Interference of Extracts Obtained from Baccharis Coridifolia D. C. (Asteraceae-Astereae) Action of Antibiotics

The family Asteraceae is the most numerous group within the Angiosperms. Plants from this family are studied for their chemical composition and biological act ivity. In these studies, the allelopathic, antimicrobial, cytotoxic and anti-inflammatory effects are highlighted. The aim of this study was to evaluate the action of the aqueous (AF1) and the oily (OF2) fract ions obtained from the ethanol extract of Baccharis coridifolia D. C. on the antimicrobial act ivity of antib iotics used in clinical treatment. The drugs chloramphenicol (30 μg), norfloxacin (10 μg), sulfamethoxazole / trimethoprim (75/23 μg), vancomycin (30 μg) and amoxicillin (10 μg) were saturated with the fractions AF1 and OF2, and used pure for further comparison. The strains Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 25923 were used. The results showed that both fractions interfered on the activity of the antibiotics tested. Synergistic, antagonistic or indifferent behavior was observed in the interactions, with various degrees of susceptibility to the antib iotics tested. These results show that the use of products derived from plants can, in some cases, interfere with the effectiveness of antibiotics during clin ical therapy.


Introduction
Synthetic products started to appear more frequently in pharmacological treat ment after the Industrial Revolution and the development of organic chemistry. Obtain ing pure compounds became easier, with the development of structural change processes to produce safer and more active medicaments, and also because of the growing economic power of the large pharmaceutical co mpanies. Even so, natural products have not lost their place in therapy, being considered by the population as safe med icines, as 25% of d rugs prescribed world wide are of natural orig in [1].
The Asteraceae family is the most numerous systematic group within the Angiosperms, as it co mprises about 1,100 genera and 25,000 species. These plants are extremely varied in appearance, main ly including small herbs or shrubs, and rarely trees [2]. About 98% of the genera are small p lants, which are found in all types of habitats, but ma inly in the mountainous tropical regions of south America [3].
Plants in this family are extensively studied for their chemical co mposition and biological activity, with so me supporting the development of new drugs, insecticides and other products [2,4,5]. The genus Baccharis (tribe Astereae) is represented by over 500 species largely distributed in the higher regions of Brazil, Argentina, Co lo mbia, Ch ile and Mexico. In Brazil, 120 species of Baccharis have been described, with most of them located in the Southeast [6,7].
Baccharis species are generally shrubs and are in average 0.5 to 4 m tall. They have a high socio-economic value, with a wide dispersion in the Brazilian states of Santa Catarina, Paraná, São Pau lo, Rio Grande do Sul, and other regions of the country. These plants are main ly consumed as teas, with indications for stomach, liver and prostate problems, as well as anemia, inflammations, diabetes, and body detoxification. In Brazil and Argentina, for example, B. crispa and B. coridifolia are used to heal wounds and inflammat ions. B. trimera, B. uncinella and B. articulata are also well-known in alternative med icine. B. genistelloides is a medicinal herb widely used in Brazil for a variety of diseases, including digestive and liver disorders, malaria, ulcers, d iabetes, anemia, diarrhea, urinary t ract inflammat io ns, tonsillitis, verminosis and Hansen's disease [8].
The flavonoids quercetin, apigenin, hispidulin and eupatorin have already been isolated from B. coridifolia. In this species, the flavonoid rutin is h igher in plants harvested during summer, when compared to plants harvested in winter and spring [11].
Some medicaments are made fro m flavonoids, in particular for the treat ment of circu latory diseases and hypertension, acting as a vitamin C cofactor. Other studies suggest that some flavonoids show considerable antitumor action, and may also have antiviral, anti-hemorrhagic, hormonal, anti-inflammatory, antimicrobial and antio xidant effects [12].
The term "drug interaction" refers to the interference of one drug on the action of another, or of a food or nutrient on the action of a drug. It is important to remember that some drug interactions are beneficial or desirable in the treat ment of concomitant diseases, reducing adverse effects and prolonging desired effects, preventing or delaying bacterial resistance, increasing adherence to treatment, increasing efficiency or enabling dose reduction. The undesirable interactions are lin ked to a reduced drug action or increased contrary results, incidence and range of adverse effects, as well as therapy costs, without increasing the therapeutic benefit. Interactions that lead to a reduced drug activity and, therefore, a loss in effectiveness, are difficult to detect and may be responsible for therapy failure or disease progression. Age, general health, kidney and liver function, alcohol consumption, smoking, diet, as well as genetic and environmental factors influence the susceptibility to drug interactions [16].
B. coridifolia (Figura 1) products are widely marketed in Brazil, and are being consumed in an irrat ional and indiscriminate manner together with other drugs, [17]. In this context, th is study aims to evaluate the effects of B. coridifolia ethanol extract on the antimicrobial activity of antibiotics used in clinical therapy.

Collection Site
The plant material was collected in the countryside in the town of Dois Vizinhos, Paraná, Brazil (25°46'23''S, 53°03'15''; elevation 620m). The specimens were identified and stored in the collection at Un ião de Ensino do Sudoeste do Paraná -UNISEP, Campus of Dois Vizinhos, PR. When the collections arrived at the laboratory, material was selected by separating the leaves, trunk and roots, which were dried in a stove for seven days at 60ºC ±4. The material was then ground and stored in paper bags in a dry environment. A ll of the collections were authorized by IBAMA under the permit number13.234-2 (1 August, 2006).

Extracti on by Maceration of the Plant Material
The 30% (w/v) hydroalcoholic ext ract was prepared by the maceration and trituration method, using a 70% ethanol solution as the extractor. The extract was filtered in absorbent cotton pads. After ext raction, the material obtained was stored in its concentrated form. Two fract ions were obtained: a water-soluble fract ion (aqueous fraction or AF1) and a fract ion soluble in organic solvent (oily fraction or OF2). Both fractions were stored in amber bottles for further analyses.

Anti microbi al Acti vi ty of the Fracti ons
The method used was based on the disk diffusion susceptibility test. In this method, a standardized solution of a specific microorganism is inoculated on agar plates. Then, paper disks impregnated with the substances investigated for antimicrobial activ ity are placed on the agar. Such substances diffuse in the culture med iu m and, if the samp le in question exhib it inhibitory activity against the microorganis m, a clear ring is formed around the disk. After the incubation period, and under the specific conditions for the microorganism develop ment, inhibit ion zones are measured around each disk [18][19][20][21].

Determination of the Mini mum Inhi bitory
The bacterial suspensions of E. coli ATCC 25922 and S. aureus ATCC 25923 were inoculated onto plates containing Mueller-Hinton agar, with a Drigalski spatula. After this procedure, Whatman no.1 filter paper disks (6 mm d iameter) previously prepared were transferred to the plates. The antibiotic disks, with and without the added extracts of B. coridifolia, were evaluated on each plate. Subsequently, the plates were incubated at 35 ± 1 °C for 24 h. After this period, the plates were analyzed fo r the presence of inhibition zones (measured in mm). A ll assays were performed in triplicate. Inhibition zone diameters were interpreted according to the Clin ical and Laboratory Standard Institute, recommended guidelines[22].

Preparation of Paper Disks
The fractions obtained were used for preparing disks of antimicrobial activ ity. The AF1 was dissolved in water, whereas OF2 was dissolved in DMSO (Dimethyl sulfo xide) to achieve concentrations of 100, 50, 25, 12.5, 6.25 and 3.12 mg/liter. Two sets of disks were prepared with the obtained concentrations. The first set of disks were prepared with Whatman filter paper no.1 (6 mm diameter), saturated in AF1 and OF2, according to the prev iously mentioned concentrations. This procedure aimed to evaluate the antimicrobial act ivity of the aqueous and oily fractions and determine the Minimu m Inhib itory Concentration (MIC).
The second set of disks was obtained commercially (Newprov  ). These disks already contain a determined concentration of each antibiotic.

Preparation of the Cu lture Mediu m
The culture mediu m used for testing the antimicrob ial activity was the Mueller-Hinton agar (M H -Merck), prepared from the dehydrated medium, according to the manufacturer's recommendation [18].

Preparation of The Inocula
E. coli ATCC 25922 (beta-lactamase negative) and S. aureus ATCC 25923 were used. The strains were reconstituted according to supplier reco mmendations. The microbial culture was standardized at 10 8 cells/ mL, wh ich is compared to the 0.5 McFarland turbidity standard [23].

Benchmarks for Assessment
CLSI guidelines [21] were used to verify the antimicrob ial susceptibility of the ATCCs. Synergistic effect was considered when a zone of inhibition with a diameter ≥ 2 mm formed by the co mbined applicat ion of the extract p lus the antibiotic was observed, compared to the zone of inhibition formed by the action of the antibiotic alone. Antagonistic effect was considered when inhib ition zones of smaller diameter were observed around the disks with antibiotic and extract, co mpared to that developed by the antibiotic alone. Indifferent effect was considered when there was a zone of inhibit ion surrounding the disks with antibiotic and ext ract, with a d iameter equal to that resulting fro m the application of the antibiotic alone [24]. All act ivit ies were performed in triplicate.

Results and Discussion
The results shown in Table 1 were obtained by evaluation of the antimicrobial susceptibility and MIC determination in a Mueller-Hinton agar mediu m (M H-Merck  ) in different AF1 and OF2 concentrations and measurement of inhib ition zone formed.
In analy zing the data contained in Table 1, it appears that the MIC of AF1 for E. coli was 12.50 mg/ L, whereas the MIC of OF2 was 6.25 mg/ L. The M IC o f both fractions for S. aureus was 12.50 mg/ L.
The MIC values obtained served as a base to evaluate the effect of these extracts on a group of antibiotics used in clin ical therapy. Thus, a MIC equal to 12.50 mg/ L was used as a standard in the interference tests of the metabolites contained in the fract ions AF1 and OF2, when associated to mo lecules of antibiotics used in clinical therapy. Antimicrobial susceptibility tests with the antibiotics chloramphenicol (30 µg), norflo xacin (10 µg), sulfametho xazole/trimethoprim (75/23 µg), vancomycin (30 µg) and amo xicillin (10 µg) (Newprov  ), pure and with added AF1 and OF2, were carried out. Synergistic, antagonistic and indifferent behaviors were observed when comparing the inhibit ion zones formed around antibiotic disks, pure and with added AF1 and OF2. These data are summarized in Tab les 2 and 3.
The inhibition zones formed by the antibiotic disks when associated to OF2 indicate synergism on the activity of all antibiotics tested for E. coli -ATCC-25922, leading to an increase in bacterial susceptibility.
Synergistic behavior was observed between AF1 and the antibiotics amo xicillin, sulfametho xazole / trimethoprim and vancomycin, through an increase in the average size of inhibit ion zones. Norflo xacin showed an antagonistic effect, because the inhibition zones were inferior when associated to AF, in co mparison to its use alone. On the other hand, chloramphenicol show no change in its activity during the tests.  The data in Table 3 show that the behavior of the studied antibiotics on S. aureus ATCC-25923 was modified when associated to OF2 and AF1. Inhib ition of act ivity did not occurred in any case, and an indifferent behavior was observed with norflo xacin for both fractions (AF1 and OF2). Amo xicillin, sulfametho xazole/trimethoprim and chloramphenicol, when associated with both fract ions, showed synergistic behavior. Vancomycin had a synergistic behavior when combined with OF2, but showed indifferent behavior when associated to AF1. Therefore, it can be suggested that an interaction occurs between the antibiotics tested and the metabolites present in the AF1 and OF2 of B. coridifolia.
The present results corroborate Canton and Onofre [25], who used the same methodology to assessed the interference of ethanol ext racts obtained fro m B. dracunculifolia on 15 antibiotics, reporting activity changes that caused synergistic or antagonistic behaviors, and no alteration in some cases. In a study on the extracts of Alternanthera brasiliana (L.) O. Kunt. (A maranthaceae) and their effect on the activity of antibiotics used in clinical therapy, [26] found that when such extracts are associated to 18 inhib itors of protein synthesis, altered behavior and, consequently, synergism or antagonism, are observed. Indifferent behavior, though, was also observed.
It should be noted that the sale of med icinal plant products currently happens in pharmacies and health food stores, where they are marketed according to an industrialized labeling system. In general, these plant preparations have no certificate of quality and are produced fro m cult ivated plants, which mischaracterize trad itional med icine that uses almost always plants of the native flora.
The use of medicinal plants fro m the tradit ional Chinese and Hindu medicines, completely unknown to westerner civilizat ions, is increasingly co mmon. These plants are marketed through advertisements that promise safe benefits, because they come fro m a natural source. However, often the supposed pharmacological p roperties advertised have no scientific validity, because they have not been investigated, or have not had their pharmacological actions proven through scientific tests in preclin ical o r clin ical t rials [24].
Media calls fo r the consumption of products made fro m natural sources increase every day in both developing and developed countries. Herbalists promise health and long life, based on the argument that plants used for millennia are safe for the population.
In Brazil, medicinal plants fro m the native flora are consumed with little or no evidence of their pharmacological properties, and their consumption is propagated by users or dealers. Frequently, these plants are even used for med icinal purposes different fro m those originally used by the native people. Co mpared to the drugs prescribed in conventional treatments, the toxicity of medicinal p lants and herbal med icines may seem trivial, but this is not true. The to xicity of medicinal plants is a serious public health problem. Adverse effects of herbal medicines, possible adulteration and toxicity, as well as synergistic effect, occur co mmon ly.
Research to evaluate the safe use of medicinal p lants and herbal medicines is still incipient in Brazil, as well as the control of these products' sales by official bodies in public markets or health food stores.
In this sense, this work has contributed to the elucidation of the effects caused by the use of B. coridifolia, in relation to the metabolites present in two fractions, showing that such metabolites can interfere positively, causing synergism, or negatively, causing antagonism in the activity of antibiotics commonly emp loyed in clin ical therapy.
Multidisciplinary studies involving ethnobotanists, chemists, pharmacologists and agronomists are needed in order to expand the knowledge on medicinal plants. This would bring a better understanding on how their co mpounds act, what are the to xic and adverse effects, how they interact with new allopathic medicines and what are the most appropriate strategies for production and quality control of herbal medicines, observing the regulatory agencies standards and resolutions.
Raising awareness in the p rofessional team involved in the prescribing, dispensing and administering med ications is important. Professionals fro m d ifferent healthcare areas must always question and warn their patients about the non-prescribed use of med icinal herbs.

Conclusions
This study shows that the oily (OF2) and aqueous (AF1) fractions fro m extracts of B. coridifolia can interfere on the activity of the antibiotics tested. Therefore, it is confirmed that the use of products of plant orig in interfere with the effectiveness of some antibiotics during clin ical therapy.