Mechanisms of Chromosome Maintenance in Bacteria

无论何种生物体，基因组的稳定性要求整个DNA在每个细胞周期内复制一次且仅一次。即使在大肠杆菌这样一个研究得很好的系统中，保证复制启动和防止过早重新启动的复杂调控机制仍在解开。霍乱弧菌有两条染色体。染色体I的复制起源与大肠杆菌相似，而染色体II的复制起源与质粒起源相似，质粒具有重复的起始特异性启动物结合位点（在质粒中称为iteron）。我们已经确定，尽管两种弧菌起源具有不同的特征，但每细胞周期一次的复制需要两种起源在其GATC位点的腺嘌呤残基上完全甲基化。对于I号染色体，就像大肠杆菌的oriC一样，甲基化对复制起始并不重要，但对控制复制起始频率是必需的。然而，对于染色体II，甲基化是启动子与起始iteron结合的额外需要，这是启动过程本身的基本功能。虽然甲基化被广泛用于控制许多DNA交易，但它在介导起始物相互作用中的作用是一个前所未有的发现。像质粒一样，染色体II也有一个专门的抑制复制起始的位点。尽管染色体II的整个起源区域具有许多iteron型质粒起源的特征，但它的功能与染色体起源相似，因为它将复制限制在每个细胞周期一次和细胞周期的特定阶段。这两种特征都不是质粒复制的标志。我们已经发现，虽然一个典型的质粒位点仅由iterons组成，但染色体II位点具有第二类位点。迭代子需要这些位点完全甲基化才能与引发子结合。另一种类型也与引发剂结合，但不涉及甲基化。由于DNA的甲基化状态在复制周期中发生变化，迭代子作为控制位点的有效性可能依赖于细胞周期。在任何情况下，典型迭代子的控制活动似乎是相当温和的。它们主要负责抑制其他更强大的甲基化无关位点的活性。在这种情况下，在复制之后，DNA变得半甲基化，使得iteron无法与甲基化无关的位点相互作用；后者在防止过早重新启动方面仍然完全有效。当迭代子的完全甲基化被恢复时，它们可以抑制甲基化无关位点的抑制作用，从而为下一个细胞周期的复制起始奠定基础。看来，II号染色体的质粒型起源取决于甲基化，以启动复制和其控制的染色体特征。多染色体细菌（不像大肠杆菌和枯草芽孢杆菌等被充分研究的细菌）可能为研究不同染色体的协调复制和分离机制提供了机会。已获得霍乱弧菌染色体间协调的初步迹象。我们已经能够找到可以选择性地阻止其中一条染色体复制的条件。阻止染色体I的复制似乎可以阻止/延迟染色体II的复制，但反之则不然。2号染色体比1号染色体小，这种延迟可能使两条染色体同时完成复制，这可能有利于它们的分离与细胞分裂的协调。延迟的机制还有待研究。与我们对真核生物中染色体如何分离的理解相比，我们对细菌中染色体如何分离的了解要少得多。直到最近，分离研究主要是在质粒中进行的，在质粒中可以找到专门用于质粒分裂的基因（par基因）。质粒par基因的同源物现已在大多数细菌（包括霍乱弧菌）的复制起源附近被鉴定出来。弧菌的两条染色体都有自己的par基因。我们已经成功地删除了1号染色体上的par基因，而没有造成太多的分离缺陷。相反，两个par基因中的一个的缺失促进了复制。在枯草芽孢杆菌中也有类似的发现。枯草芽孢杆菌和霍乱弧菌这两种细菌在10亿多年前就已经分化，但它们都保留了par基因，并将它们用于相似的目的。par基因在复制起点附近的广泛存在表明这些基因在连接复制和分离中起着普遍的作用。par基因如何促进复制目前还在研究中。利用细菌和酵母的双杂交系统，我们正试图识别可能与Par蛋白相互作用的蛋白质，这可能为它们的作用机制提供线索。细胞分裂必须等待染色体复制完成和两个姐妹染色体运动到相反的细胞半。细菌中细胞分裂的时间控制在很大程度上是未知的。最近，在大肠杆菌和枯草芽孢杆菌中发现了一些参与感知生长培养基中葡萄糖浓度的基因来调节细胞大小。这些基因的突变会使细胞变小30%，但不会改变它们的生长速度。在一项合作研究中，我们正在确定这些较小细胞变异的复制起始时间（起始年龄）。起始年龄似乎取决于所讨论的细菌。在枯草芽孢杆菌中，突变体的年龄保持不变，这意味着，在起始时，突变体的细胞大小比野生型小。然而，在大肠杆菌中，起始被延迟到突变体达到野生型起始的大小。起始的延迟由复制延伸率的增加来补偿，从而使复制周期按时完成。限速成分被认为是这两种细菌中的起始蛋白dna。引发剂如何在一种情况下以细胞依赖的方式积累，而在另一种情况下以细胞大小依赖的方式积累仍有待研究。

Irrespective of the organism, genomic stability requires that the entire DNA be replicated once and only once per cell cycle. The intricacies of regulatory mechanisms that guarantee initiation of replication and prevent reinitiation prematurely are still being unraveled even in a well-studied system like E. coli. V. cholerae has two chromosomes. The origin of replication of chromosome I is similar to that of E. coli, while the origin of chromosome II is similar to plasmid origins that have repeated origin-specific initiator binding sites (which in plasmids are called iterons). We have determined that in spite of the distinct features of the two Vibrio origins, once-per-cell-cycle replication requires that both the origins be fully methylated at the adenine residues of their GATC sites. For chromosome I, as for oriC of E. coli, methylation was found to be inessential for replication initiation but required to control replication initiation frequency. For chromosome II, however, methylation was additionally required for initiator binding to the origin iterons, an essential function in the initiation process per se. Although methylation is widely used to control many DNA transactions, its role in mediating initiator-origin interactions is an unprecedented finding. Like plasmids, chromosome II has a dedicated locus that inhibits replication initiation. In spite of the entire origin region of chromosome II bearing many signatures of iteron-type plasmid origins, it functions like a chromosomal origin in that it restricts replication to once per cell cycle and to a particular stage of the cell cycle. Neither of these features are hallmarks of plasmid replication. We have discovered that while a typical plasmid locus consists of iterons only, the chromosome II locus has a second kind of site. The iterons require the sites to be fully methylated for initiator binding. The other type also binds the initiator but does not involve methylation. Since the methylation status of DNA changes during the replication cycle, the effectiveness of the iterons as control sites is likely to be cell-cycle dependent. In any event, the control activity of the typical iterons appears to be rather modest. They are primarily responsible for restraining the activity of the other more powerful methylation-independent sites. In this scenario, following replication, the DNA becomes hemimethylated, rendering the iterons unavailable to interact with the methylation-independent sites; the latter remain fully effective in preventing premature reinitiation. When full methylation of the iterons is restored, they become available to restrain the inhibitory action of the methylation-independent sites, thus setting the stage for replication initiation in the next cell cycle. It appears that the plasmid-type origin of chromosome II depends on methylation for initiating replication and for a chromosomal feature of its control. Multichromosomal bacteria (unlike such well-studied bacteria as E. coli and B. subtilis) may offer opportunities to investigate mechanisms for coordinating replication and segregation of the different chromosomes. An initial indication of inter-chromosomal coordination in V. cholerae has been obtained. We have been able to find conditions where replication of one of the chromosomes could be selectively prevented. It appears that preventing chromosome I replication can prevent/delay chromosome II replication but the reverse is not true. Chromosome II is smaller than chromosome I and the delay might allow the two chromosomes to complete replication at the same time, which might facilitate the coordination of their segregation with the cell division. The mechanism of delay remains to be investigated. Compared to our understanding of how chromosomes segregate in eukaryotes, much less is known about how chromosomes segregate in bacteria. Until recently, segregation studies were done primarily in plasmids, where genes dedicated to plasmid partition (par genes) could be found. Homologues of plasmid par genes have now been identified near the origin of replication in most bacteria, including V. cholerae. Both the Vibrio chromosomes have their own par genes. We have succeeded in deleting the par genes of chromosome I without causing much of a segregation defect. Rather, deletion of one the two par genes promoted replication. A similar finding has also been made in B. subtilis. The two bacteria, B. subtilis and V. cholerae, have diverged more than a billion years ago but both have retained the par genes and use them for similar purposes. The wide-spread occurrence of par genes near the replication origins suggests that the genes have a general role in connecting replication and segregation. How the par genes promote replication is under current investigation. Using both the bacterial and yeast two-hybrid systems we are trying to identify proteins that might interact with Par proteins, which might provide a clue as to their mechanism of action. Cell division must await completion of chromosome replication and movement of the two sister chromosomes to opposite cell halves. The temporal control of cell division is largely unknown in bacteria. Recently, some of the genes involved in sensing glucose concentration in the growth media have been found to regulate cell size in E. coli and in B. subtilis. Mutations in these genes make cells smaller by about 30% but do not change their growth rates. In a collaborative study, we are determining the timing of replication initiation (initiation age) in these smaller cell variants. The initiation age seems to depend on the bacterium in question. In B. subtilis, the age remains unchanged in the mutants, which means that, at the time of initiation the cell size is smaller in the case of mutants compared to the wild type. In E. coli, however, initiation is delayed until the mutants reach the size at which initiation occurs in the wild type. The delay in initiation is compensated by increase in replication elongation rate, allowing the replication cycle to complete on time. The rate-limiting component is believed to be the initiator protein, DnaA, in both bacteria. How the initiator accumulates in a cellage dependent manner in one case and cell-size dependent manner in the other remains to be investigated.