First, RNA, the master switch. In neuronal stem cells, genes for the proteins that confer neural specificity are silenced. A key mediator of this cellular gag order is the neuronal restricted silencing factor/RE-1 silencing transcription factor, or NRSF/REST. This is a zinc finger protein that binds to a specific DNA response element, the neuron restrictive silencer element (NRSE/RE-1). This element, of 21-23 base pairs, is highly conserved and found in a host of genes that regulate neuronal development and function. Gage and colleagues have found that a small double-stranded RNA (dsRNA) relieves inhibition at the silencer element site, allowing genes downstream to be transcribed.

First author Tomoko Kuwabara made the connection between this non-coding dsRNA and gene activation after he found NRSE/RE-1 complements in samples of total RNA from adult hippocampal stem cells. What would a small RNA that is complementary to a DNA response element be doing in stem cells? To answer this, Kuwabara made a construct containing the RNA and expressed it in hippocampal stem cells. Sense RNA had no effect; neither did antisense, but when expressed together they caused the cells to differentiate into large flat clusters with long processes, just like developing neurons.

So what’s the dsRNA doing? Could it be acting like a small inhibitory RNA to suppress the NRSF/REST silencer? Hardly likely, say the authors, because there is no NRSE/RE-1 element in that gene, only in the genes it silences. Yet when Kuwabara slipped luciferase reporter genes containing the response element into the stem cells, the dsRNA turned on the luciferase, indicating that it does act at the transcriptional level. It also induced transcription of other genes that are regulated by the NRSE/RE-1 element, including the acetylcholine receptor, the glutamine receptor, synapsin, sodium channel gene, and the superior cervical ganglion protein 10 (SCG10).

To further characterize the relationship between the DNA response element and its dsRNA homolog, Kuwabara examined expression of luciferase driven by the promoter region from the glutamine receptor gene. When the authors mutated the NRSE region of this promoter, luciferase was expressed, indicating relief from suppression. However, the level of luciferase was only about half that achieved by adding dsRNA to cells expressing luciferase from an intact glutamine receptor promoter. “This suggests that NRSF/REST converts to an activator in the presence of NRSE dsRNA,” write the authors.

How this switch from repressor to activator is made is unclear, but current evidence points to regulation of chromatin structure. Kuwabara found that in neural stem cells, where the neuron-specific genes are silenced, the suppressor is found bound to proteins, such as histone deacetylase and methyl-CpG binding protein, that are associated with transcriptionally inactive chromatin. In contrast, these associations are abolished in neurons. Binding of the NRSF/RE-1 suppressor to the dsRNA may be crucial for the switch, too, because the authors found that the suppressor actually binds more tightly to dsRNA than to DNA.

Overall, the data suggest a novel and potentially wide-ranging role for small double-stranded RNAs, i.e., the ability to regulate global gene expression patterns by turning repressors into activators.

RNA as self-cannibalizing enzyme. The Yale scientists describe a different switching mechanism for RNA. Working with gram-positive bacteria, first author Wade Winkler demonstrates that the messenger RNA for the enzyme glutamine-fructose-6-phosphate amidotransferase (glmS), which catalyzes the synthesis of glucosamine-6-phosphate (GlcN6P), has its own catalytic activity. It turns out that the mRNA, beyond coding for glmS, is also a nuclease, and no ordinary one at that—it digests itself.

Breaker became interested in this RNA because a substantial portion of its 5' untranslated region is highly conserved, suggesting a structure/function relationship. One possible function of RNA is to work as a riboswitch. Riboswitches, such as RNA-splicing ribozymes, can be activated by small molecules, and some have even been engineered to be ligand-dependent. But if glmS mRNA is a riboswitch, what might its effector molecule be?

To answer this question, Winkler incubated the glmS mRNA with the product of the glmS reaction, GlcN6P. Sure enough, he found that the ribozyme was a thousandfold more active in the presence of the small molecule than in its absence.

“RNA can now be the active element that switches off repressible genes in response to the concentration of a cellular metabolite,” writes Tom Cech, from the University of Colorado, Boulder, in an accompanying News and Views. When GlcN6P, the product of glmS, reaches a certain concentration in the cell, production of the enzyme is shut off because the glucosamine inactivates the enzyme’s mRNA. This makes this RNA a transcript, receptor, enzyme, and metabolic regulator—in short, a multi-tasker if ever there was one.—Tom Fagan.

References:
Kuwabara T, Hsieh J, Nakashima K, Taira K, Gage FH. A small modulatory dsRNA specifies the fate of adult neural stem cells. Cell. 2004 March 18;116:779-793. Abstract

Winkler WC, Nahvi A, Roth A, Collins JA, Breaker RR. Control of gene expression by a natural metabolite-responsive ribozyme. Nature. 2004 March 18;428:281-286. Abstract

Cech T. RNA finds a simpler way. Nature 2004 March 18;428:263-264. Abstract