Four strains (Klebsiellaoxytoca, Salmonella enterica, Shigella sonnei, and Vibrio cholera) were isolated from patients concerned by diarrhea at University hospital (CHRU) of Strasbourg (France) and obtained from the team of Dr. Gilles Prévost. Bacteria are cultivated in Mueller Hinton (MHB, Merck, Darmstadt, Germany).

Chromolaena odorata L. (Asteraceae) is collected during the rainy season in Abomey-Calavi (Benin). The taxonomic identification was obtained from a botanist and a voucher specimen was deposited at the National Herbarium of Abomey-Calavi in Benin (AA6390HNB).

Leaves were dried in the dark to avoid degradation of the active components and powdered with a blender. An aliquot (50 g) was sequentially extracted under agitation (IKA KS260 Basic, Staufen, Germany) at room temperature (20˚C) with cyclohexane (C₆H₁₂), dichloromethane (CH₂Cl₂), ethyl acetate (C₄H₈O₂) and butanol (C₄H₁₀O). After the first extraction with C₆H₁₂ (2 × 150 mL, 1 h), the extract was filtered under vacuum and the filtrate was dried in a rotary evaporator (Stuart RE 300, Staffordshire, UK). The residue of this first extraction was dried under the laminar airflow chamber and used for successive similar extractions with the three others solvents.

Each extract (20 mg) was submitted to liquid chromatography on a Sephadex G-25 column (0.9 cm × 3.5 cm, GE Healthcare, Sephadex^TM G-25 Superfine, Uppsala, Sweden) and eluted with three successive solvents: solvent 1 (CHCl₃/CH₃OH/H₂O; 60:30:4.5; v:v:v), solvent 2 (CHCl₃/CH₃OH; 2:1; v:v), solvent 3 (CHCl₃/CH₃OH/ H₂O; 40:35:10; v:v:v). Elution was performed with 5 mL of solvent 1, 2.5 mL of solvent 2 and 2.5 mL of solvent 3. Each eluted fraction was dried under a stream of nitrogen using N-EVAP model 111 (Organomation Assoc. INC, Berlin South, USA).

Dried fractions were solubilized with solvent 2 (30 - 50 µl) and loaded on glass plates, Kieselgel 60 F254 HPTLC (10 cm × 10 cm) or TLC (20 cm × 20 cm) (Merck, Darmstadt, Germany). The silica plates were developed in a pre-saturated vertical twin-through glass chamber 25 cm × 25 cm × 10 cm (Camag, Muttenz, Switzerland) containing 35 mL of solvent 4 ((C₂H₅)₂O/ CH₃OH/CH₃COOH; 90:1:2; v:v:v) for monodimensional TLC. A two-dimensional thin-layer chromatography was performed for a better separation of different compounds present. For bidimensional TLC, plates were developed in solvent 5 (CHCl₃/(CH₃)₂CO; 95:5; v:v) for the first dimension, dried and placed in solvent 4 (C₂H₅)₂O/ CH₃OH/CH₃COOH; 90:1:2; v:v:v) for the second dimension. Compounds were detected under UV light (365 nm).

Antibacterial activity of plants extracts was determined by using a microdilution bioassay. Bacteria were cultured aerobically at 37˚C in a Mueller Hinton Broth (MHB, Merck, Darmstadt, Germany) with shaking at 300 rpm for 18 h in an incubator (Tritamax 1000, Heidolph, Germany). The following bacteria strains were tested: Klebsiella oxytoca, Salmonella enterica, Shigella sonnei, and Vibrio cholera. Bacterial activity was tested by measuring the inhibition of bacterial growth [11]. Extracts (20 mg) were suspended in acetone/water (10:90 v:v, 1 mL) and diluted to a concentration of 10, 5, 2.5, 1.25, 0.625, 0.312, 0.156 mg/mL in MHB. The samples (100 µL) were incubated in 96 wells microplates with 100 µL of a midlogarithmic phase culture of bacteria with a starting absorbance of 0.001 at 620 nm. Each assay was performed in triplicate. Cefotaxime (Sigma Aldrich ChemieGmbh, Steinheim, Germany) was used as positive control. Bacterial growth in the absence of antibacterial agents is used for control and acetone/water (10:90; v:v) was also tested. After 18 h of incubation under agitation at 37˚C, the Minimal Inhibitory Concentration (MIC) was determined by addition of 40 µL of Iodonitrotetrazolium chloride (INT, Sigma Aldrich Chemie GmbH, Steinheim, Germany) [12,13] at 0.2 mg/mL and the plates were incubated for 1 h at 37˚C [13]. Bacterial growth was assessed by a reddish-pink color and the MIC was determined as the lowest concentration of plant extract for which the color does not turn red after the addition of INT.

Isolation of the active components was obtained by using the bioautography procedure [12,13]. A culture of Salmonella enterica in MHB with 0.7% agar (Gibco, Paisley, Scotland) with absorbance of 0.01 at 620 nm was sprayed on HPTLC plates. Plates were incubated for 18 h at 37˚C and 100% relative humidity. The detection of antibacterial spot was obtained after spraying of plates with an aqueous solution of INT (1 mg/mL) and incubation for 1 h. Inhibition of bacterial growth was indicated by clear zones on the bidimensional plates. The inhibition zones were scraped on TLC plates similarly prepared. This material was used for a liquid column chromatography (1 cm × 0.2 cm) and the active compounds were eluted with methanol, dried by N-Evap, suspended in a mixture of isopropanol/H₂O/trifluoroacetic acid (TFA) (30:69.9:0.1; v:v:v) and centrifuged at RT 15.000 rpm for 15 min.

In order to purify the active components, the supernatants were analyzed by HPLC using a DIONEX Ultimate 3000 system and a Vydac C18 218TP510 Grace column (particle size 3 - 20 µm and pore diameter 300 Å Davidson, USA). The elution was obtained at RT with a gradient (as indicated on the chromatogram) of Solvent A: Isopropanol/H₂O/TFA (30:69.9:0.1; v:v:v) and Solvent B: Isopropanol/H₂O/TFA (90:9.91:0.09; v:v:v) at a flow rate 1.5 mL/min. The detection was obtained at 254 nm.

Mass measurements were performed using MALDI-TOF (Ultraflex^TM MALDI-TOF/TOF, BrukerDaltonics, Bremen, Germany). This instrument was used with a positive reflectron and a maximum accelerating potential of 25 kV applied to the target and 26 kV applied to the reflectron. The value of the delayed extraction was optimized to 110 ns in order to obtain the best resolution of the peptides used for calibration. The laser used was a nitrogen laser (337 nm) operating at a frequency of 20 Hz. Data acquisition was performed in a mass range from 100 m/z to 500 m/z.

Extracts (0.5 µL) were mixed with matrix (0.5 µL) loaded on the target and dried at atmospheric pressure using the dried droplet method. In order to increase the sensivity of the technique, two matrices were used: a-cyano-4-hydroxy-cinnamic (CHCA) (BrukerDaltonics. Coventry, UK) and 2,5-dihydrobenzoic acid (DHB) (Sigma Aldrich, Steinheim, Germany). A saturated solution of CHCA matrix in H₂O/ACN, 1:1 was diluted three times with H₂O/ACN 50:50 acidified with 0.1% TFA. DHB matrix was prepared at a concentration of 10 mg/mL in H₂O/ ACN, 1:1.

MS analyses were performed using an Agilent Technologies 78901A (Santa Clara, CA, USA) gas chromatograph equipped with a split/splitless programmed temperature injector and an Agilent J&W DB5MS fused silica column (30 m by 0.25 mm; film thickness, 0.25 µm) connected to a JEOLAccut of JMS-T100GCV spectrometer (ChemSW, Inc., Fairfield, CA, USA). The mass spectrometer was operated in electron impact mode at 70 eV, and spectra were recovered over a mass range from m/z 35 to 700 with a cycle time of 0.4 scans/s. The oven temperature was programmed to increase from 80˚C to 300˚C at a rate of 10˚C/min, and then kept isothermal for 8 min. The other conditions were as follows: helium split, 1:10; constant flow rate, 1 mL/min; transfer line temperature, 280˚C and MS source temperature, 250˚C. Samples were injected with an auto-injector at 280˚C using the split mode with a ratio of 1:5.

The two isolated compounds were dissolved in CDCl₃ and analyzed using a BRUKER Avance III 500 MHz spectrometer equipped with CapNMR^TMProtasis 5 µL Probe Head (Savoy, IL, USA).

Chromolaenaodorata selection was based on an ethnopharmacological survey conducted among traditional healers of Benin. 14 plants were selected and tested in a preliminary study and C. odorata was selected as one of the most active plant. C. odorata leaves extracts were prepared and antibacterial activity extracts was determined by using iodonitrotetrazolium salt. Results demonstrate that C. odorata display an antibacterial activity ranging from 0.156 to 1.25 mg/mL against Klebsiellaoxytoca, Salmonella enterica, Shigella sonnei and Vibrio cholerae (Table 1). The best activity was obtained against V. cholerae with a MIC of 0.156 mg/mL for the dichloromethane extract and a MIC of 0.312 mg/mL for the butanol extract.

After purification of the extracts on a Sephadex G-25, TLC was performed (Figure 1). CH₂Cl₂ and C₄H₈O extracts display a high pattern of similarity. In order to improve the separation of the components, we performed a two-dimensional chromatography (Figure 2) combined to a bioautography against Salmonella. We were therefore able to detect two major inhibition zones (Cp 1, Cp2; Figure 2(a)) of bacterial growth for C₆H₁₂, CH₂Cl₂ and C₄H₈O leaves extracts. These three extracts were pooled for the isolation of molecules.

The inhibition zones named Cp 1 and Cp 2 (Figure 2(a)) were recovered and isolated by RP-HPLC showing major peaks at 12 min and 24 min respectively (Figure 2(b)).

The molecular formula of Cp 1 and Cp 2 was determined as C₂₀H₂₀O₇ at m/z 372 and C₁₉H₁₈O₆ at m/z 342

Table 1. Antimicrobial activity (MIC in mg/mL) of Chromolaena odorata leave extracts.

respectively. Structural determination of the compounds by NMR enabled us to identify two flavonoids. The complete assignments of ¹H and ¹³C NMR spectra were achieved with the help of COSY, NOESY and HSQC.

Compound 1: d 7.49 (1H, dd, J = 17, J = 4 H_Z, Hc), 7.31 (1H, d, J = 4 H_Z, He), 6.95 (1H, d, J = 17 H_Z, Hd), 6.78 (1H, s, Ha), 6.60 (1H, s, Hb), 3.97 (6H, s, 7-OCH3, 5-OCH3), 3.96 (3H, s, 3’-OCH3), 3.94 (3H, s, 4’-OCH3), 3.90 (3H, s, 6-OCH3).

Compound 2: d 7.82 (2H, d, J = 18 H_Z, Hc), 7.00 (2H, d, J = 18 H_Z, Hd), 6.79 (1H, s, Ha), 6.68 (1H, s, Hb), 3.97 (6H, s, 7-OCH3, 5-OCH3), 3.90 (3H, s, 6-OCH3), 3.87 (3H, s, 4’-OCH3).

Cp 1 has been identified as 3’,4’,5,6,7-Pentamethoxyflavone or Sinensetin and Cp 2 as 4’,5,6,7-Tetramethoxyflavone or Scutellareintetramethyl ether (Figure 3).

The synthetic molecules corresponding to our natural isolated compounds were purchased from Extrasynthese (Genay, France) and their NMR spectra were compared showing similar profiles between the two series of analysis.