生物谷報道：RNAi技術(shù)是近年來熱點,，RNAi已成為遺傳學(xué)，藥理學(xué)上重要研究手段,，而且RNAi可以進(jìn)行高通量研究,，尤其是在低等生物中，如在線蟲和果蠅中,，已有采用RNAi作為藥物篩選模型,，作為疾病篩選模型等。然而RNAi技術(shù)發(fā)展仍不是十分成熟,，尤其是在高等生物中,，因此在高等生物中研究一直局限于單基因的RNAi研究，今天剛剛出版的Natuer發(fā)表了兩篇文章,，使用大規(guī)模RNAi技術(shù)研究哺乳動物,，包括人和鼠，這一突破性進(jìn)展將為今后RNAi技術(shù)推廣應(yīng)用提供了良好的基礎(chǔ),，可以預(yù)見,，在未來幾年內(nèi)，RNAi技術(shù)將成為重要的研究手段和藥理篩選平臺的技術(shù)支撐,！在今天的Nature上,，同時刊登了一篇評論。生物谷將進(jìn)一步跟蹤最新動態(tài),，作更詳細(xì)的報道,。(附有全文，點擊full text即可下載)
RNA interference: Human genes hit the big screen
ANDREW FRASER. Nature 428, 375 - 378 (25 March 2004); doi:10.1038/428375a
Genetic screens are powerful tools for identifying the genes involved in specific biological processes. At last, RNA interference makes large-scale screens possible in mammalian cells.
One of the most intuitive ways to learn how a complicated machine works is to take it apart piece by piece — a directed 'learning by breaking'. For biologists, teasing apart the machinery underlying the form and function of an organism can be done, most simply, by removing genes one at a time and looking at the effect. One experimental method for turning genes off is known as RNA interference (RNAi; Box 1, overleaf); this has shot to prominence because it allows almost any gene of known sequence to be shut down with apparently magical ease1.
In two of the biologist's favourite model organisms, nematode worms and fruitflies, RNAi has been used to turn off almost every one of their genes2, 3. Such genome-wide RNAi surveys of gene function have remained out of reach in mammals — until now, that is, for on pages 427 and 431 of this issue Paddison et al.4 and Berns et al.5 report the generation of tools to allow RNAi mass-screening of mammalian genes. This at last makes it possible to carry out genetic screens in mammalian cells in culture.
There is a range of effective strategies for RNAi in mammalian cells (reviewed in ref. 6), and they differ principally in the method for getting the double-stranded RNA (dsRNA) that specifically interferes with the target gene into the cells. In one method, rather than synthesize the dsRNA chemically before introducing it into the cells, the interfering dsRNA is made directly by the cells themselves. A vector directing the transcription of precise short hairpin RNAs — shRNAs — by RNA polymerase III is introduced into the cells; these transcribed shRNAs are processed by the cell to give the small interfering dsRNAs (siRNAs) that turn off the target gene. shRNA-expressing vectors allow for sustained RNAi in a wide range of cell lines (including embryonic stem cells, the subject of much current research). A complete library of shRNA-expressing vectors designed to target each and every gene in a mammalian genome would thus allow genome-wide RNAi-based genetic screens in cells in culture. Put simply, for any process that we are interested in (cell division, response to DNA damage and so on), with such an shRNA library we could screen every gene in the human genome and ask if it is involved.
Both groups4, 5 have converged on the same basic shRNA library approach, each generating a retrovirus-based library capable of targeting around a third of human genes; the genes were chosen for their potential roles in disease. Different shRNAs often interfere to differing extents with a target gene, so at least three shRNAs have been cloned for most genes. This multiple coverage not only provides an internal control, but may also allow comparison of both strong and weak 'knock-downs' of a specific gene in an analogous way to a classical genetic approach7.
Berns et al.5 used their library to search for genes that affect the function of p53, a tumour-suppressor gene that kills or 'arrests' cells with damaged DNA. They screened around 8,000 human genes to find those required for a p53-dependent arrest of cell proliferation and identified six genes, including p53 itself. Further assays confirm that these genes — which include a histone acetyl transferase and a histone deacetylase, two key regulators of gene expression — do indeed play a role in p53-induced cell-cycle arrest and senescence. This ability to survey the gene functions of a full third of the human genome so rapidly is breathtaking, and the success of the subsequent assays underscores the quality of this approach.
The retrovirus-based vectors used by both groups are excellent for many cell-based screens. But they cannot be used for the stable expression of shRNAs in all cell types — this requires moving the shRNA-encoding inserts to different vectors. The shRNA library described by Paddison et al.4 incorporates an elegant system for shuttling the inserts into any destination vector simply using bacterial mating. Their sequence-verified shRNA library targets almost 10,000 human genes; the shRNAs have also been chosen to allow targeting of the mouse orthologues (equivalents) of those human genes, if possible, and over 5,000 mouse genes can be targeted using their clones. This library of easily transferable shRNAs is a beautifully designed resource, and should permit an impressive range of analyses in diverse cell types.
To increase the speed of RNAi screening, both groups4, 5 borrow a sequence identifier (bar-code) system, developed in studies on yeast, for the quantitative analysis of pools of genes8. Each shRNA construct has a unique bar-code — Berns et al. use the shRNA sequence itself, whereas Paddison et al. have an independent bar-code, which they report as being of far greater effectiveness. The abundance of each shRNA construct in a pool of constructs can be assessed by monitoring the relative levels of each bar-code using a microarray. Thus any screen for genes that confer a growth advantage (or defect) can be carried out by the simultaneous screening of large pools of shRNA-expressing vectors, greatly increasing the throughput. Bar-coding is still in its infancy but has great potential for analysing RNAi selection screens.
There are still some uncertainties surrounding mammalian cell RNAi, especially regarding both specificity and efficiency of targeting. According to one report9, a sequence identity of as few as 11–12 nucleotides between an interfering RNA and a messenger RNA may be sufficient for interference to occur. If it is, cross-reactivity is a substantial problem: far from targeting one gene, many expressed shRNAs may target several genes simultaneously. Similar analyses10 came to the opposite conclusion, however, so it remains to be seen whether this is a general problem. Even if cross-reactivity does occur, there are straightforward controls for specificity: most simply, if two independent shRNAs targeting the same gene give similar effects, it is probably safe to conclude that this is specific to the targeted gene, and not due to some 'off-target' cross-reaction. This is precisely the approach adopted by Berns et al. and the presence in each of the libraries reported here of multiple shRNAs against each gene should make these internal controls relatively easy.
As regards RNAi targeting efficiency, it is clear that — as in worms or flies — different genes in mammalian cells are turned off with differing efficiencies. For example, Paddison et al. screened their library to identify components of the proteasome, a cellular machinery that degrades many unwanted proteins and that is implicated in certain diseases. Although genes encoding some subunits (those for the 19S base, for example) were apparently easily identified, others (such as those of the 19S lid or 20S core) were harder to hit. Like any screening tool, RNAi is unlikely ever to be perfect. As the rules for predicting effective shRNAs continue to improve, however, the false-negative rate will drop, and the libraries will improve.
Despite these notes of caution, we will no doubt see an explosion in RNAi screening of mammalian cells over the coming months. As with any genetic screen, the power of each RNAi screen depends on the appropriate choice of functional readout, and that will require development of a variety of cell-based assays (such as the assay for proteasomal function reported by Paddison et al.). As no single laboratory can specialize in every aspect of gene function, the general availability of these shRNA libraries as communal resources is a major step forward, harnessing the screening expertise of the entire mammalian-cell research community. Pulling together the data from these varied RNAi screens in a common, central database will take our understanding of mammalian gene function a further giant stride forward.
Original news：
RNA interference: Human genes hit the big screen ANDREW FRASER
Genetic screens are powerful tools for identifying the genes involved in specific biological processes. At last, RNA interference makes large-scale screens possible in mammalian cells.
Nature兩篇文章,，這里提供全文下載,，補充資料需要請在后面留言索取
A resource for large-scale RNA-interference-based screens in mammals 427
PATRICK J. PADDISON, JOSE M. SILVA, DOUGLAS S. CONKLIN, MIKE SCHLABACH, MAMIE LI, SHOLA ARULEBA, VIVEKANAND BALIJA, ANDY O'SHAUGHNESSY, LIDIA GNOJ, KIM SCOBIE, KENNETH CHANG, THOMAS WESTBROOK, MICHELE CLEARY, RAVI SACHIDANANDAM, W. RICHARD MCCOMBIE, STEPHEN J. ELLEDGE & GREGORY J. HANNON
doi:10.1038/nature02370
full-text
A large-scale RNAi screen in human cells identifies new components of the p53 pathway
KATRIEN BERNS, E. MARIELLE HIJMANS, JASPER MULLENDERS, THIJN R. BRUMMELKAMP, ARNO VELDS, MIKE HEIMERIKX, RON M. KERKHOVEN, MANDY MADIREDJO, WOUTER NIJKAMP, BRITTA WEIGELT, REUVEN AGAMI, WEI GE, GUY CAVET, PETER S. LINSLEY, RODERICK L. BEIJERSBERGEN & RENÉ BERNARDS
doi:10.1038/nature02371
Full-text
