MASS MEASUREMENT OF SINGLE MOLECULES WITH THE STEM

扫描透射电子显微镜（STEM）可以提供低剂量， 未染色生物样本的高分辨率图像。 隔离对象 或者离散的图像特征可以被积分（在减去 底物）以确定分子量，从而提供直接联系 生物化学和电子显微镜之间的联系 STEM的准确性 分子量从10 kD时的约20%提高到100 kD时的约4 在100万D以上的单个粒子， 剂量为10埃/平方埃。 关于tRNA、核小体、 纤维蛋白原，RNA聚合酶，谷氨酰胺合成酶，30和50核糖体 亚基，丝状病毒和TMV的一致性很好， 理论预测。 平均值的标准误差可以通过以下方式减小： 平均超过100- 1000个粒子。 颗粒质量值的直方图可以 用于表征制剂的均匀性。 这种技术 已经被常规地应用于各种各样的用户/合作者， 在布鲁克海文STEM生物技术资源过去几年的样本。 我们建议继续这些富有成效的合作和服务 活动，同时完善技术的实际方面。 进一步 将试样温度降低到20 K有望减少质量损失 现在在120 K时发现低于0.25%/el Angstrom平方（比300 K时低8倍 K），允许更高的剂量和更高的准确性。 降低表面 在较低温度下的迁移预计会导致较少的再分布 在辐射损伤中的质量。 这一预测将被测试， 已知结构的对照样品（例如TMV）。 一个主要的新推力将涉及使用图像平均程序，以改善 低剂量图像信噪比。 这将允许扩展 已经在纤维蛋白原和动力蛋白上证明了结构域质量图， 允许检测小分子（1-5 kD）与复合物的结合 如谷氨酰胺合成酶和乙酰胆碱受体。 的颗粒 与球面或圆柱对称，这些程序也将允许 确定质量作为半径的函数。

The scanning transmission electron microscope (STEM) can provide low dose, high resolution images of unstained biological samples. Isolated objects or discrete image features can be integrated (after subtraction of substrate) to determine molecular weights and so provide a direct link between biochemistry and electron microscopy. The accuracy of STEM melecular weights improves from about 20% at 10 kD to about 4% at 100 kD and better than 1% above one million D for single particles imaged at a dose of 10 el/Angstroms squared. Practical results on tRNA, nucleosomes, fibrinogen, RNA polymerase, glutamine synthetase, 30s and 50s ribosomal subunits, filamentous viruses and TMV are in good agreement with theoretical predictions. Standard error of the mean can be reduced by averaging over 100-1,000 particles. Histograms of particle mass values can be used to characterize the homogeneity of the preparation. This technique has been applied routinely to a wide variety of users'/collaborators' samples over the past years at the Brookhaven STEM Biotechnology Resource. We propose to continue these productive collaborations and service activities while refining practical aspects of the technique. Further reduction of specimen temperature to 20 K is expected to reduce mass loss below 0.25%/el Angstroms squared now found at 120 K (8x less than at 300 K), permitting higher dose and higher accuracy. Reduction of surface migration at lower temperature is expected to lead to less redistribution of mass during radiation damage. This prediction will be tested using control specimens of known structure (e.g. TMV). A major new thrust will involve use of image averaging programs to improve signal to noise ratio in low dose images. This should permit extension of domainal mass mapping already demonstrated on fibrinogen and dynein to allow detection of the binding of small molecules (1-5 kD) to complexes such as glutamine synthetase and acetylcholine receptor. For particles with spherical or cylindrical symmetry, these programs will also permit determination of mass as a function of radius.