Hays John B, Hoffman Peter D, Wang Huixian
Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, 97331-7301, USA.
DNA Repair (Amst). 2005 Dec 8;4(12):1463-74. doi: 10.1016/j.dnarep.2005.09.002. Epub 2005 Oct 5.
Evolutionarily-conserved mismatch-repair (MMR) systems correct all or almost all base-mismatch errors from DNA replication via excision-resynthesis pathways, and respond to many different DNA lesions. Consideration of DNA polymerase error rates and possible consequences of excess gratuitous excision of perfectly paired (homoduplex) DNA in vivo suggests that MMR needs to discriminate against homoduplex DNA by three to six orders of magnitude. However, numerous binding studies using MMR base-mispair-recognition proteins, bacterial MutS or eukaryotic MSH2.MSH6 (MutSalpha), have typically shown discrimination factors between mismatched and homoduplex DNA to be 5-30, depending on the binding conditions, the particular mismatches, and the DNA-sequence contexts. Thus, downstream post-binding steps must increase MMR discrimination without interfering with the versatility needed to recognize a large variety of base-mismatches and lesions. We use a complex but highly MMR-active model system, human nuclear extracts mixed with plasmid substrates containing specific mismatches and defined nicks 0.15 kbp away, to measure the earliest quantifiable committed step in mismatch correction, initiation of mismatch-provoked 3'-5' excision at the nicks. We compared these results to binding of purified MutSalpha to synthetic oligoduplexes containing the same mismatches in the same sequence contexts, under conditions very similar to those prevailing in the nuclear extracts. Discrimination against homoduplex DNA, only two-to five-fold in the binding studies, increased to 60- to 230-fold or more for excision initiation, depending on the particular mismatches. Remarkably, the mismatch-preference order for excision initiation was substantially altered from the order for hMutSalpha binding. This suggests that post-binding steps not only strongly discriminate against homoduplex DNA, but do so by mechanisms not tightly constrained by initial binding preferences. Pairs of homoduplexes (40, 50, and 70 bp) prepared from synthetic oligomers or cut out of plasmids showed virtually identical hMutSalpha binding affinities, suggesting that high hMutSalpha binding to homoduplex DNA is not the result of misincorporations or lesions introduced during chemical synthesis. Intrinsic affinities of MutS homologs for perfectly paired DNA may help these proteins efficiently position themselves to carry out subsequent mismatch-specific steps in MMR pathways.
进化上保守的错配修复(MMR)系统通过切除-再合成途径纠正DNA复制中所有或几乎所有的碱基错配错误,并对许多不同的DNA损伤作出反应。考虑到DNA聚合酶的错误率以及体内完全配对(同型双链)DNA过度无端切除的可能后果,表明MMR需要对同型双链DNA的识别能力比对错配DNA的识别能力低三到六个数量级。然而,大量使用MMR碱基错配识别蛋白、细菌MutS或真核生物MSH2.MSH6(MutSα)的结合研究通常表明,错配DNA和同型双链DNA之间的识别因子为5-30,这取决于结合条件、特定的错配以及DNA序列背景。因此,结合后的下游步骤必须提高MMR的识别能力,同时又不干扰识别多种碱基错配和损伤所需的通用性。我们使用一个复杂但具有高度MMR活性的模型系统,即将人核提取物与含有特定错配和0.15kbp外特定切口的质粒底物混合,来测量错配修复中最早可量化的关键步骤,即在切口处引发错配诱导的3'-5'切除。我们将这些结果与纯化的MutSα在与核提取物中非常相似的条件下与含有相同错配且序列背景相同的合成寡双链体的结合情况进行了比较。在结合研究中对同型双链DNA的识别能力仅为两到五倍,而对于切除起始,根据特定的错配情况,识别能力增加到60到230倍或更高。值得注意的是,切除起始的错配偏好顺序与hMutSα结合的顺序有很大不同。这表明结合后的步骤不仅强烈区分同型双链DNA,而且其作用机制不受初始结合偏好的严格限制。由合成寡聚物制备或从质粒中切出的同型双链体对(40、50和70bp)显示出几乎相同的hMutSα结合亲和力,这表明hMutSα与同型双链DNA的高结合不是化学合成过程中引入的错掺入或损伤的结果。MutS同源物对完全配对DNA的内在亲和力可能有助于这些蛋白质有效地定位自身,以在MMR途径中执行后续的错配特异性步骤。