CHALCONES TO CONTROL Alternaria alternata IN MURCOTT TANGOR FRUITS

Alternaria brown spot (ABS), caused by the fungus Alternaria alternata pathotype tangerine, is one of the main phytosanitary problems for mandarin growers. About 15 applications per year of harmful fungicides are required for controlling ABS disease in citrus orchards. As chalcones seem to be less toxic to humans and environment than the commercial fungicides in use, this study initially aimed at synthesizing 137 chalcones through aldolic condensations between benzaldehydes and acetophenones. The resulting chalcones were screened for activity against A. alternata through a fungal growth assay that was carried out in 96-cell polypropylene plates, using the same concentration to all studied substances. The four active chalcones underwent conidia germination and mycelial growth, which confirmed the antifungal activity of the compounds. These chalcones were then poured onto Murcott tangor fruit that had been inoculated with conidia of the fungus. All four chalcones reduced the ABS progress to values significantly smaller (P≤0.05) than that observed for the control. Statistical calculations showed that the best results were afforded by two compounds, bearing a 2,4,5-trimethoxyphenyl group at position 3 of prop-2-enal and a 3-nitroor 3-hydroxyphenyl group at position 1 of the aldehyde. Such compounds reduced the incidence of the disease in Murcott tangor fruit to values that did not differ statistically from those obtained with a commercial fungicide.


INTRODUCTION
Mandarin became more valued by global consumers of fresh fruits, mainly because it is easily peeled.Therefore, the production of mandarin increased within the citrus agribusiness (ZULIAN; DÖRR; ALMEIDA, 2013).Specifically in Brazil, about 959,672 tons of this fruit were produced in 51,841 hectares in 2012 (IBGE 2015).However, mandarin growers face a number of phytosanitary problems, among which we can highlight the Alternaria brown spot (ABS), which is caused by the fungus Alternaria alternata pathotype tangerine.This disease has been a serious threat for the production of mandarin in Brazil since 2006, when it was detected in Citrus reticulata var.Ponkan and hybrid tangor Murcott (Citrus reticulata x Citrus sinensis) (CARVALHO et al., 2008;STUART et al., 2009).
ABS affects shoots, young leaves and fruits, the symptoms being characterized by brownish black necrotic spots surrounded by a yellow halo (STUART et al., 2009).As the infected fruits are undervalued (VICENT et al. 2007), fungicides are usually employed to reduce the severity of this disease under field conditions (SOLEL; OREN; KIMCHI, 1997).In growing areas with high humidity and temperature, about 15 applications of fungicides are required per year to control this disease (TIMMER et al., 2003;BOSCH;BERG;PAVELEY, 2014), thus contaminating humans and the environment with substances of high toxicity (CARVALHO et al., 2007).
Chalcones appear as possible alternatives to circumvent such problem because some compounds of this class of substances have already presented activity against plant pathogenic fungi ( SVETAZ et al., 2004;SIVAKUMAR;KUMAR;DOBLE, 2009).Furthermore, they apparently are more biodegradable and less toxic to humans and the environment than the commercial fungicides in use (COTORAS et al., 2011).Naturally occurring chalcones are key intermediates in flavonoid biosynthesis by plants, which can be easily synthesized through several procedures (MCMURRY 2000;SEBTI et al., 2001;EDDARIR et al., 2003;PETROV;IVANOVA;GEROVA, 2008).Therefore, the present work aimed at preparing 137 chalcones and evaluating their effects on A. alternata pathotype tangerine.To assist in the interpretation of results, principal components analysis (PCA) was used in order to summarize and identify correlations among the studied variables (GERALDINE et al., 2013;CARVALHO et al., 2014).

Production of conidia
The fungus A. alternata pathotype tangerine used in this study was obtained as described by Carvalho et al. (2008).Briefly, fungal plugs (90 mm Ø) were transferred to PDA plates, which were maintained for seven days at 25ºC, under constant illumination provided by Philips daylight fluorescent lamps (20W, TLT, 75RS).Then, conidia were collected in 10 mL of an aqueous 0.01 g/mL Tween 80 ® solution.The resulting mixture was filtered through sterilized cheesecloth to obtain the conidia suspension.

Fungal growth assay (FGA)
Aliquots (250 µL) of solutions obtained by dissolving chalcones up to the concentration of 500 µg/mL in aqueous 0.01 g/mL Tween 80 ® solution, and 50 µL of an aqueous suspension containing 2.6 x 10 5 conidia of A. alternata per millilitre, were mixed to obtain a final suspension.From this suspension 20 µL were withdrawn to be added to PDA (130 µL) impregnated with oxytetracycline hydrochloride 500 mg (Pfizer ® , 0.55 mg/mL) that was contained in cells of 96-cell polypropylene plates.Finally, the plates were wrapped with PVC film and incubated at 25 o C, under a 12 h photoperiod per day, for three days.According to Carvalho et al. (2011), the chalcones were considered to be active when preventing fungal growth under a visual observation.

Conidia germination assay (CGA)
To solutions (1000 µL) of chalcones D7, D8, N5 and L21, at 500 µg/mL in aqueous 0.01 g/mL Tween 80 ® solution, were added 200 µL of an aqueous suspension at a concentration of 2.6 x 10 5 conidia of A. alternata per millilitre.Part (520 µL) of each final suspension was spread on agar-water medium contained in a Petri dish, and incubated at 25 ºC, under illumination, for 12 h.Part (50) of the conidia were randomly evaluated, being considered germinated those conidia with germ tubes lengths greater than or equal to the diameter according to visual observation through a microscope.

Mycelial growth assay (MGA)
To PDA medium impregnated with tetracycline (Buncker) at 0.55 mg/mL, contained in Petri dishes (90 mm Ø), were added 500 µL of chalcones D7, D8, N5 and L21 at 500 µg/mL in aqueous 0.01 g/mL Tween 80 ® solution.Just after, a seven-day-old PDA agar plug, containing mycelium of A. alternata pathotype tangerine was transferred to the center of the Petri dish containing the PDA medium, which was incubated at 25 o C, under 12 h photoperiod, for seven days.The diameters of the colonies were then measured with a caliper rule (Pittsburg, USA).

Murcott tangor fruits assay (MTFA)
Only chalcones D7, D8, N5 and L21 were used in this experiment, which was carried out by adapting the method described by Carvalho et al. (2011).Fresh, ripe and healthy Murcott tangerine fruits were washed with sterile water and dried for 60 min under aseptic conditions in a laminar flux chamber.Four equidistant positions around the insertion point on fruits were marked using a permanent marker.Then, 20 µL of an aqueous suspension resulting from the combination of 200 µL of a chalcone solution (500 µg/mL in Tween 80 ® at 0.01 g/mL) and 400 µL of an aqueous suspension at a concentration of 1.3 x 10 6 conidia of A. alternata per millilitre, were poured on each selection point.All fruits were incubated in a humidity chamber at 25ºC, under 12 h photoperiod per day.At 8 and 12 days post inoculation (DPI), lesion diameter around each inoculation point was measured with a digital caliper rule (Pittsburg, USA).This experiment was carried out with four replications (fruits) per treatment, using the following controls: (1) Tween 80 ® at 0.01 g/mL in water; (2) the fungicide Dacobre ® at 3.5 g/L in Tween 80 ® at 0.01 g/mL; and (3) the fungicide Amistar ® at 0.16 g/L in Tween 80 ® at 0.01 g/mL.
Values were converted into injured area (mm 2 ).For statistical analysis, the mean value of injured area per fruit was considered.

Statistical analysis
Data of all in vitro experiments were converted into percentage and subjected to analysis of variance (ANOVA) and means were compared by the Scott-Knott test (P≤0.05),using the software SISVAR 5.3 (FERREIRA, 2011).To assess the multivariate effect of the variables on values obtained in the assays (CGA, MGA and MTFA), the principal component analysis (PCA) was performed using the Rcmdr and FactoMineR packages from the ´R´ software 3.1.0(R Development Core Team, Vienna, Austria).

Conidia germination (CGA) and mycelial growth (MGA) assays
When submitted to the CGA, all the selected chalcones (D7, D8, N5 and L21) inhibited the germination of A. Alternate conidia (Table 1), though better performance was observed for the commercial fungicides in this assay.Among the four chalcones, D8 and N5 were the most active in this assay, inhibiting 85.5 and 87.5% of the conidia germination, respectively.In MGA, chalcone L21 was similar to the fungicide Amistar ® , inhibiting 36.2% of the mycelial growth of A. alternata (Table 1).

Murcott tangor fruits assay (MTFA)
Once the selected chalcones (D7, D8, N5 and L21) could act against A. alternaria by both inhibiting its conidium germination and mycelial growth, these substances were submitted to the Murcott tangor fruit assay (MTFA), which was performed under conditions more similar to those observed in the field than the other assays carried out in this study.All four chalcones were able to reduce the ABS progress to values significantly smaller (P≤0.05)than that observed for the control (Tween 80 ® ).It worths to notice that an increase in the ABS lesions (mm 2 ) over time occurred for all treatments, evidencing a decrease in the fungitoxic effect over time by chalcones and the commercial fungicides used as controls (Table 1).

Principal component analysis (PCA)
The principal component analysis (PCA) of all the results revealed that 96.4% of the variance could be explained by the first two components (85.3% and 11.1%), as shown in Figure 2.
Eigenvalues of the first and second components were 3.41 and 0.44, respectively.CGA was positively correlated with MGA.Similarly, lesions at 8 DPI were positively correlated with the lesions at 12 DPI in the MTFA.

DISCUSSION
FGA was used to select chalcones from different series (A, B, C, J, L, M, N, P, R and T; Figure 1) with fungitoxic or even fugistatic effect on A. alternata.Among the 137 chalcones studied, only four were active against the fungus (D7, D8, N5 and L21), representing 2.9% of the samples.This percentage seems to be in accordance with other researches, in which a percentage around or lower than 10% of active substances has been verified in the screening for compounds with potential biological activity (LOUZADA et al., 2009;CARVALHO et al., 2011;CHIARADIA et al., 2012;MARTINS et al., 2013;FERREIRA et al., 2014).This occurs due to the variation in the chemical structures of the compounds.
Regarding chalcones from D series, the inactive components are mono-or di-substituted at positions 2 and/or 4 of the A-ring, while the active chalcones D7 and D8 contain substituents in position 3 of the A-ring, suggesting that this substitution pattern is important for achieving the desired activity against A. alternata.Using the 2,4,5-trimethoxyphenyl group as A-ring seems to be worthless, since compounds in the J series were all unable to prevent fungal growth.Similarly, a tri-or tetra-substituted A-ring (B, M and T series) confers no activity against A. alternata.
The activity of chalcones with rings condensed to the B-ring (A, C, L, N, P and R series) can also be compared with each other.Among the quinoxaline derivatives, only compound N5 (without substituent at A-ring) was active.Apparently, chemical groups with electron withdrawing or donor characteristic in any position of the A-ring cause loss of the activity to compounds in the N series.When the B-ring of N5 was replaced by a 2-naphthyl group the resulting compound (C26) showed no activity.Other chalcones with naphthyl groups as A or B-ring (A, C and R series) also did not prevent the development of the fungus.Thus, the naphthyl group seems to be useless to achieve antifungal activity.Among chalcones containing the 3,4methylenedioxy motif connected to the A-ring, only L21 (without substituent at B-ring) was active, indicating that electron withdrawing or donor groups in any position of B-ring cause loss of the activity to chalcones in L series.Surprisingly, when A and B-ring are interchanged in L21, generating P1, the activity is lost, evidencing that substitution of prop-2-enal to generate chalcones plays an important role in the activity of compounds against A. alternata.
Results obtained in CGA seem to be better than those observed by Cotoras et al. (2011) andGaucher et al. (2013), who reported only 30% and 50% inhibition of conidia germination of Botrytis cinerea and Venturia inaequalis, respectively.Apparently, this behaviour is mainly caused by the ketone group in the structures of chalcones (LÓPEZ et al., 2001), which can bind thiol groups in proteins, inhibiting fungal cell wall biosynthesis (BOWDEN;POZZO;DUAH, 1990).MGA seems to be a less sensitive assay than CGA to screen chalcones for activity against A. alternata, since conidia exposed to the samples may suffer shrinkage (CARVALHO et al., 2011), which can cause irreversible ultra structural changes (PÂRVU et al., 2008).Consequently, chalcones D7, D8, N5 and L21 were able to delay A. alternata germination, but in MGA satisfactory results were only obtained for chalcone L21, which prevented mycelial growth in a manner similar to that observed for the commercial fungicide Amistar ® .
The commercial fungicides Dacobre ® and Amistar ® belong to the cupric and strobilurine chemical groups, respectively, which can be used to control ABS in citrus (FEICHETENBERGER et al., 2005).Among these fungicides, Dacobre ® was the most efficient against A. alternata, affording results in MTFA that were similar to those obtained with chalcones D7 and D8.According to Agrios (2005), the copper ion (Cu 2+ ) in Dacobre ® is toxic to fungal cells because it reacts with sulfhydryl (-SH) groups of cystein residues of proteins, causing their denaturation.As D7 and D8 were more active against the fungus in the CGA than in the MGA, apparently the ketone group of chalcones also binds to -SH groups, causing the same effect (LÓPEZ et al., 2001;BOWDEN;POZZO;DUAH, 1990).
According to PCA, chalcones D7 and D8 were positioned close to each other (Figure 2), suggesting that both substances act in the same maner against A. alternata.Furthermore, PCA suggests that D7 and D8, positioned separately in an isolated quadrant and outlying from the lesions arrows, are better than N5 and L21.Indeed, PCA can separate effects that are not prone to separation in a mean test (CARVALHO et al., 2014).Consequently, ineffective treatments (eg.Tween 80 ® ) tend to be positioned near to the lesion arrows -MTFA (Figure 2).

CONCLUSIONS
Among the 137 chalcones synthesized and submitted to assays with the fungus A. alternata, the best results were obtained with chalcones D7 and D8, which presented results similar to that obtained with the commercial fungicide Dacobre ® .
The results suggest that maximization of the antifungal acitivity by chalcones must be done by keeping B-ring as a 2,4,5-trimehoxyphenyl group and adding substituents only to position 3 of the Aring.

Figure 1 .
Figure 1.Chemical structures of the chalcones assayed in the present work.

Figure 2 .
Figure 2. Principal component analysis (PCA) with traits.Results from the conidia germination assay (CGA), mycelial growth assay (MGA) and Murcot tangor fruits assay (MTFA) at 8 and 12 days postinoculation (DPI), were investigated for controlling of Alternaria alternata pathotype tangerine by chalcones, according to the two main components.