Phenazines—

Phenazines.—Results on the biosynthesis of microbial phenazines from shikimic acid (previously published in preliminary form cf. Vol. 5, p. 44 and Vol. 7, p. 27) are now available in full papers.53 Additional results are that 2,3-dihydro-3-hydroxyanthranilic acid (140) was not a precursor for iodinin (141), nor was 

Phenazines.—Evidence had been obtained with some micro-organisms, but not with others (Pseudomonas and related organisms), that phenazine-l,6-dicarboxylic acid (121) is the first phenazine to be formed, i.e. the one from which all others derive (cf. Vol. 10, p. 28). The negative results that were obtained with (121) were 

Ferezou, A. Quesneau-Thierry, C. Servy, E. Zissmann, and M. Barbier, J. Chem. Soc., Perkin Trans. L 1980, 1739. 

Using ether-treated cells of P. aureofaciens, dicarbonyl-14C2 phenazme-l,6-dicarboxylic acid las (121)1 was found to be an efficient and specific precursor for phenazine-1-carboxylic acid (123), and also for 2-hydroxyphenazine-l-carboxylic acid (124). The rate of growth of the organism appeared to be important, because an incorporation was also recorded of the labelled (121) into (123), albeit at a lower level, with cultures that had been grown rapidly. The position of phenazine-1,6-dicarboxylic acid (121) as a universal intermediate in the biosynthesis of phenazines now seems secure. The previously reported failure of dimethyl phenazine-1,6-dicarboxylate (122) to act as a precursor of phenazines cf. Vol. 10, p. 28 Vol. 9, p. 29) has been confirmed with ether-treated cells of P. aureofaciens. Efficient hydrolysis of (125) to (123) did, however, occur.101 

Cytochalasins.—Previous results have shown that the cytochalasins, e.g. cyto-chalasin B (126), are partially polyketide in origin cf. Vol. 7, p. 29 Vol. 6, p. 44). It has now been shown that [2-2H3,2-13C]acetic acid is incorporated into cyto-chalasin B (126) in Phoma exigua and into cytochalasin D in Zygosporium masonii. Labelling of the expected sites by 13C was observed for both metabolites, but, in keeping with results on other acetate metabolites, most of the deuterium was lost, being retained only in the polyketide chain at C-l 1, which is part of the starter acetate unit.102 

Cycloheximide.—It is known that the carbon skeleton of cycloheximide (127) is derived from malonate, with the two methyl groups (C-15 and C-l6) arising from 

Phenazines.—Shikimic acid (134) is clearly implicated as a precursor for microbial phenazines, e.g. iodinin (135), and it can act as the sole source of the carbon skeleton. Essential proof that two molecules of shikimic acid are involved in phenazine biosynthesis was provided when it was shown that on incorporation of DL-[1,6- C2, 2- H]shikimic acid [as (134)] into iodinin (135) in Brevibacterium iodinum, some (7.5%) of the molecules of (135) produced were dideuteriated. [The shikimic acid was incorporated with the usual high efficiency (similar values for and H) and the deuterium label was confined to the expected positions (see below).] 

It follows from this result and the C labelling studies that phenazine biosynthesis proceeds from two shikimic acid units linked as in (136) or (137). By determining that the sites of deuterium labelling in the iodinin (135) derived from 

Further support for the conclusion that two molecules of shikimic acid are involved in phenazine biosynthesis comes from the incorporation of -[1,6,7,- CaJshikimic acid [as (134)] into phenazine-1-carboxylic acid (138) in Pseudomonas aureofaciens with close to a fifth of the activity present in the carboxy-group, as required if two molecules of shikimic acid are involved [however, the same result would have been obtained if only one molecule of shikimic add was implicated provided that a symmetrical intermediate of type (139) was also involved in the elaboration of (138)]. 

This has been achieved for the production of (146) in D. sphaerica by means of histidine decarboxylase inhibitors. Both a-methylhistidine and a-hydrazinohistidine (inhibitors for mammalian specific histidine decarboxylase) inhibited formation of histamine with the result that more (146) was synthesized at the expense of (144) (based on the radioactivity of the products after feeding [ H]histidine, [ C]isovaleric acid and inactive animomethylimidazole [145]). a-Methyldopa, an inhibitor of mammalian non-specific decarboxylase, was without effect on the proportions of the products formed. 

Kahlson and E. Rosengran, Biogenesis and Physiology of Histamine , Williams and Wilkins, Baltimore, 1971, p. 31. 

Phenazines.— The phenazine ring system is common to some thirty microbial metabolites, some of which bear carbon substituents at C-1 and C-6, e.g. phenazine-1,6-dicarboxylic acid (167). This skeletal arrangement suggests a derivation from two molecules of anthranilic acid and this idea is made more 

Both l-carboxy-5-methyl[6,7,8,9- H4]phenazinium betaine (as 171) and 1-carboxy[6,7,8,9- H4]phenazine (as 170) were incorporated efficiently into pyocyanin (173) without loss of deuterium. Hydroxylation of (171) thus occurs, as a specific decarboxylative reaction. This specificity, together with the high level of incorporation of these two compounds, strongly suggested that they are normal intermediates in pyocyanin biosynthesis and since 1-hydroxy- 

Mothes and H. R. Schiitte, Biosynthese der Alkaloide , VEB Deutscher Verlag der Wissenschaften, Berlin, 1969. 

It was shown, in addition, that phenazine methosulphate could serve as a pyocyanin precursor but, in the light of the specificity of the hydroxylative decarboxylation of (171) and a lower incorporation than (170) or (171), is probably not a normal intermediate. 

Jerina, J. W. Daly, and B. Witkop, J. Amer. Chem. Sac., 1968, 90, 6523, and references cited therein D. M. Jerina, J. W. Daly, B. Witkop, P. Zaltman-Nirenberg, and S. Udenfriend, ibid., p. 6525. 

Phenazines.—A study using [6- C]shikimic acid [as (126)] has given results which confirm previous ones concerning the orientation of shikimic acid (126) units in the bacterial phenazine iodinin (127). 

It has been claimed that dimethyl phenazine-1,6-dicarboxylate (128) is a precursor for 1-carboxyphenazine in Pseudomonas aureofaciens cf. ref. 8. This claim has been disputed careful checking showed that neither (128) nor the corresponding acid (129) was incorporated into phenazines produced by this organism. This has been supported by the results of other workers, who have found that (128) is metabolically inert in P. aureofaciens. Moreover, neither (129) nor (128) was incorporated into phenazines in P. phenazinium. On the other hand, efficient incorporations have been recorded of (129), but not of (128), into iodinin (127) and related phenazines in three actinomycetes, i.e. Streptomyces thioluteus, Microbispora amethystogenes, and M parva  

This latter observation correlates with the observation that (129) is a precursor for lomofungin (130) in Streptomyces lomodensis. It seems clear from the combined evidence that phenazine-1,6-dicarboxylic acid (129) is a precursor for all microbial phenazines. Failure to observe incorporation of (129) in Pseudomonas, with (by contrast) positive results in actinomycetes, may be attributed to differences in permeability of the cell walls cf. ref. 127). 

The above negative results with (152) and its dimethyl ester were obtained in Pseudomonas and closely related species for metabolites bearing a single aryl-Ci substituent or none. It was still an attractive possibility that (152) could be a precursor for phenazines bearing two aryl-Ci substituents, e.g. lomofungin (155), and indeed it was found to be ineorporated into this metabolite in Streptomyces lomodensis with an efficiency which indicated that it is an intermediate in lomofungin biosynthesis. It is not yet clear whether this positive result was obtained because (152) is transported across the cell walls of Streptomyces but not Pseudomonas species or because (152) is only a precursor for phenazines with two aryl-Ci substituents. Experiments with another Streptomyces species (5. luteoreticuli) which produces the methyl ester of phenazine-1-carboxylic acid (153) should allow resolution of the problem. 

9-Dihydroxyphenazine-l,6-dicarboxylic acid (157) has been isolated as its dimethyl ester from Pseudomonas cepacia Although the acid (157) has been proposed as an intermediate in phenazine biosynthesis before phenazine-1,6-dicarboxylic acid (152), the necessary loss in vivo of two phenolic hydroxy-groups in the formation of (152) or, e.g., (153), makes this highly unlikely, and preliminary testing of the hypothesis supports this view. The dihydroxy-acid (157) could, however, be an intermediate, and more reasonably so, in the biosynthesis of compounds like lomofungin (155) at a stage after (152). 

The pyruvate derivative (159) is a likely intermediate in the formation of the amino-acid (160) in chloramphenicol biosynthesis. A search for an enzyme in cultures of a Streptomyces species which produces (161) has led only to the isolation of a non-specific amino-transferase.  

---

CgHeNj. Brown-yellow crystals m.p. 103-104 C, b.p. 256 258°C. Its solutions reduce Ag ions and it is used as a photographic developer. It is also used as a dye-precursor, for the synthesis of phenazine derivatives and for characterizing inter alia) a-diketones. 

ANTTBIOTTCS - BETA-LACTAMS - CARBAPENEMS AND PENEMS] (Vol 3) Thiozolyl blue-phenazine methosulfate... 

The mechanism of oxidative dyeing involves a complex system of consecutive, competing, and autocatalytic reactions in which the final color depends on the efficiency with which the various couplers compete with one another for the available diimine. In addition, hydrolysis, oxidation, or polymerization of diimine may take place. Therefore, the color of a mixture caimot readily be predicted and involves trial and error. Though oxidation dyes produce fast colors, some off-shade fading does occur, particularly the development of a red tinge by the slow transformation of the blue indamine dye to a red phenazine dye. 

Nitrogen Compound Autoxidation. CycHc processes based on the oxidation of hydrazobenzene and dihydrophenazine to give hydrogen peroxide and the corresponding azobenzene—phenazine were developed in the United States and Germany during World War II. However, these processes could not compete economically with the anthrahydroquinone autoxidation process. 

Many substituted quinolines are intermediates for antimalarials. The 2,4-di-substituted quinolines are produced from aniline and 1,3-diketones by the Combes quinoline synthesis (28). The reaction of aniline with nitrobenzene in the presence of dry sodium hydroxide at 140°C leads to formation of phenazine [92-82-0] and by-products (Wohl-Aue synthesis) (29). 

The parent lings are always included in phenazine [92-82-0] (4), phenoxazine [135-67-1] (5), oi phenothiazine [92-84-2] (6) ting systems. 

The first of these dyes to be made synthetically was Perkin s mauveine [6373-22 ] which is a substituted phenazine of stmcture (7). 

Azonia substitution at a naphthalene bridgehead position gives the quinolizinium ion (16). Oxonia substitution, elsewhere, forms the 1- and 2-benzopyrylium ions (17) and (18). The two most well-known monoaza systems with three aromatie fused rings are aeridine (19), derived structurally from anthraeene, and phenanthridine (20), an azaphenanthrene. The better-known diaza systems inelude phenazine (21) and 1,10-phenanthroline (22), while systems with three linearly fused pyridine rings are ealled anthyridines, e.g. the 1,9,10-isomer (23). 

A computer search of volumes 70-95 of Chemical Abstracts using the keyword Pyrazine resulted in more than 2600 references, and, after removal of fused pyrazine systems and cross-referencing the remaining references, this number increased to approximately 7000 in total. When the benzopyrazines quinoxaline and phenazine were added the number of references was in excess of 10 000, all of which might be considered to be relevant to a chapter devoted to pyrazines and their benzo analogues. 

In valence bond terms the pyrazine ring may be represented as a resonance hybrid of a number of canonical structures (e.g. 1-4), with charge separated structures such as (3) contributing significantly, as evidenced by the polar character of the C=N bond in a number of reactions. The fusion of one or two benzene rings in quinoxaline (5) and phenazine (6) clearly increases the number of resonance structures which are available to these systems. 

Phenazine also exhibits D h symmetry and numerous reports on the X-ray structure of a-phenazine have appeared (54AX129). The parameters determined at 80 K are shown in... 

Pugmire etal. have published calculated electron densities for pyrazine (68JA697), quinoxaline (69JA6381) and phenazine and the calculated total electron densities a + v) are shown in (10), (11) and (12). 

The electron density at nitrogen in phenazine is intermediate between those of pyrazine and quinoxaline and the highest electron density on the carbon atoms of the benzene rings is at C-1 (with positions 4, 6 and 9 being equivalent). ... 

The Chemical Abstracts numbering system of phenazine is used. 

Phenazine gives rise to an AA BB NMR spectrum with coupling constants /i,2 9,0, /i,3 1.67, /i,4 0 and Jx3 6.55 Hz (66CPB419). Similar coupling constants are also observed in a number of phenazines and phenazine iV-oxides. 

The NMR spectrum of quinoxaline has been measured in CDCI3 and the chemical shift values are as shown in (16) (69JA6381). Curiously, C NMR spectra of phenazine and its derivatives have been recorded in benzene solution and the chemical shift values quoted relative to benzene however, for consistency the values in (17) are quoted relative to TMS. 

A feature eommon to the pyrazine, quinoxaline and phenazine ring systems is their remarkable stability in the mass speetrometer and in all eases with the parent heterocyeles the moleeular ion is the base peak. In the ease of pyrazine, two major fragments are observed at mje 53 and 26, and these fragments are eonsistent with the fragmentation pattern shown in Seheme 1. 

Table 3 lists some of the basic physical properties of pyrazine, quinoxaline and phenazine (references are given in the main text). 

Table 3 Physical Properties of Pyrazine, Quinoxaline and Phenazine...

Determined crystallographically for a-phenazine. Phenazine exhibits polymorphism, with a-phenazine being the common polymorph. 

Electrophilic substitution reactions of unsubstituted quinoxaline or phenazine are unusual however, in view of the increased resonance possibilities in the transition states leading to the products one would predict that electrophilic substitution should be more facile than with pyrazine itself (c/. the relationship between pyridine and quinoline). In the case of quinoxaline, electron localization calculations (57JCS2521) indicate the highest electron density at positions 5 and 8 and substitution would be expected to occur at these positions. Nitration is only effected under forcing conditions, e.g. with concentrated nitric acid and oleum at 90 °C for 24 hours a 1.5% yield of 5-nitroquinoxaline (19) is obtained. The major product is 5,6-dinitroquinoxaline (20), formed in 24% yield. 

In the case of phenazine, substitution in the hetero ring is clearly not possible without complete disruption of the aromatic character of the molecule. Like pyrazine and quinoxa-line, phenazine is very resistant towards the usual electrophilic reagents employed in aromatic substitution reactions and substituted phenazines are generally prepared by a modification of one of the synthetic routes employed in their construction from monocyclic precursors. However, a limited range of substitution reactions has been reported. Thus, phenazine has been chlorinated in acid solution with molecular chlorine to yield the 1-chloro, 1,4-dichloro, 1,4,6-trichloro and 1,4,6,9-tetrachloro derivatives, whose gross structures have been proven by independent synthesis (53G327). 

Conflicting reports on the nitration of phenazine have appeared, but the situation was clarified by Albert and Duewell (47MI21400). The early work suggested that 1,3-dinitroph-enazine could be prepared in 66% yield under standard nitration conditions however, this proved to be a mixture of 1-nitrophenazine and 1,9-dinitrophenazine (24). As with pyrazines and quinoxalines, activating substituents in the benzenoid rings confer reactivity which is in accord with valence bond predictions thus, nitration of 2-methoxy- or 2-hydroxy-phenazine results in substitution at the 1-position. 

Phenazine reacts with benzenesulphinic acid in alcoholic hydrogen chloride to give 2-phenazinyl phenyl sulfone (26 Scheme 4), presumably by an intermediate 5,10-dihy-drophenazine this reaction is evidently a useful method of preparing 2-substituted phenazines, since the sulfone is readily displaced in substitution reactions. 

The ease of oxidation varies considerably with the nature and number of ring substituents thus, although simple alkyl derivatives of pyrazine, quinoxaline and phenazine are easily oxidized by peracetic acid generated in situ from hydrogen peroxide and acetic acid, some difficulties are encountered. With unsymmetrical substrates there is inevitably the selectivity problem. Thus, methylpyrazine on oxidation with peracetic acid yields mixtures of the 1-and 4-oxides (42) and (43) (59YZ1275). In favourable circumstances, such product mixtures may be separated by fractional crystallization. Simple alkyl derivatives of quinoxalines are... 

---