Acetone leaf extracts of some South African trees with high activity against Escherichia coli also have good antimycobacterial activity and selectivity index

Background Tuberculosis is a world-wide problem affecting humans and animals. There is increasing development of resistance of the pathogens to current antimycobacterial agents. Many authors have investigated activities of extracts and isolated compounds from plants. The traditional uses of plants have frequently been the criterion to select plants investigated. In this contribution, we investigate whether plant extracts with very good activity against Escherichia coli may also be active against mycobacteria. Methods The antimycobacterial activity of eight South African tree leaf extracts with high activity against Escherichia coli were determined in vitro against Mycobacterium smegmatis, M. fortuitum and M. aurum, using a serial microdilution method. The cellular cytotoxicity was also determined by the MTT assay using Vero monkey kidney cells. The selectivity index was determined by dividing the cytotoxicity of extracts by MIC. Results The antimycobacterial activity of the extracts ranged from 0.02 to 2.5 mg/ml. Mycobacterium smegmatis was more sensitive to the extracts (Average MIC = 0.96 mg/ml) and Mycobacterium aurum was comparatively resistant (Average MIC = 2.04 mg/ml). The extracts of Cremaspora triflora had strong antimycobacterial activity with a MIC of 0.05 mg/ml that compared reasonably well with that of streptomycin (0.01 mg/ml) and rifampicin (0.03 mg/ml), p > 0.05. Cremaspora triflora had the best selectivity index of 2.87 against Mycobacterium fortuitum. Conclusion The high activity of C. triflora extracts against the fast-growing mycobacteria and good cellular safety is promising. It may be interesting to investigate extracts against pathogenic M. tuberculosis, M. bovis and M. avium cultures and to isolate active antimycobacterial compounds.


Background
Tuberculosis (TB) is a highly infectious disease, caused by Mycobacterium tuberculosis. TB typically affects the lungs but can also affect other organs in the body [1]. An estimated 5-15% of the 2-3 billion people exposed to M. tuberculosis will develop the disease in their lifetime. The probability is even higher among people infected with HIV [1,2]. In 2014, about 9.6 million people were infected with TB and 1.5 million consequently died. A high proportion of the reported cases (28%) came from Africa [1]. The growing incidence of pathogenic mycobacterial multi-drug resistance to the best two first line antituberculosis drugs -streptomycin and isoniazid and extensively-drug resistance to both first and second line antituberculosis drugs that include fluoroquinolones and capreomycin highlights the critical need to search for newer anti-tuberculosis drugs. There are reports of the emergence of a 'totally drug-resistant TB' strain with a limited chance of successful treatment [1].
Natural products, either as a source of pure compounds or as standardised plant extracts, provide opportunities for new drug leads due to the high chemical diversity in plants [3]. It has been shown that extracts of Maerua edulis, Securidaca longipedunculata, Zanthoxylum capense and Tabernaemontana elegans have high activity against Mycobacteria spp. [4,5].
To select plants for investigation, the traditional use can be considered. Because traditional healers have mainly water available and antimicrobial compounds are not readily soluble in water [6] this is not such a promising approach [7]. Random screening of tree leaf extracts indicated that many plant extracts have excellent activities [8]. The cell wall of Mycobacterium tuberculosis has similar characteristics to both Gram-positive and Gramnegative bacteria [9]. In addition, M. tuberculosis is closer related to Gram-negative bacteria than Grampositive bacteria. Analysis of evolutionary distance between nearest ancestral units, suggest that M. tuberculosis is closely related to Escherichia coli and Pseudomonas aeruginosa [9]. Thus in this study we selected eight plant species with high antimicrobial activity against Escherichia coli to determine the potency, efficacy and safety of acetone tree leaf extracts against three fast growing Mycobacterium species. With one exception no previous studies were found on the antimycobacterial activity of extracts of these species.

Collection of plant material, drying and storage
Leaves of the selected tree species were collected and dried in the shade before grinding to a fine powder. The trees selected with family and voucher specimen numbers in brackets were Hypericum roeperianum G.W. Schimp.ex A. Rich. var. roeperianum, (Hypericaceae, PRU 120126), Cremaspora triflora (Thonn.) K. Schum

Preparation of extracts
Ground dry leaf powder (1.0 g) of each plant was poured into 50 ml polyester centrifuge tubes and extracted with 30 ml acetone for one hour [10]. After this they were centrifuged at 4000 x g for 10 min. The supernatants were decanted into preweighed glass vials through Whatman No. 1 filter paper and concentrated to dryness under a stream of cold air. The dried extracts were made up to a concentration of 10 mg/ml (stock solution) in acetone to be used in subsequent assays and stored at 5°C in tightly stoppered glass tubes.

Antimycobacterial activity assay Mycobacterial cultures
Mycobacterium smegmatis (ATCC 1441), Mycobacterium aurum (NCTC 10437) and Mycobacterium fortuitum (ATCC 6841) were cultured as described by McGaw et al. [3], and maintained on Löwenstein-Jensen agar slants, supplemented with glycerol. Inocula suspensions were prepared by mixing a few microbial colonies with sterile distilled water to render a concentration of cells equal to standard 1 McFarland solution (approximately 4 × 10 7 cfu/ml). The suspension was then diluted with freshly prepared Middlebrook 7H9 broth supplemented with 10% oleic acid, albumin, dextrose, and catalase (OADC) to obtain a final inoculum density of approximately 4 × 10 5 cfu/ml. A serial microplate broth microdilution technique [11] was used to obtain the MIC values of the various extracts.

Statistical analysis
All experiments were conducted in triplicate and values expressed as the mean ± standard deviation. Variations in mean were analysed using one-way analysis of variance (ANOVA), and means were statistically significant if p < 0.05.

Results and discussion
There are several reasons given to justify the constant search for new anti-TB drugs to improve the current treatment regimen by reducing therapy time and addressing drug resistance especially against multi-drug resistant (MDR) and extreme drug resistant (XDR) mycobacterial strains. Development of natural anti-TB drugs with minimal hepatotoxic and nephrotoxic effects could add to the collection of available drugs used in TB treatment. In many other human diseases very useful drugs have been discovered from plants. Not any of the drugs (isoniazid, rifampicin, capreomycin, fluoroquinolones, kanamycin and amikacin) used as first or second line drugs in the chemotherapy of TB has its origin from plant derived natural products [13].
The choice of using fast growing and non-pathogenic Mycobacterium spp. in antimycobacterial assays was based on their avirulent nature and similarity in sensitivity to pathogenic Mycobacterium strains [14,15]. In addition, there are published reports on the use of M. fortuitum as an alternative screening model to Mycobacterium tuberculosis for potential antitubercular drug development [14]. Subsequently, Aro et al. [16] found that M. aurum is the best predictor of the activity against pathogenic M. tuberculosis with a correlation coefficient of 0.9 In addition, Mycobacterium smegmatis was the best predictor strain to substitute pathogenic M. bovis and M. tuberculosis, MIC values obtained using M. fortuitum correlated well with those of M. bovis BCG [3].
In addition to the antimycobacterial activity, the total activity was also calculated by dividing the mass in mg extracted from 1 g of dried material with the MIC in mg/ml (7). The total activity indicates the volume to which the extract from one g of plant material can be diluted and still inhibit the growth of the microorganism.
The MIC of activity of the extracts against M. smegmatis ranged from 0.04 to 2.5 mg/ml. C. triflora extracts had the best activity (MIC 0.04 mg/ml), while M. lanceolata had moderate activity (MIC 0.16 mg/ml); H. roeperianum and M. mesozygia had weak activity with MIC = 0.63 mg/ml. The other three species had very low activity. C. triflora had the second best total antimycobacterial activity (TAA) value of 504 ml/g and an SI value of 1.44. TAA value of 504 ml/g infers that if 1 g of the dry acetone crude extract is diluted with 504 ml of water it will still inhibit the growth of the targeted microorganism [10]. Other extracts had relatively poor activity against M. smegmatis and SI less than 1.0 ( Table 1).
The MIC values of the extracts against M. fortuitum ranged from 0.02 mg/ml to 2.5 mg/ml. Cremaspora triflora again had the best activity (MIC = 0.02 mg/ml), while M. lanceolata and M. mesozygia had moderate activity with MICs of 0.31 mg/ml. C. triflora extracts had the best total antimycobacterial activity against M. fortuitum with a value of 1008 ml/g, and a good selectivity index value of 2.87 (Table 1).
M. aurum was much more resistant to the plant extracts and only C. triflora had good activity with an MIC value of 0.08 mg/ml, and total antimycobacterial activity value of 255 ml/g and a low SI value of 0.72 (Table 1).
These results were more or less in line with the results of McGaw et al. [3] who investigated the activity of compounds isolated from Euclea species against different Mycobacterium species. It is possible that the fast growing Mycobacterium species are more susceptible than the slow growing species. It also appeared that M. smegmatis would be the best predictor of activity against pathogenic mycobacteria [3].
Palgrave (1997) in Erastus et al. [17] reported the traditional use of the dried bark of Bolusanthus speciosus in the treatment of tuberculosis. Our results investigating the leaves did not support this claim. The acetone leaf extracts of Bolusanthus speciosus had low antimycobacterial activity with a mean MIC value of 1.67 mg/ml against the three tested microorganisms. It is not strange that in vitro studies do not confirm traditional use of plant extracts. This could be due to several reasons. Table 1 Minimum inhibitory concentration (MIC) and total antimycobacterial activity (TAA) and selectivity index (SI) of the eight selected acetone leaf extracts against Mycobacterium smegmatis, Mycobacterium fortuitum, Mycobacterium aurum. The SI and TAA values of the extracts were calculated from the cytotoxicity and percentage yield of the extracts results published in Elisha et al. [ Different extractants could be used, microbial infections could influence activity of extracts left without refrigeration for some time. Clinical identification of the disease and recovery as well as the placebo effect could also have an influence. The activity of plant extracts may also change during different seasons and there could be synergistic activities if extracts of more than one plant species is used. Maesa lanceolata, traditionally used to treat sore throat and influenza [18,19], had moderate activity against M. smegmatis and M. fortuitum with MIC values of 0.16 mg/ml and 0.31 mg/ml respectively (Table 1). This may support its use traditionally in the treatment of respiratory infections.
Cremaspora triflora had the best activity against all the Mycobacterium spp. with a mean MIC value of 0.05 ± 0.01 mg/ml and E. croceum had the weakest activity against the microbes with MIC value of 2.5 ± 0.41 mg/ml (Fig. 1). Mycobacterium smegmatis was more sensitive to the extracts than M. fortuitum and M. aurum (Fig. 2), however the difference in the mean between M. smegmatis and M. fortuitum was not statistically significant p > 0.05 (Fig. 2). Means of M. smegmatis and M. fortuitum differed significantly from that of M. aurum, p < 0.05. Our findings agree with a previous report [20] that M. smegmatis was more sensitive to the essential oil of the gall of Pistacia atlantica followed by M. fortuitum, and M. aurum.
The extracts of C. triflora have potential as a good candidate for elaborate antimycobacterial investigation. Results from this study on the good activity of the extracts of C. triflora (Table 1) against the different mycobacterial strains differs with the report that the extracts of C. triflora had moderate activity against M. smegmatis and M. aurum, with MIC values of 0.23 mg/ml and 0.10 mg/ml respectively [16]. The discrepancies on the potency report of C. triflora extracts in our study and that of Aro et al. [16], may be attributed to differences in the season and time of plant collection and storage conditions. It is however agreed that the plant has potential for development as an alternative anti-TB drug.