Quantitative and qualitative analysis of sterols / sterolins and hypoxoside contents of three Hypoxis ( African potato ) spp .

The glycoside, hypoxoside, identified and isolated from the corms of the African potato (Hypoxis hemerocallidea) has shown promising anticancer activities. The African potato is used as an African traditional medicine for its nutritional and medicinal properties. Most research has been carried out on H. hemerocallidea (formerly known as H. rooperi), with very little or nothing on other Hypoxis spp. Thin layer chromatography (TLC) was used to confirm the presence of sterols/sterolins, whereas a GC method was developed to identify and quantify sterols (especially -sitosterol) in chloroform extracts of H. hemerocallidea, H. stellipilis and H. sobolifera var. sobolifera. High performance liquid chromatography (HPLC) was used to identify and quantify hypoxoside content in these Hypoxis spp. TLC results showed that H. sobolifera var. sobolifera contained the most sterols and sterolins compared to the other two Hypoxis spp. Gas chromatography (GC) results show that -sitosterol and campesterol were the two main phytosterols present in the Hypoxis extracts. H. sobolifera var. sobolifera and H. hemerocallidea contained the most -sitosterol and hypoxoside, respectively. H. sobolifera and H. hemerocallidea contained 74.69 g of -sitosterol and 12.27 g of hypoxoside per 5 mg of chloroform extracts, respectively. These results show a significant difference in the sterol/sterolin and hypoxoside contents between species of the genus Hypoxis, which may influence their degree of biological activities.


INTRODUCTION
Hypoxis hemerocallidea Fisch., C.A. Mey. and Avé-Lall., syn. H. rooperi, (commonly known as the African potato; belonging to the family Hypoxidaceae) topped the list of the 60 most frequently traded plant species in the Eastern Cape, South Africa, when studies were conducted among street traders, traditional healers, storeowners and clinic patients (Dold and Cocks, 2002).
Sterols are amphiphilic molecules consisting of hydroxyl groups forming the hydrophilic heads and sterane skeletons with side chains forming the hydrophobic tails (Heldt, 2005). Cholesterol (C 27 H 45 OH) is the main sterol found in mammals where it plays an important role in the structure and function of cell membranes, production of bile, as precursor of hormones and a role in the immune system (De Brabander et al., 2007). Sterols found in plants are known as phytosterols and over 250 phytosterols and their related compounds have been identified (De Brabander et al., 2007) in foods like plant oils, nuts, seeds, cereals, fruits and vegetables Ostlund, 2002). Phytosterols differ from cholesterol in being alkylated at C-24 with C 1 or C 2 substituents (Buchanan et al., 2000). In nature, plants contain sterols with their associated sterolins (glucosides), which are easily destroyed by glycosidic enzymes (Pegel, 1976).
Some of the compounds (hypoxoside, -sitosterol and other sterols/sterolins) that may be responsible for the medicinal properties of Hypoxis have been identified in H. hemerocallidea but the quantity of these in different Hypoxis spp. is unknown. Various species of Hypoxis are sold and used indiscriminately without any evidence that they contain equal quantities of sterols/sterolins and hypoxoside. The main aims of this study were to identify and quantify the sterol/sterolin and hypoxoside contents of H. hemerocallidea, H. stellipilis Ker Gawl. and H. sobolifera var sobolifera (Jacq.) Nel.  Boukes et al. 1625 collected near Plettenberg Bay in the Southern Cape, South Africa. Corms of the three Hypoxis spp. were planted in the same soil type and exposed to equal amounts of sunlight, humidity and water for at least six months before they were harvested and used fresh. The plants were identified by Y. Singh from the South African National Biodiversity Institute (SANBI) and voucher specimens were deposited in the Nelson Mandela Metropolitan University herbarium.

Isolation of hypoxoside and sterols/sterolins
Corms of H. hemerocallidea, H. stellipilis and H. sabolifera were washed, peeled, grated and crushed using a mortar and pestle. Chloroform was added to the plant material in a 1:1 (v:w) ratio, vortexed (5 min), extracted (15 min) and centrifuged (3645 x g for 5 min) at room temperature. The supernatant was removed and the extracting method repeated with the same plant material. The chloroform was evaporated in vacuo and mass of extracts determined. After mass determination the extracts were redissolved in chloroform until further use.

TLC
-Sitosterol (2 g) and stigmasterol (2 g) standards (Supelco, USA) and Hypoxis extracts (500 g) dissolved in chloroform were spotted onto 20 x 20 cm silica coated aluminum plates (Merck, Germany) and air dried. Chromatogram tanks were equilibrated for one hour using toluene -diethyl ether (40:40, v/v) and chloroformethyl acetate -formic acid (5:4:1, v/v/v) as mobile phases for sterol and sterolin identification, respectively. TLC plates were developed for ±20 min or until the solvent front was ±1 cm from the top of the plate. Detection of sterols/sterolins was performed as described by Scott and Springfield (2004). In brief, TLC plates were dried at room temperature and developed by firstly dipping into a solution containing 5% sulfuric acid in 96% ethanol for 15 s followed by a solution containing 1% vanillin in 96% ethanol for 15 s and dried at room temperature. Once dried, plates were heated at 80 -100 o C for five min. Photos of the developed TLC plates were taken with the AlphaImager TM 3400 (Alpha Innotech.).

GC
GC analysis of sterols was performed using a Thermo Finnigan Focus gas chromatograph equipped with a FID and an Autoinjector AI3000, with Delta Chromatography 5.0 software. The column used for GC separation was a SAC TM -5 capillary column (Supelco, 30 m × 0.25 mm i.d. × 0.25 m film thickness). The thermal conditions were: 80 o C for 2 min; 10 o C.min -1 to 300 o C; 300 o C for 14 min. The carrier gas was He (1 ml.min -1 constant flow) and the injection volume was 2 l (splitless). A sterol mixture (containing 100 g/ml of each sterol standard) was spiked individually to identify the peaks. An increase in peak area was used as criteria to identify each peak.

HPLC
High performance liquid chromatography (HPLC) analysis of hypoxoside was performed using a Beckman System Gold high performance liquid chromatograph equipped with Solvent Module 128, Diode Array Detector Module 169. The column used for HPLC separation was a Nucleosil C18 column (Supelco, 5 m, 150 × 4.6 mm i.d.). Detection of hypoxoside was performed as described by . In brief, acetonitrile -water (20:80, v/v) was used as mobile phase in isocratic mode at a flow rate of 1 ml/min and the injection volume was 10 l. Detection was achieved in the

RESULTS AND DISCUSSION
Chloroform has been shown to be very effective in dissolving sterols  due to its non-polar nature. The presence of sterols and sterolins in the chloroform extracts of H. hemerocallidea, H. stellipilis and H. sobolifera var sobolifera was confirmed via TLC. Modification of the mobile phase, toluene -diethyl ether -1.75 M acetic acid (1:1:1, v/v/v) used by Scott and Springfield (2004), to toluene -diethyl ether (40:40, v/v) resulted in better sterol separation ( Figure 1A). The spots of the sterol standards (stigmasterol and -sitosterol) had the same Rf value of 0.53. Cholesterol, campesterol, desmosterol, ergosterol, fucosterol and stigmastenol migrated the same distance on the TLC plates (data not shown), with the spots differing only in colour ranging from pink to blue. This made it impossible to identify and quantify individual sterols in the Hypoxis extracts, using TLC. The mobile phase used for sterolin identification consisted of chloroform -ethyl acetate -formic acid (5:4:1, v/v/v). The sterolins in the Hypoxis extracts could not be identified or quantified due to the unavailability of sterolin standards ( Figure 1B). From the results obtained, different sterolins (based on Rf values) were detected and differences in sterolin composition could clearly be seen between the three Hypoxis species (Figure 1).
From GC analysis, it was clear that -sitosterol was the main phytosterol found in the chloroform extracts of H. hemerocallidea, H. stellipilis and H. sobolifera. Trace amounts (<10 g per 5 mg of Hypoxis extract) of campesterol were also found in all three Hypoxis spp. extracts, whereas trace amounts of desmosterol/ ergosterol, stigmasterol, stigmastenol were found only in certain Hypoxis spp. A standard curve of -sitosterol concentration (ranging between 10 -100 g/mL) as a function of peak height (R 2 = 0.9531, R T = 30.6 min) was used to quantify -sitosterol content. H. sobolifera contained the most -sitosterol with ±2.5 and 7.5 times more -sitosterol compared to H. hemerocallidea and H. stellipilis, respectively (Table 1).
The presence of -sitosterol and campesterol as the two major phytosterols in Hypoxis correspond to published data (Moghadasian, 2000;Pegel, 1976). According to a minireview by Moghadasian (2000), 95% of dietary phytosterols consist of sitosterol and campesterol (approximately 65 and 30%, respectively), whereas the other phytosterols (mainly stigmasterol) and stanols make up the other 5%. This is the first time that GC was used to identify and quantify the presence of sterols in Hypoxis extracts.  have used high performance liquid chromatography to determine the presence of -sitosterol, stigmasterol and stigmastenol in commercially available oral dosage forms reported to contain material or extracts of Hypoxis. Using the SAC-5 TM capillary column eliminated time-consuming preparation steps, for example extraction of lipid fraction from sample material, saponification (alkaline hydrolysis), extraction of nonsaponifiables and derivatization of the sterol standards and Hypoxis extracts .
The presence of hypoxoside has been identified in several South African species of Hypoxis (Nicoletti et al., 1992) including H. hemerocallidea , but not in H. stellipilis or H. sobolifera. The HPLC method described by  was used to quantify hypoxoside content in the three Hypoxis spp ( Figure  3). Hypoxoside was detected after 12.5 minutes and quantified from a standard curve of hypoxoside concentration (ranging between 5 -100 g/mL) as a function of peak area (R 2 = 0.9971, R T = 30.6). Of the three Hypoxis species tested for hypoxoside    (Table 2). Since chloroform, the solvent used in this study, is not the best solvent for hypoxoside extraction, the presence of this glycoside in H. sobolifera was investigated using more polar solvents. A water extract of H. sobolifera has shown no hypoxoside content, whereas ethanol, metha-  (Pegel, 1979) and methanol  to extract hypoxoside from H. hemerocallidea. Traditional healers/herbalists use water and boiling to make Hypoxis extracts, which may yield hypoxoside and sterolins  (Pegel, 1976). Due to the polar nature of hypoxoside it would be better to use more polar solvents for extractions in the future, if more emphasis is placed on the effect of hypoxoside in the Hypoxis extracts. This is the first time that hypoxoside and sterol contents were quantified in H. stellipilis and H. sobolifera, and as far as we know in H. hemerocallidea. Both the sterol and hypoxoside contents were shown to vary between the three species. Differences in the hypoxoside and sterol/sterolin contents of the three Hypoxis spp. investigated may explain the differences in anticancer activity (unpublished data) obtained against certain cancer cell lines. Hypoxis species are used indiscriminately in traditional medicine and are sold under the common name 'African potato' in herbal shops. Further investigation is required to determine the implications of these findings in the uses of Hypoxis as a traditional remedy. Consumption of the three different Hypoxis species, which have different sterols/sterolins and hypoxoside content, may have adverse or favorable effects depending on the concentration of extract consumed. Sterols/sterolins and hypoxoside may have synergistic effects, which need to be investigated.