siRNA: the dark matter of genetics

Diverse mechanisms of RNA silencing

RNA silencing was first discovered as a nucleotide sequence-specific process in plants fungi and nematodes. In each instance a foreign nucleic acid – either a transgene or an RNA – specifically silenced any similar gene that was expressed in the recipient cells. To explain the specificity of RNA silencing it seemed likely that antisense RNA would be involved.

This proposed antisense RNA was eventually found as short (21-25 nucleotide) molecules that were processed from longer double stranded RNA by an enzyme known as Dicer. Dicer is a homologue of E. coli RNAse III with double stranded (ds)RNA binding and helicase domains. The short RNAs, known as small interfering (si)RNAs, determine the sequence specificity of silencing because, by Watson-Crick base pairing, they guide a nuclease known as Slicer to a target RNA.

There are many variations and elaborations of this basic silencing mechanism. For example, in plants and nematodes, one of the most remarkable features of RNA silencing is its ability to spread between cells. Silencing can be initiated locally but manifested throughout the organism. RNA-dependent RNA polymerases are implicated in amplification of siRNAs so that the production and spread of the signal persists for a long time.

A second elaboration on the basic silencing mechanism involves nucleotide sequence specific targeting of DNA and chromatin rather than RNA. This genomic silencing mechanism was first hinted at from findings with plants indicating that transgene RNAs could direct sequence specific DNA methylation. It now seems that the transgenes generated double stranded DNA that was cleaved by Dicer into siRNAs as in silencing of RNA. However, rather than guiding Slicer, it seems that the genome-targeted siRNAs associate with different effectors of silencing acting either at the chromatin or DNA level.

Natural roles of RNA silencing

This diversification of mechanisms is likely to reflect the various natural roles of RNA silencing. For example the basic silencing mechanism is implicated in regulation of gene expression through a process that involves micro (mi)RNAs. A partially double stranded RNA miRNA precursor is processed by Dicer to generate a 21nt miRNA that can act like siRNAs to guide a Slicer nuclease to a target mRNA. In the same way double stranded viral RNA is processed by Dicer to generate viral siRNAs that are thought to guide Slicer so that viral RNA accumulation is impaired. The antiviral role is also thought to involve the mobile signal that moves either with or ahead of the virus so that the silencing mechanism impairs spread of the virus through the infected plant. A genome protection role is associated with the silencing of DNA and chromatin of transposons so that they are not transcribed, or so that their ability to transpose is impaired.

There may also be additional natural roles because there are homologues of proteins involved in silencing that do not have an assigned function. It seems likely that these proteins could be involved in novel silencing pathways. Similarly the existence of many endogenous siRNAs with unknown function, in addition to miRNAs, may also be an indicator that there are many layers of genetic regulation that are based on RNA silencing.

RNA silencing technology

The discovery of siRNAs is relevant to technology as well as to understanding of genetic regulation. The principle is simple – a fragment of a gene is introduced into a cell as dsRNA or as DNA that will give rise to dsRNA. The dsRNA activates the Dicer/Slicer process so that the properties of the affected cell reflect a loss of function in the corresponding gene. In plants the siRNA precursor RNAs can be generated by transformation or by infection with viruses carrying inserts corresponding to each of the genes in the plant genome. This approach can be used for silencing of agronomically deleterious traits in crop plants. It can also be used for identification of gene function in surveys of plant genomes. This approach has a number of advantages, including speed, over more conventional genetic methods using loss of function mutants.

In my lecture I shall summarise key discoveries in the elucidation of the silencing mechanisms. I shall also describe findings about natural silencing mechanisms and evidence for additional diverse roles in genetic regulation. In the final part of my lecture I will discuss potential advantages and disadvantages of silencing technologies for crop improvement and in functional genomics.