Small interfering RNA, usually called siRNA, is a short double-stranded RNA molecule, roughly 20 to 24 nucleotides long, that silences specific genes by intercepting and destroying their messenger RNA before it can be translated into protein. It works through a natural cellular process called RNA interference, or RNAi, which cells across many branches of life use to regulate gene activity and defend against viruses. Since its discovery in the late 1990s, siRNA has become both a workhorse of laboratory research and the basis for a growing class of medicines already approved to treat conditions ranging from nerve-damaging protein disorders to high cholesterol.
How siRNA Was Discovered
The story begins in 1998, when Andrew Fire and Craig Mello injected double-stranded RNA into the roundworm C. elegans and noticed something surprising. Single-stranded RNA had, at most, a modest effect on gene expression. But when both strands were paired together into a double-stranded molecule, the silencing was potent and specific. Even more striking, only a few molecules per cell were needed to produce the effect, which suggested the process involved some kind of amplification rather than one RNA molecule physically blocking one messenger RNA molecule.1PubMed. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans The interference showed up not just in the injected worms but in their offspring, hinting that it was a deeply embedded biological pathway rather than a laboratory curiosity. Fire and Mello received the Nobel Prize in Physiology or Medicine in 2006 for this work.
The Step-by-Step Mechanism
Understanding how siRNA actually silences a gene requires following a chain of events inside the cell. Whether the double-stranded RNA arrives naturally (from a virus, for instance) or is introduced by a researcher, the pathway follows the same core steps.
Dicer Chops Long RNA Into Short Pieces
The first player is an enzyme called Dicer, a member of the RNase III family. When Dicer encounters a long piece of double-stranded RNA, it acts like molecular scissors, cutting it into short fragments about 20 base pairs long with a characteristic two-nucleotide overhang on each end.2PubMed. The mechanism of RNase III action: how dicer dices In humans, Dicer contains two RNase III domains that work together to make precise cuts, generating those signature short duplexes.3PubMed. Homodimeric structure and double-stranded RNA cleavage activity of the C-terminal RNase III domain of human dicer The resulting fragments are what we call siRNAs. When researchers design synthetic siRNAs for experiments or drugs, they typically skip this step by delivering pre-cut molecules that are already the right size.
Loading Into RISC and Guide Strand Selection
Each siRNA duplex has two strands, but only one of them ends up doing the actual work. The duplex gets loaded into a protein complex called RISC (RNA-induced silencing complex), whose central component is an Argonaute protein. Argonaute proteins are found in nearly every human cell and contribute to gene regulation during processes as varied as stem cell maintenance, fertilization, and heart development.4PubMed Central. Structural Foundations of RNA Silencing by Argonaute During loading, the cell has to decide which strand to keep. This choice is not random. The strand whose end is less tightly paired is preferentially retained as the “guide strand,” while the other, called the “passenger strand,” gets discarded and degraded.5PubMed Central. Thermodynamic Control of Small RNA-Mediated Gene Silencing This asymmetry is a built-in quality-control step, ensuring the correct strand is positioned to find its target.6PubMed. Small RNA asymmetry in RNAi: function in RISC assembly and gene regulation
Finding and Destroying the Target
Once armed with the guide strand, RISC patrols the cell’s cytoplasm looking for messenger RNA molecules whose sequence is complementary to the guide. When it finds a match, the Argonaute protein cleaves the messenger RNA, effectively cutting it in half. The broken messenger RNA is then rapidly chewed up by the cell’s normal cleanup machinery, so the protein it encoded never gets made. Because the guide strand stays loaded in RISC after each cut, a single loaded complex can destroy multiple copies of the target messenger RNA, which explains the catalytic potency Fire and Mello noticed in those early worm experiments.
How siRNA Differs From microRNA
Conversations about small RNA-based gene silencing often blur the line between siRNA and microRNA (miRNA), since both are short RNA molecules that suppress gene expression after it has been transcribed. But the two have meaningfully different behavior. The biggest practical distinction is target specificity: an siRNA is designed to match one messenger RNA target with near-perfect complementarity, while a single miRNA can regulate dozens or even hundreds of different genes because it tolerates imperfect base pairing.7PubMed Central. siRNA Versus miRNA as Therapeutics for Gene Silencing
Their biological origins also differ. Most siRNAs silence the very gene or virus from which they were derived, making them a form of self-defense. MicroRNAs, on the other hand, are encoded in the genome at one location but silence entirely different genes elsewhere. This separation between source and target makes miRNAs more evolutionarily constrained, because their sequences must stay compatible with their many targets across generations.8Cell. Small Interfering RNAs and MicroRNAs For drug development, siRNA’s one-target precision is often the selling point: you can design a molecule to knock down a single disease-causing protein without intending to affect anything else.
The Delivery Problem and Chemical Modifications
Naked siRNA molecules face a hostile environment the moment they enter the bloodstream. They are small enough to be filtered out by the kidneys within minutes, and enzymes called nucleases are everywhere, ready to shred exposed RNA. On top of that, even siRNA that reaches the right tissue still has to cross cell membranes and escape from the internal compartments (endosomes) that swallow it during uptake. Overcoming these barriers has been the central engineering challenge of the field.
Chemical Tweaks to the Backbone
One major strategy is chemically modifying the siRNA molecule itself. Swapping in slightly altered nucleotides at specific positions along the strand can dramatically improve stability, resistance to nuclease degradation, and the duration of the silencing effect.9PubMed. Advances in structural-guided modifications of siRNA The two most widely used modifications in clinical siRNAs are 2′-O-methyl and 2′-fluoro substitutions on the sugar portion of the nucleotide.10PubMed Central. Systematic Evaluation of Position-Specific Tolerability of Seven Backbone and Ribose Modifications in Fully Chemically Stabilized siRNAs These changes make the molecule sturdier without wrecking its ability to load into RISC and find its target. More exotic modifications, like adding a small chemical group at the 4′ position of the sugar, can boost nuclease resistance even further, with some modified strands remaining more than 90 percent intact after 24 hours of exposure to a nuclease that destroys unmodified RNA within an hour.11Nucleic Acids Research. Structural basis for the synergy of 4′- and 2′-modifications on siRNA nuclease resistance, thermal stability and RNAi activity
Getting siRNA Into the Right Cells
Chemical stability alone is not enough if the siRNA cannot reach its target organ. For liver-targeted therapies, the breakthrough has been conjugating siRNA to a sugar molecule called N-acetylgalactosamine, or GalNAc. Liver cells (hepatocytes) display a receptor on their surface that eagerly grabs GalNAc, pulling the attached siRNA inside via a natural uptake process.12PubMed Central. GalNAc-siRNA Conjugates: Leading the Way for Delivery of RNAi Therapeutics This receptor-mediated approach concentrates the drug in the liver while limiting its presence elsewhere in the body.13Molecular Therapy Nucleic Acids. Ribofuranose-based GalNAc-conjugated siRNA enhances the liver-targeted delivery and elicits robust RNAi-mediated gene silencing
An alternative delivery vehicle is the lipid nanoparticle, or LNP, a tiny fat-based bubble that encapsulates the siRNA. LNPs enter cells through several uptake routes, including a process where the cell membrane engulfs them. But escape from the endosome into the cytoplasm, where RISC operates, turns out to be a bottleneck: imaging studies estimate that only about 1 to 2 percent of the siRNA inside LNPs actually makes it out of the endosome and into the cytoplasm.14Nature Biotechnology. Image-based analysis of lipid nanoparticle–mediated siRNA delivery, intracellular trafficking and endosomal escape Even with that low escape rate, enough siRNA gets through to produce clinically meaningful gene silencing, which speaks to how potent the RISC machinery is once the guide strand is loaded.
siRNA Drugs Already on the Market
The first siRNA drug to reach patients was patisiran, approved in 2018 for hereditary transthyretin amyloidosis, a rare disease in which a misfolded protein called transthyretin accumulates in nerves and the heart. Patisiran uses a lipid nanoparticle to deliver siRNA to the liver, where it silences the gene responsible for producing the faulty protein.15PubMed Central. A Review of Patisiran (ONPATTRO®) for the Treatment of Polyneuropathy in People with Hereditary Transthyretin Amyloidosis In its pivotal trial, patients receiving patisiran improved on a composite measure of nerve damage, while those on placebo worsened, producing a 34-point difference between the two groups at 18 months.16PubMed. Patisiran, an RNAi Therapeutic, for Hereditary Transthyretin Amyloidosis Follow-up data out to five years showed that about two-thirds of patients maintained stable or improved walking ability over that period.17PubMed Central. Five-Year Results With Patisiran for Hereditary Transthyretin Amyloidosis With Polyneuropathy: A Randomized Clinical Trial With Open-Label Extension
A second siRNA drug with a very different patient population is inclisiran, approved for lowering LDL cholesterol. Inclisiran uses the GalNAc conjugate strategy rather than a lipid nanoparticle. It targets PCSK9, a protein the liver makes that normally prevents LDL receptors from being recycled. By silencing PCSK9 production, inclisiran lets the liver clear more LDL from the bloodstream. Across trials, it reduced LDL cholesterol by roughly 50 percent when added to standard statin therapy.18PubMed. Small interfering ribonucleic acid for cholesterol lowering – Inclisiran Its dosing schedule is a major practical advantage: after two initial loading doses, patients need only one injection every six months, administered in a clinic, which sidesteps the adherence problems that plague daily pills.19PubMed Central. Harnessing RNA Interference for Cholesterol Lowering: The Bench-to-Bedside Story of Inclisiran Additional siRNA drugs have been approved for other liver-related diseases, with more in late-stage clinical trials.
Off-Target Effects and Immune Activation
The precision of siRNA is its headline feature, but it is not absolute. Off-target silencing occurs when the guide strand partially matches messenger RNAs it was not designed to hit. The main culprit is a short stretch at the start of the guide strand, called the seed region, spanning roughly nucleotides 2 through 8. If this seed sequence happens to match a stretch in the untranslated region of an unintended messenger RNA, RISC can suppress that gene too, mimicking the way microRNAs naturally regulate their targets.20PubMed Central. The siRNA Off-Target Effect Is Determined by Base-Pairing Stabilities of Two Different Regions with Opposite Effects Computational tools now exist to screen siRNA sequences for seed-region matches before a drug candidate ever enters a cell, helping researchers pick designs with fewer predicted off-target hits.21Bioinformatics. SeedMatchR: identify off-target effects mediated by siRNA seed regions in RNA-seq experiments
A separate concern is immune activation. Unmodified siRNA duplexes can trigger the innate immune system, mainly through receptors in immune cells called Toll-like receptors, or TLRs.22PubMed. siRNA and innate immunity When this happens, the body mounts an inflammatory response that has nothing to do with the intended gene-silencing effect and can cause side effects like fever or injection-site reactions. The good news is that the same chemical modifications used to improve stability also help dodge immune detection. In animal studies, unmodified siRNA delivered to the liver triggered a burst of inflammatory signaling molecules, while a version carrying 2′-O-methyl modifications produced no such response.23PubMed. Chemical modifications on siRNAs avoid Toll-like-receptor-mediated activation of the hepatic immune system in vivo and in vitro Modern clinical siRNAs are heavily modified in part for this reason, making immune activation a much more manageable issue than it was in the early days of the field.
siRNA in Nature and Viral Defense
Long before researchers co-opted the pathway, plants and invertebrates were using siRNA as an antiviral immune system. When a virus replicates inside a plant or insect cell, it produces double-stranded RNA as a byproduct. The cell’s Dicer recognizes this foreign RNA and chops it into virus-derived siRNAs, which then load into RISC and guide the destruction of the viral genome.24PubMed Central. Antiviral immunity directed by small RNAs This mechanism has been identified in plants, flies, and worms, and it operates as a kind of adaptive immune response that does not require the antibody-based system mammals rely on. In plants, the defense goes even further: siRNA signals can travel long distances through the plant via graft junctions, spreading the silencing signal from an infected tissue to distant, uninfected parts of the organism.25PubMed Central. An endogenous, systemic RNAi pathway in plants
Mammals have a more complex immune landscape, and while they possess the basic RNAi machinery, they rely more heavily on protein-based immune defenses like the interferon system to fight viruses. The siRNA pathway in mammalian cells seems to play a more prominent role in gene regulation and genome maintenance than in frontline antiviral defense. This evolutionary context is part of why siRNA-based drugs work so well in human cells: the machinery to process and act on small RNA duplexes is already there, just waiting for the right instructions.
Reaching Beyond the Liver
The liver has been the low-hanging fruit for siRNA therapeutics because of the GalNAc receptor system and the liver’s natural role as a blood-filtering organ that readily takes up circulating molecules. But many diseases that could benefit from gene silencing occur in tissues that are much harder to reach. The brain, protected by the blood-brain barrier, is a prime example. Researchers are actively exploring non-viral delivery strategies, including specially engineered lipid particles, polymer-based carriers, and even extracellular vesicles, to shuttle siRNA across the barrier and into brain cells for conditions like Alzheimer’s disease.26PubMed Central. Brain-targeted delivery of siRNA via non-viral delivery systems, the therapeutic strategy for Alzheimer’s disease
Agriculture is another frontier. Because RNAi is a natural defense mechanism in insects, researchers have explored spraying crops with double-stranded RNA designed to silence genes essential for pest survival. The idea is appealing because it could replace or reduce broad-spectrum chemical pesticides. In practice, though, the approach faces challenges: insects vary widely in how efficiently they take up and process external RNA, and the dsRNA itself degrades quickly in the environment.27Wiley Online Library. RNA interference in insects: the link between antiviral defense and pest control Improving dsRNA stability and finding better ways to get it inside insect cells are active areas of work, with some products already reaching regulatory review in several countries.
The lung, the eye, and tumors are also targets of ongoing clinical programs, each presenting its own set of delivery puzzles. What links all of these efforts is the same underlying logic: if you can get a well-designed siRNA into the right cell, the RISC machinery will do the rest. The bottleneck has never been the silencing mechanism itself, which is remarkably efficient and conserved. It has always been getting the molecule where it needs to go.