetone with 5 ml of formic acid and stirred in an ice-NaCl bath. To this solution 38 ml of Jones reagent, which was made of 6.79 g of CrO3, 6 ml of H2504 and 50 ml of water, was added slowly and the solution was allowed to stir for another 15 min. Large part of the solvent was evaporated under vacuum (the temperature of the bath was controlled below 40 °C) and the residue was redissolved in 300 ml of methanol followed by addition of 5 ml of concentrated sulfuric acid. After standing for 8 h at room temperature the 105 solution was further diluted with 400 ml of CH2C12 and washed first with aqueous NaOAc (25%, 2 x 200 ml), and then with water (3 x 300 ml). The solvent was removed under vacuum and the crude product was chromatographed on a silica gel column, eluted with 1% MeOH/CHZCIZ. crystallization from CHZClz-MeOH gave porphyrin 58 in a yield of 3.13 g (92%); lH-NMR 5 3.26 (4 H, t, CHZCHZCOZ), 3.53, 3.56, 3.58, 3.60 (3 H each, s, ring Me), 3.64, 3.66 (3 H each, s, CHZCHZCOZSIfl3), 3.73, 3.75 (3 H each, s, CH2C02C_113), 4.34, 4.37 (2 H each, t, _C_HZCHZCOZ), 4.90, 4.96 ( 2 H each, s, C_IizCOZ), 9.94, 9.97, 9.99, 10.00 (1 H each, s, meso), -3.94 (2 H, br 5, NH); UV-vis Xmax (rel. int.) 400 nm (1.00), 498 (0.12), 532 (0.09), 567 (0.08), 621 (0.06); MS found m / e 683.7865 for (M+H)+, C38H42N408 requires 683.78934. Dihydroxychlorin 105 and 106 Osmium tetroxide (2 g, 7.8 mmol) was added to a solution of porphyrin 58 (4.5 g, 6.6 mmol) in 20 ml of CH2C12 and 5 ml of pyridine. The reaction was allowed to proceed, under nitrogen, in the dark, at room temperature for 20 h. The solution was diluted with 100 ml of methanol and then bubbled with H25 for 15 min. The precipitated osmium sulfide was filtered on celite and the filtrate was brought to dryness under vacuum. The residue was subjected chromatographic separation on a silica gel column. Most of the unreacted porphyrin was eluted first with 1% MeOH/CH2C12 and the faster moving north diols were separated from the south isomers to a large extent. The slower moving south diols was completely eluted with 5% MeOH/ CH 2C12. The northern diol containing fractions were further separated on preparative TLC plates (6% EtOAc/ CH2C12) to give the northern diols 105 and 106, (945 mg, 20%), and the southern diols 107 and 108, (3.36 g 106 51%), combined with the fractions obtained from the column. In addition, 990 mg (22%) of porphyrin 58 was recovered. The mixture of the two northern diols could be separated on TLC plate after a prolonged developing time, with diol 105 as the faster moving band, but usually were applied without separation for the next reaction to avoid decomposition. Tetramethyl-l,2-dihydroxy-1,3,5,8-tetramethylchlorin-2,4-diacetate-6,7- dipropionate (105) 1H-NMR 1.96 (3 H, 5, Me sat.), 3.12, 3.16 (2 H each, t, CHZCflzCOZ), 3.30 (3 H, s, CHZCOZQH3 sat.),3.37, 3.39, 3.65 (3 H each, 5, ring Me), 3.70, 3.72 (3 H each, s, CHZCH2C02_C_I-13), 3.94 (3 H, s, CH2C02C_H3), 4.15 (6 H, m, _C_H_2C02 and two _C_HZCHZCOZ), 4.68, 4.78 (1 H each, d, _C_HZCOZ), 9.03, 9.19, 9.61, 9.67 (1 H each, s, meso 5, Q, y, p), -2.79 (2 H, br s, NH); UV-vis Xmax (EM) 392 (191000), 495 (13500), 520 (2500), 589 (3900), 642 (44800); MS found m/ e 717.8077 for (M+H)+, C38H44N40lo requires 717.8032. Tetramethyl-3,4-dihydroxy-1,3,5,8-tetramethylchlorin-2,4-diacetate-6,7- dipropionate (106) 1H-NMR 5 1.95 (3 H, 5, Me sat.), 3.03 (3 H, s, COZ_C_H3 sat.), 3.15 3.17 (2 H each, t, CHzgflzCOz), 3.42 (3 H, s, COZC_H3 sat.), 3.48 (6 H, 5, ring Me), 3.65 (3 H,s, ring Me), 3.67, 3.74 (3 H each, s, CHZCHZCOZCflzj), 3.95 (3 H, s, CHZCOZ_C_I-I3), 4.19 (2H, s, QHZCOZ), 4.18, 4,32 (2 H each, t, _CfizCHZCOZ), 4.78, 4.89 (1 H each, d, C_HZCOZ), 9.02, 9.16, 9.66, 9.72 (1 H each, s, meso Q, 0, Y, 5), -2.73 (2 H, br 8, NH); UV-vis Xmax (EM) 392 nm (198000), 494 (13600), 590 (4100), 642 (45700); MS found m/ e 717.8055 for (M+H)+, C38H44N40lo requires 717.8032. Porphflinone 111,112 and lactone 113,114 The mixture of north diols 117 and 118 (1.43 g, 2 mmol) was dissolved in 107 an acid medium made of 10 ml of H2504, 10 ml of FSO3H and 1 ml of fuming H2504, and the mixture was stirred, in the dark, at room temperature for 5 h. The acid solution was then frozen in a dry-ice/ acetone bath and quenched carefully with 200 ml of methanol with shaking. The solution was warmed up slowly to room temperature and allowed to stand for 6 h before further diluted with 300 m1 of CH2C12. The organic solution was partitioned and washed with aqueous sodium acetate ( 25%, 3 x 200 ml), then washed with water (3 x 300 ml), dried over NaZSO4, and brought to dryness. Separation on preparative TLC plates (6% EtOAc/CHZCIZ, or 1% MeOH/CHZCIZ) gave porphyrinone 111 as the fastest moving band followed by its regioisomer 112 and then lactones 113 and 114. Tetramethy1-2,3,5,8-tetramethylporphyrinone-2,4-diacetate-6,7- dipropionate (111) Yield, 335 mg (24%); lH-NMR 5 1.95 (3 H, s, Me sat.), 2.96 (3 H, s, CHZCOZC_H3), 3.20, 3.23 (2 H each, t, CH2_C_H2COZ), 3.46, 3.56, 3.57 (3 H each, 5, ring Me), 3.63, 3.66 (3 H each, s, CHZCHZCOzgfia), 3.78 (3 H, s, CHZCOZQH3), 3.90, 4.00 (1 H each, d, _C_flzCOZ), 5.04 (2 H, s, _CfiZCOz), 9.13, 9.89, 9.92, 9.95 Q, 5, p, Y), -2.87, -2.93 (1 H each, br 5, NH); UV-vis Xmax (EM) 406 nm (169000), 507 (9500), 544 (12000), 588 (5900), 644 (32400); MS found m/ e 699.7832 for (M+H)+, C38H43N409 requires 699.7878. Tetramethyl-1,4,5,8-tetramethylporphyrinone-2,4-diacetate-6,7- dipropionate (112) Yield, 293 mg (21%); 1H-NMR 5 1.95 (3 H, s, Me sat.), 2.97 (3 H, s, CHZCOZQHB), 3.26 (4 H, m, CHsg-13COZ), 3.46., 3.55, 3.60 (3 H each, 8, ring Me), 3.66, 3.67 (3 H each, s, CHZCHZCOZCHQ, 3.81 (3 H, s, COZQH3), 3.88, 3.98 (1 H each, d, QHZCOZ), 4.23, 4.38 (2 H each, t, _CflzCHZCOZ), 4.99, 5.09 (1 H each, d, QIZCOZ), 9.07, 9.85, 9.86, 9.94 (1 H each, s, meso p, Y, Q, 5), -2.79, 2.91 (1 H each, 5, NH); UV-vis Xmax (EM) 405 nm (17300), 506 (9600), 544 (11800), 108 586 (5900), 642 (34800); MS found m/ e 699.7845 for (M+H)+, C38H43N409 requires 699.7878. Trimethyl-2-hydroxy-1,2-(Y-lactone)-1,3,5,8-tetramethylchlorin-4-acetate- 6,7-dipropionate (113) Yield, 82 mg (6%); 1H-NMR 5 2.36 (3 H, 5, Me sat.), 2.91, 2.96 (2 H each, t, CH2C_H2COZ), 3.13, 3.34, 3.35 (3 H each, 5, ring Me), 3.49, 3,59 (1 H each, d, CHZCOZ), 3.56, 3.61 (3 H each, s, CHZCHZCOZQH3), 3.71 (3 H, s, CHZCOZ_C_H3), 3.92, 3.93 (2 H each, t, CHZCHZCOZ), 4.73 (2 H, s, C_HZCOZ), 9.09, 9.12, 9.23, 9.56 (1 H each, s, meso 5, Q, Y, B), -3.22 (2 H, br s, NH); UV-vis Xmax (rel. int.) 391 nm (1.00), 494 (0.11), 540 (0.04), 587 (0.05), 641 (0.26); MS found 685.7654 for (M+H)+, C37H41N409 requires 685.7607. Trimethyl-4-hydroxy-3,4-(Y-lactone)-1,3,5,8-tetramethylchlorin-Z- acetate-6,7-dipropionate (114) Yield, 68 mg (5%); 1H-NMR 5 2.36 (3 H, s, Me sat.), 3.03 (4 H, t, CH2_C_H2COZ), 3.28, 3.40, 3.56 (3 H each, 5, ring Me), 3.53, 3.70 (1 H each, d, QHZCOZ), 3.59, 3.63 (3 H each, s, CHZCH2C02C_H3), 3.81 (3 H, s, CH2C02CE3), 4.91, 5.01 (1 H each, d, C_H_2COz), 9.14, 9.17, 9.51, 9.77 (1 H each, s, meso Q, 0, Y, 5), -2.92 (2 H, br 5, NH); MS found m/ e 685.7683 for (M+H)+, C37H41N409 requires 685.7607. 1,3-Dione 52, porphyrinone lactone 119, hydroxymethyl porphyrinone 119,p_orphyrinone g-hydroxyacetate 129, 1,7-dione 121, 1,8-dione 122, 1.5-dione 123, 1,6-dione 124 and porphflinone acrylate 125. Porphyrinone 111 (1.75 g, 2.5 mmol) was dissolved in 150 ml of CHC13 and 20 ml of MeOH and the solution was brought to reflux for 5 min before 1.0 g of Zn(OAc)2 and 0.1 g of NaOAc in 10 ml of MeOH were added. After about 30 min refluxing, (the reaction was monitored by TLC test since the Rf 109 value of the green-colored Zn(II) complex is smaller than that of the brown-colored starting compound), the mixture was cooled, washed with water (3 x 100 ml), dried under vacuum, and crystallized from CHZCHZ-MeOH in a virtually quantitative yield. To the solution of the above zinc complex in 100 ml of CH2C12 plus 5 ml of pyridine, 960 mg (3.75 mmol) of 0504 was added, and the reaction was allowed, as usual, to proceed in the dark, under argon for 30 h at room temperature. The reaction mixture was quenched with 50 m1 of MeOH, bubbled with H25, and then filtered on celite. The filtrate was dried under vacuum and chromatographed quickly on a silica gel column protecting form light. Argon was used to pack the column and flush the diols out after the front moving unreacted zinc porphyrinone (916 mg, 46%) had been eluted. To the slower moving fractions, dilute HCl solution (10%, 200 ml) was added and the solution was shaken for a few minutes (to remove the zinc) and then washed with water. The porphyrinone diols were separated to a large extent by using a chromatotron (0~5 % MeOH/ CH2C12 gradient) under argon to give about 18% of the violet diol 115, and the green colored diol 116 and 117 in a total yield of 27%. The foregoing diol 115 (354 mg, 0.45 mmol) was treated with 10 ml of sulfuric acid containing 1 ml of fluorosulfonic acid. the mixture was stirred in the dark for 2 h before being chilled in a dry-ice/ acetone bath. To this solution was added carefully 100 ml of cold methanol and the solution was let to stand at room temperature for 6 h. Further diluted with 200 ml of CH2C12, the solution was partitioned and washed with aqueous sodium acetate (25%, 2 x 200 m1) and then washed with water (3 x 200 ml). The solvent was removed by vacuum and the residue was chromatographed on TLC plates (1% MeOH/CHZClz) to give 1,3-dione 59, in a total yield of 38 mg 110 (12%), consisting of the cis and trans isomers, along with regenerated porphyrinone 123 (62 mg, 20%), Y-lactone 118 (24 mg, 8%), Q-hydroxyacetate 119 ( 45 mg, 15%) and Q-hydroxymethyl derivative 120 (13 mg, 4%). The cis and trans isomers of 1,3-dione 59 were further separated on TLC plate, developed by 5% EtOAc/CHzClz or 1% MeOH/CH2C12 for a prolonged time, in the dark, until the two green bands were separated. The faster moving band was attributed to the trans isomer 59b (18 mg), while the slower one, cis 59a (16 mg). When the mixture of Zn(II) diol 115, 116, and 117 were treated with HZSO4-FSO3H (9:1) directly without demetallation and prior separation, more than nine compounds were isolated from the reaction mixture. Among them, four known compounds: 1,3-done 59 (6%), Y -lactone 118 (4%), porphyrinone hydroxyacetate 119 (7%), hydroxymethyl porphyrinone 120 (2%) were obtained approximately in the same ratio as that in the previous reaction, five other compounds: 1,7-dione 133 (17%), 1,8-dione 134 (6%), 1,5-dione 123 (4%), 1,6-dione 124 (9%) and porphyrinone 7-acrylate 125 (3%) were identified as the major products. Tetraethyl-3,4-dihydroxy-2,3,5,8-tetramethylporphyrinone-2,4-diacetate -6,7-dipropionate (115) 1H-NMR 5 1.64, 1.69 (3 H each, is, Me sat.), 2.92, 3.00 (3 H each, s, CH2COZQI_-I_3), 2.95 (4 H, m, CH2_C_H_2COZ), 2.90, 3.23 (3 H each, s, ring Me), 3.43, 3.54 (1 H each, d, _C_HZCOZ), 3.53 (2 H, s, C_HZCOZ), 3.59, 3.60 (3 H each, s, CHZCH2C02_C_H3), 7.46, 7.79, 8.70, 8.83 (1 H each, s, meso Q, 0, 5, Y). UV-vis Xmax (EM) 400 nm (75200), 415 (88500), 436 (91300), 540 (9700), 584 (14500), 592 (14200), 638 (15500); MS found m/e 733.7894 for (M+H)+, C38H45N40ll requires 733,8025. Cis-tetramethyl-2,4,5,8-tetramethyl-1,3-porphyrindione-2,4-diacetate-6,7- 111 dipropionate (59a) 1H-NMR 5 1.83, 1.85 (3 H each, s, Me sat.), 3.10 (4 H, m, CH2_C_H_2COZ), 3.07, 3.13 ( 3 H each, s, CH2C02C_H3), 3.26, 3.28 (3 H each, s, ring Me), 3.57, 3.61 (3 H each, s, CHZCHZCOZCH3), 3.75, 3.76 (1 H each, s, C_HZCOZ), 3.75, 3.87 (1 H each, d, QHZCOZ), 4.15 (4 H, t, CHZCHZCOz), 8.46, 8.87, 9.38, 9.56 (1 H each, s, meso 0, Q, 5, Y), -0.21 (2 H, br 8, NH); UV-vis hmax (EM) 417 nm (99500), 437 (98000), 542 (9700), 591 (15600), 637 (17000), MS found m/ e 715.7832 for (M+H)+, C38H43N4OIO requires 715.7872. Trans-tetramethyl-2,4,5,8-tetramethyl-1,3-porphyrindione-2,4-diacetate- 6,7-dipropionate (59b) 1H-NMR 5 1.80, 1.81 (3 H each, s, Me sat.), 3.07, 3.11 (2 H each, t, CHZQHZCOz), 3.17, 3.19 (3 H each, s, CH2C02C_H_3), 3.24, 3.28 (3 H each, s, ring Me), 3.58, 3.61 (3 H each, s, CHZCHZCOzgfl3), 3.77, 3.78 (1 H each, s, QIiZCOZ), 3.79, 3.90 (1 H each, d, QLLZCOZ), 4.11, 4.12 (2 H each, t, QZCHZCOZ), 8.40, 8.61, 9.34, 9.50 (1 H each, s, meso 0, Q, 5, Y), 0.15 (2 H, br 5, NH); UV-vis Xmax (EM) 416 (98000), 437 (92000), 544 (9700), 589 (15700), 638 (16900); MS found m / e 715.7826 for (M+H)+, C38H43N40lo requires 715.7872. Trimethyl-4-hydroxy-3,4-(Y-lactone)-2,3,5,8-tetramethylporphyrinone -2-acetate-6,7-dipropionate (118) 1H-NMR 5 1.75, 2.14 (3 H each, s, Me sat.), 2.86, 2.89 (2 H each, t, CHZCEZCOZ), 3.01, 3.07 (3 H each, s, ring Me), 3.17 (3 H, s, CHZCOZQH3), 3.42, 3.52 (1 H each, d, C_HZCOz), 3.53, 3.54 (3 H each, s, CHZCHZCOZQ13), 3.70 (2 H, s, CHZCOZ), 3.72 (4 H, m, C_HZCHZCOZ), 7.86, 8.39, 9.00, 9.05 (1 H each, s, meso Q, B, 5, Y), 0.85 (2 H, br s, NH); UV-vis Xmax (rel. int.) 379 (0.78), 391 (0.86), 411 (1.00), 503 (0.09), 543 (0.11), 587 (0.13), 634 (0.21); MS found m/ e 667.7489 for (M+H)+, C37H40N40lo requires 667.7453. Tetramethyl-2,3,5,8-tetramethylporphyrinone-Z-acetate-4-(Q-hydroxy)- acetate-6,7-dipropionate (119) 1H-NMR 5 1.97, (3 H, s, Me sat.), 2.97 (3 H, s, 112 CH2C02CE3), 3.17, 3.22 (2 H each, t, CH2C_H2COZ), 3.41, 3.57, 3.64 (3 H each, 5, ring Me), 3.63 (6 H, s, CHZCHZCOZCH3), 3.75 (3 H, s, CHZCOZQ-I_3), 3.92, 4.01 (1 H each, d, _Ci-IZCOZ), 4.18, 4.35 (2 H each, t, CH2CH2COZ), 6.78 (1 H, s, C_HOH), 9.18, 9.93, 9.94, 10.06 (1 H each, s, meso Q, 5, Y, 0), -2.99 (2 H, br s, NH); UV-vis Xmax (EM) 406 (185000), 506 (13000), 543 (12400), 591 (6000), 647 (35600); MS found m/ e 715.7898 for (M+H)+, C38H43N4OIO requires 715.7872. Tetramethyl-B-hydroxymethyl-1,5,8-trimethylporphyrinone-2,4-diacetate -6,7-dipropionate (120) lH-NMR 5 1.93 (3 H, s, Me sat.), 2.95 (3 H, s, CHZCOZC_H3 sat.), 3.22, 3.24 (2 H each, CHZCfizCOZ), 3.57, 3.61 ( 3 H each, s, ring Me), 3.65, 3.74 (3 H each, s, CHZCHZCOZQH3), 3.76 (3 H, s, CH2C02QH3), 3.90, 4.00 (1 H each, d, CH CHZCOZ), 5. 75 (2 H, s, _C_HZOH), 9.10, 9.90, 10. 02,10.09 ( 1 H each, s, meso Q, 0, 5, Y), -2.87, -2.95 ( 1 H each, br 5, NH); UV-vis Xmax (rel. int.) 4.05 nm (1.00), 505 (0.11), 541 (0.10), 590 (0.08), 646 (0.26); MS found m/ e 715.7837 for (M+H)+, C38H43N 4010 requires 715.7872. Tetramethyl-2,3,5,8-tetramethyl-1,7-porphyrindione-2,4-diacetate-6,7- dipropionate (121) 1H-NMR 5 1.60, 2.20 (1 H each, m, CLIZCHZCOZ), 1.81, 1.83 (3 H each, 5, Me sat.), 2.87, 3.07 (2 H each, t, CHZQHZCOZ), 315 (3 H, s, CH2C02§fl3 sat.), 3.26, 3.37 (3 H each, 8, ring CH3), 3.41 (3 H, s, CHZCHZCOZCH3 sat.,) 3. 70 (3 H, s, CHZCHZCOZCH3), 3. 75 (3 H, s, CHZCOZCH3), 3.79 (2 H, s, QZCOZ), 4.07 (2 H, t, C_HZCHZCOZ), 4.76 (2 H, s, C_HZCOZ), 8.42, 8.68, 9.23, 9.48 (1 H each, s, meso Q, 5, Y, (3), 0.25 (2 H, br s, NH); UV-vis Xmax (rel. int.) 416 nm (1.00), 437 (0.84), 546 (0.12), 587 (0.18), 639 (0.20); MS found m/ e 715.7903 for (M+H)+, C38H43N40lo requires 715.7872. 113 Tetramethyl-2,3,5,7-tetramethyl-1,8-porphyrindione-2,4-diacetate-6,7- dipropionate (122) 1H-NMR 5 1.70, 2.40 (1 H each, m, QfizCHZCOZ sat.), 1.88, 1.97 (3 H each, s, Me sat.), 2.96 (2 H, t, CHZQHZCOZ sat.), 3.10 (3 H, s, CHZCOZC_H3 sat.), 3.30 (2 H, t, CHZQHZCOZ), 3.34 (3 H, s, CHZCH2C02QH3 sat.), 3.44, 3.54 (3 H each, s, ring Me), 3.62 (3 H, s, CHZCHZCOZCHB), 3.77 (3 H, s, CH2C02C_H3), 3.81, 3.83 (1 H each, d, Q-IZCOZ), 4.18 (2 H, t, Q—I_2CH2COZ), 4.96 (2H, s, CHZCOz), 886, 9.,91 964, 9.81 (1 Heach, s, meso Y,Q, 5, p); UV-vis Xmax (rel. int.) 413 nm (1.00), 435 (0.80), 545 (0.05), 590 (0.09), 619 90.16), 685 (0.04); MS found m / e 715.7816 for (M+H)+, C38H43N40lo requires 715.7872. Tetramethy1-2,3,6,8-tetramethyl-1,5-porphyrindione-2,4—diacetate-6,7- dipropionate (123) lH-N MR 5 1.59, 2.10 (1 H each, m, _C_H_2CH2COZ), 1.89, 2.02 (3 H each, Me sat.), 3.01 (2 H, t, CHZQHZCOZ sat.), 3.05 (3 H, s, CH2C02_C_H3 sat.), 3.14 (2 H, t, CH2C_H2COZ), 3.28 (3 H s, CHZCHZCOZQH3 sat.), 3.49, 3.52 (3 H each, 5, ring Me), 3.62 (3 H, s, CHZCHZCOzCfla sat.), 3.81 (3 H, s, CHZCOZC_H3), 3.87, 3. 96 (1 H each, d, C_HZCOZ), 4. 25 (2 H, t, C_I_-I_2CH2COZ), 4. 97 (2 H, s, C_HZCOZ), 9.04, 9.11, 9.66, 9.77 (1 H each, s, meso Y, Q, 5, p), -2.71 (2 H, br s, NH); UV-vis kmax (rel. int.) 409 nm (1.00), 482 (0.04), 509 (0.06), 551 (0.06), 622 (0.04), 651 (0.05), 685 (0.50); MS found in / e 715.7865 for (M+H)+, C38H43N40lo requires 715.7872. Tetramethyl-Z, 3, 5 ,8-tetramethyl-1 ,,6-porphyrindione-2 4-diacetate-5,7- dipropionate (124) 1H-NMR 5 1. 57, 2. 21 (1 H each, m, CHZCHZCOZ sat), 1.,97 1.98 ( 3 H each, s, Me sat.), 2.97 (2 H, t, CHZQflzCOz) sat.), 3.05 (3 H, s, CH3), 3.16 (2 H, t, CH2_C_H_2COZ), 3.32 (3 H, s, CHZCHZCOZC_H3), 3.47 (6 H, 5, ring Me), 3.72 (3 H, s, CHZCHZCOZC_H3), 3.77 (3 H, s, CHZCOZCfi3), 3.85, 3.87 (1 H each, d, QHZCOZ sat.), 4.23 (2 H, t, C_HZCHZCOZ), 4.87 (2 H, s, _CflzCOz), 8.87, 8.93, 9.60, 9.74 (1 H each, s, meso p, Q, Y, 5), -2.14, -2.21 (1 H each, br s, 114 NH); UV-vis Xmax (rel. int.) 399 nm (0.40), 420 ( 1.00), 483 (0.04), 549 ( 0.06), 616 (0.05), 645 (0.04), 677 (0.27); MS found m/ e 715.7832 for (M+H)+, C38H43N4OIO requires 715.7872. Tetramethyl—2,3,5,8-tetramethylporphyrinone-2,4-diacetate-7-acry1ate-6- propionate (125) 1H-NMR 5 1.95 (3 H, s, Me sat.), 2.99 (3 H, s, CH2c02c_H3), 3.15 (2 H, t, CHZCflzCOZ), 3.40, 3.54 (3 H each, 5, ring Me), 3.67 (6 H, s, ring Me and CHZCH2C02C_H3), 3.78 (3 H, s, CHZCOZQHS), 3.89, 3.97 (1 H each, d, QflzCOz), 4.05 (3 H, s, CH=CHCH3), 4.16 (2 H, t, _C_I-_12COZCH3), 4.98 (2 H, s, QizCOz), 7.12, 9.23 (1 H each, d, _CI_-I_=_C_H_COZ), 9.08, 9.90, 9.94, 9.95 (1 H each, s, meso Q, 0, 5, Y), -2.50 (2 H, br s, NH); UV-vis Xmax (rel. int.) 409 nm (1.00), 509 (0.08), 546 (0.09), 599 (0.05), 658 (0.24); MS found 697.7743 for (M+H)+, C38H41N409 requires 697.7719. £15311 128a, trans-d1 128b, cis-iso-dl ,1;9_aLtrans-iso-d11_22h Osmium tetroxide (14 mg, 0.056 mmol) was added to a solution of cis-1,3-dione 59a (20 mg, 0.028 mmol ) in 10 ml of dry CH2C12 and 0.2 ml of pyridine. The mixture was stirred, in the dark, under argon, at room temperature for 20 h before being diluted with 10 ml of MeOH and quenched by H25 gas The precipitated black osmium sulfide was removed by filtration on celite. The solvent was evaporated and the residue was chromatographed on a small silica gel column (2 x 10 cm). The column was first eluted with 1% MeOH/CHZCIZ to collect the fast moving green band, the starting dione 59a (8 mg, 40%), then with 2% MeOH/CH2C12 to obtain the slower moving grey-colored diols. The diol containing fractions were brought to dryness under vacuum and the residue was redissolved in 10 ml of benzene. To the refluxing benzene solution, 5 drops of concentrated hydrochloric acid was 115 added, and the grey-colored solution turned gradually into bright green indicating the dehydration had occurred. The reaction was monitored by TLC until the grey-colored diols had become undetectable (about 30 min). The solvent was evaporated under vacuum and the residue was chromatographed on a preparative silica gel TLC plate, developed with 1% MeOH/CHzClz. Two bands were observed on plate: the faster moving bright green band as the major product turned out to be cis-91 128a (8 mg, 67%), whereas the slower moving pigment with a deeper green color was identified as cis-iso-d1 129a (3 mg, 25%). The trans-1,3-dione 59b (18 mg) was treated 0504 and worked up exactly as described above, giving 6 mg of trans-d] 128b, 2 mg of trans-iso-gl 129b, and 9 mg of the starting material recovered. Cis-dl(Tetramethy1-2,4,5,8-tetramethyl-1,3-porphyrindione-cis-2,4- diacetate-6—acrylate-7-propionate) (128a=4) 1H—NMR 5 1.77, 1.79 (3 H each, s, Me sat.), 3.03 (2 H, t, CH2C_H_2COZ), 3.13, 3.17 (3 H each, s, CHZCOZCH3), 3.22, 3.30 (3 H each, 5, ring Me), 3.61 (3 H, s, CHZCHZCOZC_H_3), 3.73 (3 H, s, CHZCOZCl-I3), 3.70, 3.78 (1 H each, (1, 1:17 Hz, C_HZCOZ), 3.75 (2 H, s, QHZCOZ), 4.00 (3 H, s, CH=CHCOZCH3), 4.04 (2 H, t, CHZCHZCOZ), 6.90, 8.96 (1 H each, d, J=16 Hz, QH=C_I:I_COZ), 8.26, 8.42, 9.21, 9.42 (1 H each, s, meso p, Q, 5, Y), UV-vis Xmax (rel. int.) 422 nm (1.00), 446 (0.72), 610 (0.29), 660 (0.17); MS found m/ e 713.7745 for (M+H)+, C38H41N40lo requires 713.7713. Trans-d] (Tetramethyl-2,4,5,8-tetramethyl-1,3-porphyrindione-trans-2,4- diacetate-6-acrylate-7-propionate) (128b) 1H-NMR 5 1.74, 1.78 (3 H each, 5, Me sat.), 3.04 (2 H, t, CH3C_HZCOZ), 3.21, 3.22 (3 H each, s, CHZCOZCH3), 3.22, 3.29 (3 H each, s, ring Me), 3.62 (3 H, s, CHZCHZCOZCHB), 3.71, 3.79 (1 H each, d, J=17, C_HZCOZ), 3.77 (2 H, s, C_HZCOZ), 3.99 (3 H, s, CH=CHCOZQI_:I_3), 4.04, (2 H, t, 116 CH2C_H2COZ), 6.88, 8.93 (1 H each, d, J=16 Hz, _C_I-I=_C_HCOZCH3), 8.20, 8.36, 9.18, 9.38 (1 H each, s, meso (3, Q, 5, Y); UV-vis Xmax (rel. int.) 423 nm (1.00), 445 (0.70), 611 (0.32), 660 (0.71); MS found m/ e 713.7756 for (M+H)+, C38H41N40lo requires 713.7713. Cis-iso-g1 (Tetramethyl-2,4,5,8-tetramethyl-1,3-porphyrindione—cis-2,4- diacetate-7-acrylate—6-propionate) (129a) 1H-N MR 5 1.77, 1.78 (3 H each, 5, Me sat.), 3.05 (2 H, t, CH2C_H2COZ), 3.13, 3.15 (3 H each, s, CHZCOZQH3), 3.17, 3.38 (3 H each, s, ring Me), 3.64 (3 H, s, CHzCHZCOzC_H3), 3.71 (2 H, s, QHZCOZ), 3.72, 3.80 (1 H each, d, J=17 Hz, C_HZCOZ), 3.99 (3 H, s, CH=CHCOZ_C_H_3), 4.02 (2 H, t, C_HZCHZCOZ), 6.93, 9.02 (1 H each, d, J=16Hz, gH=c_Hcoz), 8.21, 8.38, 9.26, 9.40 (1 H each, s, meso p, Q, 5, Y); UV-vis Xmax (rel. int.) 419 nm (1.00), 443 (0.75), 537 (0.12), 573 (0.14), 603 (0.15), 651 (0.19); MS found m/ e 713.7729 for (M+H)+, C38H41N4OIO requires 713.7713. 'l‘rans-iso-gi1 (Tetramethyl-2,4,5,8-tetramethyl-1,3-porphyrindione-trans- 2,4-diacetate-7-acrylate-6-propionate) (129b) 1H-NMR 5 1.74, 1.75 (3 H each, s, Me sat.), 3.05 (2 H, t, CH2C_H_2COZ), 3.13, 3.21 (3 H each, s, CHZCOZQH3), 3.22, 3.36 (3 H each, s, ring Me), 3.64 (3 H, s, CHZCHZCOZCH3), 3.73 (2 H, s, C_HZCOz), 3.73, 3.80 (1 H each, s, meso p, Q, 5, Y); UV-vis Xmax (rel. int.) 420 nm (1.00), 443 (0.70), 538 (0.09), 573 (0.15), 604 (0.17), 652 (0.19); MS found m/e 713.7784 for (M+H)+, C38H41N4OIO requires 713.7713. F. Coproporphyrin IV system (The work-up procedures were similar to the 1,3-dione 59, 82 and 91 synthesis except those otherwise described.) 4,4'-BiSig-methoxycarbonylethyD-3,3',5,5'-tetramethy l-2,2'-dipflryl- 117 methenium bromide (132) t-Butyl-3-(2-methoxycarbonylethyl)-2,4-dimethylpyrrole-S-carboxylate (131) (28.1 g, 0.1 mol) was treated with 14 ml of hydrobromide acid in 50 ml of formic acid to effect the self-condensation and worked up in the same way as described in the preparation of dipyrrylmethenium 76. Compound 132 was obtained in a yield of 18.1 g (80%). MP, 209-211 °C; MS (direct probe, 70 eV), 373 for (M+). Coproporgmyrin IV tetramethyl ester The above dipyrrylmethenium bromide 132 (5.0 g, 11 mmol) was condensed with dipyrrylmethenium bromide 66 (5.8 g, 10 mmol) in 50 ml of formic acid containing 5.2 ml of bromine under refluxing. Following the same procedure as described in the synthesis of porphyrin 60 and 61 led to the formation of coproporphyrin IV in a yield of 3.3 g (46.5%). MP, 173-174 °C (lit. 174°C). 1H-nmr 5 3.22, 3.29 (4 H each, t, CH2C_H2COZ), 3.52, 3.61 (6 H each, s, ring Me), 3.68 (12 H, s, COZCH3), 4.31, 4.40 (4 H each, t, _C_HZCHZCOZ), 9.89 (1 H, s, meso Q), 9.94 (2 H, s, meso (3 and 5), 10.01 (1 H, s, meso Y), -3.98 (2 H, s, br. NH); UV-vis Xmax (rel. int.) 399 nm (1.00), 498 (0.10), 531 (0.07), 567 (0.06), 620 (0.05); MS (direct probe, 70eV) 710 m/ e for (M+). Co ro o h 'none tetrameth lester 1 The reaction of coproporphyrin IV tetramethyl ester (3.5 g, 5 mmol) with osmium tetroxide (1.52 g, 6 mmol) gave the north diol 133, which was identified as the faster moving band on TLC plate, in a yield of 1.45 g (39%) and the south diol 0.97 g (26%). Pinacolic rearrangement of 133 was carried out in sulfuric acid for 2 h, forming the porphyrinone 135 as the major product (1.06g, 75%). 1H-NMR 5 1.50, 2.08 (1 H each, m, Q—lzCHzCOz), 2.09 (3 118 H, 5, Me sat.), 3.06 (2 H, t, CH2_C_H_2COZ sat.), 3.23 (6 H, t, CH2C_1112COZ), 3.27 (3 H, s, COz(_I_H3 sat.), 3.47, 3.56, 3.58 (3 H each, s, ring Me), 3.64, 3.67, 3.69 (3 H each, s, COZQ13), 4.23, 4.37, 4.38 (2 H each, t, _C_H_2CH2COZ), 9.14, 9.84, 9.85, 9.60 (1 H each, s, meso Q, 5, p, Y), -2.91, -3.02 (1 H each, br s, NH); NOE, a selective irradiation of the ring-B Me (3.47 ppm) caused the adjacent Q-proton at 9.14 ppm to increase in intensity (6%); UV-vis Xmax (rel. int.) 405 nm (1.00), 507 (0.08), 546 (0.08), 586 (0.05), 613 (0.03), 642 (0.21); MS found 727.8445 for (M+H)+, C40H47N409 requires 727.8419. 1,3-CoproporphflinglV)-dione dimethyl ester (199) Zinc insertion of porphyrinone 135 was carried out with Zn(OAc)2 in CHCl3-MeOH as described before. Osmium tetroxide oxidation of 1 mmol (791 mg) of Zn(II) porphyrinone and subsequent rearrangement in sulfuric acid-fluorosulfonic acid (9:1) resulted in the formation of 1,3-dione 139 (cis and trans isomers) as the major product (237 mg, 32% from porphyrinone 135), together with small amounts of 1,7-dione 140 (29 mg 4%), 1,4-dione 142 (22 mg, 3%), 1,8-dione 141 (25 mg, 3%), 7-acry1ate derivative 143 (35 mg, 5%) and other by-products. 1,3-Coproporphyrin(IV)-dione tetramethyl ester (139) 1H-N MR 5 1.74, 2.13 (2 H each, m, _C_HZCHZCOZ sat.), 1.93, 1.95 (3 H each, 5, Me sat.), 2.91, (4 H, m, CH2_CH2co2 sat.), 3.10, 3.12 (2 H each, t, CHngZCOZ), 3.33, 3.46 (3 H each, s, ring Me), 3.40, 3.43 (3 H each, s, CHZCHZCOZC_H3 sat.), 3.62, 3.71 (3 H each, s, CHZCHZCOZQH3), 4.12, 4.19 (2 H each, t, CHZCHZCOZ), 8.58, 8.77, 8.33, 9.59 (1 H each, s, meso [3, Q, 5, Y, ), -0.72 (2 H, br 5, NH); NOE, irradiating the 5-proton at 8.39 ppm caused the Me singlet (3.46 ppm) to increase in intensity. UV-vis Xmax (rel. int.) 403 nm (0.73), 419 (0.96), 439 (1.00), 543 (0.10), 593 (0.16), 636 119 (0.20); MS found m / e 743.8452 for (M+H)+, C40H47N40lo requires 743.8413. 1,7-Coproporphyrin(IV)-dione tetramethyl ester (140) 1H-N MR 5 1.70, 2.14 (2 H each, m, QflzCHzCOZ sat.), 1.93, 1.94 (3 H each, 5, Me sat.), 2.91 (4 H, m, CHZQHZCOZ sat.), 3.13, 3.15 (2 H each, t, CH2C_H2COZ), 3.31, 3.35 (3 H each, 5, ring Me), 3.41, 3.42 (3 H each, s, CHZCHzCOZQIi3 sat.), 3.59, 3.62 (3 H each, s, CHZCHZCOZC_H3), 4.14, 4,18 ( 2 H each, t, C_HZCHZCO), 8.56, 8.74, 9.35, 9.66 (1 H each, s, meso Q, 5, Y, (3); NOE, irradiating the Y-proton at 9.35 ppm caused the triplet at 4.18 ppm to increase in intensity. UV-vis Xmax (rel. int.) 418 nm (0.94), 439 (1.00), 545 (0.13), 592 (0.18), 637 (0.21); MS found m/ e 743.8743 for (M+H)+, C40H47N40lo requires 743.8413. 1,8-Coproporphyrin(IV)-dione tetramethyl ester (141) 1H-N MR 5 1.62, 2.18 (2 H each, m, C_HZCHZCOZ sat.), 1.99, 2.00 (3 H each, s, Me sat.), 3.00 (4 H, m, CHZCHZCOZ sat.), 3.16, 3.19 (2 H each, t, CHZQHZCOZ), 3.34, 3.37, 3 H each, 5, ring Me), 3.36, 3.47 (3 H each, s, CHZCHZCOZQH3 sat.), 3.62, 3.64 (3 H each, s, CHZCHZCOZQH3), 4.23, 4.34 (2 H each, t, gfizCHzCOZ), 8.94, 8.99, 9.65, 9.86 (1 H each, s, meso Q, Y, 5, B), -1.80 ( 2 H, br 5, NH); UV-vis Xmax (rel. int.) 416 nm (1.00), 435 (0.78), 592 (0.08), 624 (0.17), 690 (0.04); MS found m/ e 743.8345 for (M+H)+, C40H47N40lo requires 743.8413. 1,4-Coproporphyrin(IV)-dione tetramethyl ester (142) 1H-NMR 5 1.70, 2.25 (2 H each, m QHZCHZCOZ sat.), 1.83 (6 H each, 8, Me sat.), 3.02 (4 H, t, CH2C_H2COZ sat.), 3.12 (6 H, s, ring Me), 3.38 (6 H, s, CHZCHZCOZC_H3 sat.), 3.61 (6 H, s, CHZCHZCOZQHQ, 3.95 (4 H, t, QLIZCHZCOZ), 7.57, 9.25 (l H each, s, meso Q, Y), 8.88 (2 H, s, meso (3 and 5), 0.83 (2 H, br s, NH); UV-vis hmax (rel. int.) 406 nm (1.00), 424 (0.43), 523 (0.07), 560 (0.11), 599 (0.13), 651 (0.33); 120 MS found m / e 743.8469 for (M+H)+, C40H47N40lo requires 743.8413. 7-Acrylo-coproporphyrin(IV)—one tetramethyl ester (143) 1H-NMR 5, 1.60, 2.15 (1 H each, m, _C_HZCHZCOZ sat.), 2.08 (3 H, s, Me sat.), 3.08 (2 H, m, CH2C_H2COz sat.), 3.18 (4 H, m, CHZQHZCOZ), 3.25, (3 H, s, CHZCH2C02_C_H_3 sat.), 3.61 (6 H, 5, ring Me), 3.64 (3 H, s, ring Me), 3.62, 3.72 (3 H each, s, CHZCHZCOZQ-I_3), 4.03 (3 H, s, CH=CHCOZQH3L 4.32 (4 H, m, C_HZCHzCOZ), 6.99, 9.23 (1 H each, d, C_H=§_HCOZ), 9.17, 9.74, 9.94, 10.05 (1 H each, s, meso Q, 9, 5, Y), -2.81, -2.87 (1 H each, s, NH); UV-vis Xmax (rel. int.) 412 nm (1.00), 512 (0.10), 549 (0.09), 597 (0.06), 657 90.24); MS m/ e 725.8213 for (M+H)+, C40H45N409 requires 725.8260. Homo-d1 (6-Acrylo-1,3-copropogphyg’ndV)-dione tetramethyl ester) (139) 1,3-Dione 139 (75 mg, 0.1 mmol) was treated with OsO4 (38 mg, 0.15 mmol) under the condition as described previously, yielding 36 mg (48%) of 130, together with 3 mg of hydroxymethyl dione (145). (130) 1H-NMR 1.65, 2.20 (2 H each, m QIiZCHzco2 sat.), 1.88, 1.93 (3 H each, Me sat.), 2.83 (4 H, m, CHZCHZCOZ, sat.), 3.26 (2 H, t, CHzgizCOz), 3.36, 3.38 (3 H each, s, CHZCHZCOZQH3 sat.), 3.40, 3.43 (3 H, each, ring Me), 3.70, (3 H, s, CHZCHZCOZQH3), 4.00 (3 H, s, CH--CHC02§_H_3), 4.03 (2 H, t, C_HZCHZCOZCOz), 6.90, 9.40 (1 H each, d, Qfl=C,_I-_ICOZ), 8.45, 8.55, 9.18, 9.51 (1 H, s, meso p, Q, 5, Y); UV-vis Xmax (rel. int.) 422 nm (1.00), 446 (0.70), 608 (0.28), 658 (0.16); MS found m/ e 741.8279 for (M+H)+, C40H45N 4O 10 requires 741.8254. CHAPTER 4 ON THE STRUCTURE OF HEME d1 1. THE STERIC AND REGIO ISOMERS OF HEME $11 A. Cis- and trans-d1 -- Stereochemistry Deduced from NMR Shift Reagent The visible spectra of the diasterotropic cis-g11 128a and trans-g] 128b are virtually identical. The 1H-N MR spectra of the two have recognizable differences as shown in Table 2 and Figure 11. The peaks of meso protons of the trans isomer 128b are significantly upfield shifted and all the methyls, including those of methyl ester of the acetate, are also shifted to different positions in comparison with those of cis isomer 128a or the natural 9,. The lanthanide 1H-NMR shift reagent, such as Eu(fod)3, which can provide useful information on the configuration and conformation problems, was used to study the stereochemistry of 91 via the two precursors cis-dione 59a and trans-dione 59b. 1H-NMR were taken from a solution of either dione, to which an increasing amount of Eu(fod)3 was added until the solution was saturated, therefore no more induced shift could be observed. It is known that the chemical shifts induced by lanthanide are resulted predominately from a dipolar (pseudocontact) mechanism.91 For a porphyrin with 6,7-dipropionate substituents, the lanthanide is believed to associate with the two carbonyl oxygens, either simultaneously as a bidentate complex or, in a statistical sense, with either one or the other, thus inducing a dramatic down-field shift of the Y-meso proton.92 We inferred that the possibility of Eu(fod)3 associating with the two keto groups on the ring93 121 122 MOOgc COIN. mme emM. 128a (U) - ..2 ~- 10 3.: :.~ 3.2 3.3 2.5 2.5 2.. 2.2 2.151. rrn Figure 11 250 MHz 1H-NMR spectra of cis-g1 (128a) versus trans-(11 (128b) in CDCl3. 123 might be the same for both cis and trans isomer, but the interaction between the shift reagent and the two acetate substituents on the isomers would be different. The relative magnitude of the induced shifts of the Q-proton should establish the distance between the two acetate groups and permit the assignment of their stereochemistry. In the case of cis dione 59a, in which the two acetate groups are on the same side of the macrocycle, Eu(fod)3 should have a higher affinity and thus, the Q-meso proton should have a larger shift than that of the trans isomer. Fig1_1re 12 tabulates the Eu(fod)3 induced Al.38 Meme—s, (1.18)0 520,819 l.|3 (I.I3) MGOzc c0)". Figure 12. The Eu(fod)3 induced chemical shift (in ppm) of meso protons of the cis-dione 59a and trans 59b (in parentheses). Eu(fod)3 oonc.=0~10 mM. chemical shifts (in ppm) of meso protons of the cis-dione 59a and 59b (in parentheses). In deed, the Q-proton clearly saw the largest difference between the induced shifts, AA=1.38-1.18=0.20 ppm. In contrast, the difference observed between the isomers at the p or the 5 proton is negligible. Since 59a has the largest shift, itself and the acrylate derived from which, i. e. 128a, should have a cis conformation. Based on this result, the stereochemistry of heme dl is assigned as cis with respect to the two acetic side chains. The definite proof, of cause, must await an X-ray crystallography study. 124 B. Cis- gnd trans-iso- d1 -- Location of the Acrylic Side Chain The electronegative acrylate side chain has a prominent effect on the visible spectrum of the porphyrindione core. As shown in Figure 13, the spectrum of iso-d1 (129a) is significantly different from that of £1 (128a=4a). The bands at the Soret region have a different shape as well as a shift and the bands in visible region appear in a quite different pattern. Distinguishable 1H-NMR spectra are also observed for iso-d (Table 2). The position of the acrylate was verified by NOE measurements. For example, selective irradiation of the 8-methyl singlet (5:3.38 ppm) on ring D of cis-iso-dl (129a), resulted in a 5% increase in intensity on the neighboring 5-H singlet at 9.26 ppm, 4% and 3.5% on the acrylate double bond protons. At the time when we pr0posed the structure 1 for heme _d 1, the position of the acrylic acid in relation to the two keto groups on ring A and B was tentative, now with the side by side comparison of the spectra of 128a and 129a, combined with the result from NOE measurements, the location of this substituent is firmly established. II. ON THE NATIVE FORM OF _d_] An important question concerning the proposed structure 1 for _d_l is whether the 1,3-porphyrindione structure is the native form of the heme, or it arises from a diol by a pinacol-pinacolone rearrangement during its purification. There are several lines of evidence against a possibility of an in vitro rearrangement. It is obvious that the conditions used for the isolation of the free base methyl ester are not as harsh as is necessary for diol rearrangement in model compounds and 911 synthesis. The chiral 2 and 4 E5 2.00 1 1 I I I I I I I ‘o o a c a MOOQC C0,". 5 I .oo 129a a o a uni a < 573 603 65! 1 537 O 00 l l 4 I l l I _ 1 L00 I Absorbance b O 000 Anal 400 500 600 700 800‘ Figure 13 UV-vis absorption spectra of cis-s11 (128a) versus cis-iso-g1 (129a) in CHZCIZ. 126 Table 2. 1H-N MR chemical shifts of heme d free base and its stereo- and regioisomers. Chemical shifts at 25°C in parts/ million in CDC13 w1thCHCl3 as internal standard (5:7.24). 5(ppm) compound proton c1s- ........... trans: H# mult1phc1ty ClS-dl trans-d1 ISO-d1 ISO-d1 meso Q 8 418 8.363 8.377 8 318 1 s p 8 267 8.204 8.214 8 155 1 s g 9 429 9.384 9.403 9 347 1 s 9 211 9.180 9.258 9 214 1 s -CH=CHCOZCH3 ......... 6.9.02 ........ 6. ,8836927 ........ 6.9.09. ........ 1 ......... d ....................... J=16.2 Hz 8 962 8.934 9 019 8 991 d OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO -CHZCH2COZCH3 4 046 4 037 4.025 4 012 2 t 1:7.4 Hz 3 035 3 036 3.053 3 044 2 t 3 612 3 616 3.641 3 643 3 s 2-CH2COZCH3 .............. $35 ........ 5 7883798 ........ 3&5 ........ 1 ......... d ........................ =17.6 Hz 3.696 3 710 3.721 3 735 1 d OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 4-CH2COZCH3 3.751 3 771 3 712 3 732 2 s 3.176 3.204 3 174 3 219 3 s 5033. ............................. 17°90 ........ 1 78417841752 ........ g ........ é ......................... 4-CH3 1 776 1.735 1 774 1 747 3 s 5cH3 ............................. 3367 ........ 5 3853145 ........ 3313 ........ 5 ........ é ......................... 8-CH3 3 223 3.217 3 385 3 362 3 s 127 fi-carbons give rise to diastereomers unavoidably in porphyrindiones synthesized by rearrangement when different substituents are involved, but there is no evidence for diastereomeric resonances in the extracted d1. Similarity of the meso proton chemical shifts in the free base methyl ester and a diamagnetic Fe2+ form of Q] in a freshly prepared crude extract argues against subsequent chemical transformation.45 This argument was stringently tested by comparing the differences in meso proton shifts between free base and diamagnetic iron complexes of model compounds. As shown in Table 3, the shifts produced on going from metalloporphyrin to free base are generally small and especially small for _d_], and they do not grossly distort the profile of resonances. Above all, the synthetic Q] has been found in full function in the reconstituted _cg nitrite reductase, as will be discussed in Chapter 5, thus it leaves no doubt about the consistency of £11 structure before and after its isolation. Table 3. Meso prton chemical shift differences between hemes and free bases.a meso d1 1 ,3-OEPdione 8 protoporphyrin IX Q 0.16 0.51 0.18 p 0.12 0.66 0.26 Y -0.15 0.46 0.49 5 -0.18 0.44 0.06 aFree base shifts were in CDC13 at 20 °C. Heme shifts were in 1:1 DzO—[DS] pyridine at 20 °C. 128 111. ON THE BIOSYNTHETIC ORIGIN OF Q1 The unprecedented keto structure of heme d1 poses many challenging questions. The first would be "How is heme d1 formed in nature?" Biosynthetically, all tetrapyrrolic macrocycles found in nature derive their acid side chains and substitution patterns from a single intermediate, uroporphyrinogen III, which itself is cyclized enzymically from four units of porphobilinogen.2 In animals and plants the formation of protoporphyrin IX and chlorophylls requires the decarboxylation of uroporphyrinogen III to yield c0proporphyrinogen III, which by oxidative decarboxylation produces protoporphrinogen IX. On the other hand, an alternative pathway produces sirohydrochlorin from which another set of pigments, such as vitamin £12 and E430 are formed. Now the question is "By which route is heme d1 biosynthesized, via protoporphyrin or sirohydrochlorin (Scheme 27 ?" Since the biological existence of the gem-diol structure has been substantiated by the recently established heme d structurefiof Em it can be assumed that the keto groups in 911 could be obtained from a pinacolic rearrangement via a gem-dihydroxy derivative of protoporphyrin. Such a "protoporphyrin route" is essentially what we have followed in the laboratory synthesis. Nevertheless, owing to the difficulties we experienced in preparing d1 (extremely acidic conditions, mixture of diastereomers, etc.), we have doubt about the feasibility of this route in a natural setting. If heme 11 is derived from the "sirohydrochlorin route", its configuration should be automatically defined by the precursor. Indeed the cis stereochemistry about the two angular acetic side chains concluded by our work, as well as the overall substituent arrangement, suggests that the d1 129 biosynthesis may be just another branch of the intricate uroporphyrinogen III ~>§12 pathway. The two possibilities regarding the Q] biosynthesis may be distinguished experimentally by tracing the source of the two angular methyl groups. If the protoporphyrin-related pinacolic rearrangement is involved, the methyl groups should come from the 1,3-acetic acid side chains of uroporphyrinogen III. In contrast, if the sirohydrochlorin is the precursor, the methyl groups are transferred to uroporphyrinogen III from SAM (S-adenosylmethionine). The isotopic labeling technique, which has been successfully used in the 1112 biosynthesis,2 will produce direct evidence concerning the pathway of d1 synthesis in living system. P36 l I P A I A «_— A P \ P P P H OH Mo R P P A ?\‘ ‘7/ P P SIROHYDROCHLORIN 312 F430 Scheme 27 CHAPTER 5 PHYSICOCHEMICAL PROPERTIES OF HEME 41 AND MODEL SYSTEMS I. GENERAL CONSIDERATION Having proven the unprecedented "porphyrindione" structure of Q, we must raise questions like 'Why is this structure needed ?‘ 'Are there any obligatory roles of this structure in the dissimilatory nitrite reduction ?' An obvious approach to the answers is to compare the properties of this new macrocycle with other better known porphyrinoids. An enormous amount of effort has been devoted to document just about every conceivable properties of porphyrins. It seems that every. new spectroscopic technique introduced to chemistry and biophysics in the last three decades has been applied to porphyrins and their multitude metal complexes. We cannot, of cause, duplicate all these measurements on porphyrinones and porphyrindiones since it is unnecessary to do so. However, some preliminary results have been collected from the studies with spectroscopic techniques such as UV-vis, 1H- and 13C-NMR, IR and RR, and the molecular structures, the redox potentials and the PK3 values of these compounds have been determined. This chapter is a summary of the physicochemical studies so far carried out on the d1 and related compounds in an attempt to build up a framework for future investigation. Some results covered in this chapter are cited from literature work published previously by our group and our collaborators in order to present a general review. It is intriguing to recognize that both siroheme and heme d1 prosthetic 130 131 group have the isobacteriochlorin-type core structure, yet they differ with regard to their spectroscopic, redox and chemical properties. Porphyrinones and porphyrindiones have generally been viewed as derivatives of chlorin, isobacteriochlorin or bacteriochlorin, therefore, the differences between these two systems, whenever observed, are emphasized throughout this chapter. 11. ABSORPTION SPECTRA A. Spectral Features of Porphyrinones A typical visible spectrum of porphyrinone, that of 135, is illustrated in Figure 14, and the data taken from eighteen others, plus two hydroporphyrins as comparison, are listed in Table 4. The spectra of Cu(II)-metallated and protonated porphyrinones are given in T_abl_e_§ for general reference. Like those of their chlorin cousins, the spectra of porphyrinones are very much consistent in that the overall pattern is relatively unperturbed by electronic effects. The intense singlet near-UV band, still called Soret band, is found around 407 nm with a molar extinction efficient in the order of 1.5 to 2.0 X 105, which is 4 to 5 times more intense than the visible bands above 450 nm. The peak at the longest wavelength (band I) is observed at 643 nm a 2 nm) and is always more intense than the other bands in this region. Band II usually appears as the least intense one, whereas Band IV is broad with a shoulder on its blue side. The spectral features of porphyrinones differ from that of a typical chlorin54 in that the Soret band as well as Band I of chlorin are observed at shorter wavelength, about 15 nm toward blue, but with a similar molecular extinction coefficient as that of porphyrinone. Substituents on the porphyrinone periphery can effect the absorption 132 00m «Batu 5 mm“ ucofiismuom mo 833% zozmuomna £>->D 3 93me 62.8 66.0 0 O OO.N OOUBQJOSQV 133 Table 4. Visible spectra of various porphyrinones in comparison with chlorins. The relative intensity of the bands is indicated in parenthesis. band (nm) compound Soret IV III II I octaethylporphyrinone 26 407 510 549 586 642 (1.00) (.09) (.10) (.07) (.23) 1-mesoporphyrinone 5 407 508 547 585 642 (1.00) (.08) (.09) (.06) (.22) 3-mesoporphyrinone 18 407 508 547 585 642 (1.00) (.07) (.09) (.05) (.22) 2,3-(bis-2-chloroethyl)- 1-porphyrinone 93 407 508 547 588 645 (1.00) (.09) (.10) (.07) (.25) 1,4-(bis-2-chloroethyl)- 2-porphyrinone 94 407 508 547 587 643 (1.00) (.10) (.11) (.08) (.24) 1-porphyrinone— 2,4-diacetate 111 406 507 544 589 645 (1.00) (.09) (.10) (.07) (.23) 3—porphyrinone- 2, -diacetate 112 406 506 545 587 643 (1.00) (.09) (.09) (.07) (.23) 1- orphyrinone-Z-acetate- 4- 1-hydroxy)-acetate 120 407 506 544 591 647 (1.00) (.09) (.08) (.06) (.25) 2,4-bis-(2-chloroethyl)— 8—hydroxymethyl- 3-porphyrinone 408 508 546 590 646 (1.00) (.17) (.16) (.13) (.33) 2-(2-hydroxyethyl)- 4-vinyl-1-porphyrinone 100 412 511 550 596 652 (1.00) (.11) (.09) (.07) (.24) 4-vinyl-1-pogphyrinone -2-aecalde y e 102 412 512 548 594 651 (1.00) (.11) (.09) (.07) (.23) 1-mesoporphyrinone- 6-acrylate A 414 520 565 585 641 (1.00) (.07) (.12) (.09) (.20) Table 4 (contg) 3-mesoporphyrinone- 134 7-acrylate B 416 524 563 583 640 1 h . (1.00) (.08) (.17) (.10) (.18) -co ro or none- 6-ac1Pylafe p yn 143 417 526 565 585 641 (1.00) (.07) (.12) (.09) (.19) l-porphyrinone- 2, -diacetate-6-acrylate 415 521 563 587 643 (1.00) (.11) (.15) (.12) (.22) 2-hydroxyl-1,2—(Y-lactone)- chlorin-4-acetate 113 391 494 540 587 641 (1.00) (.11) (.04) (.05) (.26) 4-hydroxyi-3,4—(Y-lactone)- chlorin-2-acetate 114 390 493 541 587 640 (1.00) (.10) (.03) (.04) (.24) 1,2-dihydroxylchlorin- 2,4-diacetate 105 392 494 522 590 642 (1.00) (.10) (.04) (.04) (.28) 3,4-dihydroxylchlorin- 2,4-diacetate 106 392 495 521 589 643 (1.00) (.11) (.04) (.05) (.27) 1,1-dihydro-gem- octaethylchlorin c 391 495 521 589 617 643 (1.00) (.09) (.06) (.05) (.06) (.29) 1-methyl- em- octaethylc lorin D 391 496 522 589 615 643 (1.00) (.09) (.05) (.05) (.05) (.32) O H H H «60,: com. no.1: cow. A B C I-mesoporphyrin- 3—mesoporphyrin- 1,1-dihydro—gem- 1-methyl- em- one-6-acrylate one- 7-acrylate octaethylchlorin octaethylc orin 135 Table 5. Visble spectra of Cu—metallated and protonated porphyrinones in comparison with chlorins. The relative intensity of the bands is indicated in parenthesis. band (nm) compound Soret Q Cu(II)- octaethylporphyrinone 26 378 414 572 616 (.28) (1.00) (.08) (.28) Cu(II)- 3-mesoporphyrinone 18 379 414 570 616 (.25) (1.00) (.07) (.27) Cu(II)-1-methyl- gem-OEC D 388 494 530 569 612 (1.00) (.07) (.06) (.08) (.33) 3H+- octaethylporphyrinone 26 401 416 535 570 623 (1.00) (.83) (.07) (.08) (.15) 3H+- 3-mesoporphyrinone 18 399 416 535 548 569 620 (1.00) (.09) (.06) (.06) (.08) (.15) 3H+-1,1-dihydro- gem-OEC c 394 406 525 619 (1.00) (.09) (.07) (.17) 3H+-1—methyl- gem-DEC D 394 407 524 619 (1.00) (.09) (.06) (.18) —————-—-_—-——-—- :-—-- :—-—_‘ -——--——————-—-—— :—--- --: _ --— 136 spectra. An exocyclic double bond, such as a vinyl group, shifts the absorption bands to longer wavelength (up to 8 nm). Introducing a conjugated acrylic side chain would red-shift the Soret band by 7 nm and also change the pattern of visible bands: band III and IV are 16 nm red-shifted but I and II are virtually unchanged. The y-lactone compounds, such as 113 and 114, obtained as a by-product of the pinacolic rearrangement of some vic-diols have a spectrum very similar to that of a chlorin with its band III diminished and shifted to a longer wavelength. B. Spectral Features of Porphyrindiones Dominated by the relative positions of the two keto groups on the macrocycle, the isomeric diones give totally different spectra from one another. In Figure 15 and _1_6, the spectra of all five isomeric porphyrindiones are illustrated. Not only the isobacteriochlorin-type 1,3- 2,3- and 1,4-dione but all so the bacteriochlorin-type 1,5- and 1,6-dione exhibit distinct absorption spectra due to different molecular symmetries are involved in their structures. Not surprisingly even the metal complexes and protonated forms of these regioisomeric diones have quite different spectra. In Table 6, the spectral data of some Cu(II) and acid salt of diones are tabulated. As illustrated in Figure 17 and 18, the spectra of Cu(II) 1,3-dione versus Cu(II) 2,3-dione, the protonated 1,3-dione versus protonated 1,4-dione, both couples exhibit very distinct spectral features. These results suggest immediately the extensive n-overlap of the two oxo groups with the ring system, forming a marocycle skeleton different from the typical porphyrin or isobacteriochlorin system. 137 300 400 500 I‘III‘I 600 700 800 Figure 15 UV-vis absorption spectra of 2,3-dione (32), 1,3-dione (8) and 1,4-dione (33) in 138 ' moon-hum: mum-oat. ' Port No. 30029 ' n 2'00 I I I I I I I I I 409 _ .1 123 q ' o o 1: _ g Moo,c CotMg l.00 _ 3 685 o a -( < 0.00 l l L“ lmj I 2'00 I I I I T I T I I L 420 _ Home 60.”. 124 - -1 ' o 2 — ° com. — a M.°gc '2 1.00 - -+ o a _ _ 2 399 677 0.00 l l LVN A l Anm 400 500 600 700 300 Figure 16 UV-vis absorption spectra of 1,5—dione (123) and 1,6-dione (124 ) in CHZCIZ. 139 Table 6. Visible spectra of Cu(II)-metallated and protonated po hyrindiones. The relative intensity of the bands is indicated in parent esis. ——_-—— ——--—--——_—-- —- ----—- -- —--—-“_---“----—- :—--—‘--- band (nm) compound Soret Q Cu(II)-ocltaethjfl- 1,3— 0 ' one 8 391 433 501 580 621 p rp yrm (.71) (1.00) (.10) (.19) (.52) CU(II)-1,3- orphyrin- dione-2,4- ”acetate 59a 389 430 575 617 (.65) (1.00) (.18) (.54) Cu(II)-cis-dl 128a 410 438 643 (.61) (1.00) (.58) Cu(II)-octaethyl- 2,3-porphyrindione 32 395 430 441 543 581 691 (.71) (1.00) (.93) (.10) (.12) (.64) Cu(II)«octaethyl- 1.5-porphyrindione 28 381 426 524 667 701 (.48) (1.00) (.06) (.12) (.76) Cu(II)-octaethyl- 1,6-porphyrindione 29 380 420 442 542 585 666 (.80) (1.00) (.21) (.34) 3H+-octa1ethyla 413 436 596 629 1,3- 0 ione 8 p rp yrm (.80) (1,00) (.21) (.76) 3H+-octa1ethyla 413 439 618 657 2,3-porp yrm ione 32 (.80) (1.00) (.18) (.22) 3H+-octaethyl- 1,4-porphyr1ndione 33 399 420 523 568 604 648 (1.00) (.07) (.10) (.17) (.52) 3H+-octae;hyl- 1,3,5- ' trione 50 424 624 679 porp yrm (1.00) (.24) (.17) 3H+1 ,3-dime thyl-gem- octaethyl- 368 385 405 492 518 587 632 isobacteriochlorin E (.46) (.45) (1.00) (.09) (.07) (.12) (.38) ‘-- --—-—-- _— _t octaethyl- 1,3-dimethyl-gem- isobacteriochlorin Absorbance 140 ” 430 ‘ I I l I f I I I I 433 s n o _ h 8 - L- -1 620 - Ann: 400 500 600 700 800 Figure 17 UV-vis absorption spectra of Cu (11) 2,3-dione (32) and 1,3-dione (8) in CHZClz. 141 648 33 .J ‘0 o b d C 0 I l J .0 h 0 I 1 I 1 f I I fl I a a 437 < _ _ l— .1 8 - q 630 596 1 1 1 1 L 1 1 1 1 Ann1 400 500 600 700 800' Figure 18 UV-vis absorption spectra of 1,4—dione (33) and 1,3-dione (8) in Cl-{ZCI2 with CF3C02H (1%). 142 The Soret region of the 1,3-dione seems to be composed of at least four distinguishable bands: a sharpest and most intense band at 439 nm, another one at 418 nm and two shoulder bands at 402 and 388 nm. In the visible region, there are five bands observed. The molecular extinction coefficient of the Soret band (€=85,000 to 90,000) of 1,3-dione is significantly lower than that of the 2,3- and 1,4-diones. In comparison with the 1,3-dione, a typical isobacteriochlorin has a single broad Soret band at a much shorter wavelength around 376 mm. The positions of the major visible bands of isobacteriochlorin are not so much dissimilar with those of the 1,3-dione but the relative intensity of 635 nm peak is higher in isobacteriochlorin. Peripheral electronic effects are summarized in 1%!th The spectrum of the monoketone-lactone compound 118 maintains virtually the basic features of the Soret region of the 1,3-dione, but its visible bands are more scattered with Band I becoming the most intense peak. The influence of a conjugated vinyl group on the 1,3-dione spectrum is significant: a vinyl group at ring C, position 6, can diminish the multiple Soret to two bands, and can shift all visible bands about 10 nm bathchromically, giving a spectrum similar to that of £11. The presence of an acrylic side chain at this position brings about the the spectrum of $11 as we described before, having virtually a single Soret band at 422 nm with a 446 nm shoulder and all the visible bands red-shifted by 20 nm. The unusually low extinction coefficient (€=82,000) of the Soret band and the low ratio of the Soret band versus visible bands (~3.2) are the unique features of "6-acrylo-1,3-porphyrindione". III. NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 143 Table 7. Visible spectra of 1,3-porphyrindiones in comparison with isobacteriochlorins. The relative intensity of the bands are indicated in the parenthesis. band (nm) compound soret visible 1 ,3-octae thyl- porpohyrindione" 8 403 421 440 544 585 591 637 (.80) (1.00) (1.00) (.13) (.21) (.17) (.23) 1,3-mesoporphyrin- 6 402 417 438 543 584 638 dione (.86) (1.00) (.97) (.15) (.21) (.22) 2,4-bis-(2-chloroethyl) 1,3-porphyrindione" 82 418 439 545 592 637 (1.00) (.94) (.13) (.18) (.21) 2,4-bis-(2-hydroxyethyl)- 1,3-porphyrindione 88 415 437 545 588 637 (1.00) (.87) (.19) (.26) (.26) cis-1,3-porphyrindione- 2,4-diacetate 59a 417 437 542 591 637 (1.00) (.98) (.11) (.17) (.19) trans-1,3-porphyrindione— 2,4-diacetate 59b 416 437 544 589 638 (1.00) (.82) (.11) (.16) (.17) 5-(1-methoxylethyl)- 1,3-porphyrindione 52 402 419 439 546 588 639 (.80) (1.00) (.87) (.17) (.25) (.21) 5-(1-hydroxylethyl)- 1,3-porphyrindione 54 401 418 439 548 587 640 (.82) (1.00) (.93) (.17) (.28) (.25) Table 7 (contd) 144 6-vinyl-octaethyl- 1,3-porphyrindione 53 422 444 554 596 646 (1.00) (.82) (.17) (.28) (.22) 1 ,3-mesoporphyrindione- 6-acrylate 10 423 445 611 661 (1.00) (.71) ' (.32) (.21) cis-d1 128a 422 446 610 660 (1.00) (.72) (.29) (.17) trans-d1 128b 423 445 611 660 (1.00) (.70) (.32) (.71) cis-iso-dl 129a 419 443 537 573 603 651 (1.00) (.75) (.12) (.14) (.15) (.19) tran-iso-dl 129b 420 443 538 573 604 652 (1.00) (.70) (.15) (.15) (.17) (.19) 4-hydroxyl-3,4-(Y -lactone)- 1-porphyrinone- 2,4-diacetate 118 379 391 411 503 543 587 634 (.81) (.86) (1.00) (.09) (.11) (.13) (.21) 1 ,3-dimethyl-gem- octaethyl- 373 511 546 589 636 isobacteriochlorin E (1.00) (.16) (.25) (.38) (.09) sirohydrochlorin 377 482sh 513 547 589 635 (1.00) (.07) (.10) (.18) (.28) (.02) —- I 145 A. 1H-NMR Spectra Removal of the peripheral double bonds and introduction of keto groups to the macrocycle lead to a decrease in the ring current, 'as indicated by the upfield shift of the outer meso proton signals and a down field shift of the inner N-H signals. A trend of upfield shift for almost all of the peaks of the peripheral substituents, attached either to the pyrrole or to the saturated pyrroline rings, is also observed as going from porphyrin, to porphyrinone, and to porphyrindione (M98). The decrease of ring current is moderate in porphyrinone and bacteriochlorin—type dione but is very pronounced in the isobacteriochlorin-type 1,3-dione. Comparing with the features observed for dione 59a, the peaks in the spectrum of Q] are even more upfield shifted indicating the powerful conjugation of acrylate side chain with the ring TI-system. In general, the peaks of porphyrindiones are less spread out and are often not first order. They are further complicated by the magnetic nonequivalence of the methylene protons of the side chains because of the asymmetric quaternary 0 carbon atom on the pyrroline ring. An additional complicating factor in these compounds is the possible spin-spin coupling between the substituents. The two methylene protons on the acetate side chains of porphyrin 58 exhibit only a single peak, thus two peaks are observed at 3.73 and 3.75 ppm respectively. However, in porphyrinone 111, these protons on pyrroline ring A are not only upfield shifted to ~3.95 ppm, but also split into a AB-type doublet-doublet with a J-value of 17 ppm. On pyrrole ring B, they remains as a singlet with almost no shift. In the spectra of dione 59a both set of methylene peaks appear around 3.70 ppm, with splitting only coming from ring A acetate. The splitting of the methylene protons might be caused by the 146 Table 8. 1H-NMR chemical shifts of heme d related orghyrin, porph inone and orphyrindiones. Chemical slhifts at 50 in parts/ mi lion in CDC 3 w1th CHC13 as internal standard (5:7.24 ppm). compound 5(ppm) proton PHZ MK-l MK-2 1,6-DK 1,8-DK 1,3-DK 1,7-DI< 58 111 112 124 122 59a 121 meso 10.005 9.955 9.945 9.745 9.815 9.565 9.485 9.995 9.925 9.865 9.605 9.655 9.385 9.225 9.975 9.895 9.855 8.935 8.915 8.675 8.685 9.945 9.135 9.075 8.875 8.865 8.465 8.425 {HZCHZCOZCH3 4.37t 4.38t 4.38t 4.23t 4.19t 4.15t 4.07t 4.35t 4.23t 4.23t 3.14t 3.13t 4.11t 3.25t 3.26t 3.23t 3.46t 2.97t 2.97t 3.14t 2.87t 3.26t 3,20t 3.26t 2.08m 2.21m 3.06t 2.10m 1.57m 1.70m 1.69m 3.245 3.665 3.675 3.725 3.625 3.615 3.705 3.665 3.635 3.665 3.325 3.415 3.575 3.255 oCLIzCOzCL-Ig, 4.965 5.045 5de 4.875 4.965 3.87dd 4.765 4.905 4.00dd 4.99dd 3.90dd 3.88dd 3.7de 3.795 3.90dd 3.98dd 3.82dd 3.80dd 3.76s 3.88dd 3.755 3.785 3.815 3.775 3.775 3.175 3.745 3.735 2.965 2.975 3.055 3.095 3.075 3.155 CH3 3.755 3.665 3.605 3.475 3.555 3.305 3.375 3.735 3.635 3.555 3.475 3.445 3.265 3.255 3.665 3.585 3.465 1.975 1.975 1.855 1.935 3.645 1.955 1.955 1.875 1.885 1.835 1.815 N -H 4.20b -2.935 -2.915 2.215 0.06b -0.20b -2.875 -2.795 2.155 147 deformation of the adjacent saturated ring which somehow inhibited the rotation of the acetate group such that the magnetic environment of the two protons became different. In the spectrum of d], the vinylic protons on the acrylate exhibit two doublets at 6.90 and 8.96 ppm with a coupling constant of 17 Hz typical of a trans olefinic structure for the acrylate side chain. In most isobacteriochlorin,60 the usual pattern of meso protons is one going down-field (the one between the unsaturated pyrroles, Y), a pair at the intermediate shift (the two between a saturated and an unsaturated pyrrole, L3, 6), and one relatively upfield (the one between the two saturated pyrroles, 01). In 1,3-porphyrindione5, the deshielding effects of the carbonyl oxygens have distorted the usual isobacteriochlorin pattern. The 3-carbonyl oxygen has deshielded OL-proton from its furthest upfield normal position to a shift comparable to [B-proton, while the 1-carbonyl has deshielded 6-proton to a range comparable to that of y. The deshielding effect of the carbonyl group is best seen in the monoketone's case as shown in Figure 12: one of the two original upfield peaks attributed to the meso protons flanking the reduced pyrrole ring is shifted down-field to the region of the two protons by the pyrrole rings, thus giving a "3:1" pattern which has been universally observed in the spectra of all porphyrinones. The 1H-NMR chemical shifts of the meso protons from the spectra of some selected porphyrinones, porphyrindiones, chlorin, and isobacteriochlorin are listed in _T_a_bl_e_2 for comparison. B. 13C-NMR Spectra The 13C spectra data of the selected porphyrin, porphyrinone and porphyrindione are summarized in Table 10. Rather than discussing each 148 JL 47- 4 I 10.! 3.0 8.0 7.0 8.0 111.0,: €0,111. 111 Mgc C0,". THF [****1_“ -3.0 1L MU QULJ; L fifirfifl‘m15‘1‘1”“1*“‘1*‘”I‘mymfi‘fifi *17 2.0 1.0 -.I "5.121 «.121 PFn 3.121 Figure 19 250 MHz 1H-NMR spectrum of porpnyrinone 111 in CDC13. 149 Table 9. 1H-NMR schmical shifts of meso protons of porphyrindiones in comparison with their analogous bacteriochlorms and isobacteriochlorins. ' ----_-—--rr—tes—o-p;o-tons (6') compound or p Y 6 OEPone 26 9.13 9.14 9.86 9.83 1,3-OEPdione 8 8.58 8.39 9.24 9.37 2,3-OEPdione 32 9.59 8.83 9.79 8.83 1,4-OEPdione 33 8.81 7.24 9.05 7.42 1,5—OEPdione 28 9.05 9.71 9.05 9.71 1,6-OEPdione 29 8.78 8.78 9.59 9.59 MeOEC D 8.71 8.87 8.87 9.69 1,3-DMOEiBC E 6.56 7.13 8.35 7.28 2,3-DMOEiBC F 6.70 7.15 7.15 8.38 1,4-DMOEiBC G 6.44 7.29 7.29 8.35 1,5-DMOEBC H 8.52 8.52 8.67 8.67 cis-1,3-dione 59a 8.67 8.46 9.56 9.38 sirohydrochlorin 6.78 7.36 8.53 7.46 H H H 11 F 2,3-DMOEiBC 1,4-DMOEiBC 1,5-DMOEiBC 150 Table 10. 13C-NMR chemical shifts of selected porphyrin, porpnyrinone and porphyrindione. compound 5 (ppm) carbon porphynn h : one - . dione ................. dfreebase 58 porplh 59a 1 128a on and 13 131.28 136.25 129.70 132.58 131.40 131.61 132.49 133.20 (pyrrole) 136.55 137.41 133.32 133.73 131.84 135.05 134.60 135.50 138.16 138.23 134.58 135.86 136.11 137.76 136.87 140.08 143.76 137.67 138.61 139.14 144.90 167.63 167.63 141.08 146.20 145.20 145.56 151.37 153.34 161.13 165.20 163.31 -C- (tertiary) 52.15 49.42 49.51 {1120-1280013 173.48 17297 173.33 173.20 172.93 -CH2CI-1280§H3 51.63 51.42 51.78 51.77 -_ 2Q-1280C1-13 32.36 32.14 36.71 36.27 36.24 36.27 21.69 21.62 21.51 21.10 21.36 21.12 21.25 -CH=QI-180g-13 122.67 52.00 -CI-12 OCH3 171.92 171.83 171.16 170.45 170.54 170.34 170.42 012809-13 52.25 52.21 51.24 50.98 51.59 51.68 51.48 {112800-13 36.83 42.12 31.87 41.67 41.98 41.50 42.21 -CH3 (pyroole) 11.47 11.23 11.28 10.86 11.12 10.69 12.87 10.80 -CH3 (pyrroline) 23.84 23.46 23.86 23.93 23.24 -C=O (ring) 208.35 207.42 207.33 206.53 205.50 Q,p.y,6 96.80 96.64 99.38 98.13 98.88 96.57 10264 98.14 (meso) 96.44 95.97 92.96 91.76 91.37 91.18 91.54 89.97 151 spectrum in detail, the general features of 13C spectra of porphyrinone and porphyrindione are presented here. Since the relative contribution of aromatic ring current to the final chemical shift is much less for 13C than 1H, there is less ambiguity about the bond types or neighboring groups in these structures. The spectra of these compounds can be deliberately subdivided into four regions: The aliphatic carbon region with the chemical shifts in the range of 10 to 60 ppm; the meso carbon region, 90 to 105 ppm; aromatic ring carbon region, 125 to 165 ppm; and the carbonyl region in the most strongly deshielded portion of the spectrum, 170 to 210 ppm. It is evident that the methylene carbons attaching directly to the pyrrole rings have their resonances at upper field in comparison with their counterpart attaching to the saturated pyrroline rings, for example, the methyl group on pyrrole ring appears at 11.3 ppm versus 24 ppm in the saturated ring and the methylene carbon of the acetate, 37 ppm compared to 42 ppm. The meso carbon signals are closely spaced in porphyrins' case but are spread out in porphyrinones and porphyrindiones. They are quite sensitive to the conjugation effect of the acrylic acid substituent. In the spectrum of cis-dione 59a the two upper field meso peaks essentially overlap at 91.30 ppm and the two down-field resonances at 96.57 and 98.88 ppm. In cis-g1, the two upper field peaks separate by more than 1.5 ppm, 89.97 and 91.54 ppm, and the other two peaks are down—field shifted to 98.14 and 102.64 ppm respectively. Except for the 6-acrylate side chain, these two compounds are identical. The region between 130 to 170 ppm belongs to the resonances of the on and p pyrrole carbons. In the case of porphyrin, these resonances are closely 152 spaced, 130 to 146 ppm, and exhibit mainly in two sets with the (3 carbons at the upper field. Owing to the N-H tautomerism at room temperature, these peaks, especially those of the on carbons, are close to coalescence. The N-H exchange might be slower in the porphyrinones and porphyrindiones, since the on and 0 bands are observed prominently. These bands are also spread out in the whole region with the on carbon next to the keto group significantly down field shifted to 167 ppm. The signals of the carbonyl carbons on both propionate and acetate side chains are observed in the narrow region from 170 to 174 ppm, whereas the ring carbonyl carbons appear above 200 ppm. IV. VIBRATION AL SPECTROSCOPY A. Infrared Spectra The IR spectra of 1,3-porphyrindiones, including that of d1 itself, are quite similar to those of isobacteriochlorins, especially where the skeletal bands are positioned. The spectra of heme d], in the form of its free base and copper complex, are given in Figure 29, the absorption bands of £1, in comparison with its precursor dione 59a, sirohydroporphyrin and DMOEiBC are listed in Table 11. The strongest bands in the spectra of _d_] from 1713 to 1741 cm"1 and 1170 to 1205 cm'1 are characteristic of C=O and C-OR stretching vibrations of the carboxylate esters. The typical ring C=O bands are observed at 1717 cm'1 as in Cu(II) d1 and OEPdione 8, however this band is usually mingled with the ester bands around 1735 cm'1 in diones with acetate and propionate side chains. The bands in the 2953 to 2851, 1437 to 1457, and 1350 to 1380 cm-1 regions are characteristic of C-H stretching and bending vibrations of 153 000“ Vflncq.€n Gm 60: «Lu .cc UldeLQ UJQICm EO$ NJECP XGWQ £H\UQ >PHMCUDHZD UPCPW ZCQHIUHZ m m - o c e. o u. m a 1 mama 1 83 I NNS 22 1 lluulllllllllhull Sufi _ _ fl _a_ _ d _a_ _ _ W. A _ WQQPZZQCC—N on: K9 mam a 0N3 COflh Mflhc an: IHOHBNINWE UHWISHY Figure 20 FT-IR spectra of 511 and Cu (11) d1 , samples were prepared as a thin films on NaCl pellets. 154 Table 11. Infrared absorption bands of d1 and porphyrindiones in comparison with sirohydroporphyrin and ilsobacteriochlorin d (cis) d (Cun) 1,3-dione 1,3-dione siro-hydro— DMOE- 1 8a 28a 59a 8 chlorin iBC 3369w 3370w 3363w 3275w 3266w 3275w 3068w 3039w 2954m 2954m 2953m 2965m 2963m 29245 29235 2924s 29335 29625 29315 2859m 2852m 2853m 2875m 2887m 2870m 17315 17415 17355 17295 17135 17145 1 640m 1626m 1622m 1595m 1600m 1599m 1599m 1603m 16035 1576m 1574m 1570m 1550w 1560w 1546w 1541w 1537w 1539w 1511w 1516w 1511m 1464m 1453m 1453w 1451m 1457m 1457m 1452m 1437m 1437m 1437m 1415w 1408w 1397w 1377w 1375w 1376w 1380w 1375w 1381w 1353w 1351w 1351m 1324w 1328w 1332w 1318w 1318w 1271m 1279m 1278w 1 247w 1260m 1260m 1257m 1267w 1260m 1232w 1246w 1204m 1204m 1203m 1207m 1 198m 1170m 1181m 1175m 1178m 1188m 1 168w 1151w 1140w 1129w 1133w 1113w 1113w 1104w 1106w 1102w 1084w 1093m 1063w 1067w 1059w 1059w 1 030w 1021m 1024w 155 Table 11 (contd.) 1010m 1013m 1014m 1001w 1005 979w 986w 971w 965w 941w 943w 940w 948w 948m 916w 921w 919w 893w 884w 893w 898w 887w 870w 864w 864w 860w 860w 851w 835w 819w 820w 827w 800w 801w 791w 770w 768w 773w 777w 772w 755w 757w 735w 741w 728w 726w 728w 718w 710w 702w 703w 690w 691w 685w 685w 666w 668w 674w 666w 639w 617w 156 methylene and methyl groups. Heme 111 has a prominent band at 1595 cm’l, which is identified as the skeletal band common to all isobacteriochlorin-type structures, revealed by data in Table 11. Previously, it was reported by Mason”, who observed a strong band at 1598 cm“1 from the spectrum of tetrahydroOEP. The corresponding absorption bands of bacteriochlorin-type diones are determined in lower frequency region, for example, 1593 cm'1 was observed for Cu(II) 1,5-OEPdione and 1594 cm'1 for 1,6-dione free base. Chlorins and porphyrinones exhibit a medium strength band in this region with slightly higher wavenumbers, as shown in Table 12, methleEC has a band at 1610, porphyrinone 26 and 111 exhibit one at 1604 and 1606 cm'1,respectively. Porphyrins rarely have bands in the 1500 to 1700 cm'1 region, and such bands when present are usually broad and weak. Therefore, the band at 1600 cm’1 vicinity can be considered as the diagnostic absorption for porphyrindiones and isobacteriochlorins.95 The acrylate side chain should add two types of IR bands arising from the conjugated olefinic and ester groups. However, these bands are not so easily identified in the spectrum of Q] due to their overlapping with the other C=C and -COOR modes. Alternatively, the contribution from the acrylate function to the spectrum is best illustrated by comparing the spectrum of a porphyrin acrylate 146 and its precursor porphyrin 61, as shown in Figure 21. The C=C stretching mode of acrylate stands out clearly at 1623 cm'1 in the spectrum of porphyrin acrylate 146, corresponding to a band at 1626 cm'1 in the spectra of £11 free base and 1620 cm'1 of its Cu(II) complex. Notably, this band is more prominent for porphyrin acrylate since it is not marred by the strong skeletal absorption observed around 1600 cm'1 in the 911 spectra. The v(C=C) mode is commonly seen as a doublet for acrylates, resulting either 157 as: :...__.__ : I (”NW /\/\'\ W\//’\\I\ -, (WVMII. 9? i I ‘ - I 1 WW . 7 :8 g 3 {.3 E a!" E a _ : § (2' ‘ (:1 fl: g 1 J : E n.0,: 61 C0,". I I I I I 7 I I I I I 1 F I I m I- I- ~ 90.11—- ,..—.\ I 1rd,“ I In /\ A)” AA 1‘” : I )6] K‘f I N \ IV"! I I I " ....—- 1 2 1 j" N l I _ r\ I“ v I — I If V" s .. as: 5' : g :5? Cl I 1 =2 s 8 75.11: E g : E 11.6 c 66,... ._ I-t ' 145 I I I I I I I I I I I I I also 190 use sac mm Figure 21 FT-IR spectra of porphyrin 61 and acryloporphyrin 146, samples were prepared as a thin films on NaCl pellets. 2 MIN“ 2 M! TTNCE HIDIIOM STATE WHERE-BUY 158 from an overtone or due to rotational isomerism;96 however, no evidence for the second band is observed in the spectra of _d_1 and its synthetic models. Out of plane =CH deformations are expected for trans acrylates at ~980 to 974 cm'l,96 and are observed at 979 cm‘1 as a weak but clear band for both d1 and porphyrin acrylates 146. The cis acrylate should have a band around 820 cm'l, but no such absorption was ever observed in the d1 spectrum, suggesting a trans acrylate side chain. The carbonyl stretching mode, V(C=O), of the methyl acrylate group is observed at 1718 cm'l, but it is usually overlapped with the V(C=O) modes of the two ring keto groups in the spectra of Q] and the other porphyrinones and porphyrindiones. This band bifurcated at 1716 cm’1 due to the propionate ester band as can be seen in the spectrum of porphyrin acrylate 146. The CO stretching modes of the acrylate group at 1310 to 1250 cm’1 and 1200 to 1100 cm“1 are observed at 1271 and 1270 cm‘1 for 146 and _d_l respectively and around 1170 cm'1 in both cases in combination with the bands of the propionate esters. In the free base d1 spectrum, two N-H stretching bands are observed at 3275 and 3369 an1 with an overtone band of the carbonyls at 3423 crn‘l. The presence of two VN-H suggests two nonequivalent hydrogens in the cis-d1 structure, as has also been proved by X-ray study which indicated both hydrogens are located at the southern ring C and ring D pyrrole nitrogen, and they are nonequivalent due to the asymmetric structure of c_1_1. This is to be compared with isobacteriochlorins where two bands with different intensities were recorded suggesting the presence of two different tautomers.6O The N-H stretching vibrations appear as a clean single band in the region of 3335 to 3345 cm“1 for porphyrinones and in 3310 to 3340 cm‘1 for porphyrins. Table 12 gives the infrared absorption bands of some selective 159 Table 12. Infrared absorption bands of porphyrinones in comparison with chlorin. acrylo- porphyrinone porphyrinone porphyrinone MeOEC 111 26 D 3342w 3338w 3336w 3341w 2956m 2952m 29635 29635 2922m 2925m 2933m 2931m 2852m 2857m 2872m 2868m 17385 1737s 1714s 1713s 1623m 1614m 1606m 1604m 1609s 1588w 1586m 1559w 1557w 1542w 1540m 1522w 1516w 1522w 1519w 1465m 1456m 1454m 1449w 1436m 1437m 1402w 1403m 1397w 1378w 1376w 1372w 1373w 1360w 1352w 1350m 1320m 1311w 1274m 1261m 1265w 1266w 1239w 1225w 1224w 1229w 1195w 1198m 1184m 1198m 1168m 1169m 1163w 1165w 1153w 1130w 1136w table 12 (contd.) 1101w 1073w 999w 978w 949w 914w 885w 855w 838w 744w 718m 690m 672m 630w 1109w 1088w 1076w 1059w 1021m 966w 952w 914w 896w 850w 706m 672m 629w 160 1095m 1056m 998m 952m 898w 862w 846w 825w 741w 732m 715m 703w 682m 666w 608w 1097w 1085w 1056m 1014w 988m 945m 896w 885w 845w 824w 744w 732m 712m 701w 685w 664m 603w 161 porphyrinones. Methyloctaethylchlorin (MeOEC), is also included for comparison. The similarity between these two systems are easily observed. B. Resonance Raman Spectra In the metal complex of porphyrindione, there is an inherent x, y inequivalence of the macrocyclic Tl-conjugation pathways, resulting in an absorption spectrum possessing separately allowed Qx and Qy visible transitions and two Soret transitions. The visible spectrum of Cu(H) d1 is given in Figure 3c, and as seen therein, the wavelength chosen for excitation, Ar ion-laser at 457.9 nm and Kr ion-laser at 406.7 nm, are on the different side of Soret band (437 nm). The resonance Raman spectrum of extracted natural g1 methyl ester and model compound 6 and 10 in their Cu(II)-metallated form are shown in Figure 22, and the spectra of copper complexes of synthetic 511 and dione 59a are given in Figure 23. The similarity between these two sets of spectra is obvious, however, the spectra obtained upon excitation at 457.9 nm have their lower frequency modes intensified whereas those under 406.7 nm excitation have their higher frequency bands enhanced. It is noteworthy that the number of the resonance Raman active vibrations of _d_l and the 1,3-dione model compounds, either with or without the acrylate side chain, far exceeds those apparent in Soret-excited scattering from metalloporphyrins. This is due to the reduction in symmetry of the chromophore from D4h to Cs, allowing the Raman-forbidden Eu vibrational modes of the higher symmetry structure to become Raman active modes in the low symmetry. In 1981, Ching et al.97 reported the resonance Raman spectra of cytochrome Q1 nitrite reductase using selective excitation to study heme g and d1. With excitation in the Soret region of heme $11 they observed an 162 . | I l 1339 (A) 1376 ' 1385 1402 1627 O 15. .°6 1647 5 . 1609 1719 1572 , 1479 1338 . “1602 1643 [I (8) E3 1473 1560 U) P 1551 1374 2: 1376 E3 . 1392 1213 E Ed 1450 PI-I3+) of some porphyrindiones have been determined in 2.5% sodium dodecyl sulfate: 1.8 for model compound 8, 1.7 for dione 10 and 51] itself. These values are drastically less basic than that of common porphyrins (3.0-5.8) or chlorins (3.5-4.2).109 The extremely weak basicity of d1 and corresponding diones 171 relative to isobacteriochlorins is derived from the strong electron- withdrawing effect of the ring-keto groups which cause the central nitrogens less basic. Similarly, the monoketone compound 26 has been reported unable to form its cation PH3+ by reaction with chloroacetic acid in benzene, unlike its chlorin counterpart which is fully protonated under the same reaction condition.110 VIII. E>G°ERINIENTAL Visible absorption spectra (in CH2C12 or CHC13) were measured with a Cary 219 or a Shimadzu 160 spectrophotometer. Spectra were plotted directly from data stored on floppy diskettes. 1H- and 13C-NMR were obtained at 250 MHz on a Bruker WM-250 instrument. Spectra were mostly recorded in CDC13; the residue CHC13 was used as the internal standard set at 7.24 ppm. IR spectra were. obtained from KBr pellets (~1 mg compound/ 100 mg KBr) or films on NaCl pellets (by evaporating the CHC13 solution of the samples) on a Perkin-Elmer Model 1800 FT -IR instrument and a Nicolet IR/ 42 spectrometer. Raman spectra in Figure 22 were obtained with a computer controlled Jarrell-Ash scanning spectrophotometer using a Spectra-Physics 164-05 Ar laser. Spectra were collected in back-scattering geometry on anaerobic solution samples (~1 mg/ ml in CH2C12) maintained at 2 °C in a sample dewar. Alternatively, room temperature spectra were recorded on polycrystalline samples dispersed in KBr (~1 mg sample/ 100 mg KBr) pressed into the angular groove of a spinning sample holder. Spectra in Figure 23 were obtained from samples spining in cylindrical quartz cells and freshly 172 distilled CH2C12 was used as solvent. The Raman equipment included a Spex 1401 Ramalog with PMT detection and a Spex 1877 B outfitted with a EG&G model 1421 detector and OMA III computer. The laser system used is Innova 90 krypton ion laser. Cyclic voltammetry was performed using a Bioanalytical System CV-IA unit. All measurements were carried out in CH2C12 containing 0.1 M tetrabutylammonium perchlorate at a scan rate of 200 mV/ sec. ‘I—‘l'llllim CHAPTER 6 RECONSTTTU’I‘ION OF CYTOCHROMe gal WITH NATIVE AND SYNTHETIC HEME _611 I. PREVIOUS WORK Yamanaka and Okunuki110 first studied the reconstitution of cytochrome ggl from P. aeruginosa with the extractable heme 511 group as well as other porphyrin hemes. The oxidase activity of the reconstituted protein was determined by the oxidation of reduced cytochrome 9551 under aerobic conditions and the nitrite reductase activity was examined by the oxidation of cytochrome £551 anaerobically in the presence of nitrite. They obtained a 37% recovery of oxidase activity and 54% recovery of nitrite reductase activity with heme d1 reconstituted enzymes and partial activity recovery with other hemes. Hill and Wharton111 later reconstituted the same enzyme with natural heme Q] by a different method and restored almost quantitatively the original oxidase activity. They also found that, except for the heme _a which yielded 5% activity after reconstitution, no other heme showed any activity. This chapter describes our reconstitution work on nitrite reductase from P. stutzeri with both native and synthetic heme d1 groups. The spectral properties and the NO and N20 producing activities of the reconstituted enzymes give further evidence that the unusual dione structure of Q] is correct and the synthetic compound is fully functional. 173 174 II. EXPERHVIENT AL A. Preparation of Apoprotein Nitrite reductase was prepared and purified from Pseudomonas stutzeri (strain JM 300) by the method of Weeg-Aerssens et al.112 The overall procedure is shown in Scheme 28. Basically, 65 g of cell paste, harvested from 15 liter of tryptic soy broth containing 5 g of KNO3, 2 g of N aHCO3, 10 1.1g of CuSO4 and 10 mg of FeSO4.7H20, yielded about 45 mg of cytochrome 9d]. Since the significant loss of the green heme d1 concentration in each stage of chromatography and dialysis has been observed, above procedure can also be simplified by using just 3 steps: a DEAE-52 column to separate cytochrome £551 and other cytochromes, a Sephacryl-300 sizing column, and then directly to hydroxylapatite to obtain higher total yield of pure enzyme. If the purity of the enzyme is not strictly required, two DEAE columns only can offer an even higher yield with about ~90% purity with much less effort. All the purification and reconstitution operations were carried out in a cold room (4 °C). The apoprotein of nitrite reductase was prepared according to the procedure developed by Hill and Wharton.111 Typically, 2 mg of pure enzyme in 1 ml of buffer (25 mM HEPES, PH 7.3) was treated with 5 ml of cold acidified acetone (0.024 N HCl). The apoprotein was precipitated and the green heme Q] was extracted into the overlying acetone solution after shaking the mixture for 1 minute. The apoprotein was separated from acetone by centrifugation at 1000 x g for 5 minute and the precipitate was extracted once more with 3 ml of acidified acetone to ensure complete removal of heme 511. The-protein precipitate was washed once with 3 ml of phosphate buffer (25 mM, PH 7.0). The pellet was then redissolved in 175 cell of pseudomonas stutzerl sonication ( heat system-ultrasonic W-225 ) centitugatlon (after RNAase and DNAase treatment. 12,000 x g ) ammonium sulfate precipitation centrifugation ( 12,0009 x 30 min) ultracentifugatlon I (2h,100.000 x g) cell free extract of pseudomonas stutzng DEAE-52 ( 2.5 x 20 cm ) KCI 0 to 400 mM Tris 10 mM, PH 7.0 1 mainly ' crude cytochrome cg1 other cytochrome c551 ( nitrite reductase) cytochromes Sephacryl-soo ( 2.5 x 75 cm ) Tris 50 mM, PH7.0 DEAE-52 ( 2.5 x 15 cm ) Tris 50 mM, PH7.0 KCl 50 to 250 mM Hydroxyapatlto ( 1.5 x 20 cm ) phosphte 150 to 500 mm, PH 7.0 pure cytochrome 9311 ( nitrite redUctase) Scheme 28 176 phosphate buffer containing 6 M of urea with gentle stirring until a homogeneous reddish solution formed. The apoprotein solution was stored at 0 ° C for use in next step. B. Preparation of Heme (11 The native heme 511 was exacted from nitrite reductase by the procedure described above. The green acetone solution containing heme d1 was brought to near dryness in the dark by blowing the acetone off with a stream of nitrogen. The residue was dissolved in the phosphate buffer (25 mM, PH 7.3) and centrifuged to remove any remaining protein precipitate. The heme d1 solution was adjusted to PH 7.0 in an ice bath with N aOH (1 N) and stored in the dark under argon. The iron insertion of the synthetic heme d] tetramethyl ester was accomplished by the standard ferrous sulfate pyridine/ acetic acid method.126 Hydrolysis of the methyl ester was carried out by dissolving 5 g of the ferric heme d1 tetramethyl ester in 10 ml of tetrahydrofuran (THF) followed by addition of 1 ml of KOH solution (1 N). The reaction solution was stirred in the dark under argon for 10 hours or until the organic layer had become almost colorless. THF was then evaporated and the PH of the aqueous solution was adjusted to 7.0 with HCl (1 N) in an ice bath. This heme solution was also stored in the dark under argon to avoid decomposition. C. Reconstitution of Nitrite Reductase The reconstitution was carried out according to Hill and Wharton's procedure modified as follows. To maximize the yield of incorporation of heme $11 into the apoprotein, excess amount of heme d1, 10 to 1, was used. The heme d1 solution was added to the heme _c_ containing apoprotein 177 solution and the mixture was incubated with gentle stirring for 30 minute, then dialyzed with agitation for 12 hours against phosphate buffer (10 mM, PH 7.0). The dialysis medium was change twice during the period. The reconstituted enzyme was separated from the excess of heme Q] by passing the crude enzyme solution over a short DEAE cellulose column (DE-23, Whatman, 0.5 x 5 cm). Heme Q1, which has a PKa of 4.5, stick tightly at the top of the column and the reconstituted enzyme was eluted off with additional concentrated phosphate buffer (100 mM, PH 7.0). The reconstitution is very straightforward : a successful reconstitution invariably gave a well defined narrow band of the protein from the column, whereas nonbounded ones, such as that with protoheme, resulted in a very defused band. D. Estimation of Protein The concentration of protein was estimated by the bicinchoninic acid method of Smith et al113 using crystalline bovine serum albumin as the standard. The relative concentration of the reconstituted enzyme was also estimated by taking the ratio of the absorbance at 522 and/ or 555 nm versus that of intact enzyme in the reduced state.114 Both methods gave comparable values. The concentration of heme Q was estimated by using the published extinction coefficient of 32,100 M'1 cm’1 for the imidazole-ferriheme complex.“4 Absorption spectra were recorded on a Perkin Elmer 5 or a Cary 219 spectrophotometer. E. Activity Assay Activity was measured by gas evolution (NO and N20) from nitrite with NADH/phenazine methosulfate (PMS) as the electron donor system. The 178 assay contained 6 pmol NADH, 0.36 pmol PMS and 3 pmol NaN02 in a total volume of 3 m1. All stork solutions were made in HEPES buffer (50 mM. PH 7.3). A mixture of NADH and nitrite solutions was made oxygen free by repeatedly evacuating and fill with argon. The deoxygenated mixture was added anaerobically to a 25 ml serum bottle containing the PMS solution in buffer which had been flushed with argon. The reaction was initiated by addition of the enzyme, about 1 1.1g. The nitrite concentration (0.5 mM) was in excess for both NO and N 20 production.112 Enzyme activity was based on the initial rate of NO and N 20 evolution. Gas evolution was monitored on a Perkin Elmer 910 gas chromatograph equipped with 63Ni electron capture detector and Porapak Q column under conditions previously described.112 Carrier flow rate was adjusted so that the approximate retention time for NO and N20 were 1 and 2.2 minute, respectively. Under these conditions the retention times of nitric oxide and oxygen were extremely close and it was necessary to use strict anaerobic techniques for sampling and injecting the gas phase of the assay vials. We used a gastight syringe equipped with a gas lock and the needle and syringe were flushed with argon before use. 111. SPECTRAL CHARACTERIZATION The acetone extract containing green heme d1 has an absorbance maximum at 427 nm (Soret band) and a broad plateau from 520 to 520 nm with the low intensity maxima at 529, 575 and 612 nm. This spectrum is essentially the same as that from P. aeruginosa by Yamanaka and Okunuki.115 The spectrum of synthetic heme d1 in acidified acetone is indistinguishable from that of the native d1 as shown in Figure 26. The ' Absorbance 179 0.8 0.6 - 004 '- o.2 L 005 I l I I B _ 427 - 1 0.3 r 1 O.| ” '3 0.0 ‘ ' 400 ‘ 600 800 Wavelength < nm ) Figure 26 UV-vis spectra of Fe (III) d1 in acetone containing 0.024 N HCl and about 10% of water, (A) synthetic heme 511; (B) native heme :11 from W. 180 spectra of oxidized and reduced synthetic heme d1 in phosphate buffer at pH 7.3 is shown in Figure 27. The ferric heme has absorption maximum at 406 nm (Soret), a broad shoulder around 480 nm and a near infrared band at 781 nm. When Fe (III) was reduced to Fe (II) by Na28204, the Soret band appeared at 454 nm with a shoulder around 416 nm and the near infrared band shifted to 628 nm. These spectra are also very close to those observed by Yamanaka and Okunuki115 and can be used as excellent references to identify the characteristic absorbances of heme gl_1 in the spectrum of nitrite reductase. As illustrated in Figure 28, The absorption spectra of pure preparation of nitrite reductase from P. stutzeri is similar to those reported by Yamanaka and Okunuki116 and Hill and Wharton111 from P. aerpginosa nitrite reductase. There are absorbance maxima at 411 nm (Soret), 524, 644 nm and a shoulder at 360 nm in the oxidized form; and at 417 nm (Soret), 460, 522, 549, 555 nm and 625-655 nm plateau in the reduced form. The 280 nm absorbance belongs to protein and the 315 nm band is derived from dithionite. The main differences between the reduced spectrum of this enzyme and that from L aeruginosa is that the intensity of the doublet at 549 and 555 nm is reversed such that the peak at 555 nm is now more intense than the one at 549 nm, whereas in all the reported spectra of nitrite reductase from P.aeruginosa_ the Opposite trend is observed, that is, the peak at 549 nm is higher than that at 555 or 554 nm. This feature is consistent in all of our preparations and has also been recorded in an earlier study on the nitrite reductase of another strain of L stutzeri (Van Niel strain).117 In addition, the broad band in the long wavelength area in the oxidized state, 644 nm, is somewhat bathochromically shifted comparing with the one in the spectra of the enzyme from L aeflginosa that usually appears between 635 to 649 nm. The Soret band at 417 nm in the reduced spectrum is attributed to heme g 181 .vOumNu Z :33 @8560." PI «vunmvmxc .I .m& IA 3 com—5 3290.93 we 2 mad 5 an 3855?. mo «53% 0.3-5 RN Sawm— nficv fmco_o>m>> com ooh com com . 00¢ 00.0 1 . No.0 I v. (D O 1 . m~._ eoueqiosqv 182 2 mad 5 flung :80 830308 31:: mo Coachmen «an we «.5on £>.>D mu scum—m 000 .vOuwunZ 5:: 03308 I 6052.8 .1... 6.5 In on Human ouaammonm “E: o fmco_o>m>> 000 00¢ 00» hpv 0.0 m.N aoueqiosqv 183 and so is the band at 522 nm and the doublet at 549 and 555 nm. Comparison of the spectra in Figure 27 reveals that heme d1 is responsible for the weak shoulder at 460 nm and the broad band near 640 nm. The spectrum of apoprotein, Figure 29, is deprived spectral characteristics typical of heme d]. the spectra of enzymes reconstituted with the native and synthetic heme d1, shown in Figure 39a and 3012, have regained virtually all the features of heme £11. The 460 nm shoulder peak in the spectrum of the enzyme reconstituted with synthetic heme g] is less prominent, indicating a slightly lower incorporation of Q] into the protein. IV. Recovery of activity The progress curves of NO and N20 production for intact nitrite reductase and reconstituted enzymes are shown in Figure 31 and 3;. The shapes of these curves are approximately the same. Because of the pattern of gas production was not linear, the values for restoration of activity were calculated from the amount of gas produced at each time point in comparison with those for the intact enzyme. The percent recovery of activity was comparable for each time point, and the mean value is reported in Table 16. The heme g containing apoprotein remained soluble after dialysis to remove the urea and had no detectable activity during the time course of an initial rate experiment. When reconstituted with either native and synthetic heme d1 a large part of its nitrite reductase activity was restored: 80% and 78% respectively, for the native and synthetic heme d1 reconstituted enzyme. These activity recoveries are higher than the 54% obtained by Yamanaka and Okunuki110 in their experiment mentioned earlier. 184 300 400 500 600 700 xmm) Figure 29 UV-vis spectra of apoprotein of nitrite reductase from P. stutzeri after removal of heme d1, the apoprotein was dissolved in 0.25 M of phosphate buffer at PH 7.0. ---, oxidized; —, reduced with Na28204. Absorbance 185 0.60 0.36 0.24 0.00 1 400 500 800 Wavelength (nm) Figure 30 UV-vis spectra of reconstituted nitrite reductase in 0.25 M of phosphate buffer at PH 7.0, (A) native d1 reconstituted and (B) synthetic g1 reconstituted. --, oxidized; —, reduced with Na28204. 186 200 _ _ 150 nmole N 100 50 0 30 60 90 120 Time (min) Figure 31 Progress curves of nitric oxide and nitrous oxide production from 1 mM of nitrite by intact nitrite reductase from W. 187 150 l l I I 1 r l l 100 50 150 nmole N 100 ‘ 50 1 I 4 1 L l 1 0 30 60 90 120 Time < min) Figure 32 Progress curves of nitric oxide and nitrous oxide production from 1 mM of nitrite by the reconstituted nitrite reductase, (A) reconstituted with synthetic heme £11 and (B) reconstituted with native heme d1. 188 Table 16. Recovery of nitrite reductase activity after reconstitution of the apoprotein with native and synthetic heme d1. Treatment of enzyme Enzynu02cfivity meas::======= N02‘ to NOa N 02' to N 203 Intact, enzyme 100 100 Apoprotein 0 0 Reconstituted, native g 83(4.4)b 77(3.3) Reconstituted, synthetic _d] 82(3.8) 73(2.5) a Activities are expressed in relative units. The activity of the intact enzyme re resents 100%. b umbers between parentheses represent the deviation from the meanvalue of two initial rate determinations within the same experiment. V. DISCUSSION Cytochrome gd_1 nitrite reductase reconstituted with either the native and synthetic heme d1 is essentially the same in most aspects. The matching of the spectral and biological properties of the synthetic heme d1 with those of the natural prosthetic group leaves on doubt that heme d1 must have the structure as we proposed. As mentioned in chapter 1, the kinetics work to date supports the idea that heme _C_ sites of the enzyme are associated with electron-uptake while heme d1 sites are responsible for electron-donation to exogenous ligands. Thus, it is expected that the apoprotein would not exhibit any nitrite reductase activity due to the removal of Q1 prosthetic groups. The enzymatic activities of the reconstituted enzymes were about 20% lower than that of the intact nitrite reductase, this loss is understandable considering the possible denaturation of the protein during the 189 reconstitution process. The ratio of A46Onm/A555nm in the reduced spectra is 1.68 for the native and 1.48 for the synthetic heme d1 reconstituted enzymes as compared to 1.75 for the intact nitrite reductase. The differences indicate that the reconstitution was about 96% complete for native $1 and about 85% for the synthetic 911. The synthetic heme _d_] we used for the reconstitution comprised a pair of enantiomers. This might account for the lower incorporation of the synthetic heme d1, since the wrong enantiomer presumably cannot fit the protein pocket as well as the correct stereo isomer so that it may dissociate easily from the binding site during purification from excess heme Q]. As we mentioned in chapter 1, Averill and Tiedje32 proposed a pathway by which NOz’ can be converted to N20 by one enzyme (nitrite reductase), through a heme-NO+ intermediate. This proposal is supported by evidence from 15N and 180 studies that show a sequential mechanism of N02” addition and by studies that show two Km values for N02“, 1.4 uM for NO production and 59 11M for N 20 production.118 It also appears that the reaction kinetics of the nitrite reductase are perturbed by its removal from the membrane environment and by purification.118 This likely affects the fate of the enzyme-bound nitrosyl intermediate which enhances the ratio of N O/ N 20 produced by the purified enzyme (Scheme 22) NO ~02\ N20 N02" 2 E-NO+ + Scheme 29 190 The results show that the apoprotein produce neither N 0 nor N 20 from N 02', and that when reconstituted with heme _d_l the original NO and N20 producing activities were restored. Furthermore, the purified enzyme did not produce N20 when NO was added as substrate.118 This confirms that a single enzyme can carry out the entire N02' to N 20 step, and establishes that heme g1 in its nitrite reductase pocket is required for both NO and N20 production. Thus N 20 production from N 02' does not require a nitric oxide reductase. The fact that only one-half of the nitrite reductase product was N 20 can be explained by the altered environment of the purified enzyme. The heme d1 extraction and reconstitution procedure did not alter the N02' to N 2O product ratio. Presently, the relationship between the prosthetic group structure and the reductase function is not well understood. Previously, porphyrin hemes, such as protoheme, hematoheme and mesoheme, have been reported to give little or no oxidase activity in the reconstituted enzyme.111 Some preliminary data from our reconstitution experiment indicated that the structural features of dione heme apparently are very special for the enzyme activity. Several hemes examined to date, including the 2,4—diacetic acid deuteroheme 58, porphyrinone heme 111 showed no nitrite reductase activities at all. Studies are in progress to insert other dione heme analogs (for example, altered keto and acrylic substitutions) to test the essentialness of Q] structural elements. CHAPTER 7 CONCLUSION AND FUTURE WORK I. EVALUATION OF THE PRESENT WORK This thesis work has essentially reached the goals of our original proposal. The model compound study provided solid evidence on the structure we proposed and the methodologies developed toward the 1,3-porphyrindione core structure and acrylate side chain formation led ultimately the total synthesis of Q. In retrospect, the d1 synthesis might be considered as a piece of "brutal force" work since the whole pathway was a "push-through" process with a large quantity of starting materials, severe reaction conditions and also unsatisfactory yields. Nevertheless, the successful synthesis of £11, together with its stereo- and regioisomers, has unequivocally proved its structure. Our physicochemical studies have shown that the most distinctive features of porphyrindions are its electronegativity due to both inductive and conjugating effect of the two keto groups and a enlarged core size brought about by the saturation of the pyrrole rings. The study on the reconstituted cytochrome _C_d_1 indicated that the synthetic heme 511 is fully functional in the enzyme, just like the native heme as far as the nitrite reductase activity is concerned. Another important message obtained from this study concerns the nitrite reduction mechanism. The loss of N20 producing activity in the absence of heme d1 and its restoration by addition of heme d1 suggested that nitrite reduction may 191 192 convert N02‘ to N 20 without invoking a role for nitric oxide reductase in the denitrification pathway. 11. FUTURE WORK It is important to further understand the chemical, structural, spectroscopic characteristics of dioneheme, and its function in microbial denitrification. Future work should include the following: 1. The ligand coordination of heme d1 and model systems. Using synthetic d1 and its analogues, we plan to prepare a variety of 5- and 6- coordinated heme complexes containing CO, 02, NO, CN’, and N02' as axial ligand. Their formation equilibria and chelation dynamics will be studied by stopped-flow or flash photolysis techniques. 2. Further work on reconstitution of cytochrome Cd]. Work on the reconstitution of gdl enzymes with a wide variety of synthetic dioneheme analogues and other related hemes has already begun, aiming at the relationship between the prosthetic group structure and its intrinsic reactivities in the enzyme and the mechanism concerning nitrite reduction. 3. Intramolecular electron transfer in cytochrome cdl. The unique two heme arrangement with a relatively fixed distance of cytochrome g1 provides an ideal model to study the long range electron-transfer problem in biological systems.120 For our purposes, with regard to the binding and electron transfer steps required for the activation and reduction of N02’ and 02, the precise description of intramolecular electron transfer step is important. The key question is whether the slow heme g to heme d1 transfer is a result of electron tunneling over a relatively long distance (13-15 A) or whether large conformation changes are involved. The study of electron transfer rates will 193 be facilitated by using reconstituted cytochrome Q] in which the heme $11 is replaced by other electron donors, such as a Zn (II) porphyrindione. PWN psoxesn 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. REFERENCES Eschenmoser, A., Angew. Chem. Int. Ed. Engl.,1988, 22, 5. Battersby, A. R.; and MacDonald, E., Acc. Chem. Res., 1979, 1_2_, 14. Chang, C. K., |. Biol. Chem., 1985, _2_6_0_, 9250. For a general reference: Payne, W. 1., "Denitrification", 1981, John Wiley & Son, New York, and Delwiche, C. C.; Bryan, B. A., Am; Rev. Microbiol. 1979, Q, 241. Losada, M. 1., |. Mol. Catal., 1976, 1 245. Forget, P.; Dervartanian, D. V., Biochim. Biophys. Acta, 1972, 2_5_6_,600. Iwasaki, H.; Saigo, T., Plant Cell Physiol., 1980, A, 1573. Matsubara, T.; Frunzke, K.; Zumft, W. G., |. Bacteriol., 1982, 1_49_, 816. Hirio, Y.; Hi ash, T.; Yamanaka, T.; Matsubara, H,; Okunuki, K., J, Biol. Chem., 961, 2_3_6_, 944. Sawhney, y.; Nicholas, D. I. D., |. gen. Microbiol., 1978, 106, 119. Matsubara, at.; Iwasuki, H., II Biochem. (Tokyo), 1972, E, 57. Newton, N ., Biochim. Biophys. Acta, 1969, 1_8_5_, 316. Mancinelli, R. L.; Cronin, S.; Hochstein, L. 1., Arch. Microbiol., 1986, 1_45, 202 Kodama, T., Plant. Cell Physiol., 1970, 11, 231, and Weeg-Aerssens, E., 1988, Ph. D. Dissertation, Michigan State University. Horio, T., |. Biochem., 1958, 4_5_, 195. Horio, T., |. Biochem., 1958, 4_5_, 260 Horio, T,; Hi ash, T.; Matsubara, H.; Kusa, K.; Nakai, M,; Okunuki, K., Biochim. iophys. Acta, 1958, _2_9, 297. Yamanaka, T.; Ota, A.; Okunuki, K., Biochim. Biophys. Acta, 1960, g 397. Yamanaka, T.; Ota, A.; Okunuki, K;, Biochim. Biophys. Acta, 1961, 53, 294. Yamanaka, T.; Kijimoto, 5.; Okunuki, K., Nature 1962, 194, 759. 194 21. 23. 24. 26. 27. 28. 29. 30. 31. 32. 33. 35. 36. 37. 38. 39. 40. 41. 42. 43. 195 Yamanaka, T.; Kijimoto, S.; Okunuki, K., |. Biochem., 1963, 88, 416. Mitra, S.; Bershn, R., Biochemistry, 1980, 1_9_, 3200. Saraste, M.; Virtanen, 1.; Kuronen, T., Biochim. Biophys. Acta,1977, 112;, 156. Greenwood, C., in "Metallopprphyrins", Part 1, (Harrison, P. M., ed) Verlag Chernie, 1985, pp 43- . Takano, T.; Dickson, R. E.; Schichman, S. A.; Mayer, T. E., |. Mol. Biol.,1979, 188, 185. Makinen, M. W.; Schichman, S. A.; Hill, S. C., Science 1983, 2_22, 929. Schichman, C. A.; Gray, H. B., |. AmI Chem. Soc., 1981, 188, 7794. Johnson, M. K.; Thomson, A. 1.; Walsh, T. A.; Barber, D.; Greenwood, C., Biochem. |., 1980, 189, 285. Walsh, T. A.; Johnson, M. K.; Greenwood, C.; Barber, D.; Springall, I. P.; Thomson, A. 1., Biochem. |.,1977, 177, 29. Yamanaka, T., Nature 1964, E54.» 253. Timkovich, r.; Thrasher, I. 5., Biochemistry, 1988, E, 5383. Averill, B. A.; Tidjde, I. M., FEBS Letter 1983, 188, 8. Garber, A. E.; Hollocher, T. C., |. Biol.Chem., 1982, ;5_7, 4705. Kim, C.; Hollocher, T. C., |. Biol. Chem., 1984, 859, 2092. Garber, A. E.; Hollocher, T. C., |. Biol. Chem., 1982, A7, 8091. Bryan, B. A.; Shearer, G.; Skeeters, I. L.; Kahl, D., |. Biol. Chem., 1983, 288, 8613. Garber, A. E.; Hollocher, T. C., |. Biol. Chem., 1983, 2_5_8_, 3587. Aerssens, E.; Tidjde, I. M.; Averill, B. A., |. Biol. Chem. 1986, 261, 9662. Aerssens, E.; Tiedje, I. M.; Averill, B. A., I. Am. CHem. Soc. 1987, 1_0_9_, 7214. Keilin, D., Nature 1933, 132, 783. Horio, t.; Higashi, T,; Yamanaka, T,; Matsubara, H.; Okunuki, K., L Biol. Chem. 1961, 288, 944. Yamanaka, T,; Okunuki, K., Biochim. Biophys. Acta, 1963, Q, 397. Barrett, 1., Biochem. |., 1956, 6A, 626. 45. 47. 48. 49. 50. 51. 52. 53. 55 56. 57. 58. 59. 60. 61. 62. 63. 65. 196 Lemberg, R.; Barrett, J., "The Cytochromes", 1973, Acedamic Press, New York. Timkovich, R.; Cork, M. S., Taylor, P. V., J. Biol. Chem., 1984, 259, 1577. Timkovich, R.; Cork, M. S.; Talyor, P. V., |. Biol. Chem., 1984, 259, 15089. Inhoffen, H. H.; Nolte, W., Tetrahetron letter 1967, 2185. Inhoffen, H. H.; Nolte, W., lustus, Liebigs Ann. Chem.,1969, 225, 167. Bonnett, R.; Dolphin, D.; Johnson, A. W.; Oldfield, D. J.; Stephenson, G. F., Pro. Chem. Soc., 1964, 371. Johnson, A. W.; Oldfield, D. J., l. Chem. Soc., 1965, 4303. Chang, C. K.; Barkigia, K.; Hanson, L. K. ; Fajer, J., I. Am. Chem. Soc., 1986, 1%, 1352 Murph , M. J.; Siegel, L. M.; Kamin, H.; Rosenthal, D., |. Biol. Chem., 1973, _2_8, 2801. Inhoffen, H. H.; Notle, W., lustus Liebigs Ann. Chem.,1969, 7_2§, 167. Gouterman, M.; 1978, in "The Porphyrins", (Dolphin, D., ed.)Vol. III, Academic Press, New York, pp 1-165 Musselman, B. D.; Waltson, J. T.; Chang, C. K., J. Org. Mass Spectrom., 1986, _2_1, 215. Smith, K. M.; Unsworth,j. F., Tetrahetron 1975, 3_1_, 367. Janson, T. R.; Katz, J. J., Porphyrins, 1979, vol. 4, 1-54. Fischer, H.; Gebhardt, H.; Rathhass, A., lustus liebigs Ann. Chem., 1930, 412, 1. Fischer, H., Pfeiffer, H., lustus Liebigs Ann. Chem.,1944, $6, 131. Chang, C. K., Biochemistry, 1980, 1_9_, 1971. Collins, C. 1., Q. Rev. Chem. Soc, 1960, 14, 357. Fischer, H.; Eckoldt, H., lustus Liebigs Ann. Chem., 1940, §5_4, 138. Chang, GK; Sotiriou, C., I. Org. Chem., 1985, Q, 4989. Sotiriou, C.; Chang, C. K., I. Am. Chem. Soc., 1988, mg, 2264. Sotiriou, C., Ph. D. Dissertation, Michigan State University, 1987. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 78. 79. 80. 81. 82. 83. 85. 86. 197 Oster, M. Y., Ph,. D. issertation, University of Wisconsin, 1971. Fischer, H.; Zischler, H. Z., Physiol. Chem., {Hoppe-Segler'sl 1937, 245, 123. McDonald, S. F.; Stedman, R. J., Cana. J. Chem., 1954, Q, 896. Chang, C. K.; Sotiriou, C., |. Hetergycl. Chem.,1985, g, 1379. Aeglius, L.; Ballan, J. A.; Ferrito, V.; Jaccarini, V.; Murray-Rust, P.; P tera, A.; Psaila, A.F.; Schember, P. J., Pure A 1. Chem. 1979, _51, 1847. Scott, A. 1.; Irwin, A. J.; Siegel, L. M.; Shoolery, J. N ., I. Am. Chem. Soc., 1978, m, 7987. Battersby, A. B.; McDonald, E,, Bioorg. Chem.,1978, Z, 161. Witlock Jr. H. W.; Hanauar, R.; Oster, M. I.; Bowry, B. K., I. Am. Chem. Soc., 1969, _9_1, 7485. Smith, K. M.; Goff, D. A.; I. Am. Chem. Ed, 1985, 1_07, 4954. Griffiths, G. F.; Kenmer, G. W.; McCombie, S. W., Smith, K. M.; Sutton, M. J., Tetrahetron 1976, 32, 275. Fischer, H.; Orth, H., "Die Chemie des Pyrrols", Vol. III, PP290. Sparatore, F.; Manzerall, D., |. Org. Chem., 1960, 25_, 1073. Smith, K. M.; Goff, D. A., |. Org. Chem., 1985 59, 2073. Inhoffen, H. H.; Jaeger, P.; Mahlhop, P., lustus Liebigs Ann. Chem., 1971, _7_4_2, 109. Harrison, P. J.; Fookes, C. J. R.; Battersby, A. R., J. Chem. Soc. Chem. Commun. 1981, 797. Battersby, A. R.; Reiter, L. A., |. Chem. Soc. Perkin. Trans. I, 1984, 2743. Paine, J. K; Chang, C. K.; Dolphin, D., Heterocycles,1977, _7_, 831. Smith, K. M.; Parish, D. W.; Inouge, W. S., |. Org. Chem., 1985, 51, 666. Battersby, A. K; Hamilton, A. D.; Mcdonald, E.; Mombelli, L.; Wong, O. H., J. Chem. Soc. Perkin. Trans. I 1980, 1283. Kenner, G. W.; McCombie, S. W.; Smith, K. M., |. Chem. Soc. Chem. Commun. 1972, 1347. Mancuso, A. J.; Brownfain, D. 5.; Swern, D. J., |. Org. Chem., 1979, 23, 4148. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 198 Core , E. J.; Gilman, N. W.; Ganem, B. G., I. Am. Chem. Soc., 1968, 20, 16. Corey, E. J.; Schmidt, G., Tetflretra Lett., 1979, 399. Fischer, H.; Ecoldt, H., lustus Libigs Ann. Chem. 1940, 5_44_, 159. Dar‘eva and Miklukhin, |. Gen. Chem. USSR, 1959, 2_9_, 620. Sievers, R E., Ed., "Nuclear Magnetic Resonance Shift Reagents", Academic, New York, 1973. Battersby, A. R., |. Chem. Soc. Chem. Commun., 1979, 539. Storzenberg, A.M.; Laliberte, M. A., I. Org. Chem., 1987, _52, 1022. Mason, S. F., |. Chem. Soc. Chem.,1958, 976. Matthew, J. M.; Lewis, M. S.; Heery, K., J, Biol. Chem., 1973, 24_8, 2801. Bellamy, L. J., "The Infrared S ectra of Complex Molecules", 3rd Edition, 1975, Chapan and Hall, ndon. Ching, Y.; Ondrias, M. R.; Rousseau, D. L.; Muhoberac, B. B.; Whaton, D. C., FEBS Lett. 1982, 138, 239. Boktos, G.,1981, Master Thesis, Michigan State University. Cotton, T. M.; Timkovich, R.; Cork, M. 5., FEBS Lett. 1981, 133,39. Ondrias, M. R.; Carson, S. D.; Hirasawa, M.; Knaff, D. B., Biochem. Biophys. Acta,1985, 820, 159. Lutz, M.; Hoff, A. J.; Brehamet, L., Biochem. Biophys. Acta, 1982, _6_7_2, 331. Cotton, T. M.; Van Duyan, R. P., FEBS Lett., 1982, 147, 81. Anderson, L. A.; Loehr, T. M.; Sotiriou, C.; Chang, C. K., I. Am. Chem. Soc. 1987,1294 258. Chang, C. K.; Barkigia, K. M.; Hanso, L. K.; Fajer, J., I. Am. Chem. Soc., 1986, 108, 1352. Suh, M. P.; Pwepston, P. N.; Ibers, J. A., I. Am. Chem. Soc., 1984, mg, 5164. Barki ia, J.; Fajer, J.; Chang, C. K.; Willams, G. J. B., I. Am. Chem. fig, 982, E44315. Chan , C. K.; Hanson, L. K.; Richardson, P. R.; Young, R.; Fajer, J., Proc. at1.Acad. Sci. U. S. A. 1981,28, 2652. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 199 Stortzenber , A. M.; Strauss, S. H.; Holm, R. H., I. Am. Chem. Soc., 1981, m 4 63. Storztenberg, A. M.; Glazer, P. A.; Foxman, B. M., Inorg Chem., 1986, 25, 983. Yamanaka, T.; Okunuki, K., Biochem. Z., 1963, 338, 62. Hill, K. E.; Whaton, D. C., J. Biol. Chem., 1978, 2_5_3, 489. Weeg-Aerssens, E.; Tiedje, J. M.; Hausinger, R. P.; Averill, B. A., J_. Bacteriol., 1988, (in press). Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K., Anal. Biochem. 1985, m, 76. Walsh, T. A.; Johnson, M. K.; Barber, D.; Thomson, A. T.; Greenwood, C., |. Inorg. Biochem.,1980, 14, 15. Yamanaka, T.; Okunuki, K., Biochem. Biopyhs. Acta 1963, 61, 407. Yamanaka, T.; Okunuki, K., Biochem. Biopyhs. Acta 1963, 62, 379 and 406. Kadama, T., Plant Cell Physiol., 1970, 11, 231. Weeg-Aerssens, E.; Tiedje, J. M.; Averill, B. A., I. Am. Chem. Soc. 1988, (in press). Barkigia, K.; Wu, W.; Fajer, J.; Chang, C. K., unpublished results. Mayo, S. L.; Ellis, W. R.; Crutchley, R. J.; Gray, H. B., Science 1986, E3, 948. "lllllllllllllll“