High Performance Liquid Chromatographic Separation And Identification Of A Toxic Fraction From Aloe Barbadensis Miller Leaf Gel Using The Artemia Nauplii Bioassay
I Cock, D Ruebhart
aloe barbadensis miller, aloe vera, anthraquinone, artemia
I Cock, D Ruebhart. High Performance Liquid Chromatographic Separation And Identification Of A Toxic Fraction From Aloe Barbadensis Miller Leaf Gel Using The Artemia Nauplii Bioassay. The Internet Journal of Toxicology. 2007 Volume 4 Number 2.
The current study was undertaken to detect toxicity in purified Aloe vera gel fractions using the Artemia nauplii lethality bioassay, thereby allowing for the identification of compounds of interest for further investigation. The work presented here therefore seeks to not only detect toxicity in gel extracts, but also to assign this toxicity to individual fractions. Methanol extraction and RP-HPLC were used to purify fractions from Aloe vera gel leading to the isolation of 13 major components. Of these 13 fractions tested using the Artemia nauplii lethality bioassay, one proved to be toxic with a 24 h LC50 of 435 µg mL-1. Compared to the tested reference toxins, this Aloe vera gel fraction was approximately three times more toxic than the organophosphate insecticide Mevinphos (24 h LC50 1336 µg mL-1) and approximately six fold less toxic than potassium dichromate (LC50 73 µg mL-1). Of particular interest was the rapid onset of toxicity against the Artemia nauplii. Dilutions of the gel extract were capable of causing 100% mortality within 90 min. The isolated fraction induced 100% mortality within 120 min at a concentration of approximately 800 µg mL-1. In contrast, greater than 36 h was required for Mevinphos and 18 h for potassium dichromate to produce 100% mortality, even at high concentrations (2000 µg mL-1 or 800 µg mL-1 respectively). These results confirm the presence of toxic compounds in Aloe vera gel. As this bioassay correlates well with pesticidal activity and cytotoxic activity in some human tumours, this bioactive fraction may hold promise as a natural pesticide and/or antitumoral agent.
The mucilaginous gel from
In addition to the multitude of positive effects attributed to Aloe vera gel, there have also been reports of negative actions of gel components [ 3,9,10,11,12 ]. The gel is known to contain a large number of anthraquinones including aloe-emodin and aloin [ 13,14 ]. Some reports have shown these compounds to be mutagenic in
However, much of the work in this area has used whole gel or crude extracts, providing conflicting and difficult to interpret results. For example, Aloe vera gel is known to contain compounds that stimulate cell proliferation [ 19 ] as well as containing cytotoxic compounds [ 9,11,12 ]. Crude extracts would be expected to contain both these classes of compounds making interpretation of the bioactivities of these extracts difficult to understand and account for conflicting reports.
Materials and Methods
Fresh clean whole
Preparation of Crude Extract
1 g of lyophilised
HPLC Separation of Extract Components
The extract was analysed and further fractionated by RP-HPLC. All of the equipment was Shimadzu. The system consisted of twin LC-10AT pumps, a DGU-12A degasser, a SIL-10AD automatic injector using a 20 µL injector loop and a SPD-M10A diode array detector, all under the control of a SCL-10A system controller. Detection was monitored at 210 nm. All solvents were of HPLC grade and were obtained from Lab-Scan Australia.
HPLC separations were performed on a Spherisorb C18 column (50 mm × 4.6 mm). 20 µL samples of Aloe gel extract were injected and chromatographed using a gradient from 20% methanol to 60% methanol as follows: 2 min isocratically at 20% methanol followed by an 8 min gradient to 40% methanol. This was followed by isocratic elution at 40% methanol for a further 5 min. The methanol was increased to 60% over a further 10 min. The column was washed with 100% methanol before re-equilibrating to 20% methanol for further chromatograms. Samples from multiple chromatograms (10 repeats) were collected and pooled. These samples were dried by rotary evaporation in an Eppendorf Concentrator 5301 and were resuspended in 1 mL distilled water and stored at 4 ° C for further analysis.
Anthrone Assay for Carbohydrate Content
The anthrone assay was performed as described by Dische [ 20 ] with the following modifications. 0.2% anthrone was prepared by dissolving pure anthrone (Chem-Supply, Australia) in concentrated H2SO4 (AR grade, Unilab). Anthrone reagent was prepared fresh for each assay. D-Mannose (Chem-Supply, Australia, AR grade) was diluted by serial dilution in the range 1 mg mL -1 - 0.032 mg mL -1 and used as a standard. 50 µL of standard dilutions or of samples was added to wells of a 96 well plate. 100 µl of anthrone reagent was added to each well and mixed. The assays were incubated for 15 min at 22 ° C and colour development was measured at 630 nm using a microplate reader (Biotrak). All determinations were performed in at least triplicate.
Total Polyphenols Assay
Determination of the polyphenol content of the extract and HPLC purified compounds was performed by the method of Singleton and Rossi [ 21 ]. This method has been routinely used to determine the total polyphenolic levels of plant extracts [ 22 ]. Gallic acid (Sigma) was used as a phenol standard. Aloe extract was diluted 1 in 10 for the assay. HPLC fractions were tested undiluted. 100 µL of the standards, the diluted extract and the HPLC fractions were added to 2 mL of 2% Na2CO3 and 200 µl of Folin-Ciocalteu reagent (Lab-Chem) was added. The tubes were incubated for 30 min at 22 ° C. The absorbance was measured at 720 nm. All determinations were performed in at least triplicate.
Reference Toxins for Biological Screening
Potassium dichromate (K2Cr2O7) (AR grade, Chem-Supply, Australia) was prepared as a 1.6 mg mL -1 solution in distilled water and was serially diluted in synthetic seawater for use in the
Toxicity was tested using the
Lyophilisation of 100 mL of Aloe vera gel by rotary evaporation produced approximately 1g of dried fraction (1% of original weight). When further processed with methanol extraction and drying by rotary evaporation, the net result was 391 mg of dried extracted material. Resuspension of the dried fraction in 1 mL distilled water resulted in 1.2 mL of concentrated extract. This equated to an approximate 80 fold concentration of the extractable, non-volatile compounds of the Aloe vera gel compared to the original gel.
Quantitative analysis of the carbohydrate content of the extract performed using a modified anthrone assay, revealed that the extract had 201 mg total carbohydrate at a concentration of 167.5 mg mL -1 . The polyphenol content, determined by the method of Singleton and Rossi [ 21 ], of the Aloe vera extract was 46.8 mg total extractable polyphenolics.
Previous reports [ 26,27 ] express
The LC50 of the extract (1443 µg mL -1 ) was similar to the Mevinphos LC50 (1336 µg mL -1 ) at 24 h demonstrating its toxicity. However, potassium dichromate had a LC50 of 73 µg mL -1 at 24 h, nearly twenty fold more toxic than the extract. The sensitivity of the
The HPLC gradient elution system used in this study facilitated the simultaneous separation of Aloe vera components of varying hydrophobicities. Figure 2a shows the typical HPLC profile of the fractionated methanol extract.
The HPLC fractions were evaluated by chemical methods to study their chemical nature. As shown in Figure 2b, the majority of the carbohydrate components of the extract eluted early in the chromatogram. Figure 2c shows the distribution of polyphenolic compounds across the HPLC profile. Polyphenolic molecules were seen to be widely distributed with most present in the latter half of the chromatogram.
The 13 HPLC separated fractions were tested for toxicity using the using the
The current study demonstrates the ability of Aloe vera gel extract and a RP-HPLC separated fraction to induce mortality in
Aloe vera leaf gel extract was found to have high levels of extractable carbohydrates. 201 mg of total carbohydrate were extracted from 1 g dried leaf gel (approximately 20% of the total dried weight). This equates to over 50% of the total extractable solids (391 mg) extracted from the original 1 g of Aloe vera leaf gel. The sugar composition was not analysed in this study but would be assumed to be high in β-(1,4)- linked polymannose (acemannan). Previous studies have shown that acemannan is the major fraction from Aloe vera leaf gel [ 29 ]. Acemannan is claimed to have several important therapeutic properties including acceleration of wound healing [ 7 ], immune stimulation [ 3,30 ] and antiviral effects [ 5 ].
Aloe vera leaf gel also had high levels of extractable polyphenolic compounds. 46.8mg of total phenolics were extracted from 1 g of dried Aloe vera leaf gel (approximately 4.7% of the total dried weight of the leaf gel). This equates to approximately 12% of the total solids extracted from the dried leaf gel (391 mg). This is comparable to other known sources of polyphenolics. For example,
The toxic fraction obtained from RP-HPLC contains both carbohydrate and polyphenolics. This could indicate a mixture of compounds in this fraction or the presence of a glycosylated polyphenolic compound. When dried, Fraction 1 was a crystalline solid with an orange red colour with a slightly tacky consistency. The methanolic solution of the fraction absorbed UV light between 250 and 290 nm with a maxima at 265 nm in the UV region. Anthraquinones are characterised by their orange red colour and their absorbances in the UVB range. Hirata and Suga [ 32 ] have listed the UV absorbance peaks for aloe emodin as 221, 253, 266 and 289 nm and the absorbance peaks of aloin as 250-290 nm with a peak at 260 nm. A more recent paper [ 13 ] has reported UV absorbance peaks for various anthraquinones as 217-220 nm and 265-270 nm. Anthraquinones are known to be cytotoxic and have been shown to induce apoptosis in human lung squamous cell carcinoma [ 14 ]. Fraction 1 may therefore contain an anthraquinone moiety. Aloe emodin (a glycosylated anthraquinone) has been reported [ 29 ] to be one of the major constituents of the gel. Thus this compound may be present in Fraction 1. Definitive structural characterisation of this fraction was not possible. Electron ionisation mass spectroscopy (unpublished results) show complicated patterns that indicate the presence of multiple molecular species. More work is necessary to further characterise the molecular composition of this fraction.
The current report demonstrates the toxicity in the
The authors wish to thank John Gorringe of Aloe Wellness Australia Pty Ltd for the gift of