Gene Silencing: A Therapeutic Approach to Combat Influenza Virus Infections

Madhu Khanna; Latika Saxena; Roopali Rajput; Binod Kumar; Rajendra Prasad


Future Microbiol. 2015;10(1):131-140. 

In This Article

Abstract and Introduction


Selective gene silencing technologies such as RNA interference (RNAi) and nucleic acid enzymes have shown therapeutic potential for treating viral infections. Influenza virus is one of the major public health concerns around the world and its management is challenging due to a rapid increase in antiviral resistance. Influenza vaccine also has its limitations due to the emergence of new strains that may escape the immunity developed by the previous year's vaccine. Antiviral drugs are the primary mode of prevention and control against a pandemic and there is an urgency to develop novel antiviral strategies against influenza virus. In this review, we discuss the potential utility of several gene silencing mechanisms and their prophylactic and therapeutic potential against the influenza virus.


Influenza virus is one of the major causes of morbidity and mortality, especially among young children and the older adults.[1–4] The viruses are pleomorphic (variable), mostly spherical or ovoid and filamentous, 80–120 nm in diameter, having eight single-stranded negative RNA gene segments, each ranging from 890 to 2341 nucleotide (nt) bases. Each gene segment codes for a functionally important protein of the virus. The highly conserved nucleoprotein (NP) and the matrix (M) antigens form the basis of classification of influenza viruses into three types (type A, B and C). Influenza A viruses can be further divided into various subtypes depending upon the antigenic differences in the hemagglutinin (HA) and neuraminidase (NA) surface proteins. HA is the major envelope glycoprotein which enables binding of the virus to receptors on the host cell surface. It is the primary target of neutralizing antibodies. NA holds significance due to its crucial role in the release of progeny virions from the surface of virus-infected cells.[5] Antibodies against HA and NA antigens provide crucial protection against the virus. So far, 18 HA and 11 NA sub-types of influenza A virus have been identified.[6]

Influenza virus evades the host immune system and cause infection despite preexisting antibody due to its property of 'antigenic drift' that are small changes in the virus that happen over a period of time. The point mutations create substitutions of amino acids in the surface glycoproteins[7] and the new virus strains containing these changed surface glycoproteins may not be recognized by the host's immune system.[8] The other type of change is the 'antigenic shift' which is an abrupt change in the influenza virus that occurs due to reassortment of genes of the surface proteins from different viruses. Shift gives rise to influenza virus of a new subtype or a virus with a new HA or HA and NA that are very different from the subtypes circulating in humans that most of the population do not have preexisting immunity against this new virus. Influenza viruses undergo changes by antigenic drift all the time and the process is slow and gradual. Antigenic drift on the other hand occurs suddenly and if the conditions are favorable the new (novel) virus may cause a pandemic.[9]

Vaccination is the most cost effective public health intervention strategy against influenza and vaccines, if developed on time, are the best alternative to prevent influenza. However, the vaccine formulations need to be revised every year to keep pace with the rapidly evolving influenza A viruses. Antiviral drugs are also available against influenza and the two classes of drugs that are currently approved for use are: adamantane derivatives (amantadine and rimantadine) and neuraminidase inhibitors (oseltamivir and zanamivir). Amantadine and rimantadine are the M2 ion channel blockers and have been used since many years for the prophylaxis and treatment of influenza. Oseltamivir and zanamivir act by blocking viral neuraminidase that is required for the release of the progeny virions from the host cell. However, in recent years their use has been limited due to several side effects, rapid emergence of antiviral resistant strains and their ease of transmissibility.[10] The recent emergence of pandemic influenza A 2009 (H1N1) strain, the continued circulation of the highly pathogenic avian influenza (H5N1) viruses in humans and the development of resistance against currently used drugs have raised the concerns of a pandemic or a severe epidemic of influenza. Since developing vaccine during a severe epidemic or a pandemic takes around six months, antiviral drugs continue to remain as the primary intervention strategy during the emergency situations.[1] Keeping the above facts in mind, development of newer antiviral strategies to combat influenza virus is the need of the hour. Gene silencing is one such approach that has proved to be an effective and an alternate antiviral strategy to counteract the disease burden posed by influenza A viruses. RNAi is a mechanism for silencing gene expression by promoting degradation of RNA (Table 1). In this process, dsRNA that is introduced directly or via a virus or transposon, is first cleaved by ribonuclease protein Dicer to generate 20–22 nt products generally known as siRNA. The antisense strand of the siRNA is engaged by RNA-induced silencing complex (RISC). The complementary target messenger RNA (mRNA) is engaged by RISC and is sliced and destroyed by Argonaute 2, the catalytic component of the RISC complex, leading to the silencing of expression of the specific gene.[11,12] Short hairpin RNAs (shRNA) are expressed inside the nucleus of the target cell in response to a DNA construct that is delivered to the nucleus via viral or other gene therapy vectors. shRNA is required in low doses compared with siRNA and result in lower immune activation. The use of siRNA is preferred for acute disease conditions (i.e., viral injections) where higher doses could be tolerated; shRNA is used for life threatening diseases or disorders and could be used for long lasting effect.

miRNAs are another class of small non coding RNAs that occur naturally. miRNAs are detected by random cloning and sequencing or by advanced computational predictions and several thousand of miRNAs have been identified in various organisms till date. Inside the nucleus, RNAse polymerase II transcribes miRNA as pri-miRNA that are large RNA precursors containing 5' cap and poly-A tail. A microprocessor complex consisting of Drosha (RNase III enzyme) and Pasha/DGCR8, processes the pri-miRNA. After processing pre-miRNAs are formed that are approximately 70 nts in length and form stem loop structures. Pre-miRNA are exported into the cytoplasm to undergo further processing by Dicer to form miRNA, that are approximately 22 nts in length.[11] miRNAs have been actively involved in several biological processes like cell proliferation, differentiation and apoptosis and boast an impact on many diseases including cancer. A single miRNA can target multiple genes and it has been predicted that around 10–30% of all the human genes are controlled by miRNAs. Either of the processed strands of miRNA can mediate posttranscriptional gene silencing, but miRNAs may show asymmetry by predominantly loading one strand into the RISC. MiRNAs and their targets do not observe complete Watson and Crick base pairing but the complimentarily is usually high between bases 2–8.[17] Many studies have been conducted to explore the role of microRNAs in virology and have shown the utility of this approach in the efficient management of disease. Many viruses, including human immunodeficiency virus (HIV), the hepatitis B and C viruses, coronavirus, respiratory syncytial virus (RSV) and influenza A virus (IAV) have been successfully inhibited or eliminated by specific RNAi. These findings clearly indicate that RNAi has a strong prophylactic and therapeutic potential against viral infections.[11]

The advantage of miRNA is that like shRNA it can mediate silencing for a long duration and with a single application. However, a few things have to be kept in mind while using miRNAs for therapeutic purposes. Firstly, miRNAs are known to directly or indirectly effect the protein expression and, therefore, changes in miRNA expression can have an intense effect on gene regulation. Secondly, if a certain disease occurs due to reduced miRNA expression, miRNA mimic may have to be delivered. To solve this purpose, delivery methods need to be considerably improved in order to achieve desired results. Finally, it is yet to be determined whether one miRNA could be targeted without affecting the other miRNAs belonging to the same family, thus making it difficult to regulate the expression of specific miRNA.[18]

Nucleic acids apart from storing genetic information also possess fascinating catalytic properties. DNA enzymes (Dzs) are small DNA molecules that can cleave target RNA in sequence specific and catalytic manner. Catalytic efficiency of DNAenzymes is comparatively more than any other nucleic acid enzyme, despite of its small size.[19] The 2′hydroxyl group is absent in DNA making it more stable than RNA for the storage of genetic information.[20] DNA enzymes are more stable than ribozymes (Rzs) because of the DNA motifs that provide increased stability. Dzs were generated de novo by Breaker and Joyce[21] and are not reported to occur naturally. Santoro and Joyce discovered the two catalytic Dz motifs, 8–17 and 10–23 that are widely used for gene suppression. In order to make Dzs more stable, they can be cloned into suitable vectors and circularized[22] or can be ligated to an oligonucleotide sequence to make a circoenzyme, that are known to be more stable than DNAenzymes. The addition of 3–3′ inverted nts at the 3′ end of the Dz increases the stability of the molecule as well as its half-life from 70 min to 21 h in human serum.[23]

Rz is another example of catalytic nucleic acid that has shown promise as an antiviral agent. Rz function by cleaving the phosphodiester bonds of nucleic acids. Rz like Dz have been designed to target HIV1 genes and have shown very promising results.[24] Unlike Dzs, Rzs occur in nature and the first natural Rz was discovered more than 30 years ago. Artificial Rzs can, however, be generated by in vitro selection or systemic evolution of ligands by exponential enrichment 'SELEX' techniques. Several classes of Rzs are present in nature and can be distinguished by characteristic structure and mechanisms of action.[25] Hammerhead Rzs and hairpin Rzs are studied most extensively due to their small size and high rate of catalysis.[26]

Hammerhead Rz

A highly conserved catalytic core is present in the hammerhead Rz that cleaves the substrate RNA at NUH triplets (N is any nts, U is uracil and H is any nt but guanidine). However, recent studies have demonstrated the exceptions to the NUH rule.[27]

Specific binding of the Rzs to their target sequence is achieved if the 22-nt-long conserved hammerhead Rz catalytic motifs are flanked by 8-nt-long hybridizing arms that are made complementary to the target RNA sequence.[28]

Hairpin Rz

Hairpin Rz is derived from tobacco rings pot virus satellite RNA. It cleaves the phosphodiester bond at 5′ to the G in the sequence NGUC where N is any nt. The structure of hairpin Rz contains two domains connected by a hinge region.[24] Domain 1 of the Rzs binds the substrate RNA and forms two helical regions that are separated by a pair of single-stranded loops. Domain 2 on the other hand is larger and contains the primary catalytic determinant of the Rz. Antiviral hairpin Rzs have been successfully generated against HIV-1[27] and hepatitis virus.[29]