The detection of some indoles and related compounds on paper chromatograms

JOURNAL OP CHROMATOGRAPHY

THE

DETECTION

Food

Rc.wurclL

Plaral

Ji!essaYcll.

OF

SOME

ON

PAPER

INDOLES

AND

=73

RELATED

COMPOUNDS

CHROMA.TOGRAMS*

The condensation of pyrroles and indoles with aldehycles is a well known reaction which was studied extensively at the turn of the century and which aided HOPKINS AXD COLBY in the discovery of tryptophan. Later on, the indole-aldehyde reaction was studied by UURR AND GORTNER~ in connection with humin formation in protein hydrolysates. Since then, the reaction has been widely used to detect indoles, ureides and aromatic amines on paper chromatograms 8~4.The detection of simple indoles has depended largely on the Ehrlich color reaction, using p-dimetl~yl&ninobenzaldethat p-dimethylaminoliyde+(J. Recently HAWZY-MASON AND ARCHER’ reported cinnsnialdehyde (DMCA) was ten times more sensitive for indole and tryptophan than p-climetl~ylaminobenzaldel~yde (Ehrlich reagent). The great interest in the biochemistry of naturally occurring indole derivatives in both plant and animal tissues, especially in connection with indole-s-acetic acid and serotonin, warrants further study on the methods of isolation and identification of these substances. The present report deals with the chromatographic behavior of twenty-seven indoles and eighteen related compounds. Ultra-violet fluorescence before spraying and the color reactions obtained after spraying with the two aldehyde reagents are reported and the relationship of structure to color reaction is discussed briefly. lk!&?cvinls

MATERIALS AND MZTHODS

Indole, skatole, gramine, indole-J-acetic acid, indole-3-propionic acid, indole-3-butyric acid, tryptoplian, tryptamine, the aromatic amines, urea, thiourea, citrulline and quinaldic acid were obtained through the usual commercial sources. The 5-benzylosyindoles, I,z-climethylindole, z,3-dimethylindole, 7-azaindole, indole-3-aldehyde and indole-3-acetonitrile were obtained from Aldrich Chemical Co., Milwaukee, Wise. 5-Hydrosytryptophan, 5-hydroxyindole-s-acetic acid (cyclohekylamine salt), kynurenine, lcynurenic acid and santhurenic acid were obtained from the California Foundation for Biochemical Research, Los Angeles, Calif, Serotonin or s-hydrosytryptamine (creatinine sulfate comples) was kindly supplied by Vismara Terapeutici. Contribution Institute, Canah l

No. IO Food Rcscarch Institute ancl Contribution Department of .-\gricultlwe, Ottawa.

No.

338 Plant

J. Chvov~,ralog., 13 (1904)

Rcscsrch

173-180

I74

A. B. DURKEE, J. C. SIROIS

The phenols were kindly supplied to us by Dr. H. MORIT~\ of the Soils liesearch Institute, Canada Agriculture, Ottawa. Indole-3-glyosylic acid, indole-3-glyosyl chloride, indole-3-glyosylamide and methyl indole-3-glyosylate were synthesized from indole by the methods of Swnw et aZ.8.S-Cyanoindole, indole-3-aldosime, indole-3-carbosylic acid and indole-s-acrylic acid were synthesized from indole-3-aldehyde by the methods of SHAW et al. Indo@-carbosylic acid, prepared by the carbonation of the Grignard reagent formed with ethyl magnesium bromide and indole and recrystallized from aqueous ethanol was not chromatographically pure. On the other hand this acid prepared by hyclrolysis of crude 3-cyanoindole by SN.AW’S method did not require further recrystallization and gave only one spot in both isopropanol-ammonia-water (S : I : I) and butanol-acetic acicl-water (4: I : I). It decomposed at 246” (uncorr.). The procluct from the Grignard reaction decomposed at 220~--225~. S-Hydrosyquinaldic acicl was prepared by a four-step synthesis from o-anisicline ancl crotonaldehycle in good yielcl by the method of IRVING AND PINKIXGTON~~),~~. Most of the substances from commercial sources were chromatographically pure and gave sharp melting points. 2,3-Dimethylindole, gramine and 5-benzylo.sygra.mine yielded more than one spot which may have been due to decomposition during chromatography. Catechol and 3-methosycatechol were unstable giving spots and streaks on the paper. The latter compouncl became discolored even when stored in the refrigerator in 95 o/oethanol. Methods The chromatography was carried out on Whatman No. I paper using the ascending method in butanol-acetic acid-water (4: I : I) and isopropanol-ammonia-water (S : I : I). The spray reagents usecl were: (a) EkGch aZdehyde reagent, prepared by clissolving 2 g $-dimethylaminobenzaldehyde in a misture of So ml of 95 o/oethanol and 20 ml of G 1V E-ICll”, (b) ih_Di712etlzylanzi~locila?znnaa:Idelt_yde (DMCA) reagent, preparecl according to HARLEY-MASON AND ARCWXII; by dissolving 2 g of this alclehyde in a misture of IOO ml of 95 oh ethanol and IOO ml of G N HCl. The chromatograms were equilibrated with the solvent for 2 h and clevelopecl overnight (IG 11). After air-drying, they were sprayecl with one or the other reagent, dried in front of a fan and then placed in an oven set at 70” for 5 min. After this treatment they were allowed to remain at room temperature for 4s h. Papers sprayecl with DMCA turned very dark after 24 11,but the strong spots remained outlinecl on the paper. Papers sprayecl with the Ehrlich reagent had very little background color. All papers were esamined under ultra-violet light using a Blak-ray, long wave U.V. lamp before and after spraying with color reagent. For chromatography, the substances were dissolved in a suitable solvent, generally 95 Y0 ethanol or acetone ancl macle up quantitatively to a concentration of I mg per ml, and IO ~1 were spotted at the starting line with a micro-pipette. Because some of the substances did not react at this level some chromatograms were run using 20-30 /-11 per spot. ’ RESULTS AND DISCUSSION The color reactions, Rp values, U.V. data, etc. for the indole derivatives and relatecl metabolites are given in Tables I, II and III. Equations and formulae referred to throughout the test are found in Figs. I and 2.

P Y Y Y

o.GS -

04j -

-

OS1

-

O-59

o.Gg

0.70

0.p

0.36 0.7s 0.1s O.Ij 0.40

Pi

0.92 -

0.90 -

I’ P(R) P I’

B -

O.S_i o.SG

0.90

R-P P-R I? I’D G B-P

Pi-R P R

0.52 0.90

o.jG o-43 o-35

0.90 OSG

o.s3 OS7 OSG

0.50

0.90

o-93

-

0.92

o.s3 o-45

osg -

0.92 0.92

-

I

-

No reaction I I -

3

ss

SS ss

AI

11 F Jr 3c

SS

31 ss

-

Pi

-

-

B

P P B I3

P-R

B -

F

-

No reaction

-

S

-

F F F F

F

F ss

P-B -

R-P

s”

No reaction X0 reaction 3

i-5

3 4 4-S 4-5

3

3

F R (fades) F G 4-3 R F P-B F P 4-3 R-P F I F Pi G Unstable, breakdown products in chromatography F B F G P 4-3 31 F B B P 4 F B 31-F B P 4 F B B F B 4 11-F F B B B 3 B-P S No reaction

2 B I it B No immed. reaction I Br No inuned. reaction 2 Y (24 11) P-R 4 1-3 T3

3 3 3 4 3

3 -

3 3 3

DETECTION OFINDOLES

TABLE 1

Y Auorcscence

B after spray

.:.c .I

l

j = very intense; I = very faint. Rate of color development: F = fast; 11 = medium; S = slow-. l ** All the glyosylic derivatives require at least 30 /(g per spot to give a faint color with either reagent. The methyl ester and acid chloride are transformed to-indole-3-glyosylamide during chromatography in isopropanol-ammonia-water (S: I : I).

l Ip-NE& = isopropanol-ammonia-water (S: I : I) ; Bu-AC acid = butanol-acetic acid-water (4: I : I). * B = blue; Br = brown; G = green; 0 = orange; P = purple; Pi = pink; R = red; Y = yellow; it = light. Color intensity:

2,3-Dimeth~lindolc 3-Benzylosyindole 7-Azaindole Indole-3-acetic acid Indole-3-propionic acid Indole-3-butyric acid Indolc-3-acrylic acid Indole-3-carbosylic acid j-Benzylosyindole-3-acetic acid Indole-3-aldehyde Indole-3-aldosime Indole-3-acetonitrile 3-Cyanoindole Tryptophan Trrptamine 5-Hydrosgtryptophan j-Hydros\rtryptamine 3 Hydrosyindolc-j-acetic acid &amine j-Beneylosygramine Indole-3-gl yos$ic acid l l l Indole-3-gly-osylchloride’ l l IndoIe-3-glyosyIamide”* Methyl indole-3-glyosylate***

0.91 0.92

I ndole

;

SkatOk I,+Din~etl~vliadolc

T

0.14 -

0.2s

0.67 -

Kynurenine

S-Hydrosyquinaldic

0.10

Xanthurcnic

O-45 o..$

0.47

Y

No reaction I\0 reaction

X0 reactron A’0 reaction

P

-

R-P

R-P

No reaction

2

2

3 I

5

4

No reaction

Gr

R

31 F

P

R

P

R-Br

AND QUINOLINES

s

>I

F

F

UREIDES

Yellow color seen before spraying

-

-

I’ -

II

AMINES,

F

XI

S

F

F

F

B fluorescence

B fluorescence

No fluorescence

R fluorescence

cence

Strong fluores-

No fluorescence

B fluorescence

* Ip-NH, = isopropanol-ammonia-water (S: I : I) ; Bu-_k acid = butanol-acetic acid-water (4: I : I). B = blue; Br = brown; Gr = gray; P = purple; R = red; Y = yellow. Color intensity: 5 = very intcnsc; I = very faint. Rate of color deveiopmcnt: 1; = fast; 11 = medium; S = slow.

acid

O-45

Kynurenic acid

Quinaldic acid

0.30

0.10

Citrullinc

acid

2

II

0.16

0.50

Thiourea 4

I

Y

0.45

Urea Q-u

5 3

p-Aminobenzoic acid Y

Y

o.so

0.20

5

Y

O.Sj

0.45

OF AROMATIC

Anthranilic Acid

DETECTION

TABLE II

l

acid

o.jS 0.7s

l

Q-NH, = isopropanol-ammonia-water * P = purple; R = red.

?,Jr-Dihydrosybenzoic

acid

2,j-Dihydrosyphenylicetic

Syringic acid Phloroglucinol 34ethoxycatechol No reaction

Decomposed in refrigerator

MOreaction

No reaction

Decomposition

No reaction

_-_-

OF PHENOLS

Color uffh EhrlicP

DETECTIOH

P

-

-

A’0reaction

- Decomposed in refrigerator

No reaction No reaction

No reaction

X0 reaction

Color z&h DMCA’

(S: I : I) ; Bu-AC acid = butanol-acetic acid-water (4.: I : I).

-

-

-

-

-

GalIic acid

o-77

-

-

Catechol

R-P

-

-

0.60

-

-

Rp’

Eugenol

--

TABLE III

Strong blue fluorescence

17s

A.B.

DURKEE,

J. C. SIROIS

In general the new aldehyde reagent, DMCA, reacted faster and gave more intense colors with indole compounds than the Ehrlich reagent. Indole-3-acetic acid could be detected on paper chromatograms in concentrations as low as 0.5 l_cgper spot. The reaction of DMCA with aromatic amines and ureides was disappointing .Red and purple colors were given which would make difficult their distinction from indole derivatives (Table II). One exception to the non-selective action of DMCA is the color reactions given by indole and skatole (Table I). Since these compounds have approximately the same Rp value in both solvents, they can be identified by the fact that indole gives a distinct green color with DMCA whereas skatole gives a purple spot. Indole-3-acrylic acid may be easily detected, if both sprays are used. This substance gives a green spot with the Ehrlich reagent and a blue spot with DMCA. Indole-3acetonitrile reacts immediately and strongly with DMCA, but the reaction with the Ehrlich reagent is very slow and rather weak. Indole-3-carbosylic acid does not react with DMCA but gives a clear blue-purple spot with Ehrlich reagent after drying for about 10-15 min. The chromatographic behavior of the indole-3-glyosylic acid derivatives is interesting. The acid itself does not react with either aldehyde but does fluoresce under ultra-violet light (Table I). Under the conditions of chromatography in isopropanol-ammonia-water (S : I :I) the methyl ester and acid chloricle are transformecl to the amide which gives a faint pink spot with DMCA ancl a faint yellow spot with the Ehrlich reagent, Of the phenols testecl, only phloroglucinol reacts with the aldehyde reagents and gives a red-purple color with the Ehrlich reagent ancl a purple color with DMCA. The quinolines dicl not react but fluoresced under ultra-violet (Table III). The most significant observation in this particular stucly is that many 3-substituted incloles either do not react with these aldehydes or react so poorly that the color reactions cannot be used with conficlence in their identification, 3-Substituted incloles in which a carbonyl or nitrile group is attached clirectly to the P-carbon atom of the pyrrole ring are poor reactors. If the carbonyl or nitrile group is separated by a methylene group the reaction is greatly improvecl. For example, 3-cyanoindole cloes not give a color with DMCA or Ehrlich reagent but indole-3-acetonitrile reacts rapidly with DMCA and slowly with Ehrlich reagent. Indole-3-glyosylic acid (two carbonyl groups) does not react but the reaction occurs to some extent when -NI-I, is substituted for -OH (indole-3-glyosylamicle). Indole-3-aldehyde and inclole-3-carbosylic acid react faintly with both aldehydes but indole-3-acetic, inclole-3-propionic and indole-3-butyric acids react strongly. Worthy of mention is the fact that in this homologous series the longer the chain the stronger and more rapicl is the reaction. On the basis of the above observations, these differences in reactivity can be esplainecl by reference to the structural and electronic formulae of these indoles (Fig. I). If the condensation takes place at the a-position the electron-attracting CN and C=O groups tend to decrease the electron density around the z-carbon atom, making that position positive and repelling the attack of the electrophilic aldehyde. When these electrophilic groups are separated by methylene groups there is no longer a conjugated system and therefore no electromeric effect. The presence of tertiary nitrogen in the gramines and 7-azainclole may esplain their failure to react, because under the acid conditions of the test, the nitrogen becomes quaternary due to salt formation and becomes an electron-attracting group as well. J. Cllvorllnlog.,

I3

(1964)

173-180

DETECTION

OF INDOLES

ANDRELATED

I79

COMPOUNDS

(cl

!

Substituents on the benzene ring in the s-position had very little effect on the color reaction escept perhaps to enhance it. Tndole and 5-benzylosyindole gave different colors with both aldehydes than the s-substituted derivatives, The former gave a green color with DMCA and red-purple with Ehrlich reagent and the 3-substituted derivatives were mostly purple with Ehrlich reagent and blue with DMCA. FEIC;L"" suggests that indole and pyrrole react with the aldehyde in the ratio of I mole for I mole to form a colored comples of the indolidene-methane structure (33) (Fig. z) and BURR AND GORTNER~ have isolated colored salts of several z-substituted indoles with this structure.

_._ ,Axp-ICOOl-I ..n__

MOOCCl-m-I,\

-__.B,AJ

_. .__.

;-I CI-I,-N--C&I, (A)

. (13)

Purple color

Red color

Fig. 2. GHIGI~" studied

the reaction of p-dimethylaminobenzaldel~yde with tryptophan and assumed that this reaction took place in the ratio of 2 moles of the indole for I mole of aldehyde to form a colored comples of the type (A) (Fig. 2). It is probable that indoles with no substituents in the 3-position react I mole for I mole according to FEIGL and to Bunn AND GORTNER whereas the 3-substituted compounds react in the manner described by GHIGI and that the condensation takes place on the pyrrole ring at position 2. The difference in color mentioned above and the clecreased reactivity of s-substituted indoles with carbonyl and nitrile groups attachecl directly to the g-carbon atom lend support to the hypothesis that condensation takes place at the z-position and that the colored compounds are different. However, the reaction mechanism requires further study before the nature of the colored products can be reaclily esplained. J. Clwomalog., 13 (1964) 173-1~0

A.

180

I3.

DURKEE,

J,

C.

SIROIS

SUhLMARY 1. P-Dimethylaminocinnamalclel~yde (DMCA) is a more sensitive reagent, in most mstances, than the Ehrlich reagent but is less selective in its color reaction and would be rather a poor chromatographic spray when both indoles and aromatic amines are present. 2. DMCA should be used if skatole or indole is assumed to be present on the chromatogranl and in conjunction with the Ehrlich reagent if indole-3-acetonitrile and indole-3-acrylic acid are thought to be present. 3. The decreased reactivity of substituted indoles with carbonyl or nitrile groups attached directly to the 3-carbon atom suggest that the condensation takes place at the z-position. 4, The differences in color of indole and benzylosyindole from the 3-substituted derivatives, indole-3-acetic acid and tryptophan, etc., suggest that their condensation products are of different structure.

REl?ERENCES 1 F. 13.I-I~PIcINsANI) S. W. COLE,J. 2 G. 0. BJRIZ AND R. A. GoRTNER,J. 3 R. J. BLACK, E. L. DURRU~I AND d G a 7 8 0 10 11 12

Pl&ysioZ,(Londo~~z), 27 (Igo2) 418. Am. Chsm Sot.. 46 (1924) 1224. G. ZWXIG, A MccwnZ of Pa&w Cltro~natogva~hy am! Paptw ~h?ct~~op?~oresz’s, 2nc1 Ed,, ikndemic Press, New York, 1958, p. 313. W. I-Icincmann Medical Books Ltd., London, 1955 IVOR ShIITI-I, C~~1~ollzalo~Ya~lc~i~ TecJwiqaces, 1% 114. A. J. VLITOS AND w. MEUDT, cow11'ib.Boyce TJton$mn hzs/., 17 (1952) 197. I<. I<. REDDI AND E. KODJCEIC, R~iocJrem. J.. 53 (1953) 2%. J. HARLEY-MASON AND A. A. P. G. ARCHER, B~iocJmw.J., 69 (195s) 60 pp. I<. N, I?. SNAW, A. MCMILLAN, A. G. GUDMUNDSON AND M. D. ARMSTRONG, J. Oq. C/rem., 23 (19.5s)II71. F. P. DOYLE, W. ITERRIER, D. 0. HOLLAND, M. D. MEHTA AND J'.I-I. C. NA~LOR, J. CJwm..Soc., (19.56) 2553. A. B. DURKEE, J. /f&'r*. I?ood CJurn., 6 (195s) 194. I-I. IRVING AND A. IL PINNINGTON. J. CJm?t. SOL, (195~1)37Ya. M. D. AR&ISTRONG, Ii. N. I?.SIIAW, RI. J. GORTATOWSKI AND H. SINGER. J. Biol. CJwm.., 232 (JQ5s)

17.

13 I?. FmGL, S$ot Tests 2’92 Oqptzic 14 E, Gnrcr, Gazr. CJtinz. Ilul., 64

.4mzZ~~sis, 6th (1933) 411.

Ed.,

Elscvicr,

Amsterdam,

IgGo,

J. Clwonmlog.,

p. aSo.

13 (1964) r73-1840