Antibacterial/Antifungal Activity and Synergistic Interactions between Polyprenols and Other Lipids Isolated from Ginkgo Biloba L. Leaves

Polyprenols separated from lipids are promising new components from Ginkgo biloba L. leaves (GBL). In this paper, ginkgo lipids were isolated by extraction with petroleum ether, saponification, and molecular distillation. Eight known compounds: isophytol (1), nerolidol (2), linalool (3), β-sitosterol acetate (4), β-sitosterol (5), stigmasterol (6), ergosterol (7), β-sitosterol-3-O-β-D-glucopyranoside (8) and Ginkgo biloba polyprenols (GBP) were separated from GBL by chromatography and identified mainly by NMR. The separated and identified compounds 1, 2 and 3 are reported here for the first time in GBL. The 3D-DAD-HPLC-chromatogram (190–232 nm) of GBP was recorded. This study provides new evidence as there are no previous reports on antibacterial/antifungal activities and synergistic interactions between GBP and the compounds separated from GBL lipids against Salmonella enterica, Staphylocococus aureus and Aspergillus niger. Nerolidol (2) showed the highest activity among all the tested samples and of all mixture groups tested the GBP with isophytol (1) mixture had the strongest synergistic effect against Salmonella enterica among the three tested strains. A proportion of isophytol and GBP of 38.19%:61.81% (wt/wt) was determined by mixture design as the optimal proportion for the synergistic effect of GBP with isophytol against Salmonella enterica.

we report that eight known compounds and GBP can be separated from non-saponifiable lipids of GBL by saponification, refrigeration, chromatography and identified by NMR analysis. The antibacterial/ antifungal activities and synergistic interactions between GBP and compounds separated from GBL lipids are examined according to the study route followed (Figure 1).

Figure 1.
The study route of extraction, isolation and synergistic antibacterial/antifungal effects on GBP and other lipids from GBL.

Structure Determination of Separated Compounds
In total eight known compounds were separated from the different polar portion of the frozen sediment (S2) and the light distillates (S3) that were collected from the non-saponifiable lipids of GBL by chromatography and eight compounds were identified using spectroscopic techniques, mainly 1 H-and 13 C-NMR, as isophytol (1) [14], nerolidol (2) [15], linalool (3) [16], β-sitosterol acetate (4) [17], β-sitosterol (5) [18], stigmasterol (6) [19], ergosterol (7) [19] and β-sitosterol-3-O-β-D-glucopyranoside (8) [20] by comparison with the data of corresponding references ( Figure 2). GBP (contents over 98%) separated from GBL lipids were determined by HPLC using an external standard method [21] and all the retention times and the absorption wavelengths of GBP match with those of standard polyprenols (C 70 , C 75 -C 105 , C 110 , C 115 , C 120 , Figures 3 and 4).   The present investigation supports that ethnobotanical uses of GBL relying on terpene trilactones and flavonoid glycosides, the active components of GBL might be responsible for their antibacterial activity [22]. Comparing with research on other different classes of compounds occurring in GBL extracts, the non-saponifiable lipids components were reported infrequently [1,23]. The separated and identified known compounds isophytol (1), nerolidol (2) and linalool (3) were reported for the first time among the components separated from GBL. Meanwhile, the 3D-DAD-HPLC-chromatogram (190-232 nm) of GBP was established for the first time and the maximum absorption wavelengths corresponding to each polyprenol homolog were recorded from the chromatogram for the first tome too.

Antibacterial and Antifungal Activity of GBL Lipids
Antibacterial and antifungal activity of the total non-saponifiable lipids (S1), the frozen sediment (S2), the light distillates (S3), the heavy distillates (S4) isolated by refrigeration and molecular distillation, the eight individual compounds and GBP was assessed in the present study at 500 μg/mL against three animal and plant pathogenic strains (Salmonella enterica, Staphylocococus aureus and Aspergillus niger) and their potencies were quantitatively assessed by their inhibition halos (Table 1) Table 2). Analysis of variance (Tukey's test at 5% probability) indicated statistical differences (p < 0.05) among all the samples. The results from the diameters of inhibition halos indicated that nerolidol (2) showed the highest activity among all the tested samples and inhibited the growth of all the strains. MIC, MBC and MFC values for the examined strains sensitive (inhibition halos 16.2-20.1 mm) to nerolidol (2) were in the range of 3.9-15.6 μg/mL, 31.3-62.5 μg/mL and 62.5 μg/mL, respectively. GBP (inhibition halos 13.4-13.8 mm) showed the MIC values were 31.3 μg/mL, MBC and MFC were 125 μg/mL. The results from the inhibition halo diameters showed the activity could be ranked from high to low (the same below) in the following order: nerolidol (2), linalool (3), the heavy distillates (S4), the total non-saponifiable lipids (S1), the light distillates (S3), GBP and isophytol (1). These seven samples were effective in inhibiting the growth of all three examined strains. The five samples of β-sitosterol (5), the frozen sediment (S2), β-sitosterol acetate (4), ergosterol (7) and stigmasterol (6) were effective at inhibiting two of the types of bacteria examined in this study. β-Sitosterol-3-O-β-D-glucopyranoside (8) was only effective at inhibiting Staphylocococus aureus.  All the samples separated from the light distillates (S3) and the heavy distillates (S4) were effective at inhibiting the growth of the three examined strains. None of the samples separated from the frozen sediment (S2) were active against Aspergillus niger.
Nerolidol showed the highest activity among all the tested samples and inhibited the growth of all the strains used in this study. It was suggested that nerolidol possesses antifungal activity against T. mentagrophytes and the activity may lead to irreversible cellular disruption [24]. Nerolidol enhanced the susceptibility of Staphylococcus aureus to ciprofloxacin, clindamycin, erythromycin, gentamicin, tetracycline, and vancomycin. Nerolidol also sensitized Escherichia coli to polymyxin B [25]. Consequently, it would be meaningful to research on expanding the antimicrobial spectrum of nerolidol in the future. In our previous study we reported on hepatoprotective and antitumour effects of the polyprenols of GBL in rats [7,26,27]. This study supports further research as there are no previous records on the anti-microorganism activity of GBP. Especially, the polyprenols are an important component in the non-saponifiable lipids (S1, GBP content > 40%) and in the heavy distillates (S4, GBP content > 80%) of GBL [9,28]. Besides, there is new interest in finding out the other lipids in the heavy distillates S4 having higher antibacterial and antifungal activities than the polyprenols from the above results. Therefore, it can be inferred GBP have synergistic inhibitory effects on microorganisms with other lipids, so it was indispensable to research the synergistic inhibitory effect on GBP with other lipids compounds in the next step.

Synergistic Antibacterial and Antifungal Effects on GBP with Separated Compounds
The method for assessing synergistic antibacterial/antifungal effects of GBP with separated compounds was by comparing the diameters of inhibition halos between theoretical values and actual values of mixture groups (compounds 1-8:GBP, 1:1, wt/wt) by analysis of variance (Tukey's test at 5% probability) from Table 1, and the Fractional Inhibitory Concentration (FIC) indexes from Table 3. The results showed that two mixture groups C1:GBP and C5: GBP had statistically significant differences (Tukey's test, p > 0.05) compared with the corresponding theoretical values of diameters of inhibition halos among all the mixture groups against all the tested strains. The mixture group C1:GBP had the most significant statistical difference (Tukey's test, F = 690.036, p = 0.001) against Salmonella enterica compared with other mixture groups. The FIC index of C1:GBP against Salmonella enterica was 0.25, that was lowest in all mixture groups against all the tested strains. This result suggested that the GBP with isophytol (1) mixture group had the strongest synergistic effect against Salmonella enterica of all mixture groups against the three tested strains. Isophytol is an important component of many essential oils with reported antimicrobial activity [29][30][31]. No studies on synergistic antimicrobial effects of isophytol and GBP have been reported. From here we saw that it was necessary to establish the optimal proportions for the synergistic effect of GBP with isophytol against Salmonella enterica in the next study.

Optimal Proportioning Design of Synergistic Effect on GBP with Isophytol against Salmonella Enterica
The optimal proportion determination of synergistic effect on GBP with isophytol against Salmonella enterica was based on the Mixture Design (D-optimal, two mixture components, two factors) method (Table 4)

Materials
The dried GBL were collected in October 2011 from China's Jiangsu Province. This plant was identified and authenticated by Prof. Cheng-Zhang Wang at the Institute of Chemical Industry of Forestry Products, CAF in China. Three types of strains (Salmonella enterica ATCC 14028; Staphylocococus aureus ATCC 25923; Aspergillus niger ATCC 16404) bought from the China Center for Type Culture Collection (CCTCC, Wuhan, China) were used. A Bruker AV-300 NMR instrument was used for compound identification, using CDCl 3 or DMSO-d 6 as solvents and TMS as internal standard. HPTLC plates (silica gel 60) were obtained from Merck. The standard polyprenols (C 70 , C 75 -C 105 , C 110 , C 115 , C 120 ) were purchased from Larodan Fine Chemical Co., Ltd, (Malmö, Sweden). HPLC measurements was performed at room temperature with a Shimadzu SPD-20A instrument equipped with DAD detector (210 nm) and a 2.5 μm Thermo BDS HYPERSIL C18 (150 × 4.6 mm) column, using 64/36 isopropanol/methanol solvent mixture as the eluent at 0.5 mL/min for 60 min.

Extraction and Isolation
Shade air-dried (at room temperature) and pulverized (over 80 mesh sieve) GBL (10 kg) were extracted three times with 30 L (total) petroleum ether (b.p. 60-90 °C) for 24 h at 65-70 °C and concentrated to give an extract (500 g) which was mixed with 5% NaOH-EtOH solution (6 L) for 3 h at room temperature. The hydrolysate was extracted with petroleum ether (6 L) three times. The collected organic phases were washed with water to neutrality and dried with anhydrous Na 2 SO 4 . The solvent was evaporated under vacuum to give the total non-saponifiable lipid extract (S1, 350 g) which was dissolved in a solvent mixture (acetone : methanol = 85:15, v/v) for a solid-liquid ratio of 1:6-1:8 (g/mL), then refrigerated for 2 h at −15 °C [7,9,28]. As a result, the frozen sediment (S2, 35 g) was obtained by quickly filtering at low temperature the solids from the refrigerated solution and the dissolved matter was concentrated to yield a product as a brown oil (215 g). The brown oil was fractionated by molecular distillation at a feed temperature of 60 °C, distillation temperature of 280 °C, feed flow rate of 180 mL/h, scraper rate of 300 rpm, and operating pressure of 0.1-0.5 Pa to give the light distillates (S3) as a yellow oil (44 g) and the heavy distillates (S4) as a dark brown oil (168 g).

HPLC Analysis
The concentration of the GBP sample was 4.38 mg/mL and the injection volume was 5 μL.

Determination of Antibacterial and Antifungal Activity
Antibacterial and antifungal tests of selected stains were carried out using a disc-diffusion method [32]. A small sterile cotton swab was dipped into the 24-h-old culture of stains and was inoculated by streaking the swab over the entire agar surface. After inoculation the plates were allowed to dry at room temperature in laminar chamber. The filter paper discs (6 mm) loaded with 100 μL of sample were placed on the surface of the agar plates. After 5 min the plates were incubated at 37 °C for 24 h. Miconazole nitrate (Sigma SM351201, 1 g) and gentamycin sulfate (Sigma G3632, 100 mg) were used as positive control and the respective solvent as negative control. After 24 h of incubation, the diameter was observed for inhibition halos (measured in mm including disc size). All tests were performed in triplicate and observed values of inhibition halos were expressed as mean value with standard error of means (SEM).

Determination of Minimum Inhibitory Concentration (MIC), Minimum Bactericidal Concentration (MBC), Minimum Fungicidal Concentration (MFC), FIC (Fractional Inhibitory Concentration) Index and Determination of the Type of Interactions of Antibacterial and Antifungal Activity
MIC, MBC and MFC were determined using the broth-dilution method. MIC was performed at seven concentrations of samples (250, 125, 62.5, 31.3, 15.6, 7.8, 3.9 μg/mL) following serial dilution technique. All the wells showing no visible growth of strains were subcultured and incubated at 37 °C (Salmonella enterica, Staphylocococus aureus) and 28 °C (Aspergillus niger) overnight. The highest dilution showing 100% inhibition was recorded as MBC or MFC [12]. The FIC is the concentration that kills when used in combination with another agent divided by the concentration that has the same effect when used alone [33]. The FIC index for the combination of A and B is the sum of their individual FIC values. The determination of the type of interactions referred to synergistic effect (0 < FIC index ≤ 0.5), additive effect (0.5 < FIC index ≤ 1), indifferent effect (1 < FIC index ≤ 4) and antagonism effect (FIC index > 4) [34][35][36].

Optimal Proportioning Design of Synergistic Effect on GBP with Isophytol against Salmonella Enterica
This design was based on the Mixture Design (D-optimal, two mixture components, two factors, the limits: 5%-95%) option in the Design Expert 7.1.3 Software that generated the experimental scheme (13 standard/run) randomly. All samples were examined at 500 μg/mL (total mass concentration). The components A and B were isophytol and GBP, respectively. The responses 1 and 2 were the FIC index and diameters of inhibition halos, respectively.

Conclusions
The eight known compounds isophytol (1), nerolidol (2), linalool (3), β-sitosterol acetate (4), β-sitosterol (5), stigmasterol (6), ergosterol (7) and β-sitosterol-3-O-β-D-glucopyranoside (8) were separated from GBL by chromatography and identified by NMR. The separated and identified compounds 1, 2 and 3 were reported for the first time from GBL. The 3D-DAD-HPLC-chromatogram (190-232 nm) of GBP was recorded for the first time. Meanwhile, this study provides the first evidence of the antibacterial/antifungal activities and synergistic effect on GBP with compounds separated from GBL lipids against Salmonella enterica, Staphylocococus aureus and Aspergillus niger. Nerolidol (2) showed the highest activity among all the tested samples, and the GBP with isophytol (1) mixture group had the strongest synergistic effect against Salmonella enterica among all mixture groups against the three tested strains. The proportion of isophytol and GBP of 38.19%:61.81% (wt/wt) was determined as the optimal proportion of synergistic effect on GBP with isophytol against Salmonella enterica. This study provides a new scientific basis for the ethnomedical use of GBL against bacterial and fungal diseases of animals and plants.