evaluation of commercial mineral water bottled in polyethylene terephthalate:
a cytogenetic approach with Allium cepa
Food Additives and Contaminants, 2000, Vol. 17, No. 12, 1037-1045 1dec00
Maria Grazia Evandri, Paolo Tucci and Paola Bolle*
Dipartimento di Farmacologia delle Sostanze Naturali e Fisiologia Generale, University of Rome `La Sapienza’ P. le Aldo Moro, 5 - 00185 Rome, Italy
(Received 24 March 2000; revised 5 July 2000; accepted 2 August 2000)
The aim of this study was to ascertain the possible toxicological effects of chemicals released into mineral water packaged in polyethylene terephthalate (PET) bottles. Two commercial mineral waters, bottled both in PET and glass and stored under different conditions, were examined using the Allium cepa test. The influence of the water samples on macroscopic (root length, colour and form) and microscopic (root tip mitotic index, chromosome aberrations) parameters was examined. The water samples were analysed after: (A) controled-condition storage (no direct light exposure and 18±2°C), (B) storage at 40°C for 10 days, in the dark (migration test in accordance with 82/711/EEC), and (C) exposure to sunlight and varying temperatures (18±38°C, mean temperature 25 § 38C). The two water samples bottled in PET induced cytogenetic aberrations regardless of the storage conditions. These signs of toxicity were evident even only 8 weeks after bottling, which is well within the recommended expiry date. Storage conditions were very important, as is suggested by the finding that chromosomal aberrations were particularly apparent after exposure to direct sunlight. However, as plant systems are not considered as primary screening tools by current inter-national guidelines for mammalian systems, extrapolation of the results from this test system to other systems and, eventually, to human beings should be based on results from a battery of assays covering various metabolic pathways.
* To whom correspondence should be addressed. e-mail: firstname.lastname@example.org
Keywords : PET bottled water, migration, storage conditions, Allium cepa test, chromosome aberrations
Polyethylene terephthalate (PET) is a relatively new packaging material prepared from terephthalic acid or its esters, and ethylene glycol. It is widely used for food and beverage bottling owing to properties (e.g. resistance, light weight, transparency and low production costs) which make it a good alternative to glass. Nevertheless, the use of plastic polymers (e.g. poly-vinyl chloride) for food packaging has raised questions regarding their safety (McGuinness 1986, Gilbert et al. 1986, Tice and McGuinness 1987, Begley and Hollifield 1989).
In fact, packaging can represent a source of contamination owing to the migration of substances into food (Begley and Hollifield 1989, Begley et al. 1990, Kim et al. 1990).
In order to control such contamination, many authorities have enacted extensive legislation. The European Economic Community’s (EEC) directives on food contact plastics (90/128/EEC, 96/11/EC, 82/ 711/EEC) take `positive lists’ of permitted substances that can be used to make retail packaging material and include Specific Migration Limits for some of these substances.
Following extensive testing, PET was judged safe and approved as a food packaging material. In spite of this, the world of finished food contact materials remains complex and often unknown: such materials contain a wide range of potential migrants (e.g. residues from the polymerization process, degradation compounds, many plastic additives) that are known to undergo transformations during processing involving substances which may not specifically be included in the positive lists (Gilbert et al. 1986, Freire et al. 1998). Moreover, packaged food may be subjected by both food retailers and consumers to treatments (e.g. different storage conditions) that may significantly influence the identity, the quantity and the toxicological profile of the migrating species. Phenomena of synergism or antagonism may also occur between substances leached into food. As there is no information concerning the toxicity and genotoxicity of a large number of these potential food contaminants (De Fusco et al. 1990), migration from food contact materials may represent a risk for human health, particularly during repeated and chronic exposure, e.g. via drinking water or foodstuffs. The most important long-term effects on human health might be mutagenicity and carcinogenicity. Although legislation exists, consumer safety might not be assured because of the difficulty in predicting and identifying substances likely to migrate into food (McGuinnes 1986). A significant part of the studies on migration of compounds from PET packaging materials into foodstuffs conducted to date have been based on the identification and quantification of PET compounds leached into real foods or simulants. Since PET is the most frequently used material for packaging mineral water, some of these studies investigated the migration process into mineral water. Results from analytical studies show that, besides the presence of known compounds, there are also chemicals whose identity has yet to be confirmed (Lo Russo et al. 1985, Begley and Hollifield 1989, De Fusco et al. 1990, Kim et al. 1990, Monarca et al. 1994, Linssen et al. 1995, Sauvant et al. 1995, Freire et al. 1998). The classical analytical methods of determination (HPLC, GC, MS, HSGC) have proved to be inexhaustive in monitoring the real status of migration from plastic packaging into food. Moreover, these analytical data do not have toxicological relevance, for which more specific investigations are required. Biological assays that are able to better evaluate the overall risk generated by one or several unknown substances which may migrate from packaging into packaged products are to be preferred (Sauvant et al. 1995).
Several plant test systems (e.g. Tradescantia, Vicia faba, Zea mays) in use since the 1970s have been found to be as sensitive as (i.e. few false negatives), and have responses comparable to, other genotoxicity short-term bioassays. Higher plants metabolize foreign compounds through a number of mechanisms (oxidation, hydrolysis, conjugation and, rarely, reduction) qualitatively similar to bio-transformation systems present in animal. The mixed function oxidase (microsomial cytochrome P-450) in higher plants (Frear et al. 1969) can activate a nonmutagenic chemical (promutagen) into an active form (mutagen) (Higashi 1988, Plewa and Gentile 1988, Grant 1994). Higher plants are eukaryotes which undergo mitosis, meiosis and mutation, and their chromosome structure is similar to that of man. By contrast, prokaryotic assays (e.g. Ames test) or mammalian cell cultures require addition of activating enzymes for the detection of promutagens.
Among plant test systems, Allium cepa is one of the most commonly used species for the study of chromosome aberrations: bulbs produce a large number of roots in a short period (the cell cycle in Allium root meristem is about 20 h) and chromosomes are relatively long (they average 10 mm). The Allium cepa test, introduced by Levan (1938), is a cytogenetic short-term bioassay that has proved to be a useful tool in basic research to evaluate the genotoxic risk of known chemicals (Fiskesjö 1988, Rank and Nielsen 1997, Steinkellner et al. 1998) and complex mixtures such as natural, drinking and industrial waste water (Fiskesjö 1985, Rank and Nielsen 1993, Nielsen and Rank 1994, Smaka-Kincl et al. 1996). The Allium cepa test allows the toxicity of aqueous samples to be evaluated through two cytological end points: root form and growth restriction, observable at the macroscopic level, and root tip meristem chromosome aberrations, scored at the microscopic level.
One of the advantages of the Allium cepa test is the possibility it offers to investigate samples not requiring any previous extraction, concentration or isolation procedure. Owing to qualities such as low cost, easy application and good correlation with mammalian genotoxicity test systems (Fiskesjö 1985, Plavica et al. 1991), the Allium cepa test represents an alter-native first-tier assay to experiments on animals for preliminary toxicity screening in accordance with the Council Directive 86/609/EEC art. 23 encouraging research on alternative techniques not requiring to animals use.
Table 1. Main physical-chemical parameters of water samples (W1 and W2) declared on bottle label.
Materials and methods
Two commercial non-carbonated mineral waters (W1 and W2) (table 1), bottled both in PET (W1p and W2p) and glass (W1g and W2g) and stored under different conditions, were examined using the Allium cepa test. All the samples, received within hours of bottling, were kept under controlled-condition storage (CCS) (no direct light exposure and 18±2°C) for 8 weeks and were, thereafter, exposed to different temperature and light conditions. Three experiments (A, B, C) were conducted. In experiment A, four samples (W1g, W1p, W2g, W2p) were studied at the end of the CCS period. In experiment B, the samples (W1g, W1P, W2g, W2P) were analysed after an additional storage at 40°C for 10 days in a fan-circulating oven, in the dark (overall migration test in accordance with 82/711/EEC). In experiment C, the samples (W1g, W1p, W2g, W2p) were exposed to sunlight at varying room temperature (18± 38°C, mean temperature 25 ± 3°C) for 16 weeks.
Glass bottles were green, whereas PET bottles were white for sample W1 and green for sample W2. An overview of exposure and storage conditions for the three experiments is provided in table 2.
Experimental design: Allium cepa test
The experiments were conducted as described by Fiskesjö (1985) with minor modifications. Equal-sized bulbs (1.5± 2 cm in diameter) of a commercial variety of Allium cepa L. (diploid; 2n = 16) were chosen. Just before use, the outer scales of the bulbs were carefully removed and the brownish bottom plates were scraped away without destroying the root primordia. The peeled bulbs, seven for each water sample, placed in distilled water during the cleansing procedure to protect the primordia from drying, were randomly placed on glass test tubes (1.5 cm in diameter, 10 cm in length) filled with the test liquids, and observed for a period of 7 days. The experiments were performed at controlled room temperature (18 ± 2°C) and in the dark. During the first 3 days, the test solutions were replaced daily. Thereafter, the test tubes were topped up with small amounts of the water samples. The bulb with the poorest growth during the initial 48 h was discarded in each group. On days two and three, respectively four and two root tips were excised from each bulb and microscopy slides were prepared in accordance with the Feulgen reaction procedure: fixation of the root tips in Carnoy reagent (ethanol 95% and acetic acid 45% , 3:1) for 20 min at room temperature, hydrolysis with 1 N HCl at 608C for 7 min followed by staining with leuco-fuchsin for 1 h at room temperature. Leuco-fuchsin was prepared by dissolving 1 g in 200 ml of boiling distilled water. The solution was stirred until its temperature was roughly 508C and then paper-filtered. HCl (1N; 30 ml) and K2S205 (3 g) were added and the solution was left for 24 h. Then activated charcoal (0.5 g) was added, the mixture stirred for 4 min and filtered. When bottled in dark glass and stored at ‡48C, the Feulgen stain prepared in this way is stable for 1 year.
Table 2. Experimental exposure and storage conditions applied on mineral water packaged in glass (W1g and W2g) and PET (W1p and W2p) bottles.
Preliminary storage time (weeks Experimental Exposure and Experiment (CCS)1 exposure time storage conditions A 8 — — B 8 10 days 40°C; dark C 8 16 weeks 18-38°C (mean 25 ± 3°C); direct sunlight 1CCS = controlled condition storage 18 ± 2°C; no direct light exposure.
When necessary, fixed root tips were kept in ethanol 70% and stored at +4°C for up to 2 weeks before hydrolysis.
From each of the well-stained root tips, the meristematic cell region was removed by cutting 2 mm from the root cap; this section was set on a clean slide, immersed in a drop of 45% acetic acid and squashed under a cover glass. In order to spread the cells evenly over the surface of the slide, the squashing process was accomplished by striking the cover glass with a pencil eraser using a bouncing 1 action. Slides were made permanent with Euparal® and then observed under 400 magnification. All slides were coded and examined blind.
Macroscopic parameter determination
For each bulb, the mean length of the roots in a root bundle was estimated on days one, two, three, four and seven, and the mean values of six measurements per experimental group, expressed in cm, were plotted. In addition, root colour and form (presence of crochet hooks or tumours) were recorded on the same days.
Microscopic parameter determination
Mitotic index (MI, number of dividing cells per 1000 observed cells) was determined on one slide per bulb in root tips excised on days two and three (six slides for each water sample/day), scoring 1000 cells. MI was measured to ensure that the toxicity was below an acceptable level for the scoring of chromosome aberrations; if the toxicity is very high, physiological cell damage might interfere with the scoring of chromosome aberrations caused by genotoxic compounds.
Chromosome aberration frequency was scored on 40 meta- and 40 ana-/telo-phases per slide (24 ‡12 slides for each sample taken on day two and three respectively).
Aberrations, classified as: (I) chromosome bridges and fragments, signs of clastogenic effect that result from chromosome and chromatide breaks; (II) vagrant, laggard chromosomes and c-mitosis, indicating spindle damage, were cumulatively considered.
Data were analysed using the statistical Sigma Stat version 2.0 (Jandel Scientific Software Corporation, Oxon, UK).
Student’s t-test was used to compare the mitotic index and root growth in glass and PET samples. Mann-Whitney U-test was used for macroscopic and chromosome aberrations.
Effect on macroscopic parameters
Data from three experiments are shown in table 3 and figure 1.
In experiment A, the study of the macroscopic parameters showed a statistically significant increase in the number of crochet hooks in PET samples stored for 8 weeks under controlled conditions (CCS) (W1p/g= 11/6; W2p/g= 10/4; p < 0:05 Mann-Whitney U-test), while root length was shorter on day seven in sample W1p (p < 0:05 Student’s t-test) but not in sample W2p. By contrast, the results obtained in experiment B (overall migration test) and in experiment C (varying storage condition test) showed no significant differences between the PET and glass samples as regards macroscopic root shape and growth.
Table 3. Number of macroscopic aberrations (hooks and/or tumours) scored in Allium cepa (n = 6) roots grown for 7 days in mineral water packaged in glass (W1g and W2g) and PET (W1P and W2P) bottles and stored under different temperature and light conditions.
Experiment . Sample A1 B2 C3 W1g 6 0 2 W1p 11* 4 6 W2g 4 4 2 W2p 10* 4 3 1CCS = 8 weeks of controlled conditions storage (18 ± 2°C; no direct light exposure). 2CCS +10 days at 40°C in the dark. 3CCS +16 weeks of varying conditions storage (18-38'C, mean 25 ± 3°C; direct sunlight exposure). *p < 0:05 Mann-Whitney U-test.
Effect on microscopic parameters
Mitotic index. The results of the mitotic index evaluation are summarized in table 4. Only in experiment A did data show a significant decrease (p < 0:05 Student’s t-test) in the mitotic index values on day two both for samples W1 and W2 bottled in PET (MI W1P/g= - 38.02% ; MI W2P/g= - 38.77% respectively) and on day three for sample W2 bottled in PET (MI W2P/g= - 38.13% ).
No differences in the mitotic index were observed in experiments B (overall migration test) and C (varying storage condition test).
Chromosome aberrations. The number of chromosome aberrations in root tips excised on days two and three was significantly higher in PET samples, regardless of the storage conditions (figure 2).
In experiment A (CCS), the number of aberrations increased by +125% (n.s.) and +250% (p < 0.05 Mann-Whitney U-test) for samples W1 and W2 respectively. In the overall migration test (experiment B), sample W1P showed a significant increase in the number of aberrations (+575% ) (p < 0.05 Mann-Whitney U-test), whereas the increase in sample W2P (+140% ) was not significant. In PET samples stored in direct sunlight and varying room temperature (experiment C), the amount of chromosome disturbances significantly increased: W1p= +71% (p < 0.05 Mann-Whitney U-test); W2p= +143% (p < 0.01 Mann-Whitney U-test).
Chromosome bridges and fragments were the most frequently observed disturbances (table 5). The results obtained in the three experiments for the two water samples were then compared to assess whether the higher incidence of chromosome aberrations in PET samples was due to water physical-chemical parameters and/or to bottle colour. No statistically significant difference was noticed. Results obtained in each experiment for the two water samples were then aggregated to ascertain whether the aberration incidence was still significant.
When PET data for the two samples in experiment A were considered together, the number of aberrations increased by +200% when compared with glass samples (p < 0.05 Mann-Whitney U-test). In the overall migration test (experiment B), when the samples were considered together, the number of aberrations in PET was +257% that in glass samples (p < 0:01 Mann-Whitney U-test).
In PET samples stored in direct sunlight and varying room temperature (experiment C), the amount of chromosome disturbances increased by + 104% (p < 0.001 Mann-Whitney U-test) (figure 3).
Figure 1. Allium cepa root length (mean ± SEM) grown for 7 days in mineral water packaged in glass (W1g = and W2g=) and PET (W1P = and W2P=) bottles and stored under different temperature and light conditions. 1 CCS=8 weeks of controlled conditions storage (18 f 2°C; no direct light exposure). 2 CCS +10 days at 40°C in the dark. 3 CCS+16 weeks of varying conditions storage (18± 38°C, mean 25 f 3°C; direct sun-light exposure). SEM =standard error mean; *p < 0:05 Student’s t-test.
Table 4. Mean ± SEM of mitotic index (number of dividing cells per 1000 observed cels) in Allium cepa root tips excised on days two and three from bulbs grown in water packaged in glass (W1g and W2g) and PET (W1p and W2p) bottles and stored under different temperature and light conditions (data from six slides for each water sample/day).
Experiment . A1 B2 C3 Sample Day two Day three Day two Day three Day two Day three W1g 26:3±1:7 26:9±2:8 24:7±1:7 19:7±2:2 28:7±3:3 18:2±2:8 W1P 16:3±3:6* 23:7±2:9 23:9±2:0 21:9±1:0 26:2±4:1 21:2±2:2 W2g 24:5±2:8 27:8±2:0 26:6±1:2 21:5±2:5 25:0±1:8 16:2±1:0 2P 15:1±2:4* 17:2±3:9* 24:1±0:7 22:2±2:5 21:3±2:0 20:0 ±2:4 1CCS = 8 weeks of controlled conditions storage (18±2°C; no direct light exposure). 2CCS +10 days at 40°C in the dark. 3CCS +16 weeks of varying conditions storage (18± 38°C, mean 25±3°C; direct sunlight exposure). SEM =standard error mean. * p < 0:05 Student’s t-test.
Some conclusions may be drawn from the data in obtained in our experiments.
Both PET bottled water samples displayed cytogenetic aberrations at the Allium cepa test, regardless of the storage conditions.
It is noteworthy that signs of toxicity were evident even only 8 weeks after bottling, which is a reasonable shelf-life period and is well within the recommended date of expiry.
Unexpectedly, the increase in the number of macroscopic aberrations and the reduction in root growth and MI were only noticed in experiment A (controlled condition storage, 18 ± 2°C). This suggests that these variations might be due to volatile, water-soluble chemicals such as acetaldehyde (b.p. 21°C), a thermal degradation product formed during the polycondensation and melt processing of PET. Its capability to migrate from PET bottles into mineral water has been demonstrated in a number of studies (Lo Russo et al. 1985, Linssen et al. 1995) and its mutagenicity for somatic cells reported (Dellarco 1988). It is highly likely, in fact, that in experiments B and C, in which the temperature was higher, all the volatile substances passed into the aerial phase on heating and were dispersed in the air on bottle-opening.
By contrast, non-volatile compounds leached from PET bottles, especially as a result of higher temperature and exposure to sunlight, might be the cause of the increased number of chromosome aberrations recorded in water samples packaged in PET.
It is interesting to note that even in Allium cepa grown glass-bottled water and exposed to direct sunlight, the incidence of chromosomal aberrations was higher than in glass-bottled water stored under controlled conditions, even if not significantly so. De Fusco et al. (1990) reported that daylight storage can increase mutagenic activity of PET-bottled water in the Ames test. Sunlight might trigger oxidative processes in glass too, followed by migration of toxic compounds into water.
In view of these considerations, in this study the Allium cepa test was used to ascertain the possible toxicological effect of chemicals released into mineral water stored in PET bottles. The Allium cepa test was conducted on commercial mineral waters bottled both in PET and glass stored under different conditions.
Figure 2. Number of chromosome aberrations in Allium cepa roots grown in water packaged in glass (W1g and W2g) and PET (W1P and W2P) bottles and stored under different temperature and light conditions. Root tips were excised on days two (four root tips per bulb) and three (two root tips per bulb). For each sample, chromosome aberration frequency was scored on 1920 dividing cells (day two) and on 960 dividing cells (day three) and aberration number cumulated. Chromosome bridges and fragments were the most frequently observed disturbances. 1CCS=8 weeks of controlled conditions storage (18±2°C; no direct light exposure). 2CCS+10 days at 40°C in the dark. 3 CCS+16 weeks of varying conditions storage (18± 38°C, mean 25±3°C; direct sunlight exposure). *p < 0:05; ** p < 0:01 Mann-Whitney U-test.
Figure 3. Cumulated data for chromosome aberrations in Allium cepa roots grown in water packaged in glass (W1 +W2)g and PET (W1 +W2)P bottles and stored under different temperature and light conditions. Chromosome aberration frequency was scored on 1920 +960 dividing cells for each sample. 1CCS=8 weeks of con-trolled conditions storage (18± 2°C; no direct light exposure). 2CCS+10 days at 40°C in the dark. 3CCS+16 weeks of varying conditions storage (18-38°C, mean 25 ± 3°C; direct sunlight exposure). *p < 0:05; ** p < 0:01; *** p < 0:001 Mann- Whitney U-test.
Table 5. Chromosome aberrations induced in Allium cepa root meristem cells by exposure to water samples stored under different temperature and light conditions: root tips were excised on days two (four root tips per bulb) and three (two root tips per bulb) and fixed in slides. Six bulbs per water sample were used. Chromosome aberration frequency was scored on 40 meta- and 40 ana-/telo-phases per slide.
Cells scored Per Per Chromosome aberration . bulb sample Bridges Fragments Laggard c-mitosis . Sample Day 2 Day 3 Day 2 Day 3 Day 2 Day 3 Day 2 Day 3 Day 2 Day 3 A1 W1g 320 160 4 2 2 0 0 0 0 0 W1P 320 160 3 5 2 4 0 0 1 3 W2g 320 160 2 0 4 2 2 0 2 0 W2P 320 160 4 12 5 10 4 0 3 6 B2 W1g 320 160 0 2 2 4 0 0 0 0 W1P 320 160 10 9 12 14 0 3 4 2 W2g 320 160 8 0 4 2 0 4 0 2 W2P 320 160 11 10 8 7 5 0 2 5 C3 W1g 320 160 8 6 5 3 4 0 1 1 W1P 320 160 10 8 12 4 8 4 0 2 W2g 320 160 7 3 5 2 0 2 4 0 W2P 320 160 15 9 17 6 5 0 1 3 1CCS = 8 weeks of controlled conditions storage (18±2°C; no direct light exposure). 2CCS ‡10 days at 408C in the dark. 3CCS ‡16 weeks of varying conditions storage (18-38°C, mean 25±3°C; direct sunlight exposure).
In view of these findings which highlight the importance of foodstuff storage, producers, distributors, dealers and users of PET-bottled beverages should be urged to introduce more controlled and thorough storing.
Although Kim et al. (1990) found the colour concentrate resin to be a source of phthalates in commercial amber PET bottles under the conditions we used, bottle colour does not appear to be involved in the migration phenomenon.
Data presented in table 5 show a homogeneous distribution of chromosome aberrations in root tips excised on days two and three of exposure. Chromosome fragments (observed both in the meta-and ana-/telo-phase) and bridges (observed in the late telo-phase) were the most frequently scored chromosome damages; this kind of alteration represents structural changes involving duplication or deletion in DNA double-strand.
This study, conducted to provide further data on PET bottle safety, proposes a new methodological approach to estimate migration from PET bottles into water. The Allium cepa test was found to be able to detect migration into water of compounds that are potentially toxic for the genetic material of somatic cells.
Nevertheless, short-term tests presently do not constitute definite evidence as to whether a sample does or does not pose a carcinogenic hazard to humans. Due to differences in efficiency of repair systems between plant and mammalian cells, an extrapolation of the results from this test system to other systems and, eventually, to human beings should be based on results from a battery of assays covering various metabolic pathways. However, the positive results in the Allium cepa and the Ames (De Fusco et al. 1990) tests suggest that the migration process may represent a biological hazard in other organisms.
BEGLEY,T.H., and HOLLIFIELD,H.C., 1989, Liquid chromatographic determination of residual reactants and reaction by-products in polyethylene terephthalate. Journal of the Association of Official Analytical Chemists, 72(3), 468-470.
BEGLEY, T. H., DENNISON, J. L., and HOLLIFIELD, H. C., 1990, Migration into food of polyethylene terephthalate (PET) cyclic oligomers from PET microwave susceptor packaging. Food Additives and Contaminants, 7(6), 797-803.
DE FUSCO, R., MONARCA, S., BISCARDI, D., PASQUINI, R., and FATIGONI, C., 1990, Leaching of mutagens into mineral water from polyethylene terephthalate bottles. Science of the Total Environment, 90, 241-248.
DELLARCO, V. L., 1988, A mutagenicity assessment of acetaldehyde. Mutation Research, 195, 1-20.
FISKESJÖ, G., 1985, The Allium test as a standard in environmental monitoring. Hereditas, 102, 99-112.
FISKESJÖ, G., 1988, The Allium test-an alternative in environmental studies: the relative toxicity of metal ions. Mutation Research, 197, 243-260.
FREAR, D.S., SWANSON, H.R., and TANAKA,F.S., 1969, N-demethylation of substituted 3-(phenyl)-l-methylureas: isolation and characterization of a microsomial mixed function oxidase from cotton. Phytochemistry, 8, 2157-2169.
FREIRE, M. T. DE A., CASTLE, L., REYES, F. G. R., and DAMANT, A. P., 1998, Thermal stability of polyethylene terephthalate food contact materials: formation of volatiles from retail samples and implication for recycling. Food Additives and Contaminants, 15(4), 473-480.
GILBERT, J., STARTIN, R., and MCGUINNESS, J. D., 1986, Compositional analysis of commercial PVC bottles and studies of aspects of specific and overall migration into foods and simulants. Food Additives and Contaminants, 3(2), 133-144.
GRANT,W.F., 1994, The present status of higher plant bioassays for the detection of environmental mutagens. Mutation Research, 310, 175-185.
HIGASHI, K., 1988, Metabolic activation of environmental chemicals by microsomial enzymes of higher plants. Mutation Research, 197, 273-288.
KIM,H., GILBERT,S.G., and JOHNSON,J.B., 1990, Determination of potential migrants from commercial amber polyethylene terephthalate bottle wall. Pharmaceutical Research, 7(2), 176-179.
LEVAN, A., 1938, The effect of colchicine on root mitoses in Allium. Hereditas, 24, 471-486.
LINSSEN, J., REITSMA, H., and COZIJNSEN, J., 1995, Static headspace gas chromatography of acetaldehyde in aqueous foods and polythene terephthalate. Zeitschrift fur Lebensmittel-Untersuchung und Forschung, 201(3), 253-255.
LO RUSSO, S., GRAMICCIONI, L., DI MARZIO, S., MILANA, M. R., DI PROSPERO,P., and PAPETTA,A., 1985, Acetaldehyde migration from poly (ethyleneterephthalate) (PET) containers. GC determination and toxicological assessment. Annali di Chimica, 75, 403-414.
MCGUINNESS, J. D., 1986, Migration from packaging materials. A need for more fundamental information. Food Additives and Contaminants, 3(2), 94-102.
MONARCA, S., DE FUSCO, R., BISCARDI, D., DE FEO, V., PASQUINI, R., FATIGONI, C., MORETTI, M., and ZANARDINI, A., 1994, Studies of migration of potentially genotoxic compounds into water stored in PET bottles. Food and Chemical Toxicology, 32(2), 783-788.
NIELSEN,M.H., and RANK,J., 1994, Screening of toxicity and genotoxicity in wastewater by the use of the Allium test. Hereditas, 121, 249-254.
PLAVICA, M., PAPES, N., and NAGY, B., 1991, 2,4 dichlorophenoxyacetic acid causes chromatin and chromosome abnormalities in plant cells and mutation in cultured mammalian cells. Mutation Research, 263, 77-81.
PLEWA,M.J., and GENTILE,J.M. (eds), 1988, Activation of promutagens by plant systems. Mutation Research, 197, 173-336.
RANK,J., and NIELSEN,M.H., 1993, A modified Allium test as a tool in the screening of the genotoxicity of complex mixtures. Hereditas, 118, 49-53.
RANK,J., and NIELSEN,M.H., 1997, Allium cepa anaphase-telophase root tip chromosome aberration assay on N-methyl-N-nitrosourea, maleic hydrazide, sodium azide, and ethyl methanesulfonate. Mutation Research, 390, 121-127.
SAUVANT,M.P., PEPIN,D., and BOHATIER,J., 1995, Chemical and in vitro toxicological evaluation of water packaged in polyvinyl chloride and polyethylene terephthalate bottles. Food Additives and Contaminants, 12(4), 567-584.
SMAKA-KINCL, V., STEGNAR, P., LOVKA, M., and TOMAN, M. J., The evaluation of waste, surface and ground water quality using the Allium test procedure. Mutation Research, 368, 171-179.
STEINKELLNER, H., MUN-SIK,K., HELMA, C., ECKER,S., MA,T. H., HORAK, O., KUNDI, M., and KNASMULLER, S., 1998, Genotoxic effects of heavy metals: comparative investigation with plant bioassays. Environmental and Molecular Mutagenesis, 31, 183-191.
TICE, P. A., and MCGUINNESS, J. D., 1987, Migration from food contact plastic. Part I. Establishment and aims of the PIRA project. Food Additives and Contaminants, 4, 267-276.