Multiple BBSRC projects aim to reduce possible cancer risk in common foodstuffs.
News that the chemical acrylamide was present in cooked and processed food products was extremely unpalatable when the discovery was first made in 2002. A known carcinogen in rodents, New Scientist magazine called acrylamide in food a "major concern" because the substance, known more as a highly reactive industrial compound, was not expected on the dinner plate (ref 1). What was it doing there? Why had this been missed for so long? And could low but lifelong dietary intake affect human health and well-being?
Acrylamide can form in a wide range of foods during frying, baking and roasting, including crisps, chips, biscuits and even coffee. The food industry responded rapidly to address the problem, but because acrylamide forms from natural precursors in a reaction that also produces desirable colours, flavours and aromas, reducing acrylamide levels in food is likely to be an ongoing problem.
It's a complex issue, and not just because of the chemistry involved. The issue is at an intersection between the science of food and the politics of public safety. In the absence of a definitive answer on risk, regulators in Europe are under pressure to bring about reductions in dietary acrylamide intake – a major challenge to commercial food producers and for plant breeders developing crops for the future.
Hence, BBSRC-sponsored researchers are using various funding mechanisms to collaborate with food growers and manufacturers to investigate whether products made from different wheat, potato and rye varieties contain different levels of acrylamide after cooking, and why. Results show that not all potatoes are created equal – products made from French fry and crisping varieties contain different levels of acrylamide – and factors such as storage duration also come into play in the acrylamide equation.
Since 2002, scientists have been looking at acrylamide levels in baked, fried and roasted foods, and whether it is present at levels that could cause disease over a lifetime's exposure. Acrylamide as a compound is classed as 'probably carcinogenic to humans' by the World Health Organisation (ref 2) and the International Agency for Research on Cancer, based on its carcinogenic action in rodents; it also has neurological, reproductive, and genotoxic effects (ref 3). (See 'Timeline').
"The acrylamide issue came out of the blue. No one expected it to be in food," says Professor Nigel Halford of Rothamsted Research, an institute that receives strategic funding from BBSRC. "Very quickly it was discovered that the Maillard reaction was responsible."
The Maillard reaction is very important to the food industry because it gives rise to colour flavour, such as the characteristic browning and distinctive, delicious 'roasty' taste of oven-cooked potatoes, as well as the oh-so pleasing look and aroma of freshly baked bread. First described in 1912 by French chemist Louis-Camille Maillard, the reaction takes place between amino groups and reducing sugars, such as glucose, maltose and fructose, during high-temperature baking, frying and roasting (but not boiling).
Acrylamide appears on the scene when the natural amino acid asparagine is involved in the last stages of the reaction (ref 4). This is a problem in potato-, wheat- and rye-based products because they contain relatively high concentrations of free asparagine, glucose and fructose.
Toxicology studies have shown that acrylamide can cause cancer in rodents, but using dosages typically 1000 times higher than would be found in human diets (ref 5). Results of epidemiological studies have been inconsistent: a recent Danish study, for example, claimed a link between acrylamide exposure and breast cancer-specific mortality (ref 6), whereas a meta-analysis of 15 studies concluded there was no clear evidence of any harm (ref 7).
It's hard to pin-point how bad the problem could be because we eat much lower concentrations of acrylamide than are fed to lab rats, but we do it for 70 years or more. And studying the effect of lifelong exposure is hard, partly because nearly everyone has been exposed to dietary acrylamide. Do you know anyone who's never eaten crisps, chips or biscuits?
Work in progress
Hence the need to tackle the problem at source by looking at what controls the concentrations of acrylamide's precursors. Halford has been researching metabolic regulatory processes involving sugars and amino acids for more than 20 years. "The acrylamide problem was an obvious thing for us to get into, and has become possibly the most important application of our research," he says. "Because I don't think we could persuade people, myself included, to abandon foods like roast potatoes, crisps or bread. Utilising BBSRC LINK projects (see 'LINKed in') and a CASE studentship he's been collaborating with academic colleagues at the University of Reading and the James Hutton Institute, as well as many industrial and commercial partners, to find ways of using plant genetics and agronomy to reduce acrylamide formation in the food we eat.
Halford is optimistic that there is lots of scope to reduce asparagine concentrations. This is because plants accumulate much more asparagine than they need, using it as a nitrogen store under stressful environmental conditions such as when sulphur levels in soil are low. This, Halford says, is particularly apparent in wheat and coaxing the plants to use another amino acid, glutamine for example, as a nitrogen store instead of asparagine might be a way around this.
However, reducing acrylamide risk in potatoes may be more complex than in wheat. Halford's previous research with Don Mottram's team at the University of Reading, which was funded by BBSRC and the Food Standards Agency (FSA), has shown that the relationship between asparagine, sugar concentrations and acrylamide formation is less clear in potatoes.
In Halford's present study, nine varieties of potato, including French fry varieties Maris Piper, Pentland Dell, King Edward, Daisy, and Markies, and crisping varieties Lady Claire, Lady Rosetta, Saturna, and Hermes grown in the UK in 2009 were analysed at monthly intervals through storage from November 2009 to July 2010 (ref 8). Halford's team results saw a big difference between potato varieties in acrylamide forming potential, from 104 parts-per-billion (ppb) in crisps from the lowest variety to 5250ppb in the highest. "That's a big difference," says Halford. "The significance for the food industry is that the European Commission's indicative level is 1000, so at 5000 you have a problem.
However, Halford cautions that the study looked at potential for acrylamide formation, whereas how much actually forms depends on cooking conditions or processing methods. The paper also covered only one growing season and more work needs to be done to understand how genetic factors interact with environmental factors from season to season.
The 1000ppb level (equivalent to 1mg per kg, or one thousandth of a gram per kilo) is neither a safety level nor a strict regulatory limit at present. Halford says that if harder evidence of harm from dietary acrylamide emerges then there may be pressure to introduce formal limits, but that currently such evidence does not exist. "The fact is that at the moment we don't know if dietary acrylamide is a safety issue," says Halford. "We know acrylamide causes cancer in rodents, and probably it will cause it in humans, but does it cause it at levels we get in the diet? We don't know."
What we do know is that regulatory bodies like the European Food Safety Authority (EFSA) and international bodies like the UN's Food and Agriculture Organisation (FAO) have stated that dietary acrylamide intake must be reduced. "And we know the European Commission is tending to regulate on hazard [how bad it could be] rather than risk [likelihood of that bad happening], so they may not wait for risk to be established," Halford explains. "This worries industry because they are dealing with a contaminant that forms from perfectly natural precursors, the concentrations of which vary in response to many factors, some of which are beyond their control."(See 'On the CASE')
Ahead of the game
Potato growers and food manufacturers are keenly aware that the politics may move ahead of the science and further reduce indicative acrylamide levels, or even move to instate binding limits. The prospect of moving goalposts made the need for industry and academia to collaborate more acute and working together enabled the partners to address both fundamental and applied aspects of the problem.
So growers and manufacturers are taking a close look at Halford's data on storage. In the UK potatoes are harvested from late July to October, most during September, and so potatoes used for most of the year come from storage. Halford and colleagues found that when potatoes are stored beyond their normal storage window, which is different for each variety, the risk of acrylamide formation goes up. Storage temperature is typically around 8-9ºC, which is cold enough to inhibit sprouting, but not so chilly that cold sweetening occurs; anything below that has serious implications for acrylamide formation because sugar concentrations rise.
"Once past their optimum storage window, potatoes should not be used for French fries or crisps," Halford warns, because acrylamide levels increase, rapidly going over the 1000ppb benchmark for crisps (600ppb for French fries). Halford states that the potato industry goes to great lengths to keep potatoes stable during storage, investing in sophisticated storage facilities and store management-systems to ensure that crops are at their optimum when delivered to customers for processing. However, he is concerned that some retailers may store at 4ºC to keep potatoes looking good, and that this could lead to high acrylamide formation in potatoes that are fried, baked or roasted in the home.
The researchers also found that some varieties, such as Markies and Maris Piper, tolerated longer storage better than others, and so could prove very useful commercially to provide potatoes from late storage. But it's likely that industry will still need new, improved varieties of potatoes in the future. Plants need asparagine to make proteins so it can't be bred out of the crop entirely, but could it be vastly reduced?
According to work by Halford's colleague Don Mottram, potato varieties that were specifically low in asparagine could still obtain the aroma, colour and flavour when cooked but with reduced acrylamide formation (ref 9, ref 10). However, it will be a challenge for potato breeders to uncouple asparagine from other free amino acids to make a potato low only in asparagine because there is a good correlation between asparagine levels and total free amino acids.
One alternative would be to knockout the gene that makes asparagine (asparagine synthetase) using genetic modification (GM). "A group in the US made a very low acrylamide potato by doing just that," says Halford. "There is currently no established biotechnology [GM] potato market at all, even in the US, but there are plans to bring low acrylamide GM potatoes to market. It will be interesting to see how the big fast-food chains react, which are all-important to the US potato industry. However, there is no prospect of a low acrylamide GM potato being developed for the European market in the foreseeable future."
Not just a potato problem
Acrylamide formation is also a major concern in wheat- and rye-based products, including breads, cakes, breakfast cereals and crisp breads. "It's not just a potato problem," Halford explains. "Wheat is a very important crop to the UK and EU, and acrylamide levels in wheat and rye products are worse than for some other cereals." The EU set different indicative levels for acrylamide in different products: breakfast cereals are 400ppb and biscuits 500ppb, while bread is 150ppb. Actual levels in most breads are much lower, in the 10s, but people eat a lot of bread so it's a major contributor to dietary intake.
In addition to the LINK potato project that ran from 2009-2012, Halford also leads a (standalone) BBSRC LINK programme on wheat. He says the aims of the wheat project are similar to those of the potato project: identify varieties and genotypes that are low in acrylamide-forming potential; understand asparagine metabolism and the underlying genetics; develop advice on best agronomic practice; provide breeders with the tools to produce new varieties, and disseminate the knowledge so that what we find out is acted upon.
On this project, he's teaming up with collaborators at another institute that receives strategic funding from BBSRC, the John Innes Centre. Together, they hope to use a technique called quantitative trait locus (QTL) analysis, which identifies regions of the genome which are associated with differences in the physical characteristics of the plant, such as asparagine levels (ref 11). The technique involves tracking many genetic markers using statistical analysis. Advanced but conventional (non-GM) techniques like this will give plant breeders the manoeuvrability to make next-generation crops that should, in theory, enable food producers to comply with regulatory limits not only in the near future but 10 years down the road and beyond.
The acrylamide issue is with us to stay. People's taste for fried, baked and roasted foods is going nowhere. The question is whether it will ever be definitively established that dietary intake is hazardous or not. As in many cases of science, there is no such thing as the 'final word'. It may be unrealistic to make food entirely risk-free, but squaring that with the public's food safety concerns, politics and the needs of industry is becoming increasingly difficult.
Nevertheless, Halford insists that it is a good news story. "Acrylamide must have been in our diet for thousands of years, so it is new knowledge, not a new risk, and levels have already been reduced as a result of that knowledge and the response of the food industry. It is a great example of public and private sector scientists and industry working together for the common good."
Jennifer Postles is studying for a PhD at Rothamsted Research's Plant Science Department in Harpenden, UK. Utilising a BBSRC CASE Studentship, she splits her time between institute and industry, in this case The Jordans & Ryvita Company, makers of well-known rye crisp bread snacks and breakfast cereals, because these training grants foster collaboration between academic and partner organisations.
What part of the acrylamide problem are you investigating for your CASE studentship PhD?
I'm looking at different ways we might be able to reduce the formation of acrylamide in rye products. This involves looking at the precursors of acrylamide, free asparagine and reducing sugars, and identifying factors that can affect their accumulation in rye grain. These factors can be genetic, so may vary between different rye varieties that are used to make crisp breads, or they may be environmental. I'm particularly interested in the effects of fertiliser application on precursor content.
How does fertiliser affect acrylamide precursors?
In the case of sulphur fertilisers, we saw no increase in free asparagine in rye grain grown under low sulphur conditions, which is a contrast with the results seen in wheat. This suggests rye either doesn't experience sulphur deprivation at the same levels as wheat, which makes sense seeing as rye is known to be a hardier crop with lower nutrient demands. Or rye may have a different mechanism in place to respond to sulphur deprivation which doesn't involve accumulating free asparagine. To investigate this I will be looking at the expression of genes in rye grown under low sulphur conditions.
What have you found in terms of results so far?
I used a field trial to compare the acrylamide forming potential of five varieties and found naturally occurring differences; some varieties accumulated lower levels of free asparagine than others. The variety that accumulated the highest amount of free asparagine in the trial had almost 50% more than the variety with the lowest asparagine concentration, and significantly more than the other three varieties. Free asparagine is the major precursor of acrylamide in rye, so these varieties may be preferable for use in baking in order to reduce acrylamide formation in rye crisp bread.
What have been the major obstacles so far?
With my work involving the genetics of rye, there has been a lot of extra work involved characterising the genes of interest. Because rye is a less economically important crop than other cereals such as wheat and barley, it isn't so well studied. Comparatively little is known about the genetics of rye, so a lot of work is done based on information from the wheat genome, which can be transferred across to rye due to their close evolutionary relationship.
I started the project in October 2009 and expect to finish at the end of 2013, so I've finished running field trials now and am finishing off the analysis of the samples.
How does your work tie in with Halford's recent paper or the wheat/potato LINK projects at Rothamsted?
As with the potato project, I am interested in the relationships between acrylamide formation and the concentrations of its precursors. These relationships are different in rye than has been shown in potatoes, and luckily for me, they are more straight-forward. Whereas in potato the concentrations of sugars are important, and the ratios of precursors to other metabolites play a role in determining the formation of acrylamide, in rye it seems that free asparagine is the determining factor. This provides an easier target for the selection of low acrylamide potential rye varieties.
What kind of collaborations do you have with your industrial partner?
Jordans-Ryvita is interested in identifying rye varieties and growing conditions that can produce grain which will form low amounts of acrylamide when baked. It is therefore important that any experiments I do to study these effects are relevant to their manufacturing process. This meant that the varieties grown in the field trial were all commercially used varieties which are known to bake well and which my sponsor is happy to use.
It's an interesting collaboration for me to be involved in as I can see how my work can be transferred to a real-world application, and have been able visit the Ryvita factory in Poole where the crisp breads are made. It was huge, and noisy, and they let me take some samples home. I really like the little cream cheese and chive ones, so I took a few packets of those with me!
The idea behind the LINK programmes was to ensure research is tied into and directly relevant to the needs of commercial industries; to do this, academic-industrial partnerships were encouraged and facilitated by the UK Government through their Department for Food and Rural Affairs (Defra).
The Defra LINK programmes are no longer open to receiving new applications. However, BBSRC operates its own 'stand-alone' LINK scheme for applications within BBSRC's remit where at least 50% of the full economic cost of the project comes from industry.
"I think the most important thing about a LINK project is that it ensures that the research addresses the needs of stakeholders because they are around the table at planning meetings," says Halford. "It has also been very useful for me to get a completely different perspective on the acrylamide problem and on broader issues faced by producers.
He adds that the commercial partners in the potato LINK project played a big part in directing the research, by informing which varieties were important to them and should be included in the study, for example. "For this paper, one of the partners provided the potatoes, which had been grown by farmers that supply them and stored in their commercial potato stores." In addition, a second paper is on the way in which a large part of the analysis was done by another partner, and in one study a field trial was replicated by one of the breeders.
- 2002: Acrylamide first detected in April by scientists working at the Swedish National Food Authority.
- 2002: The first statement (PDF) from the European Commission's Scientific Committee on Food confirms the Swedish scientists' findings and suggests "it may be possible to reduce the levels by changing the methods of production and preparation". It goes on to state that: "The Committee recommends that levels of acrylamide in food should be As Low as Reasonably Achievable (ALARA)".
- 2005: European Food Safety Authority (EFSA) statement notes there may be a potential health concern with acrylamide. The statement endorses the conclusions and recommendations of a previous risk assessment on acrylamide carried out by the Joint FAO/WHO Expert Committee on Food Additives (JECFA).
- 2007: European Commission recommends that EU Member States perform a three-year monitoring of acrylamide levels and submit data to the EFSA.
- 2009: The Confederation of the Food and Drink Industry (CIAA) (now called Food Drink Europe) updates its ' toolbox (PDF) ' that can be used selectively by food producers in line with their particular needs to lower acrylamide levels in their products. Short brochures containing information about each sector were developed, supported by and contributed to by regulators. The 'toolbox' was updated in 2011 to keep abreast of the latest scientific developments.
- 2010: JECFA (Joint FAO/WHO Expert Committee on Food Additives) evaluation of acrylamide concludes "margins of exposure to be low for a compound that is genotoxic and carcinogenic and that they may indicate a human health concern. Therefore, appropriate efforts to reduce acrylamide concentrations in foodstuffs should continue."
- 2010: EC recommends that Member States should continue annual monitoring.
- 2011: EC recommends Member States carry out investigations in cases where the levels of acrylamide in food exceed the prescribed indicative values.
- 2011 (early): EFSA issues ' indicative (PDF) ' levels for acrylamide in food in early 2011
- 2012: EFSA publishes a scientific report of Update on acrylamide levels in food from monitoring years 2007 to 2010 (PDF) , including 13,162 samples from 25 European countries. Worryingly, it finds a trend towards lower acrylamide levels could only be found in three of 22 food groups (a decrease in crackers, infant biscuits and gingerbread) but the conclusions were based on very small datasets for many product types. www.efsa.europa.eu/en/press/news/datex110420.htm
- 2012: European Commission plans to re-assess acrylamide levels by end of year.
- 2013: EFSA plans to update its European exposure assessment based on information on more recent acrylamide levels in food and new food consumption data.
- FoodDrinkEurope acrylamide toolbox (PDF)
- WHO acrylamide in food Q&A
- Chemistry, biochemistry and safety of acrylamide
- Acrylamide is formed in the Maillard reaction
- Chemistry, Biochemistry, and Safety of Acrylamide. A Review
- Pre-diagnostic acrylamide exposure and survival after breast cancer among postmenopausal Danish women
- Review of epidemiologic studies of dietary acrylamide intake and the risk of cancer
- Concentrations of free amino acids and sugars in nine potato varieties: effects of storage and relationship with acrylamide formation
- Kinetic Model for the Formation of Acrylamide during the Finish-Frying of Commercial French Fries
- Effects of sulphur nutrition during potato cultivation on the formation of acrylamide and aroma compounds during cooking
- Quantitative Trait Locus (QTL) Analysis