Xanthium strumarium L . antimicrobial activity and carboxyatractyloside analysis through electrospray ionization mass spectrometry

The aim of this work was to evaluate the antimicrobial activity of Xanthium strumarium L. leaf extracts against Staphylococcus aureus, Escherichia coli, Salmonella typhimurium, Pseudomonas aeruginosa and Clostridium perfringens, as well as to investigate the presence of the toxic compound carboxyatractyloside in different plant parts. S. aureus and C. perfringens were more sensitive to non-polar than to polar fractions, and there was no difference between extracts for the remaining bacteria. All extracts had strong antimicrobial activity against the evaluated microorganisms. Carboxyatractyloside was found in cotyledons and seeds but not in adult leaves and burrs. Thus, only Xanthium strumarium leaves in adult stage can be used for medicinal purposes.


INTRODUCTION
Cocklebur (Xanthium strumarium L.) is an herbaceous annual plant of worldwide distribution.The plant is erect, up to 2.5 m tall, and presents blotched purple stems.Its leaves are dark green on the upper surface, similar in shape to grape leaves, 15 cm in diameter and roughly textured with minute bristles.The flowers are inconspicuous, both male and female, occurring in leaf axils towards the end of the branches.
In addition, the flowers develop into hard woody burrs (fruits), 1.2 cm to 2 cm long, with numerous hooked spines.X. strumarium is extremely competitive with other crops.It has long been considered one of the worst weeds in soybean plantations (Bozsa & Oliver, 1993).
Previous studies have reported that X. strumarium induces intoxication and can be lethal to cattle (Colodel et al., 2000), sheep (Loretti et al., 1999), pigs (Stuart et al., 1981) and humans (Turgut et al., 2005).Also, the consumption of fruits (burrs) and cotyledonary-stage leaves (two-leaf stage) leads to hepatic necrosis and myocardial injury in humans.The toxic principle in X. strumarium poison was isolated and identified as carboxyatractyloside (CAT) (Cole et al., 1980), a highly selective inhibitor of oxidative phosphorylation (Scott et al., 1993).
However, there are no reports about the antimicrobial activities of X. strumarium extracts against human pathogenic microorganisms such as Escherichia coli and about the presence of the toxic compound in different plant parts.Therefore, the aim of this work was to investigate the antimicrobial activities of several extracts from X. strumarium adult leaves against Staphylococcus aureus, Escherichia coli, Salmonella typhimurium, Pseudomonas aeruginosa and Clostridium perfringens, as well as to verify the presence of CAT in different plant parts.

Plant material
Xanthium strumarium (common name: cocklebur, Family: Asteraceae) used in this work was cultivated in the experimental field of the School of Agricultural Engineering (FEAGRI), State University of Campinas (UNICAMP, Campinas, São Paulo State, Brazil).A voucher specimen was deposited at the State University of Campinas Herbarium under the number 134865 and identified by Dr. Washington M. F. Neto (curator).The leaves were separated, dried in a tray drier with air circulation at 45ºC (Marconi, model 035, Piracicaba, São Paulo State, Brazil), packed in dark plastic bags and stored in a domestic freezer at -20 ºC until extracted.Before extraction, the material was triturated in a domestic food processor (Wallita, model Master, São Paulo, SP) and the particles from 24 -48 mesh selected using a magnetic agitator (Bertel, Model 1868, Caieiras, São Paulo State, Brazil).

Extracts
Extracts were obtained using 3 different extraction methods and 4 different solvents, in triplicate.The employed methods were maceration (1), dynamic maceration (2) and soxhlet (3).Solvents were 80% ethanol (A), 80% methanol (B), ethyl acetate (C) and dichloromethane/chloroform (1:1) (D).Maceration was done with 20 g plus 100 mL of the different solvents, resulting in the extracts 1A, 1B, 1C and 1D.After 7 days under periodic agitation, the extracts were filtered through paper filter and the residue again extracted with 100 mL of the respective solvents for 10 min under agitation.Both fractions were then blended and evaporated to dryness at 38ºC under vacuum.Dynamic maceration was carried out with 20 g plus 100 mL of the different solvents, yielding the extracts 2A, 2B, 2C and 2D.After 3 h agitation, the extracts were filtered through paper filter and the residue again extracted with 100 mL of the respective solvents for 1 h under agitation.Both fractions were then blended and evaporated to dryness at 38ºC under vacuum.Soxhlet method was carried out in a soxhlet apparatus with 15 g for 5 h extraction with the solvents A, B and C, resulting in the extracts 3A, 3B and 3C.The extracts were dried to dryness at 38ºC under vacuum.All extracts were stored in a domestic freezer at -20ºC until analysis.

Antimicrobial activity
Minimal inhibitory concentration (MIC) tests were carried out according to the NCCLS (National Committee of Laboratory Standards, 2003) (31) using a tissue culture test plate (96 wells).Staphylococcus aureus (ATCC 6538), Escherichia coli (isolated from swine), Salmonella typhimurium (ATCC 14028), Pseudomonas aeruginosa (ATCC 13388) and Clostridium perfringens (ATCC 1324) were used in the tests.The microorganisms were grown overnight at 36 o C in nutrient agar plates under aerobiosis, except for Clostridium perfringens, cultivated in thioglycollate broth under anaerobiosis.The inoculum for the assays was prepared by diluting scraped cell mass in 0.85% NaCl solution, adjusted to the McFarland scale 0.5.Cell suspensions were finally diluted to 10 4 CFU mL -1 for the assays.The extracts and the antibiotics were diluted in DMSO and stored at -20ºC until used.The final concentrations of the extracts were 1.8, 1.5, 1.2, 1.0, 0.8, 0.6, 0.4 and 0.2 mg mL -1 , whereas serial dilutions of the antibiotics were prepared between 0.25 and 0.00195 mg mL -1 .The inoculum was added to all wells and the plates were incubated at 36ºC for 24 h.To verify the sterility of extracts and culture medium (Mueller-Hinton 2.1%), as well as the viability of microorganisms, a control test was done with all test plates.Antimicrobial activity was detected by adding 0.05 mL of an aqueous 0.5% triphenyl tetrazolium chloride (TTC) solution.MIC was defined as the lowest oil or extract concentration that visually inhibited growth, indicated by TCC staining (dead cells are not stained by TTC).

CAT analysis
Adult and cotyledonary-stage dried leaves, seeds and burrs were extracted with 50% methanol (100 mg to 5 mL) for 20 minutes in ultrasound and filtered in 0.5 mm membrane (FHLP13, Millipore).Plant extract samples were analyzed through direct infusion ESI-MS using a syringe pump (Harvard Apparatus) at a flow rate of 10 L min -1 .ESI-MS fingerprints and ESI-MS/MS in the negative ion mode were obtained using a hybrid high-resolution and highaccuracy (5 ppm) Micromass Q-TOF mass spectrometer.Capillary and cone voltages were set to -3000 V and -50 V, respectively, at a desolvation temperature of 100ºC.One milliliter of each sample was added to 10 mL of a solution containing 70% (v/ v) chromatographic grade methanol and 30% (v/v) deionized water and 5 L ammonium hydroxide per milliliter.Fingerprint mass spectra were obtained in the m/z range between 650 and 920.

RESULT AND DISCUSSION
Table 1 shows the antimicrobial activity of X. strumarium extracts and antibiotics.There were no differences between X. strumarium extracts, except for S. aureus and C. perfringens, which were affected by the used solvent but not by the extraction method.
The extracting solvents ethyl acetate (C) and dichloromethane/chloroform (1:1) (D) were more effective than 80% ethanol (A) and 80% methanol (B).Since S. aureus and C. perfringens are gram-positive bacteria, some cell wall active compounds have probably concentrated in the non-polar fraction; however, further studies are needed to prove this statement.S. typhimurium and C. perfringens were more resistant to X. strumarium methanolic and ethanolic extracts than E. coli and P. aeruginosa.TABLE 1. Minimal inhibitory concentration (MIC, mg mL -1 ) of Xanthium strumarium extracts and antibiotics.
Staphylococcus aureus was the most susceptible to the tested antibiotics, followed by E. coli.MIC values for S. aureus were below 0.0019 mg mL -1 for penicillin G, enrofloxacin, erythromycin, oxytetracycline, lincomycin, amoxicillin and tetracycline.Enrofloxacin was the most effective antibiotic against all tested strains, followed by tetracycline and erythromycin, respectively (Table 1).On the other hand, sulfanilamide and sulfadiazine showed the highest MIC values for S. aureus.Duarte et al. (2005) proposed a classification for plant material based on MIC results: strong inhibitors when MIC is below 0.5 mg mL -1 ; moderate inhibitors when MIC is between 0.6 and 1.5 mg mL -1 ; and weak inhibitors when MIC is above 1.6 mg mL -1 .Thus, considering this classification, all X. strumarium extracts showed potential antimicrobial activity against the evaluated microorganisms (Table 1).
Figure 1A shows the ESI-MS fingerprint of X. strumarium seed.Note the ion of m/z 769 which corresponds to the deprotonated molecule of carboxyatractyloside. Its ESI-MS/MS (Figure 1B) confirms its structure since the ion is found to dissociate nearly exclusively into HSO 4 -of m/z 97.ESI-MS/MS monitoring indicated carboxyatractyloside was present in extracts from seeds and cotyledonary-stage leaves but not in those from adult leaves and burrs.
Carboxyatractyloside is toxic and was originally isolated from the Mediterranean thistle Atractylis gummifera, a highly selective inhibitor of the cytosolic side-specific mitochondrial ADP/ATP carrier (Huber et al., 1999).As already mentioned, previous studies have reported that X. strumarium induces liver damage and can be deathful; thus, care should be taken concerning the medicinal use of X. strumarium and only its leaves in adult stage are recommended, since the latter had strong antimicrobial activity and did not present the toxic compound.