1977 Biochemical and Biophysical Research Communications Vol. 77, No.4

by David L. Vesley, William R. Graves, Timothy M. Lo, Mary Ann Fletcher, Gerald S. Levey
The balsam pear (Momordica charantia abreviata) is a plant that grows wild throughout subtropical regions of the world inclufing the…

Isolation of a Guanylate Cyclase Inhibitor from the Balsam (Momordica charantia Abreviata)

Author: David L. Vesley, William R. Graves, Timothy M. Lo, Mary Ann Fletcher, Gerald S. Levey

Type of Publication: Pre-Clinical

Date of Publication: 1977

Publication: Biochemical and Biophysical Research Communications Vol. 77, No.4, 1977

Organization: University of Miami School of Medicine , Miami , Florida

SUMMARY

The balsam pear (Momordica charantia abreviata) is a plant which grows wild thoughout subtropical regions including the Gulf Coast and Florida. The ripe fruit is highly toxic and produces mild hypoglycemia. Preparations of the ripe fruit and the leaves but not the unripe fruit or seeds inhibited guanylate cyclase activity whereas adenylate cyclase activity was unaffected. The guanylate cyclase inhibitor (GCI) has been purified about 30-fold has an estimated molecular weight of 5000 to 50,000 and is acid stable and heat labile. GCI blocked the activation of guanylate cyclase by nitroso chemical carcinoges, and therefore, may be useful in studying mechanism carcinogenesis.

INTRODUCTION

The balsam pear (Momordica charantia abreviata) is a plant that grows wild throughout subtropical regions of the world inclufing the Gulf Coast and Florida. In folk-medicine the fruit and leaves of this plant have reportedly been used as hypoglycemic agents (1), purgatives (2), emetics (2), and abortifiacients (2). The plant is highly toxic and in Miami several instances of illness following ingestion of the ripe fruit have been described in children and domestic animals such as dogs (2). Because of the toxicity and the purported medicinal uses we examined the effects of the fruit and leaves on guanylate cyclase (E. C. 4.6.1.2.) and adenylate cyclase (E.C. 4.6.1.1.) to determine if any correlation could ultimately be derived for the observed in vivo phenomena and the intracellular concentrations of cyclic nucleotides.

The results show that the ripe fruit and leaves contain a guanylate cyclase inhibitor which has the ability to impair chemical carcinogen-induced increased in guanylate cyclase activity.

METHODS

Tissues used in these experiments were obtained from male Sprague-Dawley rats, weighing 150-200 grams that had been maintained ad libitum on Purina Laboratory chow. Alumina oxide, neutral activity I for column chromatography, was obtained from E. Merck, (Darmstadt, Germany). The alpha [³² p] GTP was from New England Nuclear Corporation (Boston, Mass.) and International Chemical and Nuclear Corporation (Irvine, Calif.).

The ripe fruit, unripe fruit, leaves, and seeds were homogenized in 0.03 M Tris, pH 7.6, filtered ythrough gauze and centrifuged at 12,000 g. These extracts had a protein concentration of about 8 g/ml and were added to the enzyme preparation described below at a ration approximating 0.4 g of extract protein to 400 g of enzyme protein.

Guanylate cyclase activity was measured as previously described (3-5) utilizing a modification of the original method of White and Zener (6). The various tissues were homogenized in cold 0.03 M Tris CHI, pH 7.6 and centrifuged at 37, 000 g in a Sorval refrigerated centrifuge at 40⁰ for 15 minutes. The supernatant was assayed at 37⁰ for 10 minutes for guanylate cyclase activity, using a reaction mixture consisting of 20 mm tris HCI, pH 7.6; 5 mm MnCI₂; 2.67 mM cyclic GMP (used to minimize destruction of [³² p]- GTP; a GTP regenerating system (5 mM creatine phosphate, 11.25U creatine phosphokinase); 100 g bovine serum albumin; 20 mM caffeine; [ a-³²p]-GTP approximately 5×10⁵ cpm; and the enzyme preparation having 0.2 to 0.6 mg protein. The reaction was terminated by the addition of 10 l of 0.1 M EDTA, pH 7.6, containing about 30,000 cpm of [³H]-cyclic GMP (to estimate recovery in the subsequent steps) and boiling for three minutes. After cooling in an ice bath, the [³²p]-cyclic GMP formed is isolated by sequential chromatography on Dowex-50-H⁺ and alumina using the modification of Krishna and Krishan (7). The reaction mixtures were diluted with 0.5 ml of distilled water and transferred to a Dowex-50-H⁺ column (10×75 mM). The column were then eluted with another 0.5 ml distilled H₂O and the eluates (1 ml) were discarded. The second 1 ml water fraction eleuted from Dowex-50-H⁼ column was allowed to directly pass through a column of dry neutral alumina (10×75mM). The alumina column were then eluted with 2 ml of 0.03 M Tris-HCI buffer, pH 7.6. The above three mls of elutant from the alumina column were collected directly into scintillation vials containing 15 mls of Bray’s solution. The eluates were then counted in a Packard Tri-Carb Liquid Scintillation spectrometer. The overall recovery of cyclic GMP after the two-stage chromatographic procedure was 85 to 95%. All of the [³²p] –containing material was identifiable as cyclic GMP as determined by thin layer chromatography on a cellulose PEI, Brickman) using 1 M LiCI as solvent and Chromar sheets (Mallinchrodt, St. Louis, Momordin.) developed with absolute alcohol and concentrated NH₄OH (5:v/v-insulin)/ Proteins was determined by the method of Lowry et al. (8). Adenylate cyclase was prepared and assayed as previously described from our laboratory (9).

RESULTS

The filtrates of the ripe fruit and the leaves produced virtually complete inhibition of rat hepatic guanlate cyclase activity from a control level of 280  12 pmoles accumulated/mg protein/10 min. to 5  2 pmoles for the ripe fruit and 10  4 pmoles for the leaves (Table I). The unripe fruit and seeds were without effect. The extract from the ripe fruit produced similar reductions in guanylate cyclase activity in other rat tissues including pancreas, heart, lung, stomach, colon and kidney (Table 2). In contrast to these findings with guanylate cyclase, the extracts had no measurable effects on adenylate cyclase activity.

The guanylate cyclase inhibitor (GCI) has been purified about 30-fold using, sequentially, a 12,000 g centrifugation of an aqueous extraction of the ripe fruit, Amicon-30 ultrafiltration, dialysis at pH 7.0 and DEASE Sephadex chromatography. The resultant purified material has a molecular weight of 5,000 to 50,000 as estimated by gel filtration and is acid stable and heat labile. GCI is probably not a lipid since the material remained in the aqueous phase after a double extraction with ether: ethanol 94:1) followed by an extraction with chloroform: methanol (2:1).

In the past year a number of nitroso chemical carcinogens (3-5, 10) including nitrosoamides, nitrosoamines, and hydrazine have been shown to activate guanylate cyclase, a finding which may be of considerable importance in understanding the mechanism of chemical carcinogenesis. We therefore examined the effect of GCI on the activation of guanylate cyclase by the nitrosoamides, streptozotocin and N-methyl-N-nitroso-N’-nitroguanidine (MNNG), which are among the most potent acrcinogen stimulators of guanylate cyclase.Table 3. demonstrate that GCI almost totally abolished the 20 to 25 –fold increases in guanylate cyclase activity produced by these carcinogens. The decreases in activity were highly significant (p<0.001).

DISCUSSION

The data in this report demonstrate that the balsam pear (Momordica charantia abr.) contains an inhibitor of guanylate cyclase. This inhibitor almost completely abolished guanylate cyclase activity in all rat tissues studied and was specific in so far as it did not alter adenlate cyclase activity. The inhibitory mateial is acid-stable and heat labile, has been partially purified (30-fold), and does not appear to be a lipid. When purified and identified it should be a useful probe to aid in further understanding the controversial, complex, anf intriguing role of cyclic GMP in cell biology (11).

A number of recent investigations have demonstrated that guanylate cyclase activity is increased by nitroso chemical carcinigens (3-5,10). In addition, cyclic GMP may play a role in malignant transformation (12) and cyclic GMP levels have been reported to be increased in some tumors (13,14) including an adenocarcinoma of the human colon (14). Therefore, it is interest and potentially great importance that GCI blocks the activation of guanylate cyclase by the nitroso chemical carcinogens, streptozotocin and MNNG. Whether or not GCI will abolish or prevent tumor-induction by these agents and what rhe effects of GCI are on other tumors formulates the basis for critical future experiments.

ACKNOWLEDGEMENTS

We thank Julia F. Morton who stimulated our interest in studying the balsam pear. This work was supported by NIH grant HL 13715-07 and USPHS AM-167-63-06.

Dr. Levey is an Investigator of Howard Hughes Medical Institute. Society, Chicago, Illinois, June 8,1977.

REFERENCES

Sharma, V. N., Sogani, R. K., and Arora, R. B. (1960) Ind. J. Med. Res., 48(4), 471-477.

Morton, J. F. (1967) Economic Botany, 21(2), 57-68.

Veseley, D. L., and Levey, G. S. (1977) Biochem. Biophys. Res. Commun., 74(2), 780-784.

Vesely, D. L., and Levey, G. S. (1977) Proc. Soc. Experimental. Biol. And Med., 155(3), 301-304.

Vesely, D. L., Rovere, L. E., and Levey, G. S. (1977) Cancer Res., 37, 28-31.

White, A. A., and Zenser, T. V. (1971) Anal Biochem., 41, 373-396.

Krishna, G., and Krishna, N. A. (1975) J. Cyclic Nucl. Res., 1, 293-302.

Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem., 193, 265-275.

Levey, G. S., and Eptein, S. E. (1969) J. Clinical . Invest., 48, 1663-1669.

DeRubertis, F. R., and Craven, P. A. (1976) Science 193, 897-899.

Goldberg, N. D., Haddox, M. K., Nicol, S. E., Glass, D. B., Sanford, C. H., Kuehl, F. A., Jr., and Estensen, R. (1975) Adverse. Cycl. Nucl. Res., 5, 307-338.

Kram, R., and Tomkins, G. M. (1973) Proc. Natl. Acad. Sci. USA 70, 1659-1663.

Kimura, H.,and Murad, F. (1975) Proc. Natl. Acad. Sci. USA 72, 1965-1969.

DeRubertis, F. R., Chayoth, R., and Field, J. B. (1976) J. Clinical . Invest. 57, 641-649.