Printer Friendly

Simple partial purification of D-pinitol from soybean leaves. (Notes).

PURSUIT OF THE COMPONENT in Bougainvillea leaves that gave a hypoglycemic effect when fed to normal and diabetic mice led Narayanan et al. (1987) to pinitol as the active component. An oral dose of only 10 mg pinitol/kg body weight significantly lowered blood glucose during a period of 0.5 to 2 h following administration of the cyclitol. More recent studies have shown that a closely related cyclitol--D-chiro-inositol--administered at the same dose, has similar effects on blood glucose levels (Ortmeyer et al., 1993). myo-Inositol, a common constituent of all plant and animal cells, was not effective. Other details regarding the efficacy of certain cyclitols as potential substitutes for insulin are available in a recent review (Hansen and Ortmeyer, 1996).

Recently, there has been commercial interest in providing cyclitols as dietary supplements or as insulin substitutes (see; http://www. The latter corporation is presently marketing pinitol in tablet form under a variety of product names at a retail cost of about $10 per gram. Other corporations (personal communications) are actively seeking foods that are high in pinitol content. As the demand for pinitol as a food supplement or pharmaceutical increases, it will be increasingly important to identify simple methods to purify pinitol from plants naturally rich in this compound.

Pinitol has long been known to be a major carbohydrate in soybean plants (Phillips and Smith, 1974), where it is found in combination with small amounts of other cyclitols (Phillips et al., 1982; Streeter, 1985). In the leaves of the soybean plant, pinitol is the major low molecular weight carbohydrate, usually exceeding the combined concentration of all mono- and disaccharides (Phillips et al., 1982; Streeter and Strimbu, 1998). Typical concentrations of pinitol range from 20 to 25 mg [g.sup.-1] dry leaf blade; i.e., 2.0 to 2.5% of dry weight. This report deals with the application of a standard extraction method of soybean leaves combined with a simple solvent partitioning and a little-used method for binding of sugars to ion exchange resins to yield a white crystalline product that is composed of approximately 96% cyclitols.

Materials and Methods

Procedures described are for a typical extraction and sample processing. The amounts of solvents were kept small relative to the amount of plant tissue with the thought of reducing cost instead of maximizing recovery of carbohydrates. For the work described here, fresh leaf tissue was used; however, dried, ground soybean leaves were an equally acceptable starting point (data not shown).

Fresh soybean (cv. Flint) leaves (blades plus petioles) from greenhouse-grown plants were collected and 325 g fresh weight was extracted in 50-g batches with 400 [+ or -] 20 mL of ethanol solvent "A" (700 g ethanol [L.sup.-1] of solution) for each batch. A Waring blendor was used to macerate the tissue. The green slurry was transferred to 250-mL centrifuge bottles and centrifuged at 22 000 g for 15 min. The green supernatant was collected and chilled and the residue was resuspended in a 50 [+ or -] 5 mL of ethanol solvent "B" (630 g ethanol [L.sup.-1] of solution); the slurry was than centrifuged a second time. The combined supernatants (about 3 L) were taken to dryness in two 1000-mL flasks by means of a rotary evaporator.

Solids in each of the two evaporating flasks were dissolved in a combination of 40 mL of triple deionized (TDI) water and 15 mL chloroform. This mixture was allowed to stand overnight at 2 [degrees] C. During this time, the chloroform separates from the upper aqueous fraction with only a thin emulsion layer between the aqueous and chloroform layers. Aliquots of the aqueous fraction were analyzed for carbohydrate composition by gas-liquid chromatography. The method involves the analysis of the oxime-TMS derivatives on a packed column of OV-17 (Streeter and Strimbu, 1998). The crude extract (total vol. 80 mL) from 325 g of fresh tissue was found to contain 1.8 g of pinitol, 110 mg of chiro-inositol, 72 mg of myo-inositol, 213 mg of glucose, 86 mg of fructose, and 604 mg of sucrose. These quantities do not represent complete extraction, but the two-step extraction probably results in recovery of >95 % of leaf carbohydrates, based on prior experience (Streeter and Strimbu, 1998).

Forty milliliters (wet vol.) each of Dowex 50 X-8 and Dowex l-X8 resin (both 200-400 mesh) were placed in 60-mL sintered glass funnels. (Columns of other configurations and types should give equivalent results.) Dowex 50 was converted to the [H.sup.+] form by passage of 5 volumes of 2 M HCl. The excess HCl was washed out of the resin with TDI water until the effluent was >pH 5. The Dowex 1 resin was converted to the O[H.sup.-] form by passage of 5 volumes of 2 M NaOH and the excess NaOH was washed out with C[O.sub.2]-free water until effluent had a pH <8. Use of C[O.sub.2]-free water is important because this resin has an affinity for carbonate that is 6-fold greater than the affinity for O[H.sup.-] (Dean, 1999). (C[O.sub.2]-free water is conveniently produced by boiling TDI water and allowing the water to cool in a sealed container.) Dowex 1 O[H.sup.-] should be prepared just before use because the resin will absorb substantial amounts of C[O.sub.2] from the air over time.

Forty milliliters of the crude leaf extract (aqueous layer over chloroform) was transferred to a beaker and stirred with gentle heating for 30 to 40 min to remove traces of chloroform. The mixture of leaf solutes was passed through the Dowex 50 [H.sup.+] column and the effluent was applied to the Dowex 1 O[H.sup.-] column positioned underneath. The procedure was carried out at room temperature. Following passage of the sample through the Dowex 50 column, 50 mL of C[O.sub.2] -free water was passed through the tandem columns. Then, the Dowex 50 column was removed and an additional 100 mL of C[O,sub.2]-free water was passed through the Dowex 1 column to elute the cyclitols. The effluent was reduced in volume by rotary evaporation, transferred to a freeze-drying vessel, frozen and freeze-dried. The white, crystalline product was weighed and weighed portions were dissolved in TDI water and analyzed for carbohydrate composition as described above.

Results and Discussion

Solvents other that ethanol solvent A can be used to extract cyclitols from soybean leaf tissue; for example, hot water extraction was compared in these studies. An advantage of water is that very little lipid was present in the extract. However, water-soluble proteins are major constituents of these extracts and there was sufficient lipid so that emulsions and foaming were serious problems when concentration of these extracts was attempted by rotary evaporation. Use of ethanol solvent A gives an essentially protein-free extract but leads to extraction of much of the lipid and most of the chlorophyll from green plant tissues.

However, after removal of the ethanol by evaporation, dissolving the crude solids in chloroform + water efficiently overcame this problem. That is, virtually all of the chlorophylls and lipids were partitioned into the chloroform layer because of their high relative solubility in chloroform and because the solubility of chloroform in water is extremely low (Dean, 1999). The upper aqueous layer was an amber color, presumably because of the presence of other nonchlorophyll pigments in the tissue, e.g., flavonoids. Whatever the composition of these pigments, they were efficiently bound to the Dowex 50 [H.sub.+], giving a clear effluent from the tandem ion exchange columns.

In the crude extract of soybean leaves studied here, cyclitols and sugars accounted for about 26% of dry solids prior to treatment of the extract with ion exchange resins (Table 1). Following passage through strong cation exchanger in the [H.sup.+] form and strong anion exchanger in the O[H.sup.-] form, sugars were not detectable and cyclitols accounted for about 93.5% of dry weight. Although some variation between batches was noted, passage of the crude extract from soybean leaves through the ion exchange resins provided roughly 100% recovery of cyclitols (Table 2). The recovery of cyclitols was checked only twice. Because this procedure was not intended to provide absolute quantification of compounds in the plant tissue, recovery was not a major concern as long as it was reasonable.

The product resulting from treatment with ion exchange resins was extremely hygroscopic, thus requiring that it be weighed immediately after removal from the freeze dryer. In an attempt to improve the handling of the product and to remove other impurities, the solid was dissolved in TDI water and mixed with activated charcoal in a ratio of 1 mg charcoal per 3 mg of solid. After filtration, the clear filtrate was again frozen and freeze-dried. The resulting product was much less hygroscopic although it did absorb enough water to become an amorphous solid if exposed to room air for several days. Lower S.E. values in Table 1 represent the fact that this material could be weighed with greater accuracy. Also, another 2.5% of impurities were removed to give a product that was 85.5% pinitol; 96% of dry weight was accounted for in the four compounds reported. The impurities removed may have been materials released from the resins at the high and low pH values used in the ion exchange purification step.

The composition of the remaining 4% of the white crystalline product remains unknown. Analysis of protein by a sensitive method (Smith et al., 1985) and quantities of solid product up to 7.5 mg per assay, indicated that protein could account for <0.1% of dry weight. The possible presence of maltodextrans was checked by incubation of 500 mg quantities of the product with amyloglucosidase for 1 h at 40 [degrees] C and GLC analysis of glucose concentration in the reaction mixture. No glucose was released from the crystalline product.

Although extraction with ethanol yields an extract containing an enormous variety of plant metabolites, suspension/solution of the solids in chloroform + water and separation of the two solvents in the cold provides an aqueous fraction that is efficiently fractionated using standard ion exchange resins. Most cations have 3- to 10-fold greater affinity than [H.sup.+] on the Dowex 50 resin and only [Li.sup.+] has a lower affinity (Dean, 1999). Thus, this resin is highly efficient in binding any component in the plant extract having a positive charge at low pH (Thompson et al., 1959). Likewise, Dowex 1 has a lower affinity for O[H.sup.-] ion than for any other anion tested (Dean, 1999) and, therefore, is highly efficient in removing negatively charged species from extracts. A key feature of the combination of methods used here is based on the observation that strong anionic exchange resins in the OH- form also bind sugars (Roseman et al., 1952). The results of Roseman et al. (1952) suggest that the binding may have something to do with the reducing nature of some sugars because the affinity of the Amberlite resin for glucose and fructose was substantially greater than the affinity of the resin for sucrose. However, prior experience (Streeter, 1985) and results reported here with Dowex strong anion exchanger indicate that all three sugars are completely removed from extracts, suggesting that sugar removal depends on the formation of a weak negative charge at the ring oxygen at high pH.

A 325-g sample of soybean leaves from six plants yielded approximately 1.8 g pinitol. Taking into account that our greenhouse-grown plants were about double the size of typical field-grown plants, and assuming a plant population of 320 000 plants per hectare, a hectare of soybean leaves would potentially yield about 50 kg of pinitol. Only a small portion of the total U.S. soybean crop would provide an essentially unlimited supply of pinitol. Pinitol is found throughout the soybean plant with highest concentrations in petioles and stems (Phillips and Smith, 1974); we chose leaves because they constitute a major portion of the plant mass and are relatively easy to extract. In a commercial application of the protocol reported, extraction of whole shoots would probably be more practical.

The product obtained in these studies was not pure; only 96% of dry weight was accounted for. Thus, for commercial use, either further purification or demonstration that the product can be safely consumed by humans would be necessary. However, the cyclitols other than pinitol in the crystalline product are probably not of great concern with respect to consumption by animals. myo-Inositol is a constituent of all living cells and would be readily digested or assimilated. The cyclitol D-chiroinositol is thought also to provide a lowering of blood glucose (Ortmeyer et al., 1993) and could have therapeutic effects similar to pinitol. The other impurity, ononitol, a methylated derivative of myo-inositol, is very similar in structure and properties to pinitol, and should not be harmful. It is interesting that ononitol constitutes about 1.9% of the total cyclitols in the commercial pinitol product, based on our replicate analyses of a sample of "o-Pinitol" from the General Nutrition Center, Inc. The sample of this commercial product also contained about 1.2% myo-inositol but no v-chiro-inositol. The commercial product is derived from Bougainvillea leaves and our analysis of a sample of Bougainvillea leaves obtained from a local greenhouse also lacked any trace of D-chiro-inositol.
Table 1. Carbohydrate composition of soybean leaf extracts before
and after passage through ion exchange columns.

Fraction              Pinitol   Inositol   Ononitol   myo-Inositol

                 -- mg/100 mg product ([dagger]) --

Crude          mean    17.6       0.64        --
                                           dagger])       0.86
               S.E.     0.1       0.01        --          0.01
Purified       mean    83.3       5.02       1.95         3.26
 ([section])   S.E.     1.1       0.07       0.03         0.06
Charcoal       mean    85.5       5.21       1.93         3.40
 ([para-       S.E.     0.6       0.05       0.02         0.03

Fraction              Fructose   Glucose   Sucrose   Sum

            -- mg/100 mg product ([dagger]) --

Crude          mean      --
                       dagger])   1.76      5.16     26.0
               S.E.      --       0.02      0.08     --
Purified       mean       0       0         0        93.5
 ([section])   S.E.      --       --        --       --
Charcoal       mean       0       0         0        96.0
 ([para-       S.E.

([dagger]) Data are based on three analyses of three weighed
portions of a freeze-dried crystalline sample, i.e., 9 observations

([double dagger]) Ononitol and fructose were both present in
very small amounts and the two small peaks did not integrate
separately. Upon removal of fructose, ononitol could be quantified.

([section]) Samples from two resin purification runs gave very
similar results and the combined products were analyzed for the
results shown here. The sugars fructose, glucose and sucrose
were below detectable limits.

([paragraph]) "Purified" product treated with activated charcoal.
Table 2. Recovery of cyclitols applied to the resin columns.

                       Trial A ([dagger])

Cyclitol         Applied   Recovered   ([section])

                       -- mg --             %

Pinitol           499.0      484.1          97
chiro-Inositol     27.7       31.4         113
myo-Inositol       17.9       18.6         104

                    Trial B ([double dagger])

Cyclitol         Applied   Recovered   ([section])

                       -- mg --             %

Pinitol           849.0      839.9          99
chiro-Inositol     48.4       48.6         100
myo-Inositol       42.2       33.0          78

([dagger]) Data for Trial A are average values based on two analyses
of the aqueous fraction applied to the resin columns and two weighed
portions of the crystalline product.

([double dagger]) For Trial B, samples for analysis were the same
except three analyses of applied and recovered samples were conducted.

([section]) Recovery of sugars applied to the resin columns was
essentially zero; see Table 1.


I thank Monty Mongomery and Jim Sonowski for technical assistance.


Dean, J.A., ed. 1999. Lange's handbook of chemistry, 15th ed. McGraw-Hill, New York.

Hansen, B.C., and H.K. Ortmeyer. 1996. Inositols--Potential roles in insulin action and in diabetes: Evidence from insulin-resistant nonhuman primates, p. 333-348. In E. Shafrir (ed.) Lessons from animal diabetes. Birkhauser, Boston.

Narayanan, C.R., D.D. Joshi A.M. Mujumdar, and V.V. Dhekne. 1987. Pinitol--A new anti-diabetic compound from the leaves of Bougainvillea spectabilis. Curr. Sci. 56:139-141.

Ortmeyer, H. K., L.C. Huang, L.U. Zhang, B.C. Hansen, and J. Larner. 1993. Chiroinositol deficiency and insulin resistance. II. Acute effects of D-chiroinositol administration in streptozotocin-diabetic rats, normal rats given a glucose load, and spontaneously insulin-resistant Resus monkeys. Endocrinology 132:646-651.

Phillips, D.V., and A.E. Smith. 1974. Soluble carbohydrates in soybean. Can. J. Bot. 52:2447-2452.

Phillips, D.V., D.E. Dougherty, and A.E. Smith. 1982. Cyclitols in soybean. J. Agric. Food Chem. 30:456-458.

Roseman, S., R.H. Abeles, and A. Dorfman. 1952. Behavior of carbohydrates toward strongly basic ion-exchange resins. Arch. Biochem. Biophys. 36:232-233.

Smith, P. K., R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, and D.C. Klenk. 1985. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150:76-85.

Streeter, J.G. 1985. Identification and distribution of ononitol in nodules of Pisum sativum and Glycine max. Phytochemistry 24:174-176.

Streeter, J.G., and C.E. Strimbu. 1998. Simultaneous extraction and derivatization of carbohydrates from green plant tissues for analysis by gas-liquid chromatography. Anal. Biochem. 259:253-257.

Thompson, J.F., C.J. Morris, and R.K. Gering. 1959. Purification of plant amino acids for paper chromatography. Anal. Chem. 31:1028-1031.
COPYRIGHT 2001 Crop Science Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Streeter, John G.
Publication:Crop Science
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Nov 1, 2001
Previous Article:Endophyte infection can contribute to aluminum tolerance in fine fescues. (Notes).
Next Article:Crop Responses and Adaptations to Temperature Stress. (Book Reviews).

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters