Although RNAi constitutes a very powerful tool for studying gene functions in plants and invertebrates, its application in mammalian systems turned out to be a major problem because mammals have evolved a different and more elaborate response to dsRNA. In mammals, dsRNA longer than 30 bp mediates an interferon response, which leads to the simultaneous activation of RNase H, which unspecifically degrades all mRNA transcripts, and protein kinase R (PKR). The latter phosphorylates and, thus, inactivates transcription factor eIF2a leading to a global shutdown of protein biosynthesis and as a result to apop-tosis (94,95).
Despite the first impression that RNAi would not work in mammalian cells, several independent groups proved the existence ofmammalian RNAi pathways by the introduction of dsRNA or vectors producing dsRNA into cell lines lacking the interferon machinery, like mice oocytes or mice embryonic cancer cell lines (96,97). The most important experiment that established RNAi as the same powerful tool in mammals has been performed by Tuschl and co-workers (98), who used synthetic 21-nt duplexes (siRNAs) to trigger the RNAi pathway. They achieved the knocking down ofthe activity oftransfected and endogenous genes without induction ofthe interferon response. Obviously, the siRNAs, which act as the active intermediate of RNAi, are too short for the activation of PKR. dsRNAs shorter than 21 bp and longer than 25 bp are inefficient in initiating RNAi as well as siRNAs with blunt ends (41). Only short dsRNAs with a two-nucleotide 3-overhang resembling the naturally active products of Dicer are efficient mediators of RNAi. With this technology, even somatic primary neurons have been successfully treated to produce knock-down RNAi phenotypes (99).
Investigation ofthe minimal chemical requirements for siRNAs and a detailed mutation scan suggest that the 3'-end modification of the antisense strand usually reduces activity, and mispairing is more crucial for the first 10 nucleotides from the 5'-end of the sense strand. The 10th nucleotide seems to be important for RISC-mediated cleavage ofthe target mRNA (41,48,54,100,101). Further, the 5'-phosphate residue is essential for siRNAs to direct target-RNA cleavage, but nonphosphorylated siRNAs have been shown to be phosphorylated in vivo by a cytosolic kinase prior to their entrance into the RNAi pathway (54). Shortly after the discovery of siRNA-mediated RNAi in mammals, empirical rules were put up for the design of efficient synthetic siRNAs that are now often referred to as the Tuschl rules (56).
The recent finding that only one strand of the siRNAs enters the RISC depending on thermodynamic properties will help to design siRNAs displaying an even higher efficacy (55).
Novel rules for the design of siRNAs and algorithms, which are based on those rules and consider the accessibility of siRNA binding sites on the secondary structure of the target mRNA, are currently being developed and refined (55,56,102). They also circumvent off-target effects by evaluating the antisense sequence in a Smith-Waterman or BLAST search for possible targets.
To date, the use of siRNAs has become a state-of-the-art tool for the study of gene function in mammals and in cultured mammalian cells. Depending on the type of cells, different chemical and physical methods are used to deliver siRNAs, such as liposom-mediated transfection (98,103-105), electroporation (106), and microinjection. The most common delivery method so far is the regular transfection as described for DNA. Even though electroporation has come into focus lately, transfection is still the most reliable technique, especially for the delivery into adherent cells, such as HeLa, NIH3T3, or 293T cells. It should be mentioned that transfection by calcium phosphate is not as efficient for siRNA as it is for DNA (107,108). Following the use of calcium phosphate and polybrene, various cationic liposomal formulations were developed to increase transfection efficiency to up to 90%, depending on the cell type.
Electroporation is an alternative, which should be considered for nondividing cells or cells resistant to chemical transfection reagents (109-112). Other approaches are exploiting peptides such as the short MPG that forms stable noncovalent complexes with nucleic acids. MPG is a chimeric protein composed of gp41 (the human immunodeficiency virus [HIV-1] fusion peptide domain) and the nuclear localization sequence (NLS) of SV40 large T antigen. Using the MPG bearing a mutation in its NLS prevents nuclear entry and distributes the siRNAs throughout the cytosol (113). Other approaches use the great versatility of cell-penetrating peptides (CPPs) (114-118) with respect to cargo and cell type as a valuable tool for the introduction of siRNAs into mammalian cells and even fully grown organisms. The covalent coupling of CPPs with siRNAs yields the so-called pepsiRNAs (peptide-coupled siRNAs) (119,120).
The use of siRNAs has already helped to screen a large number of mammalian genes for their function and to unravel the molecular basis of several important biochemical processes like signal transduction, cell cycle regulation, development, cell motility, cell death, and many more.
The number of targeted mRNAs is constantly increasing together with the validation of a vast number of siRNAs. The field is constantly growing and the first genomewide studies have already been carried out but have just been published in part (121).
Until now, many gene functions responsible for embryogenesis or stem cell differentiation remain to be determined. Although these systems have hardly been accessible by traditional methods, RNAi appears as a practical approach that has already contributed to the characterization of many developmental genes and, thus, constitutes a promising tool for the elucidation of mechanisms involved in development and disease.
The siRNA-mediated silencing of a single isoform of shcA in HeLa cells predicted a crucial role ofthis protein in regulating cell proliferation (122). Likewise, siRNA-mediated silencing of the phosphatidylinositol 3-kinase causes a drastic decrease in growth and tissue invasiveness of tumor cells (123). By RNAi, the IP3 receptor-1 in germinal vesicle-intact oocytes was found to be responsible for intracellular calcium oscillations, the first steps of development after insemination (124). RNA interference was also used to silence several molecular players ofthe cell cycle and DNA replication (125-128). The identification and characterization of novel proteins intervening in embryonic (129,130)
and neuronal development (99) are further examples of insights made possible by this new technique.
Its versatily, its high specificity, and the minute amounts required to inhibit gene function render RNAi a highly promising technique to combat diseases. Because RNA silencing serves as a defense against retroviruses in plants and invertebrates it appears only challenging to reintroduce RNAi to fight viruses in mammals and humans. Many studies have been performed to treat HIV-1, herpes simplex virus, and hepatitis B and C by dsRNA-based approaches. Moreover, the high sensitivity toward point mutations qualifies RNAi as a potential tool for the cure ofcancers and inherited diseases. In a study ofthe Ras oncogene, it was possible to target mutated Ras without affecting unmutated Ras, demonstrating the exquisite specificity of RNAi as a therapeutic tool. However, the same specificity could pose problems for therapeutics against viral infection. Viral escape from RNAi selection in poliovirus by mutation of the target sequence has already been demonstrated. The high specificity is already put into doubt by recent findings that the RNAi mechanism can tolerate some sequence mismatches, particularly away from the middle cleavage site. In fact, a recent in vitro study showed that some genes with incomplete homology could be partially silenced, an effect that was more pronounced at higher concentrations of siRNA.
Site-specific delivery of siRNAs broadens the list of genes that can be silenced without inducing toxicity. To target mutated sequences in inherited diseases, the individual gene mutation needs to be determined in order to synthesize appropriate siRNAs. This patient-specific approach would be much more costly and difficult to execute. Therefore, the in vitro studies of RNAi carried out to date focus mainly on viral infection and cancer, which are likely to be the areas in which siRNAs make their way to clinical studies.
The siRNAs are further being studied as therapeutic treatment against genes involved in autoimmuninty, neurodegenerative diseases like Alzheimer's disease, infectious diseases, and inflammatory response and many more.
To date, many prerequisites still need to be fulfilled and obstacles overcome before RNAi can make its way into clinical studies. First of all, silencing is often incomplete— a knockdown rather than a knockout—and residual gene expression might be sufficient to maintain the present phenotype, especially if only low amounts of protein are required to fulfill their cellular function.
Some residual gene expression might be attributed to untransfected cells or to a low affinity of the selected siRNAs toward the target mRNA. In rapidly dividing transformed cells, the silencing effect is rather short-lived, as the transfected siRNAs are rapidly diluted.
Further obstacles ofthe clinical application of RNAi often are insufficient transfection efficiency and the limited persistence of the transient RNAi phenotype. To obtain persistent RNA silencing in cells and organisms, different plasmid and viral vectors were developed to express short hairpin RNAs under the control of RNA polymerase III (Pol III) and RNAse polymerase II (Pol II) promoters. Upon transfection of these vectors, siRNAs can be constitutively and endogenously expressed.
The currently used vectors employ short hairpin RNA (shRNA) expression cassettes that resemble pre-miRNAs and undergo processing by Dicer (131-136). They are designed according to the same rules as synthetic siRNAas to match perfectly with the target mRNA and, thus, trigger its degradation. To achieve a constitutive silencing effect, the shRNA transcript must translocate from the nucleus to the cytoplasm, where it has to be recognized by Dicer and transferred to RISC. siRNA hairpin-expressing plasmids can specifically suppress gene expression in a transient or persistent manner depending on their design. Thus, loss-of-function phenotypes can be generated if longer periods of time are required to fully deplete the cytosol of residual target protein. The potency of shRNAs to trigger RNAi in mammalian cells is comparable to synthetic siRNA duplexes.
In many cases, strong RNA polymerase III promoters, such as the human H1 and the murine U6, are employed to control the expression of the shRNAs. (131-136). In vivo, RNA polymerase III is responsible for the transcription of a limited number of genes, including 5S RNA, tRNA, 7SL RNA, U6 snRNA, and a number of other small stable RNAs that are involved in RNA processing (137).
It is assumed that long dsRNAs are a more effective trigger because they are more efficiently processed into siRNAs (49). This can be the result of a highly cooperative binding of long dsRNA strands or additional features of cleavage by Dicer, such as incorporation of the nascent siRNAs into transfer protein complexes (138).
Despite some concerns, which are discussed within the field of RNAi, it was recently shown that endogenously expressed long hairpin dsRNAs (lhRNAs) are capable ofinduc-ing RNAi in mammalian somatic cells including human primary fibroblasts, melanocytes, HeLa cells (139,140), and even whole mice (141).
To be taken up by cells or tissues, those expression vectors require classical methods like electroporation, microinjection, or liposomal transfer of the DNA precursor vector. Most cell lines are easy to transfect and recombinant cell clones permanently expressing RNAi phenotypes can be selected. However, without cell division, the shRNA (DNA) construct cannot enter the nucleus as required for DNA transcription and it passively resides in cytosol. Therefore, most nondividing cells are not susceptible to transfection, as are primary cells and stem cells. A number of important cell types have been resistant to the introduction ofboth siRNAs and shRNAs because of the lack ofan appropriate delivery system.
Even more problematic is the delivery of siRNAs directly to entire vertebrate animals. High toxicity of most cationic transfection reagents prohibits whole-body application of siRNAs aided by liposomal or chemical approaches, whereas physical techniques like the hydrodynamic transfection method was shown to be successful in mice. Naked siRNAs applied to mice via tail-vein injection caused the knock down of a reporter gene by 8090% in the liver, kidney, spleen, lung, and pancreas (142,143). Yet, the effect is rather short-lived, lasting only a few days, and not all organs and cell types are accessible.
Another method to circumvent many of those difficulties makes use of viral vectors to infect cells with the dsRNA expression construct. Retroviral vectors (144), like adenoviral vectors, and predominantly lentiviral vectors (145,146) are currently being used as viral delivery systems. By this approach, almost every cell or tissue, including stem cells and neurons, can be treated with shRNA-producing vectors.
The advantages of the lentiviral approach lies in the possiblity of systematically testing a gene function in the context of the entire organism. Further animal models can be generated in a straigthforward manner to determine which genes are important to the function of different tissues and organs and which might be effective therapeutic targets in diseases.
Even though Tuschl and co-workers (98) observed that siRNAs shorter than 30 nt do not trigger an interferon response, recent reports are concerned about nonspecific effects induced by endogenously expressed shRNAs. Both transfection of siRNAs and transcription of shRNAs leads to an interferon (IFN)-mediated activation of the Jak-Stat pathway and global upregulation of interferon-stimulated genes within the nucleus (147-149). It was found that the dsRNA-dependent protein kinase (PKR) is activated by the intracellular presence of 21-bp siRNAs and triggers the upregulation of IFN and possibly other, cellular signaling molecules. Comparative studies on interferon induction by siRNAs and their shRNA counterparts showed a significantly stronger interferon response after the application of shRNA (149), so that these side effects need to be studied more closely to settle the current controversy. It has to be kept in mind that almost all plasmid vectors can induce interferon response upon transfection independent of the type of insert they are bearing (150).
Was this article helpful?
The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.