ORGANIC CHEMISTRY.1. INTRODUCTION.THE subjects selected for review in this section include the hydrogen bond,stereochemistry, carbohydrates, steroids, and a selection of heterocycliccompounds containing nitrogen.A review is given of those methods of detection of the hydrogen bondwhich have proved of greatest use in the elucidation of organic structures.Emphasis is laid on the physical conditions of the methods by which assess-ment of structure is made, and whether these conditions are likely to preservea hydrogen-bond structure. Methods based on the measurement of inter-atomic distances by means of both X-ray diffraction and electron diffractionare discussed, examples of the former being provided by phthalocyanine,melamine, and hyperol, and of the latter by hydrofluoric acid and carboxylicacids.Various indirect or comparative methods are also discussed, such asthose based on infra-red absorption, volatility, solubility and related pro-perties, and molecular weight determinations. Reference is made to thesolute-solvent interaction revealed by the work of C. S. Marvel and hiscollaborators, and to methods for the detection of molecular association andinternal hydrogen bonds (chelate rings). A review is included of the chemicaldifferences between compounds possessing a hydrogen bond structure andothers closely related to them, e.g., o-hydroxyazo-compounds, the benzilmonoximes ; of methods of alternative syntheses leading to a single individualcompound, e.g., p-diketones, methylnaphthazarin, quinhydrones, formazylcompounds; and of the effect of chelation in stabilising certain structuresand in favouring certain reactions.The chief examples (mainly organic) of compounds exhibiting hydrogenbmds of the following types are considered: F-H-F, F-H-N, F-H-0,0-H-0, N-H-0, N-H-N, N-H-S, 0-H-S.Some emphasis is laid on theimportant part played by the hydrogen bond in deciding crystal structureand other important physical properties, e.g., density of ice, layer cleavage ofanhydrous oxalic acid, configuration of proteins. The evidence for theengagement of the CH group in hydrogen-bond formation is considered insome detail fi.e., C-H-0, C-H-N), and a short account of hydrogen bondsinvolving other elements (S, C1, Br) is given. The close connection betweenhydrogen-bond association and tautomerism (" mesohydric tautomerism ")is given brief mention.An absolute asymmetric synthesis of ethyl d-tartrate from ethyl fumaratehas been reported.Mixtures of diastereoisomeric esters have been separatedby high-efficiency fractional distillation. The sulphur-oxygen bond insulphoxides, formerly supposed to be a co-ordinate bond, is shown t o be adouble bond, so that the figure corresponding with the sulphur valencies is atrigonal bipyramid and not a " tetrahedron ". Interesting stereochemicalstudies have been made of heterocyclic compounds of sulphur, selenium, anBAKER AND IIEY : INTRODUCTION. 139tellurium, and very complete investigations of the optical activity of hetero-cyclic and spirocyclic compounds of arsenic have been described.A base,which owes its molecular dissymmetry to the presence of two ring nitrogenatoms with stable " tetrahedral " configurations, has been resolved into d-and I-forms. A beginning has been made in the direction of discussingquantitatively steric effects in replacement reactions, and continued successfuluse has been made of optical activity for the study of molecularrearrangements.I n view of the growing biological importance of the inositols and theirderivatives, the opportunity has been taken t o review the developments inthis field, and an account has been given of the rarer methylpentose sugarsand their deoxy-derivatives, which are found combined in the cardiacglycosides. The use of chromatography to separate qualitatively andquantitatively mixtures of sugars and their derivatives is discussed ; themethod has been developed on the micro-scale for the separation of methyl-ated sugars and their glycosides.The section on the oxidation of a-glycolgroups (L. N. Owen, Ann. Reports, 1943,40, 115) with periodic acid and leadtetra-acetate has been continued. Sugar derivatives containing one ormore anhydro-rings of the septano-, pyrano-, furano-, and ethylene oxidetypes have been prepared and their structures elucidated. The Waldeninversion which occurs when an oxide ring is opened has been used t o provethe configuration of the amino-group in chondrosamine and to preparederivatives of the less accessible sugars.A summary is given of the more recent constitutional work on thestructure of the polysaccharides.A noteworthy advance has been made withthe enzymatic synthesis of amylopectin, and the degradative action of (3- andother amylases on this polysaccharide has received considerable attention.The polysaccharides from gum tragacanth, the c-galactan of the larch, anddamson tree gum have received a detailed examination. Gum tragacanth hasbeen shown to be a mixture ; the c-galactan is also considered t o be a mixtureby some authors ; damson tree gum appears to be homogeneous. All thesepolysaccharides are of the ramified type and odntain a variety of sugars.Stereochemical developments during the last eight years have confirmedthe general picture of the steroid nucleus given by Ruzicka in 1933, andevidence has been adduced to show that the various possible geometricalmodifications arising from chair-boat transformations of rings A and Bmake no contribution to the stereochemistry of the steroid nucleus.Methodsavailable for determination of the orientation of nuclear substituents aresummarised, and, in a review of the various nuclear positions, examples oftheir use are given. There is still no strict proof available to show that thehydroxyl group in cholesterol and cholestanol is (p)-orientated, but on thebasis of this assumption it has been possible to establish the configurationat C, of a whole series of derivatives of androstane, androst-5-ene, and theirhomologues. Attention is drawn to the fact that the configuration is knownat C7 in the bile-acid series but not in the sterol series; also that it seemsprobable that the previous arbitrary allocation of configuration at C7 in th140 ORGANIC CHEMISTRY.steroI series must be reversed.The proof given by Reichstein that theformerly accepted structure of deoxycholic acid is incorrect in regard toconfiguration at C,, and C17 has far-reaching repercussions; the sameconclusion, namely that the hydroxyl group at C,, and the side chain a tC17 are respectively ( a ) - and (@)-orientated, has been reached independentlyand on quite different grounds by groups of workers in America. It follows,inter alia, that, as originally suggested by Reichstein and Shoppee, the Cll-hydroxyl group characteristic of the natural adreno-cortical steroids has the(@)-configuration, and that the C,,-side chain in cholesterol, progesterone,corticosterone, the steroid sapogenins, and the cardiac aglycones is (@)-orientated.A further consequence is that the Digitalis heart poisons mustcontain a cis-C/D-ring union, a view now accepted by Ruzicka, Plattner,and their co-workers who, by the synthesis of steroids with hydroxyl groupsat C, and C,,, are preparing the way to the synthetic production of thesecompounds.In the field of reduced heterocyclic rings recent examples are discussedof the synthesis of piperidines and piperidones by (i) the reductive cyclisationof y-cyano-esters, (ii) the Eisleb double alkylation method (of a reactivemethylene group by di- p-chloroethylalkylamines), and (iii) the Dieckmann-like cyclisation of di-P-carbethoxyalkylamines. Attention is drawn to thepreparative advantages arising from the use of N-benzoylated and -nitros-ated derivatives as intermediates, and some new reactions of certain piperi-dines and piperidones are mentioned. Recent syiithetic work in the reducedbicyclic field (cycloalkano-pyrrolidines, -piperidines, and -thiazoles, andbicycloaza-alkanes) is briefly reviewed, and the stereocheinistry of @-biotinand its derivatives is summarised.A section on indoles deals with recentcritical studies of the Bischler synthesis from arylamines and a-halogeno-ketones, with improved prepsrative routes to trytophan and indole-3-aldehyde, with the synthesis of 7-azaindoles, and with the chemistry ofgliotoxin, the naturally-occurring antibiotic.Recent developments in thequinoline field are reviewed chiefly from the aspect of preparative improve-ments, particularly as applied to the synthesis of 4-hydroxyquinolines, alarge number of which have been prepared by condensation of arylamineswith ethyl ethoxymethylenemalonate followed by cyclisation or by variationsof this method. Brief mention is made of progress in the chemistry ofcinnoline derivatives. A comprehensive statement is given of the chemistryof the pterins, compounds which contribute to the pigmentation of the wingsof insects and which, until recently, were regarded as of academic interestonly. The chemistry of this group, which is related to the purines, haslately been greatly clarified and interesting and important biologicalproperties have been revealed.This section includes a description of thesynthesis and proof of structure of the antianzemic " Liver L. casei factor "(vitamin B,) and of its relationship to other active compounds, which arepterin derivatives, and with a general discussion of the vitamin B, problem.W. B.D. H. HHUNTER: THE HYDROGEN BOND. 1412. THE HYDROGEN BOND.The fact that hydrogen can sometimes link two other atoms together isnow beyond all question, but the mechanism of this virtual bivalency stillremains obscure.is by no means universally accepted, and an electrostatic union, largely theresult of the hydrogen atom being a bare proton with a minimum of envelop-ing electrons, has strong support.2 The latter view is in harmony withthe fact that the atoms linked by hydrogen bonds are confined almostentirely to the electronegative elements of small atomic radius (i.e., N, 0,B’), and that hydrogen bonds between atoms other than these are of theweakest kind.As between these two views there is good evidence forbelieving that resonance, if it contributes at all to the structure of thehydrogen bond, does so to a very inconsiderable e ~ t e n t . ~The resonance mechanism suggested by N. V. SidgwickMethods of Detecting the Hydrogen Bond.It is important in this connexion to distinguish between hydrogenbonds formed intramolecularly (by chelate rings) and those formed betweenmolecules. The former will usually lead to a unimolecular condition, thelatter to molecular association, and the resultant physical properties ofthe substance concerned will be largely influenced by these alternatives.It is evident, too, that the physical conditions under which assessment ofstructure is made will determine whether or not a substance will preserveits hydrogen-bond structure ; for example, a hydrogen-bond structure presentin the solid or liquid state may well be destroyed in dilute solution or inthe vapour state.Moreover, intermolecular hydrogen bonds are usualIymore sensitive than are intramolecular to the stresses imposed by vaporis-ation or dissolution.The following is a brief summary of the chief methods in general usefor the detection of hydrogen bonds.I. Interatomic Distance Methods.-These depend on the measurement ofthe distance between the atoms linked by the hydrogen bond, and arebased on the assumption that any approach of two such atoms to a distancesignificantly less than about 3.4 A.indicates a chemical link between them.The strengths of such bonds are in the inverse order of these interatomicdistances.Applied mainly to crystals, which are mostlikely to favour a maximum display of hydrogen bonds, this method hasyielded more information than any other about hydrogen-bond structures.It is, however, confined at present to relatively simple compounda (e.g.,inorganic salts), and becomes progressively more difficult to apply as1 Ann. Reports, 1933, 30, 112; 1934, 31, 40; “ Organic Chemistry of Nitrogen ”,Oxford University Press, 1937, xvii.2 L.Pauling, “ The Nature of the Chemical Bond ”, Cornell, 1940, p. 286.3 Manse1 Davies, this vol., p. 29.* The scope of diffraction methods as a guide to molecular structure is reviewed(a) X-Ray diffraction.*by J. M. Robertson, Tilden Lecture, J., 1945, 249142 ORGANIC CHEMISTRY.molecular complexity increases. It is not surprising, therefore, that itsapplication to organic hydrogen- bond structures has been limited to simpleexamples, although phthalocyanine,4 melamine,5 and hyper016 are notableexceptions. The method gives an accurate measure of the A-H-B distance,where A and B may be in the same or in different molecules, and, thoughit does not locate the hydrogen atom within this distance, it indicates withsome certainty the presence or absence of the hydrogen bond.For example,the presence of the O-H-0 bond in sodium hydrogen carbonate,7 potassiumdihydrogen phosphate and ar~enate,~ and ammonium periodate,(NH4)2H3106,10 and its absence in ammonium hypophosphite, NH,H2P02,11supports the chemical evidence that the former are true acid salts, whereasthe latter is not.Owing to the fact that this method is usuallyapplied to vapours a t low pressures, it is unlikely to reveal intermolecularhydrogen bonds. It is for this reason that the structures of hydrogenperoxide l2 and of hydrazoic acid l3 determined by this method give noindication of hydrogen bonding, although the physical properties of thepure substances clearly point to molecular association by hydrogen bonds.14On the other hand, the F-H-F bond in hydrogen fluoride is sufficientlypowerful to persist in the vapour, which is found l5 to consist of zigzagpolymers having a F-H-F distance of 2.55 A., the F-F-F angle being about140".This agrees well with the structure of solid hydrogen fluoride deter-mined l6 by the X-ray method, and seems incompatible with previouscyclic structures. The method has also been applied to the simpler carb-oxylic acids,11. Indirect (or Cmparative) Methods.-These depend very largely onthe fact that the engagement of a group in hydrogen-bond formationmodifies the physical, and to a lesser extent the chemical, properties of thegroup involved. They consist of a comparison of properties (many ofwhich may be capable of numerical expression) displayed by substanceswhich may possess a hydrogen-bond structure, with those of closely related(Miss) I.E. Knaggs and (Mrs.) K. Lonsdale, Proc. Roy. SOC., 1940, A, 177, 140;(b) EZectron di,ffruction.*the dimeric structure of which survives vaporisation.222 ORGANIC CHEMISTRY.and (LXXX), which were quantitatively reconverted into the parent acidsby hydrolysis with potassium carbonate. The C,,-hydroxyl group and theC1,-side chain must therefore lie on the same side of the general plane ofthe ring-system in these 12-epi-acids and on opposite sides in natural deoxy-cholic acid. This conclusion was supported by an examination 133 of thebehaviour of the methyl esters of the 12-epi-acids (LXXVII), (LXXIX),20-n- bisnordeoxycholic acid (LXXXI), and 20-isobisnordeoxycholic acid(LXXXIII) with phenylmagnesium bromide.Dehydration of the diphenyl-carbinols obtained from the esters of the 12-n-acids (LXXXI) and(LXXXIII), which results in disappearance of the centre of asymmetry atC,,,, gave the.same diphenylethylene (LXXXII) but no trace of a cyclicoxide. Similar dehydration of the diphenylcarbinols resulting from the12-epi-acids (LXXVII) and (LXXIX) gave the diphenylethylene (LXXXV)together, in the case of (LXXVII), with the cyclic oxide (LXXXIV).Me CPh, Me CO,HC. The @-Biotin Problem.-Work which has appeared since the lastreview in these reports42 has been mainly directed towards the elucidationof the steric configuration of p-biotin (LXII).Of the four racemates whichcan theoretically arise from the presence of three asymmetric centres (0)in (LXII), the three previously isolated by Harris and his co-workers43-dt-biotin, dl-allobiotin, and dl-epiallobiotin-have been studied from thisaspect in some detail.P4 It was shown that each of the alternative routes(LXIII) -+ (LXIV) ---+ (LXII) + (LXV) and (LXIII) --+ (LXVI) --+(LXVII) --+ (LXV) gave rise to the same dethiobiotin (LXV) when appliedto a given racemic bisacylamidothiophan derivative (LXIII) ; in this waydl-allo- (LXIII) and dl-epiallo- (LXIII) both gave dl-dethioalto- biotin (biologic-ally inactive), and dl-(LXIII) [the precursor of dl-biotin, of which natural(d-) p-biotin is one component] gave dl-dethiobiotin (biologically active),thus confirming that, in the three racemates (LXII), two have a trans-linkage a t C,-C, and one has a cis-, or vice versa.Now in the reaction(LXIV)-+ (LXII), the yield is almost theoretical in the case of dl-biotin,but this does not apply to the dl-alZo- and dl-epiallo-compounds. Thissuggests that dl-biotin has the cis-configuration at C,-C, (LXVIII), and thatthe dl-allo- and dl-epialbracemates are trans-compounds (LXIX) . This isstrikingly confirmed by the behaviour of the three racemates towards boiling(CXV.)The naturally-occurring pterins are mostly colourless or yellow com-pounds which show marked fluorescence in solution a t or near pH 7. Theyare found in the wings of various species of butterflies and ~ a ~ p ~ in the skin 134 and eyes 135 of fishes, and in the urine and liver of mam-m a l ~ .~ ~ ~ - ~ ~ ~ At an early stage of the researches on these compounds,which were carried out in the laboratories of H. Wieland and C. Schopf, i twas recognised 131 that a close relationship exists between the purines andleucopterin, one of the commonly-occurring pterins, and structure (CXV)was proposed 131 for this substance. Soon afterwards,132 it became necessaryto discard the formula C,,H,,O,N, in view of the isolation of guanidineas a hydrolytic product of a leucopterin derivative, and the loss of & of thetotal nitrogen of leucopterin on treatment with nitrous acid; to reconcilethese facts with the analytical data, a C19-N,, formulation was adopted,and this persisted, with minor variations, for some years.During thisperiod the chemistry of leucopterin and of its analogue, xanthopterin, waspainstakingly developed with small quantities of material ; and, althoughit was noticeable that many of the reactions (acetylation, chlorination withphosphorus pentachloride, action of nitrous acid, formation of glycol deriv-atives on oxidation with chlorine in various media) occurred in triplicate( L e . , implied the presence of three similarly-reacting groups in the pteririmolecules), simplified molecular formulz were not warranted in face of theanalytical data. The practical difficulties of the problem were unusuallyformidable. Apart from the labour of the isolation of pterins, whichinvolved the collection and manipulation of many thousands of butterfliesof a given species, these substances are insoluble in organic solvents, aredifficult to crystallise and purify, and decompose without melting ; theyoccur as hydrates which retain water very tenaciously and give spuriousanalytical data under ordinary conditions.128 R.Tschesche and H. J. Wolf, 2. physiol. Chem., 1937, 248, 34.129 M. Polonovski, R.-G. Busnel, and M. Pesson, Helv. Chim. Acta, 1946, 29, 1328.130 H. Wieland and C. Schopf, Ber., 1925, 58, 2178.131 C. Schopf and H. Wieland, ibid., 1926, 59, 2067.132 H. Wieland, H. Metzger, C. Schopf, and M. Biilow, Annalen, 1933, 507, 226.133 C. Schopf and E. Becker, ibid., p. 266; 1936, 524, 49; E. Becker and C. Schopf,134 R. Huttel and G. Sprengling, ibid., 1943, 554, 69; M.Polonovski, R.-G. Busnel,135 A. Pirie and D. M. Simpson, Biochem. J., 1946, 40, 14.136 W. Koschara, 2. physiol. Chem., 1936, 240, 127.13' Idem, ibid., 1943, 277, 159.ibid., p. 124.and M. Pesson, Compt. rend., 1913, 217, 163.138 Idem, ibid., p. 284; 279, 44252 ORGANIC CHEMISTRY.In 1940 the analytical difficulties were recognised and largely over-come,139 and the formula of leucopterin was disclosed as (C,H,O,N,),,where x = 1, 2, or 3. Synthesis of the pterin by fusion of 2 : 4 : 5-triamino-6-hydroxypyrimidine (CXVI; R = NH,) with oxalic acid l*O restrictedthe possible structures to (CXVII; R = NH,), (CXVIII), (CXIX), and(CXX). Of these (CXVIII) was excluded by the results of further applic-ations of the oxalic acid synthesis.Condensation of 4 : 5-diamino-2 : 6-OH OH OHN/\NH, N/\/%oH N'\NH,(CXVI.) (CXVII.) (CXVIII.) MeR!\N)'NH2 R\N/\N/ I 'I IOH ()AN/ I IINH,(CXXI.)(CXXII.)dihydroxypyrimidine (CXVI; R = OH) and of 4 : 5-diamino-6-hydroxy-3-methyl-2-pyrimidone (CXXI) with oxalic acid a t about 250" 141 gave,respectively, (" deaminoleucopterin ")(CXVII; R = OH) and the 3-methyl analogue (CXXII), which was notidentical with 3-methylxanthine-8-carboxylic acid (CXXIII),142 thusexcluding the alternative ring-closure of (CXXI). Structure (CXVIII) forleucopterin is thus eliminated if (as is very probable) the oxalic acid con-densations to give leucopterin, deaminoleucopterin, and (CXXII) all pro-2 : 6 : 8 : 9-tetrahydroxypteridineceed in the same sense; and this was proved for leucopterin and deamino-leucopterin by the conversion of the former into the latter by means ofnitrous acid.132 Further weight is given to this argument by the synthesisof 6-deoxyleucopterin (CXXIX ; see below), a transformation product ofleucopterin, by the oxalic acid method.Furthermore, purine-8-carboxylicacids are readily decarboxylated , whereas leucopterin does not decomposebelow 400".139 H. Wieland and R. Purrmann, Annalen, 1940, 544, 163.140 R. Purrmann, ibid., p. 182.142 W. Traube, ibid., 1923, 432, 266.141 Idem, ibid., 1941, 546, 98SIMPSON : HETEROCYCLIC COMPOUNDS. 253The choice between (CXVII; R = NH,), (CXIX), and (CXX) as thestructure of leucopterin was finally settled by the elucidation of the con-stitution of xanthopterin, another naturally-occurring pterin, and of itsrelationship to leucopterin.Fusion of (CXVI; R = NH,) with dichloro-acetic * acid gave the amide (CXXIV), which on cyclisation under mildconditions yielded xanthopterin ; 141 this is therefore 2-amino-6 : S-dihydroxy-pteridine (CXXV). Now xanthopterin takes up an atom of oxygen in con-tact with platinum in weakly acid solution, yielding leucopterin 139 (thesame result is obtained by treatment of xanthopterin with methylene-blueand an enzyme preparation,145 and by the prolonged action of hydrogenperoxide),lM and, as leucopterin is devoid of peroxidic properties,l43 thereaction can only be represented by hydroxylation of xanthopterin at C,.Leucopterin is therefore (CXVII; R = NH,).A third naturally-occurring pterin is 8-deoxyleucopterin or isoxantho-pterin (CXXVII), which was synthesised by hydrolysis, followed by decarb-oxylation, of the ester (CXXVI), obtained from (CXVI; R = NH,) anddiethyl ketoma1onate.l4 When this pterin was first isolated,l32 it wasgiven the name of anhydroleucopterin ; the above synthesis, however,discloses its relationship to leucopterin, and it has been obtained fromleucopterin by electrolytic reduction.145 The reverse reaction, vuiz., oxid-ation of isoxanthopterin to leucopterin, has not yet been achieved, butthe action on isoxanthopterin of nitrous acid and of chlorine water givesthe leucopterin derivatives (CXVII; R = OH) and (CXXXII; R = H)respectively. 146Several other ‘f natural pterins have been described.Uropterin, isolatedfrom urine,l36 has been proved to be ~anth0pterin.l~’ Another urinarypterin, u r ~ t h i o n , l ~ ~ is more complex. Its molecular formula is CllH,303N5S,.Both sulphur atoms are inert, and no thiol group is present. The pigment,unlike other pterins, is opticany active. Although there is as yet no rigid143 H. Wieland and R. Purrmann, Annalen, 1939, 539, 179.1 4 4 R. Purrmann, ibid., 1941, 548, 284.145 H. Wieland and R. Liebig, ibid., 1944, 555, 146.1 4 6 H. Wieland, A. Tartter, and R. Purrmann, ibid., 1940, 545, 209.* The use of the bisulphite compound of barium glyoxylate in sulphuric acid,instead of dichloroacetic acid, gives a greatly improved yield.137An interesting general account of the occurrence of pterins in wing-pigments isgiven by (Sir) F.G. Hopbs (Proc. ROY. soC.9 1942, B, 130, 359). It should be notedthat the purple pigment, rhodopterin, which is there discussed, is not a true pterh,but a condensation product of leucopterin and xanthopterin-9-carboxylic acid, andthat it has been re-named pterorhoh (R. Purnmnn and M. Maas, Annahm, 1944,660, 186)254 ORGANIC CHEMISTRY.proof that the molecule of urothion contains the pteridine skeleton, theexpression (CXXVIII) has been advanced on the basis of the foregoingdata and the following evidence. Urothion yields a tetra-acetyl derivative,which gives satisfactory cryoscopic molecular weight values and can behydrolysed to a monoacetyl derivative. The pigment is amphoteric, andamino-nitrogen estimations suggest the presence of a guanidine residue ;its ultra-violet absorption spectrum resembles those of xanthopterin, ribo-flavin, and other isoalloxazines ; and periodic acid oxidation yields form-aldehyde and a product, C,,H,O,N,S, (urothionaldehyde).The pigmentalso gives, with concentrated sulphuric acid, the red colour (thiophenolreaction) characteristic of compounds containing a thiol, or potential thiol,group attached to an aromatic ring.(also known as fluorescyanine),l29is likewise a pterin of unknown structure. It is reduced by fuming hydriodicacid with liberation of iodine, and on dilution the leuco-compound isreoxidised by the iodine. This extremely easy reversible oxidation-reductionis shown only by isoxanthopterin 146 and the acid 144 obtained by hydrolysisof (CXXVI). Xanthopterin is also reduced under the same conditions,l46but the dihydro-compound is not reoxidised by iodine, although it can beoxidised to the pterin by a variety of other reagents.14' Leucopterin, onthe other hand, is much more difficult to reduce,145* 146 but under appropriateconditions yields isoxanthopterin 145 or dihydr~xanthopterin.~~' * A con-trolling factor affecting the redox potential thus appears to be the pointof hydroxylation of the pyrazine ring; for this reason, and also becausethe absorption spectra of isoxanthopterin and ichthyopterin are very similar,it has been suggested 134 that this marine pterin is a derivative of 9-hydroxy-pteridine.It is to be noted, however, that the suggested molecular formula,C,H,O,N,, implies that it is a dihydro-derivative of this ring-system.Properties of Pterins.-Excluding differences in elementary composition,the criteria by which individual pterins can most readily be distinguishedare basicity (this may be very slight or considerable; acidic properties arewell-marked), fluorescence and the effect of pH thereon, absorption spectra,and the redox reaction already noted. The fluorescence of pterins has beenstudied in some detail ; lM, 1489 149, leucopterin exhibits its maximumfluorescence in strongly alkaline solution,149 but under these conditionsxanthopterin fluoresces only slightly, the intensity increasing rapidly withfall in pH.14, Various measurements of the ultra-violet absorption spectraof pterins have been recorded (frequently with similar data for purines andThe fish-skin pigment, ichthyopterin147 J.R. Totter, J . Biol. Chem., 1944, 154, 105.lPs P. Decker, 2. physiol. Chem., 1942, 274, 223.l50 M. Polonovski, S. Guinand, M. Pesson, and R. Vieillefosse, Bull. SOC. chim.,1945, 12, 924.* The reduction of leucopterin to dihydroxanthopterin, followed by oxidationof the latter with alkaline silver nitrate, enables a convenient synthesis of xantho-pferin to be effected from (CXVI ; R = NH,) and oxalic acid."'W. Jacobson and D. M. Simpson, Biochem. J . , 1946,40, 3 , 9SIMPSON : HETEROCYCLIC COMPOUNDS. 255flavins), but no systematic study has yet been made.l36, 138, 150-162, 162A useful, but not exhaustive, summary of the chemical and optical data isgiven by W.Jacobson and D. M. S i m p s ~ n . l ~ ~When leucopterin is treated with phosphorus pentachloride a mono-chloro-derivative is obtained. That the 6-position is involved in thisreaction was shown by reduction of the chloro-compound to the deoxy-derivative (CXXIX), which was synthesised from 2 : 4 : 5-triaminopyrimidineand oxalic acid.146 Application of the reaction to deaminoleucopterin(CXVII; R = OH) gave, similarly, the 2 : 6-dichloro-compound,153 butunder different conditions of isolation 2 : 6 : 8 : 9-tetrachloropteridine (CXXX)c1 c1(CXXIX. ) (CXXX.)was obtained, and it was found that the 2 : 6-dichloro-derivative had beenformed * by partial hydrolysis of (CXXX) under the conditions of isolation.In contrast, drastic alkaline hydrolysis was needed to convert the dichloro-derivative (2 : 6-dichloro-8 : 9-dihydroxypteridine) into (CXVII ; R = OH) ;the corresponding dichloropyrimidine (CXXXI) is also resistant .153 Inci-dentally it may be noted that the production of a tetrachloro-derivativefrom deaminoleucopterin is not possible on the basis of the purine-8-carb-oxylic acid structure (CXVIII) for leucopterin ; its formation thus constitutesan independent proof of the correctness of (CXVII; R = NH,).Before the constitution of leucopterin and xanthopterin had been settledby synthesis, a number of degradation products had been isolated duringattempts to unravel the complexities of the supposed C,, structures.Formul-ation of these reactions on the basis of (CXIV) illustrates clearly the closeparallel between them and well-known purine transformations.Oxidationof leucopterin with chlorine water or chlorine in methanol leads to the glycol(CXXXII; R = H) or its dimethyl ether (CXXXII; R = Me) respect-i ~ e 1 y . l ~ ~ Hydrolysis of (CXXXII; R = H) results, as with uric acid, inthe conversion of the pyrimidine into a hydantoin ring and in the isol-ation of derivatives of the latter, vix., (CXXXIII), (CXXXIV), and(CXXXV).139,154 The product formed by the action of chlorine in aceticacid on leucopterin is (CXXXVI),132, 139 corresponding to the formation oflS1 H. K. Mitchell, J . Amer. Chem. SOC., 1944, 66, 274.lS2 H. Fromherz and A. Kotzschmar, Annalen, 1938, 534, 283.lS3 C.Schopf and R. Reichert, ibid., 1941, 548, 82.154 H. Wieland and A. Kotzschmar, ibid., 1937, 530, 152.* It is of considerable interest that (CXXX) is apparently more readily hydrolysedin alkaline than in acid solution ; this is in direct contrast to recent evidence (see, forexample, C. K. Banks, J . Amer. Chem. Soc., 1944, 66, 1127, 1131 ; A. J. Tomisek andB. E. Christensen, ibid., 1945, 67, 2112; C. K. Banks and J. Controulis, ibid., 1946,88, 944) that hydrolysis of chloro-heterocyclic compounds is acid-catalysed, and suggest8that a different mechanism may be operative in the hydrolysis of (CXXX)256 ORGANIC CHEMISTRY.5-hydroxypseudouric acid from uric acid ; 155 analogously, oxidation ofdeaminoleucopterin (CXVII; R = OH) with chlorine in methanol givesOR&dkaliNH-=O(CXXXV.) HN//jN/JNH*CO*C02HH H (CXXXIV.)(CXXXVII).166 The glycol ether (CXXXII; R = Me) is extremelyunstable; in aqueous solution at 40" it yields the monoether (CXXXVIII),which readily decomposes further into (CXXXIX), guanidine, and carbondioxide by hydrolytic fission.132 Reference has already been made to theformation of leucopterin by hydrogen peroxide oxidation of xanthopterin ;OMethe reaction is not, however, quantitative, and under suitable conditionsimino-oxonic acid (CXL) is also formed.139 This reaction resembles theoxidation of uric to oxonic and, indeed, (CXL) is also formed fromH.Biltz and M. Heyn, Annalen, 1917, 413, 7.lS6 H. Wieland and A. Tartter, ibid., 1940, 543, 287.16' F.J. Moore and R. M. Thomas, J . Amer. Chem. SOC., 1918, 40, 1120; H. Biltzand R. Robl, Bw., 1920, 68, 1967SIMPSON : HETEROCYCLIC COMPOUNDS. 257the purine (CXLI) .l39 Oxidation of xanthopterin with hot sodium chlorateand acid, or with cold nitrosylsulphuric acid, brings about disruption ofthe pyrimidine as well as of the pyrazine ring, and oxalylguanidine (CXLII)is formed; this is also produced by similar treatment of (CXVI; R =NH,).15*Other Synthetic Pterins.-M. Polonovski, R. Vieillefosse, and M . Pesson 159have prepared, from (CXVI; R = SH) and 1 : 2-dicarbonyl compounds,three non-fluorescent 2-mercaptopterins (CXLIII; R = SH; R’ = H,CO,H, Ph; R” = H, OH, Ph), which were converted into 2-hydroxy-analogues l6O, by alkaline hydrogen peroxide ; these authors have alsoshown that the mercaptopterins undergo S-ethylation, and they conclude,from the fluorescence shown by the 8-alkyl- and the hydroxy- (in contrast tothe 2-thiol) derivatives, that the characteristic fluorescence of pterins dependson the preservation of an intact * aromatic structure in the pyrimidine ring,i.e., the 2-hydroxy-compounds exist as such whereas the 2-thiol derivativesexist in the tautomeric form.Condensations between (CXVI; R = NH,and OH) and a-diketones have been extended to include phenanthraquinoneand acenaphthenequinone.162 The original use 132 of the term “ isoleuco-pterin ” now appears unwarranted; 146 instead, the name is given to thesynthetic pterin (CXLIV) .145 This substance, unlike leucopterin, fails toreact with nitrous acid (isoguanine and guanine are similarly differenti-ated); 145 on the other hand, the xanthopterin molecule is disrupted by thisreagent and does not yield the deaminoxanthopterin (CXLIII; R = R’ =OH; R” = H) obtainable from (CXVI; R = OH) and the bisulphitecompound of glyoxylic 15*The Vitamin B, ProbZem..F-Casual observation of progress in this fieldhas hitherto been somewhat difficult owing to the apparent complexity ofthe problem.At an early stage in the purification of the one or moregrowth factors having antianaemic and/or microbiological (L. casei E,S. Zactis R, 8. fmcalis R) growth-stimulating properties it became clear thatthe biological characteristics of the product were dependent on the source(liver, yeast, spinach, and other vegetable sources).Thus Peterson and15* C. SchBpf and A. Kottler, Annalen, 1939, 539, 128.169 Bull. SOC. chim., 1945, 12, 78; see also ref. 150.l 6 0 R. Kuhn and A. H. Cook, Ber., 1937, 70, 761.161 K. Ganapati, J . Indian Chem. Soc., 1937, 14, 627.162 C. K. Kain, BI. F. Mallette, and E. C . Taylor, jun., J . Amer. Chem. SOC., 1946,68, 1996.* In the opinion of the Reporter, a direct correlat.ion of fluorescence with “ aromatic-ity ” produced via prototropy seems to be an over-simplification. Arguments whichmay have some bearing on this point, and which are certainlyrelevant to the wholequestion of the fine structure of pterins and other hydroxy-heterocyclic nitrogen com-pounds, have recently been advanced by F.Arndt (Rev. Pac. Sci. Univ. Istanbul, 1944,9, 19), who discusses the conception that the “ aromaticity ” of such compounds iscompatible with their existence in the keto-dihydro- (CO-NH) form by virtue of anelectromeric shift to ~--C&H-, and is thus potentially independent of tautomerism.REP.-VOL. XLIII. IThis problem is dealt with later in its biochemical aspect (p. 296)258 ORGANIC CHEMISTRY.his co-workers 163 obtained from liver and from yeast a " norite eluatefactor " essential for growth of L. casei (A. helveticus), which also appearedto be vital to the growth of chicks; 164 and the preparation from spinachof a factor, designated folic acid, having growth-stimulating properties forL. casei, 8. lactis R, and other bacteria, was reported almost simultaneouslyby H.K. Mitchell et Later, J. J. Pfiffner et ~ 1 . l ~ ~ described the isolationof a crystalline antianEmic factor from liver extracts, which, following anearlier suggestion,167 was named vitamin B, ; identity of this substancewith the norite eluate factor was claimed,166 and identity with folic acidwas suggested.166 It was then found by J. C . Keresztesy et aL168 that'' various types of extracts and liver preparations " yielded a substancewhich, although much more active than folic acid for 8. Zactis R, was inactivefor L. cusei, whereas folic acid is equally effective for each micro-organism.On the other hand, E. L. R. Stokstad, working with crystalline preparationsfrom liver and from found that the liver factor was equally activefor L.casei and for 8. lactis R, but that the yeast factor was only half asactive as the liver preparation for S. Zactis R, whereas both preparationswere equally effective for L. casei; and a new L. casei factor from an undis-closed source 170 (later described 171 as a fermentation residue; the factoris named the fermentation L. casei factor) 171, 174 was stated to be 85-90% as active as the norite eluate (liver) factor for L. casei, but only 6%as active for S . Zmtis R.It is clear from these results that several factors are involved, and thisconclusion is substantiated by the results of antianBmic studies. Followingthe original observation that monkey anmnia could be cured by yeastextracts (" vitamin M "),172 it was found that chicken anEmia could like-wise be cured by a crystalline yeast factor and also by vitamin B,, whichwas chemically distinct from the yeast f a ~ t 0 r .l ~ ~ Vitamin B, thus possesses163 E. E. Snell and W. H. Peterson, J . Bact., 1940, 39, 273; B. L. Hutchings,N. Bohonos, and W. H. Peterson, J. Biol. Chem., 1941, 141, 521.16* B. L. Hutchings, N. Bohonos, D. M. Hegsted, C. A. Elvehjem, and W. H.Peterson, J . Biol. Chem., 1941, 140, 68l-.165 H. K. Mitchell, E. E. Snell, and R. J. Williams, J . Amer. Chem. SOC., 1941, 63,2284; 1944, 66, 267. See also E. H. Frieden, H. K. Mitchell, and R. J. Williams,ibid., 1944, 66, 269; H. K. Mitchell and R. J. Williams, ibid., p. 271; H. K. Mitchell,ibid., p. 274.166 J. J. Pfiffner, S. B. Binkley, E. S. Bloom, R.A. Brown, 0. D. Bird, A. D. Emmett,A. G. Hogan, and B. L. O'Dell, Science, 1943,97, 404.167 A. G . Hogan and E. M. Parrott, J . Biol. Claem., 1940, 132, 507.I b 8 J. C. Keresztesy, E. L. Rickes, and J. L. Stokes, Science, 1943,97, 465.Isa E. L. R. Stokstad, J . Biol. Chem., 1943, 149, 573.l70 B. L. Hutchings, E. L. R. Stokstad, N. Bohonos, and N. H. Slobodkin, Science,1944, 99, 371; see also E. S. Bloom, J. M. Vandenbelt, S. B. Binkley, B. L. O'Dell,and J. J. Pfiffner, ibid., 100, 295.171 R. B. Angier et al. (for names see ref. 174), ibid., 1945, 102, 227.172 P. L. Day, W. C. Langston, and W. J. Darby, Proc. Soc. Exp. Biol. Med., 1938,J. J. Pfiffner, D. G . Calkins, B. L. O'Dell, E. S . Bloom, R. A. Brown, C. J. Camp-38, 860.bell, and 0.D. Bird, Science, 1945, 102, 228SIMPSON : HETEROCYCLIC COMPOUNDS. 259both antianaemic and microbiological (L. casei E) activity; the crystallineyeast factor (known as vitamin B, conjugate), on the other hand, has verylittle microbiological activity (L. casei, 8. fcecalis), but yields vitamin B,on enzymic digestion.173From this seemingly confused background the following clarificationshave emerged as a result of recent work : (a) proof of structure and synthesisof the liver L. casei factor (pteroylglutamic acid); (b) identification ofpteroylglutamic acid with vitamin B, ; (c) establishment of the chemicalrelationship between vitamin B,, vitamin B, conjugate, and the ferment-ation L. casei factor.Structure of L. casei Factor.-The constitution of this substance hasbeen proved to be (CXLV) by a group of sixteen workers in the followingFission with sulphurous acid of the fermentation L.caseifactor yielded an amine fraction (a) together with 2-amino-6-hydroxy-pteridine-&aldehyde (CXLVI ; R = CHO). The orientation of the aldehydewas determined (i) by its conversion under anaerobic alkaline conditionsinto the corresponding acid (CXLVI; R = C0,H) and (CXLVI; R = Me),followed by vigorous hydrolysis of the latter, by the method of J. Weijlardl74et u Z . , ~ ~ ~ to the known 175 2-amino-5-methylpyrazine (CXLVII) ; (ii) by theconversion of the known acid derived from (CXXVI) 144 into (CXLVI;R = C0,H) by means of phosphorus pentachloride and hydriodic acid;and (iii) by decarboxylation of (CXLVI; R = C0,H) and synthesis of theresultant 8-deoxyxanthopterin from (CXVI; R = NH,) and glyoxal.The$-methyl derivative (CXLVI; R = Me) was also obtained by decarboxyl-ation of (CXLVI ; R = CH,CO,H), itself prepared by condensation of(CXVI; R = NH,) and ethyl 2-keto-3 : 3-dimethoxy-n-butyrate. Acidhydrolysis of the amine fraction (a) gave p-aminobenzoic acid and glutamicacid (3 mols.).The fermentation L. casei factor was converted by anaerobic alkalinehydrolysis into the liver L. casei factor and an a-amino-acid (2 mols.) ; mobicalkaline hydrolysis, on the other hand, gave (CXLVI; R = C0,H) and anamhe fraction from which p-aminobenzoic acid was obtained by further17' R. B. Angier, J. H. Boothe, B. L. Hutchings, J. H. Mowat, J.Semb, E. L. R.Stokstad, Y. SubbaRow, C. W. Waller, D. B. Cosulich, M. J. Fahrenbach, M. E. Hult-quist, E. Kuh, E. H. Northey, D. R. Seeger, J. P. Sickels, and J. M. Smith, jun., ibid.,1946, 103, 667.17' J. Weijlard, M. Tishler, and A. E. Erickson, J . Amer. Chm. Soc., 1946, 07, 802260 OROANIU OHEMISTRY.hydrolysis. Consideration of these results, together with those obtainedby the sulphurous acid degradation, indicated that the liver L. casei factorhas the structure (CXLV), and that the introduction of two more glutamicacid residues into this molecule produces the fermentation L. casei factor.Synthesis of the liver L. casei factor was achieved by two methods as shownin the accompanying scheme, the yield in each case being ca. lSy,. Theintermediate (CXLIX) was derived from I(+)-glutamic acid. It will benoted that each synthesis proceeds through a dihydro-derivative andsubsequent in situ oxidation.(a) (CXVI ; R = NH,)yHBr*CH2Br + H2N<I>O*NH*CH*[CH2]2*C0,H (CXLIX.)+ 7 0 2 3 3CHO I acetate buffer(CXLVIII .) (UXLIX) + NaOMe + 4,(CXLV) fJ.J. Pfiffner et al. have shown 176 that vitamin B, conjugate consists ofthe vitamin combined with six glutamic acid residues in peptide form;comparison of the vitamin itself with pteroylglutamic acid showed that thetwo substances are identical. The biological specificity of vitamin B, con-jugate, of pteroylglutamic acid, and of the fermentation L. cusei factorthus depends on the nature of the acid side chain (or chains) attached toa common nucleus; and this conception is strengthened by the observationof Angier et al.174 that, if p-aminobenzoic acid is used instead of (CXLIX),syntheses (a) and (b) lead to a product which is active for 8.fceculis R butdevoid of activity for L. m e i and for chicks.It has also been shown that, for. antianmnic activity, the presence of aside chain is unnecessary. Thus nutritional anzmia of rats128 and offish 179 can be cured by administration of xanthopterin, and ichthyopterin(fluorescyanine) is curative in riboflavin-deficient rats and in aneurin-deficient rats and ~ i g e 0 n s . l ~ ~ A number of other synthetic pterins alsopossess this interesting dual biological activity for the rat and the pigeon;certain micro-organisms (Glaucoma, Polytomella C ~ c a ) , however, are moreexacting in their requirements, and are unable to utilise pterins in place ofa n e ~ r i n . l ~ ~In conclusion, two points of nomenclature should be mentioned. Inthe first place, recent comments on the synthesis of pteroylglutamic176 J. J. Pfiffner, D. G . Calkins, E. S. Bloom, and B. L. O’Dell, J . Amer. Chem. SOC.,17’ K. A. Jensen, Dansk Tidsskr. Farm., 1946, 20, 219; Lancet, 1946, i, 969;1946,68, 1392.1946, ii, 532, 680SIMPSON : HETEROCYCLIU COMPOUNDS. 261acid refer to this substance as folic acid, whereas the American workersconsistently l717 17*,l78 use the names pteroylglutamic acid or liver L. caseifactor when referring to their synthetic product. This distinction shouldclearly be retained for the present, because no announcement has beenmade of formal proof that the folic acid of H. K. Mitchell et is identicalwith pteroylglutamic acid; indeed, B. L. Hutchings et al. have pointedout 170 that absorption spectra measurements indicate that folic acid isnot identical with vitamin B,, vitamin B, conjugate, or the fermentationL. casei factor, and no modification of this view has appeared in literatureavailable t o the Reporter. Secondly, the American workers have departedfrom the established numbering of the pteridine nucleus (based on analogywith the purine ring-system), and have used a method 1627 174 based onlumazine 175 as the parent nucleus; this introduces an unnecessary com-plication, and the established nomenclature has been used throughout thisReport. J. C. E. S.W. BAKER.D. H. HEY.L. HUNTER.M. M. JAMISON.J. K. N. JONES.M. S. LESSLIE.C. m7. SHOPPEE.J. C. E. SIMPSON.E. E. TURNER.178 R. €3. Angier, Dansk Tidsskr. Farm., 1946, 20, 288.179 R. TVY. Simmons and E. R. Norris, J . Biol. Chem. 1941, 140, 679