Br. J. Pharmac. (1969), 37, 748-755.

Anti-inflammatory activity of the steroid

alkaloid glycoside, tomatine
R. B. FILDERMAN AND B. A. KOVACS

The Harry Webster Thorp Laboratories, Division of Immunochemistry and Allergy,

Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada.

1. Tomatine, isolated from extracts of crown gall-infected tomato plants or

obtained commercially, was tested for anti-inflammatory activity using three
different methods.
2. Tomatine administered to intact rats intramuscularly in a dose range of
1-10 mg/kg or orally in doses of 15-30 mg/kg exerted a significant dose
dependent inhibition of carrageenan induced paw oedema. The inhibitory
effect of tomatine when given in a dose of 10 mg/kg intramuscularly to intact
rats lasted more than 24 hr.
3. In adrenalectomized rats significant dose-related inhibition of paw oedema
was obtained with tomatine and the inhibition at each dose level (0.5-10
mg/kg) was found to be greater than that found in intact animals.
4. Tomatine administered subcutaneously to intact rats daily for 7 days in
doses of 5 or 10 mg/kg exerted a significant, dose dependent inhibition of
granulation tissue formation induced by the subcutaneous implantation of
carrageenan impregnated cotton pellets.
5. Tomatine administered to intact mice in a dose of 10 mg/kg subcutaneously
1 hr before the intraperitoneal injection of acidified saline and intravenous
pontamine sky blue significantly decreased the leakage of the protein bound
dye into the peritoneal cavity.
6. Tomatidine, the aglycone of tomatine, was not effective at dose levels of
10-20 mg/kg in any of the three tests.

The isolation of two antihistamine-like substances from crown gall-bearing

tomato stems has been recently reported (Wakkary, Goodfriend & Kovacs, 1966)
and one of the active principles has been identified as tomatine (Calam & Callow,
1964; Kovacs, Wakkary, Goodfriend & Rose, 1964; Wakkary et al., 1969a). Tested
on isolated organ preparations, tomatine was found to exert a non-specific effect in
antagonizing, equally well, the contractions induced by histamine, bradykinin,
SRS-A, 5-hydroxytryptamine and acetylcholine (Wakkary, Kovacs, Goodfriend &
Rose, 1967). When injected into guinea-pigs, tomatine was found to be highly active
in preventing the effects of intradermally injected histamine and bradykinin on
capillary permeability and it exerted some protection against the effects of a lethal
histamine aerosol (Wakkary et al., 1967; Calam & Callow, 1964).
It is a widely accepted theory that, at least in the initial stage of inflammation,
Anti-inflammatory activity of tomatine 749
the vasoactive amines (histamine, kinins, etc.) might play an important role as
mediators (Wilhelm, 1962; Whitehouse, 1965). Since tomatine inhibited the effects
of these substances it seemed of interest to investigate its possible action on inflam-
matory reactions.
The results presented in this paper deal with the effect of tomatine and its
aglycone, tomatidine, in three different types of experimentally induced inflammation,
and the data obtained are compared whenever possible with the effect of hydro-
cortisone or dexamethasone.
Methods
Male Wistar rats weighing 125-185 g, intact, or adrenalectomized 24 hr before
use, and male albino CF No. 1 mice weighing 25-30 g were obtained from Canadian
Breeding Laboratories (St. Constant, Quebec, Canada). All animals were fed on
a standard Purina cube diet (except animals used for oral drug trials, which were
fasted for 24 hours before testing) and water (or 1 % saline if adrenalectomized)
ad libitum both before and during experiments.
Tomatine produced in our laboratory and recrystallized at least five times or
commercially prepared (Koch-Light Laboratories, Colnbrook, England) was
dissolved in acidified distilled water (adjusted to pH 2.0 with 12 N hydrochloric acid)
to give a concentration of 0 1-10 mg/ml. The final pH of the solution was 3.0 and
it was stored at -20° C until required. Tomatidine (Koch-Light Laboratories,
Colnbrook, England) was freshly suspended in a mixture of 1 part absolute ethanol
and 2 parts distilled water (adjusted to pH 2.0 with 12 N hydrochloric acid) to give
a concentration of 20 mg/ml. Hydrocortisone acetate suspension (Hydrocortone,
Merck, Sharp & Dohme, Montreal, Canada) was diluted with sterile physiological
saline from an initial concentration of 25 mg/ml. to 5 mg/ml. Dexamethasone
(Decadron, Merck, Sharpe & Dohme, Montreal, Canada) was diluted with Decadron
Vehicle (Merck, Sharp & Dohme, Montreal, Canada) from an initial concentration
of 4 mg/ml. to 0.1 mg/ml.

Rat paw oedema formation

0.05 ml. of 1% (w/v) solution of carrageenan (Viscarin, Marine Colloids Inc.,
Springfield, New Jersey, U.S.A.), in sterile physiological saline was injected into the
plantar surface of the left hind paws 1 or more hr after the administration of the
drug or vehicle to groups of rats. Paw volumes were measured immediately and
again 3 hr later, using the method of Harris & Spencer (1962) with slight modifica-
tion. The animals were lightly anaesthetized with ether to ensure paw flaccidity.
1% aqueous Triton X-100 (A and C American Chemicals, Montreal, Canada) was
used as the wetting agent. On the reservoir an additional line was drawn perpen-
dicular to the horizontal line described by Harris & Spencer. The paw was so
placed in the reservoir that the tip of the foremost toenail came to the intersection
of these lines while the heel just rested against the reservoir wall. This procedure
helped to ensure uniformly reproducible paw volume measurements.
Cotton pellet induced granuloma formation (Bush & Alexander, 1960)
Cylindrical cotton pellets (8 mm x 8 mm) cut from 14 cm absorbent cotton dental
rolls were each soaked for 1 min in 1% (w/v) aqueous carrageenan (Viscarin,
750 R. B. Filderman and B. A. Kovacs

Marine Colloids Inc., Springfield, New Jersey, U.S.A.), dried in air overnight on
glass plates, paired+2 mg and autoclaved for 1 hr at 30 lb/in2. Pellets were
implanted subcutaneously one into each axillary fold of groups of male rats under
light ether anaesthesia. Each animal received 100,000 i.u. penicillin G potassium
intramuscularly both on the day of implantation and 1 day later. The compound
or the vehicle was then administered subcutaneously once daily for 7 days, the
first dose being given immediately after implantation. Twenty-four hours after
the final dose, the animals were killed and the pellets, together with the surround-
ing granulomatous tissue, were dissected out, dried at 600 C for 48 hours and
weighed. If either granulomatous pellet was found to contain serum or blood,
both were discarded. The extent of granulomatous tissue formation was calculated
by subtracting the original pellet dry weight.
Peritoneal capillary permeability test (Northover, 1963)
Groups of mice received the compound or vehicle subcutaneously. Three hours
later each mouse received 4 ml. of 0.05 N acetic acid in 0.9% saline intraperitoneally.
followed as quickly as possible by 0.1 ml. of 4%/0 pontamine sky blue intravenously.
One hour after the administration of the dye the animals were killed and the
peritoneal cavity was opened and drained. The exudate was centrifuged for 10 min
at 2,000 rev/min and 0.5 ml. of the supernatant was diluted with 4 5 ml. of 0.9%
saline. Dye concentrations were measured in a Perkin-Elmer spectrophotometer
at 625 m,u, saline being used as a blank. The mean percentage of light absorption
was then calculated for both the treated and control groups.
Student's t test was used to evaluate the results in all experiments.
Results
Rat paw oedema
Groups of twenty intact rats were given tomatine in doses of 0-1, 1, 5 or 10 mg/kg;
tomatidine in a dose of 20 mg/kg, hydrocortisone in a dose of 5 mg/kg, or dexa-
methasone in a dose of 0-1 mg/kg intramuscularly, 1 hr before the administration
of carrageenan into the left hind paw. Identical numbers of animals received the
corresponding vehicles. The results obtained in these experiments are summarized
in Table 1. Tomatidine given in doses of 1 mg/kg or more induced a significant dose
dependent inhibition of paw swelling. In this test, the anti-oedema effects of

TABLE 1. Comparative inhibitory effects of intramuscularly administered tomatine, tomatidine,

hydrocortisone and dexamethasone on carrageenan-induced paw oedema formation in intact rats
Number Mean paw volume %
of animals Increase in ml.+S.D. Inhibition
Dose - of paw P
Drug (mg/kg) Test Control Test Control swelling value
Tomatine 10 20 20 029±011 051±0-14 44-6 <0001
Tomatine* 5 20 20 0-26±0t12 0.39±0 12 316
35.4
<0-01
<0-01
Tomatinet 5 20 20 0-31±0 16 0.48±0t15 0.05
Tomatine 1 20 20 0.43±0 11 0.53±0 18 19 3 <
Tomatine 0*1 20 20 0-40±0 12 0.37±0 13 -7.9
Hydrocortisone 5 20 20 0.23±0t15 0-32±0t12 28-6 <0.05
Dexamethasone 0.1 20 20 0.18±0-12 0 55±0.16 67.6 < 0 001
Tomatidine 20 19 20 040±0-13 045±0t12 11.1 N.S.
* Tomatine isolated from crown
gall-infected tomato stalks in ourlaboratory.
t Commercial tomatine (Koch-Light Laboratories).
Anti-inflammatory activity of tomatine 751

5 mg/kg of tomatine and hydrocortisone were approximately equal. Tomatidine,

on the other hand, in the dose administered, did not exert significant anti-oedema
effects.
In order to see if the effect of tomatine was mediated through stimulation of the
adrenals, a second set of experiments was performed. Table 2 summarizes the
results obtained when groups of twenty adrenalectomized rats received tomatine
in doses of 01, 0 5, 1, 5 or 10 mg/kg, hydrocortisone in a dose of
5 mg/kg, or the corresponding amount of vehicle (controls) intramuscularly 1 hr
before the carrageenan administration. Again, tomatine administered in doses of
0.5 mg/kg or more brought about a significant dose dependent inhibition of
oedema formation. Further, in these experiments tomatine appeared to be much
more potent than hydrocortisone.
The next series of experiments, the results of which are shown in Table 3, was
designed to determine the duration of the oedema inhibitory effect of tomatine.
Carrageenan was injected 8, 24 and 48 hr after tomatine 10 mg/kg or the vehicle
(controls) was given intramuscularly to one or more groups of twenty rats. The
activity of tomatine was not significantly reduced after 8 or 24 hr and a slight
inhibitory effect could perhaps be observed even after 48 hr.
To find out if tomatine would also be effective when administered orally, groups
of twenty animals received 15 or 30 mg/kg of tomatine or the same volume of
vehicle by stomach tube 1 hr before the carrageenan injection.

TABLE 2. Comparative inhibitory effects of intramuscularly administered tomatine and hydrocortisone

on carrageenan-induced paw oedema formation in adrenalectomized rats
Number Mean paw volume %
of animals Increase in ml.±s.D. Inhibition
Dose of paw P
Drug (mg/kg) Test Control Test Control swelling value
Tomatine 10 20 20 0.1340-12 042±0 10 68-4 <0001
Tomatine 5 20 20 0.26±0-10 0.47±0d11 44-4 < 0 001
Tomatine 1 20 20 0-16±0-12 0-38±0-10 57.5 < 0001
Tomatine 0.5 20 20 0.38±0-13 0 48±0 11 20.5 < 0-02
Tomatine 0-1 20 20 0-34±0-11 0.33±0.09 -0.9
Hydrocortisone 5 20 20 0-25±0-12 0 38±0l15 34-2 <0 01

TABLE 3. Effect ofpretreatment time on the inhibitory action of intramuscularly administered tomatine
on carrageenan induced paw oedema in intact rats
Pre- Number Mean paw volume %
treatment of animals Increase in ml. + S.D. Inhibition
time Dose , D-- of paw P
Drug (hr) (mg/kg) Test Control Test Control swelling value
Tomatine 8 10 20 20 0.26±0-13 0-46±0-16 43.5 <0001
Tomatine 24 10 20 20 0.24±0.13 0.36±0-14 33-3 < 0 01
Tomatine 48 10 30 28 0.37+0.20 0-39±0.20 5-1 N.S.

TABLE 4. Inhibitory effect of orally administered tomatine on carrageenan-induced paw oedema in

intact rats
Number Mean paw volume
of animals Increase in ml. ±S.D. Inhibition
Dose of paw p
Drug (mg/kg) Test Control Test Control swelling value
Tomatine 30 20 20 0.28±0 13 043±0t13 36.3 <0001
Tomatine 15 20 20 0.33±0*13 0.41±0.13 19-5 =005
752 R. B. Filderman and B. A. Kovacs

> ooZo
vv V
v~
0 c:
C -- 0
0-
o ~
~ ~
0 11,11
'I cc gn 00
~ 1*

.0
:3 -~E k.. C:q I..
rA
N~
kA
.O
4- "

.2 1.0.

0*0000
-
*
C -H-H-H-H
o U o .---..
U mCNt
mc,.A
.4
Fo
o- _ _
I-1 -O
-H0 .

0 -H. 0000
=-
r-

0 -H -H-H -H
o]Ne_ "
0_
-000
.
-
. *

4)
0(3
qc 0

4)

1.
TTq
4)

(E
Cd
0
C
U
0

- roO

z
0
9) 00
w
',
o -- o r. o6
to ~ ~
In EE

U
>h 0

E 0 EEE K

v08x
Anti-inflammatory activity of tomatine 753
The results obtained are summarized in Table 4. As can be seen, orally adminis-
tered tomatine in the dose range used exerted a significant dose dependent inhibitory
activity upon oedema formation.
Cotton pellet induced granuloma formation
Groups of ten rats were treated daily for 7 days with tomatine in a dose of 5 or 10
mg/kg, tomatidine in a dose of 10 mg/kg, dexamethasone in a dose of 01 mg/kg,
or the corresponding vehicle subcutaneously. Table 5 summarizes the results
obtained. At both dose levels tomatidine induced a significant inhibition of
granulation tissue formation. The effect was dose dependent, but as the data show,
in this test tomatine was much less potent than dexamethasone. The daily
injections were well tolerated by the animals and at the time of pellet removal no
signs of local purulence were found at the site of the tomatine injection. Tomatidine
in the dose administered did not inhibit granulation tissue formation. Animals
treated with dexamethasone lost about 15% of their original body weights, whereas
all of the other animals which were treated with tomatine, tomatidine or the various
vehicles showed an average weight gain of about 12%.
Peritoneal capillary permeability test
Groups of 11 mice were injected subcutaneously with 1, 5 or 10 mg/kg of
tomatine, 10 or 20 mg/kg of tomatidine or the corresponding vehicle in a total
volume of 015 ml. Table 6 summarizes the absorption at 625 mg for the diluted
peritoneal fluids. Only at a dose of 10 mg/kg, did tomatidine significantly inhibit
the leakage of dye into the peritoneal cavity while tomatidine in the dose range
administered did not exert any significant inhibitory activity.
Discussion
The results obtained in these experiments indicate that tomatidine not only
exhibits a fairly potent anti-inflammatory activity in the tests used but also mani-
fests several rather unusual properties not previously described for any other anti-
inflammatory compound in general use. The anti-inflammatory activity of tomatine
was most extensively studied in the carrageenan-induced rat paw oedema test
because it permits a valid quantitative assay and because most of the known anti-
inflammatory drugs seem to possess anti-carrageenan activity (Winter, Risley &
Nuss, 1962; 1963 ; Niemegeers, Verbruggen & Janssen, 1964; Winter, 1964).
In its oedema inhibiting activity, tomatine showed a good dose-response relation
in the dose range used. In adrenalectomized animals at each dose level tested, the

TABLE 6. Influence of tomatine and tomatidine on capillary permeability in the mouse peritoneum
Number %
Drug of animals Absorbance at 625 mni±S.D. Inhibition
administered Dose o __-_-- ___ of dye P
subcutaneously (mg/kg) Test Control Test Control leakage value
Tomatine 10 20 19 0.21640-095 03199±0*139 34-4 < 0 01
Tomatine 5 21 21 0.444+0.157 0470+0e158 5-5 N.S.
Tomatine 1 19 19 0.406±0t142 0377±0*182 -7.7 N.S.
Tomatidine 10 20 20 0408+0.178 0*457±0*231 10-7 N.S.
Tomatidine 20 21 21 0371+0.144 O-409+0147 8.8 N.S.
754 R. B. Filderman and B. A. Kovacs

relative percentage inhibition brought about by tomatine was higher than in intact
animals. These results are very unusual and at present cannot be explained.
Another unusual result was the long duration of action of a single dose of tomatine
in inhibiting carrageenan-induced rat paw oedema. This effect was, however,
anticipated since Wakkary et al. (1967) have previously demonstrated a very long-
lasting but reversible effect of tomatine in the inhibition of histamine or bradykinin
induced smooth muscle contractions.
In order to evaluate the possible anti-inflammatory activity of tomatine and
tomatidine on the later stages of inflammation, the widely used carrageenan impreg-
nated cotton pellet granuloma test was selected. Granulation tissue formation has
been shown to be depressed by nearly all anti-inflammatory drugs, but the steroids
are most active while phenylbutazone and salicylates show poor activity and chloro-
quine is almost inactive (Whitehouse, 1965; Adams & Cobb, 1967). The effect of
tomatine in suppressing granulation tissue formation seems to be similar to that of
steroids, since the inhibition of the granuloma formation was directly proportional
to the dose of tomatine used. Unlike the dexamethasone treated animals, which lost
about 15% of their original body weights by the end of the experiment, the animals
treated with tomatine and tomatidine showed no change in body weight as compared
with controls. Tomatine therefore appears to be effective in suppressing granulation
tissue growth without overt toxic effects.
Tomatine has a steroid nucleus, and yet the entire molecule may be functioning
in a manner which differs from that of the steroid compounds. Consequently, as a
further test, the mouse peritoneal capillary permeability test of Northover (1963) was
chosen, since it has been shown (Northover, 1963, 1964) that steroids are ineffective
in this situation while non-steroid substances are generally active.
Tomatine was found to exert a significant inhibition of dye leakage when given in
a dose of 10 mg/kg subcutaneously. Therefore, according to the criteria of the
method in this test, tomatine seems to act like a non-steroid drug.
Interestingly, tomatidine, the aglycone of tomatine, showed no significant inhibi-
tory activity in any of the three tests. Previously, Calan and Callow (1964) also
observed that tomatidine, unlike tomatine, did not protect animals against the lethal
effects of a histamine aerosol.
The present studies did not include systemic toxicological investigations, but
Wilson, Poley & De Eds (1961) showed that tomatine is a relatively non-toxic agent.
It has also been shown by several groups of workers that tomatine exhibits anti-
fungal and antibacterial activity (Irving, Fontaine & Doolittle, 1945; Schuster &
Tarrade, 1969). Should tomatine prove to be effective as an anti-inflammatory
agent in humans without exerting the serious side effects observed with both the
steroid and non-steroid anti-inflammatory drugs, it may be of use in the treatment
of chronic inflammatory diseases.

This work was supported by a grant from the John A. Hartford Foundation (New York).
We are greatly indebted to Dr. B. Rose, Director of the Division of Immunochemistry and
Allergy, for his constant interest in the progress of this study. The authors also wish to thank
Miss S. McPhail, Mr. G. Szentirmay and Mr. I. De Thokoly, Jr., for valuable technical
assistance. Data from a thesis presented to McGill University, Montreal, by Dr. R. B.
Filderman in partial fulfilment of the requirements for the degree of Master of Science, June
1969. Work done during support of Dr. R. B. Filderman by Medical Research Council
Anti-inflammatory activity of tomatine 755
(Canada) Fellowship Grant. We wish to thank Marine Colloids Inc., Springfield, N.J., U.S.A.,
and Merck, Sharp and Dohme, Montreal, Canada, for the generous supply of carrageenan
(Viscarin) and Decadron and Decadron Vehicle.

REFERENCES
ADAMS, S. S. & COBB, R. (1967). Non-steroidal anti-inflammatory drugs. Progress in Medicinal
Chemistry, 5, 59-138.
BUSH, I. E. & ALEXANDER, R. W. (1960). An improved method for the assay of anti-inflammatory
substances in rats. Acta Endocr., 35, 268-276.
CALAM, D. H. & CALLOW, R. K. (1964). Histamine protection produced by plant tumour extracts.
The active principle of tomato plants infected with crown gall. Br. J. Pharmac. Chemother.,
22, 486-498.
HARRIS, J. M. & SPENCER, P. S. J. (1962). A modified plethysmographic apparatus for recording
volume changes in the rat paw. J. Pharm. Pharmac., 14, 464-466.
IRVING, G. W., JR., FONTAINE, T. D. & DOOLITrLE, S. P. (1945). Partial antibiotic spectrum of
tomatine, an antibiotic agent from the tomato plant. J. Bact., 52, 601-607.
KoVACs, B. A., WAKKARY, J. A., GOODFRIEND, L. & RoSE, B. (1964). Isolation of an antihistaminic
principle resembling tomatine from crown gall tumors. Science, N. Y., 144, 295-296.
NIEMEGEERS, C. J. E., VERBRUGGEN, F. J. & JANSSEN, P. A. J. (1964). Effect of various drugs on
carrageenan-induced oedema in the rat hind paw. J. Pharm. Pharmac., 16, 810-816.
NORTHOVER, B. J. (1963). The permeability to plasma proteins of the peritoneal blood vessels of
the mouse and effect of substances that alter permeability. J. Path. Bact., 85, 361-370.
NORTHOVER, B. J. (1964). The action of anti-inflammatory compounds in mice with peritonitis.
J. Path. Bact., 87, 395-404.
SCHUSTER, G. & TARRADE, A. (1969). Antibiotic activity of tomatine. Fedn Proc., 28, 264.
WAKKARY, J. A., GOODFRIEND, L. & KOVACS, B. A. (1966). Antihistamine activity of tomatine-
free crown gall extracts. Fedn Proc., 25, 692.
WAKKARY, J. A., KOVACS, B. A., GOODFRIEND, L. & ROSE, B. (1967). Isolation of antihistamine-
like substances from crown gall of tomato. Excerpta Med. int. Congr., Ser. No. 144, 4
WHrTEHOUSE, M. W. (1965). Some biochemical and pharmacological properties of anti-inflammatory
drugs. Progress in Drug Research, 8, 321-429.
WILHELM, D. L. (1962). The mediation of increased vascular permeability of inflammation.
Pharmac. Rev., 14, 251-280.
WILSON, R. H., POLEY, G. W. & DE EDS, F. (1961). Some pharnacologic and toxicologic properties
of tomatine and its derivatives. Tox. appl. Pharmac., 3, 39-48.
WINTER, C. A. (1964). Anti-inflammatory testing methods: comparative evaluation of indomethacin
and other agents. In Non-steroidal Anti-Inflammatory Drugs, ed. Garattini, S. and Dukes,
M. N. G., pp. 190-202. Amsterdam: Excerpta Medica.
WINTER, C. A., RISLEY, E. A. & Nuss, G. W. (1962). Carrageenan-induced edema in hind paw of
the rat as an assay for anti-inflammatory drugs. Proc. Soc. exp. Biol. Med., 111, 544-547.
WINTER, C. A., RISLEY, E. A. & NUss, G. W. (1963). Anti-inflammatory and antipyretic activities
of indomethacin, 1-(p-chlorbenzoyl)-5-methoxy-2-methyl-indole-3-acetic acid. J. Pharmac. exp.
Ther., 141, 369-376.
(Received July 10, 1969)