Coumarin and Cinnamaldehyde in Cinnamon Marketed in Italy: a Natural Chemical Hazard? Silvia Lungarini, Federica Aureli, Ettore Coni
To cite this version:
Silvia Lungarini, Federica Aureli, Ettore Coni. Coumarin and Cinnamaldehyde in Cinnamon Marketed in Italy: a Natural Chemical Hazard?. Food Additives and Contaminants, 2008, 25 (11), pp.1297-1305. 10.1080/02652030802105274. hal-00577395
HAL Id: hal-00577395 https://hal.archives-ouvertes.fr/hal-00577395 Submitted on 17 Mar 2011
HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Food Additives and Contaminants
For Peer Review Only
Coumarin and Cinnamaldehyde in Cinnamon Marketed in Italy: a Natural Chemical Hazard?
Journal: Food Additives and Contaminants
Manuscript ID: TFAC-2008-011.R1
Manuscript Type: Original Research Paper
Date Submitted by the 01-Apr-2008 Author:
Complete List of Authors: Lungarini, Silvia; Istituto Superiore di Sanità, National Centre for Food Quality and Risk Assessment Aureli, Federica; Istituto Superiore di Sanità, National Centre for Food Quality and Risk Assessment Coni, Ettore; Istituto Superiore di Sanità, National Centre for Food Quality and Risk Assessment
Methods/Techniques: Clean-up, Exposure assessment, HPLC, Method validation
Additives/Contaminants: Flavourings, Natural toxicants
Food Types: Bakery products, Beverages, Biscuits, Confectionary
http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 1 of 23 Food Additives and Contaminants
1 2 Coumarin and cinnamaldehyde in cinnamon marketed in Italy: a natural 3 4 5 chemical hazard? 6 7 8 9 10 SILVIA LUNGARINI, FEDERICA AURELI and ETTORE CONI 11 12 13 14 15 National Centre on Food Quality and Risk Assessment, 16 For Peer Review Only 17 Istituto Superiore di Sanità, Viale Regina Elena 299, 18 19 00161 Rome, Italy 20 21 22 23 24 25 26 27 Correspondence: Ettore Coni. E-mail: [email protected] 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 2 of 23
1 2 Abstract 3 4 Some plants that are processed into foods often contain natural substances that may be hazardous to 5 6 7 human health. One example is coumarin that is known to cause liver and kidney damage in rats, 8 9 mice and probably in humans. The main source of coumarin in the diet is cinnamon. The name 10 11 12 cinnamon is correctly used to refer to Ceylon Cinnamon, also known as “true cinnamon”. However, 13 14 other plant species are sometimes sold with the label of cinnamon. This is the case of Cinnamomun 15 16 aromaticum (Cassia).For In recent Peer years, due Reviewto its cheaper price, CassOnlyia is replacing true cinnamon in 17 18 19 the European food market being largely used in the preparation of some kinds of sweets. Several 20 21 European health agencies have recently warned against consuming high amounts of Cassia, due to 22 23 its high content of coumarin. In this study, thirty-four samples of cinnamon and fifty samples of 24 25 26 cinnamon-containing foodstuffs were collected from the Italian market. Quantitative determinations 27 28 of coumarin and cinnamaldehyde were performed by HPLC with diode array detector (DAD). The 29 30 analytical method was in house validated assessing recovery, repeatability, linearity, LOD and 31 32 33 LOQ. The results showed that about 51 % of cinnamon samples consisted of Cassia, 10 % were 34 35 probably a blend of cassia and Ceylon cinnamon whereas only 39 % were actually Ceylon 36 37 38 cinnamon. As far as cinnamon-containing foods are concerned, the samples often exceeded the 39 -1 40 maximum level fixed in the European Flavourings Directive of 2 mg kg . 41 42 43 44 45 46 47 Keywords: Coumarin; cinnamaldehyde, cinnamon; high-performance liquid chromatography. 48 49 50 51 52 53 54 55 56 57 58 59 60
2 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 3 of 23 Food Additives and Contaminants
1 2 Introduction 3 4 High dietary consumption of fruit and vegetables is promoted by nutritionist because this is 5 6 7 considered healthy. This notwithstanding, some foods of vegetable origin can contain natural 8 9 compounds that are hazardous to human health. One recognized example is coumarin, a natural 10 11 12 occurring flavouring substance which is found in relatively high concentrations in a wide variety of 13 14 plants (Hoult and Paya, 1996). Coumarin was first isolated from Tonka beans, and is present at high 15 16 levels in some essentialFor oils, Peer specifically cassiaReview leaf oil, cinnamon Only leaf and bark oils, lavender oil 17 18 19 and peppermint oil. Coumarin is also found in fruits, green tea and other vegetables, such as 20 21 chicory. However, the main contribution of coumarin in human diet is surely given by cinnamon. 22 23 24 25 26 Besides pepper and vanilla, cinnamon is, in fact, the best-known and most commonly used spice in 27 28 the world. The name cinnamon is correctly used to refer to Ceylon Cinnamon, also known as “true 29 30 cinnamon” (from the botanical name Cinnamomum verum ). In spite of this, other species of the 31 32 33 same genus are sometimes sold with the label of cinnamon. This is the case of Cinnamomun 34 35 aromaticum (Cassia) and Cinnamomum burmanii . The greater part of the spice sold as cinnamon 36 37 38 in the United States and Canada (where true cinnamon is still generally unknown) is actually cassia. 39 40 In some cases, cassia is labelled "Chinese cinnamon" to distinguish it from the more expensive true 41 42 cinnamon, which is the favourite form used in Mexico and Europe. "Indonesian cinnamon" can also 43 44 45 refer to Cinnamomum burmanii , which is also commonly sold in the United States, simply labelled 46 47 as cinnamon. 48 49 50 51 52 In recent years, due to its cheap price, Cassia is replacing true cinnamon also in European food 53 54 market being largely used in the preparation of some kinds of desserts, biscuits, cakes, chocolate, 55 56 57 spicy confectionery, tea, hot cocoa and liqueurs. This considered, the German Federal Institute for 58 59 Risk Assessment (BfR 2006) has recently warned against consuming high amounts of cassia, due to 60
3 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 4 of 23
1 2 its high content of coumarin which is present at much lower concentrations in Cinnamomum verum 3 4 and in Cinnamomum burnanii . 5 6 7 8 9 Coumarin is known to cause liver and kidney damage in rats and mice and there are isolated 10 11 12 incidents of similar hepatotoxicity in humans (WHO 1995). In particular, coumarin is a clear 13 14 carcinogen in rats and possibly in mice for oral exposure, noting that adenomas and carcinomas of 15 16 the liver and bile Forducts and Peerkidney adenomas Review have been observed Only in rats whereas adenomas and 17 18 19 carcinomas of the lung and liver adenomas in mice. 20 21 22 23 Cytochrome P450 (CYP)2A6 is the major enzyme involved in metabolizing coumarin to 7- 24 25 26 hydroxycoumarin. A reduction in CYP2A6 activity will lead to shunting of coumarin into other 27 28 metabolic pathways. In particular, coumarin is metabolized by CYP3A4 to form 3- 29 30 hydroxycoumarin, the major metabolite in mice and rats. It has been seen that an increase in the 3- 31 32 33 hydroxycoumarin ratio is associated with an increased production of the significant cytotoxic 34 35 product o-hydroxyphenylacetylacetaldehyde (O-HPA), suggesting that a shunting of coumarin 36 37 38 metabolism away from 7-hydroxylation is the most important cause of toxicity. Poor CYP2A6 39 40 metabolizers are, thus, more likely to metabolize coumarin via the cytotoxic pathway. Since its 41 42 toxicity, the Scientific Panel on Food Additives, Flavourings, Processing Aids and Materials in 43 44 45 Contact with Food of European Food Safety Authority (EFSA) has recently established a tolerable 46 47 daily intake (TDI) for coumarin of 0 – 0.1 mg kg -1 bw based on the no-observed-adverse-effect- 48 49 level (NOAEL) for epatotoxicity (EFSA 2004). 50 51 52 53 54 On the other hand, the Directive 88/388/EEC lays down maximum levels for coumarin in foodstuffs 55 56 -1 57 (EEC 1998). In particular, the amount of coumarin is limited to 2 mg kg for foodstuffs and 58 59 beverages with the exception of alcoholic beverages, certain types of caramels and chewing gums 60 (10, 10 and 50 mg kg -1, respectively). The European Commission did not maintain these levels in a
4 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 5 of 23 Food Additives and Contaminants
1 2 proposed new Regulation but a recent communication from the Commission to the European 3 4 Parliament has reconsidered the coumarin issue owing to new analytical data from BfR (EC 2008). 5 6 7 Germany requested re-introduction of maximum levels and the other Member States agreed with 8 9 this re-introduction. However, the food industry argues that the European regulations are far too 10 11 -1 12 strict and that, based on new toxicological data, a TDI for coumarin of 0. 64 mg kg bw poses no 13 14 health threat (Felter et al. 2006). For these reasons the European Commission organised two 15 16 technical meetingFor with Member Peer State experts Review and stakeholders Onlyfor identifying possible maximum 17 18 19 levels. Furthermore the Commission asked EFSA to provide additional scientific information about 20 21 a possible TDI review. In the meantime some leading manufactures of cookie goods have chosen to 22 23 change their recipes in order to reduce the amounts of coumarin in their products while other 24 25 26 producers are developing cinnamon analogue flavours with very low coumarin contents. 27 28 29 30 Besides high coumarin levels, Cassia has considerably higher levels of cinnamaldehyde than 31 32 33 Ceylon cinnamon. The Joint FAO/WHO Expert Committee on Food Additives (JEFCA) promoted 34 35 several toxicological assessments of cinnamaldehyde as a food ingredient in recent decades. 36 37 38 Following the setting up at the its eleventh meeting of a conditional Acceptable Daily Intake (ADI) 39 -1 -1 -1 -1 40 of 0 - 1.25 mg kg b. w. day , a temporary ADI of 0.7 mg kg b. w. day was established at its 41 42 twenty-third which was extended at its twenty-fifth and twenty-eighth meetings. This ADI was 43 44 -1 45 referred to a No Observable Effect Level (NOEL) of 125 mg kg b. w. calculated in a study on rats 46 47 (Hagan et al. 1967). At its thirty-fifth meeting no further extension of the value was, however, 48 49 undertaken because the required data were not available. At its twenty-fifth meeting an animal 50 51 -1 52 experiment NOEL of 620 mg kg b. w. was established on the basis of the evaluation of 53 54 cinnamaldehyde and other 54 structurally related flavouring substances. The safety margins were 55 56 57 also calculated to the above said exposure in the USA and Europe without establishing an ADI 58 59 (JEFCA 2001). The conclusion regarding current consumption was of "no safety concern". A 60 critical consideration of the JECFA assessment does, however, seem to be appropriate.
5 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 6 of 23
1 2 Cinnamaldehyde may, in fact, likely constitute a further risk for human health considering that 3 4 some experimental studies indicate it as possible genotoxic agent (Mantovani et al. 1989; Skidmore- 5 6 7 Roth 2004; Westra et al. 1998). 8 9 10 11 12 In this framework, the present work presents the monitoring of these two toxic compounds in 13 14 cinnamon and cinnamon-containing foodstuffs marketed in Italy. Moreover, the development and 15 16 validation of a For rapid and Peer simple HPLC Review method for the Only determination of coumarin and 17 18 19 cinnamaldehyde in food matrices has been also carried out. 20 21 22 23 Materials and methods 24 25 26 Reagents and chemicals 27 28 All solvent were from Mallinkrodt Baker (Phillipsburg, NJ, USA) and were HPLC grade. Coumarin 29 30 (purity > 99 %) and cinnamaldehyde (purity > 99 %) standards were purchased from Sigma-Aldrich 31 32 33 (Milan, Italy) and Carlo Erba (Milan, Italy), respectively. Stock standard solutions were prepared by 34 35 dissolving 10 mg of each standard in ethanol and adjusting the final volume to 10 mL. These 36 37 38 solutions were stable at 4 °C for three months. Working standard solutions were prepared by 39 40 diluting stock solution with ethanol/water (1 : 1 v/v). Working solutions were stable at 4 °C for four 41 42 weeks. 43 44 45 46 47 Apparatus 48 49 An Alliance 2690 HPLC system from Waters (Waters S.p.A., Vimodrone, Milan, Italy) was used 50 51 52 throughout this study. It was assembled as follows: a pump with autosampler device fitted with a 53 54 Vertex column (250 × 4.6 mm, Eurospher 100-5 m C18) and a RP18-e guard cartridge (40 × 55 56 57 4mm), both packed by Knauer (LabService Analytica S.r.l., Anzola Emilia, Italy); an HPLC column 58 59 heater (Croco-cil, Cluzean-infolabo SA, Saint-Foy-la-Grande, France); a DAD spectrophotometric 60 detector (mod. 996 from Waters).
6 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 7 of 23 Food Additives and Contaminants
1 2 3 4 Sample collection and treatment 5 6 7 Thirty-four samples of cinnamon (20 of powdered bark and 14 of whole quill bark), all from 8 9 different importers and fifty-two samples of cinnamon-containing foodstuffs, all from different 10 11 12 producers, were collected from twelve big distribution warehouses located in seven Italian regions.. 13 14 The whole quill barks were powdered by means of a ball-mill and sieved (0.5 – 1 mm). The 15 16 cinnamon-containingFor food Peer samples were Review lyophilized before Only powdering and sieving. The 17 18 19 lyophilisation greatly make easier the subsequent crushing, while at the same time allowing dilution 20 21 to be minimised. Moreover, lyophilised foodstuff samples with a high fat content were 22 23 preliminarily defatted by n-hexane extraction. Powdered and lyophilized samples were transferred 24 25 26 into polypropylene conical tubes and stored at - 20 °C until they were analysed. 27 28 29 30 Analytical procedure 31 32 33 Several methods are reported in literature on the isolation and analysis of compound of this class 34 35 that include gravimetric, titrimetric, photocolorimetric, and polarographic analysis, and the more 36 37 38 recently developed spectroscopic (UV and IR), fluorimetric, and chromatographic (GC and HPLC) 39 40 techniques (Areias et al. 2000; Cabral et al. 2001; He et al. 2005; Jayatilaka et al. 1995; Lozhkin 41 42 and Sakanyan 2006; Miller et al. 1995). The sample pre-treatment procedure and the 43 44 45 chromatographic conditions, adopted in this study, were similar to those described by Martino et al. 46 47 (2006) and Celeghini et al. (2001), respectively, except for some modifications both in the 48 49 extraction procedure and in the chromatographic conditions. 50 51 52 53 54 Briefly, 10 mL of ethanol/water (1:1 v/v) was added to the homogenised sample (about 1 g). After 55 56 57 agitation on vortex for 2 minutes and sonication for 30 minutes at room temperature, the sample 58 59 solution was centrifuged at 3000 × g for 10 minutes at room temperature. The use of hydro- 60 alcoholic solution as extracting solvent produced extracts that could be directly analyzed by HPLC
7 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 8 of 23
1 2 without further clean-up steps. After centrifugation, 1 mL of the resulting supernatant was, in fact, 3 4 filtrated by a 0.45 µm nylon filter and 20 µL of the clear solution was injected into the 5 6 7 chromatographic system for coumarin determination. Since cinnamaldehyde is present in cinnamon 8 9 at very higher concentrations, a 1:10 dilution of supernatant was necessary for cinnmaldehyde 10 11 12 determination. The eluents were: 1 mM ortophosphoric acid (A), methanol (B), acetonitrile (C). The 13 -1 14 isocratic elution consisted in A : B : C (80 : 20 : 20) at a flow rate of 0.7 mL min and column 15 16 temperature of 25For °C. Coumarin Peer and cinnamaldehyde Review were detected Only at 275 nm and peak area was 17 18 19 used for signals evaluation. Their contents in cinnamon and cinnamon-containing foods were 20 21 quantified using the respective matrix calibration curves corrected for recoveries. 22 23 24 25 26 Method validation 27 28 Before sample analysis, the described HPLC method was in house validated following the 29 30 requirements fixed by the International Standards Organization (ISO) guidelines. In absence of any 31 32 33 certified reference material (CRM) for coumarin and cinnamaldehyde in food matrices, method 34 35 accuracy was evaluated on the basis of recoveries obtained from in-house standard materials 36 37 38 (fortified samples). Precision, expressed as repeatability, was calculated by repeated analyses on the 39 40 same sample sets used for recovery tests, with the only difference that independent samples were re- 41 42 extracted and analyzed on two other occasions for calculating interday repeatability. 43 44 45 46 47 In order to calculate accuracy and precision in agreement with the ISO 5725-1 (ISO 1994), a 48 49 cinnamon sample at the lowest natural contents of coumarin and cinnamaldehyde (sample 11 in 50 51 52 Table 4), was fortified with coumarin and cinnamaldehyde standards at 8 (5, 10, 25, 50, 100, 250, 53 54 500 and 1000 mg kg -1) and 3 (5, 10 and 15 g kg -1) levels, respectively. Nine cinnamon samples for 55 56 57 each level (three for three different days) were prepared and analyzed for a total of 99 fortified 58 59 cinnamon samples. 60
8 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 9 of 23 Food Additives and Contaminants
1 2 The same procedure was adopted for cinnamon-containing food samples. The analogous food 3 4 samples without cinnamon as ingredient, were used as blank matrices and fortified with coumarin 5 6 -1 -1 7 and cinnamaldehyde standards at 4 (5, 10, 25 and 50 mg kg ) and 3 (0.5, 1 and 2.5 g kg ) levels, 8 9 respectively. Food samples were only fortified at the lowest levels because the concentration of 10 11 12 coumarin and cinnamaldehyde in cinnamon-containing foods are, obviously, much lower than those 13 14 measured in cinnamon. 15 16 For Peer Review Only 17 18 19 Limit of detection (LOD) and limit of quantification (LOQ) for cinnamon and cinnamon-containing 20 21 foods were calculated on the basis of the standard deviation (SD) and the slope (S) of calibration 22 23 curves as reported on the ISO 11843-1 standard (ISO 1997). Linearity was determined by means of 24 25 26 standard and matrix calibration curves obtained by HPLC analyses of standard solutions and the 27 28 above-mentioned fortified samples, respectively. The response factor (r.f. = peak area – 29 30 intercept/[analyte concentration]) test was applied for the purpose. The deviation of the r.f. of each 31 32 33 point of the calibration curve must be within ± 3 % of the experimental slope. 34 35 36 37 38 Finally, in order to comply with internal quality control (IQC) procedures, two control samples 39 40 (house reference materials) were inserted in each analytical batch made up of six samples. As 41 42 regards cinnamon analysis, two powdered cinnamon samples at two different levels of analytes 43 44 45 (approximately at the upper or lower decile) were used as control samples (samples 11 and 18 in 46 47 Table 4). In the case of cinnamon-containing food samples, the two control materials were 48 49 constituted by the analogous food samples without cinnamon as ingredient, fortified with coumarin 50 51 -1 -1 52 and cinnamaldehyde standards at two levels (5 and 50 mg kg for coumarin and 0.5 and 2.5 g kg 53 54 for cinnamaldehyde). The individual values obtained for control samples were plotted on a 55 56 57 Shewhart control chart during the entire duration of the study. 58 59 60 Results and discussion
9 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 10 of 23
1 2 Method validation parameters 3 4 Under the described chromatographic conditions, the retention times of coumarin and 5 6 7 cinnamaldehyde are about 17.3 and 29.9 min., respectively, as shown in Figure 1. The accuracy and 8 9 the precision, expressed as intra- and interday repeatability, of the method are reported in Table 1 10 11 12 and 2. Recovery data were satisfactory with values ranging from 90 to 100.0 % and from 95 to 110 13 14 % for coumarin and cinnamaldehyde in cinnamon, respectively while in the case of cinnamon- 15 16 containing foodstuffsFor the recoveryPeer values Review ranged from 90 to Only 91 % and from 91 to 97 % for 17 18 19 coumarin and cinnamaldehyde, respectively. These values fall within the guideline range from – 10 20 21 to + 10 % considered acceptable for the investigated mass fractions (Thompson 2006). 22 23 24 25 26 The coefficients of variation (CV) values for intraday and interday repeatability ranged from 1.1 to 27 28 7 % and from 2.8 to 10.3 % for coumarin in cinnamon and cinnamon-containing foodstuffs, 29 30 respectively. As far as cinnamaldehyde is concerned the CV values varied from 0.4 to 2.9 % and 31 32 33 from 0.5 to 3.8 % in cinnamon and cinnamon-containing foodstuffs, respectively. These data 34 35 indicate that both intraday and interday repeatability are good, since all CV values are below the 36 37 38 recommended limits proposed by Horwitz on the basis of the following equation: CV(%) = 0.67 ⋅ 39 40 2(1-0.5logC) where CV is proposed acceptable limit for repeatability and C is the concentration of 41 42 analyte in the sample as a decimal fraction (Horwitz 1982). 43 44 45 46 47 For coumarin, the LOD was 0.7 mg kg -1 both in cinnamon and in cinnamon-containing foodstuffs 48 49 -1 50 while the LODs for cinnamaldehyde were 0.5 and 0.02 g kg in cinnamon and cinnamon- 51 - 52 containing foodstuffs, respectively. Consequently, the calculated LOQ for coumarin was 2.3 mg kg 53 54 1 both in cinnamon and in cinnamon-containing foodstuffs while in the case of cinnamaldehyde 55 56 -1 57 LOQs were 1.6 and 0.07 g kg in cinnamon and cinnamon-containing foodstuffs, respectively. The 58 59 observed differences for LOD and LOQ values of cinnamaldehyde in cinnamon and cinnamon- 60
10 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 11 of 23 Food Additives and Contaminants
1 2 containing foodstuffs are due to the different calibration curves employing for calculating their (see 3 4 below). 5 6 7 8 9 Two standard calibration curves for coumarin and cinnamaldehyde in hydro-alcoholic solution 10 11 12 (ethanol/water 1:1 v/v) and four matrix calibration curves for coumarin and cinnamaldehyde in 13 14 cinnamon and cinnamon-containing food, respectively, were calculated. The relative linear 15 16 regression equationsFor and squared Peer correlation Review coefficients (r 2) are Onlyreported in Table 3. 17 18 19 20 21 All curves showed a good linearity in the whole range of the tested concentrations (0-1000 mg kg -1 22 23 for coumarin and 0.5-15 g kg -1 for cinnamaldehyde) with r 2 values ranging from 0.9823 to 1. 24 25 26 Moreover, r.f.values were within the above-mentioned limit ( ± 3 %), since the deviations of all 27 28 points of the curves were in the range of – 1.9 to + 1.6 of the slopes of the corresponding equations. 29 30 31 For the quantification of coumarin and cinnamaldehyde calibration curves corrected for recoveries 32 33 in cinnamon and cinnamon-containing food samples were used. Finally, as regards control samples, 34 35 concentration values exceeding the “action limits” were not observed during the 4-month period of 36 37 38 the method application. 39 40 41 42 Cinnamon samples 43 44 45 Analytical results for all cinnamon samples are reported in Table 4. As expected, coumarin contents 46 47 ranged from a few mg up to thousands mg per kg of cinnamon. This because, as above mentioned, 48 49 50 cassia cinnamon has significant amounts (up to 5 %) of coumarin, whereas Ceylon cinnamon only 51 52 has traces (about 0.004 %) and so coumarin contents in commercial spices vary depending on the 53 54 different type of cinnamon employed. Also cinnamaldehyde levels were very different in cinnamon 55 56 -1 57 samples ranging from 0.5 to 25.8 g kg but there was not, as expected, a direct relationship with 58 59 coumarin contents. In samples 4, 8, 9, 15, 19 and 24, the coumarin concentrations were rather high 60 whereas cinnamaldehyde contents were among the lowest. For samples 12, 22 and 31, it was just
11 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 12 of 23
1 2 the opposite. The highest content of cinnamaldehyde was found in sample 33 (25.8 g kg -1) where, 3 4 however, coumarin level was not at the highest concentrations. Probably, cinnamaldehyde content 5 6 7 is not always suitable for distinguishing Ceylon cinnamon from cassia cinnamon. 8 9 10 11 12 Regarding ground cinnamon, coumarin levels let suppose that about 70 % of the tested samples 13 14 derives from cassia cinnamon whereas 85 % of whole quill bark samples are referable to Ceylon 15 16 cinnamon. This apparentFor discrepancy Peer can Review be explained with theOnly fact that whole quill barks of 17 18 19 cinnamon allow the identification of their origin and, so, the consumer can recognize the finest 20 21 “true cinnamon”. This is, in fact, light reddish-brown, rolled into a single quill whereas cassia is 22 23 dark brown and rolled from both sides toward the centre so that they end up resembling scrolls. 24 25 26 Moreover, cassia cinnamon is about 1.5 mm thick while cinnamon is less than 0.08 mm thick. 27 28 29 30 On the contrary, it is virtually impossible for the average consumer to distinguish between Ceylon 31 32 33 cinnamon and cassia cinnamon when the spices are ground. Since the ground cinnamon is almost 34 35 always cassia, a large use of this flavouring powder in plain cooking can pose health risk for 36 37 38 consumers, above all for children. 39 40 41 42 Cinnamon-containing foodstuff samples 43 44 45 A summary of the contents of coumarin and cinnamaldehyde found in all analyzed foodstuff 46 47 samples is given in Table 5. As regards coumarin, most tested foods (about 70 %) exceeded the 48 49 maximum level (2 mg kg -1) stated in the European Flavourings Directive 88/388/EEC (EEC 1988 ). 50 51 52 This is a clear indication that the food industry processes cassia cinnamon because the percentage of 53 54 cinnamon, employed as ingredient in foodstuff production, rarely exceeds 1 %. Consequently, only 55 56 57 the use of cassia cinnamon, which has considerably high coumarin levels, can explain the high 58 59 concentrations of this natural toxicant in tested foods. 60
12 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 13 of 23 Food Additives and Contaminants
1 2 On the other hand, the high contents of coumarin found in this survey give cause for concern, being 3 4 in many cases significantly higher than the statutory limit. Since children for average body weight 5 6 7 and dietary habits certainly represent a sensitive group of population, an assessment of coumarin 8 9 and cinnamaldehyde diet intake has been attempted. Children consume a greater amount of sweets 10 11 12 than adults and expressing this amount on a body weight basis, the relative consumption increases. 13 14 The calculated probabilities of exceeding coumarin exposure limits are therefore much higher for 15 16 children. For Peer Review Only 17 18 19 20 21 Based on the Italian average daily consumption of sweets (biscuits, cakes and desserts, ice creams 22 23 excluded), chocolate and confectionery (93.5, 11 and 41 g/person, respectively) and the contents of 24 25 26 coumarin found in these foodstuffs (highest values), the estimated intake of coumarin was about 27 28 0.21 mg kg -1 b.w. for a 6 year old child with an average body weight of 20 kg (D’Amicis et al. 29 30 2002; Turrini et al. 2001; Turrini et al. 2002). This intake is twice the TDI value for coumarin (0.1 31 32 -1 33 mg kg bw) established by EFSA. Obviously, the calculated intake represents the worst case 34 35 scenario where all consumed sweets, chocolate and confectionery contain coumarin at the highest 36 37 38 levels. However, the possible risk must not be underestimated, also considering that the mentioned 39 40 daily consumption is referred to average children. There are so called “big consumers” that can 41 42 double the amount of sweets daily consumed. 43 44 45 46 47 The exposure calculation revealed, thus, that in the worst case scenario children who eat a lot of 48 49 cinnamon-containing foods can be at risk of health adverse effects. On the contrary, 50 51 -1 - 52 cinnamaldehyde levels in all foodstuffs investigated ranged from < LOD (0.020 g kg ) to 3.20 g kg 53 54 1 and, thus, cinnamaldehyde intakes from these foodstuffs are unlikely to pose any significant risk to 55 56 57 health also considering a precautionary ADI of 0.7 mg/kg body weight/day. Probably, the high 58 59 volatility of cinnamaldehyde accounts for the low levels found in foodstuffs. Many tested products 60
13 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 14 of 23
1 2 being, in fact, either baked (biscuits, cakes) or undergone at a moderate heat for their fusion 3 4 (chocolate, candies) can release cinnamaldehyde during production processes. 5 6 7 8 9 Conclusion 10 11 12 Although cinnamon is not widely used in Italian cookery, several manufactured foods from retail 13 14 outlets, have coumarin contents that could pose risk to human health. With regard to this problem, it 15 16 is opportune to stressFor that strengthening Peer theReview official food control Only in the European Union would be 17 18 19 sufficient to reduce the possible risk, since many marketable foods exceeded the maximum level (2 20 21 mg kg -1) stated for coumarin in the Flavourings Directive 88/388/EEC. Concerning this the results 22 23 of the present study have been submitted to the Italian Minister of Health so that it can adopt all 24 25 26 necessary measures to strengthen the official food control on this hazard and to demand to the 27 28 national food industry a re-examination of the self-control plans in accordance with the HACCP 29 30 system. 31 32 33 34 35 On the other hand, the aim of risk analysis is to protect the most sensitive groups of consumers from 36 37 38 health risks related to a high intake of natural toxicants. In this framework besides risk assessment 39 40 and risk management, the risk communication associated with some plant constituents is important, 41 42 too. Consumers should be informed about possible hazards so that they can choose their dietary 43 44 45 habits on the basis of their individual sensitivity and need for well-being. The official advice about 46 47 moderate consumption of cinnamon-containing foodstuffs have also been issued with particular 48 49 reference to sensitive consumers such as children and pregnant women. 50 51 52 53 References 54 55 56 57 Areias FM, Valentão P, Andrade PB, Moreira MM, Amaral J, Seabra RM. 2000. HPLC/DAD 58 59 analysis of phenolic compounds from lavender and its application to quality control. J Liq Chrom 60 & Rel Technol. 23:2563-2572.
14 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 15 of 23 Food Additives and Contaminants
1 2 BfR Health Assessment, No. 043/2006, 16 June 2006. Consumers, who eat a lot of cinnamon, 3 4 currently have an overly high exposue to coumarin. Available from: http://www. 5 6 7 .bfr.bund.de/cd/template/index_en 8 9 Cabral LM, Dos Santos TC, Alhaique F. 2001. Development of a profitable procedure for the 10 11 12 extraction of 2-h-1-benzopyran-2-one (coumarin) from Mikania glomerata. Drug Dev Ind Pharm. 13 14 27:103 -106. 15 16 Celeghini RMS, VilegasFor JHY, Peer Lanças FM. Review 2001. Extraction andOnly Quantitative HPLC Analysis of 17 18 19 Coumarin in Hydroalcoholic Extracts of Mikania glomerata Spreng: ("guaco") Leaves. J Braz 20 21 Chem Soc. 12:706-709. 22 23 D’Amicis A, Intorre F, Maccati F, Pettinelli A, Martines S, Forlani F, Battistini N, Tunfio O, 24 25 26 Ponziano M, Carbini L, Lantini T, Nieddu MJ, Peretti M, Podda C, Giacchi M, Gulino M, 27 28 Corsinovi E, Cuda C, Leonardi F, Calvo GM, Portelli G, Di Bella MG, Sculati O, Villa M, Rosati 29 30 S, Bertazzoli S, Anelli N. 2002 Studio sui consumi alimentari e ripartizione dei pasti degli scolari 31 32 33 dell’obbligo in Italia (SCARPS). Rivista Italiana di Scienza dell’Alimentazione 31(3):235-48. 34 35 EC 2008. Communication from the Commission to the European Parliament pursuant to the second 36 37 38 subparagraph of Article 251 (2) of the EC Treaty concerning the common position of the Council 39 40 in the adoption of a Regulation of the European Parliament and the Council on flavourings and 41 42 certain food ingredients with flavouring properties for use in and on foods and amending Council 43 44 45 egulation (EEC) No 1576/89, Council Regulation (EEC) No 1601/91, Regulation (EC) No 46 47 2232/96 and directive 2000/13/EC. COM 142 final, Brussels, 11.3.2008. 48 49 EEC. 1998. Council Directive 88/388/EEC of 22 June 1988 on the approximation of the laws of the 50 51 52 Member States relating to flavourings for use in foodstuffs and to source materials for their 53 54 production. Official Journal of European Community L 184, 15/07/1988:0061-0066. 55 56 57 EFSA. 2004. Opinion of the Scientific Panel on Food Additives, Flavourings, Processing Aids and 58 59 Materials in Contacts with Food on a equest from the Commission related to Coumarin. The 60 EFSA Journal 104 (2004).
15 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 16 of 23
1 2 Felter SP, Vassallo JD, Carlton BD, Daston GP. 2006. A safety assessment of coumarin taking into 3 4 account species-specificity of toxicokinetics. Food Chem. Toxic. 44: 462-475. 5 6 7 Hagan EC, Hansen WH, Fitzhugh OG, Jenner PM, Jones WI, Taylor JM, Long EL, Nelson AA, 8 9 Brouwer JB. 1967. Food flavourings and compounds of related structure. II. Subacute and 10 11 12 chronic toxicity. Food Cosmet Toxicol. 5(2):141-157. 13 14 He Z, Qiao C, Han Q, Cheng C, Xu H, Jiang R, But PP, Shaw P. 2005. Authentication and 15 16 quantitative analysisFor on the Peer chemical profile Review of cassia bark (cor Onlytex cinnamomi) by high-pressure 17 18 19 liquid chromatography. J Agric. Food Chem. 53: 2424-2428. 20 21 Horwitz W. 1982. Evaluation of Analytical Methods for Regulation of Foods and Drugs. Anal 22 23 Chem. 54:67A-76A. 24 25 26 Hoult JR, Paya M. 1996. Pharmacological and biochemical actions of simple coumarins: Natural 27 28 products with therapeutic potential. Gen Pharmacol. 27 (4):713-722. 29 30 ISO. International Organization for Standardization. 1994. Standard ISO 5275-1:1994. Accuracy 31 32 33 (trueness and precision) of measurement methods and results – Part 2: Basic method for the 34 35 determination of repeatability and reproducibility of a standard measurement method, ISO, 36 37 38 Geneva, Switzerland. 39 40 ISO. International Standard Organisation. 1997. Standard ISO 11843-1:1997. Capability of 41 42 detection, ISO, Geneva, Switzerland. 43 44 45 Jayatilaka A, Poole SK, Poole CF, Chichila TMP. 1995. Simultaneous micro steam 46 47 distillation/solvent extraction for the isolation of semivolatile flavor compounds from cinnamon 48 49 and their separation by series coupled-column gas chromatography. Anal Chim Acta 302:147- 50 51 52 162. 53 54 JECFA Safety evaluation of certain food additives and contaminants. 2001. WHO Food Additives 55 56 57 Series 46 (2001). 58 nd 59 L. Skidmore-Roth 2004. Mosby’s Handbook of Herbs and Natural Supplements 2 Ed. St Louis, 60 Mo Mosby.
16 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 17 of 23 Food Additives and Contaminants
1 2 Lozhkin AV, Sakanyan EI. 2006. Natural coumarins: Methods of isolation and analysis. Pharm 3 4 Chem J. 40:337-346. 5 6 7 Mantovani A, Stazi AV, Macrì C, Ricciardi C, Piccioni A, Barellino E. 1989. Pre-natal (segment II) 8 9 toxicity study of cinnamic aldehyde in the Sprague-Dawley rat. Food Chem Toxicol. 27:781-786. 10 11 12 Martino E, Ramaiola I, Urbano M, Bracco F, Collina S. 2006. Microwave-assisted extraction of 13 14 Coumarin and related compounds from Melilotus - Officinalis (I.) pallas as an alternative to 15 16 soxhlet and ultrasoundFor assisted Peer extraction. Review J Chromatogr A. 1125:147-151. Only 17 18 19 Miller KG, Poole CF, Chichila TMP. 1995. Solvent-assisted supercritical fluid extraction for the 20 21 isolation of semivolatile flavor compounds from cinnamons of commerce and their separation by 22 23 series-coupled column gas chromatography. J High Resol Chromatogr. 18:461-471. 24 25 26 Thompson M. 2006. The International Harmonised Protocol for the Proficiency Testing of 27 28 Analytical Chemistry Laboratories. Pure and Appl Chem. 78:145-196. 29 30 Turrini A, Saba A, Perrone D, Cialfa E, D’Amicis A. 2001 Food Consumption Patterns in Italy: the 31 32 33 INNCA Study 1994-96. European Journal of Clinical Nutrition 55(7):571-88. 34 35 Turrini A, Martines S, Orsini S, De Carli A, D’Amicis A. 2004 Abitudini alimentari: Tendenze 36 37 38 evolutive nella popolazione e nei giovani. In: Atti del Convegno Informazione statistica e 39 40 politiche per la promozione della salute. Roma, 10-12 settembre 2002. Roma: ISTAT; p. 45-60. 41 42 Westra WH, McMrray JS, Califano J, Flint PW, Corio RL. 1998. Squamous cell carcinoma of the 43 44 45 tongue associated with cinnamon gum use: A case report. Head Neck 20:430-433. 46 47 WHO. 1995. Coumarin: a srong association with hepatotoxicity. World Health Organization Drug 48 49 Information 9 (1995) 159. 50 51 52 53 54 55 56 57 58 59 60
17 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 18 of 23
1 2 3 Table I. Performance of the analytical method for the determination of coumarin in cinnamon and 4 cinnamon-containing foodstuffs 5 6 a 7 Fortification level Measured content Recovery Intraday Interday 8 9 (mg kg -1) (mg kg -1) (%) repeatability b repeatability c 10 11 12 CV (%) CV (%) 13 14 Cinnamon 15 16 5 For Peer 4.5 Review 90 Only 4.3 7.0 17 18 19 10 9 90 6.2 6.6 20 21 25 23 92 2.9 5.4 22 23 24 50 50 100 3.1 3.9 25 26 100 91 91 2.7 2.9 27 28 250 239 96 3.2 3.7 29 30 31 500 452 90 1.3 1.7 32 33 1000 966 97 1.1 3.7 34 35 Cinnamon-containing foodstuff (biscuit) 36 37 38 5 4.5 90 4.5 5.1 39 40 10 9 90 6.4 5.3 41 42 43 25 22 90 2.8 10.3 44 45 50 45 91 5.6 3.8 46 47 aValues are mean ± SD for nine samples. bValues are referred to three independent samples 48 c 49 analyzed in one day. Values are referred to nine independent samples analyzed in three different 50 days (three samples analyzed each day). 51 52 53 54 55 56 57 58 59 60
1 http://mc.manuscriptcentral.com/tfac Email: [email protected] Page 19 of 23 Food Additives and Contaminants
1 2 3 Table II. Performance of the analytical method for the determination of cinnamaldehyde in 4 cinnamon and cinnamon-containing foodstuffs 5 6 7 Fortification level Measured Recovery Intraday Interday 8 9 g kg -1 content a (%) repeatability b repeatability c 10 11 -1 12 g kg CV (%) CV (%) 13 14 Cinnamon 15 16 5.00 For 4.5Peer Review 95 Only 2.4 2.9 17 18 19 10.00 11.0 110 0.8 2.6 20 21 15.00 14.5 97 0.4 1.8 22 23 24 Cinnamon-containing foodstuff (biscuit) 25 26 0.50 0.45 91 3.5 3.8 27 28 1.00 0.97 97 1.2 3.4 29 30 31 2.50 2.35 94 0.5 2.4 32 33 aValues are mean ± SD for nine samples. bValues are referred to three independent samples 34 analyzed in one day. cValues are referred to nine independent samples analyzed in three different 35 36 days (three samples analyzed each day). 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
2 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 20 of 23
1 2 Table III. Linear regression equations obtained for coumarin and cinnamaldehyde in standard and matrix calibrations curves. 3 4 2 5 Matrix Analyte Test range Regression equation Linearity (r ) 6 a 7 Aqueous solution Coumarin 5 – 1000 y = 11855x - 67999 1 8 9 Cinnamaldehyde 5 – 15 b y = 6044884x + 82416158 0.9823 10 For Peer Review Only 11 a 12 Cinnamon Coumarin 5 – 1000 y = 11399x -137311 0.9987 13 14 Cinnamaldehyde 5 – 15 b y = 9856675x + 46609263 0.9964 15 16 Cinnamon-containing foodstuffs Coumarin 5 – 50 a y = 10129x -3138.6 0.9999 17 18 b 19 Cinnamaldehyde 0.5 – 2.5 y = 19250332x + 871420 0.9972 20 21 aValues are in mg kg -1 22 bValues are in g kg -1 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 3 45 http://mc.manuscriptcentral.com/tfac Email: [email protected] 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 21 of 23 Food Additives and Contaminants
1 2 3 Table IV. Coumarin and Cinnmaldehyde contents in Cinnamon marketed in Italy 4 5 6 Cinnamon Coumarin Cinnamaldehyde Cinnamon Coumarin Cinnamaldehyde 7 8 sample (mg kg -1) (g kg -1) sample (mg kg -1) (g kg -1) 9 10 11 Powder 18 3094 19.9 12 13 1 14 8.9 19 2215 9.4 14 15 2 1625 12.3 20 2071 11.9 16 For Peer Review Only 17 18 3 1654 10.5 Rolled 19 20 4 1424 5.7 21 8 9.9 21 22 5 14 3.51 22 3 11.4 23 24 25 6 1970 13.3 23 4445 21.6 26 27 7 5 11.7 24 3297 8.4 28 29 30 8 973 5.0 25 3 11.6 31 32 9 2213 9.0 26 5 9.2 33 34 10 2635 19.0 27 380 14.1 35 36 37 11 6 0.5 28 9 11.1 38 39 12 16 6.8 29 12 18.7 40 41 13 1233 10.3 30 83 7.2 42 43 44 14 2336 12.8 31 7 12.1 45 46 15 1758 8.8 32 75 6.5 47 48 16 1886 9.1 33 736 25.8 49 50 51 17 1977 11.9 34 11 12.5 52 53 54 55 56 57 58 59 60
4 http://mc.manuscriptcentral.com/tfac Email: [email protected] Food Additives and Contaminants Page 22 of 23
1 2 Table V. Coumarin and cinnmaldehyde contents in some cinnamon-containing foodstuffs marketed in Italy 3 4 5 Cinnamon-containing foodstuff Number of Coumarin Cinnamaldehyde 6 -1 -1 7 samples (mg kg ) (g kg ) 8 9 mean median range mean median range 10 For Peer Review Only 11 c 12 Biscuit 10 12 7 1 - 23 < LOQ < LOD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Fig.1. Typical chromatogram (HPLC-DAD) of the hydroalcoholic extract of a cinnamon sample. 33 282x211mm (72 x 72 DPI) 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 http://mc.manuscriptcentral.com/tfac Email: [email protected]