The phenomenon of silencing sequence complimentary genes by the introduction of double stranded RNAs into C. elegans was mentioned in the 1998 report by Nobel laureates Fire and Mello (Börner, K.Grimm, 2010). This phenomenon known as RNA interference (RNAi) has revolutionized research, by finding therapeutic applications for various disorders like cancer, infectious diseases caused by viruses, bacteria and parasites and respiratory diseases (R.K.M. Leung, P.A. Whittaker, 2005). RNA interference (RNAi) results in post-transcriptional gene silencing by making use of a highly conserved intracellular mechanism (M. Yang, J. Mattes, 2008). Despite the mechanism of RNAi being well-known, the application of this technology faces numerous challenges that needs to be overcome (Castanotto et al, 2009). RNAi mediated through small interfering RNA (siRNA) technology is gaining a lot of prominence amongst the existing gene silencing techniques because of its higher reliability, efficiency, and specificity than previous methods (S.J. Tebes, P.A. Kruk, 2005). This article discusses the mechanism of RNA interference through siRNA, the promises and pitfalls, delivery strategies and therapeutic applications of siRNA.

What is siRNA?
siRNA are oligonucleotides of around 21–23 nucleotides (nt) in length. The secondary structure of mRNA determines the efficacy of a complimentary siRNA to access its mRNA target (Holen et al., 2002, Renata Servan de Almeida et al, 2008). Long (> 500 bp) dsRNAs have been known to mediate specific and effective silencing of genes in C. elegans and D. melanogaster (Mello and Conte et al, 2004). However, in mammalian systems, 21–23 bp siRNAs are used since dsRNAs > 30 bp has been long known to trigger the ?-interferon (IFN) pathway (M.P. Gantier, B.R.G. Williams, 2007)

Mechanism of RNAi
Long dsRNAs are cleaved into small 19–21 nucleotide duplexes (termed siRNAs) in the cytoplasm by an endogenous cytoplasmic RNase III-like enzyme, Dicer. This leaves two nucleotide overhangs in the 3´ position (Jones, de Souza and Lindsay. 2004) (Fig. 1). Subsequently, the siRNA duplex associates with several proteins to form the RNA induced silencing complex (RISC). Argonaute family of proteins in the RISC complex bring about removal of sense strand. This results in antisense strand being annealed to RISC complex thereby facilitate silencing of target mRNA following hybridization of the antisense strand with the target mRNA.  The target mRNA is then cleaved by the RISC complex approximately 12 nucleotides from the 3´ terminus of the complimentary siRNA. The RISC is released from the inactivated mRNA and the targeted mRNA is degraded by cellular exonucleases. If the target sequence match is inexact, only translation of the mRNA is inhibited (Hammond et al. 2001, S. Geley, C. Müller. 2004).

Advantages of siRNA delivery
The various advantages associated with siRNA therapeutics are schematically represented in fig. 2. The highly specific nature of siRNA therapeutics enables efficient targeted delivery for almost any gene. This is especially of advantage in the fields of oncology and genetic neurological disorders where disease is often caused by a dominant mutation in a single allele. It is also possible to target disease-specific alleles that differ from the normal allele by only one or few nucleotide substitutions because of the high specificity offered by siRNA based therapeutics (Lars A et al. 2007). RNAi is an innate biological response and hence a natural strategy for manipulating gene expression, RNAi-mediated inhibition is more potent than that achieved with antisense oligonucleotides even when site selection was optimized for antisense effectiveness. (Achenbach et al, 2003).

It is possible to target multiple sequences within an individual gene, or a group of genes, simultaneously, and hence RNAi-mediated inhibition is more versatile and can be achieved with ease thereby providing the benefit of combination therapy. It is comparatively more difficult to identify efficient antisense oligonucleotide target sequences than that of siRNA (Scherer et al, 2003). siRNA needs to be delivered into the cytoplasm only (site of action) and does not require penetration into the nucleus which could otherwise act as an additional barrier that is often difficult to traverse (Peixuan G. et al. 2010). During target identification, RNAi can be used to execute high-throughput RNAi library screens to assay the direct correlation between the modulation of gene expression and functional phenotypes (Moitreyee C-K et al. 2005). Prolonged therapeutic benefits can be obtained with siRNA as compared to that of antisense oligonucleotides because siRNAs are more resistant to nuclease degradation than antisense oligonucleotides (S. Zhang. et al. 2011).

Challenges in siRNA delivery
siRNAs are more stable in serum and mammalian cells than antisense oligonucleotides and ribozymes siRNA molecules are more resistant to nuclease degradation than antisense molecules. Nevertheless, some serum nucleases can degrade siRNAs (Bertrand et al, 2002).  Hence various chemical modifications have been attempted to improve the stability of siRNAs. It has been shown that boranophosphate siRNAs were more effective at silencing than phosphothioate siRNAs and displayed 10 times more nuclease resistance over the unmodified siRNAs. siRNAs containing 2´O- methyl and 2´-fluoro nucleotides portrayed improved stability and showed increased potency (Jonathan et al. 2008; Hall et al, 2004; Allerson et al, 2005). Stability of siRNAs can also be achieved by complexing them with polyethyleneimine (PEI) (Urban-Klein et al, 2005, Seung-Young Lee et al. 2010). siRNA can be delivered into the cell by use of DNA encoding short hairpin RNA (shRNA) expression cassettes. This gives rise to intracellular expression of shRNAs, which are further processed into active siRNAs by the host cell (Rao et al. 2009a, Paddison et al, 2002). In order to achieve systemic delivery, one needs to look at stabilization of the siRNA, targeting to the correct tissue, and facilitation of cellular uptake. In order to improve stability and cellular uptake of siRNA drugs, various approaches like chemical modification of nucleic acid, packaging into viral or non-viral vehicles and targeting with different ligands and antibodies are being investigated (Sorensen et al, 2003). Viral vectors like lentivirus and adeno-associated virus (AAV) vectors are also being considered for clinical delivery of siRNAs although its side effects cannot be neglected (Clayton, 2004). The exposure of cells to any exogenous molecule (siRNA or delivery agent) can disturb normal cellular functions and needs to be carefully controlled. Issues like off-target effects, non-specificity and saturation of RNAi machinery can occur. Two mRNA molecules may have sequence homologies due to which a siRNA duplex may target more than 1 mRNA molecule.  Hence, it is essential to know the minimum level of homology required for siRNAs to mediate inhibition of a gene (Vickers et al, 2003). The delivery vehicle and the siRNA effector may themselves trigger unexpected cellular responses like immune and interferon responses that can be cytotoxic (Kariko et al, 2004). This calls for cautious design and screening of each  siRNA and delivery vehicle in order to minimize all potential adverse effects. siRNA delivery can cause saturation of intracellular RNAi processing machinery leading to alterations in the expression of miRNAs or their processing machinery thereby becoming the cause of several neurological diseases (Gong et al, 2005) and cancers (Lu et al, 2005). A wide variety of issues will determine the efficacy of siRNA delivery. Resistance of particular RNAs to RNAi-mediated degradation has also been observed in cases where accessibility of the target sequence was restricted. For viral RNAs, resistance can also be related to intracellular location and/or nucleocapsid association of genomic RNA molecules (Hu et al, 2002). Resitance may occur in case of cancers wherein the processing machinery required for siRNA activity is not functionally normal in some tumor cells (Lu et al, 2005). siRNA can slow the process of tumor progression, but it cannot totally eliminate it. When delivered as a drug, siRNAs and shRNA-expressing DNA templates function transiently. When transfected into cells in culture, the inhibitory effect usually peaks at day 2 post-transfection and decreases thereafter which is disadvantageous in case of chronic conditions (Susan, 2005).

Delivery strategies
siRNA performs its action at the cytoplasm. The most challenging of its delivery is its stability in plasma to nucleases and its ability to cross the cellular barrier to be able to enter cytoplasm. High pressure injection is one of the oldest strategies to deliver siRNA intravenously wherein saline containing siRNA is injected into the tail vein in case of rats resulting in distribution of siRNA to various organs like liver, kidney, lung, etc. Another approach wherein siRNA is directly delivered into target organs and tissues is by use of electroporation (M. Yang, J. Mattes. 2008). Research is also focused on the development of viral and non-viral based delivery of siRNA in vivo. Viral mediated delivery is based on the principle that viruses can enter cells by their natural way and by conjugating them to siRNA one can be assured of siRNA getting delivered into cytoplasm. The various viral vectors include adenovirus, lentivirus, retrovirus, Sendai virus or adeno-associated virus, carrying encapsulated siRNA (Paul et al, 1998). Non-viral mediated delivery can be achieved by use of cationic lipids (lipoplexes, liposomes), cationic polymers (polyplexes, nanoparticles), polymer-lipid hybrids, etc. siRNA can be complexed with polyethyleneimine (PEI) and this complex destabilizes in cytoplasm by proton sponge effect thereby releasing siRNA into cytoplasm (A. Aigner. 2006). Successful delivery of siRNA in hepatitis C virus (HCV) infections have been achieved by encapsulating siRNA into cationic liposomes composed of three lipids: a cationic lipid, phosphatidylcholine (PC), and lactosylated pho phatidylethanolamine (T. Watanabe et al. 2007). Transferrin conjugated to a cyclodextrin-polycation polymer has been attempted to deliver siRNAs to target the Ewing’s sarcoma Ews–Fli1 fusion mRNA by means of the transferrin receptor in mice48, resulting in inhibition of tumour progression (Daniela C. et al. 2009).

Protiva Biotherapeutics and Alnylam have developed nanoparticles composed of a lipid–PEG conjugate that can encapsulate and protect nucleic acids for systemic delivery. These stable nucleic acid lipid particles (SNALPs) have been successfully used to administer siRNAs to a non-human primate (D.D. Rao et al. 2009).

Therapeutic potential
siRNA has profound applications in various areas of drug delivery because of its attractive features like specificity, potency and versatility. siRNA has been effectively used to treat viral infections like human immunodeficiency virus, rotavirus, dengue virus and influenza virus. siRNA can also be directed against essential cellular genes, whose products are required for virus replication. For example, silencing of Tsg101, a protein required for budding of virions from infected cells, greatly reduced the release of HIV from transfected 293T cells (J.A. Taylor, N.V. Naoumov 2005). In ovarian cancer, siRNA technology was shown to inhibit Her-2/neu expression in vitro resulting in decreased cell proliferation, increased apoptosis, increased G0/G1 arrest, and decreased tumor growth (Stephen et al, 2005). The vascular endothelial growth factor (VEGF)-targeted siRNA Bevasiranib was used for the treatment of age-related macular degeneration (S. Zhang et al 2011). Intranasal administration of siRNA/chitosan polyplexes have shown good potential in downregulation of nflammation associated with allergic rhinitis (H. de Martimprey et al 2009). RNAi holds great promise to treat neurological disorders as well. Silencing of the mutant doublecortin (subcortical band heterotopias); amyloid precursor protein (familial Alzheimer’s disease), tau (Parkinsonian frontotemporal dementia); torsinA (DYT1 dystonia); survival motor neuron protein (spinal muscular atrophy), MJD1 (Machado-Joseph disease;), and prion (prion-based disease;) are areas wherein RNAi could be therapeutically beneficial for dominantly inherited neurodegenerative disorders (D.R. Thakker et al 2006).

Conclusion
The immense potential that siRNA holds to eradicate various diseases is clearly evident. However, the plethora of challenges needs to be successfully addressed to effectively take this unique technology from bench side to bed side.