This page has been archived and is no longer updated

 
The Elaboration of the Central Dogma 
Unit 4: Expanding the Central Dogma
Progress
Loading ...

4.1  RNA FunctionsCitation

 

RNA
RNA
© 2009 Nature Education All rights reserved. View Terms of Use
The central dogma of molecular biology suggests that DNA maintains the information to encode all of our proteins, and that three different types of RNA rather passively convert this code into polypeptides. Specifically, messenger RNA (mRNA) carries the protein blueprint from a cell's DNA to its ribosomes, which are the "machines" that drive protein synthesis. Transfer RNA (tRNA) then carries the appropriate amino acids into the ribosome for inclusion in the new protein. Meanwhile, the ribosomes themselves consist largely of ribosomal RNA (rRNA) molecules.
However, in the half-century since the structure of DNA was first elaborated, scientists have learned that RNA does much more than simply play a role in protein synthesis. For example, many types of RNA have been found to be catalytic--that is, they carry out biochemical reactions just like enzymes do. Furthermore, many other varieties of RNA have been found to have complex regulatory roles in cells.
Thus, RNA molecules play numerous roles in both normal cellular processes and disease states. Generally, those RNA molecules that do not take the form of mRNA are referred to as noncoding, because they do not encode proteins. The involvement of noncoding mRNAs in many regulatory processes, their abundance, and their diversity of functions has led to the hypothesis that an "RNA world" may have preceded the evolution of DNA and proteins (Gilbert, 1986).

Noncoding RNAs in Eukaryotes

In eukaryotes, noncoding RNA comes in several varieties, most prominently transfer RNA (tRNA) and ribosomal RNA (rRNA). As previously mentioned, both tRNA and rRNA have long been known to be essential in the translation of mRNA to proteins. For instance, Francis Crick proposed the existence of adaptor RNA molecules that were able to bind to the nucleotide code of mRNA, thereby facilitating the transfer of amino acids to growing polypeptide chains. The work of Hoagland et al. (1958) indeed confirmed that a specific fraction of cellular RNA was covalently bound to amino acids. Later, the fact that rRNA was found to be a structural component of ribosomes suggested that like tRNA, rRNA was also noncoding.
In addition to rRNA and tRNA, a number of other noncoding RNAs exist in eukaryotic cells. These molecules assist in many essential functions, which are still being enumerated and defined. As a group, these RNAs are frequently referred to as small regulatory RNAs (sRNAs), and, in eukaryotes, they have been further classified into a number of subcategories. Together, these various regulatory RNAs exert their effects through a combination of complementary base pairing, complexing with proteins, and their own enzymatic activities.

Small Nuclear RNAs

One important subcategory of small regulatory RNAs consists of the molecules known as small nuclear RNAs (snRNAs). These molecules play a critical role in gene regulation by way of RNA splicing. snRNAs are found in the nucleus and are typically tightly bound to proteins in complexes called snRNPs (small nuclear ribonucleoproteins, sometimes pronounced "snurps"). The most abundant of these molecules are the U1, U2, U5, and U4/U6 particles, which are involved in splicing pre-mRNA to give rise to mature mRNA.

MicroRNAs

Another topic of intense research interest is that of microRNAs (miRNAs), which are small regulatory RNAs that are approximately 22 to 26 nucleotides in length. The existence of miRNAs and their functions in gene regulation were initially discovered in the nematode C. elegans (Lee et al., 1993; Wightman et al., 1993). Since the time of their discovery, miRNAs have also been found in many other species, including flies, mice, and humans. Several hundred miRNAs have been identified thus far, and many more may exist (He & Hannon, 2004).
miRNAs have been shown to inhibit gene expression by repressing translation. For example, the miRNAs encoded by C. elegans, lin-4 and let-7, bind to the 3' untranslated region of their target mRNAs, preventing functional proteins from being produced during certain stages of larval development. Most miRNAs studied thus far appear to control gene expression by binding to target mRNAs through imperfect base pairing and subsequent inhibition of translation, although some exceptions have been noted.
Additional studies indicate that miRNAs also play significant roles in cancer and other diseases. For example, the species miR-155 is enriched in B cells derived from Burkitt's lymphoma, and its sequence also correlates with a known chromosomal translocation (exchange of DNA between chromosomes).

Small Interfering RNAs

Small interfering RNAs (siRNAs) are yet another class of small RNAs. Although these molecules are only 21 to 25 base pairs in length, they also work to inhibit gene expression. Specifically, one strand of a double-stranded siRNA molecule can be incorporated into a complex called RISC. This RNA-containing complex can then inhibit transcription of an mRNA molecule that has a sequence complementary to its RNA component.
siRNAs were first defined by their participation in RNA interference (RNAi). They may have evolved as a defense mechanism against double-stranded RNA viruses. siRNAs are derived from longer transcripts in a process similar to that by which miRNAs are derived, and processing of both types of RNA involves the same enzyme, Dicer (Figure 1). The two classes appear to be distinguished by their mechanisms of repression, but exceptions have been found in which siRNAs exhibit behavior more typical of miRNAs, and vice versa (He & Hannon, 2004).

Small Nucleolar RNAs

Inside the eukaryotic nucleus, the nucleolus is the structure where rRNA processing and ribosomal assembly take place. Molecules called small nucleolar RNAs (snoRNAs) were isolated from nucleolar extracts because of their abundance in this structure. These molecules function to process rRNA molecules, often resulting in the methylation and pseudouridylation of specific nucleosides. These modifications are mediated by one of two classes of snoRNAs: the C/D box or H/ACA box families, which generally mediate the addition of methyl groups or the isomerization of uradine in immature rRNA molecules, respectively.
Scitable by Nature Education Nature Education Home Learn More About Faculty Page Students Page Feedback