影響高溫加熱處理食品中丙烯醯胺形成與風險之研究

Study on the formation and health risk of acrylamide in high temperature processed foods

林佳儒(Jia-Ru Lin)

指導教授 ： 吳焜裕

國立臺灣大學/公共衛生學院/食品安全與健康研究所/碩士(2020年)

https://doi.org/10.6342/NTU202003765

摘要

丙烯醯胺(Acrylamide)為工業化學品，亦可於高溫處理的食品中加熱生成。丙烯醯胺在動物實驗中對大鼠及小鼠多個器官或部位引起腫瘤，且大量攝取丙烯醯胺經代謝轉化為具有化學反應性和基因毒性的環氧化物縮水甘油醯胺(Glycidamide)。但因流行病學研究尚未有具體結論，因此被國際癌症研究機構（International Agency for Research on Cancer）在1994年將它列為2A級致癌物。過去研究顯示富含澱粉之食品，例如:馬鈴薯，在高溫(120℃以上)烹調之下，會與天門冬醯胺及還原糖經梅納反應生成丙烯醯胺，因此民眾擔心每天經飲食攝食到丙烯醯胺，會對健康造成不良的影響。本研究擬針對高溫處理食品的加工過程中，影響丙烯醯胺形成的因素，如處理時間及溫度，其次食材中所含天門冬醯胺、游離胺基酸、葡萄糖、果糖含量等作系統性研究，並評估這些因素對健康風險的影響。因此首先建立形成丙烯醯胺程的化學動力學模型，以預測模擬在不同條件下丙烯醯胺形成量，經與實測值比對驗證後，研究控制或是降低丙烯醯胺含量的處理條件。
利用模式模擬在165℃下，加熱處理300秒形成的丙烯醯胺濃度為的0.0104 mmol/kg fat-free dry weight (相當於0.216 ppm)，介在文獻模擬值的0.015 mmol/kg fat-free dry weight (相當於0.312 ppm) 及實驗值 0.007 mmol/kg fat-free dry weight 之間(相當於0.145 ppm)，這些結果作為驗證，顯示可以利用此化學動力學模式模擬高溫處理食品中丙烯醯胺的含量。並根據模擬得到的丙烯醯胺含量，整合毒理資料與動物實驗數據，執行劑量反應關係利用(The Benchmark Dose Software 2.7.0.4)估算得到BMDL10為0.265 mg/kg/day，致癌係數為0.377 (mg/kg/day)-1。台灣的國家攝食資料庫在2019年調查19~65歲國人消費者的薯條薯餅食物品項之每天平均的攝食量為76.05 g/day。利用蒙地卡羅方法進行10000次抽樣，計算透過攝食薯餅薯條類暴露丙烯醯胺的致癌風險(Cancer risk)與暴露限值(Margin of exposure, MOE)。19~65歲國人丙烯醯胺消費者之致癌風險(Cancer risk)中位數為8.64×10-5，95%的信賴區間上限為4.75×10-4，男性及女性丙烯醯胺消費者致癌風險中位數則分別為7.71×10-5 及9.31×10-5；19~65歲國人、男性及女性之暴露限值中位數分別為1.15×103 、1.28×103 、1.08×103 。經由對致癌風險進行敏感度分析，影響最大的因素為暴露到丙烯醯胺的食物攝食量，與致癌風險呈正相關且影響程度占93 ％。
從製程參數對於致癌風險的關聯性，可以觀察到體重為-0.0826、攝食量為0.31、天門冬醯胺為0.2354、時間為0.2123、水分為-0.282、溫度為0.7918，除了可從攝食量改善外，從溫度著手甚為重要。
本研究目的為開發出丙烯醯胺在食品中化學動力學模型，探討影響丙烯醯胺形成的因素，並藉由模型參數改善加工製程。模擬馬鈴薯類食品在油炸過程中丙烯醯胺之產生，並針對油炸馬鈴薯類食品製造過程中產生丙烯醯胺進行健康風險評估。

關鍵字

丙烯醯胺 ； 化學動力學 ； 風險評估 ； 馬鈴薯 ； 高溫加熱處理食品

並列摘要

Acrylamide is an industrial chemical and exists in high temperature processed food. When exposed, it is metabolized into chemically reactive and genotoxic epoxides －　glycidamide. Evidence of its carcinogenicity has been shown via animal studies that observed tumors in multiple organs of rats and mice. However, evidence in epidemiological studies still remain inconclusive. Therefore, International Agency for Research on Cancer (IARC) listed acrylamide as a Class 2A carcinogen in 1994. The Maillard reaction between amino acids and carbohydrates has been recognized as the most probable process for producing acrylamide. Previous studies have reported that when starch-rich foods, such as potatoes, are processed in high temperature (above 120 °C), asparagine and reducing sugars would react to acrylamide, and partially decompose to degradation products. Thus, exposure to acrylamide through daily diet have become a common public concern. In this study, a systematic study was conducted to understand the effect of high-temperature processing on the formation of acrylamide. Food processing involves numerous factors that can affect the formation of acrylamide, such as processing time and temperature, followed by asparagine, free amino acids, glucose, fructose content, etc. The impact of these factors on health risks was evaluated. The objective of this study was to establish a chemical kinetic model of the food processing to discuss how the formation of acrylamide can be predicted, controlled and reduced.
We use the model to predict frying for 300 seconds at 165°C, the acrylamide concentration is 0.0104 mmol/kg fat-free dry weight (equivalent to 0. 216 ppm), which is between 0.015 mmol/kg fat-free dry weight (Equivalent to 0. 312 ppm) of the simulated valueand and 0.007 mmol/kg fat-free dry weight (equivalent to 0.145 ppm) of the experimental value in the paper. These results were used as verification and show that this chemical kinetic model can be used to simulate the content of acrylamide in high-temperature processed food. Based on the simulated acrylamide concentration, integrating toxicological data and animal experiment data of acrylamide, we used The Benchmark Dose Software (version 2.7.0.4) to calculate dose response relationship. The BMDL of humans was 0.265 mg/kg/day, and the CSF was 0.377 (mg /kg/day)-1. In 2019, the National Food Database of Taiwan surveyed the consumption pattern of fries and potato foods of 19 to 65 years-old consumers and estimated the average intake to be 76.05 g per day. Monte Carlo simulation with 10,000 iterations was carried out to conduct a probabilistic estimate of the cancer risk and Margin of exposure of exposure to acrylamide through ingestion of fries potato foods.
The median cancer risk of 19~65 years-old consumers was 8.64×10-5, and the upper limit of the 95% confidence interval was 4.75×10-4; the median cancer risk of 19~65 years-old male and female consumers were 7.71×10-5 and 9.31×10-5, respectively; the median MOE for of 19~65 years-old consumers, male and female were 1.15×103, 1.28×103, and 1.08×103, respectively. The sensitivity analysis of cancer risk showed that the most influential factor was the intake rate of acrylamide, which 93% positively correlated to cancer risk.
From the correlation of process parameters to cancer risk, it can be observed that body weight is -0.0826, food intake is 0.31, asparagine is 0.2354, time is 0.2123, moisture is -0.282, and temperature is 0.7918, except that food intake can be improved. It is very important to start with temperature.
The purpose of this research was to develop a chemical kinetic model of acrylamide in food, explore the formation mechanism of acrylamide, and improve the processing process by adjusting the model parameters. Simulate the production of acrylamide during the frying process of potato foods, and conduct a health risk assessment.

並列關鍵字

Acrylamide ； Chemical Kinetic Model ； Risk Assessment ； Potato ； High Temperature Processed Foods

參考文獻

Rice, J. M. (2005). The carcinogenicity of acrylamide. Mutation Research/Genetic

Toxicology and Environmental Mutagenesis, 580(1-2), 3-20.

International Agency for Research on Cancer. (1994). Some industrial chemicals. IARC

Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, 60.

Alison, R. H., Capen, C. C. and Prentice, D. (1994). "Neoplastic lesions of questionable significance to humans." Toxicologic Pathology 22(2): 179-186.