Antimicrobial Properties of Ethanolic and Methanolic Extracts of Finger millet ( Eleusine coracana (L.) Gaertn.) Varieties Cultivated in Sri Lanka

The emerging significance of natural antimicrobial agents creates an imperative need to identify novel plant sources with antimicrobial activities. The objective of the present study was to evaluate antibacterial, antifungal and β-lactamase enzyme inhibitory activities of ethanolic and methanolic extracts of Ravi, Rawana and Oshadha finger millet varieties. Flours of whole grains of the finger millet varieties were extracted with absolute ethanol and methanol separately. Antibacterial activities against six antibiotic-sensitive and four antibiotic-resistant pathogenic bacterial strains were evaluated using the resazurin reduction assay. Antifungal activities against six antimicrobial-sensitive pathogenic fungal strains were evaluated using the agar tube dilution method. β-Lactamase enzyme inhibitory activity was evaluated using a standard method. Both ethanolic and methanolic extracts of the three finger millet varieties showed dose-dependent inhibitory activities against the tested antibiotic-sensitive and antibiotic-resistant bacterial strains while exhibiting high inhibitions against Gram-positive antibiotic-sensitive bacterial strains when compared to Gram-negative antibiotic-sensitive bacterial strains. The findings revealed the antibacterial potential of both ethanolic and methanolic extracts of the three finger millet varieties against antibiotic-sensitive Staphylococcus aureus (ATCC ® 6538™) and Bacillus subtilis (ATCC ® 23857™) strains and the minimum inhibitory concentrations of the extracts against S. aureus and B. subtilis were found to be 2.1 and 1.8 mg/ml, respectively. However, none of the extracts can be considered as significantly active against the tested antibiotic-sensitive and antibiotic-resistant bacterial strains when compared to the standard drugs. In addition, none of the extracts can be considered as active against the tested fungal strains at the tested concentrations. Nevertheless, all extracts showed more activities against the tested bacterial strains when compared to the tested fungal strains. Since all extracts showed less than 40% β-lactamase inhibitory activities even at 2 mg/ml concentration, they do not qualify as promising sources of β-lactamase inhibitors at the tested concentration.


INTRODUCTION
The incidences of resistance of pathogenic microorganisms to antimicrobial agents are increasing at an alarming rate (Khan et al., 2009;Saravanan and Parimelazhagan, 2014;Khan et al., 2015). Indiscriminate use of antimicrobial agents has become the major factor for the emergence and dissemination of multi-drug resistant strains of several groups of microorganisms.
Consequently, the efficiencies of the currently available antimicrobial agents are declining (Khan et al., 2009;Silva et al., 2015). This has become a great threat in the effective management of infectious diseases and has become a global concern to public health (Khan et al., 2015). In addition, many of the currently available antimicrobial agents are expensive and sometimes induce serious side effects (Silva et al., 2015). The antimicrobial agents commonly used for food preservation are mostly synthetic chemicals and now there is an emerging demand for natural products with antimicrobial activities for preserving foods (Banerjee et al., 2012). Therefore, identifying novel therapeutically active antimicrobial drugs and edible antimicrobial agents, preferably from natural sources, is receiving attention worldwide (Khan et al., 2009;Banerjee et al., 2012;Khan et al., 2015;Silva et al., 2015).
Kingdom Plantae has been regarded as an important source of natural products with antimicrobial activities (Su et al., 2014). Antimicrobial activities of plants are attributed to the presence of bioactive components such as polyphenols including flavonoids (Khan et al., 2015). Several authors have reported the antimicrobial effects of plant polyphenols and have recognized plant polyphenols as potential inhibitors of the growth of a wide spectrum of microorganisms (Viswanath et al., 2009;Banerjee et al., 2012;Saravanan and Parimelazhagan, 2014;Su et al., 2014;Singh et al., 2015).
Several studies have confirmed the presence of phenolic compounds in cereal grains. Polyphenols are the most important phytochemicals of millets due to their nutraceutical potentials (Singh et al., 2015). Finger millet (Eleusine coracana (L.) Gaertn.) grains are rich in polyphenols when compared to other cereals such as rice, wheat, maize and barley (Viswanath et al., 2009;Chandra et al., 2016). Several therapeutic properties of finger millet including antibacterial and antifungal activities have been reported previously (Mathanghi and Sudha, 2012;Srivastava and Sharma, 2012;Singh et al., 2015). Finger millet is commonly cultivated and consumed in Sri Lanka since ancient times (Kumari et al., 2015). It is considered as a medicinal plant and used in the treatment and management of several diseases in the traditional system of medicine (Hettiarachchige and Edirisuriya, 2013). However, there is a scarcity in scientific information on antibacterial and antifungal activities of the finger millet varieties which are commonly cultivated in Sri Lanka. Since there is a genetic diversity in finger millet varieties worldwide and there are varietal differences in composition with respect to every constituent (Kaluthanthri and Dasanayaka, 2016), it is important to study the antimicrobial potential of the finger millet varieties which are commonly cultivated in Sri Lanka. Therefore, the objective of the present study was to evaluate ethanolic and methanolic extracts of Sri Lankan finger millet varieties for antibacterial activities against six antibiotic-sensitive and four antibiotic-resistant pathogenic bacterial strains, antifungal activities against six antimicrobial-sensitive pathogenic fungal strains and β-lactamase enzyme inhibitory activities.

Sample collection and preparation
Finger millet varieties which are recommended for cultivation by the Department of Agriculture, Sri Lanka namely Ravi, Rawana and Oshadha were collected from Field Crop Research and Development Institute (FCRDI), Mahailuppallama, Sri Lanka. These varieties were cultivated in experimental plots in the Low Country Dry Zone, at FCRDI, Mahailuppallama and the seeds were certified by the Seed Certification Service of Department of Agriculture, Sri Lanka. Finger millet seeds were dehulled (TM 05C, Satake Corporation, Japan) and flours from whole grains were obtained by milling (Pulverisette 14, Fritsch, Germany) and passing through a 0.5 mm sieve. Flours of whole grains were extracted with absolute ethanol and methanol separately. Whole grain flour (100 g) was extracted with the solvent (400 ml) by soaking overnight at room temperature (28 ± 2 o C) while stirring continuously using a magnetic stirrer and centrifuging (Sigma 318 K, Sigma Laborzentrifugen GmbH, Germany) at 4000 rpm for 20 min. The supernatant was collected separately and the residue was re-extracted twice under the same conditions. The supernatants were combined and evaporated to dryness under reduced pressure at 40 o C using a rotary evaporator (R-114, Büchi Labortechnik AG, Switzerland). The solvent-free extracts were stored in airtight glass containers at -20 o C until use for the analysis.

Determination of antibacterial
activity Antibacterial activity was evaluated using the resazurin reduction assay as described by Osaka and Hefty (2013) with slight modifications. Using aseptic techniques, a glass bottle containing sterilized Mueller-Hinton broth (10 ml) was inoculated with one colony of the bacterial strain, incubated at 37 o C for 24 hrs and adjusted to the 0.5 McFarland turbidity index by diluting with Mueller-Hinton broth. A known amount of finger millet extract was dissolved in DMSO. Dissolved finger millet extract (10 µl), Mueller-Hinton broth (183 µl) and bacterial suspension (7 µl) were added to a tissue culture treated 96-well microplate, covered with a parafilm and incubated at 37 o C for 20 hrs. Parafilm was removed, the microplate was covered with a sealing film and preplate reading was recorded at 570 nm and 600 nm using a microplate reader (SpectraMax M5e, Molecular Devices Inc., USA). Resazurin sodium salt (20 mg) was dissolved in 100 ml of sterile distilled water and filtered. In dark, the prepared dye (20 µl) was added to the wells, the microplate was covered with aluminium foil and incubated at 37 o C in a shaking incubator (80 rpm) for 2 hrs. Aluminium foil was removed, the microplate was covered with a sealing film and absorbances were recorded at 570 nm and 600 nm. Ampicillin and ofloxacin were used as standard drugs. DMSO was used as the control. Inhibition percentage was calculated using the following equations.
Where, εOXλ1 is the molar extinction coefficient of oxidized form of resazurin at 570 nm (80,586), εOXλ2 is the molar extinction coefficient of oxidized form of resazurin at 600 nm (117,216), ASλ1 is the absorbance of sample at 570 nm, ASλ2 is the absorbance of sample at 600 nm, ACλ1 is the absorbance of control at 570 nm and ACλ2 is the absorbance of control at 600 nm.
Antibacterial activities of the extracts were evaluated at three different concentrations (150, 600 and 2400 µg/ml) and the extracts which showed no color change from blue to pink or colorless were subjected to the determination of minimum inhibitory concentration (MIC) values as described by Coronado-Aceves et al. (2016). To determine the MIC value, the resazurin reduction assay was performed using seven different concentrations (600, 900, 1200, 1500, 1800, 2100 and 2400 µg/ml) of finger millet extract and the lowest concentration at which the blue color of the dye remained unchanged was recorded as the MIC value.

Determination of antifungal activity
Antifungal activity was evaluated using the agar tube dilution method as described by Atta-ur-Rahman et al. (2001), Uddin et al. (2011) and Thadhani et al. (2012). A known amount of finger millet extract was dissolved in DMSO. Using aseptic techniques, dissolved finger millet extract (66.6 µl) was added to a glass tube containing sterilized Sabouraud dextrose agar (4 ml) and the tube was allowed to solidify in a slanting position at room temperature to prepare an agar slant. From each extract, four agar slants were prepared having four different concentrations (400, 800, 1600 and 3200 µg/ml). Each tube was inoculated with a piece of fungus having a diameter of 4 mm taken from a seven-day old culture of fungus and incubated at 27 o C for 7 days. For A. niger, amphotericin B was used as the standard drug and miconazole was used for the other fungal strains. DMSO was used as the control. Growth of the fungus was evaluated by measuring the linear growth (in mm) and growth inhibition percentage was calculated with reference to the control using the following equation.
Inhibition % = 100 -Linear growth in test (mm) × 100 Linear growth in control (mm) 2.6 Determination of β-lactamase enzyme inhibitory activity β-Lactamase enzyme inhibitory activity was determined according to the method described by Atta-ur-Rahman et al. (2001) with slight modifications. A known amount of finger millet extract was initially dissolved in DMSO and diluted with 100 mM phosphate buffer (pH 7.4). Diluted finger millet extract (50 µl) was added to a 96-well microplate, mixed with 40 µl of penicillinase enzyme (1000 U/ml) and incubated at 25 o C for 20 min. Preplate reading was taken at 482 nm using a microplate reader (SpectraMax Plus 384 , Molecular Devices Inc., USA). The reaction was initiated with the addition of 2 mM nitrocefin (10 µl) and the change of absorbances was monitored at 482 nm for a period of 10 min at 30 s intervals and the maximum velocity (Vmax) value was recorded. Reaction mixture without the sample was used as the control. Potassium clavulanate was used as the standard. β-Lactamase enzyme inhibitory activity as percentage inhibition was calculated using the following equation.

Data analysis
Data of all experiments were statistically analyzed by one-way analysis of variance (ANOVA) and Tukey's test using the IBM SPSS Statistics (Version 20) software. Statistical significance was set at 95% confidence level.

Antibacterial activity
Antibacterial activities of the extracts were evaluated using the resazurin reduction assay. Resazurin is an oxidation-reduction based cell viability indicator widely used to evaluate cell growth in various assays (Sarker et al., 2007;Osaka and Hefty, 2013). It is a blue color non-fluorescent and non-toxic dye, which becomes pink and fluorescent when reduced to resorufin by oxidoreductases produced in viable cells. Resorufin is further reduced to hydroresorufin which is colorless and non-fluorescent (O'Brien et al., 2000;Sarker et al., 2007). The oxidized and reduced forms of resazurin can be measured separately and used to evaluate the reduction capabilities of the cells, which reflect the cell viability (Osaka and Hefty, 2013). There is a direct correlation between the reduction of resazurin and the proliferation of living organisms, ranging from bacteria to mammalian cells (O'Brien et al., 2000).
Antibacterial activities of plant extracts vary with the solvents use for the extraction (Sunder et al., 2011). Silva et al. (2015) and Khan et al. (2015) reported that methanol is the best solvent to investigate antibacterial activities of plant extracts. Sunder et al. (2011) have stated that both ethanol and methanol are commonly used to extract antibacterial compounds. In the present study, ethanolic and methanolic extracts of Ravi, Rawana and Oshadha finger millet varieties were used. Antibacterial activities of the extracts against Gramnegative antibiotic-sensitive bacteria at 150, 600 and 2400 µg/ml concentrations are shown in Figure 1. All extracts showed dose-dependent inhibitory activities against the tested antibioticsensitive bacteria. However, none of the extracts showed more than 40% inhibition against S. flexneri and S. typhi at the tested concentrations. Both extracts of the three varieties showed more than 50% inhibitory activities against E. coli and more than 60% inhibitory activities against P. aeruginosa at 2400 µg/ml concentration. However, all extracts showed significantly (P < 0.05) low inhibitory activities, against the tested Gram-negative antibiotic-sensitive bacteria when compared to the standard drugs.
Both ethanolic and methanolic extracts of the three finger millet varieties showed dose-dependent inhibitory activities against the tested Gram-positive antibiotic-sensitive bacterial strains (Figure 2). Both extracts of the three varieties showed more than 50% inhibitions against S. aureus and B. subtilis at 600 µg/ml concentration. Furthermore, they showed more than 70% inhibitions against S. aureus and B. subtilis at 2400 µg/ml concentration. Nevertheless, both extracts of the three varieties cannot be considered as significantly active against the tested Gram-positive antibiotic-sensitive bacterial strains when compared to the standard drugs.
According to the findings of Banerjee et al. (2012), different extracts of an Indian variety of finger millet showed considerable proliferation inhibitory activities against E. coli, S. aureus, P. aeruginosa, Bacillus cereus, Listeria monocytogenes, Streptococcus pyogenes, Proteus mirabilis, Serratia marcescens, K. pneumonia and Yersinia enterocolitica. Singh et al. (2015) have reported antibacterial activities of ethyl acetate extract of an Indian variety of finger millet against E. coli, S. aureus, P. aeruginosa, K. pneumoniae, Shigella dysenteriae, Enterococcus species and Salmonella species. In addition, Shukla et al. (2015) have reported antibacterial activities of hexane extracts of three Indian finger millet varieties against E. coli, S. aureus, P. aeruginosa, Enterococcus species and Salmonella species. However, those results cannot be compared with the results of the present study due to the differences in the extracts and the methods used to determine antibacterial activities. Phenolic compounds including flavonoids have been shown to act as antibacterial agents by interacting with the proteins on bacterial cell wall and thereby disrupting the functional and structural integrity (Silva et al., 2015). Therefore, the phenolic compounds, which are present in finger millet extracts, may be responsible for the antibacterial activities observed in the present study as well.
Generally, Gram-negative bacterial strains are more resistant to inhibition than Gram-positive bacterial strains (Nair and Chanda, 2007). Gram-negative bacteria have a complex cell wall structure when compared to Gram-positive bacteria. Due to the thick murein layer and periplasmic space, the outer membrane of the cell wall of Gram-negative bacteria restricts the diffusion of antibiotics and plant extracts. Multidrug efflux pumps also contribute to create a high intrinsic resistance by pumping out the antibacterial agents through the active efflux processes (Khan et al., 2015). Banerjee et al. (2012) have reported that finger millet extracts are more active against Gram-positive bacteria when compared to Gramnegative bacteria. In agreement with aforementioned studies, ethanolic and methanolic extracts of Sri Lankan finger millet varieties also exhibited high inhibitions against Gram-positive antibiotic-sensitive bacterial strains when compared to Gram-negative antibioticsensitive bacterial strains.

(d) Pseudomonas aeruginosa (ATCC ® 10145™)
150 μg/ml 600 μg/ml 2400 μg/ml  150 μg/ml 600 μg/ml 2400 μg/ml MIC value is the lowest concentration of an antimicrobial agent that completely inhibits the growth of a microorganism as detected by the unaided eye (Balouiri et al., 2016). When both extracts of the three finger millet varieties were tested against the antibiotic-sensitive strains of S. aureus and B. subtilis at 2400 µg/ml concentration, no color changes from blue to pink or colorless were observed. Therefore, MIC values of the extracts against S. aureus and B. subtilis were determined using seven different concentrations (600, 900, 1200, 1500, 1800, 2100 and 2400 µg/ml) and the MIC values of the extracts against S. aureus and B. subtilis were found to be 2100 and 1800 µg/ml, respectively (Table 1)

(a) Staphylococcus aureus (ATCC ® 6538™)
be higher than 2400 µg/ml. Crude plant extracts are generally a mixture of active and non-active compounds. Therefore, higher MIC values were expected when compared to the standard drugs. Antibacterial activities of the ethanolic and methanolic extracts of the three finger millet varieties against antibiotic-resistant bacterial strains are shown in Figure 3. Both extracts showed dose-dependent inhibitory activities against the tested antibiotic-resistant bacterial strains. However, none of the extracts of the three varieties showed more than 45% inhibition against the antibiotic-resistant bacterial strains even at 2400 µg/ml concentration. The inherent resistance mechanisms of these antibiotic-resistant bacterial strains which function through a restrictive outer membrane barrier and transenvelope multidrug resistance pumps (Girish and Satish, 2008) (Webster et al., 2008). In the present study, ethanolic and methanolic extracts of Ravi, Rawana and Oshadha finger millet varieties were evaluated for antifungal activities. Although none of the extracts of the three finger millet varieties was active against the tested fungal strains at 400, 800 and 1600 µg/ml concentrations, the extracts showed less than 10% and 30% inhibitions against F. lini and A. niger, respectively at 3200 µg/ml concentration (Table 2). However, none of the extracts can be considered as active against the tested fungal strains at the tested concentrations.

β-Lactamase enzyme inhibitory activity
The common antibiotics which are extensively used for the treatment of many infectious diseases such as pencillins, cephalosporins, carbapenems and monobactams are β-lactam antibiotics which contain a β-lactam ring in the molecular structure (Atta-ur-Rahman et al., 2001;Drawz and Bonomo, 2010). β-Lactamases such as pencillinases, cephalosporinases and carbapenemases are plasmid or chromosomally encoded bacterial enzymes, which can inactivate βlactam antibiotics by efficiently hydrolyzing the amide bond of the βlactam ring. These enzymes are produced by bacteria as a part of the resistance mechanism against β-lactam antibiotics (Atta-ur-Rahman et al., 2001). Since lifesaving antibiotic drugs are losing their efficacy due to β-lactamase-mediated bacterial resistance, there is an increasing demand to identify novel β-lactamase inhibitors. In the present study pencillinase was used as a β-lactamase enzyme and nitrocefin, a chromogenic cephalosporin, was used as the substrate. Nitrocefin is sensitive to hydrolysis by all known β-lactamase enzymes produced by both Gram-positive and Gram-negative bacteria and exhibits a distinctive color change from light yellow to red with the hydrolysis of the amide bond in the β-lactam ring by a βlactamase (Bidya and Suman, 2014 Oshadha finger millet varieties in inhibiting the β-lactamase enzyme. However, all extracts showed less than 40% inhibitions even at 2000 µg/ml concentration ( Figure 4). Therefore, they do not qualify as promising sources of βlactamase inhibitors at the tested concentration.

CONCLUSIONS
To the best of our knowledge, this is the first report revealing antibacterial, antifungal and β-lactamase enzyme inhibitory activities of any extract of any finger millet variety cultivated in Sri Lanka. Ethanolic and methanolic extracts of Ravi, Rawana and Oshadha finger millet varieties showed dose-dependent inhibitory activities against the tested antibiotic-sensitive and antibiotic-resistant bacterial strains. Both extracts exhibited high inhibitions against the tested Grampositive antibiotic-sensitive bacterial strains when compared to the tested Gram-negative antibiotic-sensitive bacterial strains. The results of the present study provided evidence for the antibacterial potential of both ethanolic and methanolic extracts of the three finger millet varieties against antibiotic-sensitive S. aureus (ATCC ® 6538™) and B. subtilis (ATCC ® 23857™) strains. However, none of the extracts of the three varieties can be considered either as active against the tested fungal strains or as promising sources of β-lactamase inhibitors at the tested concentrations.