Validation of a commercial receptor kit Sulfasensor® Honey for the screening of sulfonamides in honey according to Commission Decision 2002/657/EC

The Sulfasensor® Honey kit is a receptor test dedicated to the screening of sulphonamide residues respectively in different matrices. The aim of this project was to evaluate and validate this kit according to the Community Reference Laboratory (CRL) guideline for the validation of screening methods to achieve the French control plan for honey. The test is robust, quick (90?min for 40 samples), easy to perform and easy to read. The false-positive rate was estimated to be 12.5%. The detection capabilities CCβ of the kit were lower than or equal to 25?µg?kg^?1 for sulfamethazine, sulfamerazine, sulfathiazole and sulfapyridine, and between 25 and 50?µg?kg^?1 for sulfadiazine and sulfadimethoxine, 150?µg?kg^?1 for sulfaquinoxaline, and 1000?µg?kg^?1 for sulfamethoxazole and sulfamethizole. Sulfanilamide was not detected by the kit. The kit was applicable to a wide variety of honeys (different floral and geographical origins, liquid or solid). This kit was used to implement the French control plan for the detection of antibiotic residues in honey in 2010 in parallel with an HPLC method. However, in 2011 the kit was replaced by an LC-MS/MS method for the screening and confirmation of sulfonamide residues in honey, which detects all the sulfonamides of interest.     

Evaluation and validation of two microbiological tests for screening antibiotic residues in honey according to the European guideline for the validation of screening methods

Two microbiological kits based on Bacillus stearothermophilus (Eclipse 50® and Premi®Test) have been evaluated and validated according to the European guideline for the validation of screening methods (January 2010) and in relation to the concentrations recommended by the EU-RL in 2007. Both tests are robust, a fast method and easy to implement. Both tests are applicable to a very large variety of honeys from different floral and geographical origins (rosemary, lavender, scrub, heath, alder, forest, lemon, acacia, chestnut, raspberry, mountain and flowers) as well as honey of different colours (from blank honey to brown honey, including yellow and orange honey). A satisfactory false-positive rate of 5% was obtained for the Eclipse 50® test. The observed detection capabilities CCβ of the Eclipse 50® kit were: chlortetracycline (>75?µg?kg^?1), oxytetracycline (≤200?µg?kg^?1), tetracycline (>100?µg?kg^?1), cloxacillin (≤40?µg?kg^?1), tylosin (≤200?µg?kg^?1), desmycosin (>400?µg?kg^?1), sulfadiazine (≤300?µg?kg^?1), sulfadimethoxine (≤250?µg?kg^?1), sulfamerazine (>300?µg?kg^?1), sulfamethazine (>1000?µg?kg^?1), sulfamethizole (>75?µg?kg^?1), sulfamethoxazole (≤25?µg?kg^?1), sulfanilamide (?1000?µg?kg^?1), sulfaquinoxaline (>75?µg?kg^?1), sulfathiazole (≤250?µg?kg^?1) and lincomycin (>1500?µg?kg^?1). These levels were all higher than the recommended concentrations where they exist. Due to its lack of sensitivity, it cannot be recommended for reliable routine use. The observed CCβ of the Premi®Test kit were: chlortetracycline (10?µg?kg^?1), oxytetracycline (>10?µg?kg^?1), tetracycline (≤10?µg?kg^?1), cloxacillin (≤5?µg?kg^?1), tylosin (≤10?µg?kg^?1), desmycosin (≤15?µg?kg^?1), sulfadiazine (≤25?µg?kg^?1), sulfadimethoxine (≤25?µg?kg^?1), sulfamerazine (≤25?µg?kg^?1), sulfamethazine (≤25?µg?kg^?1), sulfamethizole (≤25?µg?kg^?1), sulfamethoxazole (≤10?µg?kg^?1), sulfanilamide (≤25?µg?kg^?1), sulfaquinoxaline (≤10?µg?kg^?1), sulfathiazole (25?µg?kg^?1) and lincomycin (≤25?µg?kg^?1). The Premi®Test kit could be recommended for reliable use in routine control due to its low detection capabilities (except for aminoglycosides), but the disadvantage is a high false-positive rate of 14%.     

Validation of a liquid chromatography-tandem mass spectrometry method for the simultaneous determination of sulfonamides,trimethoprim and dapsone in muscle,egg, milk and honey

A quantitative multi-residue method that includes 13 sulfonamides, trimethoprim and dapsone was developed and validated according to Commission Decision 2002/657/EC for muscle, milk egg and honey samples. For all matrices, the same extraction procedure was used. Samples were extracted with an acetone/dichloromethane mixture and cleaned up on aromatic sulfonic acid (SO₃H) SPE cartridges. After elution and concentration steps, analytes were identified and quantified by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Data were acquired according to the multiple reaction-monitoring approach (MRM) and analytes were quantified both by the isotope dilution and the matrix-matched approaches calculating the response factors for the scanned product ions. The developed method shows good linearity, specificity, precision (repeatability and within-laboratory reproducibility), and trueness. Estimated CCβ for sulfonamides ranged between 5.6 and 8.2 µg kg^?1 for eggs, between 11.1 and 69.9 µg kg^?1 for milk, between 64.7 and 87.9 µg kg^?1 for muscle, and between 2.7 and 5.3 µg kg^?1 for honey. CCβ values for dapsone were 3.1, 0.6, 0.7 and 1.5 µg kg^?1 and for trimethoprim were 3.1, 6.7, 81.7 and 3.0 µg kg^?1 calculated for eggs, milk, muscle and honey, respectively. Recovery for all matrices was in the range from 89.1% and 109.7%. In matrix effect testing, no significant deviations were found between different samples of muscle and milk; however, a matrix effect was observed when testing different types of honey. The validation results demonstrate that the method is suitable for routine veterinary drug analysis and confirmation of suspect samples.     

Oxytetracycline,tetracycline, chlortetracycline and doxycycline in pasteurised cow’s milk commercialised in Brazil

The aim of this study was to investigate the presence of tetracycline residues in pasteurised cow’s milk using high-performance liquid chromatography coupled with UV/VIS detection to determine the exposure of Brazilian’s population to antibiotic residues. One hundred samples collected from the State of Paraná, Brazil, were analysed. Three of these samples were contaminated at the following concentrations: 121.8 µg·kg^?1 for oxytetracycline, 93.5 µg·kg^?1 for tetracycline and 134.6 µg·kg^?1 for chlortetracycline (61.6 µg·kg^?1) and doxycycline (73.0 µg·kg^?1). The median tetracycline residue concentration found in the samples was 42.3 µg·kg^?1, and the estimated daily intake (EDI) was 0.05 µg Kg^?1 bw day^?1 in Brazil. These results demonstrate that the occurrence of tetracycline in Brazilian milk was low (3%) and only for 2% above the maximum residue limit, so the risk to the population from the presence of these residues in milk was low (<1% of the acceptable daily intake).     

Rapid multi-class multi-residue method for the confirmation of chloramphenicol and eleven nitroimidazoles in milk and honey by liquid chromatography-tandem mass spectrometry (LC-MS)

A confirmatory method was developed to allow for the analysis of eleven nitroimidazoles and also chloramphenicol in milk and honey samples. These compounds are classified as A6 compounds in Annex IV of Council Regulation 2377/90 (European Commission 1990) and therefore prohibited for use in animal husbandry. Milk samples were extracted by acetonitrile with the addition of NaCl; honey samples were first dissolved in water before a similar extraction. Honey extracts underwent a hexane wash to remove impurities. Both milk and honey extracts were evaporated to dryness and reconstituted in initial mobile phase. These were then injected onto a liquid chromatography-tandem mass spectrometry (LC-MS/MS) system and analysed in less than 9 min. The MS/MS was operated in multiple reaction monitoring (MRM) mode with positive and negative electrospray ionization. The method was validated in accordance with Commission Decision 2002/657/EC and is capable of analysing metronidazole, dimetridazole, ronidazole, ipronidazole and there hydroxy metabolites hydroxymetronidazole, 2-hydroxymethyl-1-methyl-5-nitroimidazole, and hydroxyipronidazole. The method can also analyse for carnidazole, ornidazole, ternidazole, tinidazole, and chloramphenicol. A recommended level of 3 µg l^?1/µg kg^?1 for methods for metronidazole, dimetridazole, and ronidazole has been recommended by the Community Reference Laboratory (CRL) responsible for this substance group, and this method can easily detect all nitroimidazoles at this level. A minimum required performance level of 0.3 µg l^?1/µg kg^?1 is in place for chloramphenicol which the method can also easily detect. For nitroimidazoles, the decision limits (CCα) and detection capabilities (CCβ) ranged from 0.41 to 1.55 µg l^?1 and from 0.70 to 2.64 µg l^?1, respectively, in milk; and from 0.38 to 1.16 µg kg^?1 and from 0.66 to 1.98 µg kg^?1, respectively, in honey. For chloramphenicol, the values are 0.07 and 0.11 µg l^?1 in milk and 0.08 and 0.13 µg kg^?1 in honey. Validation criteria of accuracy, precision, repeatability, and reproducibility along with measurement uncertainty were calculated for all analytes in both matrices.     

Removing τ‐fluvalinate residues from press‐extracted honey

Honey is obtained from the hives of European honeybees. Honeybees can be infested by Varroa parasitism, which affects honey production. The acaricide τ‐fluvalinate used in beehives to control Varroa mites, can leave its residues in honey, pollen and wax. A monitoring study was undertaken to determine the τ‐fluvalinate residues in honey samples taken from treated hives. The τ‐fluvalinate residues were determined by gas chromatography using a short, non‐polar packed column with electron capture detection. Analysis of 50 samples showed an average residue level of 15 µg kg⁻¹. The same samples were filtered through Whatman filter paper to separate the honey from residual suspended particles. The filtered honeys showed no acaricide residues above the detection limit of 3 µg kg ⁻¹, while the average residue level determined in the sediments was 392 µg kg ⁻¹. Thus the residues of τ‐fluvalinate in honey can be eliminated by separating the honey from residual suspended particles that contain acaricide residues. Copyright © 2003 Society of Chemical Industry     

Determination of Antibiotic Residues in Honey by High-Performance Liquid Chromatography with Electronspray Ionization Tandem Mass Spectrometry

The aim of the present study was to develop a rapid and simple method for the detection and quantification of antibiotic and antibacterial residues in honey using liquid chromatography with electronspray ionization tandem mass spectrometry. Two different extraction methods were used. The first method uses water and 1% formic acid in acetonitrile for the determination of sulfonamides while the second uses phosphate buffer, 10% trichloroacetic acid, and acetonitrile as the extracting solvent for the determination of tetracyclines, amphenicols, fluoroquinolones, penicillin g, trimethoprim, and tiamulin. The multi-residue method was validated in a thyme honey matrix. Thirty-six different antibiotics and residues from four different families (sulfonamides, tetracyclines, amphenicols, fluoroquinolones) and some individual antibiotics (penicillin g, trimethoprim, and tiamulin) were tested in 20 honey samples originating from Cyprus and Greece. The decision limits (CCα) were from 0.1 to 9.2 μg kg^?1; the detection capabilities (CCβ) were from 0.3 to 27.6 μg kg^?1 while recoveries were from to be between 65.0 and 116.1%. The method was successfully applied to commercial samples from different types of honey from Greece and Cyprus. Among them, oxolonic acid, sulfathiazole, and sulfadimethoxine were found in three honey samples. Finally, proficiency testing was applied to the proposed method while analysis of certified samples showed good method performance characteristics.     

Application of a luminescent bacterial biosensor for the detection of tetracyclines in routine analysis of poultry muscle samples

Tetracyclines are extensively used in veterinary medicine. For the detection of tetracycline residues in animal products, a broad array of methods is available. Luminescent bacterial biosensors represent an attractive inexpensive, simple and fast method for screening large numbers of samples. A previously developed cell-biosensor method was subjected to an evaluation study using over 300 routine poultry samples and the results were compared with a microbial inhibition test. The cell-biosensor assay yielded many more suspect samples, 10.2% versus 2% with the inhibition test, which all could be confirmed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Only one sample contained a concentration above the maximum residue limit (MRL) of 100 µg kg^?1, while residue levels in most of the suspect samples were very low (<10 µg kg^?1). The method appeared to be specific and robust. Using an experimental set-up comprising the analysis of a series of three sample dilutions allowed an appropriate cut-off for confirmatory analysis, limiting the number of samples and requiring further analysis to a minimum.     

Determination and surveillance of nine acaricides and one metabolite in honey by liquid chromatography-tandem mass spectrometry

A simple and accurate analytical method for the determination of acaricides in honey was developed and validated in accordance with Japanese validation guidelines. Analytes – amitraz, N-2,4-dimethylphenyl-N-methylformamidine (DMPF), etoxazole, fenpyroximate, fipronil, hexythiazox, propargite, pyridaben and spirodiclofen – were extracted with ethyl acetate under basic conditions and subsequently cleaned up using an InertSep^® MA-1 polymer-based anion-exchange column. The method was validated by fortified recovery tests at three different concentrations (1, 5 and 10 µg kg^?1) performed with three samples daily on five different days. The method exhibited recoveries of 77–116% and precision (relative standard deviations – RSDs) of repeatability and within-laboratory reproducibility ranged from 2% to 22% and from 3% to 23%, respectively. The sample solution was successfully cleaned up to enable quantification using external solvent calibration curves. The limits of quantification (LOQs) were estimated to be 1 µg kg^?1 for all analytes. The method was applied to honey samples commercially available in Tokyo, Japan. Analysis of 250 honey samples indicated that amitraz was present in 127 samples, and that its residual concentration was less than 20 µg kg^?1. Propargite was detected in 23 samples at concentrations less than 1 µg kg^?1.     

Feed additives diclazuril and nicarbazin in egg and liver samples from Croatian farms

In total 307 egg and 275 liver samples were examined for nicarbazin and 365 eggs for diclazuril over a 30-month period. Enzyme-linked immunosorbent assay methods used for quantification were validated according to European Commission Decision 2002/657/EC. Non-compliant samples were confirmed by LC-MS/MS. Mean diclazuril concentrations in egg samples were 0.31?µg?kg^?1, which is below the MRL. In only one egg sample, 2.26?µg?kg^?1 was determined by enzyme-linked immunosorbent assay, although confirmation by LC-MS/MS gave a value of 1.6?µg?kg^?1. Mean nicarbazin levels determined were 1.85?µg?kg^?1 in egg and 21.1?µg?kg^?1 in liver samples. Four samples, one egg and three livers, yielded elevated concentrations of nicarbazin, but only in the egg sample the LC-MS/MS method confirmed nicarbazin (106?µg?kg^?1) above the MRL value.     

Solid-phase microextraction/gas-chromatographic/mass spectrometric analysis of p -dichlorobenzene and naphthalene in honey

Protection of honeycombs from the Wax moth, Galleria mellonella, involves the use of physical, biological or chemical control methods. As chemical control may result in residues in the extracted honey, the presence of p-dichlorobenzene and naphthalene residues was investigated by solid-phase microextraction (SPME) coupled to gas-chromatographic/mass spectrometry (GC/MS). The method was linear between 5 and 200 µg kg^–1 honey for p-dichlorobenzene and 1 and 200 µg kg^–1 for naphthalene. Limits of detection were 1 and 0.1 µg kg^–1, respectively, for p-dichlorobenzene and naphthalene, while relative standard deviations were 2.6 and 7.9%, respectively. Application of the method to 90 unifloral Greek honeys revealed that, in 25.6% of the samples, the concentration of either one of the pesticides exceeded the maximum residue level (MRL). Maximum concentrations were 163.03 µg kg^–1 honey for p-dichlorobenzene and 193.74 µg kg^–1 honey for naphthalene. Naphthalene was found in traceable amounts in 78.9% of the samples, but only 5.6% of them contained concentrations above the MRL, which indicates the use of pre-contaminated honeycomb foundations or built combs. Nevertheless, because naphthalene is naturally present in some plant species growing in Greece, the contribution of nectar from such a floral source should not be overlooked.     

Survey of the anticoccidial feed additive nicarbazin (as dinitrocarbanilide residues) in poultry and eggs

A survey was carried out on the occurrence of dinitrocarbanilide (DNC), the marker residue for nicarbazin, in poultry produced in Ireland during 2002–2004. Liver (n?=?736) and breast muscle samples (n?=?342) were tested. DNC residues were found in 40 and 26% of liver and breast muscle samples at levels greater than 12.5 and 5?µg?kg^?1, respectively. DNC residues were found at >200?µg?kg^?1 in 12 and 0% of liver and muscle samples, respectively. Samples of breast muscle (n?=?217) imported from 11 countries were also tested for DNC residues. A lower incidence of DNC residues (6%) was found in imported breast muscle. Egg samples (n?=?546) were tested and DNC residues were found in nine samples, with levels ranging between 14 and 122?µg?kg^?1. Analysis of poultry, carried out as part of official food inspection in the period 2004–2006, indicated a reduction in the number of broiler liver samples containing DNC at >200?µg?kg^?1, to approximately 7%. Low levels of DNC residues continue to be found in <2% of egg samples.     

Doxycycline and sulfadimethoxine transfer from cross-contaminated feed to chicken tissues

During feed preparation at feed mills or during feed mixing in bins at farms, the accidental contamination of feed at trace levels by veterinary drug residues, commonly known as carry-over, can accidentally but frequently occur. To evaluate the concentrations of residual antimicrobials in poultry edible tissues, due to contaminated feed, sulfadimethoxine and doxycycline were administered for 10 days to chickens in poultry feed incurred at the contamination levels frequently found during national feed monitoring programmes (1–5?mg?kg^?1). Sulfadimethoxine and doxycycline residual concentrations detected in muscle (<LOD and 31?µg?kg^?1, respectively), liver (13 and 56?µg?kg^?1, respectively) and kidney (56 and 115?µg?kg^?1, respectively) were compared with their maximum residue limits (MRLs) fixed by EC 470/2009 and EU 37/2010 Regulations for a preliminary risk evaluation.     

Distribution of semduramicin in hen eggs and tissues after administration of cross-contaminated feed

Semduramicin is an ionophore coccidiostat used in the poultry industry as a feed additive. Cross-contamination of feeds for non-target animals with semduramicin is unavoidable. However, it is not known whether undesirable residues of semduramicin may occur in food after cross-contaminated feed is administered to animals. The aim of the work was to determine the levels of semduramicin in hen eggs (yolks and albumen) and tissues (liver, muscle, spleen, gizzard, ovarian yolks and ovaries) after administration of feed contaminated with 0.27 mg kg^?1 of this coccidiostat. The residues were determined using LC-MS/MS. The distribution pattern confirmed the high lipophilicity of semduramicin. Residues were found mainly in egg yolks (28.8 µg kg^?1), ovarian yolks (19.5 µg kg^?1) and liver (2.57 µg kg^?1), while hens’ muscle was free from semduramicin (LOD = 0.1 µg kg^?1). Among edible tissues, the maximum level (2 µg kg^?1) was exceeded only in the liver.     

Characterization of key aroma compounds in a representative aromatic extracts from citrus and astragalus honeys based on aroma extract dilution analyses

The aroma and aroma-active compounds of astragalus and citrus honeys were determined by GC-MS-olfactometry (GC-MS-O). The honey extracts were obtained by two different extraction techniques including liquid–liquid extraction (LLE) and solid phase extraction (SPE), and compared to determine the more representative odour aromatic honey extract. According to sensory analysis, the aromatic extract (40 g honey with 22 mL water) obtained by LLE was the most similar of honey odour. The amount of aroma compounds of citrus honey (14391.6 µg kg^?1) was higher than astragalus honey (4270.1 µg kg^?1). Terpenes and acids in citrus honey and esters and aldehydes in astragalus honey were the most dominant aroma compounds. Aroma extract dilution analysis (AEDA) was used to determine the aroma-active compounds of honey samples. By application of AEDA on the aromatic extracts isolated by LLE, 12 aroma-active compounds in astragalus honey and 23 in citrus honey could be detected within a flavour dilution (FD) factor range of 8–128. On the basis of the FD factor, phenylethyl alcohol in both honeys was the most powerful key compound and was described as the flowery odour.     

Investigation of pyrrolizidine alkaloids including their respective N-oxides in selected food products available in Hong Kong by liquid chromatography electrospray ionisation mass spectrometry

ABSTRACT

This study determined the levels of pyrrolizidine alkaloids (PAs), including their respective N-oxides, in foodstuffs available in Hong Kong by liquid chromatography-electrospray ionisation tandem mass spectrometry. A total of 234 samples (48 food items) were collected randomly from a local market and analysed. About 50% of samples were found to contain detectable amount of PAs. Amongst the 48 food items, PAs were not detected in 11 food items, including barley flour, beef, cattle liver, pork, pig liver, chicken meat, chicken liver, milk, non-fermented tea, Melissa tea and linden tea. For those found to contain detectable PAs, the summed PA content ranged up to 11,000 µg kg^?1. The highest sum of PA content among the 37 food items calculated with lower bound was cumin seed, then followed by oregano, tarragon and herbs de Provence with ranges of 2.5–11,000, 1.5–5100, 8.0–3300 and 18–1300 µg kg^?1 respectively. Among the samples, the highest sum of PA content was detected in a cumin seed sample (11,000 µg kg^?1), followed by an oregano (5100 µg kg^?1), a tarragon (3300 µg kg^?1) and a herbs de Provence (1300 µg kg^?1). In general, the results of this study agreed well with other published results in peer-reviewed journals, except that the total PAs in honey and specific tea infusion in this study were comparatively lower.     

Aflatoxin M1 contamination in cow and buffalo milk samples from the North West Frontier Province (NWFP) and Punjab provinces of Pakistan

A total of 178 milk samples (94 of buffalo and 84 of cow) were randomly taken from Punjab and the North West Frontier Province (NWFP) of Pakistan (n?=?89 in each province) and analysed for the presence of aflatoxin M1 (AFM₁) by HPLC-FLD. From Punjab about 46% of buffalo's and 49% of cow's milks were contaminated with AFM₁ as compared with 52% and 51% for milk samples from NWFP, respectively. Overall, the mean AFM₁ concentration was 0.046?µg?kg^?1 with a maximum of 0.350?µg?kg^?1. All samples complied with the Codex Alimentarius limit of 0.50?µg?kg^?1 for AFM₁ in milk, but 16.3% of samples exceeded the European Union maximum level of 0.05?µg?kg^?1. Another set of 415 buffalo's and cow's milk samples (213 morning milks and 202 evening milks) were analysed. Statistical analysis revealed significant differences (p?<?0.05) between mean AFM₁ concentrations in milk during the morning (0.043?µg?kg^?1) and the evening (0.028?µg?kg^?1) lactation times.     

Validation of two ELISA kits for the screening of tylosin and streptomycin in honey according to the European decision 2002/657/EC

Antibiotics are mixed with the food of bees to fight against diseases. No maximum residue limits have been set for honey. Recommended concentrations (RCs) have been published by European Union Reference Laboratories for tylosin and streptomycin. The objective of this project was to select and validate enzyme-linked immunosorbent assay (ELISA) kits for the screening of tylosin and streptomycin/dihydrostreptomycin residues to be implemented in the French honey control plan. Four ELISA kits for tylosin and five ELISA kits for streptomycin were evaluated. At the end, one kit each was selected and validated for tylosin (TECNA AB620) and streptomycin (Europroxima). Both ELISA kits for tylosin and streptomycin are specific, robust, fast and easy-to-use tests. The detection capability CCβ of tylosin A was less than or equal to 10?µg?kg^?1 (half the RC). The CCβ of desmycosin (the hydrolysed product of tylosin A in acidic conditions) is approximately 200?µg?kg^?1, which is five times the RC for tylosin (20?µg?kg^?1). Thus, this kit is fit for the screening of tylosin A but is unsuitable to detect desmycosin. The detection capability CCβ of streptomycin was less than or equal to 10?µg?kg^?1 (one fourth the RC). The cross-reactivity with dihydrostreptomycin was equal to 136%. Both ELISA kits were applicable to a wide variety of honey (single flower and multiflower, different floral origins, different geographic origins, different consistencies [liquid or solid] and different colours).     

Determination of pyrrolizidine alkaloids in tea,herbal drugs and honey

Honey was previously considered to be one of the main food sources of human pyrrolizidine alkaloid (PA) exposure in Europe. However, comprehensive analyses of honey and tea sampled in the Berlin retail market revealed unexpected high PA amounts in teas. This study comprised the analysis of 87 honey as well as 274 tea samples including black, green, rooibos, melissa, peppermint, chamomile, fennel, nettle, and mixed herbal tea or fruit tea. Total PA concentrations in tea ranged from < LOD to 5647 µg kg^?1, while a mean value of about 10 µg kg^?1 was found in honey samples. Additionally, herbal drugs were investigated to identify the source of PA in teas. Results suggest that PA in tea samples are most likely a contamination caused by co-harvesting of PA-producing plants. In some cases such as fennel, anise or caraway, it cannot be excluded that these plants are able to produce PA themselves.     

Characterisation of the phenolic and flavonoid fractions and antioxidant power of Italian honeys of different botanical origin

BACKGROUND: Twenty‐seven Italian honey samples of different floral origin were analysed for total phenolic and flavonoid contents by a spectrophotometric method and for antioxidant power and radical‐scavenging activity by the ferric‐reducing/antioxidant power (FRAP) and 1,1‐diphenyl‐2‐picrylhydrazyl (DPPH) assays respectively. In addition, the phenolic and flavonoid profiles were analysed using high‐performance liquid chromatography with UV detection (HPLC‐UV). RESULTS: The results of this study showed that honey contains copious amounts of phenolics and flavonoids. HPLC‐UV analysis showed a similar qualitative polyphenolic profile for all honey samples analysed. The main difference among samples was in the contribution of individual analytes, which was affected by floral origin. Total phenolic and flavonoid contents varied from 60.50 to 276.04 mg gallic acid equivalent kg^?1 and from 41.88 to 211.68 mg quercetin equivalent kg^?1 respectively. The antioxidant capacity was high and differed widely among samples. The FRAP value varied from 1.265 to 4.396 mmol Fe²⁺ kg^?1, while the radical‐scavenging activity expressed as DPPH‐IC₅₀ varied from 7.08 to 64.09 mg mL^?1. Correlations between the parameters analysed were found to be statistically significant (P < 0.05). CONCLUSION: The present study shows that honey contains high levels of phenolics and flavonoids and that the distribution of these compounds is influenced by the honey's floral origin. Copyright © 2009 Society of Chemical Industry