Muramoto K, Makishima S, Aizawa S I, Macnab R M
Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520-8114, USA.
J Mol Biol. 1998 Apr 10;277(4):871-82. doi: 10.1006/jmbi.1998.1659.
Frameshift mutations in the fliK gene of Salmonella result in abnormal elongation of the hook and the failure to assemble filament (polyhook phenotype). Second-site suppressor mutations restore filament assembly, but the cells often remain defective in hook-length control (polyhook-filament phenotype). Where the suppressor mutations are intragenic, the second mutation restores the original frame, generating a region of frameshifted sequence, but restoring the natural C terminus. Some of these frameshifted sequences contain a UGA (opal) termination codon. These cells have few flagella and swarm poorly. We suspected that readthrough of UGA by tRNATrp might be the reason for the partial function. When the UGA codon was changed to the Trp codon UGG, flagellar assembly and function were restored to wild-type levels. Conversely, underexpression of the wild-type fliK gene, achieved by changing the sole Trp codon in the sequence (Trp271) to UGA, decreased both the number of flagella and the ability to swarm. These results validate the readthrough hypothesis and indicate that low levels of FliK sustain some degree of flagellation and motility. At low levels of FliK, most flagella had polyhooks. With increasing amounts, the morphology progressively changed to polyhook-filament, and eventually to wild-type hook-filament. When FliK was overproduced, the hook length was slightly shorter (46(+/-7) nm) than that of the wild-type strain (55(+/-9) nm). FliK levels were measured by immunoblotting. Wild-type levels were about 40 to 80 molecules/cell. FliK synthesized by UGA readthrough could be detected when overproduced from plasmid fliK-W271opal, and the levels indicated a probability of readthrough of 0.002 to 0.01. This value was used to estimate the cellular level of underexpressed FliK, which could partly restore function to a fliK mutant, at about 0.07 to 0.8 molecule/cell. These results suggest that FliK does not form a large structure in the cytoplasm and may function as a regulatory protein for protein export. A model for hook-length control is presented that involves feedback from the assembly point to the export apparatus.
沙门氏菌fliK基因中的移码突变会导致钩状结构异常延长，且无法组装鞭毛丝（多钩状表型）。第二位点抑制突变可恢复鞭毛丝的组装，但细胞在钩状结构长度控制方面通常仍存在缺陷（多钩状 - 鞭毛丝表型）。当抑制突变是基因内的时，第二个突变会恢复原来的读码框，产生一段移码序列区域，但会恢复天然的C末端。这些移码序列中的一些含有UGA（乳白）终止密码子。这些细胞的鞭毛很少，群游能力很差。我们怀疑tRNATrp对UGA的通读可能是部分功能得以实现的原因。当UGA密码子被改为色氨酸密码子UGG时，鞭毛组装和功能恢复到野生型水平。相反，通过将序列中唯一的色氨酸密码子（Trp271）改为UGA来实现野生型fliK基因的低表达，会减少鞭毛数量并降低群游能力。这些结果验证了通读假说，并表明低水平的FliK能维持一定程度的鞭毛形成和运动能力。在低水平的FliK情况下，大多数鞭毛具有多钩状结构。随着FliK数量的增加，形态逐渐变为多钩状 - 鞭毛丝，最终变为野生型钩状 - 鞭毛丝。当FliK过量表达时，钩状结构的长度比野生型菌株（55（±9）nm）略短（46（±7）nm）。通过免疫印迹法测量FliK水平。野生型水平约为40至80个分子/细胞。当从质粒fliK - W271opal过量表达时，可以检测到通过UGA通读合成的FliK，其水平表明通读概率为0.002至0.01。该值用于估计低表达的FliK的细胞水平，其可部分恢复fliK突变体的功能，约为0.07至0.8个分子/细胞。这些结果表明，FliK在细胞质中不形成大的结构，可能作为一种蛋白质输出的调节蛋白发挥作用。本文提出了一种钩状结构长度控制模型，该模型涉及从组装点到输出装置的反馈。
