A phytochemical and pharmacological investigation of indigenous agathosma species

Abstract

As part of an investigation of the biological activities of South African plants and due to their extensive traditional use and lack of scientific evidence, a phytochemical and pharmacological investigation was performed on 17 indigenous Agathosma species (19 samples). The chemical composition of the essential oils was determined using gas chromatography coupled to mass spectroscopy (GC-MS). Analysis resulted in the identification of 333 compounds. To evaluate the chemical similarities and differences, cluster analysis was used to assess the essential oil composition of the samples. The results showed qualitative and quantitative differences amongst the taxa. The essential oils of Agathosma hirsuta and A. zwartbergense are particularly rich in citronellal, hence they are tightly clustered in the dendrogram obtained from the cluster analysis. Linalool, myrcene and limonene are the major constituents of both A. capensis (Gamka) and A. capensis (Besemfontein). Qualitative and quantitative differences are noted in the chemical compositions of the leaf oils of Agathosma capensis (Gamka) and A. capensis (Besemfontein). Agathosma arida and A. lanata are united in a single cluster due to the compounds β-pinene, linalool and spathulenol being major components in both species. The presence of 1,8-cineole in large quantities in both Agathosma namaquensis (23.5%) and A. ovalifolia (9.7%), unites them in a single cluster. A wide chemical variability for the essential oils of indigenous Agathosma species has been demonstrated. There was considerable variation in the percentage oil yield of the essential oils. Agathosma hirsuta produced the highest yield (1.15%) whilst A. ovalifolia produced the lowest yield (0.16%). vi Previous studies have revealed that the coumarin and flavonoid components of Agathosma species are responsible for their biological activities. High performance liquid chromatography (HPLC) was used to document the non-volatile composition of Agathosma species and to establish if phenolic patterns were present amongst the species. All species were found to be rich in flavonoids (i.e. flavones and flavonols). Many of the compounds detected were common to most of the species. A pure coumarin, puberulin, was identified in the diethyl ether extract of Agathosma ovata (round-leaf) and detected in the dichloromethane and methanol (1:1) extract of A. namaquensis. Agathosma species have been used traditionally to treat a wide variety of infections. They has been used as a cough remedy, for the treatment of colds and flu, kidney and urinary tract infections, for the treatment of cholera and other stomach ailments. Based on the extensive use and lack of scientific evidence, a study was embarked upon to determine its bioactivity. Using the disc diffusion assay as a preliminary screening and thereafter the minimum inhibitory concentration (MIC) assay, the antimicrobial activity of the essential oils and non-volatile compounds was assessed on two Gram-positive bacteria, Staphylococcus aureus and Bacillus cereus, one Gram-negative bacterium, Klebsiella pneumoniae, and one yeast, Candida albicans. All of the extracts proved to be active against the four pathogens tested with the exception of Agathosma bathii which showed poor activity against Klebsiella pneumoniae (MIC value of 32mg/ml). The extracts exhibited stronger activity against the pathogens as compared to the essential oils. Both the essential oils and extracts exhibited higher activity towards the Gram-positive bacteria than the Gram-negative bacterium, with the extract of Agathosma ovata (round-leaf) displaying the greatest vii activity against Staphylococcus aureus (MIC value of 0.156mg/ml) and Bacillus cereus (MIC value of 0.125mg/ml). The extract of Agathosma parva displayed the greatest activity against Candida albicans and Klebsiella pneumoniae (MIC value of 1.5mg/ml). Amongst the essential oils, Agathosma pungens proved to be the most active against the Gram-positive pathogen, Bacillus cereus (MIC value of 3mg/ml). Agathosma collina was the most active against Candida albicans (MIC value of 3mg/ml) whilst A. zwartbergense proved to be the least active against most of the tested pathogens. The antimicrobial activity of the essential oils may be ascribed to oxygenated constituents, such as 1,8-cineole, linalool and carvacrol. The activity of the extracts may be ascribed to constituents such as flavonoids, coumarins and alkaloids. Due to the availability and accessibility of Agathosma ovata, a seasonal variation study was performed on the chemical composition of the essential oils and how this may impact on the antimicrobial activity. Furthermore, this species has recently been earmarked for commercial development by the flavour and fragrance industry and information on variability is required to establish the harvesting protocol. Ten samples were harvested in total. There was a substantial variation in the oil yield throughout the year, ranging from 0.23% in early Spring to 0.85% in late Autumn. A higher yield was observed during the flowering season as compared to the non-flowering season. Oil yields were low during Summer (0.44%-0.48%) which may have been due to the low oil content in stems and higher proportion of stems after flowering. The proportion of oil-rich green leaves also decreased markedly, hence affecting the yield. Overall the yields were dependant on the season harvested and proportion of plant parts distilled. viii The chemical composition of the essential oils was determined using GC-MS and resulted in the identification of 145 compounds in 10 of the samples. All samples contained a large number of common monoterpenes and had very similar compositions, with minor quantitative variation. Some components common to all samples include: sabinene, p-cymene, β-pinene, α-pinene, α-thujene, myrcene, limonene, linalool and terpinen-4-ol. Sabinene was found to be the most dominant component in all samples, ranging between 25.6% and 44.4%. Myrcene levels dropped sharply between the beginning of Spring and end of Summer, from 14.9% to 1.0%. β-pinene followed a similar trend, peaking during Spring and decreasing during the Summer months. The lowest levels of linalool (4.3%), myrcene (1.0%), β-pinene (3.9%), limonene (1.9%) and sabinene (25.6%), occurred during the Summer months when the temperatures were high. There was a Springtime increase in the levels of β- pinene, terpinen-4-ol, linalool, sabinene, limonene and p-cymene in the non-flowering Agathosma ovata. These changes may have been due to the higher proportion of young leaves during Spring, which may have oil compositions slightly different to those of mature leaves. A rare thiol derivative (tr) that could not be identified was detected in the March sample. Many of the changes were associated with flowering and the results obtained reveal that the chemical composition of the essential oil of Agathosma ovata is subject to seasonal variation. Using the MIC assay, the antimicrobial activity of the essential oils was assessed on two Gram-positive bacteria, Staphylococcus aureus and Bacillus cereus, one Gramnegative bacterium, Klebsiella pneumoniae, and one yeast, Candida albicans. The study demonstrated differences in the potency of antimicrobial activity of the essential oils distilled each month. The Winter samples were more active against Bacillus cereus, Staphylococcus aureus and Klebsiella pneumoniae. Activity in mid Spring ix was greater against Staphylococcus aureus (MIC value of 3mg/ml) and Klebsiella pneumoniae (MIC value of 3mg/ml), whilst activity decreased in Summer. There was a correlation between the concentrations of the active compounds each month and the oils antimicrobial activity. The results reveal that the antimicrobial activity of the essential oil of Agathosma ovata may not depend on the level of one component but rather the ratio of several components. ‘Buchu’ has been used traditionally as a general tonic and medicine. Tonics generally have a high anti-oxidant content in order to promote the overall well-being of the user. The anti-oxidant properties of the essential oils and non-volatile compounds was investigated using the 2, 2-diphenyl-1-picrylhydrazyl (DPPH) and 2, 2'-azinobis(3- ethylbenzothiazoline-6-sulfonic acid) (ABTS) assays. Only the non-volatile compounds exhibited activity. Their activities may be ascribed to the flavonoid components. Most of the species portrayed moderate to poor activity in the DPPH assay with the exception of Agathosma capensis (Gamka) (IC50 value of 24.08 + 4.42μg/ml) and A. pubigera (IC50 value of 35.61 + 0.86μg/ml) which were two of the most active species, although their activities were inferior when compared to vitamin C. The results from the ABTS assay differed from that of the DPPH assay. All extracts showed greater activity in this assay with Agathosma namaquensis (IC50 value of 15.66 ± 4.57μg/ml) and A. capensis (Besemfontein) (IC50 value of 19.84 ± 0.09μg/ml) being the most active species. This may be due to the ABTS assay having an additional reaction system. ‘Buchu’ has been used traditionally as an antipyretic, topically for the treatment of burns and wounds and for the relief of rheumatism, gout and bruises. The antix inflammatory activity of the essential oils and non-volatile compounds was assessed using the 5-lipoxygenase (LOX) assay. Only the essential oils exhibited activity. All proved to be active with the exception of Agathosma stipitata which was UV active and caused interference. This was due to its major compounds neral (39.9%) and geranial (10.1%) which absorbed strongly at 234 nm and hence rendered its spectrophotometric measurement impossible. The essential oil of Agathosma collina displayed the most promising activity (IC50 value of 25.98 ± 1.83μg/ml). It is well known that many herbal medicines can have adverse effects, in which case it is necessary to evaluate the benefit-risk profile. The toxic effects of Agathosma species have been poorly studied and no information is available in this regard. Hence the toxicity profile of the non-volatile compounds and essential oils was assessed on transformed human kidney epithelium (Graham) cells using the microculture tetrazolium (MTT) cellular viability assay. The extracts of Agathosma lanata (IC50 value of 26.17 ± 9.58μg/ml) and A. ovata (round-leaf) (IC50 value of 25.20 ± 6.30μg/ml) proved to be the most toxic, whilst the extracts of Agathosma bathii, A. capensis (Besemfontein), A. betulina, A. crenulata and A. namaquensis did not prove to be toxic at the concentrations tested. Serial dilutions displayed different inhibitions of cell growth and the species proved to be toxic in a dose-dependant manner. The essential oils of all 19 species proved to be much more toxic (IC50 values < 0.0001μg/ml) than a plant-derived compound that is considered relatively safe, namely quinine (IC50 value of 136.06 ± 4.06μg/ml). The toxicities of the essential oils may be due to compounds like methyl chavicol, eugenol, methyl eugenol, pulegone and methyl salicylate whilst the toxicities of the extracts may be due to the alkaloid and coumarin components.