Chem.-Biol. Interactions, 63 (1987) 39--45 39 Elsevier Scientific Publishers Ireland Ltd.

CHARACTERIZATION OF TANNIC ACID METABOLITES FORMED IN VITRO BY RAT LIVER MICROSOMES AND ASSAY OF THEIR CARCINOGENICITY BY THE MICROSOMAL DEGRANULATION TECHNIQUE

MADAN M. GUPTA and HARINDER M. DANI*

Department of Biochemistry, Panjab University, Chandigarh-160014 (India) {Received February 2nd, 1987) (Revision received February 18th, 1987) {Accepted April 6th, 1987)

SUMMARY

Tannic acid is converted to four metabolites on incubation with isolated rat liver microsomes. Bis-(3,4,5-trihydroxyphenyl)methanone (I); 3,4,6,7,9,10- hexahydroxy-2,11-epoxy- 1,12-(epoxy-methano)- 14H, 16H- benzo[b]{1,4)benzodioxepino [3,2-g]{1,4)benzodioxepin-14,16,17-trione {If) and 1,1,2-trimethyl-l~thanyl-2-ylidene tris{3,4,5-trihydroxybenzoate) (Ill) were isolated from the post-microsomal supernatant of the incubate while 6,12,18,26,27,29,31,33-octahydroxy-22,22-dimethyl-2,8,14,20-tetraoxapenta- cyclo (22.2.2.24.7.210.1s.2x6.19) tetratriaconta-4,6,10,12,16,18,24,26,27,29,31,33- dodecene-3,9,15,21,23-pentone (IV) was found attached with the microsomal fraction. Metabolite (III) was found to be a potential carcinogen on the basis of microsomal degranulation technique.

Key words: Tannic acid -- Carcinogenicity -- Microsomes -- Metabolites

INTRODUCTION

In addition to its natural occurrence, tannic acid is of wide commercial use [1]. I t has already been reported to be a hepatocarcinogen [2]. I t disag- gregates polyribosomes [3] and degranulates microsomal membranes in vivo [4] and in vitro [5]. Many carcinogens have been reported to degranulate microsomes in vitro and the technique has been used for the short-term prediction of carcinogenicity of various chemicals [6--8]. Microsomal mixed function oxidases are known to convert many chemicals into electrophiles

*To whom all correspondence should be sent.

0009-2797/87/$03.50 © 1987 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

40

which might act as carcinogens. Activation of tannic acid to potential car- cinogen(s) and its mode of action in the detachment of ribosomes from reti- cular membranes have not been studied so far. We have identified four moieties in the incubate of tannic acid with rat fiver microsomes. Three of these metabofites remain in the supernatant and one is bound to microso- mal membranes. Screening of all the four metabolites for their carcinogenic- ity by the microsomal degranulation technique showed that fraction III was a potent carcinogen while fraction II might have milder carcinogenicity.

M A T E R I A L S AND M E T H O D S

Tannic acid IC76H5204~) w a s incubated with rat fiver microsomes (500/~g/ ml suspension) prepared as described earlier [5] in the presence of 1 mM NADPH at 20°C for 2 h. After the activation, the supernatant and the microsomes were separated at 11 000 × g [5]. Metabolic products of tannic acid in the supernatant (I, II and III) and in the microsomes dissolved in SDS {IV) were separated by thin-layer chromatography [9] and their R~ val- ues recorded as given in Table I. These products were then isolated by preparative chromatography using silica gel G [9] for elemental and spectral analyses. All these fractions were eluted in acetone, passed through sintered glass crucible and dried under vacuum before analyses.

IR spectra were recorded on Perkin-Elmer model 137 spectrophotometer using KBr pellets. UV spectra were measured using Perkin Elmer model 200 spectrophotometer. PMR measurements were made on a PMR Varian 390 (90 MHz) spectrometer in deuterated chloroform using tetramethyl- sflane (TMS) as an internal standard. Chemical shift values have been reported in &scales. Mass spectra were taken on Vg micromass 7070 spec- trometer.

Carcinogenicity of all the four metabolites was studied by the microsomal degranulation technique [5].

TABLE I

Rf VALUES OF M E T A B O L I T E S OF TANNIC ACID A F T E R INCUBATION WITH RAT LIVER MICROSOMES

Metabolites a Rf value

I 0.47 II 0.33 I l l 0.23 IV 0.17

aFor s t ructures and nomenclature consult Fig. 1 and Introduction, respectively.

41

RESULTS

The structure of bis43,4,5-trihydroxyphenyl)methanone (I) in Fig. 1 was characterised by elemental analysis (Calculated for ClsH,001: C, 56.13; H, 3.59; found C, 56.00; H, 3.67) and spectral data: IR (KBr) 3300, 2800,1735 and 1100 cm -* UV :MeOH nm (log E) 221; PMR; (d, CDCI~) 1.25 (6 H, s, OH

^ m a x

HO OH

OH

(I)

0

HO

(TT) . 0

H O - - - ~ C ~ 0 ~

HO

OH CH 3 CH) 0 /

c--c--o--c--i[ )2-- OH

CH s 0 ~ C ---0 OH

H O ~ O H

OH

(IK)

~ C ~ C ~ O - - - ( ( b ~ - C - r ' - v ' ~ t l y ' ~ , ' - r - ~ J/

(Tv) Fig. 1. Structures of metabolites of tannic acid formed on in vitro incubation with rat liver microsomes. Bis(3,4,5-trihydroxyphenyl)methanone (I); 3,4,6,7,9,10-hexahydroxy-2,11-epoxy- 1,12.~ep~xymethan~b14//,16H-benz~{b]~1,4~benz~di~xepin~[3~2.g](1,4~benz~di~xepin-14,16,17- trione (II) and l,l,~-trimethyl-l-ethanyl-2-ylidane tris(3,4,5-trihydroxybenzoate) (Ill) were isolated from the post-microsoma] supernatant of the incubate while 6,12,18,26,27,29,31,33- octahydroxy-22,22-d/methyl-2,8,14,20-tetraoxapentacyclo (22.2. 2-2.4'7.2]°'13.21sJ 9)tetratriac°nta" 4,6,10,12,16,18,24,26,27,29,31,33-dodecene-3,9,15,21,23 -pentone (IV) was found attached with the microsomal fraction.

42

disappeared in D~O, 7.35 (4H, s aromatic protons); mass spectrum showed molecular ion peak m/e 278 and other prominent peaks as m/e 153, 125.

Similarly elemental analysis (calculated for C2]H60~s: C, 50.60 and H, 1.20; found C, 50.75 and H, 1.28) and spectral analysis (IR (KBr) q 3350, 1710, 1220 and 1120 cm-~; UV ~MeOH nm (log e) 223; PMR (d, CDC1 s) 2.55 mAY (6 H, s, - O H disappeared in D~O; base peak was obtained at m/e 498 and other prominent peaks were at m/e 450, 267) suggested the structure of (II) as 3,4,6, 7,9,10-hexahydroxy-2,11-epoxy- 1,12-(epoxymethano)- 14/-/, 16H-ben- zo[b](1,4)benzodioxepino[3,2-g](1,4)benzodioxepin -14,16,17-trione (Fig. 1).

Elemental analysis of metabolite (III) calculated for C2~H2~O~5: C, 54.16; H, 4.17; found C, 54.29; H, 4.07) and its spectral analysis (IR (KBr) ~ 3350, 2850, 1735, 1100 cm-~; UV AMmaxeOHnm (log ~) 220 nm; PMR (d, CDCl~) 1.35 (6H, s, aliphatic protons), 1.7 (3H, br s, aliphatic protons), 2.7 (9H, s, -OH) disappeared in D20 and 7.4 (6H, s, aromatic protons), base molecuar ion peak mYe 576 and other peaks m/e 407, 238, 169) established the structure of 1,1,2-trimethyl-l-ethanyl-2-ylidene tris(3,4,5-trihydroxybenzoate) as pre- sented in Fig. 1.

The structure of 6,12,18,26,27,29,31,33-octahydroxy-22,22-dimethyl: 2 ,8 ,14 ,20- te t raoxapentacyclo (22.2.2.247.21°'13.216'19) t e t ra t r i acona ta - 4,6,10,12,16,18,24,26,27,29,31,33-dodecene-3,9,15,21,23 pentone (IV) as depicted in Fig. 1 was revealed by elemental (calculated for C32H22017" C, 56.64; H, 3.24; found C, 56.32; H, 3. 30) and spectral data (IR (KBr) q 3400, 2880, 1735, 1450, 1120 cm-~; UV ~maxlMeOH nm (log ~) 218; PMR (d, CDC1 s) 1.35 (6H, s, aliphatic protons), 2.7 (4H, s, OH) disappeared in D20; 3.1 (4H, s, - O H ) disappeared in D20, 7.4 (8H, s, aromatic protons); molecular base ion peak at m/e 678 and other peaks were at m/e 650, 610, 580 and 335).

Data for determining the carcinogenicity of the 4 metabolites of tannic acid employing the microsomal degranulation technique [5] are presented in Table II. Metabolite I I I was found to be a potent carcinogen as it degranu- lated the microsomes substantially [6] both on the bases of RNA/protein and RNA/phospholipid ratios in the absence as well as presence of NADPH. Metabolite II was found to have milder carcinogenicity as it degranulated the microsomes slightly above 5% level, which is an arbitrary minimum for deciding the carcinogenicity of a compound [6]. These low values of degranu- lation were obtained only on the basis of RNA/protein ratio when degranula- tion experiment was performed in the absence of NADPH and on RNA/ phospholipid basis in the presence of NADPH. Metabolites I and II failed to cause detachment of ribosomes more than 5% level and therefore seem to lack any carcinogenic potentials.

D I S C U S S I O N

Tannic acid has already been reported to be a carcinogen on the basis of its potentiality to detach ribosomes from rough microsomes [5]. Attempts have been made in the present study to identify the possible metabolites of tannic acid on incubation with rat liver microsomes and study their poten-

TA

BL

E I

I

EF

FE

CT

S O

F T

AN

NIC

AC

ID M

ET

AB

OL

ITE

S O

N R

IBO

SO

MA

L D

ET

AC

HM

EN

T

IN R

AT

LIV

ER

MIC

RO

SO

ME

S (

n --

3)

Com

posi

tion

of

Wit

ho

ut

NA

DP

H

Wit

h N

AD

PH

in

cuba

tes

RN

AJp

rote

in

RN

A/p

hosp

holi

pid

RN

A]p

rote

in

RN

A/p

hosp

holi

pids

±

S.D

. ±

S.D

. ±

S.D

. ±

S.D

.

(1)

Mic

roso

mes

(2

) M

icre

som

es +

m

etab

olit

e I a

(3)

Mic

roso

mes

+

met

abol

ite

II a

(4)

Mic

roso

mes

÷

met

abol

ite

Ili

a

(5)

Mic

roso

mes

+

met

abol

ite

IV'

0.24

1 +

0.00

55

0.86

7 +

0.01

35

0.21

3 -t-

0.0

189

0.83

5 ±

0.02

82

0.22

3 ±

0.01

86

0.82

8 ±

0.04

94

0.22

6 ±

0.01

13

0.90

0 ±

0.00

40

(3.3

1) b

(4.

49)

(0.0

) (0

.0)

0.22

7 ±

0.02

32

0.87

2 ±

0.03

9 0.

240

± 0.

009

0.82

3 ±

0.01

04

(5.8

0)

(0.0

) (0

.0)

(5.2

6)

0.14

2 ±

0.01

19

0.50

1 ±

0.04

1 0.

139

± 0.

0065

0.

518

± 0.

0145

(4

1.07

) (4

2.21

) (3

4.74

) (3

7.96

) 0.

241

± 0.

0094

0.

868

± 0.

0075

0.

207

± 0.

0062

0.

831

± 0.

024

(0.0

) (0

.0)

(2.8

1)

(0.4

8)

a F

or s

truc

ture

s an

d n

omen

clat

ure

cons

ult

Fig

. 1

and

Intr

oduc

tion

, re

spec

tive

ly.

b V

ltlu

es i

n pa

rent

hese

s in

dica

te %

det

ach

men

t of

rib

osom

es f

rom

mic

roso

mes

.

44

tia]s to cause microsomal degranulation for the first time. The predictive efficiency of the microsomal degranulation technique for detecting the car- cinogenicity of various chemical compounds has been recently reported to be around 80% [8,11]. Results presented in Table II indicate that the carcinogenicity of tannic acid might be mainly due to metabolite III {1,1,2- trimethyl-l-ethanyl-2-ylidene tris(3,4,5-trihydroxybenzoate), as both of them detach around 40% of r ibosomes from the microsomes [5].

The carcinogenicity of most of the chemical compounds is based on their electrophilic interactions with the cellular nucleophiles like proteins and nucleic acids [10]. Metaboli te I I I could behave as a s trong electrophile as carbon 1 could act as a ter t iary carbonium ion in a basic medium or due to the action of an esterase. Similarly carbon 2 could act a carbonium ion due to the formation of an unstable diol which might ult imately lose a water molecule and the resulting ketonic group could acquire a partial positive charge due to electrometric effect. These two carbonium ions will enable this metabolite to act as a s t rong electrophile which could interact with nucleo- philic r ibosomes and detach them from the reticular membranes. I t will be interesting to find whether metaboli tes I, II and III are at tached to the stripped ribosomes or exist freely in the supernatant. Our results do not indicate the a t tachment of these metaboli tes to the detached ribosomes as preparative chromatography without any solubilisation of ribosomes has enabled us to separate these fractions.

ACKNOWLEDGEMENTS

We are thankful to Dr. J o y E. Merrit, Senior Associate Editor of Chemi- ca] Abst rac t Service at Ohio Sta te University, Columbus for providing us the nomenclature of tannic acid metaboli tes and to CSIR, New Delhi, for financial assistance to MMG.

REFERENCES

1 E. Ward led.), Tannic acid and Tannins, IARC Sci. Publ. No. 10 Lyon, IARC 253, 1976. 2 J.K. Reddy and D.J. Savoboda, The relationship of nuclear segregation to ribonucleic acid

synthesis following the administration of selected hepatocarcinogens, Lab. Invest., 19 {1968) 132.

3 J.K. Reddy, M. Chiga, C.C. Harris and D.J. Savoboda, Polyribosome disaggregation in rat liver following adminiAtration of tannic acid, Cancer Res., 30 (1970) 58.

4 W.M. Butler, Early hepatic parenchymal changes induced in the rat by afiatoxin B 1, Am. J. PathoL, 49 (1966) 113.

5 M.M. Gupta and H.M. Dani, A simplified short term test for detection of carcinogens: Microsomal degranulation by tannic acid, Ind. J. Exp. Biol., 17 {1979) 1144.

6 I.F.H. Purchase, E. Longetaff, J. Ashby, J.A. Styles, D. Anderson, P.A. Lefevre and F.R. Westeood, An evaluation of 6 short-term tests for detecting chemical carcinogens, Br. J. Cancer, 37 {1978} 873.

7 E.G. Fey, H.A. White and B.R. Rabin, Development of the degranulation test system, in: F.J. de Sarres and J. Ashby (Eds.), Evaluation of Short-term Tests for Carcinogens, Elsev- ier/North-Holland, New York, 1981, pp. 236--244.

45

8 M.M. Gupta and H.M. Dani, Efficient prediction of chemical carcinogenicity by microso- real degranulation, ToxicoL Lett., 30 (1986) 167.

9 E. Stahl and P.J. Schorn, Hydrophilic constituents of plants, in: E. Stahl (Ed.), Thin Layer Chromatography, A Laboratory Handbook, Academic Press, New York, 1965, pp. 371--391.

10 E.C. Miller and J.A. Miller, Mechanism of carcinogenesis, Cancer, 47 (1981) 1055. 11 S.K. Jagota and H.M. Dani, Improved detection of carcinogens by degranulation of micr~

somes prepared at low 'g' force by glutathione, Aust. J. Exp. Biol. Med. Sci., 63 (1985) 683.