【作者】 张书文；

【导师】 吕加平；

【作者基本信息】 中国农业科学院 ， 农产品质量与食物安全， 2012， 博士

【摘要】 引起UHT乳发生质量缺陷的因素有很多，但来源于原料乳中的耐热酶是最主要因素。本课题对原料乳中的耐热蛋白酶在加工过程中的变化规律其对UHT乳品质的影响进行了研究，并构建了以原料乳品质来预测UHT乳货架期的数学模型。本论文的主要结果如下：采用硫酸铵沉淀、DEAE-Sepharose F.F.离子交换层析、Sephacryl S-100HR凝胶过滤层析方法对从原料乳中分离出一株Pseudomanas fluorescens BJ-10胞外蛋白酶进行纯化及特性研究。纯化后蛋白酶的分子量为47kDa，经纯化后蛋白酶比活提高了近61.38倍；最适温度和pH分别为30℃和7.0；二硫苏糖醇对蛋白酶活性具有一定的抑制作用，Fe2+可以促进该酶活性的提高；该酶具有较强耐热性，原酶液经130℃热处理3min后，残留酶活仍超过原酶液的47.67%；该蛋白酶氨基酸组成中甘氨酸、丙氨酸和谷氨酸的含量占明显优势，其中甘氨酸含量最高，摩尔分数高达42%。嗜冷菌在冷藏过程中仍可大量生长繁殖。Pseudomanas fluorescens BJ-10在牛乳中的生长速度及产酶活性明显超过营养肉汤培养基；6℃贮藏，当嗜冷菌数超过109、酶活性超过0.775U/mL时，蛋白水解开始明显加快；28℃培养，当嗜冷菌数超过108、酶活性达到6.49U/mL时，κ-casein被完全水解。嗜冷菌蛋白酶对酪蛋白的水解次序依次为κ-casein> β-casein> αs-casein。乳清蛋白对其不敏感。纤溶酶（PL）水平与奶牛品种、泌乳期、挤奶时间及饲养环境等因素紧密相关。荷斯坦牛乳中PL最高，为7.81U/mL；水牛乳中，摩拉、尼里拉菲、一代杂交、高代杂交乳中PL分别为6.49、5.9、6.46、6.07U/mL；牦牛乳中PL为5.7U/mL；娟姗牛乳中PL最低，仅为4.6U/mL。2、3胎原料乳中PL高于1、4胎，但不同胎次原料乳中PL差异不显著（p>0.05）。牧场的原料乳中PL要低于养殖小区。原料乳中PL随着热处理温度的升高而逐渐下降，巴氏杀菌（75℃、15s）可以使原料乳中PL下降25%左右。半胱氨酸对原料乳中PL具有一定的抑制作用。纤溶酶、纤溶酶原（PG）、纤溶酶原激活剂（PA）主要附着在酪蛋白胶粒上；PL主要水解β-casein、as2-casein、as1-casein，不水解κ-casein、β乳球蛋白和a-乳白蛋白；原料乳中PG是PL的4倍。嗜冷菌蛋白酶对PG具有一定的激活效果，但嗜冷菌蛋白酶和尿激酶联合作用于PG的激活效果比尿激酶单独的激活效果更加明显。嗜冷菌蛋白酶在牛乳中和缓冲液中，充当的是PA的角色，将PG转化成PL。根据原料乳品质建立了UHT乳货架期数学模型，得到UHT乳货架期（y）与SCC（x₁）、嗜冷菌数（x₂）、贮藏温度（x3）以及耐热酶添加量（x4）四因素在编码空间的回归方程为：y=225.069167＋0.324417x1－0.026167x2－8.776042x3－5.722143x4－0.000983x1×x1－0.000600_x2×x1－0.001083x2×x2－0.004688x3×x1＋0.009375x3×x2＋0.111979x3×x3＋0.004429x4×x1＋0.005143x4×x2＋0.054464x4×x3＋0.034422x4×x4。F回=8.87>F0.01（14,15）=4.94，方程回归极显著，该方程可用于设计范围内的预测。从因子分析可知嗜冷菌蛋白酶添加量（x₄）和贮藏温度（x₃）对UHT乳货架期的影响极显著（p<0.01）；而体细胞数（x₁）和嗜冷菌数（x₂）对货架期的影响差异显著不显著（p>0.05），这和理想的结果存在一定偏差，主要和试验过程中难以精确控制有关。从本文结果可以看出，耐热蛋白酶对原料乳及UHT乳品质的影响较大，在实际生产中对其控制具有一定的难度。因此，实际生产加工中要加强原料乳卫生管理，尽量降低原料乳中的微生物数量及其耐热酶活力、使用健康奶牛所产原料乳，避免由耐热酶引起产品变质情况的发生。更多还原

【Abstract】 Heat-stable protease can not be completely inactivated during UHT sterilization. The activities ofprotease that survive the heat treatments may cause some quality defects.The objective of this paper wasto carry out comprehensive research in the field of effect of processing on proteases and effect ofheat-stable proteases on the quality of UHT milk. The main results of this paper are as follows:Crude protease from Pseudomanas fluorescens BJ-10was purified by using ammonium sulfatefractionation, ion-exchange, and gel filtration chromatography. The purified enzyme was a monomerwith a molecular weight of47kD by SDS-PAGE. The specific activity of the enzyme increased61.38-fold. The optimum pH and temperature was7.0and30℃, respectively. The purified protease waspartially inhibited by DL-dithiothreitol, with little increase upon Fe2+addition. The protease showedtypical heat-stable behavior. After treatment at130℃for3min, more than47.67%activity wasremained. The purified protease had comparatively high content of glycine, alanine and glutamic acid.Glycine content was the highest, mole fraction up to42%.Growth rate and enzyme production of Pseudomanas fluorescens BJ-10in milk was significantlymore than in nutrient broth medium.When the counts exceeding109cfu/mL, the enzyme activity morethan0.775U/mL at6°C cultivation, milk protein began to be hydrolyzed. When the counts exceeding108cfu/mL, the enzyme activity more than6.49U/mL at28°C cultivation, κ-casein was completelyhydrolyzed. The protease preferentially degrade caseins in the following order κ-casein> β-casein>α-casein and have low activity on whey proteins.Plasmin activity were closely related to the breeds, stage of lactation, feeding and environmentalfactors. Holstein milk showed a high content (7.81U/mL) of plasmin in comparison with milk fromother breeds. The plasmin of buffalo milk such as Murrah, Nili, crossbreed F1, multi-crossbreed FHmilkwere6.49,5.9,6.46,6.07U/mL, respectively. The plasmin activity of yak milk was5.7U/mL. Theplasmin activity of Jersey milk was lowest (4.6U/mL) in the measurement of breeds. Plasmin activity ofHolstein milk was higher in2,3parity than in1,4parity, but the difference was not significantly (p>0.05).Plasmin activity in the Pasture was lower than in the farming community. Upon pasteurization of milk at75°C for15s, plasmin decreased by about25%. Cysteine could inhibited plasmin activity in raw milk.Plasmin (PL), plasminogen (PG), plasminogen activator (PA), which were associated with caseinmicelle. Plasmin hydrolyzes β-, αs2-, and αs1-caseins, but has little or no activity on the whey proteinsβ-LG and α-LA. In fresh milk, PG was the predominant form, where its concentration was4times thatof PL. Pseudomanas fluorescens BJ-10protease had activation effect on PG, but effect of BJ-10proteaseand urokinase (uPA) on PG activation was better than urokinase activate PG alone. Results of this studyconfirmed the interaction of the BJ-10protease with the plasmin system by acting as a PA.The objective of this study was to establish the forecasting model for predicting the shelf life ofUHT milk. The model base on somatic cell count, the number of psychrophilic bacteria, heat-stableprotease activity of raw milk and preservation temperature of UHT milk. Through the quadratic orthogonal rotatable design and analysis, we obtained regression equation:y=225.069167＋0.324417x1－0.026167x2－8.776042x3－5.722143x4－0.000983x1×x1－0.000600x2×x1－0.001083x2×x2－0.004688x3×x1＋0.009375x3×x2＋0.111979x3×x3＋0.004429x4×x1＋0.005143x4×x2＋0.054464x4×x3＋0.034422x4×x4. The results showed that regression equation was significant. It canbe used within the design range forecast. Analysis of variance of showed main factors have extremelysignificant impact on shelf life. Protease addition (x4) and storage temperature (x3) had siginificanteffect on UHT milk shelf life (p-<0.01); while SCC (x1) and psychrophile counts (x2) on the shelf lifewas not significant (p>0.05), there was a certain deviation with desired results, mainly due to thedifficult to control the whole experiment precisely.It can be concluded that heat stable proteases had a greater impact on the quality of UHT milk, itwas difficult to control it in the actual production. So, use of high-quality raw milk was of utmostimportance for achieving a long shelf-life of UHT milk. Storage of raw milk at low temperature (≤4°C)for a minimum period of time (≤48hours) minimizes growth of psychrotrophic bacteria and,consequently, the amount of extracellular bacterial proteinases produced in the milk before heattreatment. Furthermore, the use of milk with a low somatic cell count ensures a minimum level ofplasmin and plasminogen in the milk.更多还原