The pathogen Xanthomonas axonopodis pv. citri 306 strain (Xac 306) whose genome had been sequenced [11] , was used in all experiments. This bacterium was stored in 30% (v/v) glycerol in liquid nitrogen. The renovation of the cells was made each 12 months on nutrient agar, at 28˚C during 48 h; to maintain of biological viability.

The non-pathogenic antagonist bacteria, Pseudomonas sp. strain LN, was isolated from leaves with citrus canker lesions collected in Astorga, PR, Brazil [12] . The procedure of storing and cell maintenance was the same as for Xac 306, except the culture medium that was tryptic soil agar (TSA) plus copper chloride (100 mg∙L⁻¹). Cain and collaborators [13] suggests that some bacteria isolated from culture media added copper might produce antagonistic substances.

The strain LN inoculum was prepared from cultures grown for 48 h at 28˚C on TSA plus copper chloride. Cells were removed from the surface of the growth medium with a sterile loop and suspended in sterile phosphate buffer (KH₂PO₄ 1 g; K₂HPO₄ 1.5 g; distilled water 1000 mL, pH 7.0). Inoculum concentration was determined spectrophotometricaly (λ = 590 nm) and adjusted to a final density of 10⁸ CFU∙mL⁻¹ using sterile phosphate buffer.

Pathogen inoculum was prepared from cultures of Xac 306 grown on nutrient agar for 48 h at 28˚C. Xac 306 was prepared as described above, and the pathogen final concentration was adjusted to 10⁸ CFU∙mL⁻¹.

Bacteria strain LN was grown in TSB plus copper (Cain et al., 2000) (1500 mL) and incubated in a shaker at 28˚C and 100 rpm, during 15 days. After bacterial growth, the medium was centrifuged in a refrigerated centrifuge (9000 rpm at 4˚C), to obtain the secondary metabolites in the supernatant. The solid residue was discarded.

Aliquots of 500 mL were extracted with ethyl acetate (500 mL) in separation funnel. This process was repeated five times. The organic phase, called acetate fraction (AF 0.031 g) was concentrated in rotatory evaporator and lyophilized. The lyophilized powder was suspended in methanol and filtered; the filtrate was lyophilized to yield the acetate methanol fraction (AMF 0.022 g). The AMF was suspended in ethyl acetate, filtered and the filtrate was lyophilized; originating the acetate-ethyl acetate fraction (AEF 0.020 g).

VLC was performed using a glass column (20 mm diameter × 140 mm height), silica gel 60 (0.063 - 0.200 mm, Merck), vacuum at ~380 mm Hg and the organic solvents were hexane, dichloromethane, ethyl acetate, methanol, methanol plus distilled water 1:1 (v/v), and distilled water; in this order [14] . Nine grams of silica gel were packaged in the glass column; afterwards in the top of the column 1.55 g of AEF fraction was added. Twenty milliliters of each solvent was added four times. The respective organic phases were collected and dried originating the fractions hexane (VLC1 0.0014 g), dichloromethane (VLC2 0.1541 g), ethyl acetate (VLC3 0.5659 g), methanol (VLC4 0.7217 g) water: methanol mix v/v (VLC5 0.0197 g) and distilled water (VLC6 0.0022 g). All organic phases were concentrated in rotatory evaporator and lyophilized.

The experiments were performed utilizing four replications per treatment (fractions). Petri plates containing nutrient agar were inoculated with a suspension of 10⁸ CFU of Xac 306 in pour plate. A 150 µL aliquot of AMF, EAF and all fractions of VLC were tested in a concentration of 0.01 at 0.0001 mg∙mL⁻¹ in wells of 9 mm diameter. The plates were incubated at 28˚C during 48 h and control was used distilled water for each tested fraction.

In this step the experiments was performed in duplicate. The fractions, which showed more antimicrobial activity against Xac 306 (EAF, VLC2, VLC3 and VLC4) in the item 2.3, were selected to the next step. In this test a 24-well tissue culture plate containing 1.9 mL of nutrient broth was inoculated with of Xac 306 (10⁸ CFU∙mL⁻¹) per well. Two-fold dilution of fractions concentration such as 5000; 2500; 1250; 625; 312.5; 156.25; 78.12; 39.06; 19.53; and 9.76 µg∙mL⁻¹ were tested. Distilled water as used as control for each fraction tested. The minimal inhibitory concentration (MIC) was defined as the lowest concentration of FAM, FAE and VLC fractions which had non-bacterial growth after incubation. To determine the minimal inhibitory concentration, 50 µL of each wells with no visible growth were picked up from each well, and inoculated on nutrient agar and incubated at 28˚C during 48 h to check viable cells of Xac 306 cells observing the presence or absence of Xac 306 colony in Petri dishes.

Epithelial cells of human larynx carcinoma ATCCL CCL-23 (HEp-2) were grown in DMEN (GiBco™), supplemented with 10% fetal bovine serum and treated with 100 µg∙mL⁻¹ of streptomycin, 100 units∙mL⁻¹ of penicillin and 2.5 µg∙mL⁻¹ of fungizone. Cells were kept at −80˚C.

The fractions effects on cell viability were analyzed using the colorimetric MTT (3-[4, 5-dimethilthiazol-2- yl]-2, 5-diphenil tetrazolium bromide) method (MTT-based assay kit, Sigma Chem. Co.) according to the manufacturer’s instructions. Cells were grown in microplates with 96-wells for 48 h. After attachment to plates, the supernatant of each well was removed carefully and replace with 100 µL of fresh DMEN containing different concentrations of EAF, VLC2, and VLC3. All the fractions were tested in the same concentration (50; 10; 5; 1; 0.5 and 0.05 µg∙mL⁻¹). The plates were incubated for 72 h at 37˚C in a 5% CO₂ chamber with humidified atmosphere. The absorbance of each well was measured in a microplate reader at 490 nm and 630 nm. The experiment was performed in four replicate per fraction.

HEp-2 cells were grown as adherent monolayers in 25 cm², sterile disposable flasks utilizing Dulbecco’s Modified Eagles’s Medium (DMEM F12) (Gibco BRL) supplemented with 10% fetal bovine serum (Gibco BRL) for the fortification of growth factors. The cultures were maintained in a BOD incubator at 37˚C. The cells were grown for two complete cell cycles (24 h) before being treated according to the protocol for each test.

The alkylating agent ethylmethane sulfonate (EMS) (Acros Organics) is a well-known chemical mutagen. It was used for the positive control in the determination of genotoxicity and cytotoxicity and to cause mutations at the hgprt locus. EMS was dissolved in Ca²⁺ and Mg²⁺ free phosphate-buffered saline (PBS), pH 7.4, just before use. The final concentration of this agent in the cultures was 155 µg∙mL⁻¹ in the comet assay.

The general procedure for the genotoxicity test (comet assay) followed the method described by Speit and Hartmann [15] wich is based on the original work of [16] with some modifications. Approximately 1.8 × 10⁵ cells were grown in 2.5 mL culture tubes for 24 h. After incubation, the cells were washed with 5 mL PBS and treated in medium without serum at 37˚C for 3 h. At the end of treatment, the cells were washed, trypsinized and re-suspended in the fresh medium. The cells were then centrifuged (900 rpm) for 5 min to remove the supernatant leaving 300 µL for re-suspension of the pellet. Liquified agarose, 120 µL, was then added to a 20 µL aliquot of cell suspension. The sample was applied to a microscope slide previously gelatinized and covered with a coverslip. After 20 min at 4˚C, the coverslip was removed and the slides were immersed for 60 min at 4˚C in lysis solution containing 89 mL of lysis stock solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, ~8 g NaOH to obtain a pH of 10, 890 mL of deionized water, and 1% sodium lauryl sarcosinate), 1 mL Triton X-100 (Merck) and 10 mL DMSO.

The slides were then placed in an electrophoresis chamber in a chilled bath and maintained submersed in alkaline running buffer (pH > 13) (5 mL 200 mM EDTA, 30 mL 10 N NaOH and 965 mL of deionized water at 4˚C) for 20 min, to examine for single strand breaks. The slides were electrophoresed for 20 min at 25 V and 300 mA. The slides were then covered with neutralizing solution (0.4 M Tris-HCL, pH 7.5) for 15 min at 22˚C. Finally, the slides were submersed in absolute ethanol for 10 min at 22˚C. All the procedures were carried out under low light to avoid extra DNA damage by the action of the light. The material was stained with 80 L ethidium bromide (20 µg∙mL⁻¹) and examined with fluorescence microscopy using an excitation filter of 515 - 560 nm and an emission filter of 590 nm.

The extend and distribution of DNA damage indicated by the comet assay was evaluated by examining at least 100 randomly selected and no-overlapping cells on the slides per treatment. These cells were scored visually according the presence or absence of nucleus tail. The results were evaluated by the statitical tests analysis of variance (ANOVA) and Tukey’s test at p < 0.05, with the experimental criterion being the significance of the response in relation on the negative control in genotoxicity assay, and in relation to the positive control.

Plants of C. sinensis cv. Valencia were used in the greenhouse experiments under the following conditions: 28˚C/22˚C and 10 h/14 h day/night respectively and 80% relative humidity. Plants were watered three times at each seven days and fertilized twice a month with Hewitt solution for non-legume [17] .

Orange leaves with 15 days old were treated with EAF, VLC2, and VLC3 fractions which showed high antimicrobial activity against Xac 306 in a antimicrobial activity assay in 24-well tissue culture plate. All the fractions were applied at the concentration 0.01 g∙mL⁻¹ diluted in distillated water and applied with a hand-sprayer in the abaxial and adaxial leaf surface. The application was realized before or after leaves inoculation with Xac 306.

Plants were subjected to a moist chamber for 24 h before and after inoculation and kept at greenhouse, in order to stimulate the stomata opening and improve the efficiency of bacterial infection and the action of the treatments. Control plants were sprayed with distilled water. The plants were watered three times a week. The number of citrus canker lesions per leaf was determined 21 days after inoculation to harvest 18 leafs per plant. The average number of lesions was calculated by the equation: ∑no lesions in 18 leaves/∑area of 18 leaves. The treatments with the fractions were done before (pre-treatment) and after (post-treatment) the pathogen spraying, in five replications. Analysis of variance was performed on root square-transformed data and means were compared by Tukey multiple range test at P < 0.05. Data were transformed by.

In this experiment antimicrobial activity as tested in all fraction obtained from supernatant of LN culture extracted by ethyl acetate (EAF and AMF) and obtained from EAF fractionation in vacuum liquid chromatography extracted with n-hexane (VLC1), dichloromethane (VLC2), ethyl acetate (VLC3), methanol (VLC4), water: methanol mix v:v (VLC5), and distilled water (VLC6). In this experiment only EAF, VLC2, VLC3, and VLC4 showed antimicrobial activity against Xac 306 (Table 1). However different effects were observed. EAF and VLC2 fractions had the highest antimicrobial effect were found halo with 30 mm of diameter on 0.01 mg∙mL⁻¹,

(-) no inhibition zone.

in the same concentration VLC3 presented halo with 20 mm, and VLC4 10 mm. The 0.001 mg∙mL⁻¹ the EAF, VLC2 and VLC3 showed the same antimicrobial effect, and VLC4 did not have any effect (Table 1).

In this experiment was tested the bactericidal effect only a fractions which showed antimicrobial effect against Xac 306. EAF fraction showed bactericidal effect in concentration of 5000 until 312.5 µg∙mL⁻¹. On the other hand in VLC2 and VLC3 the bactericidal effect was observed also in 156.25 µg∙mL⁻¹. However VLC4 showed bactericidal effect only in 5000 and 2500 µg∙mL⁻¹ (Table 2).

The cytotoxicity capacity of EAF, VLC2 and VLC3 fractions were tested. In the ten-fold dilutions curves the three fractions showed similar cytotoxicity effects on the different concentrations evaluated (Figure 1). Similar

(+) growth inhibition and (−) no inhibition.

results were obtained for CC₅₀ which was calculated using the regression equation from each one, and the three fractions EAF, VLC2 and VLC3 showed CC50 values approximated 36.98; 30.09 and 32.22 µg∙mL⁻¹, respectively.

Figure 2 present the data for genotoxicity, for the concentrations 1, 0.1 and 0.05 µg∙mL⁻¹ of the fractions EAF, VLC2 and VLC3. No DNA migration was found at the EAF, VLC2 and VLC3 fractions on its concentrations tested, compared with positive control. The total cells with damage in the comet assay did not differ from the negative control, except for VLC3 on the high concentration (1 µg∙mL⁻¹). Antimicrobial activity test in orange leaves.

The fractions EAF, VLC2, and VLC3 were tested in the greenhouse experiments to check its capacity to reduce the area of citrus canker lesions formed on leaves. The strategies used to treat the leaves with fractions before or after infestation with Xac 306 did not show any inhibitory effect (data not show). However, the leaves treated with CLV3 fraction had 40% less lesions area when compared with control. The fractions EAF and VLC2 did not differed significantly to the control (Figure 2).