Antitubercular evaluation of root extract and isolated phytochemicals from Lophira lanceolata against two resistant strains of Mycobacterium tuberculosis

Abstract Context: The roots of Lophira lanceolata Van Tiegh. Ex Keay (Ochnaceae) have numerous medicinal values in the Central African region. Even though the MeOH extract of the roots has shown antimycobacterial activities, the constituents responsible for this inhibitory activity remain unknown. Objective: Phytochemical investigation of the MeOH root extract of L. lanceolata and determination of the antimycobacterial activities of that extract and constituents against the growth of Mycobacterium tuberculosis. Materials and methods: Column chromatography was used to provide bioactive phytoconstituents. Those compounds were elucidated using MS and NMR spectroscopic data. Antimycobacterial screening of the extract (4.882–5000 µg/mL in DMSO during 24 h at 37 °C) and isolated compounds (0.244–250 µg/mL in DMSO during 24 h at 37 °C) was performed by microplate alamar blue assay (MABA) against two mycobacterial strains. Results: The investigation of L. lanceolata MeOH roots extract provided of mixture of unseparated biflavonoids with a newly described one, dihydrolophirone A (1a) associated to lophirone A (1b). The bioactive compounds that effectively inhibited the growth of M. tuberculosis AC45 were found to be compounds 1 and 2. They exhibited MIC values of 31.25 and 15.75 µg/mL, respectively, and their MIC was found to be 62.5 µg/mL against resistant strain AC83. Discussion and conclusions: It is clearly evident from the results obtained that the mycobacterial activity of L. lanceolata could be related mainly to its steroid and flavonoid contents. Therefore, this study suggests the potential of the above-mentioned classes of compounds as promising candidate agents for developing new anti-tuberculosis drugs.


Introduction
Tuberculosis (TB) is a chronic contagious and deadly disease that spreads through the air. The number of patients infected with the disease is rising worldwide. In 2015, there were an estimated 10.4 million cases of TB with over 400,000 deaths resulting from TB disease among people living with HIV (World Health Organization [WHO] 2016). The proportion of TB cases living with HIV was highest in the African region despite the implementation of the standardized control strategy (Directly Observed Treatment Short course [DOTS]) (WHO 1996). Thus, the incidence, prevalence and mortality rates of TB in Africa have continued to be on increase (WHO 2016). In addition, the high susceptibility of human immunodeficiency virus-infected persons to this illness, the proliferation of HIV/AIDS, and the emergence of multidrug-resistant strains of Mycobacterium tuberculosis (MDR) are contributing to the worsening impact of this disease (Bloom 2002). Currently, the first-line regimen for treating TB is considered old and prescribes rifampicin (RMP) and isoniazid (INH) as component drugs. These antibiotics overtime affect the rise of multi-drug resistant (MDR). This increasing drug resistance incidence has led to an urgent need to develop new antitubercular drugs with low toxicity to overcome these limitations and increase the armamentarium of the existing therapeutic drugs.
In continuation of our phytochemical studies on Lophira lanceolata, other chemical constituents of the roots of this species have been investigated; newly isolated compounds could support the traditional uses of the plant, based on their medicinal interest. Although there is no information on the antitubercular activity of this plant or its constituents, previous reports on antitubercular activity of steroids and biflavonoids have designated immense prospects in this field (Camacho-Corona et al. 2009;Thakur and Gothwal 2015). It has been claimed that several plant natural products inhibit several species of mycobacteria (Okunade et al. 2004;Pauli et al. 2005). The Lophira genus through the richness of its components could be considered as an important source for new antitubercular agents.
We report herein the isolation and identification of a new biflavonoid derivative and other constituents from the methanol extract of Lophira lanceolata with their antitubercular properties.

General procedures
Melting points were uncorrected and were measured on a Mettler Toledo instrument. IR spectra were recorded on an Alpha FT-IR Spectrometer from Bruker, while 1D and 2D NMR spectra were obtained on a Bruker DRX 500 (500 MHz for 1 H and 125 MHz for 13 C spectra) spectrometer (Bruker, Rheinstetten, Germany) with chemical shifts reported in d (ppm) using TMS (d H ) as an internal standard. The HR-ESI-MS were obtained on LTQ-FT instrument (Thermo Scientific, Waltham, MA). UPLC-MS were measured by a Shimadzu UPLC-MS system using a L À column 2 ODS (I.D. 2.1 Â 100 mm, Chemical Evaluation and Research Institute, Tokyo, Japan), at a flow rate of 0.2 mL/min, a detection wavelength of 350 and 300 nm, and FMW (HCOOH/MeCN/H 2 O ¼ 1:12:87) as eluent, ESI þ 4.5 kV, ESI À 3.5 kV, 250 C. Optical rotations were measured on a Perkin-Elmer 341 polarimeter. Silica gel 60 (230-400 mesh E. Merck, Darmstadt, Germany) was employed for column chromatography (CC), the solvent mixing systems for elution were mainly CH 2 Cl 2 -MeOH with increasing polarity. Analytical thin layer chromatography (TLC) was carried out on Merck silica gel (Merck, Darmstadt, Germany) 60 PF 254 plates (0.25 mm layer thickness).

Plant material
The stem roots of L. lanceolata were collected at Balamba, near Yaound e in the Center region of Cameroon (4 26 0 . 00 00 N, 11 14 0 . 00 00 E) in July 2014 and authenticated by Mr. Victor Nana a botanist. A voucher specimen (N 45596 HCN) was deposited at the National Herbarium in Yaound e, Cameroon.

Extraction and isolation
Dried and powdered stem roots of L. lanceolata (854 g) was extracted with methanol (6.0 L, three times) at room temperature.

Mycobacterium tuberculosis strains
For the present study, the mycobacteria (M. tuberculosis) used were clinical isolated strains resistant to isoniazid and rifampicin codified AC45 and AC83, respectively (these strains were obtained from Sangmelima district's Hospital in South Region of Cameroon). The genetic profile of the resistance has been carried out at Laboratory for Tuberculosis Research (Biotechnology Centre, University of Yaound e I) through Line probe Assay method.

Preparation of extract and phytochemicals for antitubercular test
The activity of all phytochemicals (extract and pure compounds) against the aforementioned M. tuberculosis strains was tested using the microplate alamar blue assay (MABA) as described previously by Collins and Franzblau (1997) and Jim enez-Alleranes et al. (2003,2007). In 96-well microplates, all wells received 100 mL of supplemented Middlebrook 7H9 broth, then working metabolites solutions (100 mL) were poured into the first well of each row, from which twofold dilution series were made through the microplate column. The test inoculum (100 mL) was added to all testing wells, as well as to the drug-free control wells. The final concentration of DMSO in wells was 7% v/v.

Determination of minimum inhibitory concentration
MIC values were determined using the MABA; rifampicin and isoniazid were employed as references. The mycobacteria strain has been cultured at 37 C for two weeks in Middlebrook 7H9 (Himedia, Mumbai, India) supplemented with 0.05% (v/v), 2% glycerol and 10% OADC (oleic acid-albumin-dextrose-catalase of Liofilchem s.r.l, Roseto degli Abruzzi, Italy). The 96-well plates received 100 lL of Middlebrook 7H9 broth and serial dilution of compounds was made directly on the plate with drug concentrations of 0.244-250 lg/mL and 5000 to 4.882 lg/mL for extracts. Plates were covered and sealed with parafilm and incubated at 37 C for 14 days. Then, 40 lL of freshly prepared 1:1 mixture of alamar blue reagent and 7% Tween 80 (Himedia, Mumbai, India) was added to the plate and incubated for 24 h. A blue colour in the well was interpreted as no bacterial growth and pink colour was scored as growth. The MIC was defined as the lowest drug concentration, which prevented colour change from blue to pink. The results of antitubercular activity are depicted in Tables 2 and  3. Minimum inhibitory concentration (MIC) was determined by the broth micro-dilution method according to the guidelines of the Clinical and Laboratory Standards Institute (2011).

Determination of minimum bactericidal concentration
The minimal bactericidal concentration (MBC) determination, 50 lL of each wells which concentration was ! MIC was Table 1. 1 H and 13 C NMR spectroscopic data a of compounds 1a and 1b (500 and 125 MHz in DMSO-d 6 ) d in ppm, with those of reference (compound 8 c ). sub-cultured in 150 lL of Mbk 7H9 medium and incubated at 37 C for 10 days, then mycobacterial growth was carried out by addition of 40 lL of alamar blue. MBC was defined as the lowest concentration of extract at which no visible growth of the germ was observed. Minimum bactericidal concentration was determined by the broth micro-dilution method according to the guidelines of the CLSI 2011.

Results and discussion
Crude MeOH extract from Lophira lanceolata was evaluated against two strains of M. tuberculosis. The strain AC45 of M. tuberculosis was more susceptible to MeOH root extract than AC83, with MIC values of 312.5 and 1250 mg/mL (Tables 2  and 3), respectively. Suggesting that the strain AC83 is more resistant than AC45. This extract was chromatographed on a column of silica gel eluted with CH 2 Cl 2 -MeOH in increasing polarity to afford compounds 1-7. Compounds 2-7 were identified as known; stigmasterol (2) (Ghogomu et al. 1989a(Ghogomu et al. , 1989b. Those structures were deduced from comparison of their spectral data with those reported in the literature. Compound 1 was obtained as an orange amorphous powder and a single peak in the UPLC profile. However, two sets of signals with almost the same intensities in the 1 H NMR spectrum suggested that 1 was a mixture of two unseparated components, 1a and 1b, which were present in an approximate 2:1 ratio. This is strengthened by the HR-ESI-MS recorded in the negative and    Figure 2. The 13 C NMR of 1a showed the resonances of 30 carbons consisting of 10 quaternary aromatic carbons at d C 114.3 (C-12), 115.8 (C-4), 130.9 (C-19), 133.1 (C-25), 156.0 (C-22), 157.6 (C-28), 159.0 (C-9), 165.0 (C-7), 165.1 (C-13) and 165.6 (C-15), along with two carbonyl carbons at d C 190.6 and 205.4, fourteen sp 2 methine carbons (d C 102.6-132.9), two sp 3 carbons at d C 71.5 (C-1) (methylene) and at d C 49.4 (C-2) (methine). Moreover, signals of two sp 3 methine carbons at d H 49.1 (C-10) and 55.3 (C-18) indicated the presence of partial structure that was in accordance with ethyl group as shown in Figure 1. The cross-peaks of H-1 with H-2 and H-10 in the 1 H-1 H COSY spectrum indicated the presence of a partial structure that was consistent with a chromone derivative group. All these signals were in good agreement with those of the Lophirone A skeleton (Ghogomu et al. 1987). The complete assignment of all proton and carbon resonances was achieved after careful analysis of COSY, HSQC and HMBC techniques. The chemical structure of 1a was established with the aid of the HMBC spectrum; the chromone moiety (ring AC) was assigned by the long range correlations between H-2 (d H 2.76) and C-1 (d C 71.5), C-3 (d C 190.6) and C-10 (d C 49.1). Additional correlations concerned H-1 (d H 3.39) and C-2 (d C 49.4), suggesting that the ring C of the chromone moiety was hydrogenated, and H-5 (d H 7.60) with C-3 (d C 190.6), C-4 (d C 115.8) and C-9 (d C 159.0) indicating the attachment of ring C to C-10. Further elements from HMBC predict correlations between H-17 and C-11 (d C 205.4), C-15 (d C 165.6), and between H-18 with the ketone carbonyl of C-11 (d C 205.6) corroborating the presence of an ethyl moiety in 1a. This ethyl substructure was linked to C-18 of the molecule based on the HMBC correlations between H-20 (d H 7. 28), H-26 (d H 7.29), with C-18 (d H 55.3) as shown in Figure 2.
In addition, further elements from NOESY between H-10 (d H 4. 04) and H-17 (d H 7.39) indicated the attachment of ethyl group to benzoyl moiety (ring A 0 ). Moreover, HMBC correlation between an hydroxyl group at d H 12.6 with carbon at d C 165.1 (C-13), other hydroxyl groups at d H 9.48 and d H 9.07 correlating with carbons at d C 157.6 (C-28) and 156.0 (C-22) respectively indicated the attachment of these hydroxyls to ring A 0 , B and B 0 separately. The relative configuration of carbon C-2 was determined from coupling constant of H-2 proton and from NOESY results (Figure 2). A 10.9 Hz value of this proton suggested the trans-relationships with H-10 (10.6 Hz). The NOESY correlation observed between H-2 (d H 2.76) and H-18 (d H 3.75) indicated the common orientation on the same side of the molecule for these protons. The structure of 1a was therefore assigned as (2R Ã )-2,3-dihydrolophirone A. The 1 H and 13 C NMR spectra of 1b were very similar to those of 1a. In the 1 H NMR spectrum, signals due to two 1,2,4-trisubstituted benzene, two 1,4-disubstitued benzene, two methine protons involved in trans-relationships at d H 3.80 (1H, d, J ¼ 12.0 Hz, H-18) and 4.98 (1H, d, J ¼ 12.0 Hz, H-10), and one sp 2 methine singlet at d H 8.06 (H-1) were evident, and these signals were in good agreement with those reported in literature (Ghogomu et al. 1987). The slight difference in the chemical shifts between H-18 for 1a (d H 3.75) and 1b (d H 3.80) might be explained by the dehydrogenation of chromone moiety for compound 1a. However, 13 C NMR spectrum showed significantly different chemical shift values from those of 1a from chromone moiety. The appearance of carbons at d C 155.6 (C-1) and d C 125.4 (C-2) in 1b instead of the methylene and methine carbons (d C 71.5, C-1 and 49.4, C-2) in 1a was evident. The structure of 1b was also determined through the interpretation of the HMBC spectrum, which showed long-range correlations from H-1 (d H 8.06) to C-2 (d C 125.4), C-3 (d C 174.5) and C-9 (d C 159.0) and from H-10 (d H 4.98) to C-2 (d C 125.4) and C-3 (d C 203.0). All these results were very similar to those of reported lophirone A. Furthermore, the NOESY experiment of compound 1b showed cross-peaks from proton signal at d H 4.98 (H-10) to proton signal at d H 7.39 (H-17), indicating the attachment of the benzoyl moiety (nucleus A 0 ) to methine carbon (C-10) of ethyl group as described for compound 1a. Therefore, the structure of 1b was identified as the known lophirone A.
In order to assess the biological effects of the compounds isolated from MeOH extract, antimycobacterial assays were performed. The antimycobacterial effect was assessed by testing compounds against two resistant strains of M. tuberculosis. Moreover, the methanol extract of L. lanceolata showed higher antimycobacterial activity in M. tuberculosis AC45 (MIC 312.5 mg/mL) than M. tuberculosis AC83 (MIC 1250 mg/mL) in the MABA assay, and it indicates that these results can be associated with differences in pathogenesis and virulence of the two strains or different pathogenic phenotypes between AC45 and AC83.
Although plants are an excellent source of diverse molecules, there is no plant based drug for the treatment of TB (Tiwari et al. 2013). The present study has brought out the antitubercular activities of compounds 1, 2, 4, 5, 6 and 7 against two clinical isolate strains of M. tuberculosis AC 45 and AC 83 in MABA assay. From these results (listed in Tables 2 and 3), compounds 1 and 2 exhibited the most potent antitubercular activity at MIC 31.25 and 15.75 mg/mL, respectively, against M. tuberculosis AC45. There are few studies describing the antimycobacterial activity of biflavonoids especially chalcone derivatives and steroids. In this respect, stigmasterol and the biflavonoids (dihydrolophirone A and lophirone A) can be considered as promising isolated compounds according to Cantrell et al. (2001). Compound 6 possessed antitubercular activity at MIC 250 mg/mL against the two tested strains and was considered inactive (Tables 2 and 3). In addition, compound 2 was more active than analogues (4 and 6) against M. tuberculosis suggesting an impact of polar substituents which showed a decreased inhibitory potency for antimycobacterial effect, implying an important role of the substitution patterns at C-3 (Figure 1). In assessing structure-activity relationships, the higher activity of compound 1 (MIC 31.25 mg/mL) compared to that of compound 8 (a pure sample of lophirone A, MIC 62.5 mg/mL) might be explained by a higher lipophilicity of the pair (1a and 1b) which destabilizes the cytoplasmic membrane of microorganism, thereby reducing the pH gradient across the membrane and induces the cell death (Zengin and Baysal 2014). Nevertheless, the antimycobacterial activity of 1a was not well evaluated because it appeared as mixture knowing that the position of hydroxyl group in phenolic ring is not recognized to strongly influence the degree of antibacterial activity (Shan et al. 2005), this is exemplified by the MIC values of compounds 7 and 8 at 62.5 mg/mL. However, according to Castellar et al. (2011), in all researches, no correlation on structure activity of flavonoids against mycobacteria could be drawn. Ergosterol (3) could not be tested due to its insufficient quantity. The antimycobacterial properties of some derivatives such as ergosterol-5,8-endoperoxide isolated from Ajuga remota are documented (Cantrell et al. 1999). The activity detected for 2 (MIC 15.75 mg/mL) and 4 (MIC 62.5 mg/mL) is twofold better than those previously reported (Thakur and Gothwal 2015). No antimycobacterial activity was reported earlier for the test compounds except for compound 4 which showed similar results against the resistant strain to isoniazid AC 45 (Evina et al. 2017;Tiam et al. 2017).
It was observed that all the compounds possessed very weak and/or no activity against M. tuberculosis AC83. Regarding the MBC values of both extracts tested (Tables 2 and 3), it seems that they could be similar to their MIC values against the tested organisms. These results also imply that all tested compounds exhibit bactericidal action against the studied strains (Peterson and Shanholtzer 1992).
Many plants of our continent are used locally in the treatment of TB, but their anti-tubercular properties have not been investigated due to the lack of appropriated material and scientific facilities. Therefore, it is imperative for African governments in general and in Sub-Sahara Africa in particular to make an effort to fund anti-tubercular drug research.
This study is the first report of antimycobacterial activity of constituents isolated from L. lanceolata against M. tuberculosis.

Conclusions
The data illustrate that MeOH extract of the roots of L. lanceolata has exhibited less pronounced in vitro antimycobacterial activities and a number of active plant derived compounds belonging mainly to two chemical classes, steroid and bioflavonoid derivatives, exhibited good to moderate antimycobacterial activities. These bioactive phytochemicals can be used or utilized as promising candidate substances to develop new tools in antitubercular research. Additionally, there is a strong positive correlation between the antimycobacterial activity results and the ethnomedical/traditional usage on this plant against TB and TB-related diseases.