Is AMF inoculation an alternative to maximize the in vitro antibacterial activity of Libidibia ferrea extracts?

Arbuscular mycorrhizal fungi (AMF) are known to provide plant species with several benefits, such as an increased production of bioactive compounds. However, it is yet to be defined whether extracts of mycorrhizal plants are more efficient in vitro antibacterial actions when compared to non-mycorrhizal plants. We tested the hypothesis of whether or not, methanolic extracts of Libidibia ferrea fruits, from plants established in the field and inoculated with AMF, have higher antibacterial action when inoculated with Acaulospora longula, Claroideoglomus etunicatum or Gigaspora albida. In addition, native L. ferrea fruits collected from the Caatinga area were also tested. The extracts of L. ferrea fruits inoculated with A. longula had higher in vitro antibacterial action in relation to the extracts of fruits from noninoculated plants (p <0.05) thus characterizing the first record of different antibacterial actions of plant extracts due to inoculation with AMF. The extracts of L. ferrea fruits inoculated with A. longula were more efficient in inhibiting growth of Gram-negative bacteria. The zone diameters of inhibition ranged from 2.48 % to 7.56 % larger than the zones of the non-inoculated L. ferrea fruit extracts. The inoculation of L. ferrea with AMF may represent an alternative way of producing fruits with different antibacterial activity.


Introduction
As documented in many studies, arbuscular mycorrhizal fungi (AMF) are microorganisms that provide plant species with several benefits (Wu et al., 2016;Amiri, Nikbakht, Rahimmalek, & Hosseini, 2017;Tavarini et al., 2018). However, only a few papers have recorded whether 'products' with higher mycorrhizal efficiency in plants have a different action, such as mycorrhizal plant extracts.
Some studies reported that the use of AMF inoculation in plants of medicinal interest enhances the production of bioactive compounds (Oliveira, Alves, Silva, & Silva, 2015;Amiri et al., 2017). This may provide the species with medicinal properties, such as antioxidant (Giovannetti et al., 2012), antitumor (Nakamura et al., 2002), and antimicrobial activities (Silva et al., 2013b).
Even though the plant organs accumulate more compounds whenever the plants are mycorrhizal, only one study has reported whether mycorrhization with Funneliformis mossae (T.H. Nilcolson & Gerd.) C. Walker & Schuessler, in addition to increasing the concentration of essential oils extracted from mycorrhizal Anethum graveolens L., var. Hanák, maximizes the fungicidal effect of these essential oils against Colletotrichum nymphaeae (Pass.) Aa strain CCh32; the fungus which causes anthracnose in strawberries (Karimi et al., 2016).
However, no records have been found in literature about the efficiency of medicinal plant extracts demonstrating a higher concentration of phenolic compounds resulting from mycorrhization. Therefore, the following question arises: Can mycorrhization, in addition to increasing the production of phenolic compounds in medicine plants, also maximize their medicinal actions?
It is crucial to develop investigations that seek to prove the efficiency of mycorrhizal medicinal plant extracts as alternatives to phytochemical studies associated with AMF inoculation. This is because the focus of the field rather the mycorrhizal efficiency at producing phytochemicals (Lima et al., 2015;Almeida, Sawaya, & Andrade, 2018) than the efficiency of mycorrhizal plant extracts. The confirmation of this hypothesis could benefit the phytotherapeutic drugs industry in regards to the preparation of the raw-material used in the production chain of more efficient phytotherapeutic drugs as well as to collaborate with health practices who are seeking alternatives for the use of phytotherapeutic drugs associated with traditional medicine (WHO, 2013).
In this context, considering the increasing worldwide number of reports about the emergence of multidrug resistant bacteria (Rice, 2018), one alternative to mitigate this problem could be the development of medicinal plant extracts with efficient antimicrobial activity. Organs of L. ferrea have shown antimicrobial activity (Silva et al., 2013a;Araújo et al., 2014;Biasi-Garbin et al., 2016) and the extracts of mycorrhizal plants have increased the production of compounds such as flavonoids, tannins and gallic acid (Silva et al., 2014a(Silva et al., , 2014bSantos et al., 2017). However, it is yet to be confirmed if these extracts could present a higher antibacterial activity than those of non-mycorrhizal plants. Therefore, our aim was to verify the in vitro antibacterial activity of methanolic extracts of L. ferrea fruits, from plants inoculated or not with AMF; considering the strains of Grampositive and Gram-negative bacteria resistant or not to antibiotics. We tested the hypothesis that the extracts of L. ferrea fruits from plants inoculated with AMF have higher antibacterial action.

Material and Methods
We used L. ferrea fruits, mycorrhizal or not, maintained since February 2013, in the Experimental Field of the Laboratory of Mycorrhizal Technology (LTM/UPE), University of Pernambuco, Campus Petrolina (9º23'54.1" S; 40º28'49.0" W). After 32 months of field transplant, fruits were collected and used for the testing of antibacterial activity in laboratory.

Field experiment
For the field experiment, seedlings produced in a greenhouse were inoculated or not with soil inoculum containing 200 spores of each AMF isolate. We used Acaulospora longula After 225 days, the seedlings were transplanted to the experimental field (Silva et al., 2014a).
In the field, the seedlings were distributed using a spacing of 5 m (5 x 5) apart from each other with irrigation through semi-automatic dripping (8.4 L H2O plant -1 h -1 ). Before the transplant, each pit (40 x 40 x 40 cm) received five liters of vermicompost and 150 g of simple superphosphate. The plants were distributed in six blocks containing 16 plants each, totaling 96 plants in the field area. Surrounding the field, we developed a border which consisted of non mycorrhizal L. ferrea plants (Silva et al., 2014b). Two central plants of each portion were used for the analyses.

In vitro antibacterial activity tests
The in vitro antibacterial activity analyses were carried out in the Laboratory of Microbial Resistance (LRM), University of Pernambuco -Campus Santo Amaro.

Bacterial strains
The strains Escherichia coli (Migula) Castellani and Chalmers (ATCC ® 25922) and Staphylococcus aureus subsp.  The zones of inhibition which were formed surrounding the wells were measured using digital pachymeter (Lee tools, Ltda) (Araújo et al., 2014, modified). Two measures from each well were taken to generate an average for each sample. This test was conducted in technical triplicate for each sample.

Minimum inhibitory concentration (MIC)
The minimum inhibitory concentration was determined by microtritation in Mueller-Hinton broth using microdiluition plates (CLSI, 2012). In the first well of each raw was placed with 200 µL of the extract (100 mg mL -1 of initial concentration) from inoculated and non-inoculated plants. From the second well, the extract was mixed with 100 µL of Mueller-Hinton broth, totaling 200 µL per well. Eight dilutions of the extract were made, the concentrations varied between 100 mg mL -1 and 0.78 mg mL -1 , in technical triplicate for each sample. Each well received 5 µL standardized inoculum of 0.5 McFarland diluted (1:10, in saline solution), the plants were incubated for 20 to 24 h at 37 °C.

Statistical analysis
The data was subjected to ANOVA and the averages compared by using a Duncan test (5 %) on software Assistat (7.7).

Results
The methanolic extracts of L. ferrea fruits inoculated with A. longula had higher in vitro antibacterial activity when compared to those from non-inoculated plants. This was for all the bacterial strains studied ( Table 2).

Treatments
Bacterial strains

Discussion
This is the first record of different effects of mycorrhizal L. ferrea extracts in regards to the bacteria. This effect was probably due to the increased concentration of phenols (26.4 %) and tannins (39.77 %) observed and compared to the methanolic extract of non-inoculated L. ferrea fruits (Table 1). Similar reports have also shown an association of bacterial growth inhibition with the presence of phenols in fruit extracts from plants of Caatinga (Sampaio et al., 2009;Silva et al., 2013a). In addition, the inhibitory effect of tannins on growth of S. aureus-MRSA strains has also been reported (Okuda, 2005;Okuda & Ito, 2011).
Similarly, regarding the phenolic compounds, Karimi et al. (2016) also reported that essential oils extracted from mycorrhizal A. graveolens with F. mosseae had higher action against C. nymphaeae than essential oils of non mycorrhizal plants. The authors attributed such a benefit to the contents of the essential oils carvone and limonene.
However, the presence of FMA does not always increase the efficiency of plant extracts in inhibiting bacterial growth ( Table 2). The zones of inhibition produced by some of the methanolic extracts of mycorrhizal L. ferrea fruits tested against E.
coli (ATCC ® 35218) were similar to those formed by extracts of native L. ferrea fruits collected in the Caatinga region (Table   2) and the MIC test of these strains showed bacterial growth independent of the concentration tested, what did not occurred with Gram-positive strains (S. aureus ATCC® 25923) and S. aureus (ATCC® 33591) which showed growth in the highest concentration (Table 2), this fact, demonstrates the importance of complementary tests to verify the antibacterial activity.
It is likely that, different phytochemical and nutritional profiles of the plant extracts may have interfered in the bacterial growth, such as the phosphorus concentration which is an important nutrient for bacterial growth and has different values in the fruits used in this study according to the AMF inoculated (Table 1). The alterations in the phosphorus concentrations activate, or not, the regulation systems which influence the microorganism growth (Behrendes et al., 2014;Shimizu, 2014). In turn, this may have influenced the bacterial growth regarding the extracts of L. ferrea fruits inoculated with A. longula. In addition, the inhibitory potential of the bacterial growth may vary according to the bacterial species thus giving rise to different efficiencies of the extracts . As verified in our study (Table 2).
AMF inoculation can be efficient at maximizing the in vitro antibacterial activity of extracts of L. ferrea fruits and represents an alternative to the production of raw materials for the material manufacture of phytotherapeutic drugs based on L.
ferrea fruits. The inhibition zones produced by the methanolic extracts of L. ferrea fruits inoculated with A. longula varied from 21.35 to 32.63 mm (Table 2) and were higher than the zones produced by extracts of L. ferrea fruits collected in the Caatinga, which varied between 11 and 18.5 mm Silva et al., 2013a;Conde et al., 2015).

Conclusion
Thus, the inoculation of L. ferrea with A. longula can be an alternative to producing fruits with different antibacterial activities thus contributing to the World Health Organization guidelines, which seek to integrate alternative and traditional medicine (WHO, 2013).
Our tests are still in their primary stage with regards to the production chain of raw materials which can be used in the phytotherapeutic drugs industry. Therefore, it is necessary to develop studies to verify the efficiency of different concentrations of methanolic extracts of L. ferrea fruits, in addition to toxicity tests on the extracts of mycorrhizal plants (Braquehais et al., 2016;Santos et al., 2017). Complementary studies should include assessing the synergic effect of extracts of mycorrhizal L.