Chemical Constituents from the Fruits of Forsythia suspensa and Their Antimicrobial Activity

Lignans and phenylethanoid glycosides purified from Forsythia suspensa were reported to display various bioactivities in the previous literature, including the antimicrobial activity. Therefore, the present research is aimed to purify and identify the chemical constituents of the methanol extracts of fruits of F. suspensa. The methanol extracts of fruits of F. suspensa were fractionated and further purified with the assistance of column chromatography to afford totally thirty-four compounds. Among these isolates, 3β-acetoxy-20α-hydroxyursan-28-oic acid (1) was reported from the natural sources for the first time. Some of the purified principles were subjected to the antimicrobial activity examinations against Escherichia coli to explore new natural lead compounds.


Introduction
Food safety is an important public health issue continuously attracting researchers from various fields. The use of biopreservatives and pathogen antagonists had been completed as a means of protecting the microbiological safety of fresh and processed products [1][2][3][4]. Lignans and phenylethanoid glycosides are widely distributed among plant bioresources and those purified from Forsythia suspensa have already been reported to exhibit antimicrobial bioactivities in the previous literature [5][6][7][8][9][10][11][12][13]. Although these natural compounds did not exhibit better inhibition of the bacterial growth, they were not very toxic while compared with the synthetic antibiotics. F. suspensa (Oleaceae) is an important original plant of the crude drug "rengyo" (Forsythiae Fructus) which has been used for anti-inflammatory, diuretic, drainage, and antimicrobial purposes in Oriental medicine [6,8]. Previous phytochemical investigations of Forsythia genus afforded a series of steroids, triterpenoids, lignans, and phenylethanoid glycosides [5][6][7][8][9][10][11][12][13][14][15][16]. In our continuous program aimed to the bioactive principles from natural sources, the fruits of F. suspensa were selected as the target due to their antimicrobial potential in our preliminary bioassay (Table 1). In the present study, we wished to report the structural characterization of one new triterpene, 3 -acetoxy-20 -hydroxyursan-28-oic acid (1), along with thirty-three known compounds, as well as their antimicrobial effects against E. coli. We hoped to explore new lead compounds which could be performed for further investigation of the new antibiotic agents.

General Procedure. Melting point was determined by a
Fisher Scientific melting point measuring apparatus without corrections. The IR spectrum was obtained, as a KBr disc, on a Bruker Tensor 27 FT-IR spectrometer. Optical rotation was measured with an Atago AP-300 automatic polarimeter. 1 H-and 13 C-NMR, COSY, HMQC, HMBC, and NOESY spectra were recorded on the Varian Unity 400 and Bruker AV 500 NMR spectrometers, using tetramethylsilane (TMS) as the internal standard. Standard pulse sequences and parameters were used for the NMR experiments and all chemical shifts were reported in parts per million (ppm, ). The low and high-resolution FAB mass spectra were obtained on a JEOL JMS-700 spectrometer operated in the positive-ion mode. All the chemicals were purchased from Merck KGaA (Darmstadt, Germany) unless specifically indicated. Column

Determination of the In Vitro Antimicrobial Activity.
The antimicrobial activities against E. coli of different concentrations of tested samples were determined by the microtiter plate method described by the United States Pharmacopeia [17]. A twofold microdilution broth method was used to determinate the minimum inhibitory concentrations (MIC) value for each test substance [18][19][20][21]. Each well contained 10 6 CFU/mL of test bacteria and LB medium (100 L). 100 L of MeOH solutions of tested samples (5 mg/mL for pure compounds and 20 mg/mL for the fractions) was added to wells of the first row. Dilutions were used to dispense 100 L into the other sterile 96 wells of an ELISA plate using a multichannel micropipette, resulting in eight concentrations to be tested for each compound. A negative control containing inoculated growth medium and methanol was prepared. Each experiment was performed in triplicate.

Minimum Inhibitory Concentration (MIC) Determination.
The MIC value is a measure to define the antibacterial activity of a compound and is defined as the lowest concentration of drug that inhibits visible growth. The amount of growth in the wells containing test samples was compared with the amount of growth in the control wells when determining the growth end points. When a single skipped well occurred, the highest MIC was read.

Structural
Elucidation of Compound 1. The purified white powder 1 was visualized by spraying with 1% (w/v) Ce(SO 4 ) 2 in 10% (v/v) aqueous H 2 SO 4 followed by heating at 120 ∘ C and displayed purplish black spots on TLC plate. It also displayed positive responses against the Lieberman-Burchard test. These results suggested compound 1 to be a triterpenoid [50]. The molecular formula of 1 was established as C 32 H 52 O 5 by the pseudomolecular [M+H] + ion peak at m/z 517.3896 in HR-FAB-MS analysis and was further supported by its 13 C-NMR spectrum which showed signals for all the 32 carbons of the molecule including one set of acetyl group ( C 171.0, , and one acetyl methyl group at 2.05 (3H, s, and CH 3 -32), respectively. The spectroscopic data indicated compound 1 to possess oleanane type basic skeleton. In the downfield region, one oxygenated proton at 4.48 (1H, dd, J = 10.4, 5.6 Hz, H-3 ) was located at C-3 which was further established by the NOESY correlations between CH 3 -23 and H-3. The 2 J, 3 J-HMBC correlations from 4.48 (H-3) to C 21.3 (C-32) and 171.0 (C-31) also evidenced the presence of acetoxy group at C-3. The substitution of tertiary alcohol at C-20 was also determined with the HMBC analysis of correlations from CH 3 -30 to C-21 ( C 31.2) and C-20 ( C 90.1). The 2 J, 3 J-HMBC correlation peak between 2.60 (m, H-16) and C 176.8 (C-28) supported the carboxylic acid group to be attached at C-17. The complete assignments of 1 H and 13 C NMR signals of 1 were furnished from the NOESY and HMBC spectra. Therefore the chemical structure of 1 was established as 3 -acetoxy-20 -hydroxyursan-28-oic acid and shown in Figure 1.

The Antimicrobial Effects of Isolated Compounds against
Escherichia coli. The crude extracts, partially purified fractions, and some of the purified principles ( Figure 2) were subjected to the examinations for the inhibitory effects against E. coli [17][18][19][20][21]. The MIC data of the fractions were presented in Table 1. The MIC value of crude extracts (FS) was 4.25 mg/mL and demonstrated inhibition of the bacterial growth. Comparatively, the chloroform fraction (FSC) displayed more significant inhibitory effects against E. coli (BCRC-11634) than the water fraction (FSW) with MIC values of 6.25 and 12.50 mg/mL, respectively. When studying the influence of the concentration of compounds on the antimicrobial activities against E. coli, twofold microdilution broth method was used for the purified principles from the chloroform fraction (FSC), including triterpenoids 1, 2, 6, and 10; diterpenoids 11 and 12; and lignan 16. It was observed that as the concentration increased, the inhibition of the bacterial growth was also increased. All of the tested samples demonstrated the inhibitory effects in a concentration-dependent manner. The MIC data of the examined compounds were presented in Table 2  between 1.20 and 5.00 mg/mL against E. coli (BCRC-11634). Among the tested compounds, triterpenoids betulinic acid (6) and -taraxasterol (10) exhibited the most significant inhibition against E. coli with MIC values of 1.20 mg/mL. These principles should be responsible for the bioactivity of the chloroform fraction (FSC). The results exhibited that the triterpenoids from the methanol extracts of fruits of F. suspensa possessed antibacterial activities against the common bacteria. It also provided evidence for the traditional uses of the fruits of F. suspensa as herbal medicines in the treatment of bacterial diseases. Although these purified compounds did not display better inhibition of the bacterial growth compared with the reported synthetic antibiotics, the extracts and principles from the natural sources usually possessed lower toxicity. Further structural modification could be performed to improve the activity and maintain the safety of these compounds. Therefore, it would be potentially useful in developing new antimicrobial therapeutic agents.