Effects of Ginsenosides, Lectins and Momordica
charantia Iinsulin-Like Peptide on Corticosterone
Production by Isolated Rat Adrenal Cells
Author: T. B. Ng, W.W. Li, H. W. Yeung
Type of Publication: Pre-Clinical
Date of Publication: 1987
Publication: Journal of Enthnopharmacology Vol. 21, pp 21-29
Organization/s: The Chinese Universtiy of Hong Kong Shatin
Abstract: Ginsenosides Rb1, Rb2, Rc and Rgl inhibited steroidogenesis induced by a
maximally active dose of corticotropin in isolated rat adrenal cells. The galactosebinding lectins from Momordica charantia seeds and Trichosanthes kirilowii tubers
and mannose-binding concanavalin A did not affect basal corticosterone productions.
The lectins potentiated steroidogenesis induced by a submaximal dose of
corticotropin but were without effect on steroidogenesis induced by maximally active
dose of corticotropin. Momordica charantia insulin-like peptice did not affect
steroidogenesis.
Introduction
Ginsenosides Rb2, Rc ab=nd Rgl, Momordica charantia and Trichosanthes kirilowii
lectins, concanavalin A and an insulin-like peptide from Momordica charantia seeds
have been studied for their effects on lipid metabolism in charantia seeds have been
studied. For their effects on lipid metabolismin isolated rat adipocytes (Ng et al.,
1985, 1986b, 1987). The ginsenosides were shown to suppress hormone-induced
lipolysis and lipogenesis. The non-competitive nature of ginsenoside inhibition of
corticotropin-induced lipolysis and the ability of ginsenosides to inhibit lipolysis
induced by a maximal dose of dibutyryl cyclicAMP suggest that the site of action of
ginsenosides is beyond the production of cyclic AMP, the intracellular mediator of the
action of lipolytic hormones (Ng et al., 1986b). The galactose-binding M. charantia
lectin (Ng et al., 1986a), the T. kirilowii isolectins (Yeung et al., 1986) and mannosebinding concanavalin A (Hedo et al., 1981, Katzen et al., 1981) have been shown to
exhibit antilipolytic and lipogenic activities. The lectins appear to express their
insulin-like activities by interacting with insulin receptors on adipocytes or plasma
membrane components closely associated with insulin receptors (Hedo et al., 1981;
Katzen et al., 1981). An (missing text).from M.charantia seeds with a protocol
similar to the mammalian insulin isolation procedure (Ng et al., 1987).
Ginseng saponin administration has been shown to elevate plasma levels of
corticotropin and corticosterone (Hiai et ai., 1987b). Differences in the sugar moiety
between different ginsenosides did not engender any major differences in their
activities. Rather, the sapogenin moiety of the ginsenosides was indispensable for
the expression of the corticosterone-stimulating activity since glycyrrhizin and
saikosaponin, whose sapogenin moieties are different from those of ginsenosides,
were devoid of such activity (Hiai et al., 1979b). Ginseng saponins increased the
adrenal camp level in intact but not in hypophysectomized rats, indicating that its
action on the adrenal glands was an indirect one via the pituitary (Hiai et al., 1979a).
The ability of dexamethasone, a synthetic glucocorticoid, to block the effect of
ginsenosides on pituitary corticotropin and adrenal corticosterone secretion further
supports the contention that the CNS-hypophyseal complex rather than the adrenal
cortex is the site of action of the ginsenosides (Hiai et al., 1979b). However, no
direct studies of the effect of ginsenosides on the adrenal have been carried out.
Since we have available to us an assay system which measures the steroidogenic
response of isolated rat adrenal cells to corticotropin and which has been successfully
employed to purify corticotropin-like materials and to demonstrate their existence in
various vertebrate tissues (Hon and Ng, 1986a,b; Ng et al., 1986a,b), the effect of
various ginsenosides onadrenal cells was examined.
Hormones which stimulate adrenal steroidogenesis, including corticotropin, melanotropin, -melanotropin and -lipotropin (Li et al., 1982) are also lipolytic (Ng
and Yeung, 1982). In view of the effects of the lectins and M. charantia insulin-like
peptide on lipid metabolism, they were also tested for a possible effect on
corticotropin-induced steroidogenesis.
Materials and Methods
Isolation of ginsenosides, lectins and insulin-like peptide
Ginsenosides were isolated as described by Nagai et al. (1971) from dried
ginseng roots (Panax ginseng CA. Meyer) obtained from a local commercial source.
The identity of individual ginsenosides was confirmed by comparison with authentic
samples supplied by Drs. J. Shoji and O. Tanaka using thin-layer chromatography
and high-performance liquid chromatography. The commercial preparation (Ginsana)
used in this study was a gift of, Pharmaton Ltd. (Switzerland).
Momordica charantia lectin was isolated from a saline extract of the seeds by
a procedure involving ammonium sulfate precipitation and affinity chromatohraphy
on Lacto-gel (Ng et al., 1986a). Trichosanthes kirilowii isolectins 1 and 2 were
isolated from tubers by a procedure involving saline extraction, acetone
fractionation, ion exchange chromatography on DEAE-Sepharose (Yeung et al.,
1986).
Concanavalin A and wheat germ agglutinin were purchased from Sigm
Chemical Co., St. Louis, MO.
The insulin-like peptide employed in this study was isolated from M. charantia
seeds by extraction with acidic ethanol. The ph of the extract was adjusted to 8 and
the resulting precipitate was chromatographed of CM Sepharose CL-6B and then
desalted on Sephadex G-10. A number of factors exhibiting antilipolytic and lipogenic
activities in rat adipocyte were obtained. Two of the factors were demonstrated to be
peptides with similar aminoacid compositions and molecular weights of
approximately 8000. They were designated C1-3 and C6-1, respectively. The former
peptide was unadsorbed on Cm Sepharose in 50 mM phosphate buffer (pH 6.4) an
retarded on Sephadex G-10. The C6-1 peptide was absorbed on CM-Sepharose and
unretarded on Sephadex G-10, and it was utilized in this investigation because of its
higher yield.
Steroidogenesis assay using rat decapsular cells
Adrenal decapsular cells were prepared by the method of Li et al. (1982) with
minor modifications. Male Sprague-Dawley rats weighing 400-450 were killed by
cervical dislocation. Adrenal glands were carefully remove and trimmed free of fat.
The capsules, consisting mainly of glomerulos cells, were separated from the inner
zones made up mainly of fasciculate and reticularis cells. The decapsulated glands
were minced and suspended in Krebs-Ringer bicarbonate buffer (1 ml/gland)
containing collagenase (Sigma type II, 3 mg/ml), glucose (2mg/ml) and bovine
serum albumin (Sigma fraction V, 4 mg/ml) in polypropylene culture tubes. The
tubes were the saturated with 95% O2/5% CO2 and incubated in a Dubnoff
metabolism shaker at 37o C with gentle shaking (65-75 cycles/min) for 1 h. The
tubes were then allowed to stand upright until the tissue had settled. The incubation
medium (Krebs-Ringer bicarbonate buffer) containing 0.4% bovine serum albumin,
0.2% d-glucose and 0.1% lima bean trypsin inhibited (Sigma), was then added (0.5
ml/adrenal) and the adrenal cells were dislodged from the tissue by repeatedly
drawing the suspension into and out of a Pasteur pipette. The pieces of the tissues
were then allowed to settle and the supernatant containing the dispersed cells were
collected and filtered through four layers of cheesecloth. The procedure was repeated
at least twice to ensure completeness of cell dispersal. The cell suspension was
centrifuged at 50 x g for 5 min at room temperature in an MSE GF-centrifuge. The
supernatant was removed by aspiration and the cells were resuspended in the same
volume of incubation medium, washed twice and the cell concentration was adjusted
to about 10 cells/ml. Aliquots of the cell suspension were transferred to
polypropylene culture tubes containing corticotropin and a solution of the fraction
(ginsenoside, lectin or insulin-like peptide) to be assayed so as to make up final
assay volume of 250 l. The (missing text) metabolic shaker with moderate shaking
in an atmosphere of 95% O2/5% CO2 at 37o C for 2 h. At the end of the incubation,
the tubes were frozen at – 20o C until corticosterone determination by
radioimmunoassay.
Corticosterone radioimmunoassay was performed using 1,2,6,7-3H
corticosterone serum (Miles-Yeda). The buffer used was 50 mM Tris-HCI (pH 7.4)
conataining 0.01% bovine serum albumin. The sample of standard (100 l) was
incubated with 40 000 dpm 3H corticosterone (100l), bacitracin (100 l) and
corticosterone antiserum (100 l) at 4o C for 16-24 h. Separation of free and bound
steroid was achieved by Norit A charcoal. An aliquot of the supernatant was taken for
liquid scintillation counting. The rate of steroidogenesis was computed as the
amount of corticosterone produced per 2 h per 300 000 adrenal decapsular cells.
The sensitivity of the assay was 5 pM porcine ACTH and a maximal response
occurred with 5 mM porcine ACTH. A variety of peptides were tested including βendorphin, β-lipotropin, dynorphin (1-13), leucine enkephalin, methionine enkephalin
and vasoactive intestinal peptide and shown to be inactive at the dose of 1 μM. It
has been shown that the potency of α- and β-melanotropins was only 10 –6 that of
corticotropin (Li et al., 1982).
Results
Ginsenosides Rb1, Rb2, Rc and Rg inhibited corticosterone production induced
by a submaximally active dose of corticotropin in a dose-dependent manner (Table
1, Expt.2). Ginsenoside Rc at a dose of 400 μg/ml and ginsenoside Rg1 at a dose of
1 mg/ml did not affect basal steroidogenesis which occurred at a very low rate (Table
2). Although all three doses (10 nM, 100 nM and 1 μM) of ACTH maximally
stimulated steroidogenesis, at any one dose of ACTH the higher dose of ginsenosides
(1 mg/ml) inhibited steroidogenesis to a greater extent than lower dose (400 μg/ml).
Furthermore, at any dose of the ginsenosides, 400 μg/ml or 1 mg/ml, the greatest
inhibition occurred at the lowest dose of corticotropin (10 nM) and the least inhibition
at the highest dose of corticotropin (1 μM). Similarly, Ginsena inhibited corticotropininduced lipolysis (Table 2).
All of the lectins tested did not affect basal steroidogenesis except wheat
germ agglutin. This lectin slightly enhanced basal steroidogenesis. The lectins had no
effect on steroidogenesis induced by a maximally active dose of corticotropin (Table
3, Expt.1) but potentiated the stimulatory effect of a submaximally active dose of
corticotropin (Table 3, Expt.).
Discussion
The present investigation demonstrated a direct inhibitory effect of the
ginsenosides on corticosterone production by rat adrenal decapsular cells. The
inhibitory effect was due to an adverse effect on cell viability because basal
steroidogenesis, albeit low, was not affected and also because there was no increase
in the number of cells stainable with trypan blue. The effect of the ginsenosides was
not exerted on the corticotropin receptor because the inhibition occurred at a
maximally active dose of the hormone. The present results are consistent with the
previous demonstration that ginsenosides administered to hypophysectomized rats
did not affect the level of adrenal cAMP, the intracellular messenger of corticotropin
(Hiai et al., 1979a). The vitro inhibitory effect of the ginsenosides on corticosterone
production occurred at a high dose as utilized in this study. The demonstration that
in vivo administration of ginseng saponin augmented plasma levels of corticotropin
(Hiai et al., 1979b) suggest that the indirect action of ginsenosides on the adrenal
through the pituitary is much stronger than and therefore masks any direct effects of
ginsenosides on the adrenal.
In contrast to their pronounced effects on lipid metabolism in isolated rat
adipocytes (Katzen et al., 1981; Ng et al., 1986a), the lectins did not greatly affect
basal steroidogenesis in isolated rat adrenal decapsular cells. The slight potentiation
of corticosterone production by the lectins at a submaximally active dose of
corticotropin may be due to interaction of the lectins with the glycoprotein
corticotropin receptor on the adrenal cell since the inhibitory effect of lectins on
lipolysis induced by a submaximally active dose of a lipolytic hormone is attributed to
the interaction of lectins with the carbohydrate moiety of insulin receptor or a moiety
closely associated with the insulin receptor (Hedo et al., 1981; Katzen et al., 1981).
The inability of the lectins to potentiate sterodoigenesis at a maximally active dose of
corticotropin is consistent with the suggestion that the lectins act on the corticotropin
receptor.
The Momordica charantia insulin-like peptide lacks hemagglutinating (lectin),
hemolytic (saponin), abortifacient (Ng et al., 1987) and corticosteroidogenic
activities. Further studies are required to establish whether it possesses
hypoglycemic activity.
Acknowledgment
We thank Pharmaton Ltd. (Switzerland) for the gift of the standardized
ginsenoside preparation Ginsana, Dr. J. Shoji (Showa University, Japan) and Dr. O.
Tanaka (Hiroshima University, Japan) for their gifts of gimsenosides, and Miss
Jennifer Chiu for expert secretarial assistance.
References
Hedo, J. A., Harrison, L. C. and Roth, J. (1981) Binding of insulin receptors to lectins:
evidence for common carbohydrate determinats on several membrane receptors.
Biochemistry 20, 3385-3393.
Hiai, S., Sasaki, S. and Oura, H. (1979a) Effect of ginseng saponin on rat adrenal
cyclic AMP. Planta Medica 37, 15-19.
Hiai, S., Yokoyama, H., Oura, H. and Yano, S. (1979b) Stimulation of pituitaryadrenocortical system by ginseng saponin. Endocrinologica Japonica 26, 661-665.
Hon, H. K. and Ng, T. B. (1986a) Opiate-like and adrenocorticotropin-like materials
in equine pancreas. General Pharmacology 17, 397-404.
Hon, W. K. and Ng, T. B. (1986b) Presence of corticotropin-like and opiate-like
hormones in rat and bovine placental tissues. Comparative Biochemistry and
Physiology, 84C, 277-281.
Katzen, H. M., Vicario, P. P., Mumford, R. A. and Green, B. G. (1981) Evidence that
the insulin-loke activities of concanavalin A and insulin are mediated by a common
insulin receptor linked effector system. Biochemistry 20, 3800-5809.
Li, C. H., Ng, T. B. and Cheng, C. H. K. (1982) Melanotropins: aldosterone and
corticosterone stimulating activity in isolated rate adrenal cells. International Journal
of Peptide and Protein Research 19, 361-365.
Nagai, Y., Tanaka, O. and Shibata, S. (1971) Chemical studies on the oriental plant
(missing text) XXIV: Structure of ginsenoside-Rg, a neutral saponin root.
Tetrahedron 127 (missing text) –891.
Ng, T. B. and Li, C. H. (1982) Corticotropin-inhibiting peptide and a synthetic bmelanotropin analog inhibit the lipolytic activity of corticotropin and melanotropin.
International Journal of Peptide and Protein Research 19, 1562-1564.
Ng, T. B. and Yeung, T. B. (1986) Scientific basis of the therapeutic effects of
ginseng. In: (missing text) Steiner (Ed.), Folk Medecine, The Art and the Science,
American Chemical Society, Washington, DC, pp.139-151.
Ng, T. B., Wong, C. M. and Yeung, H. W. (11985) Effect of Trichosanthes kirilowii
lectin on lipolysis and lipogenesis in isolated rat hamster adipocytes. Journal of
Ethanopharmacology 14, 93-98.
Ng, T. B. Wong, C. M., Li, W. W. and Yeung, H. W. (1986a) Isolation of galactosebinding lectin with insulinomimetic activities from the seeds of the bitter gourd
Momordica charantia (family Cucurbitaceae). International Journal of Peptide and
Protein Research 28, 163-172.
Ng, T. B., Wong, C. M. and Yeung, H. W. (1986b) Effect og ginsenosides Rg1, Rc and
Rb2 on hormone-induced lipolysis and lipogenesis in rat epididymal fat cells. Journal
Ethnopharmacology, 16, 191-199.
Ng, T. B., Womg, C. M., li, W. W. and Yeung, H. W (1987) Peptides with antilipolytic
and lipogenic activities from seeds of the bitter gourd Momordica charantia (family
Cucurbitaceae). General Pharmacology, in press.
Yeung, H. W., Wong, D. M., Ng, T. B. and Li, W. W. (1986) Purification of three
isolectins from root tubers of Trichosanthes kirilowii (Trianhuafen). International
journal of Peptide and Protein Research 27, 325-333.