A03050: An investigation of the stability of BADGE in foods and the reaction products formed

Study Duration: September 2004 to September 2006

Contractor: Central Science Laboratory (Lead Contractor) TNO Quality of Life (Additional Scientific Contact)

The finding in the mid-1990's of bisphenol A diglycidyl ether (BADGE) migration from certain can coatings led to an explosion of publications on the measurement of BADGE and its then-known reaction products. These products are formed by BADGE reacting with hydrochloric acid in heated polyvinyl chloride lacquers to form chlorohydrins, and with water to form the corresponding mono- and diols. Although the formation of these reaction products is well documented, it is believed that they do not account for all the products that can derive from BADGE when it migrates into some foods. For example, when BADGE was spiked into tuna and then processed under typical canning conditions of 121
C for 30 minutes, the added BADGE disappeared almost completely and could not be accounted for by the known reaction products. It was postulated that the BADGE epoxy groups could be reacting with nucleophiles in the food.
This work aimed to confirm that BADGE levels decayed in some canned foods and that new reaction products formed between BADGE and food components. In addition it was hoped that new knowledge and understanding could be gained on the extent of these interactions and the identity and concentration of reaction products.

BADGE may be considered as an example of a more general case, especially for epoxy additives but also for other reactive substances, for which information on reaction products and transformation products is desirable.

BADGE was added at known concentrations ('spiked') into foodstuffs (sunflower oil, water, tuna in sunflower oil, tuna in water and apple pur
e), which were then canned, sterilised and stored at room temperature or at 50˚C. BADGE was also 'spiked' into bottled beers (ale, lager and stout) under an atmosphere of carbon dioxide. The beers were recapped and stored at 40˚C. The spiked samples were analysed at defined times and the concentrations of BADGE and its hydrolysis products were measured.

Modelling studies were carried out reacting BADGE with free amino acids, proteins and sugars. High pressure liquid chromatography (HPLC) was optimised to detect these reaction products. The BADGE-protein mixtures were also digested using enzymes prior to analysis.

Deuterium-labelled BADGE was synthesised and spiked into tuna in sunflower oil; apple pur
e; ale; lager; and stout samples. The samples were extracted and analysed as for non-labelled samples and also using Fourier Transform Mass Spectrometry (FT-MS) which enabled the exact mass and hence the elemental composition of the peaks to be determined. Enzyme digests were carried out to try to release any BADGE bound to proteins.

Analysis of food and beer samples spiked with unlabelled BADGE confirmed the literature reports. Additional peaks, believed to be BADGE-food reaction products, were detected in the chromatograms. BADGE gives a strong fluorescent response but the signal intensity of these additional peaks did not account for all of the BADGE that had been 'spiked' into the foods.
Analysis of the foodstuffs to which labelled BADGE had been added confirmed the identity of BADGE related peaks. Only a small percentage of BADGE 'spiked' into foods could be accounted for, this ranged from 1-62% depending on food type, storage time and extraction solvent.

The reaction of BADGE with amino acids in model systems was confirmed. The reaction rate is in the order Tyr < His < Met < Lys < Cys. For these reactions, all of the fluorescent response associated with the BADGE starting material was accounted for. It was also found that BADGE reacts with glucose and therefore it is expected that it will react with other hydroxy substances such as food sugars, in the same way.

MS analysis of BADGE reacted with the low molecular weight protein insulin found the [insulin + BADGE] reaction product at the expected molecular weight. For three other, higher MW proteins tested (bovine serum albumin (BSA), myosin and actin) the direct MS analysis did not detect any BADGE-protein products.

Two digestion protocols were followed, using trypsin and pronase. The tripsin digests of insulin and BSA (but not myosin and actin) showed peptide fragments with BADGE attached. All these tryptic peptides contained cysteine thereby supporting the idea that the reactivity of cysteine is higher than for other amino acids. Similar digestion experiments were conducted on some of the food samples. Some evidence was found for the presence of BADGE-related reaction products, i.e. BADGE reacted with proteins, in tuna. The pronase digests showed the presence of BADGE.H2O.Cys.

This work has confirmed that:

• BADGE levels in 'spiked' canned food decay
• New reaction products are formed by interaction with food components

This work has also provided more knowledge and understanding of:

• The extent of the interactions and the concentrations of the reaction products that may accumulate in canned foods and beverages
• The chemical nature/identity of some of the reaction products

Much of the reacted BADGE in the food and beverages studied was bound irreversibly to high molecular weight food components. All of the products identified arose from reaction with, and therefore loss of, the epoxide group of BADGE. Most of the bound residue could not be liberated using model digestion procedures.

The final report is available from the Agency's Information Centre.
To obtain a copy, please contact the Enquiry Desk, Information Services, Food Standards Agency (tel: 020 7276 8181/8182 or email: infocentre@foodstandards.gsi.gov.uk)

Contact: For any enquiries concerning this research project, please contact the relevant Programme contact or email: science@foodstandards.gsi.gov.uk