The Carissa edulis medicinal plant was collected from its natural habitat in Gitoro forest of Meru, Eastern Province, Kenya. The roots were collected with acceptable bio-conservation methods and a voucher specimen (Mungai, Rukunga and Tolo; No. 076) is on deposit at the East African Herbarium, National Museums of Kenya, Nairobi, Kenya.

Extraction and activity guided phytochemical analysis was carried out concurrently. Briefly; 150 g of powdered root bark was extracted successively using diethylether and methanol and the extracts dried in vacuo. The yields for the diethylether extract was loaded into a column packed with silica gel and eluted gradient wise with increasing concentrations of ethylacetate in hexane. The fractions of 10 ml each were collected and monitored on pre-coated thin layer chromatography (TLC) plates. Similar fractions were pooled according to the TLC characteristics. One such pooled fraction, coded CEE-3, had reasonable antiviral activity in vitro than the others and was a mixture of three compounds. Further purification of this fraction was carried out using column chromatography packed with reverse phase (RP)-C18 Silica gel and eluted with increasing concentrations of acetonitrile in methanol, starting from 5 to 10 % acetonitrile. Fractions were monitored on RP-C18 TLC analytical plates and fractions pooled as appropriate. Three pure compounds coded; CEE-3A, CEE-3B and CEE-3C were subsequently obtained.

The methanol extract was loaded into a column packed with sephadex LH-20 and eluted isocratically using methanol (MeOH). Ten milliliter (10 ml) fractions were obtained and monitored through TLC. Following bioactivity, one of the fractions was purified using a column packed with RP-C18 and eluted with MeOH:H₂O (35:65). One pure compound coded as CEM-1E was isolated.

These four compounds; CEE-3A, CEE-3B, CEE-3C and CEM-1E were identified by comparison of their spectroscopic profiles with published data and found to be lupeol ¹⁷ , β-amyrin ²⁴ , carissol ²⁴ and oleuropein ^4,14 respectively.

The Vero E6 cells were donated by the Virology department, Toyama Medical and Pharmaceutical University, Toyama, Japan. The cells were grown in 5% CO₂ at 37 ^o C in MEM supplemented with 5% FBS for cell growth, and 2% FBS for cell maintenance.

The herpes simplex virus (HSV) strains used in the study were wild-type 7401H HSV-1 ¹¹ , thymidine kinase deficient (TK−) B2006 HSV-1 ⁵ and acyclovir resistant (AP ^r ) 7401H HSV-1 ¹³ . The viruses were donated by the Virology Department, Toyama Medical and Pharmaceutical University, Toyama, Japan. The virus stocks were prepared from infected Vero E6 cells. The infected cultures were frozen and thawed three times to lyse the cells, and centrifuged at 3000 rpm for 15 min. Their supernatants, containing HSV, were harvested and stored at −80 ^o C until use.

The compounds were examined for extent of inhibition of plaques on HSV infected cells as a measure of anti-viral activity in vitro using the method described by Kurokawa et al. ^11,12 . Briefly, Vero E6 cells were cultured to a confluent monolayer in MEM supplemented with 5% FBS in 5% CO₂ at 37 ^o C. The cells were infected with 100 plaque forming units (PFU) of HSV and left to adsorb for 1 h at room temperature. The cells were then overlaid with MEM supplemented with 2% FBS and 0.8% Methylcellulose (MC) containing various concentrations of the compounds and incubated at 37 ^o C in 5% CO₂ for 2 days. The infected treated cells were fixed in formalin, washed and stained with 0.03% methylene blue solution. The plaques, appearing as transparent dots against a blue background, were counted using a dissecting microscope and the percent plaque inhibition calculated against control. The effective concentration inhibiting formation of plaques by 50% (EC₅₀) was determined from two independent experiments.

The cytotoxic concentration causing 50% cell lysis and death (CC₅₀) was determined for the compounds by the method described elsewhere ²² . Briefly, Vero E6 cells were seeded at a concentration of 2.5×10 ⁴ cells/well in 24-well plates and grown in 5% CO₂ at 37 ^o C for 2 days. The culture medium was replaced by fresh medium containing compounds at various concentrations, and cells further grown for 24 h. The cells were treated with trypsin and the number of viable cells determined by the trypan blue exclusion method. The concentration of compound reducing cell viability by 50% (CC₅₀) was determined from a curve relating percent cell viability to the concentration of compounds.

Lupeol was compared for its anti-viral activity with acyclovir on the growth of the wild type strain of HSV-1 in the virus yield reduction assay as described elsewhere ²² . Briefly, confluent monolayers of Vero E6 cells were infected with HSV-1 at multiplicity of infection of 5 (5 M.O.I). Separate dishes each containing infected cells were treated with lupeol at various concentrations (0-10 µg/ml). The cells were incubated in 5% CO₂ at 37 ^o C for 24 h and lysed. The lysed cultures were collected in 15 ml tubes and centrifuged at 3000 rpm for 10 min and supernatant harvested. A serial dilution of each supernatant was prepared in MEM and the virus titre determined by the plaque inhibition assay.

Balb/C mice, reared at the Kenya Medical Research Institute’s (KEMRI) animal facility, were used in the experiments. The animals were handled according to the guidelines laid down by the Animal Care and Use Committee of the KEMRI. The in vivo evaluation of the efficacy of lupeol was carried out using the method described by Kurokawa et al. ¹² . Briefly, female Balb/C mice of 7 weeks old and weighing approximately 20 g were acclimatized for one week in the experimental room. The animals were fed with food (Mice cubes, Unga feeds, Kenya) and water ad libitum. The mid flank of each mouse was shaved using an electric hair trimmer (Wahl super taper, England) and hair completely removed by applying a chemical hair remover (Shiseido, Co., Ltd., Tokyo, Japan) on the shaved area. The mice were randomly divided into treatment groups each comprising of 5 mice. The shaved mid flank of each mouse was scratched using a bundle of G27 needles and the scarified area infected with 1×10 ⁶ PFU HSV-1. The groups were administered with the test drug or acyclovir or placebo orally three times a day for seven consecutive days. The development of skin lesions and mortality were continuously monitored every 8 h daily and scored as follows: 0-no lesion; 2-vesicles in local region; 4-erosion and/or ulceration in the local region; 6-mild zosteriform; 8-moderate zosteriform and 10-severe zosteriform and death. The mice were fed and observed for 30 days before the experiment was terminated.

In the in vitro assays, the mean number of plaques was used to calculate the percentage of plaques inhibited. A curve relating percent plaque inhibition against concentration (a dose response curve) was used to calculate the inhibitory concentration for 50% (EC₅₀). In the cytotoxicity experiments, the number of viable cells in the untreated control was used to calculate the percent of cells lysed in the treated cells. The cell cytotoxic concentration (CC₅₀) was determined from a plot relating percent cell viability against each tested concentration.

In the in vivo experiments, the Student’s t-test was used to evaluate the significance of differences between control and compound treated mice in mean survival times and mean times at which skin lesions were initially scored 2 (vesicles in local region) or 6 (zosteriform lesions) after infection. The repeated measure analysis of variance (ANOVA) with Dunn’s procedure as a multiple comparison procedure was used to analyze the interactions between treatment and progression of infection in mean lesions scores from day 3 to 10 following HSV-1 infection. A p-value of  0.05 was defined as statistically significant.

The 1998 StatView computer software by SAS Institute Inc. USA was used for the statistical analysis.

The in vitro antiviral activities of the compounds against sensitive and resistant viral strains of HSV are presented as Table 1. Of the four compounds, lupeol exhibited the most potent antiviral activity; EC₅₀ at 2.98 µg/ml for 7401H HSV-1, 3.66µg/ml for AP ^r 7401H HSV-1 and 4.2 µg/ml for the TK ^– B2006 HSV-1. The EC₅₀ concentrations were far below the toxic level (CC₅₀ >160.0 µg/ml) giving lupeol a selectivity index of >38. The 7401H HSV-1 strain was sensitive to oleuropein and carissol; EC₅₀ at 4.83 and 3.76 µg/ml respectively, however, no activity was registered for oleuropein in both the acyclovir resistant viral strains at 5 µg/ml. β-amyrin did not have activity against the 7401H HSV-1 strain at the highest tested concentration of 5.0 µg/ml and was therefore not examined any further. Acyclovir, the reference drug, had activity for 7401H HSV-1; EC₅₀ at 0.45 µg/ml, but did not show any activity for the other strains at the highest tested concentration of 10.0 µg/ml.

Figure 1

Table 1: antiviral activity of the compounds

The bar graph Figure 1 compares the virus yields of 7401H HSV-1 in Vero cells treated with varying concentrations of lupeol or acyclovir. Lupeol reduced the virus yields in a dose dependent manner, the highest concentration of 10.0 µg/ml reducing the yields to less than 1x10 ⁵ PFU/ml (~ 98.3% virus yield reduction). At 2.5 µg/ml, lupeol reduced the virus yields by half the amount registered in untreated cells. On the other hand, acyclovir reduced the viral yields more sharply, the 2.5 µg/ml concentration reducing the yields significantly (to less than 1x10 ² PFU/ml) and completely at 10.0 µg/ml. The mean virus yields at each tested concentration for both lupeol and acyclovir were statistically significant (p≤0.05 test verses control by Student’s t-test).

Figure 2

Figure 1: The virus yields in 7401H HSV-1 infected Vero E6 cells treated with lupeol or acyclovir . The error bars indicate the levels of deviation within the mean of three independent experiments at each tested concentration. * = ? 0.05 (Test verses control by StudentÂ’s t-test).

Table 2 presents the efficacy of lupeol on 7401H HSV-1 cutaneous infection in mice. Three doses (5, 10 and 20mg/kg) were orally administered to mice over a period of 7 days and observations for onset of infection, progression, mean survival times and mortality recorded. Vesicles in local region (score 2) were observed on the 4 ^th , 5 ^th and 6 ^th days for 5, 10 and 20 mg/kg treatments respectively. Vesicles appeared on controls by the 4 ^th day while the group on acyclovir at 5mg/kg on the 7 ^th day. Mild zosteriform lesions (score 6) were observed on 7 ^th day for groups receiving treatment at 5 and 10 mg/kg while for the 20 mg/kg and acyclovir groups, lesions were noted on 8 ^th day. The mean survival times for lupeol treatments at 5 and 10 mg/kg were similar to the control group (approximately 8 days) while the group on 20 mg/kg was 9 days (p0.05 test Vs control by Student’s t-test). A high mortality was observed in all the groups that received lupeol treatment. No toxicity was observed in uninfected mice treated with the highest dose of lupeol (results not shown).

Figure 2 outlines the interaction line plots for progression of 7401H HSV-1 infection in mice following oral treatments with lupeol at 5, 10 and 20 mg/kg. The progression of infection in mice treated with 5 mg/kg was similar to that of control, where once the infection was on progresses, it continued uninterrupted until the final lethal stages. The 10 mg/ml treatment delayed the progression of infection in the treatment phase, however on withdrawal of treatment on the 7 ^th day, the infection progressed faster to the lethal stages. The mice on 20 mg/kg had a much slower progression of infection than the 10 mg/kg treatment. The influence of treatment on progression of infection at 10 and 20 mg/kg were statistically significant {p0.05 test Vs control by repeated measures ANOVA (Benferroni/Dunn)}. Acyclovir treatment contained progression of infection to below mild zosteriform lesions, with a statistical significance {p0.05 test Vs control by repeated measures ANOVA (Benferroni/Dunn)}.

Figure 3

Table 2: The efficacy of lupeol in a cuteneous 7410H HSV-1 infection in mice

Figure 4

Figure 2: The interaction line plots for progression of 7401H HSV-1 infection in Balb/C mice following oral treatments with increasing doses of lupeol. In the figure, = ? 0.05 Test Vs control by repeated measures ANOVA (Benferroni/Dunn). Error bars are deviations within the mean of 5 mice in each group.

The authors are grateful to the Japan International Cooperation Agency (JICA) for the financial and technical support in the study and to the Director, KEMRI, for authority and permission to publish these results. We also thank Mr. L. Ogutu, Ms. M. Muthoni, Ms. G. Nginya, Mr. D. Odira and Mr. N. Mwangi for their excellent technical assistance.