Chloroesters in foods: An emerging issue

April 2009

Process-based food contaminants have existed for many thousands of years, ever since prehistoric man first threw a haunch of wooly mammoth on a fire and produced polycyclic aromatic hydrocarbons. By comparison, it has been considerably less than 100 years since food safety agencies began conducting risk assessments of food contaminants. At this point, there are more knowledge gaps than facts about most potentially suspect compounds.

One such group of food contaminants—the chloropropanols—is of growing significance to the fats and oils community, particularly in their esterified (or bound) state. The chloropropanol most commonly found in food, in either its free or bound form, is 3-MCPD (3-monochloropropane-1,2-diol), although others are also of interest, including 2-MCPD (2-monochloropropane-1,3-diol), 1,3-DCP (1,3-dichloro-2-propanol), and 2,3-DCP (2,3-dichloro-1-propanol).

FORMATION AND TOXICOLOGY      

Current thinking suggests that 3-MCPD is formed as a result of a reaction between a source of chlorine (chlorinated water or sodium chloride) in a food or a food contact material and a lipid. The mechanisms of its formation are not fully understood. Two basic pathways have been proposed: thermally driven and enzyme-catalyzed (generally lipase) reactions. Direct precursors are thought to be glycerol and chloride. Recent work has also suggested glycidol (2,3-epoxy-1-propanol) as a precursor. (Glycidol is highly reactive and has been found to be a multisite carcinogen in both sexes in animal models, as well as a genotoxin in vitro and in vivo.)

Fatty acid esters of 3-MCPD (see Scheme 1) were identified in the early 1980s in adulterated Spanish rapeseed oil treated with aniline and refined with hydrochloric acid. To date, however, the majority of the scientific investigations and regulatory actions involving chloropropanols have come as a result of the detection of high levels of free 3-MCPD in acid-hydrolyzed vegetable protein (acid-HVP) and nonfermented soy sauces made from acid-HVP.

Ingestion of free 3-MCPD induces carcinogenic and benign tumors in experimental animals after long-term intake; it has not been found to be genotoxic in vivo. Both the European Commission’s (EC) Scientific Committee on Food (now the Scientific Panel on Contaminants in the Food Chain, or CONTAM) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA) set a tolerable daily intake (TDI) of 2 μg/kg of body weight in 2001. (The TDI is an estimate of the amount of a substance in air, food, or drinking water than can be taken in daily over a lifetime without appreciable health risk.) In 2008, JECFA set the maximum allowable content of free 3-MCPD in foods at 0.4 mg/kg (400 μg/kg) for liquid condiments.

Germany, supported by the EC and Canada, requested in 2008 that JECFA undertake full toxicological and exposure assessments on 3-MCPD esters. JECFA agreed to include fatty acid esters of 3-MCPD in its priority list, but not to assign a high priority to them because of the lack of available data. No toxicological or risk assessment work has yet been done on esterified chloropropanols.

The presence of 3-MCPD esters in processed food was first described in 2004, followed by the determination of 3-MCPD esters in refined, bleached, and deodorized vegetable oils in 2006, and in human breast milk in 2008. So far, no or only traces of 3-MCPD esters have been found in unrefined and native vegetable fats and oils. Only traces of 3-MCPD esters appear to be present in unrefined animal fats. The list of food and ingredients in which 3-MCPD esters have been detected includes bread, toasted bread, coffee and coffee creamer, non-HVP seasonings, cheese, cooked meat, salami, infant formula, margarine, french fries (chips), and doughnuts.

In November 2007, the German food safety agency (Bundesinstitut für Risikobewertung: BfR) called for levels of 3-MCPD esters to be reduced in oil-containing foods such as infant formula and margarine as well as in deep-frying fats, despite the absence of any indication of risk from bound 3-MCPD. In its assessment, BfR assumed that 100% of the bound 3-MCPD (and related isomers) is released from its esters during digestion in the gut. This assessment was based largely on an in vitro study using human intestinal Caco-2 cells that showed a high percentage of 3-MCPD esters were hydrolyzed.

However, based on an in vitro lipolysis model, Walburga Seefelder and coworkers at the Nestlé Research Center in Lausanne, Switzerland, propose that “the potentially slower release of 3-MCPD from 3-MCPD diesters, and the mono- to diester ratio, suggest 3-MCPD esters may in fact contribute only marginally to the overall dietary exposure to [free] 3-MCPD.” In addition, “the analysis of 11 different samples of fat mixes typically employed in food manufacturing demonstrated that a maximum of about 15% of the total amount of 3-MCPD bound in esters is present in the monoesterified form,” Seefelder and coworkers noted.

Scheme 1 (R = alkyl)

  3-MCPD                  3-MCPD diesters              3-MCPD monoesters

ANALYTICAL CHALLENGES

Adding to the uncertainty is the lack of a validated, fully collaboratively studied method for the detection of 3-MCPD esters. This makes it difficult to compare data from previously published articles as well as results obtained from analytical laboratories. Those working to mitigate the formation of 3-MCPD esters in vegetable oils have expressed frustration over the lack of reproducibility of analytical results. In fact, 3-MCPD can be either formed or lost during the process of extraction from food products.

One industry source, who wishes to remain anonymous, reports his company has sent identical samples on a weekly basis to a number of laboratories for 3-MCPD analyses, with widely varying results. "We have also sent portions of the same sample to the same lab over time, with widely varying results," he said. "Even as we are doing all this work, though, it is important to note that both free and bound 3-MCPD have been found in a great number of foods. It is unfair to highlight refined oils and fats too strongly," he stressed.

As a result of the BfR report in 2007, the European oil industry trade association, FEDIOL, developed an action plan. "In close collaboration with the European food industry . . . , FEDIOL continues to monitor the scientific information available on the analytical methods to determine 3-MCPD esters, their potential toxicity, and bioavailability, as well as investigate the origins of the 3-MCPD esters in foods . . . ," the group said in a written statement.

The first results of pilot-plant trials conducted by FEDIOL suggest that 3-MCPD esters are formed primarily at high temperature during oil deodorization, although some also are formed during bleaching. The degree of formation "is largely determined by crude oil characteristics and pretreatment conditions (especially crude oil type and bleaching conditions)," according to Gerrit van Duijn of Unilever. The preliminary FEDIOL work also found that the free fatty acid level at the start of deodorization had no effect on 3-MCPD ester formation. In addition, no significant difference in formation has been found between chemical and physical refining.

FEDIOL has established that in some oils, 3-MCPD esters have already formed at the minimum effective deodorization temperature of 180°C. "The level more or less doubles if the temperature is increased to 260°C," van Duijn said. The deodorization time, however, seems to be of "less importance."

"Our samples were analyzed by Eurofins using two methods," he clarified. "One uses NaCl in the sample preparation and one uses (NH₄)₂SO₄. The results with the NaCl-based method indeed showed a doubling of the 3-MCPD ester level when the temperature increased from 180 to 260°C. However, the results with the ammonium sulfate-based method showed hardly any effect of the temperature increase," he said.

Similarly, Katrin Hoenicke of Eurofins in Hamburg, Germany, has confirmed recent findings by Jan Kuhlmann of SGS Institut Fresenius in Berlin, Germany, suggesting that the 3-MCPD concentrations observed in some fats and oils were lower when sodium chloride was replaced by ammonium or sodium sulfate during sample preparation. The results can differ up to a factor of two or three, Hoenicke said. "In addition, other substances-such as glycidol-that may also be present in oils and fats can be converted into 3-MCPD during the sample preparation step when using the method published by the German official food laboratory (Chemisches und Veterinär Untersuchungsamt in Stuttgart)," Hoenicke noted. "This underscores the need to identify and validate methodology that accurately quantifies both free and bound 3-MCPD in fats and oils," she concluded.

Elsewhere, the Malaysian Palm Oil Board reports that it is conducting extensive research on 3-MCPD esters in a number of oils, looking at all stages of production. And the US Food and Drug Administration, which so far has been concerned only with free 3-MCPD in hydrolyzed proteins and soy sauces, told inform that the agency "will continue its collaborations to ensure that appropriate data are developed on the formation of 3-MCPD esters in foods and their potential impact on human health."

Catherine Watkins is associate editor of inform and can be reached at cwatkins@aocs.org

ILSI Europe holds workshop on 3-MCPD esters

A workshop on 3-MCPD esters in food and food ingredients (see main article on page 200 for background) was organized by two International Life Science Institute (ILSI) Europe task forces (Risk Assessment of Chemicals in Food and Process-Related Compounds/Natural Toxins) in cooperation with the Directorate General for Health and Consumer Affairs (DG Sanco) of the European Commission (EC). Held February 5-6, 2009, in Brussels, the workshop was attended by more than 70 participants from 20 countries, including scientists from government (US Food and Drug Administration, UK Food Standards Agency, and the European Food Safety Authority), academia, industry, and representatives of the EU Member States.

The primary aim of the workshop, according to ILSI Europe, was "to review all available data required for risk assessment, to identify the key data gaps for risk assessment, to define experimental research strategies to fill the data gaps, and to propose an action plan."

The two-day review of available data demonstrated that the issue of 3-MCPD esters in food is more complex than previously thought, with numerous molecular species requiring characterization and risk assessment. In addition, the lack of reproducibility of analytical results makes the development of a validated method vital for future work.

Presentations by key European regulatory and risk assessment agencies made it clear that, for the EC, the first priority is to reduce levels of 3-MCPD esters in food by risk mitigation measures taken by food business operators. Possible maximum allowable levels of 3-MCPD esters in food may be considered after more knowledge is gathered on the pathways of formation and on what levels are achievable by applying appropriate risk mitigation measures. "There is much work to be done," said Frans Verstraete of DG Sanco, "and that needs a global, comprehensive, and coordinated approach."

Participants agreed that:

• The issue of chloroesters in food needs to be examined from different perspectives: analytical, technological, and toxicological. In the meantime, there is no evidence that current food consumption patterns should be changed for public health reasons because of the presence of 3-MCPD esters.
• The first priority is for a ring-tested, fully collaboratively studied method for determination of chloroesters in food as well as a clear and coordinated global approach to answering the many questions surrounding chloroesters.
• Work on characterizing the formation of chloroesters is required to determine how to mitigate formation of 3-MCPD esters during refining and food processing.
• Because 3-MCPD esters in vegetable oils are probably formed during deodorization-a high-temperature process-the amount of esters as well as the ratio between 3-MCPD and 2-MCPD esters is most likely dependent on the deodorization temperature.
• Unintended consequences of mitigation must be elucidated carefully.
• The lack of exposure data and information about the in vivo oral bioavailability of 3-MCPD esters makes risk assessment impossible at this time. Further, a better understanding is needed of the ratio of monoesters of 3-MCPD to diesters, which appear to be hydrolyzed preferentially to sn-2 monoesters. Finally, a determination about whether all chloroesters (and their mono- and diesters) need individual toxicological evaluation must be made.
• The significance of glycidol esters must be established through quantitative work (the only findings to date are qualitative). In addition to esters, the significance of other forms of MCPD-adduct formation in foods, such as 1,3-dioxolans, must be settled.

"In addition to developing a web page and communications tool in partnership with ILSI Europe, AOCS Technical Services is ready to help industry and our constituency develop the analytical tools and resources needed to meet this global challenge," said Richard Cantrill, director of AOCS Technical Services and a participant in the ILSI workshop.

A web page available at www.aocs.org/tech/3_MCPD.cfm  includes background information as well as an extensive reference list and the presentations from the ILSI Europe workshop. Also, the analysis of 3-MCPD esters will be discussed at the 100th AOCS Annual Meeting & Expo in a session on contaminant analysis scheduled for Tuesday, May 5, from 1:55 to 5:00 p.m., tentatively scheduled in Gatlin E4.

information 

AOCS has developed a web-based reference list with explanatory material on 3-MCPD esters. The information is available at www.aocs.org/tech/3_MCPD.cfm , and includes a link to the presentations made at the ILSI Europe workshop on 3-MCPD esters in food products held February 5-6, 2009, in Brussels.