Article pubs.acs.org/JAFC
Evaluating the Antibacterial Properties of Polyacetylene and Glucosinolate Compounds with Further Identification of Their Presence within Various Carrot (Daucus carota) and Broccoli (Brassica oleracea) Cultivars Using High-Performance Liquid Chromatography with a Diode Array Detector and Ultra Performance Liquid Chromatography−Tandem Mass Spectrometry Analyses L. Hinds,† O. Kenny,*,†,‡ M. B. Hossain,† D. Walsh,§ E. Sheehy,∥ P. Evans,∥ M. Gaffney,† and D. K. Rai† †
Department of Food Biosciences and §Department of Food Safety, Teagasc Food Research Centre, Ashtown, Dublin D15 KN3K, Ireland ‡ Institute for Global Food Security, School of Biological Sciences, Queens University Belfast, Belfast BT9 5BN, United Kingdom ∥ Centre for Synthesis and Chemical Biology, School of Chemistry, University College Dublin, Dublin D04 V1W8, Ireland ABSTRACT: Ongoing consumer concerns over using synthetic additives in foods has strongly influenced efforts worldwide to source suitable natural alternatives. In this study, the antibacterial efficacy of polyacetylene and glucosinolate compounds was evaluated against both Gram positive and Gram negative bacterial strains. Falcarinol [minimum inhibitory concentration (MIC) = 18.8−37.6 μg/mL] demonstrated the best overall antibacterial activity, while sinigrin (MIC = 46.9−62.5 μg/mL) was the most active glucosinolate compound. High-performance liquid chromatography with a diode array detector analysis showed falcarinol [85.13−244.85 μg/g of dry weight (DW)] to be the most abundant polyacetylene within six of the eight carrot (Daucus carota) cultivars investigated. Meanwhile, sinigrin (100.2−244.3 μg/g of DW) was the most abundant glucosinolate present within the majority of broccoli (Brassica oleracea) cultivars investigated using ultra performance liquid chromatography−tandem mass spectrometry analysis. The high abundance of both falcarinol and sinigrin within these respective species suggests that they could serve as potential sources of natural antibacterial agents for use as such in food products. KEYWORDS: polyacetylene, glucosinolate, antibacterial activity, Brassica oleracea, Daucus carota, falcarinol, sinigrin
1. INTRODUCTION At present, the economic loss incurred as a result of bacterial food spoilage is of great concern in both developed and developing countries worldwide.1 In many instances, a limited shelf life can often affect the globalization of particular foods and food products.2 While efforts have been made to increase the shelf life of such food products using synthetic antibacterial agents [e.g., butylated hydroxyanisole (BHT) and butylated hydroxyanisole (BHA)], the introduction of natural antibacterial agents from plant extracts has proven popular among consumers.3,4 In keeping with this rising trend, the onus has been leveled at the food industry and researchers alike to investigate the potential of natural product compounds as suitable antibacterial agents for inclusion in food products to further extend their shelf life. Among the myriad of bioactive properties associated with plants and plant extracts, detailed knowledge regarding their antibacterial efficacy, specifically their minimum inhibitory concentration (MIC), is often lacking.5 For example, polyacetylenes (Figure 1), which are compounds generally found in Apiaceae species, including carrot (Daucus carota), are better known for their purported anticancer6,7 and antiinflammatory8,9 activities, while information regarding their antibacterial potential is somewhat limited. Similarly, glucosinolates (Figure 1), which are synonymous with many © XXXX American Chemical Society
Brassicaceae species, such as broccoli (Brassica oleracea), are precursors to potent bioactive phytochemicals known as isothiocyanates. Although isothiocyanates are regarded as having strong anticancer and antibacterial properties,10 less is known about the antibacterial properties associated with glucosinolates themselves. To this end, there is a duty to further investigate the possible antibacterial capabilities of these compounds with a view to exploiting them as natural functional ingredients to prolong shelf life within food products. The purpose of the present study was to determine the MIC for a number of polyacetylene and glucosinolate compounds against common foodborne pathogens as well as methicillinresistant Staphylococcus aureus (MRSA). Furthermore, quantitative high-performance liquid chromatography with a diode array detector (HPLC−DAD) (polyacetylene) and ultra performance liquid chromatography−tandem mass spectrometry (UPLC−MS/MS) (glucosinolate) analyses of these compounds were carried out using various carrot and broccoli cultivars, respectively, to determine a viable source for active compounds. Received: May 4, 2017 Revised: July 27, 2017 Accepted: July 31, 2017
A
DOI: 10.1021/acs.jafc.7b02029 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Figure 1. Chemical structures of the polyacetylene and glucosinolate compounds investigated for their antibacterial activity.
Table 1. Identification of Glucosinolate Compounds Using UPLC−MS/MS compound
a
retention time (min)
molecular weight (Da/mol)
MRM transitions (m/z)
cone voltage (V)
collision energy (eV)
→ → → → → → → → → → → →
40 40 40 40 40 40 40 40 40 40 40 40
16 16 14 14 15 15 15 15 17 17 18 18
glucoiberin
0.88
422.18
singrin
1.22
358.02
gluconapin
1.24
371.98
progoitrin
1.50
387.78
glucotropaeolin
3.37
408.10
neoglucobrassicin
3.98
478.50
421.18 421.18 357.02 357.02 370.98 370.98 386.78 386.78 407.10 407.10 477.45 477.45
a
96.92 357.89 97.04a 195.10 96.92a 195.16 96.85a 195.10 98.98a 195.10 96.92a 446.45
MS/MS MRMs used for the quantification of each compound. sample was decanted, while the pellet was resuspended in fresh ethyl acetate (1 mL) and re-extracted twice more. For each sample, the supernatants were pooled and subsequently dried under nitrogen. 2.4. HPLC Analysis of Polyacetylene Extracts. Reverse-phase high-performance liquid chromatography (RP-HPLC) of polyacetylene extracts from carrot varieties was carried out using an Agilent 1100 Series HPLC system with a diode array detector (DAD) (Agilent Technologies, Stuttgart, Germany). Polyacetylene extracts were dissolved in acetonitrile (150 μL), centrifuged at 10 000 rpm, and syringe-filtered using 0.22 μm polyvinylidene diflouride (PVDF) filters. Separation was achieved using a 5 μm Zorbax RX-C18 column (4.6 × 12.5 mm), where the injection volume of sample for each run was 20 μL and the flow rate was 1 mL/min. The total run time for each sample was 20 min. The starting condition for each run was 30:70 mobile phase A (water)/mobile phase B (acetonitrile). This ratio of A/B was kept isocratic for 11 min before being ramped to 15:85 (A/ B) over 1 min, followed by a final ramp to 2:98 (A/B) and held for 5 min, before re-equilibration at 18 min to initial starting conditions of 30:70 (A/B) and held for 2 min. 2.5. Glucosinolate Extraction from Broccoli Varieties. Freezedried broccoli powders (100 mg) were extracted using 1 mL of boiling aqueous methanol (70%, v/v). Samples were shaken for 10 min, followed by centrifugation for 2 min at 13 000 rpm. After the supernatant was decanted for each sample, the pellet was resuspended
2. MATERIALS AND METHODS 2.1. Chemicals and Reagents. Müller−Hinton broth (MHB) nutrient agar and the HPLC-grade solvents, acetonitrile, methanol, ethyl acetate, water, and acetone, were purchased from Fisher Scientific, Ltd. (Dublin, Ireland). Dimethyl sulfoxide (DMSO), formic acid (MS grade), and iodonitrotetrazolium chloride (INT) were obtained from Sigma-Aldrich, Ltd. (Wicklow, Ireland). Glucosinolate standards, glucoiberin, sinigrin, glucotropaeolin glucobrassicin, phenethyl glucosinolate, and neoglucobrassicin, were purchased from ChromaDex (Irvine, CA, U.S.A.). Polyacetylenes, (S)-falcarinol, (R)acetoxyfalcarinol, and (S)-panaxjapyne, were prepared as described by McLaughlin et al.11 2.2. Plant Material and Preparation. Eight carrot (D. carota) and six broccoli (B. oleracea) varieties were grown locally in Dublin, Ireland (Table 1). After harvesting, cultivars of carrot and broccoli were freeze-dried (A12/60 freeze-dryer, Frozen in Time, Ltd., York, U.K.) and blended (Robot Coupe-Blixer 2, Isleworth, U.K.) to a fine powder. Powders were individually vacuum-packed and stored at −80 °C prior to extraction. 2.3. Polyacetylene Extraction from Carrot Varieties. Freezedried carrot powders (100 mg) were extracted using ethyl acetate (1 mL). Samples were mixed for 2 h at 150 rpm using a MaxQ 6000 shaker (Thermo Scientific, Waltham, MA, U.S.A.), followed by centrifugation for 5 min at 10 000 rpm. The supernatant of each B
DOI: 10.1021/acs.jafc.7b02029 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry in fresh solvent (1 mL) and re-extracted twice more. For each sample, the supernatants were pooled and subsequently dried under nitrogen. 2.6. UPLC−MS/MS Analysis of Glucosinolate Extracts. The quantification of glucosinolates present in aqueous methanol extracts from six broccoli varieties was carried out using UPLC coupled to a tandem quadrupole detector (TQD) mass spectrometer (Waters Acquity, Waters Corporation, Milford, MA, U.S.A.). The dried extracts were reconstituted in methanol (1 mg/mL) prior to UPLC−MS/MS analysis. Separation was achieved using an AQUITY UPLC BEH C18 column (2.1 × 50 mm, 1.7 μm), where the injection volume of each sample was 5 μL and the flow rate was 0.1 mL/min. The total run time for each sample was 7.5 min. The starting condition for each run was 100:0 mobile phase A (water + 0.1% formic acid)/mobile phase B (methanol + 0.1% formic acid) and held for 4 min. The gradient was then ramped to 85:15 (A/B) over 1 min and held for 0.5 min. This was followed by another ramp to 10:90 over 1 min, before reconditioning for 1 min to the starting conditions. The mass spectrometer was operated in negative electrospray ionization (ESI) mode. The source temperature was set at 120 °C, and the desolvation temperature was set to 350 °C. Capillary voltage was set at 2.8 kV, while the cone voltage and collision energy were optimized for each of the compounds using IntelliStart software (Waters Corp., Milford, MA, U.S.A.) during setup of the multiple reaction monitoring (MRM) parameters (Table 1) Nitrogen gas was used as both sheath gas and auxiliary gas (800 and 50 L/h), respectively. 2.7. Bacterial Strains and Culture Conditions. The antibacterial activity of glucosinolate and C17 polyacetylene compounds was measured against the following strains of bacteria: Bacillus cereus NCTC 7464, Escherichia coli DSM 1103, MRSA clinical isolate (Causeway Hospital, Coleraine, Ireland), Salmonella typhimurium SARB 69, and S. aureus NCTC 8178. All strains were stored on ceramic beads in glycerol (−80 °C) prior to use. One bead of each strain was then streaked on a nutrient agar plate and incubated for 18 h at 37 °C. Subsequently, one colony was isolated and inoculated into sterile MHB (25 mL) and then incubated for 22 h at 37 °C. Cultures were then vortexed, and aliquots were diluted appropriately in sterile MHB (log 6.0 ± 0.5 cells/mL). Cell numbers were confirmed by plate counting. 2.8. MIC Assay. The MIC of each compound was measured using a microdilution method as described by Kenny et al.12 The 96-well plate contained the following controls and blanks: 100 μL of MHB (blank), 50 μL of gentamicin (0.2 mg/mL) (positive control), 50 μL of aqueous acetone (10%, v/v) solution (negative control), and 50 μL of sterile water (4% DMSO) (negative control). Compounds (0.4 mg/ mL) were prepared by dissolving in sterile water containing either 4% DMSO (glucosinolates) or 10% acetone (polyacetylenes). A serial dilution of each compound was carried out across corresponding lanes of the 96-well plate, where the final volume of each well was 50 μL. All wells, with the exception of the MHB blank wells, were inoculated with 50 μL of bacteria, where the final volume of each well was 100 μL. This procedure was carried out for all bacterial strains. Each plate was incubated at 37 °C for 24 h using a Stuart SI50 orbital incubator (Rhys International, Ltd., Bolton, U.K.) at 30 rpm. After incubation, 40 μL of INT was added to each well. Plates were further incubated at 37 °C for 1 h. The MIC of each sample against a bacterial strain was determined as the lowest sample concentration at which no pink color appeared.13 This process was repeated in triplicate on separate days for each bacterial strain. 2.9. Statistical Analysis. HPLC and UPLC−MS/MS samples of carrot and broccoli cultivars were analyzed using one-way analysis of variance (ANOVA) tests, followed by Tukey comparison, using GraphPad Prism, version 5.01.
Table 2. Quantitative HPLC−DAD Analysis of Polyacetylene Abundancies within Eight Cultivars of Carrot (D. carota), with Values Expressed as Micrograms of Compound per Gram of DW Extract variety
falcarinol (μg/g of DW)
falcarinol-3-acetate (μg/g of DW)
falcarindiol (μg/g of DW)
Flyaway Nairobi Narbonne Purple Haze Resistafly St. John’s Purple White Belgian White Satin
114.64 152.65 116.59 85.13 120.71 244.85 49.39 92.39
8.54 7.91 7.38 n/d 21.09 19.78 8.35 8.77
93.64 128.89 62.79 18.65 17.55 100.53 225.13 203.30
of falcarinol (49.39−244.85 μg/g) and falcarindiol (18.65− 225.13 μg/g) varied across all samples analyzed, falcarindiol-3acetate was present at considerably lower levels (7.38−21.09 μg/g) and was not detected within the Purple Haze variety. Falcarinol was the most abundant polyacetylene found in varieties of Resistafly (120.71 μg/g), Flyaway (114.64 μg/g), Nairobi (152.65 μg/g), Nabonne (116.59 μg/g), St. John’s Purple (244.85 μg/g), and Purple Haze (85.13 μg/g), while falcarindiol was the most abundant in White Belgian (225.13 μg/g) and White Satin (203.30 μg/g). Previous studies have sought to investigate the effects of various processing techniques and storage conditions on the polyacetylene content within carrot varieties.17−19 Incidentally, the polyacetylene content of fresh and non-processed carrots has also been reported by these studies. For example, Rawson et al.18 reported higher levels of falcarinol [449.8 μg/g of dry weight (DW)] compared to falcarindiol (245.2 μg/g of DW) within the D. carota L. cultivar. Similarly, Aguiló-Aguayo et al.19 reported higher levels of falcarinol (25.04 μg/g of DW) in fresh carrot juice (D. carota L.) when compared to falcarindiol (13.24 μg/g of DW) and falcarindiol-3-acetate (16.01 μg/g of DW), respectively. 3.2. UPLC−MS/MS Analysis of Glucosinolate Extracts. Aqueous methanol (70%) extracts from six varieties of B. oleracea var. italica were investigated for the presence of glucosinolate compounds, namely, glucoiberin, gluconapin, glucotropaeolin, progoitrin, and sinigrin, using UPLC−MS/ MS analysis (Table 3). Glucoiberin (291.0−761.4 μg/g) was the most abundant compound found within these extracts, excluding the Marathon variety, where high levels of neoglucobrassicin (178.6 μg/g) were present. In relation to sinigrin, appreciable levels of this compound were observed in Cardinal (100.2 μg/g), Green Magic (106.9 μg/g), Pantheon (222.1 μg/g), TZ6002 (233.6 μg/g), and TZ5055 (244.3 μg/g) broccoli varieties. A study by Ares et al.22 carried out quantitative UPLC−MS analysis on three broccoli varieties, including Parthenon, for the presence 15 glucosinolte compounds. In relation to the Parthenon variety, Ares et al.22 detected 6 glucosinolates, including glucoiberin (130−173 μg/g), glucobrassicin (85−102 μg/g), and neoglucobrassicin (19−29 μg/g). While the present study showed sinigrin to be antibacterial and present in appreciable amounts within the Pantheon variety (222.1 μg/g), this compound was not detected by Ares et al.22 in any of the broccoli varieties under investigation. These results suggest that sinigrin may be exclusively found within broccoli florets, used in the present study, rather than broccoli leaves. This is further
3. RESULTS AND DISCUSSION 3.1. HPLC−DAD Analysis of Polyacetylene Extracts. Qualitative and quantitative analyses of three polyacetylene compounds (falcarinol, falcarindiol, and falcarindiol-3-acetate) from ethyl acetate extracts of eight carrot varieties were carried out using HPLC−DAD analysis (Table 2). Although the levels C
DOI: 10.1021/acs.jafc.7b02029 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Table 3. Quantitative UPLC−MS/MS Analysis of Glucosinolate Abundancies within Five Cultivars of Broccoli (B. oleracea), with Values Expressed as Micrograms of Compound per Gram of DW Extract variety
glucobrassicin (μg/g of DW)
glucoiberin (μg/g of DW)
neoglucobrassicin (μg/g of DW)
sinigrin (μg/g of DW)
Cardinal Green Magic Marathon Pantheon TZ5055 TZ6002
15.9 13.4 49.9 34.9 20.6 40.4
398.2 291.0 39.3 372.1 761.4 502.0
8.4 2.1 178.6 7.1 1.2 8.4
100.2 106.9 1.9 222.1 244.3 233.6
supported by a number of previous studies, which were also unable to detect sinigrin in leaf and stalk extracts of various broccoli varieties.23−25 3.3. Antibacterial Activity of Polyacetylene Extracts. The antibacterial activity of polyacetylenes (falcarinol, acetoxyfalcarinol, and panaxjapyne) was evaluated against three Gram positive (S. aureus, MRSA and B. cereus) and two Gram negative (E. coli and S. typhimurium) bacterial strains, where activity was measured as the MIC against bacterial growth in each case (Table 4).
Although, in the present study, falcarinol demonstrated an antibacterial effect against Gram positive bacteria, there was no effect seen against Gram negative E. coli. The disparity between the present study and that of Kobaisy et al.14 may be due to the use of different E. coli strains. Schinkovitz et al.15 investigated the antimycobacterial activity of polyacetylenes, isolated from a dichloromethane extract of Levisticum officinale L. roots, against strains of Mycobacterium fortuitum and M. avium. Falcarinol exhibited an antimycobacterial effect against M. fortuitum (MIC = 30.4 μM) and M. avium (MIC = 60.8 μM), while falcarindiol was active against both strains at MICs as low as 16.4 μM, respectively. Schinkovitz et al.15 also measured the antimycobacterial effect of dehydrofalcarindiol, previously isolated from a chloroform extract of Artemisia monosperma, against both aforementioned strains.16 However, this compound was deemed inactive at the highest concentrations (128 μg/mL) tested. Schinkovitz et al.15 suggested that the lack of activity associated with dehydrofalcarindiol in comparison to falcarinol and flacarindiol and may be due to structural differences between these related polyacetylene compounds, in particular, the absence of a terminal methyl group in dehydrofalcarindiol. While the polyacetylenes in the present study all contain a terminal methyl group, panaxjapyne does not contain a double bond in the 1C position, which may be responsible for the lack of activity seen against these Gram positive strains. 3.4. Antibacterial Activity of Glucosinolate Extracts. A total of five glucosinolate compounds (glucoiberin, gluconapin, phenethyl, progoitrin, and sinigrin) were evaluated for their antibacterial activity (Table 4). Sinigrin was the only compound to exhibit an antibacterial effect against the Gram positive strains, S. aureus (MIC = 46.9 μg/mL), MRSA (MIC = 46.9 μg/mL), and B. cereus (MIC = 62.5 μg/mL), while all five compounds demonstrated no activity against both Gram negative strains tested. Although isothiocyanates, the hydrolyzed products of glucosinolates, have previously been shown to exhibit an antibacterial effect,10 there are limited studies available detailing the efficacy of glucosinolates. Aries et al.20 measured the antibacterial effect of three glucosinolate compounds (sinigrin, glucotropaeolin, and gluconasturtiin) using a disc diffusion method. All three compounds proved ineffective at preventing bacterial growth when measured against a range of Gram positive and Gram negative bacteria, which included strains of S. aureus and E. coli. Additionally, Aries et al.20 measured the antibacterial efficacy of numerous isothiocynates against the same bacterial strains, with zones of inhibition ranging from approximately 9 to 60 mm for Gram positive and from approximately 7 to 70 mm for Gram negative bacteria at concentration doses of 3 μM. While Aries et al.20 suggest that only glucosinolate derivatives are antibacterial-active, the results of the current study have shown sinigrin to be active against
Table 4. Antibacterial Activies of Polyacetylene and Glucosinolate Compounds, Represented as MICs, against Strains of S. aureus, MRSA, and B. cereus, with Values as Means of Replicate Assays (n = 3) MIC (μg/mL)n compound (R)-acetoxyfalcarinol (S)-falcarinol (S)-panaxjapyne
S. aureus NCTC 8178 Polyacetylenes 50 18.8
MRSA clinical
B. cereus NCTC 7464
50 37.6
50 12.5
46.9
62.5
Glucosinolates glucoiberin gluconapin glucotropaeolin progoitrin sinigrin
46.9
Overall, both falcarinol and acetoxyfalcarinol demonstrated strong antibacterial activity against Gram positive bacteria at relatively low concentrations (MIC = 12.5−50 μg/mL), while panaxjapyne was shown to be inactive against all three strains. Falcarinol demonstrated the best antibacterial activity against all Gram positive bacteria when compared to acetoxyfalcarinol. The marked higher potency of falcarinol in comparison to acetoxyfalcarinol suggests this polyacetylene compound to be a more likely antibacterial target. A study by Kobaisy et al.14 investigated the antibacterial activity of falcarinol and falcarindiol, isolated from Oplopanax horridus, against a number of Gram positive (Bacillus subtilis and MRSA), Gram negative (E. coli and Pseudomonas aeruginosa), and mycobacterial (Mycobacterium tuberculosis and Mycobacterium avium) strains as well as the yeast Candida albicans. The results showed falcarinol to be more active against all Gram negative and positive strains (MIC = 3.1−50 μg/mL) than falcarindiol (MIC = 6.25−50 μg/mL) and similarly against C. albicans (MIC = 6.25 and 25.0 μg/mL). Meanwhile, results of a disk diffusion assay showed both compounds to exhibit large zones of clearing (
