泌尿道致病性大腸桿菌於臺灣市售牛肉之盛行率及其於生牛肉與熟牛肉之生長預測模型

根據流行病學研究統計，臺灣女性得到泌尿道感染 (UTIs) 的發生率為每十萬人口250.94人，而全世界每年則有1.3-1.75億的人受到感染，其中有65-75%的感染源來自於泌尿道致病性大腸桿菌 (UPEC)，由於復發率高且具有抗藥性，進而造成嚴重的社會負擔，於美國一年的經濟損失將近35億美元。許多證據顯示UPEC是一種新興的食源性病原菌，而肉品是UPEC最有可能的貯存庫，其中牛肉為了保持口感通常不會完全煮熟，甚至有許多餐廳提供生牛肉料理，若在食物鏈中受到污染可能會使消費者暴露於風險中。本實驗的目的即為研究臺灣市場中牛肉產品受到UPEC感染的盛行率，並進一步建立數學模型預測UPEC在牛肉中的生長行為。從傳統市場及超市的牛肉產品中分離出E. coli並使用聚合酶鏈反應 (PCR) 進行UPEC鑑定、演化型分類及毒力因子檢測。生牛肉中的帶有UPEC的肉品盛行率為25%，而由肉品所單離的大腸桿菌中，UPEC菌株之盛行率為13.9%，且即食牛肉也檢測出UPEC的存在。本實驗分別接種2-3 log CFU/g的UPEC於生牛肩肉與熟牛肩肉上，在4、10、20、25、35℃的等溫條件下建立預測模型，結果顯示使用No lag phase以及Huang square root model較佳。進一步使用生牛肩肉、熟牛肩肉、生牛絞肉、熟牛絞肉進行30℃的獨立實驗以驗證模型可靠度，在相對應的熟度與牛肩肉下能夠可靠的進行預測 (RMSE, 0.37-0.45 log CFU/g; adjusted R2, 0.96-0.97; pRE, 0.88)；當使用脂肪含量不同的牛絞肉時預測誤差會增加 (RMSE, 0.98-1.37 log CFU/g; adjusted R2, 0.93-0.96; pRE, 0.60-0.80)；若模型與基質生熟不同則無法可靠地進行預測 (RMSE, 1.35-1.83 log CFU/g; adjusted R2, 0.43-0.86; pRE, 0.38-0.63)。本實驗建立的模型可應用於牛肉產品的品質管控，並提供量化數據以建立定量微生物風險評估，促進食品安全。

關鍵字

泌尿道致病性大腸桿菌；預測微生物學；牛肉；生長模型；食品安全；泌尿道感染

並列摘要

According to epidemiological studies, the incidence rate of urinary tract infections (UTIs) in females in Taiwan is 250.94 per 100,000 people, while 130-175 million people are infected in the world each year. Uropathogenic Escherichia coli (UPEC) accounts for 65-75% of UTIs, causing a huge burden due to high recurrence rates and antibiotic resistance and costing nearly 3.5 billion a year in the United States. Much evidence suggests that UPEC is an emerging foodborne pathogen, and meats are the most likely reservoirs. Since the beef is preferred not fully cooked to keep the tenderness for taste, many restaurants even serve raw beef dishes. Contamination of raw beef with UPEC in the food chain may expose consumers to the risk of foodborne UTIs. The first objective of this study was to investigate the prevalence of UPEC in beef products in Taiwan. The secondary objective was to establish a mathematical model to predict the growth behavior of UPEC in beef. E. coli was isolated from beef products sampling from traditional markets and supermarkets. The isolates were analyzed by the polymerase chain reaction (PCR) method for UPEC identification, phylogroup classification, and virulence genotyping. The prevalence of UPEC-contaminated raw beef meats was 25%, and UPEC in E. coli isolates was 13.9%. More importantly, UPEC had also been detected in ready-to-eat beef. To obtain the growth curves, UPEC was inoculated on raw or cooked beef chuck with 2-3 log CFU/g initial concentration and stored under the isothermal conditions at 4, 10, 20, 25, and 35 ℃. The growth curves were fitted with primary and secondary models in IPMP2013 to establish the predictive models for the growth of UPEC in raw and cooked beef. The results showed that the combination of the no-lag phase and Huang square root models was better than the others. Further independent experiments at 30 °C were conducted using raw or cooked beef chuck or ground beef to verify the reliability of the developed models. The validation results showed that UPEC growth in raw and cooked chuck could be reliably predicted by the raw (RMSE, 0.37 log CFU/g; adjusted R2, 0.97; pRE, 0.88) and cooked (RMSE, 0.45 log CFU/g; adjusted R2, 0.96; pRE, 0.88) beef models, respectively. The predictions variations increased when the beef type changed to raw (RMSE, 1.37 log CFU/g; adjusted R2, 0.93; pRE, 0.60) or cooked (RMSE, 0.98 log CFU/g; adjusted R2, 0.96; pRE, 0.80) ground beef. However, the raw beef model did not predict UPEC growth in cooked chuck or ground beef efficiently (RMSE, 1.46–1.76 log CFU/g; adjusted R2, 0.54–0.67; pRE, 0.38–0.60). Similarly, the cooked beef model could not be reliably applied to predicting the growth in raw and raw ground beef (RMSE, 1.35–1.83 log CFU/g; adjusted R2, 0.66–0.86; pRE, 0.60–0.63). The model established in this experiment can be applied to the quality control of beef products and provides quantitative data to build quantitative microbial risk assessment and promote food safety.

並列關鍵字

uropathogenic E. coli ； predictive microbiology ； beef ； growth models ； food safety ； urinary tract infections

參考文獻

Ahmed, O. B., Asghar, A. H., Elhassan, M. M. (2014). Comparison of three DNA extraction methods for polymerase chain reaction (PCR) analysis of bacterial genomic DNA. African Journal of Microbiology Research, 8(6), 598-602. https://doi.org/10.5897/AJMR2013.6459

Google Scholar

Allocati, N., Masulli, M., Alexeyev, M. F., Di Ilio, C. (2013, Dec). Escherichia coli in Europe: An Overview. International Journal of Environmental Research and Public Health, 10(12), 6235-6254. https://doi.org/10.3390/ijerph10126235

Google Scholar

Awati, R. (2022). Definition extrapolation and interpolation. Retrieved Jul. 31, 2022 from https://www.techtarget.com/whatis/definition/extrapolation-and-interpolation

Google Scholar

Baranyi, J., Roberts, T. A. (1995, Jul). Mathematics of predictive food microbiology. International Journal of Food Microbiology, 26(2), 199-218. https://doi.org/10.1016/0168-1605(94)00121-l

Google Scholar

Baranyi, J., Tamplin, M. L. (2004). ComBase: a common database on microbial responses to food environments. Journal of Food Protection, 67(9), 1967-1971. https://doi.org/10.4315/0362-028x-67.9.1967.

Google Scholar

主題瀏覽