Pollen-imprinted polyurethanes for QCM allergen sensors

Molecularly imprinted polymers for detecting plant pollen were designed as artificial recognition materials for quartz crystal microbalances in the gaseous phase. Imprints of birch (diameter, 25 μm) and nettle (diameter, 15 μm) pollen can be generated by polydimethylsiloxane stamping technique as proven by atomic force microscopy. If pollen grains are able to access the cavities and thus are incorporated, the resulting sensors display Sauerbrey-like negative frequency shifts. Non-Sauerbrey behaviour can be observed as soon as pollen is prevented from entering the selective hollows: this results in grain mobility on the electrode surface leading to frequency increases. Access to the cavities is determined by the diameter ratio between pollen grains and imprints as can be revealed during cross-selectivity measurements of nettle and birch pollen imprinted layers. When the amount of pollen grains on the electrode surface exceeds the number of available imprints, the excess particles move freely, resulting in positive, non-Sauerbrey frequency shifts. Dedicated to Prof. Dr. Ernst Kenndler on the occasion of his 65th birthday.     

Purification and Characterization of Superoxide Dismutase （SOD） from Camellia Pollen

A superoxide dismutase（ SOD ） was purified to homogeneity from fresh camellia pollen by means of ammonium sulfate precipitation and column chromatography with DEAE-cellulose（ DE52 ）, Sephadex G-100 and phenyl sepharose^TM 6 Fast Flow columns. Its specific activity could reach to 4034 U/mg protein and it was determined to be Cu/ Zn-SOD according to its different sensitivities to different inhibitors. The molecular weight of the SOD and its subunit were 69500 and 34700, respectively, based on sodium dodecyl sulfate-polyacrylamide gel electrophoresis （SDS- PAGE）, which implicates that the SOD in camellia pollen is a dimmer composed of two identical subunits. The isoelectric point of the enzyme was determined to be 4. 1 by isoelectric focusing electrophoresis and the N-terminal amino acid was identified to be Gly by the DNS-Cl method. Its α-Helix was also calculated to be approximately 21.8% according to the circular dichroism（CD） spectra.     

Template Synthesis and Adsorption of Hollow Zr02 Microspheres

以油菜花粉为生物模板,通过温和易控的水浴-陈化法制备了纳/微米结构ZrO2中空微球.利用扫描电子显微镜（SEM）、红外光谱（FTIR）、X射线能谱（EDS）、X射线衍射（XRD）、比表面孔隙度分析、热分析等对所制备的产物和前驱体进行了表征,并对产物的吸附性能进行了初步的研究.结果表明,ZrO2中空微球的球壳由纳米粒子构筑并形成介孔结构.花粉模板前处理方式不同,其模板作用不同,可以获得两种不同球壳厚度、表面形貌和比表面积的ZrO2中空微球.其中＂镂空＂结构的ZrO2微球对铬黑T有良好的吸附性能.对ZrO2中空结构形成的机理进行了分析和讨论.     

不同微肥处理下油菜产量与高光谱反射率关系分析

油菜是我国第一大国产植物油来源,大田生产中需要施加适量的微肥以提高产量和品质。筛选出一种可提高油菜产量的微肥配方需要经过复杂的大田统计和产量测定,因此构建出能快速筛选微肥的模型十分重要。以高油酸油菜“帆鸣1号”为试验材料,使用地物波谱仪测定了不同微肥条件下全生育期的光谱反射率,并用乙醇提取法准确测定叶绿素含量。将光谱反射率、叶绿素含量和最终产量性状两两间进行相关性分析。产量测定表明,施加微肥可以提高油菜产量和蕾薹期时叶绿素的含量,使单株产量最高提高2%。光谱参数与叶绿素相关性分析表明,蕾薹期时叶绿素含量与光谱参数550和720 nm相关性较高,表明蕾薹期光谱参数可用于预测产量进而筛选出能提高油菜产量的微肥。叶绿素含量和产量相关性分析表明,蕾薹期时,叶绿素含量与产量相关性较高。光谱参数与产量相关分析表明,550和720 nm的光谱反射率与产量之间均呈显著负相关性。光谱参数与产量相关分析表明,550和720 nm的反射率与产量之间均呈显著负相关性。综合分析施肥量、光谱参数、产量和叶绿素变化可知,蕾薹时光谱参数550和720 nm与产量相关系数模拟的线性方程可用于微肥的筛选,线性方程分别为y=-32.362x+33.097,y=4.069 5x+35.386,y=28.849x+23.735,y=-19.023x+31.005,y=12.447x+24.586,R2均大于0.6。综合分析施肥量、光谱参数、产量和叶绿素变化,油菜生长至蕾薹期时光谱参数550和720 nm与产量相关系数模拟的线性方程R2≥0.6时的微肥配比可以使产量提升。本研究结果表明,蕾薹期光谱参数可用于预测产量进而筛选出能提高油菜产量的微肥,可增加样本量进一步检测相关性并开展后续验证。鉴于地物波谱检测技术具有过程高效,不使用化学试剂,无需对样本进行破坏性取样,成本低,该模型的建立对开展大规模高油酸油菜微肥配方的快速筛选具有重要意义,为筛选油菜微肥和促进油菜产量研究提供了理论基础。     

用荧光探针法研究茶叶花粉黄酮对氧自由基致鼠红细胞膜氧化损伤的保护作用

应用1,6-二苯基-1,3,5-已三烯(DPH)为荧光探针,超氧阴离子自由基和羟基自由基致鼠红细胞膜氧化损伤为模型,研究了茶叶花粉黄酮对鼠红细胞膜氧化损伤的影响.结果表明,超氧阴离子自由基和羟基自由基均能引起鼠红细胞膜脂质过氧化反应,脂质过氧化产物丙二醛含量显著升高,膜脂流动性下降.将鼠红细胞膜预先用茶叶花粉黄酮处理后,膜脂的MDA含量明显下降,呈现剂量与效应关系,膜脂流动性显著提高,表明茶叶花粉黄酮对超氧阴离子自由基和羟基自由基引起的鼠红细胞膜的氧化损伤有保护作用.     

Morphology‐controlled Synthesis of SiO2 Hierarchical Structures Using Pollen Grains as Templates

Rape‐, sweet potato‐ and camellia‐pollen‐templated SiO₂ hierarchical structures with species‐specific surface morphologies were prepared through a facile surface sol‐gel and succedent calcination process. Various‐morphology SiO₂ replicas with complex structures, such as about 50‐µm‐length and 20‐µm‐width high‐fidelity ellipsoidal replicas with about 5‐µm uniform meshes, about 40‐µm spheres with 2‐µm taper‐like protuberances and 15‐µm spheres with troughs, were obtained from using rape‐, sweet potato‐ and camellia pollen grains as templates, respectively. The results of N₂ adsorption‐desorption isotherms indicate that both rape‐ and camellia‐pollen‐ templated SiO₂ replicas show type IV‐like isotherm with the characteristics of mesoporosity and corresponding Brunauer‐Emmett‐Teller (BET) surface areas are 10.5 and 5.2 m²·g^?1, respectively. The Barrett‐Joyner‐Halenda (BJH) pore size distribution of the rape‐pollen‐templated SiO₂ replicas presents smaller pores centered at 3.6 nm and bigger pores centered at 8.6 nm with a broad distribution whereas the camelliapollen‐templated SiO₂ replicas display narrow BJH pore size distribution centered at 20 nm. The biofabrication procedure using ubiquitous, inexpensive and environment‐friendly pollen grains as a biotemplate holds a promise as a simple route to construct controlled complex porous oxides via simple chemical methods.     

Micro Salting-Out Assisted Matrix Solid-Phase Dispersion: A Simple and Fast Sample Preparation Method for the Analysis of Bisphenol Contaminants in Bee Pollen

In the present work, a novel sample preparation method, micro salting-out assisted matrix solid-phase dispersion (μ-SOA-MSPD), was developed for the determination of bisphenol A (BPA) and bisphenol B (BPB) contaminants in bee pollen. The proposed method was designed to combine two classical sample preparation methodologies, matrix solid-phase dispersion (MSPD) and homogenous liquid-liquid extraction (HLLE), to simplify and speed-up the preparation process. Parameters of μ-SOA-MSPD were systematically investigated, and results indicated the significant effect of salt and ACN-H₂O extractant on the signal response of analytes. In addition, excellent clean-up ability in removing matrix components was observed when primary secondary amine (PSA) sorbent was introduced into the blending operation. The developed method was fully validated, and the limits of detection for BPA and BPB were 20 μg/kg and 30 μg/kg, respectively. Average recoveries and precisions were ranged from 83.03% to 94.64% and 1.76% to 5.45%, respectively. This is the first report on the analysis of bisphenol contaminants in bee pollen sample, and also on the combination of MSPD and HLLE. The present method might provide a new strategy for simple and fast sample preparation of solid and semi-solid samples.     

基于卷积神经网络与投票机制的蒲黄炮制品近红外判别方法

蒲黄炭是由香蒲花粉炮制而成,具有止血、化瘀、通淋等多种功效,被广泛应用于临床抗血栓,创面和出血。然而蒲黄炭在炒炭过程中,常常会出现炭化过轻或者炭化过重的现象,从而出现不同炭化程度的蒲黄炭药品,主要为轻度炭化、标准炭化与重度炭化三种不同的蒲黄炭药品。由于炭化程度不同,蒲黄炭的凝血效果优劣不等,其中标准炭化的蒲黄炭药品药效最优。目前,鉴别蒲黄炭药品的方法多为人工凭借肉眼与经验进行判别。基于人工的蒲黄炭药品判别方法判别效率低,受主观因素影响大,判别结果不稳定,难以区分出标准炭化的蒲黄炭。为有效地对不同炭化程度的蒲黄炭进行识别,提出一种基于卷积神经网络与投票机制的蒲黄炮制品近红外判别方法。该方法创新性地结合深度学习与机器学习算法,有效利用卷积神经网络强大表征提取能力的同时通过投票决策提升算法模型的泛化能力与鲁棒性。首先通过近红外光谱技术获取蒲黄炭的近红外光谱,并通过卷积神经网络分别提取样本经过四种预处理方法所得到光谱图的高阶特征,并计算预测结果。按照样本准确率与损失值为四种预处理方法分配相应权重得到蒲黄炮制品预测模型。该模型将所得到的四种预测结果结合权重共同投票出样本的最终结果,从而鉴别出蒲黄炭的炭化程度。实验结果表明所提方法可以有效判别蒲黄炮制品的炭化程度。当训练集所占样本比例为80%时,预测准确率达到95.4%。所提方法与传统卷积神经网络方法、线性判别分析方法以及标准正太变量变换-线性判别分析方法相比预测准确率分别提高8.6%,4.3%和2.6%。同时,所提方法具有一定的稳定性,当训练集所占样本比例大于70%时,测试准确率高于90%;当训练集比例仅占10%时,预测准确性仍然能够达到约80%。