高效液相色谱法同时测定发酵液中赤藓糖醇和L-赤藓酮糖的含量

建立了采用高效液相色谱(HPLC)同时测定发酵液中底物赤藓糖醇和产物L-赤藓酮糖含量的方法。采用Lichrospher 5-NH2色谱柱(250 mm×4.6 mm),柱温30 ℃,以乙腈-水(体积比为9:1)为流动相,流速1.0 mL/min。用示差折光检测器检测赤藓糖醇,检测器温度为35 ℃。用紫外检测器在室温下检测L-赤藓酮糖,检测波长为277 nm。所得赤藓糖醇的线性范围为1.00～100.00 g/L,相关系数为0.9985,检出限为0.10 g/L,定量限为0.45 g/L;所得L-赤藓酮糖的线性范围为1.00～100.00 g/L,相关系数为0.9958,检出限为0.50 g/L,定量限为0.87 g/L;赤藓糖醇的日内和日间相对标准偏差(RSD)分别小于3.28%和5.30%, L-赤藓酮糖的日内和日间RSD分别小于2.16%和2.25%;回收率均大于99%。取不同时间的发酵液样品分别用上述方法测定,结果表明所建立的HPLC法不受发酵液中其他组分的影响,可同时测定底物赤藓糖醇和产物L-赤藓酮糖的含量。     

siPLS-LASSO的近红外特征波长选择及其应用

近红外技术广泛应用于食品、药品等生产过程和产品质量检测,具有样品无需预处理、成本低、无破坏性、测定速度快等优点。但是,全光谱数据维数高、冗余信息多,直接应用于建模会导致模型复杂性高、稳定性差等问题。siPLS是最常见的光谱数据降维方法,但是难以处理光谱数据的共线性问题。LASSO是一种相对新的数据降维方法,但在小样本应用中具有不稳定性。针对siPLS和LASSO在近红外光谱数据应用中存在的问题,提出了基于siPLS-LASSO的近红外特征波长选择方法,并将其应用于秸秆饲料蛋白固态发酵过程pH值监测。该方法首先采用siPLS算法,实现对光谱波长最佳联合子区间的优选;然后,对优选联合子区间使用LASSO算法进行特征波长选择,在此基础上建立PLS校正模型。同时,将siPLS-LASSO方法与其他传统特征波长选择方法进行了对比。结果表明：建立在siPLS-LASSO方法优选33个特征波长基础上的PLS模型预测结果更好,其预测方差(RMSEP)和相关系数(R_p)分别为0.071 1和0.980 8;所提siPLS-LASSO方法有效选取了特征波长,提高了模型预测性能。     

Ultrasonic monitoring of malolactic fermentation in red wines

The progress of malolactic fermentation in red wines has been monitored by using ultrasonic techniques. The evolution of ultrasonic velocity of a tone burst 1 MHz longitudinal wave was measured, analyzed and compared to those parameters of oenological interest obtained simultaneously by analytical methods. Semi-industrial tanks were used during measurements pretending to be in real industrial conditions. Results showed that the ultrasonic velocity mainly changes as a result of the conversion by lactic acid bacteria of malic acid into lactic acid and CO₂. Overall, the present study has demonstrated the potential of the ultrasonic technique in monitoring the malolactic fermentation process.     

发酵过程混合建模与优化控制新方法

针对当前微生物发酵过程存在因为生物传感器不具备足够的准确性和灵敏性,实验时的菌液和产物浓度等生化指标难以实时监测和控制等缺点,提出了采用量子粒子群优化算法(QPSO)优化最小二乘支持向量机(LSSVM)参数的QPSO-LSSVM混合建模新方法,并用于多粘菌素的发酵过程建模;同时,基于此模型,采用QPSO算法对pH值与溶解氧浓度Do控制轨线进行优化研究。首先,利用LSSVM进行发酵过程的建模,然后采用QPSO对LSSVM建模过程中的重要参数进行优化调整,形成QPSO-LSSVM混合建模与优化控制方法。仿真结果表明,该方法得到的模型能取得更好的预测效果,优化后的pH值与Do浓度控制轨线能够提高最终的产物浓度。该方法用于发酵过程的建模和重要参数的优化控制是可行的、有效的。     

高效液相色谱法同时测定生物质乳酸发酵液中有机酸及糖类化合物

建立了高效液相色谱(HPLC)同时测定生物质乳酸发酵液中有机酸及糖类的分析方法。使用Bio-Rad Aminex HPX-87H色谱柱,以5 mmol/L的H2SO4为流动相,在柱温55 ℃,流速0.6 mL/min条件下,采用示差折光检测器进行检测。结果表明,该方法可在17 min内实现发酵液中各种有机酸和糖类化合物等的完全分离与定量,6种有机酸和3种糖类化合物在0.15～5.19 g/L范围内的线性关系良好,回归方程的线性相关系数在0.9998以上。将该法用于米根霉发酵液的检测,两个水平的加标回收率为96.91%～103.11%,相对标准偏差(n=6)为0.81%～4.61%。该法适用于微生物发酵液中多种有机酸和糖类的快速、高效分离和定量测定。     

近红外光谱结合ELM快速检测固态发酵过程参数pH值

pH值是固态发酵过程关键参数之一,为此提出基于近红外光谱技术的秸秆蛋白饲料固态发酵过程参数pH值检测方法。利用近红外光谱系统获取140个固态发酵过程产物样本在10 000～4 000cm-1范围内的近红外光谱数据,通过酸度计测得近红外光谱预测模型的参考测量值;运用ELM算法建立pH值的预测模型,在模型建立过程中由交互验证法确定最佳主成分因子数和ELM网络隐含层节点数。试验结果显示:最佳ELM网络模型的拓扑结构为10-40-1,模型预测集相关系数(Rp)和预测均方根误差(RMSEP)分别为0.961 8和0.104 4。研究结果可为固态发酵过程参数的在线检测提供技术基础。     

普洱茶及其原料多糖分子组成及光谱学特性研究

研究了普洱茶发酵过程中多糖分子组成及光谱学特性的变化规律。结果显示,普洱茶及其原料多糖主要组分TPS1和TPS2的分子组成及光谱学特性差异显著。TPS2含有较高糖醛酸,而TPS1含有较高的中性糖和蛋白质。TPS1和TPS2均由半乳糖(Gal)、阿拉伯糖(Arb)、甘露糖(Man)、葡萄糖(Glu)、木糖(Xyl)、鼠李糖(Rha)六种单糖组成,其分子摩尔比分别为23.6∶5.9∶24.2∶1.1∶1.8∶3.2和26.9∶3.2∶19.3∶5.5∶1.3∶2.7。TPS2和TPS1的重均分子量分别为1.68×104和1.21×104道尔顿。TPS1和TPS2水溶液在200～400 nm之间无特征吸收峰。红外光谱图显示,TPS1和TPS2的信息基本相同,都是含有吡喃环的多糖。在三维AFM图中,TPS1形成的聚集体高度约为4 nm, 长宽约为0.2～0.4 μm,TPS2形成的聚集体高度约为40 nm,长宽约为0.5～0.8 μm。SEM图片显示,TPS1呈表面光滑的鳞片状聚集体,TPS2呈表面粗糙的片状聚集体。引起普洱茶及其原料多糖分子组成及光谱学性质发生变化的动力主要为微生物作用和湿热作用。     

生物制造2,3-丁二醇:回顾与展望

2,3-丁二醇是生物制造产品体系中一种重要的精细化工原料和潜在平台化合物,广泛应用于材料、医药、食品及航空航天等领域。利用生物质可再生资源为原料生产2,3-丁二醇符合当前发展低碳经济的国家需求。本文回顾了生物制造2,3-丁二醇的研究历史,分析了微生物合成2,3-丁二醇的代谢机理,总结了提高生物制造2,3-丁二醇经济性的有效途径,包括廉价原料的替代、菌株选育与遗传改造和发酵过程控制等,并对2,3-丁二醇的各种下游分离过程进行了对比分析;指出今后研究重点应着眼于努力提高生物质的利用效率,同时实现高效的2,3-丁二醇生物转化两方面,并在此基础上开发2,3-丁二醇的系列高值衍生物,以进一步拓展其应用领域。     

Effect of agitation and aeration on production of hexokinase by Saccharomyces cerevisiae

A batch culture of Saccharomyces cerevisiae for the production of hexokinase was carried out in a 5-L fermentor containing 3 L of culture medium, which was in oculated with cell suspension (about 0.7 g/L), and left ferm entingat 35°C and pH 4.0. The aeration and agitation were adjusted to attain k _La values of 15, 60, 135, and 230 h⁻¹. The highest hexokinase productivity (754.6 U/[L h]) and substrate-cell conversion yield (0.21 g/g) occurred for a k _La of 60 h⁻¹. Moreover, the formation of hexokinase and cell growth are coupled events, which is in accordance with the constitutive character of this enzyme. Hexokinase formation for k _La>60 h⁻¹ was not enhanced probably owing to saturation of the respiratory pathway by oxygen.     

Modeling of saccharide utilization in primary beer fermentation with yeasts immobilized in calcium alginate

Immobilized beer fermentation was studied using an industrial bottom-fermenting yeast strain Saccharomyces cerevisiae. The yeast cells were immobilized in 2.5% calcium alginate gel and used for brewing in a five-vessel cascade reactor. The fermentation was performed at 15°C at various flow rates. A nonstructured mathematical model was developed to simulate the performance of continuous primary fermentation of lager beer. The model was based on the following variables: maltose, maltotriose, glucose, fructose, ethanol, and cell concentration. Experimental values of these variables were determined in samples taken at regular intervals. For experimental data fitting a nonlinear regression was used. Substrate consumption was characterized by specific substrate consumption rate and saturation constant. The values of these two parameters were optimized for all four substrates. Inhibition effects of substrates and product were analyzed using various inhibition patterns. Only the inhibition effect of maltose on maltose consumption was clearly identified. A good-fitting relationship for maltose inhibition was found, and inhibition constants were calculated.