Study of anti-microbial and anti-pectinase activity of extracts and compounds isolated from Siphoneugena densiflora ( Myrtaceae ) and Vitex polygama ( Lamiaceae )

Crude hydroalcoholic extracts (HA) and twelve compounds isolated from Siphoneugena densiflora O. Berg and Vitex polygama Cham. were tested in vitro as growth inhibitors of the fungus Leucoagaricus gongylophorus (Singer) Möller, that lives in symbiosis with the leaf-cutting ant Atta sexdens, as well as inhibitors of the fungus’ pectinases. Also, the effects of the extracts and the compounds were assessed in a panel of bacteria and yeasts similar to those living in symbiosis with the ant. HA extracts from S. densiflora displayed significant antimicrobial results, mainly against Pseudomonas aeruginosa. Caffeoyl 6-O-β-D-glucopyranoside was the most promising compound in the control of L. gongylophorus growth. Revista Fitos, Rio de Janeiro, Vol. 8(1): 1-72, Jan-Mar 2013 43


Introduction
Leaf-cutting ants from Atta genus are regarded as important damaging agriculture pests in South America.Atta sexdens rubropilosa Forel (Hymenoptera: Formicidae), one of the most ordinary species of Atta, lives in obligate symbiosis with the basidiomycete Leucoagaricus gongylophorus (Singer) Möller and in association with several other microorganisms, for example the bacterium Burkolderia sp. and yeasts such as Candida, Cryptococcus and Trichosporon (Carreiro et al., 1997;Santos et al., 2004;Erthal et al., 2009).The basidiomycete provides the ants with nutrients (Martin and Weber, 1969) and plant polysaccharide degrading enzymes (Silva et al., 2006), like the pectinases, while the ants supply the symbiotic fungus with a variety of substrates (Weber, 1972) and also stimulate its growth on plant material inside their nests, in chambers called "fungus garden" (Martin et al., 1975).The rupture of that symbiosis either by the employment of antimicrobial agents or by the use of pectinases activity inhibitors can contribute to controlling the insects in the field.
As part of our search for plant secondary metabolites to be used as insecticides and antibiotics (Gallo et al., 2006a;Cazal et al., 2009;Bicalho et al. 2012), hydroalcoholic extracts and isolated compounds from Siphoneugena densiflora O. Berg (Myrtaceae) and Vitex polygama Cham.(Lamiaceae) were tested against several microbes, on L. gongylophorus growth, and as inhibitors of the fungal pectinases activity.Chemical interactions between the test compounds and the reagents of the enzymatic assay were evaluated as well.The extracts and the isolated compounds (1 to 12) were obtained and identified according to procedures formerly described (Gallo et al. 2006a;Gallo et al. 2006b;Gallo et al., 2008).

Antimicrobial assay
Antimicrobial activity of extracts was determined using the paper disk diffusion method (Bauer et al., 1966) (Salvat et al., 2001).
In order to harvest the fungal pectinases, the worker ants of A. sexdens rubropilosa were placed in a frozen chamber at -15 o C for 10 minutes to be anesthetized.Subsequently, each ant had its abdomen softly squeezed by calipers to release the inside liquid, which was collected with a Pasteur capillary pipette and transferred to a flask containing 50 mM sodium phosphate buffer at pH 6.At the end of the harvest, the obtained solution was diluted with deionized water by 1:500 (v/v) and named faecal fluid whose composition was mainly of fungal pectinases.
The enzymatic inhibitory activity of extracts and isolated compounds was measured as following: 4 µL of solution of the plant extract or the isolated compound (50 µg/µL in DMSO) were added to 150 µL of pectin solution (2.0 g in100 mL of 50 mM sodium citrate/ phosphate buffer at pH 5), 75 µL of faecal fluid and 71 µL of deionized water.For the solvent control, 4 µL of DMSO was used.In addition, an experiment without solvent or extract was conducted, using 150 mL of deionized water and 150 mL of pectin solution.The experiment was carried out in duplicate; the reaction mixture was incubated at 37 ºC and shaken for 30 minutes.Aliquots of 50 mL of each above mentioned mixture were collected during the initial and the final incubation time, and were added to 100 µL of DNSA reagent (3,5-dinitrosalicilic acid;Miller, 1959) and 100 µL of deionized water.The resultant solutions were boiled during 5 min, followed by ice-refrigerating and centrifuging.The optical density (OD) of the collected supernatant was measured at 540 nm.The pectinases inhibition activity (I) was calculated through the formula I = [1 -(TC -1 )] x 100, where T and C are the pectinases activity, obtained as OD values, in the presence and in the absence of a given plant extract or compound, respectively.
The influence of the tested tannins on the assay reagents was assessed by mixing the tannins separately with each one of the biochemical assay reagents, and the procedure for OD measurement was carried out as above mentioned.
Differences between treatment means were performed by Mann Whitney or Student's t tests (Snedecor and Cochran, 1989).Results are given in the text as probability values, with P < 0.05 adopted as the criterion of significance.

L. gongylophorus growth inhibition
The fungus was isolated from an A. sexdens rubropilosa nest and kept in culture media (Pagnocca et al., 1990).Compounds and extracts were tested against L. gongylophorus growth according to established protocols (Bigi et al., 2004).Hydroalcoholic extracts of twigs (SD-HAT), stem (SD-HAS), leaves (SD-HAL) and root bark (SD-HAR) from S. densiflora were tested at 1000 μg/mL.Isolated substances were tested at 50 or 100 μg/mL.The assays were run twice (two sets of five tubes for each concentration).Fungal growth was estimated macroscopically on the basis of mycelia surface and density after 30 days of incubation and the modal value among the five replicates was registered.

Antimicrobial assay
Table 1 summarizes the obtained results when HA extracts from S. densiflora were assessed against some yeast, fungus and bacterium species through the disk diffusion method (DDM).The extract SD-HAL displayed potent activities against Pseudomonas aeruginosa and Micrococcus roseus, causing an inhibition zone diameter (IZD) of 11 mm on both species, in comparison to those exhibited by the standard drug (IZD of 10 and 12 mm, respectively).In addition, the other extracts that showed IZD higher than 8 mm on the tested bacteria and fungus species were considered active, albeit they caused no effect toward the yeasts development.The antibacterial activity of HA extracts was corroborated by the determination of the minimal inhibitory concentration values (MIC, Table 2).The Gram-negative P. aeruginosa, known as one of the most prevalent causes of nosocomial infection in the world, was found to be particularly susceptible to the HA extracts (MIC < 31.3 μg/mL), indicating them as promising sources of antibiotic compounds against that species.The antimicrobial activity of some of the isolated compounds, 1 to 8, 10 and 11, was also evaluated by DDM.They showed no effect against the microorganisms when tested at 50 μg/disk (data not shown), except for gallic acid (5), which presented an IZD of 15 mm on Staphylococcus aureus growth.In general, flavonoids, triterpenes and hydrolysable tannins, the more representative classes of the compounds tested, are known for displaying antimicrobial activities (Serrano et al., 2009;Buzzini et al., 2008;Harborne and Williams, 2000;Li et al., 2002).For example, the flavone luteolin (2) performed strong activity against some filamentous fungi, presenting IC 50 values from 38 to 81 μg/mL (Wang et al., 2010).Flavone 2 is also known as an effective phytoalexin in sorghum seeds (Du et al., 2010), acting as a competitive inhibitor of enzymes in the biosynthesis of melanin, a fungal virulence factor (Brunskole et al., 2009).Conversely, luteolin caused no effect against twenty microbe strains Revista Fitos, Rio de Janeiro, Vol.8(1): 1-72, Jan-Mar 2013 when tested at 2.5 mg/mL (Schinor et al., 2007).The phenolic acid 5 showed an IZD of 7 mm on S. aureus and C. albicans growth when tested at 100 μg/disk (Fogliani et al., 2005) (Taguri, Tanaka and Kouno, 2004).The triterpene arjunolic acid (6) was found to be active against E. coli (NEU 1006; IC 50 of 3 µg/spot) (Djoukeng et al., 2005) as well C. albicans (ATCC 90028; MIC of 50 µg/mL) (Bisoli et al., 2008), and its antifungal activity was highly potentiated when in mixture with asiatic acid (Masoko et al., 2008), a constituent also isolated from S. densiflora leaves (Gallo et al., 2006b).O'May and Tufenkji (2011) showed that ellagitannins can block P. aeruginosa swarming motility by disturbing its quorum-sensing and biosurfactant production that are important requisites for the formation of biofilms and antibiotic resistance.Unfortunately, the similar tannins isolated from S. densiflora (Gallo et al., 2006b) could not be tested due to their small available amount and/or instability when in contact with air, yielding a thin insoluble pale yellow layer on the flask surface.Comparing literature data with the obtained results, it seems that the ellagitannins present in S. densiflora may be responsible for the high activity of HA extracts along with other constituent compounds that present antioxidant properties, like the flavonoids, so acting synergically, which might explain the MIC values of HA extracts against the bacteria in contrast to the ineffectiveness of the tannins that  were tested.This fact suggests the employment of HA extracts from S. densiflora, rather than a single compound, against the A. sexdens symbiotic bacteria in order to control that common agriculture pest and also draw attention to this plant as a promising source of antibiotics.

Pectinases inhibition assay
Since tannins can bind with proteins or react with weak alkalis or acids, yielding non-specific enzymatic inhibition effect (Chalton et al., 2002;Khambabaee and Ree, 2001), the influence of this class of substances on the reagents of the enzymatic assay was previously assessed in order to check possible chemical interactions between them.The mixture containing one of the tannins (9, 10, 11 or 12) along with either pectin or buffer did not become cloudy or precipitated after the test had been performed, which might indicate absence of chemical interactions between the tannins and those reagents and, consequently, no influence in the final results.Conversely, the mixture containing either one of the tannins and the pectinases or DNSA revealed a significant OD increase, proving the occurrence of reaction between the tested tannins and those reagents.
Table 3 summarizes the enzymatic inhibitory activity of HA extracts from S. densiflora and some of the isolated compounds.The SD-HAS and SD-HAR extracts as well as the compounds 9, 11, and 12 exhibited potential outcomes.In addition, it was observed a significant difference between the effects of tannins 10 and 11, indicating the importance of a gallic ester group for a better activity.On the other hand, compounds 3 and 8 caused an increase in the pectinases activity.It is known that pectinases can present glycosidase activity, which induces glycosilated compound hydrolysis yielding the respective aglycone (Versari et al., 1997).This might be the reason why compounds 3 and 8 enhanced pectinases activity rather than causing its inhibition.However, it was noteworthy that tannins 9, 11 and 12 were able to inhibit pectinases activity significantly (Table 3), even when the preliminary tests predicted an opposite effect.Thus, one may speculate that the total tannin amounts were not completely hydrolyzed, and that the remaining molecules were capable of causing the observed result.Consequently, a higher amount of the test compound would be required to cause the complete enzyme inhibition or the hydrolytic products might be presenting a supplementary effect.Though all these assumptions need further experiments to be corroborated, the results suggest tannin 12 has potential to be explored as inhibitor of L. gongylophorus pectinases.

L. gongylophorus growth inhibition
In general, the HA extracts from S. densiflora assayed against L. gongylophorus growth showed none or just moderate activity (Table 4), as observed in SD-HAS effect (20% inhibition).After several chromatographic separation steps, SD-HAS yielded the tannins 9 and 10.Compound 10 displayed no effect when tested at the concentration of 50 μg/mL (data not shown), but exhibited the same activity of the original extract at 100 μg/mL (Table 4).Compound 9 was not tested due to its insufficient amount.Compounds 5, 7 and 11 revealed a lack of activity.However, it is interesting to notice that either syringaldehyde, which is structurally different from syringic acid (7) just by a functional aldehyde group, or vanillic acid, which lacks a hydroxy group but bears an extra methyl group in comparison with gallic acid (5), caused 80% of fungal growth inhibition at the same tested concentration (Souza et al., 2005).The glycosilated flavonoid 4 and triterpene 8 displayed a weak effect on the fungal development, whereas compounds 1, 2 and 6 were twice more active, indicating that an additional sugar molecule may reduce the fungal growth inhibitory activity.Djoukeng and coworkers (2005) observed analogous trend assessing glycosilated triterpenes and their respective aglycones against some bacteria.On the other hand, the glycosylated phenolic acid 3 and tannin 12 caused 60 and 80% of inhibition on L. gongylophorus growth, respectively.Ellagic acid glycosides related to compound 12 also displayed significant activity against Magnaporthe grisea spore development (Zhou et al., 2007), which points out those compounds as promising lead antifungal substances.c Compound isolated from S. densiflora.Tested concentration: 50 μg/ mL, except for compounds 10 and 12 (100 μg/mL).
d Control with or without dilution solvent: 0% inhibition; the dry weight of the fungal suspension presented a mean of 7.2 mg/mL.

Conclusion
The results obtained with the antimicrobial, growth inhibition and enzymatic inhibition assays point out the use of HA extracts of S. densiflora as a promising alternative method in the control of A. sexdens rubropilosa in the field.Also, the results showed HA extracts of S. densiflora as a prospective source of antibiotics.

V
. polygama and S. densiflora were collected in July 2000, in the city of Poços de Caldas, Minas Gerais, Brazil.Voucher specimens were deposited at the Herbarium of the Faculdade de Filosofia, Ciências e Letras of the University of São Paulo (SPFR), campus Ribeirão Preto, under the acquisition number 9968, and at the Herbarium of the Department of Botany of University of São Paulo, São Paulo, Brazil, respectively.

Table 3 -
Inhibitory activity of hydroalcoholic extracts and isolated compounds from Siphoneugena densiflora and Vitex polygama on Leucoagaricus gongylophorus pectinases.P values ≤ 0.05 were considered significantly different from control.
b Compound isolated from S. densiflora.cCompound isolated from V. polygama.d

Table 4 -
Inhibitory activity of hydroalcoholic extracts and isolated compounds from Siphoneugena densiflora and Vitex polygama on Leucoagaricus gongylophorus growth.