Schwartz R, Curran J F
Department of Biology, Wake Forest University, PO Box 7325, Winston-Salem, NC 27109, USA.
Nucleic Acids Res. 1997 May 15;25(10):2005-11. doi: 10.1093/nar/25.10.2005.
Others have recently shown that the UUU phenylalanine codon is highly frameshift-prone in the 3'(rightward) direction at pyrimidine 3'contexts. Here, several approaches are used to analyze frameshifting at such sites. The four permutations of the UUU/C (phenylalanine) and CGG/U (arginine) codon pairs were examined because they vary greatly in their expected frameshifting tendencies. Furthermore, these synonymous sites allow direct tests of the idea that codon usage can control frameshifting. Frameshifting was measured for these dicodons embedded within each of two broader contexts: the Escherichia coli prfB (RF2 gene) programmed frameshift site and a 'normal' message site. The principal difference between these contexts is that the programmed frameshift contains a purine-rich sequence upstream of the slippery site that can base pair with the 3'end of 16 S rRNA (the anti-Shine-Dalgarno) to enhance frameshifting. In both contexts frameshift frequencies are highest if the slippery tRNAPhe is capable of stable base pairing in the shifted reading frame. This requirement is less stringent in the RF2 context, as if the Shine-Dalgarno interaction can help stabilize a quasi-stable rephased tRNA:message complex. It was previously shown that frameshifting in RF2 occurs more frequently if the codon 3'to the slippery site is read by a rare tRNA. Consistent with that earlier work, in the RF2 context frameshifting occurs substantially more frequently if the arginine codon is CGG, which is read by a rare tRNA. In contrast, in the 'normal' context frameshifting is only slightly greater at CGG than at CGU. It is suggested that the Shine-Dalgarno-like interaction elevates frameshifting specifically during the pause prior to translation of the second codon, which makes frameshifting exquisitely sensitive to the rate of translation of that codon. In both contexts frameshifting increases in a mutant strain that fails to modify tRNA base A37, which is 3'of the anticodon. Thus, those base modifications may limit frameshifting at UUU codons. Finally, statistical analyses show that UUU Ynn dicodons are extremely rare in E.coli genes that have highly biased codon usage.
其他人最近表明，在嘧啶3'上下文环境中，UUU苯丙氨酸密码子在3'（向右）方向极易发生移码。在此，我们采用了几种方法来分析此类位点的移码情况。研究了UUU/C（苯丙氨酸）和CGG/U（精氨酸）密码子对的四种排列组合，因为它们在预期的移码倾向方面差异很大。此外，这些同义位点可以直接检验密码子使用能够控制移码这一观点。对嵌入在两种更广泛上下文环境中的这些双密码子进行了移码测量：大肠杆菌prfB（RF2基因）编程移码位点和一个“正常”的信息位点。这两种上下文环境的主要区别在于，编程移码在滑码位点上游包含一个富含嘌呤的序列，该序列可与16 S rRNA的3'端（反Shine-Dalgarno序列）进行碱基配对，以增强移码。在这两种上下文环境中，如果滑码的苯丙氨酰tRNA能够在移位后的阅读框中进行稳定的碱基配对，移码频率最高。在RF2上下文环境中，这一要求没那么严格，就好像Shine-Dalgarno相互作用能够帮助稳定一个准稳定的重新定相的tRNA：信息复合物。先前的研究表明，如果滑码位点下游的密码子由稀有tRNA读取，RF2中的移码会更频繁发生。与早期的这项研究一致，在RF2上下文环境中，如果精氨酸密码子是CGG（由稀有tRNA读取），移码发生的频率会显著更高。相比之下，在“正常”上下文环境中，CGG处的移码仅比CGU处略高。有人提出，类似Shine-Dalgarno的相互作用在翻译第二个密码子之前的暂停期间会特异性地提高移码，这使得移码对该密码子的翻译速度极为敏感。在这两种上下文环境中，在一个无法修饰位于反密码子3'端的tRNA碱基A37的突变菌株中，移码都会增加。因此，这些碱基修饰可能会限制UUU密码子处的移码。最后，统计分析表明，在密码子使用高度偏向的大肠杆菌基因中，UUU Ynn双密码子极其罕见。
