Addressing the Challenge of Acrylamide in the Bakery Sector

Transcription

1 Addressing the Challenge of Acrylamide in the Bakery Sector Colin G Hamlet, Peter Sadd, Li Liang BCCC 2009 Technology Conference: Drivers For Success 26th and 27th March 2009 Stratford-Upon-Avon, Warwickshire

2 Outline Why acrylamide is important and how it is formed Reduction options (pilot scale) Raw material selection Changes in key precursors during processing Pilot scale reduction in sweet biscuit and cracker systems Effects of ingredients/additives Interactions with other contaminants Conclusions

3 Why is Acrylamide Important? Neurotoxic (repeated high dose in humans) Long term low level exposure? Mutagenic & carcinogenic observed in rats Evidence for genotoxic mechanism of carcinogenicity (mediated by glycidamide) increasing H 2 N O H 2 N O O DNA Acrylamide glycidamide

4 Why is Acrylamide Important? JECFA estimate pastry & sweet biscuits contribute 10 20% exposure 1 Biscuits 2 Crispbread 2 Gingerbread 2 Other bakery ware 2 Total number μg/kg μg/kg μg/kg μg/kg Min % Median % % Max MOEs for both morphological nerve changes and carcinogenicity are relatively low (<10000 = concern for substances which are mutagenic & carcinogenic) 1 efforts to reduce acrylamide should be continued ALARP 1) 64 th JECFA 2) Data from

5 Outline Chemistry Amino acid (all) Amino acid (other) Other Maillard products (brown colours, aromas, flavours) Intermediate Sugar Asparagine Acrylamide Decay products Acrylamide is formed from asparagine (Asn) in a reaction mediated by sugars

6 Asn is the key determinant of acrylamide in cereals Yield of acrylamide from added sugars in dough is circa g/100g Yield of acrylamide from added Asn in dough is circa 0.1 g/100g

7 Mitigation options Asn reduction by Raw material selection Yeast fermentation Complexing agents Asparaginase Reaction mechanism Amino acid addition (competitive inhibition, decay) ph adjustment (Maillard) Ammonium salt replacement (reactive intermediates) Other Dilution (surface area/volume effect) Oven profiling/rf heating

8 Precursors in Wheat Berry Fractions Highest levels of both sugars and Asn are found in the germ and bran Germ fraction Bran fraction Wholemeal White flour Sugar (g/100g) Fructose Glucose Sucrose Maltose Amino acid (mg/kg) ASN GLU ASP GLN SER GLY THR ARG ALA TYR VAL MET TRP PHE ILE LEU LYS So more refined flours give less acrylamide (but at the price of other health benefits) Asn accumulates in plants under conditions of stress, so it is a variable source of acrylamide

9 Amino Acids in Commercial Bread Wheat Flours Roughly ordered by total amino acid levels then Asn Asn was as variable within as between varieties No link with the total free amino acid levels Selection by growing variety is not enough Malacca Claire Solstice Robigus German Canadian French Einstein ASN GLU ASP GLN SER GLY THR ARG ALA TYR CYS VAL MET TRP PHE ILE LYS Amino Acid Level (mg/kg)

10 Asn variation in other flours Asn (mg/kg) Rice Wheat (white) Barley Maize Wheat (w/meal) Oats Rye

11 Biscuit/confectionery Flours: Asn v Reducing Sugars Reducing sugars and Asn vary in tandem Oats is outlier, but only 2 samples Grist selection may offer opportunities for acrylamide reduction But not for bread flours, possibly due to selection on protein quality/quantity. Fructose+Glucose (g/100g) Barley Maize Oats Rice Rye Wheat Wheat (wholemeal) All data (excluding oats) y = 0.001x R 2 = Oats Oats Asparagine (mg/kg)

12 Products Produced (kg scale) Savoury biscuits (cream crackers) Sweet biscuits C.f. market product

13 Recipes Flour Salt Water Fat Yeast Strong Short dough biscuit Invert syrup NaHCO HN 4 HCO Tartaric acid Soft Castor sugar 0.9 Straight dough cracker Sponge & dough cracker Sponge dough Simplified commercial recipes 1 Invert syrup was 36% glucose, 32% fructose Other additives: CaCl 2, CaCO 3 Ca propionate ( %, 2-4x standard level when used as a preservative) Phytic acid (0.5-1%, comparable to the levels found in wheat) Propionic acid 1. Derived from: Whitely, P. R., (1971), Biscuit manufacture; fundamentals of in-line production. London: Elsevier Publishing Company Ltd.

14 Pilot Scale Manufacturing Processes Short dough biscuits Doughs z-blade mixed, sheeted to 3 mm and docked Baked in pilot commercial oven with control doughs as sets of 3x3 on stainless steel mesh (rising oven profile) Crackers (straight dough/sponge & dough processes) Doughs spiral mixed, proved, laminated, sheeted to 3mm and docked Baked in pilot commercial oven with control doughs as sets of 3x3 on stainless steel mesh (rising oven profile) Measured final moisture, ph, acrylamide

15 Relative Reactivity of Sugars Data from sugar add back experiments on model bread doughs As expected monomers are most reactive Fructose could be created in situ by inversion of sucrose in acid conditions Extra Acrylamide (μg/kg /% added) Sucrose Glucose Maltose Fructose

16 Release of Sugars in Fermented Cracker Doughs Sucrose rapidly consumed by yeast 10 More fructose produced than corresponding loss of sucrose (glucofructans?) Glucose and maltose rise as expected (starch damage + yeast action) Sugar (mmol/100 g flour basis) Virtually all sugars 2 gone after 18hrs (0.1% total) e.g. in sponge process fructose maltose glucose sucrose Fermentation time (min)

17 Levels of all amino acids fall during proof However.. Effect of Yeast on AAs in Cracker Doughs Amino acid level (mg/kg flour basis) lys leu ile phe trp met val tyr ala arg thr gly ser gln asp glu 0 (Flours) (Straight doughs) (Sponge) Fermentation time (min) asn

18 Effect of Yeast on AAs in Cracker Doughs Levels of all amino acids fall during proof However, yeast selectively consumes Asn After extended fermentation times (c.18h) all the Asn was consumed (not shown) Asparagine level (mg/kg flour basis) asparagine (mg/kg, flour) asparagine (% all amino acids) Asparagine (% all amino acids) Fermentation time (min)

19 Effect of Yeast on acrylamide in Cracker Doughs Despite initial fructose surge early in fermentation, acrylamide levels fall rapidly 3-MCPD, generated from yeast glycerol, does not increase with time Glycerol formation may be limited by sugars availability Acrylamide (μg/kg) acrylamide 3-MCPD MCPD (μg/kg) Fermentation time (min)

20 Low Gassing Yeast in (Bread) Dough For short fermentation products, doubling yeast significantly cuts Asn after 1 h Low gassing bakers yeast is almost as effective as the standard in consuming Asn May be possible to optimise yeast addition / gassing and Asn reduction Asparagine/Initial Asparagine (%) Low gassing bakers yeast (2.7% addition) Control bakers yeast (2.7% addition) Control bakers yeast (5.4% addition) Proof time (min)

21 Effect of Dough Age in Sweet Biscuits More acrylamide is formed in doughs which have been allowed to age Increase is ~35% over three hours Extra acrylamide is not due to a rise in sugar levels Amino acid levels rise with age for at least the first hour after mixing Acrylamide or asparagine Increase in Asn is enough to explain the extra acrylamide 0 50 Acrylamide (ug/kg) Asparagine (mg/kg) Glucose (%) Fructose (%) Dough age (hr) Sugar (%)

22 Effect of Amino Acid Addition on Savoury Biscuits 5.6 µmol/kg of three amino acids added separately to basic recipe ( %) Severe (c.40%) dilution of amino acids 15-20% drop in acrylamide formation Simple AA dilution Lysine Glycine Cysteine FSW dough % Change in Acrylamide

23 Effect of Amino Acid Addition on Savoury Biscuits II Dough spiked with 100% 626ppm 13 C 3 90% acrylamide 80% No loss of acrylamide during bench proof (six 70% 20 C) 60% 1556 mg/kg glycine 50% added to FSW recipe Cooked at 160 C for 40% 10-80min 30% Decay of the 13 C 20% labelled acrylamide was slightly slowed by 10% the extra glycine 0% Labelled acrylamide remaining (%) FSW dough FSW dough + glycine Time at 160 C (min)

24 Effect of Metal Ions in Savoury Biscuits Doughs made from pure flour compared to (UK) normal and enhanced calcium fortification when cooked Acrylamide levels fell with both increasing calcium and magnesium concentration Both metals produced comparable effects for equal quantities of their metal ions % change in acrylamide level Base flour Normal Ca 1% added 1.38% added fortification [1] CaCl2.2H20 MgCl2.6H O 2 2 O (flour (flour weight) wt) (flour (flour weight) wt) Ca+Mg (%) [1] SI 1998 No 141: g/100g CaCO 3

25 Effect of Metal Ions: Theory Multivalent cation crosslinking / ionic interactions H O Reduce solubility of Asn Stabilise complexes at high temperature H HO H H O H O O O - Ca 2+ O NH 3 + O OH O H H H O - OH O H OH O H O - O NH2

26 Effect of Additions on Sweet and Savoury Biscuits Phytic acid and calcium reduced acrylamide as expected Calcium propionate was NOT beneficial Propionate alone had little effect Cracker doughs were typically twice as sensitive to additions as sweet doughs Type and level of addition to dough Calcium carbonate (1.0%) Calcium chloride (1.0%) Cracker doughs Sweet doughs Calcium chloride (2.0%) Calcium propionate (0.75%) Calcium propionate (0.70%) Calcium propionate (0.38%) Calcium propionate (0.35%) Propionic acid (0.3%) Phytic acid (1.0%) Phytic acid (0.5%) Fructose (1.0%) Change in acrylamide from control (%)

27 Effect of Raising Agents in Sweet Biscuits (B) + sodium bicarb + tartaric acid + double ammonium bicarb (B) + sodium bicarb + tartaric acid + double ammonium carbonate (B) + sodium bicarb + tartaric acid + ammonium carbonate CONTROL: (B) + sodium bicarb + ammonium bicarb + tartaric (B) + double sodium bicarb Recipe (B) + sodium bicarb + tartaric (B) + sodium bicarb + tartaric sodium bicarb + (B) no raising agents (B) no raising agents (B) no raising agents (B) -90% -80% -70% -60% -50% -40% -30% -20% -10% 0% 10% 20% 30% 40% % Change in Acrylamide Any raising agent increases acrylamide, but ammonium based agents are worst

28 Effect of ph and NH 3 in Sweet Biscuits ph shifts can explain non NH 3 raised data 40% 30% 20% 10% No ammonium salts All doughs Effect of ph is linear and in line with work on bread (17% vs 12%) % Acrylamide change 0% -10% -20% -30% -40% -50% -60% -70% y = x % -90% ph

29 Effect of NH 3 in Sweet Biscuits Effect of ph removed As expected NH 3 strongly increased acrylamide Plateau in acrylamide possibly due to increased decay at high NH 3 levels ph adjusted % Acrylamide change 90% 80% 70% 60% 50% 40% 30% 20% 10% y = x x 0% Moles NH 3 /kg

30 Effect of NH 3 in Savoury Biscuits Ammonium bicarbonate added to savoury biscuit formulation Acrylamide levels again more sensitive than with sweet doughs Increase was insensitive to bake time or addition level 0.5% added ammonium salt: 10 min bake 0.5% added ammonium salt: 20 min bake 1.5% added ammonium salt: 20 min bake 0% 50% 100% 150% 200% 250% 300% 350% Change in acrylamide from control (µg/kg)

31 Crust Colour v Contaminants (in Bread) Data from loaves made in breadmaking machines, but similar results seen in other processes Crust colour (L) is a good indicator for all contaminants Hence oven profiling (where the heat is put into the product early and the end of the oven is run cooler) can be helpful. Contaminant (µg kg -1 ) MCPD Acrylamide Ethyl carbamate 100 R 2 = R 2 = R 2 = Crust colour (L)

32 Effect of Dough ph (bread) Low ph is beneficial for acrylamide But not for 3-MCPD Practical ph range is small Similar trends with cracker doughs Acrylamide (µg/kg dough) Acrylamide MCPD (µg/kg dough) 3-MCPD Dough ph (before cooking)

33 Predicted Acrylamide v 3-MCPD (in Bread) 3-MCPD levels (dashed contours) are static or reduced whenever acrylamide is reduced (solid contours) Hence baking changes such as modified oven profiles are safe

34 Acrylamide Reduction Methods: Conclusions Raw material selection may be helpful for (low sugar) products In real bakery processes yeast has a protective effect giving lower levels of acrylamide in the final food than would be expected from the ingredients Recipe or process changes which affect yeast may have unexpected effects on final acrylamide levels as it is a filter/amplifier for acrylamide precursors Sugars can be released or consumed Asn can be consumed Low gassing bakers yeasts may offer a reduction route in some cases Extra amino acids have a modest effect on acrylamide at commercial levels Best practice should avoid allowing sweet biscuit doughs to age Calcium supplementation is beneficial (statutory fortification), but interactions with other ingredients (especially propionate) need investigation Ammonium salt replacement is the priority for chemically raised products Low ph is beneficial for acrylamide, but limited by dough buffering low phs promote 3-MCPD formation Temperature and moisture changes to reduce acrylamide will not promote 3-MCPD

35 Acknowledgements The support of the UK Foods Standards Agency for this work is gratefully acknowledged For more information see: Updated (17/2/09) CIAA toolbox (incorporating information from GMA members) & pamphlets 1 Hamlet,C.G.; Sadd,P.A.; Liang,L. Correlations between the amounts of free asparagine and saccharides present in commercial cereal flours in the united kingdom and the generation of acrylamide during cooking. J. Agric. Food Chem., 56 (15), (2008) Sadd,P.A.; Hamlet,C.G.; Liang,L. Effectiveness of methods for reducing acrylamide in bakery products. J. Agric. Food Chem., 56 (15), (2008) Konings, E.J.M., Ashby, P., Hamlet, C.G., Thompson, G.A.K., (2007) Acrylamide in cereal and cereal products: a review on progress in mitigation. Food Addit. Contam., 24(S1), Sadd,P.A.; Hamlet,C.G. The Formation of Acrylamide in UK Cereal Products in: Chemistry and Safety of Acrylamide in Food, M.Friedman and D.S.Mottram, Editors, Springer, New York; Advances in Experimental Medicine and Biology, Vol. 561, pp (2005) 1 available at: