The contribution of natural products particularly medicinal plants to discovery of novel drugs have been well highlighted [1] and Newman et al [2] reported that research on natural products accounted for about 48 % of all natural product-derived compounds between 1981 and 2002. Furthermore, Jones et al [3] revealed that majority of plant-related drugs on the pharmacy shelf as at 1993 actually emanated from ethnomedicinal application of most species that were indicated for the same therapeutic uses. Crude herbs which have long been used as the basis for traditional medicines worldwide are still invaluable now [4]. The role of traditional medicine in mitigating disease burden and poverty in the African continent has been emphasized and extended to include increasing the economic well-being and healthcare accessibility of communities.

Lobelia commonly known as Indian tobacco is a genus of flowering plants comprising about 415 species (family: Campanulaceae) with subtle distribution in tropical-warm temperate regions of the world [5]. L. inflata is regarded as the most important variety in this family consisting of several species including L. cardinalis, L. erinus, L. spitaca, L. siphilitica, L. puberula and L. appendiculata [6]. Sixty-nine species occur in South Africa, found throughout the country growing in a variety of habitats [7,8]. One of the commonest species found in the Eastern Cape of South Africa is L. flaccida (C. Presl) A.DC, family: Campanulaceae known as itshilizi in Ixhosa language [8,9] and is the specie investigated in this study. Traditionally, different Lobelia species has been used for bronchitis and asthma, topically for myositis and rheumatic nodules, as a diuretic, antidote and as carcinostatic agents for stomach cancer in Chinese folk medicine [10,11]. Lobelia siphilitica and L. cardinalis were historically used in the treatment of syphilis in North America [12]. Although several studies have been undertaken to evaluate lobeline (an alkaloid from Lobelia) as anti-smoking drug, the results obtained showed that it was ineffective in helping people quit smoking with conflicting results [13].

Reports on the chemical constituents of Lobelia species confirmed alkaloids such as lobeline, norlobelanine; flavonoid compounds (apigenin, luteolin, quercetin), and coumarins; and essential oil [14]. Biological studies on Lobelia species showed several activities including antimicrobial [15], antidepressant [16] anticancer [11] and antiepileptic [17] and analgesic and antivenom potentials [18].

The people of the Eastern Cape of South Africa depend largely on the use of medicinal plants born out of strong belief that they are effective to treat many diseases [19]. The Department of Science and Technology, South Africa, initiated the Indigenous Knowledge System (IKS) Project in 2002 in its bid to bridge the gap between the herbal practitioners and the academia, improve the health care system in the rural areas and advance herbal practices in the community.

This is the first study reporting chemical composition and biological activities of this particular species from the Eastern Cape and the aim was to investigate the chemical composition of the essential oil; screen the infusion extract for secondary metabolites; and then determine the acute toxicity profile, anti-inflammatory and anticonvulsant [8,11] activities of the crude infusion extract of Lobelia flaccida. Thus, this study is expected to generate new information and to validate the use of this plant in South African herbal practice.

Plant collection, and extraction of essential oil and infusion extract

Dried leaves of Lobelia were collected from the Herbal Practitioner on 3^rd July 2014 at Ginsberg, King Williams Town, Eastern Cape, South Africa. The leaves were used for the extraction of the essential oil and infusion extract. The dried leaves of Lobelia were identified by Dr. K Immelman, Herbarium Unit, Department of Botany, Walter Sisulu University (WSU), Mthatha. Essential oil of Lobelia flaccida dried leaf was obtained by hydrodistillation using a cleavenger-like apparatus and stored in a vial, protected from light until sent for analysis. Infusion extract of the leaves was obtained according to herbal practitioner’s directive. Briefly, 100 g of the dried plant material was grounded into coarse powders and soaked in boiling water with continuous stirring and allowed to infuse for 24 h. Thereafter, it was filtered using Whatman filter paper; a portion was kept for the phytochemical screening, while the remaining portion was dried in an oven at 35 ^oC.

Phytochemical screening

For qualitative screening, aqueous extract of Lobelia flaccida leaf was used while the powdered leaf was used for quantitative screening according to the method described by Mir et al [20].

Qualitative screening

The infusion filtrate was used for the qualitative analysis to detect the presence of secondary metabolites including tannins, saponins, flavonoids, terpenoids, alkaloids, phenols, phytosterols, glycosises, anthraquinones, phlobatanninsand proteins/amino acids in the aqueous leaf extract of Lobelia flaccida [20].

Analysis of essential oil of Lobelia leaf

Gas chromatography- flame ionization detector (GC-FID) analysis

GC analysis was carried out on a Perkin-Elmer 8500 gas chromatograph with a FID detector with a SGE BP X5 column that is 30 m in length with a film thickness of 0.25 µm and diameter 0.25 mm ID. The operating conditions were as follows: carrier gas, nitrogen with a flow rate of 3.0  ml/min; column temperature, 60 - 275 ^oC at 4 ^oC/min; injector and detector temperature, 280 ^oC -; volume injected 0.1 µl of the oil; split ratio, 1:50.

GC-MS analysis

GC/MS analysis of the oil was performed on a GCMS-QP2010 Gas Chromatography- mass spectrometer system operating in EI mode at 70 eV, equipped with a HP-5 MS fused silica capillary system with a 5 % phenylmethylsiloxane stationary phase. Capillary column parameter was 30 m by 0.25 mm, film thickness 0.25 µm. The initial temperature of the column was 70 ^oC and was heated to 250 ^oC at a rate of 5 ^oC/min. The final temperature was kept at 450 ^oC and run time of 68 min.  Helium was used as the carrier gas at a flow rate of 1 ml/min. The split ratio was 100:1. Scan time was 68 min with a scanning range of 35 to 450 amu. One microliter (1 µl) of the diluted oil (in hexane) was injected for analysis. N-alkane of C8 to C30 was run under the same condition for Kovat indices determination.

Experimental animals

Mice and rats were obtained from the South African Vaccine Initiative, Johannesburg and kept at the Animal Holding Facility, Zoology Department, Walter Sisulu University WSU. Male and female Wistar rats (200-300 g) were randomly selected (n=6) for the anti-inflammatory test. Swiss mice of both sexes (25-35 g; n=6) were also selected for the acute toxicity and the anticonvulsant tests. They were kept under standard laboratory conditions and had free access to rat chow and water. Food was however withheld overnight prior to experiments while water was provided ad libitum. This study was approved by the Department of Higher Education, WSU (Ref: DVC (AA&R) DRD/SREC) The study was carried out in accordance with the "Principles of Laboratory Animal Care", NIH publication no. 85-23, revised 1985.

Acute toxicity

Acute toxicity effect of the infusion extract of the leaves of Lobelia flaccida was assessed in mice using the oral route (p.o.) according to Lorke’s method [21]. The method involved using only thirteen (13) animals and was divided into 2 phases. The first phase consists of three (3) sub-groups (n=3) at dose levels of 10, 100 and 1000 mg/kg. The second phase involved four (n=1) dose levels of 1000, 1600, 2900 and 5000 mg/kg respectively. Immediately after treatment, each mouse was placed inside the Plexiglas cage and observed for immediate effects up to 30 min and thereafter for 24 h for lethal effects culminating into death. The LD₅₀ of the extract was estimated as the geometric mean of the lowest dose causing death and the highest dose causing no death according to Eq 1.

LD₅₀ = √(AB) . …………………………. (1)

where A is the maximum dose producing 0 % death and B is the lowest dose that produced 100 % death.

Anti-inflammatory activity studies

The anti-inflammatory activity was evaluated by the carrageenan-induced paw oedema test in rats 22]. The left hind paw was injected with 0.2 ml of normal saline and used the control. Normal saline (10 ml/kg), the extract of Lobelia (250 and 500 mg/kg) and aspirin (100 mg/kg) were orally administered 1 h prior to carrageenan injection (0.1 ml 2 % w/v carrageenan in normal saline) to 4 different groups of rats (n=6). The paw oedema sizes were recorded for each rat before carrageenan injection and at 1, 2, 3 and 4 h post carrageenan injection using Vernier Callipers.

Anticonvulsant activity studies

The anticonvulsant effect was evaluated using the pentylenetetrazole (PTZ)-induced convulsion model. Pentylenetetrazole (85 mg/kg) was used to induce tonic-clonic convulsions [23]. Five different groups (n=5) of mice were randomly selected and orally pre-treated for 1 h as follows: group I received normal saline (10 ml/k), groups II-III received extract (500 and 1000 mg/kg) respectively and group IV diazepam (1 mg/kg). Thereafter, anticonvulsant assessment was carried out with pentylenetetrazole (85 mg/kg, i.p.). The animals were observed for the onset of convulsion, time of death and mortality. Animals that survived beyond 30 min post-pentylenetetrazole injection were regarded to be protected.

Statistical analysis

The results are expressed as mean ± standard error of mean (SEM) and analysed using one-way analysis of variance (ANOVA) followed by post hoc test using Dunnett’s test for comparison between the treated groups and control at p < 0.05; while the ratio of mortality or protection for each group was expressed in percentage. The software used are GraphPad Instant R version 3.0.10.0 (UK) and GraphPad Prism Version 5 2013 (GraphPad Software Inc).

Yield of the essential oil and infusion extract

The yield of the essential oil was 0.22 % w/w, which was light yellowish in colour with characteristic aromatic odour. The infusion extract obtained from the dried leaf of Lobelia weighed 25.54 g (25.50 % w/w). 25.54 g (25.5 % w/w).

Phytochemical profile of the infusion extract of Lobelia flaccida leaf

The results obtained from the qualitative screening of Lobelia leaf indicated the presence of alkaloids, flavonoids, phenols, phytosterols, proteins, saponins and tannins; while carbohydrates, glycosides, phlobatannins and steroids were not detected. The quantitative analysis of Lobelia leaf infusion extract showed that the percentage of alkaloid, saponins and flavonoids constituents were 2.40, 0.14 and 0.40 % w/w respectively while its tannin constituent was not recoverable.

Composition of Lobelia essential oil

GC/MS analysis of Lobelia essential oil indicated the presence of 25 compounds out of which 23 were fully identified constituting about 98 % of all the constituents (). The major compounds identified include acetophenone (26.37 %), caryophyllene (8.11 %), borneol (7.46 %) and anethole (6.81 %).

The classes of compound identified in the oil were sesquiterpenes (8.11%), phenyl alkenes (6.81 %), monoterpenes (7.46 %) and a simple ketone (26.37 %). The major compounds obtained were from different classes and they play different roles in the Lobelia but the major compound is an aromatic ketone acetophenone (26.37 %).

Acute toxicity effect of the infusion extract of Lobelia dried leaf

There was no mortality at all the doses of the infusion extract up to 5000 mg/kg, p.o. The LD₅₀ was therefore estimated to be ≥ 5000 mg/kg through the oral route in mice.

Effect of infusion extract of Lobelia leaf on carrageenan-induced rat paw oedema

Rats pre-treated with normal saline had increased paw sizes throughout the 4-hour period of observation. Infusion extract of Lobelia, 250-500 mg/kg caused significant (p < 0.01) reduction in paw oedema sizes at 1, 2, 3  and 4 h post-carrageenan injection compared to control group. Also, aspirin (100 mg/kg) similarly caused significant (p < 0.01) reduction in the rats’ paw oedema ().

Effect of the infusion extract of Lobelia leaf against PTZ-induced convulsion

All the mice in the control group convulsed and died within 30 min post-PTZ (85 mg/kg, i.p.). The extract delayed the onset of convulsion and time of death after convulsion compared to vehicle although not statistically significant. Diazepam, (a standard anticonvulsant drug used) caused significant (p < 0.01) delay in onset of convulsion and protected the mice against PTZ-induced death. However, Lobelia at 500 and 1000 mg/kg offered 40 and 20 % protections against the PTZ-induced mortality respectively compared to the group’s 0 % and diazepam’s 100 % protection at 30 min observation period ().

Phytochemical screening of Lobelia leaf confirmed the presence of several metabolites including alkaloids, saponins, flavonoids, phenols, phytosterols, proteins and tannins. These phyto-constituents have variously been shown to be responsible for many bioactivities displayed by plants and new compounds can be discovered by probing them further. It must be noted however, that plant extract contains low concentrations of active compounds and a large number of promising compounds especially Lobelia species [17,18]. The essential oil analysis revealed the major compounds present in the Lobelia flaccida dried leaf to be acetophenone (26.37 %) caryophyllene (8.11 %), borneol (7.46 %) and anethole (6.81 %). There was scanty information in literature on the chemical composition of essential oil of this Lobelia specie. However, Joshi et al [15] reported that perilla ketone (25.61 %), camphorquinone (12.16 %), dibutyl phthalate (10.66 %) and allylnonanoate (8.47 %) were the main constituents of essential oil of Lobelia pyramidalis (Wall) from India. The oil composition in the present study is different in the class of compounds constituting it. Our report showed that there were more monoterpenes and sesquiterpenoids present as compared to that of Joshi et al [15].

Acute toxicity profile of the Lobelia flaccida dried leaf aqueous extract indicated an LD₅₀ of ≥5000 mg/kg, p.o., indicating that it is non-toxic [21]. This toxicity test result can be used to explain why the plant is a popular medicinal agent that has been used extensively over the years without report of serious adverse effects.

Carrageenan-induced paw oedema continues to be a valuable experimental animal model for acute inflammation which is strongly speculated to be biphasic viz. early phase (1-2 h) mediated by histamine and serotonin, and late phase (>2 h) mediated by bradykinin, leukotrienes and prostaglandins [26]. In this study, Lobelia extract significantly inhibited paw oedema induced by carrageenan in both phases, suggesting possible inhibition of release of inflammatory mediators and cyclooxygenase synthesis similarly to nonsteroidal antiinflammatory drugs such as aspirin, whose mechanism of action is inhibition of the cyclooxygenase pathway. The results obtained here support the ethno-medicinal claims ascribed to this particular plant species found in South Africa by the herbal practitioner. The screening of this plant extract showed high flavonoid content (4 % w/w) which has been previously linked to the antiinflammatory activity of L. chinensis [10], it can therefore be postulated here that the antiinflammatory activity of this Lobelia extract could be associated with the high flavonoids and or saponins present in it [27,28].

The anticonvulsant test indicated that Lobelia infusion extract did not significantly alter the onset of convulsion or time of death compared to the vehicle, but it did prolong insignificantly the latency and time of death caused by the PTZ compared to the vehicle. Previous studies have indicated that the extract of this plant showed anticonvulsant potentials in many models of convulsion and lobeline isolated from the leaf of L. nicotianaefolia demonstrated significant anticonvulsant activity [29]. Considering the fact that it was just the infusion extract that was used in this study, extraction with other solvents such as ethanol would probably have demonstrated greater anticonvulsant activity because more bioactive substances would be extracted. Also, the essential oil component which was not tested against convulsion due to low quantity obtained in this study may exhibit greater activity since essential oils of many plants demonstrate anticonvulsant activities [43].

The major component found in the essential oil of this L. flaccida species is caryophyllene. The infusion extract is non-toxic, possesses strong anti-inflammatory activity but mild anticonvulsant activity. Thus justifies the plant’s use in traditional medicine to treat pain and rheumatism.