Bioassay Guided Isolation of Antibacterial Compounds from Andrographis paniculata

Problem statement: Chronic disease-causing bacteria of medical import ance have developed resistance to antibiotics, hence, necessi tating distinct and constant need for safe and efficient therapeutic agents. Plants are considered potent candidate for this aim. A way out of reduci ng antibiotic resistance and adverse effects on host is the employment of antibiotic resistance inhibitors of plant origin. Approach: About 5 kg pulverized Andrographis paniculata whole plant was macerated with MeOH at room temperature to get 305 g freeze d ri d MeOH extract. The bioautography of MeOH extract using Staphylococcus aureus and Proteus mirabilis as indicator organisms revealed the presence of two potent antibacterial compounds . MeOH extract was further fractionated and purified by silica gel column chromatography which led to th e isolation of a diterpene lactone and an entlabdane diterpene glycoside upon crystallization wi th absolute ethanol. Results: Two antibacterial compounds viz., 3-Oβ D-glucosyl-14-deoxyandrographolide and 14-deoxyand rographolide were successfully isolated and characterized. Their stru ctu es were exclusively elucidated through spectroscopic methods (UV, IR, Hand C NMR). Conclusion: A. paniculata possesses antibacterial activity and could be potential source of a new cla ss of antibiotics that might be useful for infectio us disease chemotherapy and control.


INTRODUCTION
Andrographis paniculata (Burm.f.) Wall. ex Nees., (Family-Acanthaceae) (English name-King of Bitters, Local malay name-Hempedu bumi) is an annual herbaceous plant and is extensively cultivated in Southern Asia, China and some parts of Europe. In traditional medicine, A. paniculata is widely used to get rid of body heat, dispel toxins from the body, prevent common cold, upper respiratory tract infections including sinusitis and fever (Gabrielian et al., 2002) and as an antidote against poisons of snakes and insects (Samy et al., 2008). A. paniculata has been reported to exhibit various mode of biological activities in vivo as well as in vitro viz., antibacterial (Singha et el., 2003;Mishra et al., 2009;Parvataneni and Koduru, 2010;Roy et al., 2010;Abubakar et al., 2011), antiviral (Wiart et al., 2000), anti-inflammatory (Wen et al., 2010), antihuman immunodeficiency virus (HIV) (Calabrese et al., 2000), immunomodulating/immunostimulatory (Iruretagoyena et al., 2005) and anticancer (Li et al., 2007;Geethangili et al., 2008). The characteristic secondary metabolites encountered in this plant have considerably enhanced its importance in the arena of medicinal plants. It is specifically rated high in therapeutic action in curing liver disorders, common cough and colds in human (Niranjan et al., 2010).
Biodiversity is a precious source for modern biotechnology. It is a source which potentially holds innovative and sustainable solutions to a broad range of important problems for modern society. Improved cooperation between the natural product chemists and the microbiologists is a constructive step to speed up the process of evaluating these potentialities. Moreover, microbiologists and natural product chemists in tropical countries including Malaysia, with the richest flora and fauna placed right at their door step have a very central position. They are essential for building up international scientific cooperation, with the objective of expanding our understanding of biological and biochemical diversity and based on this bringing forward more biological solutions. The entire process is built on a principle of fairness and equity in sharing of the benefits and respecting the State's sovereign right to its own resources. After structure elucidation of secondary metabolites, it is considered crucial to know how useful these molecules might be in terms of medicinal properties. During the past 40 years, numerous novel compounds have been isolated from different plants and marine organisms and many of these have been reported to have core biological activities, some of which are of interest from the point of view of potential drug development (Gerald, 2001;Houghton, 2001). In this context, A. paniculata could be a potential source to develop new efficacious drugs. A. paniculata has already been reported for its significant antibacterial potential (Singha et al., 2003, Mishra et al., 2009Parvataneni and Koduru, 2010;Roy et al., 2010;Abubakar et al., 2011). However, no attempt has ever been made to identify and isolate active principles responsible for unleashing its antibacterial activity. Identification and isolation of active principles from A. paniculata might prove promising antibacterial agents through foreseeable future endeavors. Hence, this study is a conscientious attempt to identify and isolate pure antibacterial compounds from the methanol extract of the whole plant of A. paniculata through bioassay guided isolation method.

Preparation of methanol (MEOH) extract:
The fresh whole plant (15 kg) of A. paniculata was cleaned and dried in a protech laboratory air dryer (LDD-720) at 40°C for 7 days and pulverized to powdered form (5.6 kg, 37.33%) using Fritsch Universal Cutting Mill-PULVERISETTE 19-Germany. It was then stored in a desiccator at 2°C until further use. The air dried powder of whole plant (5 kg) of A. paniculata was extracted by macerating in double distilled methanol (20.0 L) at room temperature for 24 h, filtered and evaporated under reduced pressure. The whole process was repeated thrice to ensure maximum yield of methanol soluble compounds from the plant powder. Each time, filtrate was evaporated under reduced pressure (Buchi Rotary Evaporator, R-210) and combined. The dark blackish green residue so obtained was further freeze dried to yield 305 g (6.1%) MeOH extract and was stored at 2°C in a labeled sterile bottle until further antibacterial evaluation and isolation of antibacterial compounds.

Preparation of standard bacterial suspensions:
The average number of viable, S. aureus, S. pyogenes, M. luteus, P. mirabilis and P. aeruginosa organisms per mL of the stock suspensions was determined by means of the surface viable counting technique (Hedges, 2002). About 10 7 -10 8 CFU/mL was used. Each time, a fresh stock suspension was prepared; the experimental conditions were maintained constant so that suspensions with very close viable counts could be obtained successfully.

In vitro antibacterial activity test for MeOH extract:
The cup-plate agar diffusion method was adopted according to Kokoska et al. (2002) to assess the antibacterial activity of the MeOH extract. 0.6 mL of standardized bacterial stock suspensions corresponding to 10 7 -10 8 CFU/mL was thoroughly mixed with 60 mL of sterile nutrient agar. 20 mL of the inoculated nutrient agar were distributed into sterile labeled Petri dishes. The agar was left to set at room temperature and in each of these plates, 3 cups 6 mm in diameter were punched using a sterile cork borer allowing at least 30 mm between adjacent wells and the agar discs were removed. Fixed volumes of the plant extract (1000, 500-250 µg mL −1 ) were then introduced into each wells using microtiter pipette and allowed to diffuse at room temperature for 2 h. In separate wells, 30 µg each of gentamicin and vancomycin were added as positive controls whereas 10% DMSO was taken as negative control. The plates were then incubated in the upright position at 37°C for 24 h. Three replicates were carried out for the extract against each of the test organism. After incubation the diameter of the results and growth inhibition zones were measured, averaged and the mean values were recorded.

Determination
of Minimum Inhibitory Concentration (MIC): Micro broth dilution method was used for the determination of MIC values for each plant extract showing antibacterial activity against test pathogens (EUCAST, 2003;Jana et al., 2004). Serial dilutions of the extracts were carried out in 10% DMSO (which had no inhibitory activity against test microorganisms) to make 500 µg mL −1 final concentration, this was then two fold serially diluted by adding to the broth media in a 96-wells microtiter plates to obtain 250, 125, 62.5, 31.3, 15.6 and 7.81 µg mL −1 . Thereafter, 100 µL inoculum (10 8 CFU mL −1 ) was added to each well. Bacterial suspensions were used as negative control, while broth containing standard drug (vancomycin and gentamicin) were used separately as positive controls. The microtiter plates were incubated at 37°C for 24 h. Each extract was assayed in duplicate; one was kept for incubation while the other was kept at 4°C for comparing the turbidity in the wells of microtiter plate. The MIC values were taken as the lowest concentration of the extracts in the well of the microtiter plate that showed no turbidity after incubation. The turbidity of the wells in the microtiter plate was interpreted as visible growth of microorganisms.

Antibacterial Activity Index (A b I):
Antibacterial index (A b I) of MeOH whole plant extract of A. paniculata was calculated separately as the average value of zone of inhibition against the Gram-positive and Gram-negative bacteria, respectively (Mbwambo et al., 2007).

Bioassay guided isolation:
To sterilized 8×4 cm silica gel 60 F 254 TLC plates (Merck, Germany), 10 µL of MeOH extract was applied as small spots and the plates were developed in Hexane: Acetone (2:1) in duplicate (a TLC plate was used as the bioautogram while the other served as a chromatogram for reference in comparison with the bioautograph). The TLC plates were dried in an oven at 25°C for 7 h to activate the plates by absorbing the moisture content from the plates and removing all residual solvents (Veronica and Scott, 2005).
Bioautography technique: S. aureus and P. mirabilis were used as the indicator microorganisms for the bioautography of antibacterial compounds from the MeOH extract of A. paniculata. 200 µL each from broth cultures of S. aureus and P. mirabilis (adjusted to 10 8 CFU mL −1 ) were mixed with 35 mL molten Mueller-Hinton agar (MHA) at 30°C separately. The suspensions of agar and bacteria were spread aseptically onto the already developed TLC plates in square Petri dishes (8×4 cm), allowed for 30 min to solidify and the plates were incubated at 37°C for 24 h. At the end of incubation time, 0.5% piodonitrotetrazolium violet (INT) was uniformly sprayed over the TLC plates. The active antibacterial compounds in the plant extracts formed clear zones of inhibition on the TLC plates against a deep pink back ground of bacterial growth, allowing the chromatographic Retention factors (Rf) observation by viewing under UV light at 254 nm (Short wave) and 366 nm (Long wave) and comparing with the reference chromatogram (already sprayed with vanillin reagent and heated at 120°C) to note the antibacterial compounds. Vanillin reagent was prepared by dissolving 15 g of vanillin in ethanol (250 mL) and H 2 SO 4 (2.5 mL). Vanillin reagent gives different colored spots with different compounds on TLC plate upon heating at 120°C (Rahalison et al., 2007).
3  (Poonam et al., 2010) of the same compound, AB-2 was unambiguously identified as 14deoxyandrographolide (Fig. 2). Minimum Inhibitory Concentration (MIC) of Isolated Compounds: The minimum inhibitory concentrations of the isolated compound AB-1 and AB-2 were determined using the agar dilution method following the standard protocol of the European Committee for Antimicrobial Susceptibility Testing (EUCAST, 2003). The compounds were dissolved in 10% DMSO and 2-fold diluted in MHA to obtain 250, 125, 62.5, 31.3, 15.6 and 7.81 µg mL −1 . The mixture of the media and compounds were thoroughly mixed and poured onto pre-labeled sterile Petri dishes on a level surface. Additional Petri dishes containing only the growth media were prepared in the same way so as to serve for comparison of growth of the respective bacteria. The plates were then set at room temperature and dried. The suspensions of the respective bacteria (corresponding to 10 8 CFU mL −1 ) were inoculated onto the series of agar plates. The plates were then incubated at 37°C for 24 h. The experiments were performed in duplicate and MIC values expressed as the lowest concentration of the plant extracts that produced complete suppression of colony of respective bacteria.

Statistical analyses:
The experimental results were expressed as mean ± Standard Deviation (STD) of triplicate experiments. Statistical differences between the antibiotics and inhibition zones formed by the plant extracts were detected by Analysis Of Variance (ANOVA) using SPSS 19.0 statistical software (SPSS, Chicago, Illinois, USA) followed by the Tukey test for multiple comparisons between means. P values lower than 0.05 (p<0.05) were considered significantly different whereas P values lower than 0.01 (p<0.01) were considered highly significant.

RESULTS
The results of the cup-plate agar diffusion method showed that MeOH extract of the whole plant of A. paniculata do possess antibacterial activity against all 5 bacteria taken into account in vitro (Table 1). Maximum antibacterial activity was observed against S. aureus (19.67 ± 0.76 mm) at 1000 µgmL −1 and the lowest activity was detected against P. aeruginosa (7.00 ± 1.50 mm) at 250 µgmL −1 . MIC values for MeOH extract and isolated compounds are shown in Table 2. MIC of MeOH extract ranged from 125-250 µg mL −1 with the highest MIC value exerted by the extract against S. pyogenes, P. mirabilis and P. aeruginosa (250 µg mL −1 ) and the lowest against S. aureus and M. luteus (125 µg mL −1 ). The bioassayguided isolation of antibacterial compounds from MeOH extract led to the identification and subsequent isolation of an ent-labdane diterpene glycoside (AB-1) and a diterpene lactone (AB-2) as the main active principles. Both compounds were active against S. aureus (Fig. 5-6) and P. mirabilis which were used as indicator organisms by the bioautography technique on TLC plates forming clear zones against pink background of the living microorganisms when compared to the reference chromatogram. MIC values for both isolated compounds ranged from 15.6-250 µg mL −1 . Highest MIC value was exerted by compound AB-1 against P. aeruginosa (250 µg mL −1 ) while the lowest was exerted by compound AB-2 against S. aureus (15.6 µg mL −1 ), however, no activity was exerted by compound AB-1 against M. luteus ( Table  2). The MeOH extract's antibacterial index (A b I) was found to be the best against Gram-positive strains tested as compared to the Gram-negative strains with mean inhibition zones of 13.9 mm and 10.4 mm, respectively (Table 3).

DISCUSSION
Antibiotics provide the main basis for the therapy of chronic bacterial infections. However, the high genetic variability of bacteria enables them to rapidly evade the action of antibiotics by developing antibiotic resistance. As resistance becomes more common, there becomes a greater need for alternative treatments. However despite a push for new antibiotic therapies there has been a continued decline in the number of newly approved drugs (Bachi, 2002;Nagi et al., 2010). According to the World Health Report on infectious diseases 2000, overcoming antibiotic resistance is the major issue of the WHO for the next millennium. Hence, the last decade witnessed an increase in the investigations on plants as a source of human disease management (Paul et al., 2006). A. paniculata is common throughout Southeast Asia and India and is extensively used by traditional healers for the treatment of a wide variety of ailments (Coon and Ernst, 2004). The antibacterial activity of A. paniculata extracts are well known (Singha et al., 2003;Mishra et al., 2009;Parvataneni and Koduru, 2010;Roy et al., 2010;Abubakar et al., 2011). Whilst many studies have isolated and characterized A. paniculata compounds, no study has ever determined the antimicrobial activity of isolated compounds so far. In the present experiment, the MeOH extract of the whole plant of A. paniculata showed broad spectrum antibacterial activity. 3-O-β-Dglucosyl-14-deoxyandrographolide and 14deoxyandrographolide were isolated as active principles, which may serve as lead for the development of new pharmaceuticals that might address the unmet therapeutic needs. The obvious fields where the natural product chemist can harvest benefits from a cooperation with the microbiologists are development of bioassay for efficient monitoring of isolation and purification of new compounds; bioassay fingerprinting to help early de-selection of known compounds (hereby supplementing the chemical data and giving additional avenues for tapping into the computerized data bases); activity spectrum to help de-selecting the very toxic compounds; obtaining a sharper focus in the natural product chemistry work on biologically active compounds. Novel and potentially useful may be of more interest than to go exclusively for just novelty (Houghton, 2001). Bio-autography provides more information about plant compounds requires a smaller weight of sample and can be used for the bioassayguided isolation of biological active compounds, simplifying the process of the identification and isolation of the active compounds (Rahalison et al., 2007).
The antibacterial activity measured by the cupplate agar diffusion method was more pronounced on the Gram-positive bacteria (S. aureus, M. luteus and S. pyogenes) than the Gram-negative bacteria (P. mirabilis and P. aeruginosa). Gram-positive bacteria were more susceptible to growth inhibition by MeOH extract of A. paniculata whole plant. The greater susceptibility of Gram-positive bacteria has been previously reported for South American (Paz et al., 1995), African (Kudi et al., 1999) and Australian (Palombo and Semple, 2001) plant extracts. Susceptibility differences between Grampositive and Gram-negative bacteria may be due to cell wall structural differences between these classes of bacteria. Gram-negative bacteria have an outer phospholipid membrane carrying the structural lipopolysaccharide components. This makes the cell wall impermeable to antimicrobial chemical substances.
The Gram-positive bacteria tested were more susceptible to the plant extracts because it is well known that all Gram-positive bacteria have an outer peptidoglycan layer which is not an effective permeability barrier. The cell walls of Gram-negative organisms are more complex in lay out than the Grampositive ones acting as a diffusion barrier and making them less susceptible to the antimicrobial agents than are Gram-positive bacteria (Nikaido, 2003). In the present study, after the first chromatography of the MeOH extract of the whole plant of A. paniculata on a silica gel column, the antibacterial activity of the collected fractions were tested against S. aureus and P. mirabilis using bio-autography on a TLC plate. This revealed that all the fractions except nine were active against S. aureus and P. mirabilis. Isolation of these compounds in pure form was achieved by repeated washing of the crystalline matter off the green coloring material with toluene and repeated recrystallization with absolute ethanol and final washing of the crystals with cold methanol. The purity of the sample at every stage of recrystallization was monitored through TLC and HPLC.

CONCLUSION
The TLC bioautography-guided strategy was used to separate the antibacterial compounds from the MeOH plant extract. Two antibacterial compounds were successfully isolated from MeOH extract of the whole plant of A. paniculata for the first time. The isolated 3-O-β-D-glucosyl-14-deoxyandrographolide and 14-deoxyandrographolide demonstrated significant antibacterial activities against the selected microbial strains. Quantitative HPLC and TLC analysis confirmed that these isolated compounds are predominant components in whole plant MeOH extract, indicating their significant contribution to the overall antibacterial activity. Further investigation of the activities of these compounds and their potential use in the treatment of bacterial diseases are still warranted. This is the first report on the isolation of antibacterial components through bioassay-guided isolation from A. paniculata.