14 Biological Chemistry Part (iii) Tetrapyrroles and their Biosynthesis By D. G. BUCKLEY Chemistry Department Queen Mary College Mile End Road London El 4NS 1 Introduction In the three years since porphyrins and related compounds were last reviewed in Annual Reports,’ a great deal of new work has been published much of which relates to the biosynthesis of porphyrins chlorins and corrins and to details of their biological activity. Most of this work relates directly to results published since the late 1960’s and where appropriate a brief summary of this earlier work will be included.* The publication of the authorititative ‘Porphyrins and Metallopor- phyrins’ edited by K. M. Smith,* and based on Falk’s original was most welcome as was the appearance of a full account4 of the important Royal Society discussion on the biosynthesis of porphyrins chlorophyll and vitamin BI2held in London in February 1975.The informal ‘Tetrapyrrole Discussion Group’ has been founded recently as a result of the increased activity in this area.? CO,H 2 Me Me I I CO H CO H (1) Porphyrin (2) Cobyrinic acid * The period on which this review is based is January 1975 to March 1978. t Hon. Secretary/Treasurer Dr. S. B. Brown Department of Biochemistry University of Leeds 9 Hyde Terrace Leeds LS2 9LS U.K. ’ A. H. Jackson Ann. Reports (B),1974,71 519 ‘Porphyrins and Metalloporphyrins’ ed. K. M. Smith Elsevier Amsterdam 1975. J. E. Falk ‘Porphyrins and Metalloporphyrins’ Elsevier Amsterdam 1964. Phil. Trans. Roy.SOC.,1976 B273,pp. 75-357. 392 Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis Both porphyrins and chlorins are based on the porphyrin (porphin) ring system (l),while the cobalamins are corrin derivatives e.g. cobyrinic acid (2). The newer and more systematic numbering system will be used in this article; the older descriptions for the four meso carbon atoms (a,p 7 and 6) are given for reference in (1). 2 Biosynthesis of Tetrapyrroles General Remarkable progress has been made in our understanding of the biosynthesis of the haems chlorophylls and corrins over the past thirty years and a summary of what was known’ at the end of 1974 is given in Scheme 1. Porphobilinogen (PBG) ALA J e- Me (4) Copro’gen-11I (3) Uro’gen-111 >)< 1 The Cobalamins (5) Proto’gen-IX (6) Protoporphyrin-IX Here and elsewhere A = CH2C02H Y kg P = CH2CH,C02H The Haems The Chlorophylls Scheme 1 Reviewed by A.R. Battersby and E. McDonald in ref. 2 pp. 61-122. 394 D. G. Buckley normally derived from glycine and succinyl coenzyme A by way of 5-amino-laevulinic acid (ALA) is condensed to give uroporphyrinogen-I11 (abbreviated to “uro’gen-111”) (3) as the next discrete (i.e. detectable) intermediate. Modification of the side-chains then occurs to give protoporphyrinogen-IX (“proto’gen-IX”) (5) by way of coproporphyrinogen-I11 (“copro’gen-111”) (4); oxidation of the porphy- rinogen system follows to give protoporphyrin-IX (6) which either leads directly to the iron-containing haems or else to the chlorophylls after insertion of magnesium followed by further modification.The pathway to the cobalamins is thought to branch at the uro’gen-I11 stage; one of the bridging methylene groups in uro’gen- I11 (3) must be removed to give the corrin ring system cf. (2). Re~ently,~ efforts have been directed towards elucidating the details of various steps in the above biosynthetic pathway; for clarity this work will be divided into four main sections (a) the specific formation of uro’gen-111; (b) modification of side-chains and subsequent aromatization of protoporphyrin-IX; (c) the iron and magnesium branches; and (d)the biosynthesis of vitamin B,*. 3 Biosynthesis of Uro’gen-111 the ‘Type-111 Problem’ Early work established the crucial role of PBG in the biosynthesis of protohaem and in turn PBG was shown to be synthesized from glycine and succinyl coenzyme A by way of ALA.’ Recent work6 has shown that ALA synthetase catalyses the condensation of glycine and succinyl-CoA by specific removal of the pro-2R- hydrogen of glycine; the pro-2s-hydrogen of glycine was shown to occupy the pro-5s-position in the derived ALA (Scheme 2).Further have now revealed the fate of the two hydrogen atoms at C-5 in ALA during its conversion into PBG by ALA dehydratase. C-5 of one ALA molecule provides the aminomethyl side-chain of PBG and the substituents at C-5 are incorporated intact. Position 2 of PBG is derived from C-5 of the second ALA molecule; specific loss of the pro-5R-hydrogen of ALA occurs during the aroma- tization step to yield PBG that has the pro-5s-hydrogen of ALA at C-2 (Scheme 2).6-8 These results suggest that the removal of the pro-R-hydrogen atoms from an intermediate such as (7)occurs while the species is enzyme-bound and not after its release into the medium [(7)+ (7a) +PBG] as is implicated in the otherwise excellent mechanistic sequence proposed earlier by Nandi and Shemin.’ The pioneering studies of Shemin Granich Bogorad Neuberger and Riming- ton5 established that the co-operative action of two enzymes porphobilinogen deaminase (‘deaminase’) and uroporphyrinogen-I11 cosynthetase (‘cosynthetase’) was required to catalyse the conversion of faur molecules of the monopyrrole PBG into uro’gen-111 (3) and ammonia (Scheme 3).In the absence of cosynthetase deaminase catalyses the conversion of PBG into uro’gen-I (8) the isomer that would be expected from single head-to-tail combination of four PBG units. However uro’gen-I (8) is not transformed into uro’gen-111 (3) by cosynthetase alone nor by the complete deaminase-cosynthetase system. These results sum- ‘M. M. Abboud P. M. Jordan and M. Akhtar J.C.S. Chem. Comm. 1974,643. ’ M. Akhtar M. M. Abboud G. Barnard P. Jordan and 2. Zaman in ref. 4 pp. 117-136. a M. M. Abboud and M. Akhtar J.C.S. Chem. Comm. 1976 1007. D. L. Nandi and D. Shemin J. Biol. Chem. 1968,243 1236. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis C02H Hs synthetase 0 [2,2-3H2]glycine (5S)-[5-3H1]ALA H N -H,C 10 H H2N (11S)-[2,1 l-3H2]PBG + Enz-NH2 H,N-H,C H H Scheme 2 deaminase yinaseit;;ynthetase a,on/ A 'a: +4NH3 + :a A +4NH3 P (8) Uro'gen-I (3) Uro'gen-I11 Scheme 3 396 D.G. Buckley marize the 'type-I11 problem' how does the complete enzyme system bring about the molecular rearrangement required to produce the unexpecfed type-111 isomer and what are the intermediates in this intriguing transformation? The Nature of the Rearrangement.-Until 1973 there were no answers to these questions although some 25 hypothetical mechanisms had been prop~sed.~,~~ However in a series of 13Cn.m.r. experiments," Battersby and McDonald were able to deduce what happened during the biosynthesis of uro'gen-111 (3) from PBG i.e.they were able to deduce the origin of each of the carbon atoms which make up the porphyrin macrocycle. Full details af this work have now been published" and are summarized below. [2,1 l-'3C2]Porphobilinogen (9) was prepared from 90 atom YO '3C-labelled ALA (1O),I3 using the enzyme ALA-dehydratase. No dilutions were made throughout so 81% of the PBG molecules were doubly labelled as in (9). This PBG (1 part) was diluted with normal unenriched PBG (4parts) and the resultant PBG sample was incubated with a PBG-free enzyme system from the alga Euglena gracilis. The product was protoporphyrin-IX [Iabelled (6)]; this was formed because the biological system contained all the enzymes necessary to effect the conversion of the uro'gen-111 [labelled (3)] as it was formed through the steps in Scheme 1 as far as [labelled (6)].After I3C n.m.r. analysis this product was converted chemically into the labelled diketone (1 1) as shown in Scheme 4. (9) Reagents i Enzyme preparation from E. gracilis; ii HBr-HOAc HzO; iii CH2N2; iv Na2Cr207-HZS04-HZO. Scheme 4 lo Ref. 5 pp. 85-87. See also E. Margoliash Ann. Rev. Biochem. 1961,30,551;J. H. Matheson and A. H. Corwin J. Amer. Chem. SOC.,1961,83 135; E. Bullock Nature 1965 205 70; E. B. C. Llambias and A. M. del C. Batlle Biochem. J. 1971,121,327;R. Radmer and L. Bogorad Biochemistry 1972 11 904. A. R. Battersby E. Hunt and E. McDonald J.C.S. Chem. Comm. 1973,442. l2 A. R. Battersby G. L. Hodgson E. Hunt E. McDonald and J. Saunders J.C.S. Perkin I 1976 273.l3 A. R. Battersby E. Hunt E. McDonald and J. Moron J.C.S. Perkin I 1973 2918. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis 397 The low-field part of the 'H-decoupled 13C n.m.r. spectrum of the labelled protoporphyrin-IX (6) contained signals from the meso-carbon atoms C-5 C-10 and C-20 as 5.5Hz doublets while that for C-15 was a 72Hz doublet. The chemical shifts for C-5 C-10 C-15 and C-20 were assigned14 by rational total synthesis of (6) that was specifically labelled at C-10 C-15 and C-20 respectively. Also the couplings of 72 Hz and 5.5 Hz were proved to correspond respectively to direct coupling of two adjacent 13Catoms and to I3C-l3Ccoupling through three bonds as in Scheme 5 by the unambiguous preparation12 of the appropriate doubly labelled porphyrins.The spectrum of the derived diketone (ll) in which the 13C n.m.r. signals for the four meso-carbon atoms are even more clearly differentiated showed similar features. Thus it was clearly established that the major components in the labelled sample of protoporphyrin-IX (6) that was derived from the incubation experiment were the four doubly labelled species shown in Scheme 5. , Me\ O ' U/ Me ii ii C0,Me C0,Me C0,Me C0,Me 5.5 Hz 5.5 Hz C0,Me C0,Me C0,Me C0,Me 5.5 Hz 72 Hz Scheme 5 Exactly the same conclusion was derived from studies" of the incorporation of the same diluted [2,11-'3C,]PBG (9) into protoporphyrin-IX [labelled (6)] by a mixed enzyme preparation from chicken blood and beef mitochondria.Further- more it has been shown15 that during enzymic conversion of uro'gen-I11 (3) into l4 A. R. Battersby G. L. Hodgson M. Ihara E. McDonald and J. Saunders J.C.S. Perkin I 1973 2923; A. R. Battersby M. Ihara E. McDonald J. Saunders and R. J. Wells ibid. 1976 283. 15 B. Franck D. Gantz F.-P. Montforts and F. Schmidtchen Angew. Chem. Internat. Edn. 1972 11 421; A. R. Battersby J. Staunton and R. H. Wightman J.C.S. Chem. Comm. 1972 1118; A. R. Battersby E. McDonald J. R. Redfern J. Staunton and R. H. Wightman J.C.S. Perkin I 1976 266. 398 D. G. Buckley protoporphyrin-IX (6) (see Scheme l) the macrocycle remains intact and unscrambled. All the results obtained for protoporphyrin-IX (6) therefore hold good for uro’gen-I11 (3).From these experiments the nature of the rearrangement process by which type-I11 porphyrins are biosynthesized has been defined and it is characterized by the following features (a) the three PBG units which form ring A and its attached C-20 bridge ring B and the C-5 bridge and ring c with its C-10 bridge are all incorporated intact without rearrangement; (b),the PBG unit which forms ring D is built in with rearrangement which is intramolecular with respect to that PBG unit; and (c) the rearranged carbon atom forms the bridge at C-15. It is of considerable interest that exactly the same features characterize the biosynthetic process which generates type-I11 porphyrins in such widely different organisms as an alga (Euglena gracilis) a chicken and a bacterium (Propionibacterium sher- manii).l6 The Role of Cosynthetase Timing of the Rearrangement.-The known’ deaminase-catalysed conversion of PBG into uro’gen-I (8) is usually con~idered~~’*~* to occur by simple head-to-tail combination of four PBG units without rearrangement.Recent studiesI6 with [2,11-13C2]PBG(9) and a purified preparation of deaminase from E. gracih have confirmed this view; 13C n.m.r. analysis of the recovered I3C-labelled uroporphyrin-I octamethyl ester showed that each of the four PBG units had been incorporated intact into uro’gen-I (8). A plausible sequence for the interaction of PBG with deaminase is given in Scheme 6. The fact that deaminase alone catalyses head-to-tail coupling of four PBG units to produce uro’gen-I (8) while deaminase and cosynthetase together catalyse the formation of uro’gen-I11 (3) from three intact PBG units and one intramolecularly rearranged unit suggests that cosynthetase must bring about this rearrangement either by operating on an intermediate produced by deaminase or by modifying the way in which deaminase brings about one of the coupling steps.Cosynthetase might operate at any stage in the overall process i.e. at the monopyrrole dipyrrole tripyrrole or tetrapyrrole level and it should be possible to make a distinction by testing the hypothetical intermediates (12)-(15) of Scheme 6 (and rearranged isomers) as precursors of uro’gen-I11 (3). Early work5,” suggested that rearrangement does not occur as the first step although incorporation studies have not been reported.Dipyrrolic Intermediates in Uro’gen-I11 Biosynthesis.-Pyrromethanes e.g. (16) have been prepared 18-20 in several laboratories and tested for incorporation into type-XI1 porphyrins in the presence of the deaminase-cosynthetase system. The l6 A. R. Battersby E. McDonald R. Hollenstein and D. C. Williams unpublished work Cambridge 1976; personal communication. l7 A. T. Carpenter and J. J. Scott Biochem. Biophys. Ada 1961,52 195. A. R. Battersby D. A. Evans K. H. Gibson E. McDonald and L. N. Nixon J.C.S. Perkin I 1973 1546; A. R. Battersby J. F. Beck and E. McDonald ibid. 1974 160. l9 B. Frydman and R. B. Frydman Accounts Chem. Res. 1975,8 201. 2o e.g. J. Bausch and G. Miiller Enzyme 1974,17,47;J. M. Osgerby J. Pluscec Y. C.Kim F. Boyer N. Stojanac H. D. Mah and S. F. MacDonald Canad. J. Chem. 1972,50 2652. 399 Biological Chemistry-Part (iii) Tetrapyrroles and their Biosynthesis +-i-=7-NH3 PBG uroporphyrinogen-I (8) NH3 or Jh X =NHz or group X group of deaminase Scheme 6 on enzyme initial results of such were confusing and led to some disagreement between the groups at Buenos Aire~,’~.’~ and (later) Yale.24 Cambridge,5*zz*z3 All researches on the enzymic incorporation of PBG and pyrromethanes into porphyrinogens are complicated by concomitant non -enzymic conversion of sub-strates into porphyrinogens (often accompanied by some prior rea~rangement),’~ and this results in a troublesome blank. This is a major difficulty encountered by all workers in the area and undoubtedly is an important factor in the differences in results indicated above.The interested non-expert must remember that a I4C-labelled precursor has not been proved to be incorporated until (a)the precursor is shown to be chemically (including isomerically) pure; (b)the isolated products are proved to be chemically (including isomerically) and radiochemically pure; and (c) the incorporations observed are proved to be specific and the site(s) of labelling are determined. Recent results from the Cambridge groupz6 with both 13C- and 14C-labelled aminomethylpyrromethanes and with 13C-labelled bilanes (see later) have B. Frydman R. B. Frydman A. Valasinas E. S. Levy and G. Feinstein in ref. 4 pp. 137-160. z2 A. R. Battersby D. A. Evans K.H. Gibson E. McDonald L. N. Mander and J. Moron J.C.S. Chem. Comm. 1973 768. 23 A. R. Battersby and E. McDonald in ref. 4 pp. 161-180. 24 A. I. Scott K. S. Ho M. Kajiwara and T. Takahashi J. Arner. Chem. Soc. 1976,98 1589. ’’G. W. Kenner A. H. Jackson and D. Warburton J. Chem. SOC.,1965 1328. 26 (a) Reviewed by A. R. Battersby in the Paul Karrer Lecture University of Zurich (July 1977) Experientia 1978,34 1; (6)See also A. R. Battersby and E. McDonald Accounts Chem. Res. 1978 11,in the press. 400 D. G. Buckley clarified the role of pyrromethanes in uro'gen-I11 (3) biosynthesis and their results are presented first. Preliminary had suggested that only the pyrromethane AP.AP (16) was involved in uro'gen-I11 biosynthesis. When the intact enzyme systems from E.grucilis or avian erythrocytes were used radioactivity was found in the isolated protoporphyrin-IX (6) from ['4C]AP.AP (16) (range 1.6-9.0% with duck eryth- rocytes and 0.2-0.5% with E. grucilis)and from [14C]PA.AP (18) (range 0.06- 0.12% from duck erythrocytes and 0.01-0.015% from E. gracilis); even lower maximum incorporations were obtained from incorporation studies with ['4C]AP.PA (17) (0.05%) and [14C]PA.PA (19) (0.09°/~).The incorporations of the rearranged pyrromethanes (17) (18) and (19) were so low that the labelling pattern could not be checked by degradation. AflAm NH HN /' NH HN /A H,NJ H,NJ (16) AP.AP (17) AP.PA p&-PJNH HN /' P mNHHN / A H2N J H,NJ (18) PA.AP (19) PA.PA In contrast (Scheme 7) the relatively high level of incorporation of radioactivity from the unrearranged isomer (16a) into the isolated protoporphyrin-IX dimethyl ester (21a) and haemin ester (20a) using the enzyme system from duck erythrocytes allowed oxidative degradation of the latter to the four biliverdins-IXa -IX& -IXy and -1XS [the IXa isomer (22) is illustrated in Scheme 7].27 The haemin ester isolated after incubation of [14C]AP.AP (16a) with enzymes but without PBG was degraded to the four biliverdins which showed that the I4C activity of the haemin ester (20) was distributed equally between C-5 and C-15.It followed that the [14C]AP.AP (16a) had been specifically incorporated into uro'gen-I11 (3) to label C-5 and C-15 equally.22,28 This specific incorporation of AP.AP (16) without PBG into uro'gen-I11 (3) was also found2* with the enzymes from E.grucilis. The initially formed uro'gen-I11 (3) was again enzymically converted in situ into protoporphyrin-IX (6); incubation of [bridge-methylene-'3C]AP.AP (16b) led to recovery of [5,15-'3C2]protoporphyrin-IX ester (21b) whereas [~minomethylene-'~C]AP.AP (16c) gave [10,20-'3C2]pro- toporphyrin-IX ester (21c) as in Scheme 7. The sites of labelling were read directly from the I3Cn.m.r. spectra using the ear!ier unambiguous assignment^.'^ The Frydmans have that the above incorporations with AP.AP (16) may not be leading to protoporphyrin-IX (6) as described but to an unnatural '' R. Bonnett and A. F. McDonagh Chem. Comm. 1970,237. A. R. Battersby Special Lecture 23rd Znternat.Congr. Pure Appl. Chem. 1971,5 1. 29 R. B. Frydman and B. Frydman F.E.B.S. Letters 1975 52 317. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis 401 enzymes b (16a) 14Cat * (16b) llIC at 0 (16c) Cat ji!kZEtion (20)X = Fe"'-CI (20a) 14cat * (21) -P X t = H H (21a) 14cat * (21b) EC at (21c) Cat rn Uro'gen-IV [plus isomers] [enzymes \ \ COzH CO,H C0,Me C0,Me (23) LJ$abelled (22) Biliverdin-IXa ester (23a) Cat * 14cat * (23b) ll:C at 0 [plus IXp IXy and 1x6 isomers] (23c) Cat Scheme 7 isomer protoporphyrin-XI11 (23). Their view was based on the known chemical formation (by rearrangement) of some uro'gen-IV from AP.AP (16),22,29 which could then be converted enzymically via copro'gen-IV into protoporphyrin-XI11 (23) (Scheme 7); the terminal enzymes of protoporphyrin-IX (6) biosynthesis (see Scheme 1)are known5 to act on these unnatural isomers.This suggestion cannot be correct. It is clear that the product obtained by the Cambridge group is unquestionably protoporphyrin-IX (6) for the following reasons. First the recently proved structure of protoporphyrin-XI11 (23)30has a plane of symmetry which renders C-10 and C-20 equivalent [cf. (21)]; the four 30 H.M. G. AI-Hazimi A. H. Jackson D. J. Ryder G. H. Elder and S. G. Smith J.C.S. Chern. Cornrn. 1976 188; L. Mombelli E. McDonald and A. R. Battersby Tetrahedron Letters 1976 1037; G. Buldain J. Hurst R. B. Frydman and B. Frydman J. Org. Chem. 1977,42 2953. 402 D. G. Buckley different 13C n.m.r.signals observedZ2 from the bridge carbon atoms of the labelled biosynthetic products (2 lb) and (21c) could not have arisen from protoporphyrin- XI11 (23). Second the haemin ester derived from a protoporphyrin of structure (23) could not be degraded27 to any of the known biliverdin-IX isomers obtained22’28on degradation of the protoporphyrin recovered from the incubation experiments (see Scheme 7). Third protoporphyrin-IX and protoporphyrin-XI11 dimethyl esters are separable by h.p.l.c. and the product from the above enzymic experiments was distinguished in this ‘kay from isomer XI11 and identified with isomer IXZ6’ These experiments prove conclusively that two molecules of unrearranged AP.AP (16) can combine and rearrange in a highly specific manner to generate uro’gen-I11 (3) in the presence of the deaminase-cosynthetase system.The prob- lem of determining the extent of chemical versus enzymic formation of uro’gen-I11 (3) in the incubation experiment26 was solved by the development of methods for isolating the derived uroporphyrins from highly reactive pyrr~methanes~’ and for quantitative ~eparation~~ of the recovered porpbyrin isomers. The isolated mixture of uroporphyrin esters was decarboxylated with hot to the cor- responding mixture of coproporphyrin isomers; a two-stage h.p.1.c. separation of the tetramethyl esters allowed the four coproporphyrin isomers to be collected and assayed.32 Incubation3’of [14C]AP.AP (16a) at pH 7.2 for 16h without enzyme gave a ca. 30% yield of uroporphyrins having the isomer composition shown in Table 1.When a strictly parallel experiment was run but with the addition of purified deaminase-cosynthetase from E. gracilis the isomer ratio was dramatically altered (see Table l) type-I11 formation being vastly increased (3’/0+54~/0) at the expense of type-I formation (68%+15%). The competing chemical formation of the type-IV isomer was essentially unaffected by deaminase-cosynthetase. Table 1 Radioactivity of the uro’gen isomers formed from AP.AP (16a) Uro’gens formed (“10of total) ~ r 7 Type-I Type-II Type-111 Typ-IV Blank run 68*2 1*L 3*2 28*2 Enzyme run 15*1 2*1 54*2 29*2 In addition the doubly 13C2-labelled form of AP.AP (16d) synthesized18*20 from [2,ll- 13C2]PBG (9) was with purified deaminase-cosynthetase and then worked up as in the 14Cseries31 to give coproporphyrin-111 tetramethyl ester.The 13Cn.m.r. signal that arises unambiguously from C-15 was a 72 Hz doublet centred on a smaller singlet. It followed that the labelling pattern around C-15 was that shown in Scheme 8 for uro’gen-IIIC(3d) and (3e)] and the signal from C-5 was consistent with the illustrated arrangement around ring B. The ratio of split to 31 A. R. Battersby D. G. Buckley E. McDonald and D. C. Williams J.C.S. Chem. Comm. 1977 115. 32 A. R. Battersby D. G. Buckley G. L. Hodgson R. E. Markwell and E. McDonald in ‘High Pressure Liquid Chromatography in Clinical Chemistry’ ed. P. F. Dixon C. H. Gray C. K. Lim and M. S. Stoll Academic Press London 1976 pp.63-70. 33 Ref. 2 p. 825. 34 A. R. Battersby D. W. Johnson E. McDonald and D. C. Williams J.C.S. Chem. Comm. 1977 117. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis NH HN /p m H,N-/ A NH HN /’ H,NJ (16d) 13Cat 0 (1 part) + (16) Unlabelled (3 parts) deaminase-cosynthetase [plus unlabelled (3)] Scheme 8 unsplit signals from C-15 was in agreement with an intramolecular rearrangement of ring D exactly as had been found earlier for PBG These results show clearly that deaminase-cosynthetase does catalyse the formation of uro’gen-I11 (3) when AP.AP (16) is provided and further that the enzymic conversion involves an intramolecular rearrangement. Nevertheless they do not prove that free AP.AP (16) is a biosynthetic precursor which is normally dimerized by the enzymes.Eventually it became clearz6 that two molecules of AP.AP (16) might be reacting chemically to produce the unrearranged tetrapyrrole system (26) which as it is formed could be transformed enzymically into uro’gen- 111 as in Scheme 9 (see below). At this stage it is appropriate to consider the results of other workers. In their work with pyrromethanes the Frydman~’~~~~ employed enzymes from wheatgerm and used paper chromatography to separate the isomeric porphyrins prior to radioassay. They found that AP.AP (16) in the presence of PBG was converted (ca. 1%) into uro’gen-I (8) by deaminase alone but they did not observe incorporation of AP.AP (16) into uro’gen-I11 (3) with deaminase-cosynthetase when they used a short incubation (1h).In the light of the results quoted pre- viously it seems probable that the decisive differences between the experiments of the groups at Cambridge and Buenos Aires were the duration of the incubation and 35 B. Frydman S. Reil A. Valasinas R. B. Frydman and H. Rapoport J. Amer. Chem. Soc.,1971 93 2738; R. B. Frydman A. Valasinas and B. Frydman Biochemistry 1973,12,80; R. B. Frydman A. Valasinas H. Rapoport and B. Frydman F.E.B.S. Letters 1972,25 309. 404 D. G. Buckley the concentration of pyrromethane. When the rearranged [‘4C]PA.AP[labelled (18)J was incubated with deaminase-cosynthetase together with PBG radioac- tivity (ca. 0.7%) appeared in the uro’gen-111 (3). Later incubated the ‘headless’ pyrromethane (24) with PBG and deaminase-cosynthetase and found that there was 0.15-0.89% yield of radioac- tivity in the type-I11 porphyrins.Both sets of results were thought to prove that the (24) 14Cat * synthesis of type-I11 porphyrin is controlled at the pyrromethane level. However the evidence given above (and some of what follows) is so overwhelming that this cannot be correct The small transfers of radioactivity from rearranged pyr- romethanes into uro’gen-I11 (3) could reasonably be explained as arising from various well-known minor chemical ~ide-reactions.~~ The amounts of radioactivity found in the type-111 porphyrins in both of these studies with rearranged pyr- romethanes are of the same order ( < 1YO)as were found by the Cambridge groupz3 for the three rearranged aminomethylpyrromethanes (17)’ (18) and (19).Furthermore none of the foregoing weakly radioactive type-111 porphyrins from experiments with rearranged pyrromethanes has been degraded so as to locate the labels.* Proof that Rearrangement Occurs at the Tetrapyrrole Level.-Groups at Stutt- ga~t~~ and Cambridge37 have reported the involvement of the bilane (26) in .~~ uro’gen-I11 biosynthesis. Muller et ~1prepared the bilane (26) and reported that it was transformed into a uro’gen mixture having a type-I type-111 ratio of 84 16 by deaminase-cosynthetase from P. sherrnanii. From a different synthesis Battersby and McDonald et al.37prepared the crystalline lactam ester (25) which was hydrolysed to the same bilane (26) corresponding to unrearranged head-to- tail joining of four PBG units.This cyclked chemically (without enzyme) at pH 7.2 to give virtually pure (>95%) uro’gen-I (8) isolated as uroporphyrin-I and analysed by h.p.1.c. as the coproporphyrin isomer as described previou~ly.~~ However incubation of the bilane (26) with purified deaminase-cosynthetase gave a markedly different result the products after aromatization were uroporphyrin-I11 (70%) derived from (3) and uroporphyrin-I (30%) derived from (8). The differences in the isomer ratios from the two experiments were probably due to the different concentrations of purified enzymes but both sets of results strongly suggested that the unrearranged bilane (26) is a key precursor in the biosynthesis of uro’gen-I11 (3).The Cambridge group have now established beyond doubt that this is the case.26 *The Reporter has attempted to set out clearly the relevant work in this area but must declare his interest in having been involved in the Cambridge effort. 36 H.-0. Dauner G. Gunzer I. Heger and G. Muller 2.physiof. Chem. 1976,357 147. 37 A. R. Battersby E. McDonald D. C. Williams and H. K. W. Wurziger J.C.S. Chem. Comm. 1977 113. Biological Chemistry-Part (iii) Tetrapyrroles and their Biosynthesis [ 15-13C]Bilane (26a)37 was converted into a product consisting of 80% uro’gen-111(3a) and 20% uro’gen-I (8a) by the deaminase-cosynthetase system as shown by h.p.1.c. analysis of the derived coproporphyrin esters. The crystalline copro- porphyrin-I11 ester was shown by I3C n.m.r.to be labelled specifically at C-15 by comparison with the spectrum of unambiguously synthesized [ 15-13C]copropor-phyrin-I11 tetramethyl ester. This established specific incorporation of the bilane (26a) into uro’gen-I11 (3a). The two [13C,]bilanes (26b) and (26c) were prepared3* in such a way that the enrichment was 90 atom% at each labelled site and so 81% of the molecules of each bilane sample were doubly labelled as illustrated in Scheme 9. After dilution with unlabelled bilane the two mixtures were incubated separately with purified deaminase-cosynthetase. The locations of labels in the resultant samples of uro’gen-I11 (3b) and (3c) were shown to be as illustrated in Scheme 9 by the usual isolation decarboxylation and 13C n.m.r.analysis; [20-13C]coproporphyrin-III tetramethyl ester was synthesized to allow 0 Me0,C (25) Unlabelled (26) %labelled (25a) 13C at C-15 (26a) C at C-15 (25b) lltC at (26b) lltC at 0 (2%) Cat A (26c) Cat A enzymic J 1 (8) lJ$abelled (3) Unlabelled (8a) C at one bridge (3a) 13C at C-15 (8b) ::Cat0 (3b) 13Cat. (8c) Cat A (3c) 13cat A Scheme 9 ’’ A. R.Battersby C. J. R. Fookes,E. McDonald and M. J. Meegan J.C.S. Chem. Comm. 1978 185. 406 D. G. Buckley the assignment of C-20 to be confirmed.26 Most importantly both I3C n.m.r. spectra showed a 72Hz doublet for the major resonance from the labelled bridge carbon atom;38 the two 13Catoms in each case have therefore become directly bonded as illustrated for (3b) and (3c) via an intermolecular reaction.The foregoing combined results prove that intact incorporation of unrearranged bilane (26) into uro’gen-III(3) occurs with inversion of ring D by an intramolecular process exactly as was found for PBG. Thus it is now certain that the biosynthesis of the natural type-I11 porphyrins chlorins and corrins involves the following steps. Four PBG units are joined together head-to-tail and the resultant bilane bound covalently or by physical forces to the enzyme system is then converted into uro’gen-I11 (3)by an intramolecular rearrangement which directly affects only ring D and the two carbon atoms which become C-15 and C-20.266 Battersby and McDonald have suggested26 that in the absence of cosynthetase deaminase catalyses the cyclization of the bilane to join C-20 and C-19 to give uro’gen-I (8) as shown in Scheme 6 but that in the presence of cosynthetase the combined enzyme system causes cyclization between C-20 and C- 16 rather than C-20 and C-19 (Scheme 10).This postulated attack would produce the spiro- intermediate (27); the labelling arising from (25b) and (2%) is shown. Frag-mentation of the C-15-C-16 bond followed by cyclization as shown would generate uro’gen-I11 (3) (Scheme 10). The spiro-system (27) is related to that proposed by Mathewson and C~rwin~~ in 1961 and its intermediacy is consistent with all the evidence available. Uro’gen-I11 (3) Scheme 10 4 The Pathway from Uro’gen-I11 to Protoporphyrin-IX Extensive studies’ over the past 25 years had indicated that the enzymic decar- boxylation of the four acetic acid side-chains of uro’gen-I11 (3)to give copro’gen- I11 (4) is a stepwise process and intermediates with seven six and five carboxy- groups had been detected in several laboratories.However it seemed that a single enzyme (or a highly organized enzyme complex) is able to effect the four successive decarboxylations although it must have a rather low substrate specificity. 39 J. H. Mathewson and A. H. Convin J. Amer. Chem. Soc. 1961,83 135. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis Pure heptacarboxylic acid porphyrins have been isolated from various sources including haemolysed avian erythrocyte^,^**^' the urine of patients with por- ~hyria,~~ and rat It now seems very probable that these porphyrins are identical and they have been variously named phyriaporphyrin porphyrin-208 and pseudouroporphyrin.From a series of spectroscopic analyses and by synthesis of three of the four possible isomers Battersby and McDonald showed that the heptacarboxylic acid porphyrin isolated from chicken erythrocytes had the struc- ture (28);40this assignment was confirmed subsequently by using the pyrromethene method.43 The heptacarboxylic acid porphyrin isolated from the faeces of rats poisoned with hexachlorobenzene was shown by Jackson et to have the same structure (28) by synthesis in this case using the b-oxobilane This work was part of the Cardiff group’s investigation of the porphyrins excreted during both normal and abnormal metaboli~m,~~ details of which are given in the following two sections.(28) Phyriaporphyrin-III Intermediates between Uro’gen-111 and Copro’gen-111.-Four heptacarboxylic six hexacarboxylic and four pentacarboxylic acid porphyrins might be formed by successive decarboxylations of the acetic acid side-chains during the conversion of uro’gen-III(3) into copro’gen-I11 (4) and a major aim of the Cardiff group was to discover whether or not there was a specific (or preferred) route and whether this was the same in both normal and abnormal metab01ism.~~ After extensive pre- liminary work on isolation procedures and analytical techniques it was possible to carry out preparative-scale separations of the porphyrins excreted in the faeces of rats that had been poisoned with hexachlorobenzene.Ip this way 10-20 mg of each of the octa- hepta- hexa- and penta-carboxylic fractions were obtained all of which were shown to be type-I11 porphyrins by decarboxylation to coproporphyrin- 111. The octacarboxylic acid was readily shown to be uroporphyrin-111; the other fractions were essentially homogeneous and their structures were then deduced and confirmed by synthesis. The heptacarboxylic acid had the structure (28) (see above) which is now generally termed phyriap~rphyrin-III.~’-~~ Of the six possible isomeric hexacar- 40 A. R. Battersby E. Hunt M. Ihara E. McDonald J. B. Paine 111 F. Satoh and J. Saunders J.C.S. Chem. Comm. 1974,994. 41 A. R. Battersby E. Hunt E. McDonald J.B. Paine 111 and J. Saunders J.C.S.Perkin I 1976 1008. 42 A. H. Jackson H. A. Sancovich A. M. Ferramola N.Evans D. E. Games S. A. Math G. H. Elder and S. G. Smith in ref. 4 pp. 191-206. 43 R. L. N. Harris A. W. Johnson and I. T. Kay J. Chem. SOC.(C) 1966,22. 44 cf. A. H. Jackson and K. M. Smith in ‘The Total Synthesis of Natural Products’ ed. J. Apsimon Wiley London and New York 1974 p. 144. 408 D. G. Buckley boxylic acid porphyrins which can be derived from uro’gen-I11 (3) only structure (29) was comEatible with the findings of a detailed spectroscopic analysis. This assignment was confirmed4* by synthesis of the hexamethyl ester by the MacDon- ald and by comparison with the ester derived from the natural compound. The pentacarboxylic acid was deduced to have the structure (30) by a combination of spectroscopic and biosynthetic arguments; other porphyrins derivable from structure (30) had been identified earlier.45 Synthesis of the pentamethyl ester of (30) and comparison with the ester of the natural compound confirmed this assignment .(29) R=A (30) R=Me Most of the possible isomeric penta- hexa- and hepta-carboxylic acid porphy- rins have been synthesized in Cardiff,42 and some have been synthesized indepen- dently by other groups.41y46 Biosynthetic studies with the porphyrinogens derived from the isomeric penta- hexa- and hepta-carboxylic acid p~rphyrins~~ showed that the porphyrinogens corresponding to the isolated porphyrins (28) (29) and (30)were metabolized at least as fast as those from the synthetic isomers.The fact that each of the isolated porphyrins was homogeneous strongly suggests that the parent porphyrinogens did not arise by a diversion from the normal metabolic pathway; the low specificity of the enzyme suggests that this would be most unlikely. All these combined with the that the heptacarboxylic acid porphyrins from haemolysed chicken erythrocytes and from humans with porphyria are the same as that from rat faeces strongly suggest that the porphyrinogens (31) (32) and (33) are the major intermediates in both the normal and abnormal metabolic conversion of uro’gen-I11 (3) into copro’gen-I11 (4). The major effect of hexachlorobenzene which is well known to induce porphyria in animals and humans,47 seems to be simply to slow down the overall rate of decarboxylation by reducing the activity of the enzyme.Thus the biosynthesis of copro’gen-I11 (4) from uro’gen-III(3) does involve a series of discrete steps as predicted by previous inve~tigators,~*~~ and the preferred (and possibly specific) pathway is shown in Scheme 11. One striking feature of the preferred pathway [(3) +(31) +(32) +(33) +(4)] is that decarboxylation occurs in a clockwise fashion starting with the acetic acid side-chain on ring D of uro’gen-I11 (3) and proceeding by successive decarboxyl- 4s M. S. Stoll G. H. Elder D. E. Games P. O’Hanlon D. S. Millington and A. H. Jackson Biochem. J. 1973,131,429. 46 P. S. Clezy T. T. Hai and P. C. Gupta Austral. J. Chem.1976,29 393. 47 Reviewed briefly in ref. 42. 409 Biological Chemistry -Part (iii) Tetrapyrroles arzd their Biosynthesis ..flP NH HN (4) Copro’gen-111 (33) Scheme 11 ations of the acetic acid residues on rings A B and C to form copro’gen-111 (4) (Scheme 11). It has been suggested that the uro’gen-III (3) in effect ‘performs a cartwheel’ on the enzyme surface as the side-chains are de~arboxylated.~~ However this intriguing suggestion is largely speculative and as yet the nature of the enzyme (or group of enzymes) involved is undefined. In contrast to the above result~,~*~~* which show that decarboxylation of the acetic acid side-chains of uro’gen-I11 (3) starts with ring D and finishes with ring c during its converion into copro’gen-I11 (4) only the acetic acid group on ring c of uro’gen-I11 (3) is decarboxylated during the formation of the corrin nucleus of vitamin BIZ,cf.(2). Thus the pathway to the corrins diverges from that to the chlorins and natural porphyrins immediately after the formation of uro’gen-I11 (3).48 Akhtar’s have defined the stereochemistry of the decarboxylation of the acetate side-chains of the cand D rings of uro’gen-111. (2R)-[2-2Hl,2-3H1]Succinic acid (34)49was incubated with a haemolysed chicken erythrocyte preparation and was incorporated into the acetate residues of uro’gen-III(35) (Scheme 12); this on decarboxylation generated chiral methyl groups in copro’gen-I11 (36) and then in haemin (37). The labelled haemin (37) was degraded’ to ethyl methyl maleimide and haematinic acid (38) which are derived from rings A and B and rings C and D respectively.Ozonolysis of (38)gave labelled acetic acid (39)’ which was converted 48 For recent reviews see (a)ref. 26a and (b) A. I. Scott Accounts Chem. Res. 1978,11,29. 49 G. F. Barnard and M. Akhtar J.CS. Chem. Comm. 1975,494. 410 D. G. Buckley (35) Uro'gen-I11 (36) Copro'gen-I11 (partial) 1 HT R = CT=CDT PI=CDTCDTC02H Scheme 12 into labelled malic acid (40).50a Incubation of (40) with fumarase and analysis of the results using the method5' of Cornforth and Arigoni showed that the labelled acetic acid (39) had the (S) configuration as illustrated in Scheme 12. Thus the decarboxylation reaction converting the uro'gen-III(3) into copro'gen-III(4) must have occurred as in (35)+(36) (Scheme 12) i.e.with retention of configuration. It is highly likely that this result holds good for decarboxylation of all four acetic acid side-chains although the above result only refers to acetic acid residues of rings c and D. Akhtar that the mechanism of the decarboxylation reaction involves enzymic protonation of an intermediate such as (35b) formed by enzyme-catalysed decarboxylation of the enzymically protonated porphyrinogen (35a) (Scheme 13). 5" (a)A. I. Rose J. Biof.Chem. 1970,245,6052; (b)J. W. Cornforth J. W. Redmond H. Eggerer W. Buckel and C. Gutschow Nature 1969,221 1212; J. Luthy J. Ritey and D. Arigoni ibid. p. 1213. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis 41 1 Biosynthesis of Protoporphyrin-IXfrom Copro’gen-111.-As was noted earlier the enzymes involved in the transformation of uro’gen-I11 (3) into protoporphyrin-IX (6) are not highly specific,’ and thus results from the testing of putative precursors must be interpreted carefully.Initial had suggested that hardero’gen-111 (4 1) was a natural intermediate between copro’gen-I11 (4) and proto’gen-IX (5) although (41) was incorporated only 6-10 times more efficiently than its isomer (42) using an enzyme preparation from E. grucilis and the absolute incorporations were low. A recent investigation by the Cardiff group has established5’ that hardero’gen-I11 (41) is indeed an intermediate on the pathway to proto’gen-IX (5).The isomeric porphyrins (43) and (44)were prepared labelled with tritium in the meso-positions and were reduced with sodium amalgam to the corresponding porphyrinogens (41) and (42). The latter were incubated separately with haemolysates of mature chicken erythrocytes and after work-up [which included dilution with unlabelled (6)] the isolated protoporphyrin-IX dimethyl ester was purified by h.p.1.c. The specific incorporations of (41) and (42)were found to be 34 and 0.8% respectively but after allowing for the stereospecificity of the later enzyme-mediated aroma- tization reaction (5)-+(6),53 it is clear that the absolute incorporations of hardero’gen-I11 (41) and its isomer (42) must have been ca. 70 and 1.5% respec-tively. Mea;e Me I \ P P (4) R’=R~=P (5) R’= R2= CH=CH2 (41) R’= CH=CH2 R2= P (42) R’= P R2 = CH=CH2 (6) R’=R2= CH=CH;! (43) R’= CH=CH2 R2 = P (44) R’ = P R2= CH=CH2 Further support for the intermediacy of (41) came from a kinetic study of the conversion of (4) into (5).In either the same avian system or in experiments with rat liver homogenates (using h.p.1.c. analysis of the porphyrin esters obtained on work-up) only hardero’gen-I11 (41) was formed as an intermediate and none of the isomer (42) could be detected. Thus the normal biosynthetic route to proto’gen-IX (5) involves formation of the vinyl side-chain on ring A before modification of the propionic acid residue on ring B. Oxidative Decarboxylation of the Propionic Acid Side-chains. Early work in Cambridge and Southampton showed’ that when the propionic acid residues on ’’ J.A. S. Cavaliero G. W. Kenner and K. M. Smith J.C.S. Chem. Comm. 1973 183; J.C.S. Perkin I 1974,1188. D. E. Games A. H. Jackson J. R. Jackson R. V. Belcher and S. G. Smith J.C.S. Chern. Comm. 1976 187. ’’ (a)A. H. Jackson D. E. Games P. Couch J. R. Jackson R. V. Belcher and S. G. Smith Enzyme 1974 17 81 88; (6) R. Poulson and W. J. Polglase J. Biol. Chem. 1975 250 1269; (c) A. R. Battersby E. McDonald J. R. Redfern J. Staunton and R. H. Wightman J.C.S. Perkin I 1976 266. 412 D. G. Buckley both rings A and B of copro'gen-I11 (4) were oxidatively decarboxylated to give proto'gen-IX (9,only the pro-S-hydrogen atom from the methylene group adjacent to the macrocycle was removed and that each of the two vinyl groups was derived from the remaining three hydrogen and two carbon atoms without rear- rangement.The original work by Akhtar54 which had defined the stereospecificity of hydro-gen removal was based on incubation studies with labelled succinic acid. Recent work by the Cambridge group which has confirmed the previous findings was based on incubation of PBG stereospecifically labelled at C-8. This latter approach circumvented the difficulties encountered in the earlier work due to the incorpora- tion of label into each of the eight side-chains of protoporphyrin-IX (6). (8S)-[8-3Hl]PBG (44)was synthesized26" as shown in Scheme 14. The initial hydrogenation step was accomplished using the new chiral homogenous catalyst (44) Scheme 14 developed by Knowle~;~~ the stereoselectivity was determined to be as shown by using model compounds and ca.92% of the derived amido-ester had the (2s) configuration. The labelled PBG (44) was mixed with ''C-labelled material to give a 3H:14C ratio of 8.7 1. This sample was converted enzymically into pro- toporphyrin-IX (6a) which had a 3H:14C ratio of almost one half of that of the PBG. Further degradation as illustrated in Scheme 15 (after hydrogenation of the vinyl groups) clearly showed that tritium had largely been lost during biosynthesis of the vinyl groups. The values found corresponded closely to those expected (shown in brackets) for ca. 92% configurational purity. This shows that it is the pro-8s-hydrogen atom of PBG which is removed during the formation of the vinyl groups of protoporphyrin-IX (6).Experiments using PBG specifically deuteriated in the propionic acid side-chain allowed the Cambridge group to determine the overall stereochemistry of the " Z. Zaman M. M. Abboud andM. Akhtar J.C.S. Chem. Comm. 1972 1263. 55 W. S. Knowles M. J. Sabacky B. D. Vineyard and D. J. Weinkauff J. Amer. Chem. SOC.,1975 97 2567. Biological Chemistry-Part (iii) Tetrapyrroles and their Biosynthesis .CO,H enzymic 3H I4C = 8.7 C02H CO,H (6a) 3H:I4C= 4.7 (4.7) H H A+B C+D 3H:14C= 0.64 H :I4C= 8.7 (0.7) (8.7) Scheme 15 oxidative decarb~xylation.~~ The required labelled PBG was prepared by reduc- tion of the dideuterio-acrylic ester (45) with di-imide which by its syn-stereo- specificity fixes the relative configuration at centres X and Y.This arrangement was not affected by the subsequent reactions which produced the racemate (46) + (47). An analysis of the possible outcome of overall anti- or syn-elimination (irrespec- tive of the precise mechanism) from the molecules (46) and (47) is given in Scheme 16; note that it is the pro-S-hydrogen atom which is lost from position X. By using a 'H n.m.r. analysis of the hydrogen atom at centre x in the derived proto- porphyrin-IX [structures (48)-(5 l)] neither product (SO)nor (5 1) from enan- tiomer (47) will register and a clear 'H-'H coupling pattern should be observed from products (48) or (49). Under suitable conditions separate signals could be seen from H on each of the vinyl groups of the protoporphyrin-IX dimethyl ester.So this ester derived enzymically from the labelled PBG [(46) and (47)] was examined by 'H n.m.r. in this way and two doublets with trans-coupling (18 Hz) were observed from the two H [see (48)]. Thus both vinyl groups of pro- toporphyrin-IX (6) are formed by an overall anti-elimination and the absolute stereochemistry of this process is as shown in Scheme 17. Also illustrated there is the important reaction which generates the vinyl group of isopentenyl pyrophos- hate;^' the correspondence is striking.26" Of all the mechanisms which have been proposed for formation of the vinyl groups of proto'gen-IX (5),only the two given in Scheme 18 are compatible with A. R. Battersby E. McDonald H.K. W. Wurziger and (in part) K. J. James J.C.S. Chem. Cumm. 1975,493. 57 J. W.Cornforth R. H. Cornforth G.Popjak and L. Yengoyan J. Bid. Chem. 1966 241 3970. 414 D. G. Buckley CO,CH,Ph HD "'y C0,CH2Ph + PhCH20eD N\ N H H \I DH (45) ,$+,C02CH2 Ph /% D"H JH H$ x' D (48) (49) Haem, PorphyrinHc + %HD Porphyrin < -+ chlorophyll and H~ cytochromes CO H Hc ?H HI)-Steroids and terpenoids Hc Scheme 17 the above data.* These represent possible mechanisms for the two known types of enzymes one of which requires oxygen while the other operates in anaerobic conditions. For this latter enzyme mechanisms involving hydride abstraction have * Variations of the mechanisms given in Scheme 18 are possible (see ref.5 pp. 102-104) but those given represent the two main types. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis been proposed but there is no experimental evidence in .this area. It has been suggested that the oxygen-requiring enzyme coproporphyrinogenase acts by first hydroxylating the propionic acid side-chain and then by catalysing the elimination of carbon dioxide and water; antiperiplanar elimination would require retention of configuration in the hydroxylation step (Schemes 17 and 18). Hydride acceptor H Then ring B Proto’gen-IX (as ring A) (5 ) Jwvwwy (4) Copro’gen-111 (41) Hardero’gen-I11 (partial) Tenrymic (4) H CO,H Porphyrin Porph yrin (52a) (52b) (53) ‘Porphyrin S-411’ Scheme 18 Incubation studies with derivatives of copro’gen-111 (4) hydroxylated in the side-chain of either ring A or ring B were inconclu~ive.~ However indirect evi- dence for the involvement of the ring-A-hydroxylated porphyrinogen [as (52)J has been presented by both and Cle~y.~~ They independently deduced the structure of ‘porphyrin S-411’ from meconium to be (53) and Jackson suggested that this probably arises by the more favoured chemical elimination of water [from (52b)l rather than carbon dioxide and water [from (52a)l.A consideration of the Newman projections (52a) and (52b) indicates strong steric interaction between the porphyrin ring and the carboxy-group in (52a) which is absent in the conformation (52b) required for anti-elimination of water to give the trans-acrylic acid side-chain of (53).It thus seems that the normal metabolic process requires enzymic assis- tance to hold the side-chain in the correct conformation for fragmentation to the ’* P. W.Couch D. E. Games and A. H. Jackson J.C.S. Perkin I 1976,2492. s9 I. A. Chaudhry P. S. Clezy and V. Diakiw Austral. J. Chem. 1977,30,879. 416 D. G.Buckley vinyl group whereas the formation of porphyrin S-411in meconium may be due to non-enzymic loss of water. Alternatively porphyrin S-411may be an artefact produced on work-up from the hydroxypropionate porphyrin (52). In either case the involvement of the hydroxylated derivative (52) is likely.58 Arornatization of Proto'gen-IX (5). Although the autoxidation of porphyrinogens occurs readily in daylight to give the corresponding porphyrins clear evidence has been obtained for enzymic aromatization of proto'gen-IX (5) in oiz~o.~~ The prod- uct protoporphyrin-IX (6),is the last intermediate common to the biosyntheses of both the haems and the chlorophylls.As a result of the work described above its biosynthesis has now been defined and it is of considerable interest that the same pathway is followed in organisms as diverse as bacteria higher plants birds and mammals. 5 The Iron and Magnesium Branches The Haemoproteins.-An enzyme ferrochelatase is known to catalyse the chela- tion of Fe2+ by protoporphyrin-IX (6) to give protohaem (54). Protohaem (54) itself is the prosthetic group of haemoglobin myoglobin and the cytochromes-b.Protohaem is also the precursor of the cytochromes-c in which the protein is covalently bound to the haem by the addition of an -SH residue to each vinyl side-chain. The cytochromes-a have as their prosthetic group a considerably modified protohaem carrying a long hydrocarbon side-chain and they have been shown to be derived from protohaem (54)in Staphylococ~us.~ The structure of porphyrin C derived from the cytochrome-c of yeast and of horse heart has been shown to be 3,8-di(ar-S-cysteinylethyl)deuteroporphyrin-IX (55) by a series of degradative synthetic and n.m.r. studies.60 A recent synthesis of haemin C (56)has been achieved61 from haemin (20) as shown in Scheme 19. The CO,H I CO,H H-C-NHZ I I H-C-NH, I (20) Haemin (55) + X t= H H Porphyrin C (56) X =Fe"'-Cl Haemin C + Reagents i Cysteine n-C16H33NMe3 Br- room temp.02,NaBH4 pH 8.1 Scheme 19 J. T. Slama H. W. Smith G. C. Willson and H. Rapoport J. Amer. Chem. Soc. 1975,97 6556. 61 S. Kojo and S. Sano J.C.S. Chem. Comm. 1977 249. Biological Chemistry-Part (iii) Tetrapyrroles and their Biosynthesis iron atom of haemin was found to play an important role in the synthesis but other observations suggested that the reactive intermediate was not a simple Fe"-pro- toporphyrin [as (54)j. Cytochrome-c oxidase is a key respiratory enzyme and its prosthetic group(s) have attracted many studies since Warburg's isolation62 of the first crude metallo- porphyrin preparation called haem A. A block to further progress in this area had been the difficulty of demonstrating homogeneity for haem A preparations or more importantly of isolating the corresponding metal-free porphyrin(s) A.New work by groups in Australia and Cambridge has that the 'porphyrin A' dimethyl ester from beef heart (whether under acidic or basic condi- tions) was a mixture of two similar compounds in roughly equal amounts. Investi- gation of the more polar compound has not been reported but the crystalline less polar ester (designated porphyrin A dimethyl ester) has been shown to have the structure (57).63 This confirms that the structure of haem A is (58) as was suggested earlier.64 The of porphyrin A dimethyl ester (57)was part of Clezy's investigation of biologically significant porphyrins much of which has been reported re~ently.~~ HO <CO,H <CO,H <CO,R CO R< (54) Protohaern (57) R=Me +X+ =H,H (58) R = H,X = Fe" Haem A The ChlorophyUs.-Research on the late stages of the biosynthesis of chlorophylls has been hindered by the insolubility of intermediates and relevant enzymes.The pathway first outlined by Granick66 in 1951 has been generally supported by the numerous investigations since then,5 and recent work by Jones67 has confirmed that the steps between protoporphyrin-IX (6) and chlorophyllide-a (63)in higher plants are as given in Scheme 20. The enzyme which catalyses the conversion of chlorophyllide-a (63) into chlorophyll-a (64) has not been characterized. However chlorophyllase the 62 0. Warburg and H.-S.Gewitz Z. physiof. Chern. 1951,288 1. M. Thompson,J. Barrett E. McDonald A. R. Battersby C. J. R. Fookes I. A. Chaudhry P. S. Clezy 63 and H. R. Morris J.CS. Chem. Comm. 1977 278. 64 See ref. 63 and ref% 2-4 cited therein. 65 See P. S. Clezy and V. Diakiw Austral. J. Chem. 1975,28 2703; P. S. Clezy and C. J. R. Fookes ibid. 1977.30 1799; and refs. therein. 66 S. Granick Ann. Rev. Pfunr Physiol 1951,2 115. 67 0.T. G. Jones,in ref. 4 pp. 207-225. 418 D. G.Buckley Protoporphyrin-IX + Mg Protoporphyrin-IX + (6) (59) CO,H CO,Me (60) 1 (61) R=C=CHZ 1 (63) R = H Chlorophyllide-a (62)R = CH2CH3 Protochloroph yllide- a (64) R.= phytyl Chlorophyll-a (65) Chlorophyll4 (66) Bacteriochlorophyll-a Scheme 20 Biologica1 Chemistry-Pa rt (iii ) Te trap yrroles and their Biosy nthesis 419 enzyme which catalyses the reverse (hydrolysis) reaction of chlorophyll-u (64) to give (63) and phytol (67) may also catalyse the biosynthetic esterification which generates chlorophyll-a (64); this would probably occur in a lipid environment within the cell.’ An alternative e~planation~~ may be that phytol (67) is not the natural substrate for the synthetic reaction catalysed by chlorophyllase and that late stages in phytol (67) biosynthesis occur after some other C, alcohol [e.g.all- trans-geranylgeraniol (69)68] has been esterified. Although the biosynthesis of the chlorophyll macrocycle has been studied exten- si~ely,~,~~,~~ much less was known about the pathway to phytol(67) until recently.” Reasons for this lack of information included difficulties in both obtaining pure phytol (67) and devising specific degradations of the labelled alcohol (67) obtained from feeding experiment~.~~ Battersby et aL70 have now defined conditions for the isolation of pure phytol (67) from cells of Euglena grucilis.The method involves separation of the phytol carbamate (68) from the corresponding derivative of geranylgeraniol which gives both C, alcohols as their correspondiog crystalline biphenyl-4-ylcarbamates (68) and (70). This allowed them to study the incorporation of [l-I3C]acetate into (67) R=H Phytol (68) R=CONH (69) R = H Geranylgeraniol (70) R=CONH phytol (67) and their results are summarized in Scheme 21.The labelling pattern found supports the operation of the normal terpenoid pathway to geranylgeranyl pyrophosp hate7* with subsequent reduction. These results do not distinguish between initial formation of the geranylgeranyl ester of protochlorophyllide-a (62) or chlorophyllide-a (63) followed by reduction or prior reduction of (69) to phytol (67) before esterification.68 However the methods developed should facilitate further studies of this pr~blem.~’ The Bacteriochlorophylls.-Unlike the higher plants in which only chlorophylls-a (64) and -b (65) are normally found the green photosynthetic bacteria use a wide ‘* J. J. Katz H. H. Strain A. L. Harkness M.H. Studier W. A. Svec T. R. Janson and B. T. Cope J. Amer. Chem.SOC.,1972 94 7938; C. Liljenberg Physiol Plant. 1974 32. 208 and refs. cited there. ‘’ G. W. Kenner J. Rimmer K. M. Smith and J. F. Unsworth in ref. 4 pp. 255-276. ’O E. H. Ahrens jun. D. C. Williams and A. R. Battersby J.C.S. Perkin I 1977 2540. 71 See ref. 70 and refs. 7-9 cited therein. 72 G. Popjak and J. W. Cornforth Biochem. J. 1966 101 553 and refs.cited therein; G.P. Moss in ‘Terpenoids and Steroids’ ed. K. H. Overton (Specialist Periodical Reports) The Chemical Society London 1971 Vol. 1 p. 221 and refs. cited therein. 420 D. G. Buckley :? / * RO (64) R =chlorophyllide-a residue .=13c Scheme 21 range of different chlorophylls for their energy-gathering activities. These bacteriochlorophylls differ in the substituents around the periphery of the macro- cycle and although it appears that bacteriochlorophyll-a (65) arises by modification of chlorophyllide-a (63) in Rhodopseudomonas spheroides,’ recent work has shown that (63) is not an intermediate in the biosynthesis of other bacteriochloroph y Kenner ef al.69have studied the structures of the various Chlorobium chloro- phylls (650) and (660) from certain Chloropseudomonas species; the designations (650)and (660) refer to the wavelength (in nm) of the red electronic absorption band.Distribution between hydrochloric acid and ether on Celite columns separates both the (650)and (660) chlorophylls into six components (or ‘bands’) in each case. The structures of the six components of the (650) chlorophylls had been determined beyond (see Table 2) and although the structures of the (660) series were less certain,69 bands 5 and 6 of the chlorophylls (660) had been shown to be as in Table 2.Many of the differences between the Chlorobium chlorophylls and chlorophylls-a and -b involve simple modifications around the periphery of the macrocycle which are common to all of the former group. There has been considerable confusion as to the substituent on the S-meso- carbon (C-20) in bands 1-4 of the (660) series. It is now clear that all of the chlorophylls (660) carry a methyl substituent at C-20,69but the structures given in Table 2 for bands 1-4 of the (660) series are less certain. Kenner ef al. have suggested these structures on the basis of a series of synthetic degradative and biosynthetic studies.Their for the biosynthesis of the (660) chloro-phylls are outlined in Scheme 22. They suggest that the pentacarboxylic acid porphyrinogen (33) is the branch from normal chlorophyll metabolism (Scheme 20) for bands 2 4 and 5 of chlorophylls (660); this suggestion was made before it was known4* that (33) is a normal intermediate in porphyrin biosynthesis (see above). 73 J. L. Archibald D. M. Walker K. B. Shaw A. Markovac and S. F. MacDonald Canad.J. Chem. 1966,44 345; R. A. Chapman M. W. Roomi T. C. Morton D. T. Krajcarski and S. F. MacDonald ibid. 1971 49 3544. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis 421 Table 2 Structures of the Chlorobium chlorophylls R2 R3 (650) series R'= H (660) series R'= Me P Bu' Prn R2.R3 Et Et band 1 2 R2 Pr"(Bu')? Bu' R3 Me Et Bu' Et Me Et 3 4 Bu'(Pr")? Pr" Me Et Pr" Me 5 Et Et Et Me 6 Et Me The carboxylic acid groups on the substituents at C-8 and C-12 would activate the side-chain for methylation by S-adenosylmethionine. This could occur first on the substituent at C-12 to give (71) after decarboxylation of the C-12 residue which would lead to (660) band 5. Alternatively methylation of (72) on the propionic acid side-chain at C-8 either once or twice would lead to (660) band 4 or (660) band 2 respectively. Scheme 22 also shows the favoured for the formation of chlorophylls (660) bands 1 3 and 6; it was this proposal that led to the structures given for bands 1 and 3 in Table 2.Methylation of (73) once or twice would give after decarboxylation either n-propyl or isobutyl substituents at C-8 whereas simple decarboxylation of (73) would lead to the 8-ethyl substituent known73 for chloro- phyll (660) band 6. Kenner et al. have also pointed out that the (650) chlorophylls which differ from the (660) series in that meso-methylation does not occur contain fractions which have 8-n-propyl- 12-methyl (band 5) and 8-isobutyl- 12-methyl (band 3) substituent arrays; rneso-methylation of these would lead to the structures proposed for chlorophylls (660) bands 1 and 3. Another bacteriochlorophyll has been isolated from both Chlorobium phaeobac- teroides and C. phaeouibrioides and has been designated bacteriochlorophyll-e.74 This was shown to be a mixture of the three structures shown in (74).These are clearly related to the Chlorobium chlorophylls (660),69 the only significant difference being the formyl group at C-7 in (74) and now the Chlorobium chloro- phylls (650) and (660) are termed bacteriochlorophylls-d and -c respe~tively.~~'~~ It H.Brockmann jun. in ref. 4 pp. 277-285. 422 D. G.Buckley P Me -Uro'gen-111 (3) PP (33) Copro'gen-111 (4) I P Me V Me Me Me Me PP (71) (73) V Me band 6 band I?A.band 3? P P (72) V =CH=CH2 Scheme 22 H HO..kMe (74) R =Et Pr" or Bu' Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis is of interest that the various bacteriochlorophylls-c -d and -e all have the (R)-configuration at the hydroxyethyl side-chain at C-3.74,75 6 The Biosynthesisof Vitamin B, Coenzyme B, (75) is the biologically active form of vitamin B, (76) and this complex molecule presents formidable challenges to chemists.Determination of the and the total of the vitamin are among the great achievements in organic chemistry. The elucidation of the biosynthesis of vitamin BIZand the determination of its role and mechanism of action in biology78 are two further challenges which are under active inve~tigation.~’~~ Me HO OH (75)Coenzyme B12 (76)Vitamin BIZ X=CN 2 2 HO The origin of the complex side-chain attached at C-17 has been but this Report will be concerned only with the biosynthesis of cobyrinic acid (Z) the natural corrin nucleus. In common with the porphyrin nucleus this consists of four pyrrole rings built into a macrocycle with the important difference of a direct C-1 to C-19 link in the corrins.Furthermore the acetate and propionate side-chains of ’’ H. Brockmann jun. and N. Risch Angew. Chem. In&rnaf.Edn. 1974,13,664. ’‘ D. C. Hodgkin J. Pickworth J. H. Robertson K. N. Trueblood R. J. Prosen and J. G. White Nature 1955,176,325; R. Bonnett J. R. Cannon A. W. Johnson I. Sutherland A. R. Todd and E. L. Smith ibid. 1955 176 328. 77 (a) R. B. Woodward Pure Appf. Chem. 1973 33 145; (b) A. Eschenmoser XXIIIrd IUPAC Congress Boston Pure Appl. Chem. Suppl. 1971,2,69; (c)A. Eschenmoser Naturwiss. 1974,61,513; (d)A. Pfaltz B. Hardegger P. M. Muller. S. Farook B. Krautler. and A. Eschenmoser Helu.Chim. Acta 1975 58. 1444. T. C. Stadtmin Science 1971,171,859. 79 P. Renz and R. Weyhenmeyer F.E.B.S. Letters 1972 22 124; S. H. Lu and W. L. Alworth Biochemistry 1972 11,608 and refs. cited therein. 424 D. G. Buckley cobyrinic acid (2) are clearly arranged as in the natural type-I11 porphyrins with the characteristic ‘inversion’ in ring D. By late 1974 the existence of a shared biosynthetic pathway to porphyrins and corrins had been demonstrated by the incorporation of 13C-and l4C-labe1led ALA and PBG into cobyrinic acid.5 It was also known that methionine acted as the source of seven of the eight methyl groups of cobyrinic acid (2) including that at C-1; the eighth the pro-12s-methyl group was known to arise by decarboxylation of the C-12 acetic acid ~ide-chain.~.~~”~’ However even after several sets of experiments in various laboratories it was not certain that uro’gen-I11 (3) was an intermediate in the biosynthesis of cobyrinic acid (2).5,48 Proof that vitamin BI2(76) is derived from uro’gen-III(3) has now been obtained from various experiments with bacterial whole-cell and cell-free preparations81 carried out in several laboratories (see Scheme 23).48,82-84 Scott showed that CO,H enzymes ___* < < CO,H CO,H CO,R C02R (2) R =H unlabelled (3) Unlabelled (77) R =Me unlabelled = l3C (77g) m = l3C (3g) = I4C (3h) 0 = 14C (77h) R=Me Scheme 23 incubation82 of the doubly labelled uro’gen-I11 (3f) (3H :14C=4.10 1) in the cell-free system from Propionibacterium shermanii gave after dilution with carrier and purification of the recovered heptamethyl cobyrinate (77) (‘cobester’) a radio- chemically pure sample of cobester (77f) (3H 14C=4.05 1).Any randomization via fragmentation and recombination would have led to a profound change in the 3H 14C specific activity ratio of this unsymmetrically substituted substrate. Further evidence was obtained on administrations2 of [5 15-’3C2]uro’gen-III (3g) to resting A. I. Scott in ref. 4 pp. 303-318. ” A. I. Scott B. Yagen and E. Lee J. Amer. Chem. SOC.,1973 95 5761; A. R. Battersby European Symposium in Bio-organic Chemistry Gregynog Wales May 1973; see also ref. 836. ’* A. I. Scott B. Yagen N. Georgopapadakou K. S. Ho S. Klioze E. Lee S. L. Lee G. H. Temme 111 C.A. Townsend and I. M. Armitage J. Amer. Chem. SOC.,1975,97,2548; 1976,98,2371. (r3 (a)A. R. Battersby M. Ihara E. McDonald F. Satoh and D. C. Williams J.C.S. Chem. Cornm. 1975 436; (b)A. R. Battersby E. McDonald R. Hollenstein M. Ihara F. Satoh and D. C. Williams J.C.S. Perkin I 1977 166. 84 H.-0. Dauner and G. Miiller 2.physiol. Chem. 1975 356 1353. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis 425 whole cells of P. shermanii. In this case pure vitamin B12(76) was recovered; this was shown to be labelled in the corrin nucleus as illustrated in (77g) by a 13C n.m.r. analysis (4.5 YOspecific incorporation). The Cambridge group used the specifically labelled uro’gen-I11 (3h)83 in their studies which demonstrated S-8% specific incorporations of (3h) into cobester (77h) using a similar cell-free preparation.” Dauner and Muller have reporteds4 high incorporations of a mixture of uro’gens-I (8) -11 -111 (3) and -1V (labelled equivalently in each pyrrole ring) using a cell-free system from Clostridium tetunomorphum.They have also achieved similar incorporations with Battersby’s [12-methylene-14C]uro’gen-III (3h),83b using the same cell-free preparation from C. tetanomorphum. These results firmly establish that uro’gen-I11 (3) is a specific biosynthetic precursor of cobyrinic acid (2) and thus of vitamin BI2(76) itself. By inspection the biosynthetic conversion of uro’gen-I11 (3) into cobyrinic acid (2) requires the following structural changes (a) introduction of seven methyl groups from methionine at C-1 C-2 C-5 C-7 C-12 (pro-R),C-15 and C-17; (b) decarboxylation of the acetic acid side-chain at C-12; (c) extrusion of C-20; (d) possible adjustments of the oxidation level; and (e) insertion of cobalt.Scott suggested8’ that the first step could be decarboxylation to give the ring-c- methyl heptacarboxylic acid porphyrinogen (78) and subsequently reporteds2 a 0.1% incorporation of the 14C-labelled form (78a) into cobester (77) of unknown labelling pattern using the cell-free system from P. shermanig1 This result was interpreteds2 as confirmation that uro’gen-I11 (3) suffers decarboxylation (at C-12) prior to the necessary reductive methylation steps. Independent work by the Cambridge groups3 also demonstrated that incubation of the 14C-labelled porphyr- inogen (78b) leads to labelled cobester (77) after the usual work-up.However they concluded that the ring-c-methyl porphyrinogen (78) was probably not a normal intermediate in corrin biosynthesis. They reached this conclusion mainly because the porphyrinogen (78) was 30-50 times less effective as a precursor of cobyrinic acid (2)than was uro’gen-I11 (3) while the ring-c-methyl porphyrinogen (78) and uro’gen-I11 (3) were almost equally well incorporated into copro’gen-I11 (4) in the same cell-free The incorporations of l4C-labe1led (78) were so low in both that the recovered cobester (77) could not be degraded to locate the site(s) of labelling. Groups in Zurichs6 and Cambridge83b*87 have shown that each of the seven methyl groups in cobyrinic acid (2) derived from methionine is incorporated intact into the corrin nucleus.The Yale groups8 independently determined that the methyl groups at C-12 (pro-R)and C-1 of vitamin B12 (76) are incorporated intact. Intact incorporation had been demonstrated previously by the Cambridge group for the methyl groups at C-12 (pro-R)and C-7,89 but the most important result concerned the methyl group at C- 1. 85 A. I. Scott E. Lee and C. A. Townsend Bioorg. Chem. 1974 3 229; A. I. Scott Tetrahedron 1975 31 2639. a6 M. Imfeld C. A. Townsend and D. Arigoni J.C.S. Chem. Comm. 1976,541. 87 A. R. Battersby R. Hollenstein E. McDonald and D. C. Williams J.C.S. Chern. Comm. 1976 543; see also ref. 836. 88 A. I. Scott M.Kajiwara T. Takahashi I. M. Armitage P. Demou and D. Petrocire J.C.S. Chem. Comm. 1976,544. 89 A. R. Battersby M. Ihara E. McDonald J. R. Stephenson and B. T. Golding J.C.S. Chem. Comm. 1973,404; 1974 458; J.C.S. Perkin I 1977 158. 426 D. G.Buckley P Me Me P A-@+M;;R Me p (78) Unlabelled (79) (78a) rn = 14C (79a) R=H =14C (79b) R =CH20H (78b) The question of loss or retention of the methyl hydrogen atoms is of particular importance in the case of this 1-methyl group of vitamin B12 (76) as it bears critically on current theories concerning the mechanism of formation of the cor- rinoid C-1-C-19 bond.” Each of the groups used essentially the same methods in their recent studies they incubated [Me-13CD3]methionine (90 atom ‘/o 13C 98 atom YO D) with intact cells of P.shermanii and examined the recovered vitamin B, (76) by 13C n.m.r. spectroscopy. The Cambridge and Zurich groups used deuterium-noise-decoupled 13C Fourier-transform n.m.r. spectroscopy which showed that each of the seven methionine-derived methyl groups including that at C-1 had been incorporated without exchange of D with H of the medium; none of the signals of the 13C-enriched carbon atoms displayed the large splitting (Jca. 125 Hz) expected for 13C-’H spin-spin coupling. The Yale group reached the same conclusion about the incorporation of the 1-methyl group using a different n.m.r. method. These results rule out the intermediacy of any system in which the potential 1-methyl group is temporarily a methylidene residue,” e.g.(80) although the existence of an intermediate seco-corrin such as (79) would be compatible with the above results. In principle C-20 could be lost’ from the system at any oxidation level between methanol and carbon dioxide and this could occur at various stages in the forma- tion of the corrin ring e.g. before bond formation between C-l and C-19 OCCU~S,~’ or as a consequence of it. Recent work by indicates that C-20 is lost as formaldehyde possibly from a species such as (79b); electrocyclic ring closure of the product (79a)90 would lead to a dehydrocorrin which may then be reduced by insertion of Co’+ rather than CO~+.~” However there are many other possibilities and further experimental work is required. Sirohydrochlorin.-The prosthetic group of the widespread class of enzymes which catalyse the six-electron reduction of sulphite to sulphide was first characterized in 1973 by Kamen and Siegel’l and was named sirohaem.Removal of the iron afforded an orange fluorescent compound sirohydrochlorin (8l) which was shown to be an isobacteriochlorin. It was suggested9’ that sirohydrochlorin might arise by two C-methylations at C-12 and C-18 of uro’gen-III(3) and it was also noted that 90 A. Eschenmoser Chem. SOC.Rev. 1976,5 377; see also refs. 26a 486 77d,and 85. 9’ L. M. Siegel M. J. Murphy and H. Kamen J. Biol. Chem. 1973,248 251. 92 M. J. Murphy L. M. Siegel H. Kamen and D. Rosenthal J. Biol. Chem. 1973,248 2801. Biological Chemistry -Part (iii) Tetrapyrroles and their Bios yn thesis 427 sirohydrochlorin could represent an early intermediate in the biosynthesis of vitamin B12(76).92 In 1976 Bykhovsky’s group in Moscow the isolation of a methylated tetrapyrrole which contained cobalt from P.shermanii. The metal-free form of this compound appeared to be similar to sirohydrochlorin incorporated radioactive methionine into its methyl groups and furthermore increased the rate of vitamin B12 production in the organism. Muller’s group in Stuttgart then reportedg4 the isolation of ‘Factor 11’ from P.shermanii and showed that labelled forms of ‘Factor 11’ were incorporated into cobyrinic acid (2) by an enzyme preparation of C. tetanomorphum. The Cambridge and Moscow groups working in collaboration sho~ed~’,~~ that the isobacteriochlorin from P.shermanii was the same as the sirohydrochlorin obtained from the sulphite-reducing organism Desulphovibrio gigus. Finally ‘Factor 11’ was shown26u to be identical with sirohydrochlorin. Chemical and spectroscopic work by’the Cambridge and Moscow groups led to the postulation of two alternative structures for sirohydrochlorin; the A-B structure (81) and an isomeric B-c structure although the former was Recent work by the Cambridge group has rigorously established the structure (81) for sirohydrochlorin and its specific incorporation into cobyrinic acid (2) has also been dem~nstrated.~~ The origin of the two C-methyl groups of sirohydrochlorin was confirmedY7 by incorporation of (2S)-[Me-’3C]methionine,using D. gigas and examination of the product @la) by 13Cn.m.r.spectroscopy. Incubation of (2S)-[Me-’4C]methionine with P. shermanii under suitable conditions gave labelled sirohydrochlorin (81b). The purified ester (82) was proved to be radiochemically pure and the derived labelled acid (81b) was then incubated in the broken cell from P. sher-manii; the incorporation into cobyrinic acid (2b) isolated as cobester (77b) was in the range 3-7’/0. Ozonolysis of the purified cobester (77b) gave the fragments (83) (84) and (85),98the radioactivities of which confirmed that the labelling pattern was shown in Scheme 24 (* = 14C). These results when combined with those noted above e~tablish~~-~~ that siro- hydrochlorin has the structure and absolute configuration (81) and that sirohaem is the corresponding iron complex.The proven97 efficient enzymic conversion of sirohydrochlorinI-(81) into cobyrinic acid (2) confirms the Cambridge group’s earlier findings36 that the heptacarboxylic acid (78) is not the next intermediate beyond uro’gen-I11 (3) en route to vitamin B, (76) contrary to what had been thought.” The middle section of the biosynthetic pathway to vitamin B, can now be defined as uro’gen-I11 (3)-+ (81)-+ (2) -+ vitamin BI2(76). ?They envisage that the biosynthetic intermediate may be a dihydro-derivative of (81) which would be the product expected from simple methylation in rings A and B of uro’gen-I11 (3) (see refs. 26a and 486). 931 V. Ya. Bykhovsky N. I. Zaitseva A. V. Umrikhina and A. N. Yavorskaya Priklad. Biokhim. Mikrobiol.1976,12 825 and refs cited therein. 94 R. Deeg H.-P. Kriemler K.-H. Bergmann and G. Miiller 2.physiol. Chem. 1977,358 339. 95 A. R. Battersby E. McDonald H. R. Morris M. Thompson D. C. Williams V. Ya. Bykhovsky N. I. Zaitseva and V. N. Bukin Tetrahedron Letters 1977 2217. 96 A. R. Battersby K. Jones E. McDonald J. A. Robinson and H. R. Morris Tetrahedron Letters 1977 2213. 97 A. R. Battersby. E. McDonald M. Thompson and V. Ya. Bykhovsky J.C.S. Chem. Comm. 1978 150. 98 D. Arigoni E.T.H. Zurich; personal communication to A. R. Battersby quoted in ref. 97. 428 D. G. Buckley Uro’gen-111 enzymic he’ \ CO R CO R (82) R =Me * = 14C (8 1) R =H Unlabelled @la) R =H * = 13C (81b) R =H * = 14C lenrymic C0,Me I 2 Me 0 < C0,Me CO R t.0:R (84) (2b12 R =H * = l4G (77b) R =Me * = 14C Scheme 24 In his recent review,48b Scott describes the work of the Yale group with siro- hydrochlorin which confirms both the structure as (81) and the intermediacy of sirohydrochlorin* in the biosynthesis of vitamin BIZ.Interestingly they found that (81) and its dihydro-derivative were both efficiently incorporated into cobyrinic acid (2).7 Miscellaneous New Techniques.-The extensive survey of laboratory methods made available early in 1975 by Fuhrhop and Smith in the book edited by K. M. Smith2 has now Of been published ~eparately.~~ the three most important new techniques * He also envisages that the intermediate may be a dihydro-derivative of (81); see footnote on p.427. as J.-H. Fuhrhop and K. M. Smith ‘Laboratory Methods in Porphyrin and Metalloporphyrin Research’ Elsevier Amsterdam 1975. Biological Chemistry -Part (iii) Tetrapyrroles and their Biosynthesis developed since the early 1970s only the use of 13C-labelling was sufficiently well advanced to be included in detail in ref. 2.5The crucial role played by '3C-labelling and the associated I3C n.m.r. techniques in the elucidation of various parts of the biosynthesis of porphyrins chlorins and corrins has been published subsequently and is described abo~e.~~,~~~*~~ The Cambridge' 1*12*23*41 and Cardiff42745 groups have used 'H and 13C n.m.r. spectroscopy of porphyrin esters in the presence of various amounts of shift reagents"' to show the distribution of the side-chains which flank the four meso positions.This technique has both facilitated the identification of various natural porphyrin~~~,~~ and the confirmation of their structure by comparisons of the esters of the natural and synthetic compounds. The use of shift reagents and of other new techniques which change n.m.r. spectra as probes for the study of chlorophyll chemistry has been reviewed in a recent Meldola Medal Lecture."' Other important new work in the area of chlorophyll chemistry using the n.m.r. method has been reported by Katz et al."' High-performance liquid chromatography (h.p.1.c.)"' has been used very suc-cessfully by various groups in the past four years for the efficient separation of different porphyrin type^''^ and more importantly for the otherwise very difficult separation of isomeric p~rphyrins.~ 1*32,5'v5' The efficient separation of all four coproporphyrin isomer^,^^.^' which was not possible until re~enfly,''~ was one of the crucial developments by the Cambridge group in their successful work on the 'type-I11 pr~blem'.'~ Much of the recent work by the Cambridge and Cardiff groups has relied heavily on h.p.1.c.for both analytical and preparative separations of porphyrins (see above). Bile Pigments.-O'Carra has reviewed the biochemistry of haem cleavage and associated chemical analogues. lo5 Oxidative cleavage of protohaem (54) at the a-rneso-carbon (C-5) and loss of the iron atom normally gives biliverdin-IXa (86) although in isolated cases oxidative attack does occur at one of the other meso positions.In mammals the biliverdin-IXa (86) is immediately converted into bilirubin (87) (sometimes named bilirubin-IXa) by enzymic reduction (Scheme 25). The structures shown for biliverdin-IXa (86) and bilirubin (87) with the (2) configuration about each of the three or two exocyclic double bonds are what might reasonably be expected from their mode of formation; the particular tautomers shown have been accepted on the basis of various spectroscopic studies. O6 The first X-ray analysis of any naturally occurring linear tetrapyrrole has been reported recent1y.lo6 This study of bilirubin confirms that the structure is as shown in (87) thus confirming the gross chemical structure assigned to bilirubin by Fisher loo J.K. M. Sanders Chem. SOC.Rev. 1977,6 467. lo' H. Scheer and J. J. Katz in ref. 2 pp. 399-524; J. J. Katz W. Oettmeier and J. R. Norris in ref. 4 pp. 227-254; M. J. Wasielewski U. H. Smith B. T. Cope and J. J. Katz J. Arner. Chem. SOC.,1977,99 41 72. lo2 L. R. Snyder and J. J. Kirkland 'Introduction to Modern Liquid Chromatography' Wiley New York 1974. '03 Ref. 32 and accompanying papers. Ref. 2 pp. 859-861. lo' P. O'Carra in ref. 2 pp. 123-153. lo6 R. Bonnett J. E. Davies and M. B. Hursthouse Nature 1976 262 326. 430 D. G. Buckley Protohaem (54) O2+reductant knzymic 1 co+iron 2H CO H CO,H C0,H C0,H (86) Biliverdin-IXa (87) Bilirubin Scheme 25 et al.lo’ There is considerable intramolecular hydrogen bonding in the crystalline bilirubin (Figure 1); the low solubility of bilirubin in water which has important biological consequences can be understood readily in terms of this pattern of hydrogen bonding.lo6 Since bilirubin (87) is derived directly from biliverdin-IXa (86) and the former has the same configuration about the two exocyclic double bonds as is found in protohaem (54) it is clear that biliverdin-IXa (86) has the illustrated (2)configuration at the C-5 and C-15 double bonds. The stereo-chemistry of the C-10double bond of biliverdin-IXa has not been determined but presumably it also has the (2)configuration shown in (86). c-0 0 .. .. H ‘0-c Figure 1 Structure of bilirubin in simplified form. (Reproduced by permission from Nature 1976,262,326) lo’ H.Fischer H. Plieninger and 0.Weissbarth 2.physiol. Chem. 1941 268 231. Biological Chemistry-Part (iii) Tetrapyrroles and their Biosynthesis 43 1 Haemoglobin/Myoglobin Model Studies.-The respiratory haemoproteins haemoglobin and myoglobin transport and store molecular oxygen and thus are essential to the life of all vertebrates.”’ The most significant property of haemo-g10bin’~’ is co-operative oxygen binding the oxygen affinity of the biologically active tetramer increases with increasing saturation.lOg.llo Co-operativity is required for the transfer of oxygen from the carrier haemoglobin to the receptor myoglobin as well as for responses to other physiological requirements. The way that the protein in haemoglobin and myoglobin regulates oxygen affinity controls axial ligation of oxygen and provides kinetic stability to the iron-dioxygen group is under active investigation by the preparation and full characterization of synthetic model compounds.Since Collman’s synthesis of his ‘picket-fence’ porphyrin,’ which led to the isolation and characterization of a series of crystalline dioxygen-iron porphyrin complexes l1 many different groups have synthesized a wide variety of different iron porphyrins.’ l2 The different approaches used in these various studies have already helped to define the degree of hindrance required around the iron-dioxygen grouping to allow reversible oxygen binding,’ ” but further work is necessary before these model studies can make a significant contribution to our understanding of the chemistry of the binding of oxygen to haemoglobin and myoglobin in uivo.For recent reviews see N. W. Makinen in ‘Techniques and Topics in Bioinorganic Chemistry’ ed. C. A. McAuliffe Wiley New York,1975 Part 1 Chapter 2; J. H. Prat ibid. Part 2 Chapter 7. M. F. Perutz Brit. Med. Bull. 1976,32 193. ‘lo J. M. Baldwin Brit. Med. Bull. 1976 32 213. J. P. Collman R. R. Gagne T. R. Halbert T.-C. Marchon and C. A. Reed J. Arner. Chem. SOC.,1973 95,7868; J. P. Collman R. R. Gagne C. A. Reed T. R. Halbert G. Lang and W. T. Robinson ibid. 1975,97 1427. J. P. Collman Accounts Chem. Res. 1977 10 265.