Journal of Bacteriology & Parasitology

Infectious diseases result in >17 million deaths worldwide annually, mostly in children and the elderly [1]. The morbidity and mortality associated with infectious diseases has remained significant, particularly food-borne illnesses including diarrhea among children and respiratory infections such as tuberculosis, despite the advances in antimicrobial chemotherapy and supportive care. To make matters worse, the haphazard use of antimicrobials in the treatment of many infectious diseases has inevitably led to the emergence of multiple drug resistant microorganisms [2]. For example, in 1990, almost all cholera isolates in New Delhi (India) were sensitive to furazolidone, ampicillin, co-trimoxazole and nalidixic acid. In 2000, these drugs became largely obsolete in the treatment of cholera. The use of natural products, such as medicinal plants as therapy against infectious diseases is a traditional therapeutic measure especially in developing countries as they contain a combination of potential antimicrobial compound(s) instead of a single purified molecule [3]. Khat, Catha edulis, from the family Celastraceae is a natural stimulant that is found as a flowering evergreen tree and is widely cultivated in Africa and southern parts of Arabia [4-6]. The chewing of young shoots and leaves of C. edulis is a traditional and social habit in some countries of East Africa and Arab Peninsula [5,6]. It contains the alkaloid called cathinone (an amphetamine-like stimulant) and other polyphenolic compound such as tannins and flavonoids, which generally occur as glycosylated derivatives and are known for their antioxidant effects [7]. Although, the stimulating effects of C. edulis are well known, the antimicrobial effects of C. edulis remain incompletely understood. The present study was designed to investigate the antimicrobial properties of C. edulis extract against Gram-positive bacteria, Gram-negative bacteria, yeast and protists. The results revealed that C. edulis extract possess potent antibacterial properties against Gram-positive bacteria tested as well as anti-amoeba activity.

Plant materials

Relatively fresh plant leaves and young shoots near the top of the leaves of Catha edulis were purchased from a Somali shop on Edgware Road, London, UK. Approximately, 55 g of the plant material were ground and soaked in 100 mL of methanol (100 %) in a water bath at 25°C for 24 h with continuous shaking using Cole Parmer Orbital shaker at 40 g. The extract was filtered, concentrated at 40°C under reduced pressure, and methanol was dried in a freeze-dryer [8]. Finally, the extracts were re-suspended in 20 mL methanol and stored at 4°C until tested for antimicrobial activities.

High performance liquid chromatography (HPLC)

The crude methanolic extract were applied to a reverse phase C18 column with flow rate of 1 mL per min. Using HPLC, methanolic extracts were eluted with 0, 50 and 100 % methanol. Fraction collected from the crude methanolic extract were freeze dried. The lyophilised fractions were reconstituted in 1 mL of methanol and tested for antimicrobial activities [9].

Microbial strains

The microbial strains used in the present study were bacteria [Escherichia coli K1 strain RS218 (O18:K1:H7), a cerebrospinal fluid isolate from a neonate with meningitis, and environmental isolates of Brevundimonas diminuta, Bacillus magaterium, Micrococcus luteus], yeast (Aspergillus varicolor, Penicillium solitum, P. brevicompactum), and the protist (Acanthamoeba castellanii, a keratitis isolate belonging to the T4 genotype). All isolates were available in the University microbial collection (stored at -80°C and available upon request) except A. castellanii that was obtained from the American Type Culture Collection (ATCC 50492).

Antibacterial test

The antibacterial activities of C. edulis extract were determined using the antibiotic disc diffusion assay (“Kirby-Bauer” assay) [10]. Briefly, bacterial strains (E. coli, B. diminuta, B. magaterium and M. luteus) were grown at 37°C in Luria-Bertani broth and their optical density adjusted to 1.0 at 595 nm. The bacterial cultures were evenly spread on the surface of nutrient agar plates. Sterile Whatman filters of 60 mm size were impregnated with 100 μL (0.1 mg protein) of methanolic extract and placed on agar containing bacteria. For negative controls, filters were impregnated with 100 μL of the solvent alone and for positive controls filters were impregnated with 100 μg per mL gentamicin. The plates were incubated at 37°C overnight, following this the zones of inhibition were measured. An increase in the diameter of the inhibition zone was used to determine whether C. edulis extract have the potential to target the bacteria investigated.

Antifungal activity test

Sporing fungus (A. varicoluor, P. solitum and P. brevicompactum) were suspended in 1 mL of sterile water and evenly spread on the surface of potato dextrose agar (PDA). The antifungal activity of the C. edulis extract was determined by disk diffusion method. Sterile Whatman filters of 60 mm size were impregnated with 100 μL (0.1 mg protein) of methanol extracts and placed on PDA containing yeast. For negative controls, filters were impregnated with 100 μL of the solvent alone and for positive controls filters were impregnated with 100 μg per mL amphotericin B. Plates were incubated at 25°C for 24 h, following this the zones of inhibition were measured.

Amoebistatic and Amoebicidal assays

Acanthamoeba castellanii were grown to confluency in 24-well plates. The next day, plates were washed with PBS to remove unbound amoebae. Amoebae were incubated with various concentrations of methanol extracts of C. edulis (50 μg and 100 μg protein) or solvent alone and the plates incubated at 30°C for 24 h. The effects on amoeba growth and viability were determined using haemocytometer counting and Trypan blue exclusion testing. The normal growth rates of A. castellanii were determined using growth medium alone, i.e., PYG. For positive control, amoebae were treated with 100 μg per mL of chlorhexidine (anti-amoeba drug).

C. edulis extracts exhibited antibacterial but no antifungal properties

The crude metanolic extract of C. edulis were tested against bacteria and yeast using the disc diffusion assay [10]. The results revealed that C. edulis extract inhibited growth of B. diminuta, M. luteus and B. magaterium. In the presence of extract, the results showed that the zone of inhibition for B. diminuta was 19 mm ± 2.3; 22 mm ± 3.1 for M. luteus and 16 mm ± 0.7 for B. magaterium. Notably, C. edulis extract had no effect against the clinical isolate of E. coli. The zone of inhibition for E. coli around the C. edulis extract disc was <11 mm, even though the zone of inhibition for E. coli around the gentamicin antibiotic disc (positive control) was >18 mm. The breakpoint inhibition zone was considered 14 mm as per EUCAST guidelines. In contrast, C. edulis extract had no effect on A. varicolor, P. solitum and P. brevicompactum in the disc diffusion assay even though the zone of inhibition around the amphotericin B disc (positive control) was >15 mm.. The results are representative of at least three independent experiments performed in duplicate.

C. edulis extract showed potent amoebicidal and amoebistatic effects

To determine the effects of crude methanolic extract of C. edulis on A. castellanii growth and viability, amoebicidal and amoebistatic assays were performed. 100 μg per mL of chlorhexidine exhibited 100% kill. For C. edulis extract, the results revealed that C. edulis extract exhibited amoebicidal effects with an initial reduction in the viability of amoeba numbers as evidenced by the uptake of Trypan blue dye. However, a sub-population of A. castellanii remained viable, but cultures remained static over longer incubations (i.e., 48 h) even in growth medium. At 100 μg, the addition of C. edulis extract produced 55 % ± 6.7 reduction in viable cell numbers within 24 h. At 50 μg, the addition of C. edulis extract produced 23.5 % ± 3.3 reduction in viable cell numbers of the original inoculum. The results are representative of at least three independent experiments performed in duplicate.

HPLC fractions of C. edulis extract showed anti- Acanthamoebic activities

A total of 35 fractions were collected from the crude methanolic extract and were tested for antimicrobial activity against A. castellanii. The results revealed that fractions eluted with 50 % methanol of C. edulis extract exhibited amoebicidal effects corresponding to crude methanolic extract. The addition of 100 μg of C. edulis extract produced 63.4 % ± 5.4 amoebicidal effects within 24 h. The residual amoebae, although viable, remained static in growth medium for up to 48 h. In contrast, fractions eluted with 0 and 100 % methanol (i.e., PBS alone) of C. edulis extract did not exhibit amoebicidal and/or amoebistatic effects, even though both fractions contained approximately similar amount of protein (100 μg). The results are representative of at least three independent experiments performed in duplicate.

Although C. edulis is widely studied for its psychoactive properties, little work has been done to test its antimicrobial properties. Our findings revealed that the methanolic crude extract of C. edulis exhibit antibacterial properties against B. diminuta, M. luteus and B. magaterium but not E. coli and they have no antifungal properties against A. varicolor, P. solitum and P. brevicompactum using the disc diffusion assay. These results are consistent with previous findings, which demonstrated the antimicrobial properties of C. edulis extract against bacteria (Porphyromonas gingivalis, Tannerella forsythensis, Streptococcus pyogenes with the zone of inhibition in the range of 10 to 14 mm at a concentration of 10 mg per cm³ but showed no effect against either Staphylococcus aureus), or yeast (Candida albicans) [11]. The selective targeting of certain bacterial species by C. edulis extract is not clear. Given the potent antibacterial activities of C. edulis extract against some Gram-positive but none of the Gram-negative bacteria tested, suggest that they possibly target(s) the bacterial outer membrane and this needs to be investigated in future studies.

More importantly, for the first time the results revealed potent anti-Acanthamoebic properties of C. edulis extracts. This is interesting as C. edulis extracts had no effect against the eukaryotic yeast (Candida albicans) suggesting its selective targeting. Although, the mechanisms of C. edulis-mediated A. castellanii death is unknown, previous studies have shown that C. edulis-induced apoptosis in various human leukaemia cell lines including HL-60, NB4, Jurkat cells in a caspase-1 and -8 dependent manner [4]. As the single-celled A. castellanii is not yet known to undergo programmed cell death, future studies are needed to understand the molecular mechanisms of C. edulis-mediated A. castellanii death as well as its antibacterial properties. Further investigations are in progress to identify the active component(s) and to assess their clinical relevance.

In conclusion, this study showed the selective antibacterial properties against Gram-positive but not Gram-negative bacteria tested. For the first time, it is shown that C. edulis extracts possess anti- Acanthamoebic but not antifungal properties, albeit weak compared to that of the positive control (chlorhexidine).

This work was partially supported by grants from The Aga Khan University, Life Sciences Research Fund, and the University of London.