时近岁末，各大杂志接连进行了年终盘点，上周出版的《Nature》杂志也对2011年进行了回顾：365 days: Nature's 10 ，评点了2011年的科技进展，科技政策以及重要人物。其中《Nature Methods》也盘点了年度技术，选出了2011年最受关注的技术成果：人工核酸酶介导的基因组编辑（genome editing with engineered nucleases）技术。
方法技术从来都是科学进步的推动力，在生命科学领域更是如此，去年《Nature Methods》将这一殊荣颁给了光遗传学（optogenetic），原因是专家们认为这一技术工具在神经科学，以及细胞生物学信号通路研究方面具有革命性的促进作用。而今年《Nature Methods》将年度技术颁给了基因组编辑技术，理由是认为这种技术能通过在某些物种基因组中进行靶向特异性的突变，从而解答并提出更多精确的生物学问题。
也许了解一个基因或者蛋白的功能最可靠的方法就是特异性干扰其作用，并监控带来的影响，这种逆向遗传学分析方法目前已经在许多物种中成为了常规性方法，但是还是存在一些例外，某些物种很难，甚至可以说是不可能完成内源性基因位点的靶向突变，要达到这一目的，就必需采用一些更为间接的方法：对应的异源位置修改基因的过量表达，或者敲除这一基因（常常进行部分敲除），比如RNA干扰技术。
而人工锌指核酸酶介导的基因组编辑技术，或者称为基因组定点修饰技术，则不同，这种技术原则上能在任何物种基因组的任何位置上进行设计切除，从而能在内源性序列上引入特异性修改。
从技术发展的角度来看，实际上人工核酸酶技术已经成为了在多种细胞类型和生物体内进行高效、位点特异性的基因修饰的一个常用工具。这种酶包括有三个主要的类型——锌指核酸酶 (ZFN)，转录激活因子样效应物核酸酶(transcription activator-like effector ucleases，TALENs)，以及归巢核酸内切酶（engineered meganucleases），这些技术都建立在非常基础的研究之上，比如首次利用ZFN进行物种内源性基因修改，就用到了三方面的知识：限制性内切酶的作用机理，细胞如何修复DNA缺口，以及DNA结合蛋白如何在如此多基因组序列中特异性定位。
这些技术也毫无疑问能用于临床上，比如纠正单基因决定的人类疾病中错误突变等。今年在这方面就获得了重要进展——来自费城儿童医院等处的研究人员研究人员，利用遗传工程改造后的ZFNs，在基因组上特定位置诱导双链断裂，并且在实验小鼠活体中以具有临床意义的水平刺激基因组编辑，从而诱导高效的基因修正。
这种方法利用了锌指核酸酶独具在染色体上精确定位的优势，避免了传统基因疗法存在插入诱变（insertional mutagenesis）的风险，首次取得了基因组编辑在基因疗法上的突破。
这是临床上基因组编辑技术2011年取得的一大突破，而在技术进展方面，今年TALENs也获得了新成果——虽然ZFNs是最佳的工具，但是很难设计，因此难以实际操作。而今年在植物病原菌TAL效应子方面的成果促进了基因组编辑技术向前迈开了一大步，这项研究表明TAL效应子能显著简化DNA绑定代码，从而减轻了一些目前存在的，工程改造工具中的问题。
除此之外，令人高兴的是可以看到商业用途ZFNs的价格降低了，比如今年Sigma宣布调整了CompoZr 锌指核酸酶（ZFN）产品的价格——CompoZr ZFN 产品全线降价50%。而且一些产品能组合工具，帮助研究人员定制所需的核酸酶。同时TALENs以及归巢核酸内切酶也有了商品化产品。到底哪种技术最终将取胜，目前还不得而知，毕竟还存在许多未知数，尤其是一些物种中TALEN的功能。
以下是英文原文：
In our annual toast to biological research methods and to the scientists who develop them, we have chosen genome editing with engineered nucleases as our Method of the Year 2011.
Perhaps the most reliable way to learn about the function of a gene or protein is to specifically perturb it and monitor what happens. This reverse genetic approach is routinely applied in many species, but, with a few exceptions, it is challenging or even impossible to make targeted changes at endogenous genomic loci, arguably the most elegant method of genetic perturbation. Instead, the experimenter must settle for more indirect approaches: overexpressing the modified gene from a heterologous location or knocking it down, often only partially, with an approach such as RNA interference. Engineered nucleases—which can be designed, in principle, to cut at any location in the genome of any species and thus to introduce tailored modifications into the endogenous sequence—are set to change this. We provide a brief Primer on these tools on page 27.
From the perspective of methods development, the trajectory of the engineered nucleases has been a compelling one, as reported in a News Feature on page 23. All three major classes of these enzymes—zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) and engineered meganucleases—stand on the shoulders of very basic studies, a testament, yet again, to the necessity of basic research and the difficulty in predicting whence technological leaps will come. Engineering the first ZFN to modify an endogenous gene in an organism (the fruit fly, as it appositely turns out), required that three strands of knowledge, gained from many decades of research in many laboratories, be drawn together: how restriction endonucleases work, how DNA breaks are repaired by the cell and how DNA-binding proteins achieve specificity in the vastness of genomic sequence space.
This said, it has undoubtedly been the potential clinical utility of these tools—to correct mutations in monogenic human disease, for instance—that has primarily driven the intense effort put into developing them over the past decades. The fortuitous consequence, however, is that very powerful basic research tools have been generated in the process.
Engineered nucleases can be used to knock out or knock in genes, to make allelic mutants, to change gene-regulatory control and to add reporters or epitope tags, all in the endogenous genomic context. Matthew Porteus discusses these and other exciting research applications of these tools in a Commentary on page 28.
Although the first ZFN was reported more than a decade ago, the pace of work in this field has picked up remarkably in the past year. This is in no small part because of the development of TALENs. ZFNs remain the best-characterized tools, but they are not always easy to design, as Mark Isalan discusses in a Commentary on page 32. The recent discovery of TAL effectors in plant pathogenic bacteria and the realization that their properties—notably their apparently simpler DNA-binding code—should mitigate some of the existing problems with engineering robust tools, has given a real boost to the field. Gene-editing nucleases will achieve their full potential when they can be easily and quickly designed, in practice, to specifically modify any sequence of any genome; having more than one technology available will help achieve this goal.
In addition, commercially available ZFNs are dropping in price at the same time as methods developers are assembling tools to help researchers design their own nucleases. Meanwhile, TALENs and engineered meganucleases are also already commercially available. Which technology will dominate is not yet clear: there are still many unknowns, in particular about TALEN function as discussed in several of the pieces in this issue. But the price at which researchers in regular research labs can obtain good tools is likely to play a role.
You can also hear about genome editing with engineered nucleases in a short video, and we include in this issue, as in previous years, a section of Methods to Watch in the future (p. 35).
To all our readers, a happy, successful 2012!