You are on page 1of 5

THE CHEMISTRY OF THE "AMINOCHROMES"*

P A R T I. T H E P R E P A R A T I O N AND P A P E R C H R O M A T O G R A P H Y O F P U R E ADRENOCHROME1

ABSTRACT T h e preparation of adre~iochrornein a pure stable crystalline form has been carried out b y with the use of an anion-exchange resin the silvcr oxide osiclation of adrenaline in ~ n e t h a n o l (Dowes-l(C1-)) t o remove heavy metal ions from the reaction l n i s t ~ ~ prior to the isolation re of the product. Its paper chromatographic behavior together with t h a t of three derivatives (adrenolutin, adrcnochrome n~onoremicarbazone, and adrenochrome monoisonicotinic acid hydrazide) in six different solvent systems has been examined. \Water was found to be t h e best paper chromatographic solvent so far esamined for this series of C O I I I ~ O L I I I ~ S .

Adre~lochrome( I ) , the substa~lce mainly responsible for the red colors produced during the mild oxidation of adrenaline (11), was hrst isolated from the products of an enzymatic oxidation by Green and Richter ( I ) . Subsequently, inorganic oxidizing agents, particularly

silver oxide, were successfully employed and i t was observed that adrenochrome could be crystallized from methanol (containing a little formic acid) a t -80" (cf. Veer (2); MacCarthy (3) ; Harley-Mason (4) ; Sobotka and Austin (5)). Green and Richter reported t h a t (1) adrenochrome was unstable both in the solid state and in solutio~l ; however, Sobotka and Austin claim that the dry crystalline nlaterial can be liept indefinitely a t 0" (5). The stability of aqueous solutions of adrenochrome has been shown to bemarliedly affected by pH (cf. Zambotti and i\Joret (7)); under alkaline conditions and in the presence of certain metallic cations, particularll. zinc, adrenochrome readily rearranges into adrenolutin (N-n1ethyl-3,5,G-i11doletriol, (cf. Lund (8) ; Fischer, Derouaux, Lainbot, and 111) Lecomte (9); Harley-h11ason and Bu'Locli (10) ; Fischer and Lecomte (11)).

Manz~scriptreceived Decernber 11, 1.957. Contribzitiofz from the Psychiatric Research U n i t , U?ziversity Hospitnl, Saskntoon, Saskatchewan. T h i s investigation was supported by grants froriz the Departnzeltt of Health and Welfare, Ottawn, and the Rockefeller Foundation. T h i s work was carried out under the az~spiceso f the Saskatchewan Conzmittee on Schizophrenia Research. * T h e term "anzinochronzes" was suggested by Sobotka afzd A u s t i n (5) for the highly colored cyclic oxidation and These szibstances are often fornzulated products of P-(S,4-dihydroxypheny1)-ctltylamines related compo~~?zds. a s substituted 2,s-dilzydroindole-5,6-qz~iqiones, although there i s some ulzcertainty u s to the existence of a true o-quinonoid strzicture (4, 6 ) .
Can. J. Chem.

Vol. 3G

(1958)

854

C.-\NADIAN JOURNAL

O F CHEMISTRY. VOL. 36. l95S

Rzlany samples of adrenochrome, prepared by existing methods, that were available to the authors appeared to contain varying amounts of adrenolutin (sho\vn paper chromatographically by a methocl to be discussed later) and black water-insoluble melanin-lilie compounds. I t appeared not improbable that contamination of the products by metal in a colloidal form or metallic ions, derived from the oxidizing agent, might be respoilsible for the deterioration of the samples and for the apparent instability of their aqueous solutions. In order to eliminate or drastically reduce the contamination by silver ions, the adrenaline - silver oxide reaction mixture in methanol was filtered through a n anionexchange resin bed in the chloride form prior to crystallizatioil (at -20) in orcler to remove ionic silver from solution as silver chloride. In this manner a highly crystalline product containing only traces of silver was obtained ( k 0 . 0 1 7 0 ) . The paper chromatograpllic behavior of adrenochrome and related compouncls has recently been studied by Fischer using n-butano1:aceticacid: water (8:2:2) andi-propanol: 0.880 ammonia: water ( 8 : l : l ) solvent systems (12). T h e results reportecl were unsatisfactory, since some of the substances apparently decomposed in these solvents. This is not altogether surprising in view of the known sensitivity of these compounds toward acicls and alltalies (7, 8 , 9), aild furthermore Pavolini, Gambarin, ancl Godenigo in a study 011 the action of ammonia on some 5,G-indolequinoiles, including adrenochrome, reportecl a definite reaction in this case with the formation of an unstable substance, exhibiting a yellow-green fluorescence, \vhich rapidly turns brown (13). W e have repeated Fischer's worlc and observed with the i-propanol-ammonia system that the bright red non-fluorescent adrenochrome spot rapidly turned brownish-yellow in color and exhibited a yellowgreen fluorescence, even on exposure to amrno~lia fumes, and in contact with the solvent the spot soon became dark brown indicating that marlied decomposition had occurred. In the acid solvent system, a slower but definite decomposition of the adrenochrome spot was observed during the chromatography. We have found that dilute aqueous solutions of adrenochrome prepared by our method are relatively stable and that distilled water and 2% acetic acid in water are the most suitable solvents for paper chromatography. Pure adrenochrome gave a single spot Rf0.8&0.02 in water on previously washed Whatman No. 1 paper. Aclrenolutin had an R f of ca. 0.4 under these conclitions. Contamination of adrenochrome samples with adrenolutin can easily be observed paper chroinatographically, since the characteristic yellow adrenolutin spot showing a strong yellow-green fluorescence is sooil visible behind the red adrenochrorne spot; this can be observed very quickly as it is only necessary to allo\\l the chromatogram to run a few inches in order to see the two spots. The adrenolutin spot is quite distinct and not a diffuse "tail" to the adrenochrome spot, indicating that the ilnpurity was present initially and not formed by continuous isomerizatioil of the starting material. Polyineric melanin-type impurities s t a r in the position of the original spot. An interesting phenomenon was observed 011 clrying of the paper chromatograms; as the paper was allowed to d r y a t rooin temperature, the color of the adrenochrome spot gradually changed froin red to yello\v-bro\vn and the spot began to exhibit the characteristic adrenolutin fluorescence. The partial formation of adrenolutin was confirmed by two-dimensional chromatography, using water as running solvent in both directions, although only one spot KfO.S was observed in the initial run; after drying and rerunning a t 90' to the original direction two distinct spots with the characteristic R,'s for adrenochrome and adrenolutin appeared. Euler, in describing the paper chro~natographyof adrenaline, states that after the e chroinatograms have beell dried a d r e ~ ~ a l i nspots s h o ~ va11 apple-green fluorescence (14, p. IG). A possible expla~lationof this observation n~oulclbe the air-oxidation of the

I-IEACOCK ET

AL.: AMINOCI-IROMES.

855

spot to adrenochrome followed by isomerization to adrenolutin on the paper. Therefore it would appear that a thin film of adrenochrome 011 paper will spontaneously rearrange into adrenolutin. This type of behavior has also been observed by Austin, Chanley, and Sobotka with the related compound epinochrome (IV), which even in the crystalline form undergoes a slow spontaneous rearrangement to 5,6-dihydroxy-N-methylindole (V) (15, 16). The paper chromatography of adrenochrome, adrenolutin, adrenoxyl (adrenochrome monosemicarbazone) (VI), and adrenochrome monoisonicotinic acid hydrazide

(VII) has been studied i l l six different solvent systems: (a) water, (b) 2% acetic acid in water, (c) 75% methanol in water, (d) 75y0 ethanol in water, (e) n-butanol: acetic acid: water (8 :2: 2), and Cf) i-propanol : 0.880 ammonia: water (8:l : l ) . Although all solvents proved satisfactory for either oi the two derivatives, VI and VII, only ( a ) and (b) were suitable for adrenochrome and possibly oilly (a) for adrenolutin, since even in (b), (c), and (d) the fluoresceilce characteristics of the compou~ld seem to be irreversibly altered. An attempt to purify impure preparations of adrenochrome on cellulose columns using water as solve~lt was disappointiilg, since, although pure aqueous solutions of adrenochrome were probably obtained, it was not possible to crystallize the adrenochrome from water, and all samples prepared by freeze-drying under high vacuum (a technique which gave a very finely divided product) were shown chromatographically to be contaminated t o some extent with adreilolutiil (fortuitously formed during the drying process). In view of the very considerable interest that has developed, in the pl~ysiological activity of adrenochrome and related compounds, in recent years (see Baccl (17) and Hoffer (18) for lists of references) the necessity for working with the compounds in a pure and stable form has become of paramount importance.

Prefinratiofz o Adrenochrome f I.-Adrenaline (9.13 g.) \vas suspended in methanol (360 ml.) and '30% formic acid added dropwise with stirring until a clear solution was obtained (ca. 3.5 nil.). Freshly prepared silver oxide (36.3 8.) was added portioilwise during a period of 3 minutes, the reaction temperature being maintained between 18" and 23" during the addition of the oxidant. The reaction mixture was filtered (with suction) through a Dourex-l(C1-) (200/400 mesh) resin bed (diam. = 6.5 cm., height = 3.7 cm.)" and the deep red filtrate was allowed to crystallize slo~vly t -2.0". Pure adrenochrome (l.G g.) was obtained i l l a deep-violet needles, m.p. 112" with decomposition. Found: C , 60.30; H, 4.69; S , 7.76. Calc. for CgHg03N:C, 60.33; H , 5.02; N , 7.81y0. The ultraviolet and visible absorptioil spectra were measured in aqueous solution (A,,,: 301, 487 mp; Amln: 262, 361 mp). Samples of adrenochrome which we have prepared by repeating existing procedures contain, in our view, less well defined crystals than the product obtained as described above (rapid cooling to -80" tends to produce an ;~morphousproduct) and the products are
" T I I Eresilt zuas prepared b y e s f c r ~ s k w a s l ~ i n g :(1) milh 3 N ~ ~ y d 1 . 0 ~ ~ 2acid, i (2) 7i'ith ziratcr 1171lilt1e111r(12o z ~0r c t liti~lzls, n d (3)f i n a l l y .ioillt ,irell1,a71,olzi~clllthe roasiz,i?~gs a ruere t r a ? t s p a ~ e n to zlltraviolet light. t

856

C.-\N.ADIAN JOURN:\L

O F CHEMISTRY. VOL. 3Li, 1958

invariably Inore or less contaminated with adrenolutin and melanin. This has been demonstrated paper chromatographically (using Whatman No. 1 paper with water as running solvent). No particular advantage either in yield, purity, or crystalline form, appeared to attend the use of absolutely d r y methanol. T h e silver contents of adrenochrome sanlples prepared in this mallller and by the method of Sobotlta and Austin (5) were determined by the dithizone method after complete oxidation of the organic material in the sample with a mixture of nitric and perchloric acids (cf. 19, p. 644) and were observed to be very low (not greater than ca. 100 parts per million) whereas preparations obtained b y other routes had considerably higher silver contents (of the order of 1000 p.p.m.). Sobotka and Austin filtered the metha~iolreaction mixture through a sodium sulphate bed prior to crystallization. This ma). have fortuitously acted as an anion-exchange resin and removed silver ions from solutiorl as the sulphate. I t is conceivable that both sulphate beds would help to hold back colloidal as well as ionic the resill and the sodiu~ll silver. One possible advantage of the resin over sodium sulphate would be t h a t it \vould enable the reaction t o be carried out in aqueous solutioil if required. The mother liquors were examined chromatographically on paper using water as solvent and there mas evidence that they coiltai~ledadre~lochromeRJ0.82, a little adrenolutin 1 J0.4-0.45, a non-fluorescent dark-violet unidentified compound R 0.92, and a n o ~ l 1 fluorescent colorless zone RJO.O-0.4. All these zones gave blue colors (i.e. reduced) with the ferric chloride- potassium ferricyanide reagent (see next section). The low R j area probably contained some unchanged adrenaline, since it gave a red color when sprayed with potassium ferricyanide or ammo~iiumpersulphate solution. f Paper Chromatogra$hy o Adrenochrome and Related Com$ounds Whatman No. 1 paper was washed with distilled water for 12 hours and dried prior to use. I t was observed that unwashed paper caused some decompositio~lof the adrenochrome. The ascending method for paper chromatography was invariably employed, and 10-20 pg. of sample were applied to the paper in each case. T h e results obtaiiled are given in Table I. In all cases, the solvent was allowed to asceild a distance of ca. 9 to 12 inches. There was a co~lsiderablevariability in the time required for this to occur; water and 2% acetic acid having the advantage of giving considerably quiclter results than the other solvent ts systems. T h e approximate times taken by the various s o l v e ~ ~are given in Table 11. Detection 0 S$ots 1 All four compounds are colored and therefore are self-indicating. Adrenolutin (111) also exhibits a marked yellow-green fluoresceilce in ultraviolet light. I t has been previously stated t h a t , on drying, the spots of I undergo a slow change to 111; this substance, being a catechol derivative, has reducing properties and the chromatogram call be developed in the following manner (if a permanent record is required). The developing solution was prepared immediately before use, and coilsisted of 3% ferric chloride solutio~z(5 ml.), 3% potassium ferricya~iidesolution (5 ml.), and water (90 rnl.). The papers were dipped in the reagent and the reducing cornpou~ldsformed perma~lentblue spots. The excess reagent was washed off; the papers were dipped in 2 N hydrochloric acid and finally excess acid was washed off. Other chron~ogenic reagents for adre~lochronle include: ( a ) Ehrlich's reagent (violet spot), (b) 3% ferric chloride solutioil (gray-brown spot), (6) diazotized p-nitroaniline (red-brown spot), and (d) zinc chloride solution (yellow-green fluorescent spot in ultraviolet light).

HBACOCK

BT AL.: AMINOCHROMES.

:2clrenochron1e (I) 0.8 f 0.02 0.82~k0.02 S. clec. (0.7-0.8) S. clec. (0.7-0.8) R. clec.7 I?. dec. i\drenolutin (III):!' 0.45f 0.05 0.47+0.0'& S. tlec. (0.7-0.8) S. tlec. (0.7-0.8) R . dec. R. dec. :\clrenosyl (VI)t 0 . 4 8 f 0.05 O.5Of 0.03 0 . 6 5 f 0.05 0.85+0.05! 0.57fO.05"~.58~0.05"' -4drenochrome ~lionoisonicotinic acid hyclrazide (VII)$ 0.30

NOTE.-.-l = IVater. B = 2% ace1.i~ocid i n zvater. C = d~ethano1:roater (3:i). D = Etltanol:zvater ( 3 : l ) . E = ?L-Butano1:c~cetic acid: water (8:2:2). F = i-Propanol:0.880 arnn$on.ia: water (8:1:1). S. dec. = slozv (lec0)1?positio71. dec. = rapid decorispositiu?~. R. " = T h e edreaolzlti?~ prepared b y t l ~ errletl~odof Harley-ilfusoit and Bu'Lock (20). noas t = Szrpplied b y the Labas Co. t = Szlpplied b y the Pfizer Co. 5 = I n a l l acidic solve?~tsadrenolz~tinezllibited a pink j'lz~orescei~ce. / / = Occasio?~ally iirexplicable lower Rl's were observed for adrenoxyl wit11 this solvent s y s t e ~ u . 7 = Differing results cuere obtai7sed depettde?tt U I L the t i ~ n eteketl to r u n t l ~ ecl~ro~nzlogratit. h e r c ~ lcolor T spot slozoly faded with the for~,tntio?tof a dirty yellozo diffzise spot exhibiting a pink$zroresce?~ce. of t l ~ c *C = T i ~ e s cresz~ltsare differe?tt from tltose reborted breviozlslv (12): zue arc zrnnble to exblain these discref~s pancies. TItc R, ualzles qtiot&l i n Fable I were n'btainci i,o?lsiste;ltlY ill O I L Y i~~vest~gatio?ts. c h e r ' s clairi~tlmt cldrenoxyl resolves i n t o a d r e n o c l ~ r o ~ ~ ~ ea flzloresci?tg spot R I 0.88 zn the c ~ ~ ~ z ~ i z o n i solve?st systent F (12) and acal zuoz~ldappear to be due to a ~)zisi,tterpretatiolt of tile evidence i n uiezo of k7tozon instability of adre?toclrrorne i n nm7nonia.
'I'rIBLE I1 Solvent system .\pproximate running tillle (hours)
.\

2.5-3.0

B 2.5-3.0

C 7.5

D 15.5-18.0

E 15.5-18.0

F 18.0-20.0

ACI<NOWLEDG&IENT

The a u t h o r s w i s h to express t h e i r t h a n k s to Dr. B. M e l a n d e r , of the Iiabi Co. of Stockholm, S \ v e d e n , f o r a s s i s t a n c e i n o b t a i n i n g the m i c r o s i l v e r a n a l y s i s .


REFERENCES 1. GREEK, E. ancl RICHTER, D. D. Biochem. J. 31, 506 (1937). 2. VEER,W. L. C. Rec. trav. chim. 61, 638 (1942). 3. MACCARTHY, L. Chim. & ind. (Paris), 55, 435 (1946). C. 4. HARLEY-MASON, J. Chem. Soc. 1276 (1950). J. Ti. Son0~1c.4,H. and !\USTIN,J . J,. Am. Chcm. Soc. 73, 3077 (1951). G. HARLEY-MASON, Esperientla, 4, 307 (1948). J. 7. ZAMDOTTI, and MORET, V. V. Arch. sci. biol. (Bologna), 33, 522 (1940). 8. L r r ~ nA . Acta Pharmacol. 'Toxicol. 5, 121 (1949). , 0. FISCMER, DEROUAUS, LAMBOT, and LECOMTE, Bull. soc. chim. Beiges, 59, 72 (1950). P., G., H., J. 10. HARLEY-RIIASON, J. and Bu'Loctc, J. Nature, 166, 1036 (1950). 11. FISCAER, and LECOMTE, Bull. soc. chirn. biol. 33, 569 (1951). P. J. R. Nat~lrwiss. 443 (1957). 12. FISCHER, 46, 13. PAVOLIXI, GAMBARIN, and GODENIGO, S . Gazz. chim. ital. 81, 527 (1051). 'r., F., A. 14. ETTLER, S. VON. Norad[-e~~ali~ic-cllcniistry, U. physiology, pharmacology and clinical aspects. 1iyer:ou Press, 'Toronto, Ont. 195G. 15. I\USTIN, CHANLEY, D., ancl S O ~ ~ O T K A , J. Am. Chem. Soc. 73, 2395 (1051). J., J. H. 16. ; i u s ~ ~Js ,, CHAWLEY, D., ancl S O ~ ~ O T IH.A , J . i\m. C h e ~ n Soc. 73, 5299 (1951). . J. < . %. 17. BACQ, &I. J. Pharmacol. Esptl. 'Therap. 95 (ii), 1 (1949). 18. HOFFER, 11. J . Clin. Esptl. Psychopathol. & Quart. Rev. Psychiat. Neurol. 18 (No. I ) , 38 (1956). 19. SANDEI.I., B. Colori1iietric determination of traces of metals. Interscience Publishers, Inc., New E. Yorl;. 1950. 20. HARLEY-~IASON, Bu'LocI<, J. J. Chem. Soc. 703 (1951). J. and