Prenylated flavonoid-enriched fraction from Maclura tinctoria shows biological activity against Staphylococcus aureus and protects Galleria mellonella larvae from bacterial infection

Background The Atlantic Forest biome extends along the entire Brazilian coast and is home to approximately 20,000 plant species, many of which are endemic; it is considered one of the hotspot regions of the planet. Several of these species are sources of natural products with biological activities that are still unknown. In this study, we evaluated the antimicrobial activity of 90 extracts derived from native Atlantic Forest tree species against Staphylococcus aureus, an important human and veterinary pathogen. Methods Extracts from native Atlantic Forest tree species were evaluated for their antimicrobial activity against S. aureus by in vitro standard methods. Phytochemical fractionation of the extract from Maclura tinctoria was performed by liquid-liquid partitioning. LC-DAD-ESI-MS was used for identification of constituents in the most active fraction. Damage of cells and alterations in the permeability of cell membrane were determined by atomic force microscopy (AFM) and crystal violet uptake assay, respectively. In vivo antimicrobial activity was evaluated using Galleria mellonella larvae infected with S. aureus with survival data collected using the Kaplan-Meier method. Results Among the organic or aqueous extracts tested here, 26 showed biological activity. Eight species showed relevant results, with a minimum inhibitory concentration (MIC) below 1 mg/mL. Antibacterial activity was registered for three species for the first time. An organic extract from Maclura tinctoria leaves showed the lowest MIC (0.08 mg/mL). Fractionation of this extract by liquid-liquid partitioning led to obtaining fraction 11FO d with a MIC of 0.04 mg/mL. This fraction showed strong activity against veterinary S. aureus isolates and contributed to the increased survival of Galleria mellonella larvae infected with S. aureus ATCC 29213. The bacterial surface was not altered by the presence of 11FO d, and no cell membrane damage was detected. The LC-DAD-ESI/MS analyses identified prenylated flavonoids as the major constituents responsible for the antibacterial activity of this active extract. Conclusion A fraction enriched in prenylated isoflavones and flavanones from M. tinctoria showed in vitro and in vivo efficacy as antistaphylococcal agents. These findings justify the need for further research to elucidate the mechanisms of action of these compounds.


Background
Staphylococcus aureus is responsible for high rates of morbidity and mortality worldwide and is the most common cause of skin infections, although it may also be associated with osteomyelitis, bacteremia and sepsis [1]. In cattle, it is the primary cause of mastitis, which is currently considered the main disease of dairy herds [2]. Great adaptations to different niches and hosts are determined in part by the variety of bacterial virulence factors that contribute to the establishment of the disease.
S. aureus has a great ability to develop antimicrobial resistance. The first penicillin-resistant strains arose in the mid-1940s, and as early as the 1960s, reports described the isolation of S. aureus with broad resistance to beta-lactams, especially methicillin (MRSA) [3]. At first, the infections caused by MRSA were limited to hospitalized patients (HA-MRSA), but since the 1990s, strains associated with community-acquired infections (CA-MRSA) have also been isolated [4]. There are different strains of CA-MRSA around the world that are typically prevalent in certain regions, such as the ST 80 strain that is disseminated throughout Europe and ST 30 in Australia [4]. The USA 300 strain, which is associated with infections in hospitals and in the community, was initially reported in the US. However, it has been isolated from different continents, and it has become a global concern due to its hypervirulence, reduced susceptibility to different antimicrobials and great epidemic potential.
Plant resources can also be sources of compounds with antimicrobial properties. Several studies on plant extract libraries have been established, and the results show the importance of this strategy for the discovery and development of new drugs [5,6]. Coumarin derivatives have displayed activity against S. aureus and Bacillus anthracis DNA helicases [7]. Further reports have described the efficacy of intraperitoneal injections of a synthetic dicoumarin derivative in septicemic mice [8] and hypodermic injections of lysionotin in mouse pneumonia model [9], confirming the potential use of natural products as leads for new drugs.
The Atlantic Forest biome is the second-largest block of tropical forest in Brazil and has the second-highest biological diversity, behind only the Amazon. As an area of great potential for medical research, the forest originally covered 15% of the Brazilian territory but it presently suffers from constant deforestation pressure due to urban expansion and the withdrawal of trees for conversion into coal, which endangers the fauna and flora of the region. Considering the great diversity of the Atlantic Forest biome, which hosts plant species still unknown to science, along with the urgency to discover new antimicrobial agents, the present work presents a bioprospection of natural products in the native trees extracts from this ecosystem. The antimicrobial activity of some of these plants was demonstrated for the first time.

Methods
Microorganisms and culture conditions S. aureus ATCC 29213 was used as a test strain to evaluate the antibacterial activity of the plant extracts. Four field isolates of S. aureus (302/4, 469/4, 4130, and 4651), which originated from cows presenting mastitis, were also used to determine the MIC of the active fraction. The bacteria were maintained on Brain Heart Infusion agar (BHA, Fluka, Munich, GER) or Müeller-Hinton agar (MHA, Himedia, Mumbai, IND). For the antibacterial assays, bacteria were streaked onto MHA and cultured overnight at 37°C before being used for inoculum preparation as described below. Stock cultures were stored at − 80°C in BHI medium containing 20% glycerol.

Plant material and extraction
Plant material from 23 species (Table 1) native to the Atlantic Forest were collected from a forest fragment at the coordinates 20°48′07″ S and 42°51′31″ W, which is part of the Universidade Federal de Viçosa (UFV), Minas Gerais, Brazil. The permission issued by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) granted access to genetic heritage (permit No 010134/2014-0). A specimen voucher of all the samples was deposited in the VIC Herbarium of the UFV and given a voucher number (VIC 40.269). Identification was performed by Marcos Vinícius R.C. Simão with the aid of specialized bibliography, consultation with botanicals, and comparison with herborized material from the VIC Herbarium determined by specialists. The Brazilian Flora 2020 Project was used to check the species names and the botanical families. The leaves and stems were separated, dried in a ventilated oven at 40°C for 48 h and then subjected to extraction by serial maceration using a dry vegetable/solvent ratio of 1:5 (w/v). Each sample was first extracted with a mixture of lower-polarity solvents (dichloromethane and methanol, 1: 1) followed by water, and for both extracts, the solvents were completely removed [10]. The tree species were enumerated randomly, and the extracts were identified by the abbreviations FO (organic extract of leaves), FA (aqueous extract of leaves), GO (organic extract of branches) and GA (aqueous extract of branches). Stocks of all the plant extracts were prepared at a 50 mg/mL concentration in dimethyl sulfoxide (DMSO, Vetec, Rio de Janeiro, BR) for use in the biological tests.

Agar well diffusion assay
The antimicrobial activity was initially evaluated according to the well diffusion test. Inoculum was prepared as     The tests were performed on three independent biological replicates. The mean value ± SD for the replicates was calculated.

Determining the minimum inhibitory concentration (MIC) of the plant extracts
A 96-well microplate was filled with MH broth, with extract concentrations ranging from 0 to 10 mg/mL and 10 5 CFU/mL of a suspension of S. aureus 291213. The control wells consisted of inocula added to 100 μL of the MH broth and MH broth plus DMSO in the volume corresponding to the highest concentration of tested extract. The microplates were incubated at 37°C for 24 h. Subsequently, 5 μL of violet p-iodonitrotetrazolium (INT, I8377, Sigma-Aldrich, Dorset, UK) was added at 2 mg/mL. After 2 h of incubation, the MIC was defined as the lowest concentration of extract for which the purple color formation, which was indicative of cell viability, was not observed. The MIC of ampicillin for S. aureus 29213 was also determined. The MIC of the active fraction was estimated as described above using the reference strain and the isolates S. aureus 302/4, 469/4, 4130, and 4651. The assays were performed using three biological replicates.
Fractionation and phytochemical prospecting of the bioactive extract The extract with the lowest MIC according to the preliminary screening was chosen for further assays. The phytochemical composition was investigated by liquid-liquid partitioning in which the bioactive extract (1.0 g) was resuspended in water (w) and partitioned into a separatory funnel in hexane (h), dichloromethane (d), ethyl acetate (ac), and n-butanol (b). A volume of 100 mL of each solvent was used in this step. The solvents were evaporated to give the 11 FO h (0.087 g), 11 FO d (0.333 g), 11 FO ac (0.017 g), 11 FO b (0.092 g), and 11 FO w (0.318 g) fractions. All the fractions were subjected to phytochemical analysis and an antibacterial activity assay. A phytochemical screening was performed by thin layer chromatography in glass plates covered with silica gel (Whatman, Kent, UK). The extract and fractions were investigated for the presence of cumarins, tannins, flavonoids, anthraquinones, essential oils, saponins, alkaloids, and triterpenes/steroids using standards and specific reagents for each class of secondary metabolites [12].

LC-DAD-ESI/MS analyses
The most active fraction of the fractionated extract was subjected to analyses by LC-DAD-ESI/MS using UPLC (ACQUITY) system coupled with a photodiode array detector (Waters) and on Triple Quadrupole instrument (Waters ACQUITY® TQD) with an electrospray source (ESI) operating in the positive and negative mode. The ESI-MS/MS operation was performed using capillary voltage, 3500 V; capillary temperature, 320°C; source voltage, 5 kV; vaporizer temperature, 320°C; corona needle current, 5 mA; and sheath gas, nitrogen, 27 psi with argon as the collision gas, and the collision energy was set at 30 eV. Analyses were run in the full scan mode (100-1500 Da). Chromatographic separation was done on Acquity UPLC HSS C 18 column (50 mm × 2.1 mm i.d.; 1.7 μm; Waters, USA) in combination with an Acquity UPLC guard column (5 × 2.1 mm; 1.7 μm; Waters, USA) [13]. The mobile phase consisted of water 0.1% formic acid (solvent A) and acetonitrile 0.1% formic acid (solvent B). The elution protocol was 0-11 min, linear gradient from 5 to 95% B. The flow rate was 0.3 mL min − 1, and the sample injection volume was 4.0 μL. The UV spectra were registered from 190 to 450 nm. Phenolic compounds were identified on the basis of the typical UV absorption for each phenolic class analyzed here, in addition to the typical fragmentation patterns obtained by MS/MS in comparison to the literature data [14][15][16][17][18][19].
Effect of the 11FO d fraction on S. aureus growth The reference strain S. aureus 29213 was used to analyze the antibacterial activity of the 11FO d fraction [20]. A bacterial growth curve with an initial inoculum of 10 6 Atomic force microscopy (AFM) S. aureus ATCC 29213 was grown in BHI medium at 37°C for 12 h, centrifuged (2100 x g for 15 min), washed three times in phosphate buffer (5 mM, pH 6.5) and resuspended in the same buffer at approximately 10 8 CFU/mL. The plant extract was then added at the MIC (0.04 mg/mL) for 4 h at 37°C. Cells grown in culture media without the fraction were used as the control. After centrifugation, the pelleted cells were deposited on mica substrates for AFM (Ntegra Prima, NT-MDT, RUS). The AFM images were recorded using the intermittent contact mode and silicon probes with a 10 nm tip radius, 240 kHz resonant frequency and 11 N/m force constant [21].

Measurement of permeability with crystal violet
Alterations in the cell permeability were evaluated by crystal violet assay [22]. In brief, a culture of S. aureus 29213 was centrifuged, and the cells were resuspended in 0.5 mM potassium phosphate buffer, pH 7.4. The cells were incubated at 37°C for 120 min with CTAB (0.3, 30, and 120 μg/mL) or with the 11FO d fraction at different MIC values (1/2 to 8X). CTAB is a cationic detergent that increases cell permeability and therefore was used as a positive control. The negative control consisted of cells incubated only in PBS. The samples were centrifuged at 9300 x g for 5 min, resuspended in PBS containing 10 μg/mL of crystal violet and incubated at 37°C for 10 min. After further centrifugation, the supernatants were collected and read at an OD of 590 nm. The absorbance of the violet crystal solution was also checked. The percentage of crystal violet uptake was calculated as the (OD of samples / OD of crystal violet solution) × 100.

In vivo antimicrobial activity
Galleria mellonella was reared on an artificial diet at 25°C. The diet consisted of a mix of 400 g wheat bran, 200 g wheat germ, 120 g dried beer yeast, 80 mL glycerin, 200 g skim milk powder, and 300 mL liquid honey. Larvae weighing 250-350 mg were divided into six groups. Unless otherwise stated, all the experiments involved groups containing 10 larvae. Group 1 was injected with a suspension of S. aureus ATCC 29213 at a 10 5 CFU concentration. Group 2 received the same inoculum as Group 1 but 2 h after bacterial injection 10 μL of the 11FO d fraction were injected into larvae's last proleg. Before injection, this fraction was diluted in PBS at a dose corresponding to 200 mg/kg. Every 24 h, a new dose of the fraction was applied, and the larvae were monitored for 96 h after the last application. A total of four doses of the fraction were applied to each larva in this group. Group 3 resembled Group 2, but instead, the 11FO d fraction was replaced by gentamicin at 200 mg/ kg (G1397, Sigma-Aldrich, Dorset, UK). The other groups (4-6) consisted of uninfected controls, which were treated with a single injection containing 10 μL of PBS, the 11FO d fraction, or DMSO. The larvae were kept in petri dishes in a dark environment at 37°C. The experiments were performed using three biological replicates.

Statistical analyses
Survival data were collected using the Kaplan-Meier method, and the comparison between the control group and the treatment group was performed using the Logrank test. A p ≤ 0.05 was considered significant. Data obtained from the agar well diffusion test and the cell permeability assay were subjected to one-way analysis of variance (ANOVA) followed by Dunnett's test for comparison of more than two groups using PRISM 6 statistical software (San Diego, CA, USA). The results were considered statistically significant when p ≤ 0.05.

Screening of extracts for antimicrobial activity
Twenty-six organic and aqueous extracts tested on S. aureus ATCC 29213 by the well diffusion test exhibited antimicrobial activity ( Table 1). The largest inhibition halo was recorded for the organic extract of Miconia petropolitana leaves (24FO). The extracts from Anadenanthera peregrina led to the formation of inhibition zones ranging from 9 to 13 mm. Inhibition halos were observed when the leaf extracts from Maclura tinctoria (11FA and 11FO) were tested. Different extracts from Bathysa nicholsonii, Piptadenia gonoacantha, Tabernaemontana hystrix, Xylopia sericea, and Myrciaria glazioviana also showed activity against S. aureus. One extract from Luehea grandiflora, Ceiba speciosa, Siparuna guianensis, and Alchornea glandulosa produced inhibition halos. MIC assays were performed with all the extracts that resulted in inhibition zones and values ranged from 0.08 to 10 mg/mL ( Table 1). The 11FO organic extract prepared from Maclura tinctoria leaves had the smallest MIC, and therefore, it was chosen for later assays.

Phytochemical prospecting and bioactivity-guided fractionation of Maclura tinctoria
Coumarins, flavonoids, essential oils, saponins, tannins and triterpenes/steroids were detected in the 11FO extract. The MICs of the fractions obtained with the solvents hexane (11FO h), ethyl acetate (11FO ac), n-butanol (11FO b), and water (11FO a) ranged from 0.04 to 10 mg/mL. The MIC obtained for the dichloromethane fraction of 11FO d was reduced to half that obtained for the crude extract (Fig. 1). The MIC of fraction 11FO d for bovine S. aureus isolates was also determined, and an antimicrobial activity between 0.01 and 0.04 mg/mL was recorded (Fig. 1). TLC showed that the fraction 11FO d consisted of a mixture of flavonoids. The phytochemical profile of active fraction 11FO d was determined by UPLC-DAD (Fig. 2), which showed the presence of five major compounds. All compounds showed UV spectra with maximum absorption (λ max. ) near 264 and 270 nm, which are typical wavelength values for isoflavone or flavanone derivatives. Full scan mass spectra (100-1500 Da) were obtained for 11FO d by LC-DAD-ESI-MS in positive and negative modes, and their molecular ion and fragmentation analyses allowed us to confirm the structures of five prenylated flavonoids (Fig. 3). . Compound 5 also showed spectra suggestive of a flavanone (λ max 270 nm). The MS spectrum in positive mode was focused on a m/z 407 with an abundant product ion at m/z 351 that was attributed to a loss of 56 Da, which is characteristic of the presence of prenyl group. The spectral data suggest that compound 5 is a diprenylated flavanone.

Effect of fraction 11FO d on S. aureus ATCC 29213
After 4, 8, and 12 h of exposure to 2X MIC, MIC and 1/ 2 MIC, there was no further bacterial growth, which revealed the bactericidal effect of the fraction (Fig. 4). To explore the possible antibacterial mechanism of fraction 11FO d, the morphological changes in S. aureus were observed by AFM. No remarkable changes in the bacterial surface or cellular dimensions (1 μm × 500 nm) were apparent, although a decrease in the number of bacteria was observed after 4 h of treatment, as were changes in the cellular arrangement (Fig. 5). The alterations in cellular permeability were tested by incubating the cells with crystal violet, a hydrophobic dye that is capable of crossing the lipid bilayer (Fig. 6). In the presence of increasing concentrations of CTAB, a cationic detergent, there was greater dye entry, which is consistent with the increased permeability of cell membrane. The same trend was not observed when the cells were incubated in the presence of the 11FO d fraction, suggesting that the membrane properties were not altered by the compounds found in the fraction. Cells treated with the fraction had a lower uptake of crystal violet in relation to the cells without the added fraction (control). In vivo antimicrobial activity The 11FO d fraction used at the 200 mg/kg concentration increased the survival of insects that were infected with S. aureus ATCC 29213 when compared to those in the control groups. After 24 h, 50 % of the larvae were dead compared to the high survival seen in the group that received the fraction. After 48 h, there were still live larvae in the group treated with 11FO d. All the larvae injected with PBS, PBS and DMSO, or PBS and extract survived after 96 h of monitoring (Fig. 7).

Discussion
The relevance of the Atlantic Forest biome as a reservoir of chemical compounds with antibiotic potential was demonstrated in this work. Of the 23 species tested, 12 presented antimicrobial activity against S. aureus, with highlights for Tabernaemontana hystrix, Anadenanthera peregrina, Miconia petropolitana, Xylopia sericea, Myrciaria glazioviana, and Maclura tinctoria. This is the first report of the antibacterial activity of species such as T. hystrix, which has been described only for its anticholinesterase activity, [23] and A. peregrina, for its antioxidant property [24]. No studies on M. petropolitana have been found, although other species of this genus are known for their antiparasitic, antibiotic, analgesic, and antitumor activities [25]. Among all species evaluated in the present study, M. tinctoria showed the most promising antimicrobial activity. M. tinctoria D. Don Ex Steud is a tree species of the family Moraceae, with wide distribution in Brazil. The bioassay-guided fractionation conducted here resulted in a fraction enriched in prenylated isoflavones and flavanones, which led to a greater reduction in MIC compared to the crude extract, evidencing that these compounds are responsible for the antibacterial activity seen. The dichloromethane fraction (11FO d) also showed a bactericidal action, which contributes to a faster elimination of the pathogen, decreasing the probability of infection propagation and the emergence of resistance when compared to a bacteriostat. Prenylated flavanones and isoflavones are flavonoids that have already been identified in the Maclura genus [26,27].
Although there is already a report on the antibacterial activity of extracts from M. tinctoria [28], the present work corroborates this result with new scientific evidence, also associating the secondary metabolites responsible for this activity to this species. Studies performed with plant extracts are usually restricted to in vitro assay methods. Here, the in vivo model indicated that the fraction 11FO d conferred a remarkable degree of protection against S. aureus ATCC 29213. Additionally, this model demonstrated that the experimental dose used (200 mg/kg) was not toxic to the larvae. Further studies are underway to investigate the possible synergism of fraction 11FO d with commonly used antibiotics and the effect of other dose regimens. The use of G. mellonella to evaluate the efficacy of phytochemicals has been increasing in recent years [29], and the results show that the larvae are a good alternative for assessing the in vivo efficacy of plant extracts, although tests in standard models should be conducted. Furthermore, larvae rearing is quite inexpensive and experimentation does not require ethical approval [30].
The antibacterial compounds of M. tinctoria had no effect on the cell surface since the S. aureus cells treated with the fraction 11FO d had similar morphology to the control but lost typical grape-like clusters arrangement. Fraction 11FO is rich in prenylated flavanones and isoflavones, which are hydrophobic compounds that could, in principle, disrupt the membrane function. However, no change in cell membrane permeability was detected by the method used herein. Since prenylation can contribute to diversification of flavonoids [31] the strong antibacterial activity seen may result from the interaction of the compounds with different bacterial targets.

Conclusion
This is the first report of the antibacterial activity of T. hystrix and A. peregrine. Flavanones and isoflavones were enriched in the active fraction of M. tinctoria against S. aureus that protected infected larvae suggesting a possible use of prenylated flavanones and isoflavones as antistaphylococcal agents. Further research is ongoing to elucidate the mechanisms of action of these compounds and to evaluate their synergistic effects with other antimicrobials.