Biochemical Society Transactions

Post-Transcriptional Control: mRNA Translation, Localization and Turnover

Cytoplasmic mRNA: move it, use it or lose it!

Mark J. Coldwell, Nicola K. Gray, Matthew Brook

Abstract

Once an mRNA is synthesized and processed, the immediate translation and later destruction of the transcript is not as inevitable as the central molecular biology dogma suggests. Interest in the field of post-transcriptional control continues to grow rapidly, as regulation of these multiple steps in gene expression is implicated in diverse aspects of biology such as metabolism, neurology, reproduction and viral lifecycle regulation. Researchers who utilize various combinations of human studies, animal models, cellular, genetic, biochemical and molecular techniques were brought together at the University of Edinburgh to discuss their latest findings. In this article, we introduce the content of the related reviews presented in this issue of Biochemical Society Transactions which together illustrate a major theme of the meeting content: namely the need to understand how dynamic changes in mRNP (messenger ribonucleoprotein) complexes modulate the multifunctionality of regulatory proteins which link different post-transcriptional regulatory events.

  • gene expression
  • localization
  • microRNA (miRNA)
  • mRNA
  • repression
  • translation
  • turnover

Introduction

The broad characterization of the processes of gene expression into transcription and translation does a disservice to the many researchers working on the complex facets of these key steps, and the numerous events that precede and succeed them. Recent advances in our understanding of processes that occur following the synthesis, processing and nuclear export of an mRNA (in eukaryotic models) demonstrate the importance of post-transcriptional control in the cytoplasm in regulating the expression levels and intracellular localization of proteins and the multiplicity of proteins produced from a single mRNA. The importance of these mechanisms is highlighted by the contribution of their mis-regulation to a growing number of disease processes.

Transcripts in the cytoplasm have three broad fates, as detailed below and illustrated in Figure 1, and the meeting showcased research which continues to both surprise and encourage further endeavours in this rapidly evolving area. Although Figure 1 illustrates an mRNA as an essentially naked transcript, each mRNA interacts with a plethora of binding partners, both protein and RNA in nature, forming mRNP (messenger ribonucleoprotein) complexes, and the composition of these complexes influences the translation, localization or turnover of mRNAs.

Figure 1 The cytoplasmic life of a eukaryotic mRNA

During translation, binding of the 40S ribosomal subunit to the mRNA is facilitated via eIFs in the process of cap-dependent initiation. The ribosome may also be recruited to IRESs within the 5′-UTRs, in conjunction with ITAFs. This and subsequent steps in initiation are subject to regulation. Proteins that activate translation generally stimulate the formation or activity of eIFs, whereas translational repression may be brought about by a number of mechanisms affecting factor recruitment to the transcript. The role of miRNAs in driving this process has been the focus of much recent work. The regulated turnover of mRNA may be induced in response to signals such as a PTC in an aberrant mRNA or due to decapping or deadenylation making the transcript more unstable and a target for exo- and/or endo-nucleases. CYFIP, cytoplasmic FMR1 (fragile X mental retardation 1)-interacting protein; miRNP, microRNA–ribonucleoprotein complex.

One of the key themes to emerge is the multifunctional nature of some of the regulatory proteins involved in these processes. Some trans-acting factors have been found to influence translation in diverse manners or to affect both mRNA translation and stability or localization. Dynamic remodelling of mRNP complexes appears to link the regulatory fates.

Signals and factors that promote decoding of the mRNA

The cap and poly(A) tail of the mRNA are primary determinants of translation efficiency and have well-characterized binding partners that promote the assembly of complexes of eIFs (eukaryotic initiation factors) and ribosomal subunits in order to initiate translation. The stepwise assembly of the canonical initiation factors and the process of cap-dependent scanning to the initiation codon have been detailed elsewhere [1,2], but it is clear that our knowledge of this fundamental process is far from complete and, in this regard, Stevenson et al. [3] discuss the application of fluorescence quantification methods to single-molecule studies of the translation machinery to enable greater understanding of initiation. The simplest description of this process underlines the central importance of the eIF4F complex (Figure 1), consisting of the cap-binding protein eIF4E, the eIF4A helicase, which removes structures formed by base-pairing within the 5′-UTR (untranslated region), and eIF4G, which acts as the ‘scaffold’ that binds eIF4E and eIF4A as well as eIF3. This complex sits on the 40S ribosomal subunit and helps its recruitment to the mRNA. Surprisingly, PABPs [poly(A)-binding proteins] [4] which bind the poly(A) tail also participate in the initiation process and link translation to mRNA turnover via a functional interaction with the cap-binding complex (eIF4F) which effectively circularizes mRNAs.

Despite the fundamental nature of the cap and poly(A) tail, it has emerged that these structures are surprisingly dynamic, and Victoria Cowling [5] reviews novel evidence that the transcription factor Myc can up-regulate capping, leading to a subsequent activation of protein synthesis and hence cell proliferation and transformation. Likewise, the length of the poly(A) tail may be dynamically altered, being subject both to regulated polyadenylation in the cytoplasm, as discussed by Richter [6], and deadenylation, as discussed by Weiderhold and Passmore [7].

Within the mRNA, primary sequences and secondary and tertiary structures also influence translational efficiency. For the most part, these exist in the 5′- and 3′-UTRs and often denote binding sites for mRNA-specific regulatory protein or RNA partners, which can promote the assembly of the ‘canonical’ initiation factors, and therefore ribosomal subunit recruitment on the mRNA. However, basal components of the translational machinery can also participate in mediating mRNA-specific activation (or even repression) as highlighted by recent work on the PABPs (discussed by Burgess and Gray [4]).

Moreover, basal components of the translational machinery can also have ‘moonlighting’ roles, for instance ribosomal proteins have non-translational roles in gene expression, and their localization to sites of transcription in the nucleus is described by De and Brogna [8].

It is becoming increasingly clear that different mechanisms have evolved to permit some cellular mRNAs to maintain translation under circumstances where global translation is arrested. One well-defined mechanism is by the recruitment of the translation machinery directly to IRESs (internal ribosome entry sites). Employing this alternative mechanism usually requires extra non-canonical factors collectively termed ITAFs (IRES-specific trans-acting factors), with King et al. [9] highlighting the discovery of an increasing number of these proteins and their targets due to advances in RNA affinity chromatography and proteomics. Internal entry of ribosomes was first discovered in the Picornaviridae and continues to be observed in many other RNA viruses, including HIV, although this virus may also initiate translation via the classic cap-dependent mechanism as reviewed by Chamond et al. [10], underlining the complexity of initiation processes. In many cases, IRES-mediated translation occurs concurrently with the cleavage of initiation factors during infection, resulting in a loss of host protein synthesis and allowing the hijack of factors for the production of viral proteins. However, studies of DNA viruses, as highlighted by Derek Walsh [11], illustrate that several viral families positively regulate the expression levels, activity and assembly of components of the eIF4F complex, allowing efficient maintenance of cap-dependent initiation during infection.

Viral models have not only led to many important insights into the steps of translation initiation, but have also demonstrated how this process can be linked to a previous termination event permitting efficient use of their compact genomes. For example, Yan et al. [12] discuss how a second ORF (open reading frame) may begin translating in the absence of both a termination and an initiation codon via a co-translational ‘stop–carry-on’ event, or a –1 frameshift may occur directly after termination of translation of an ORF in order to initiate in the correct frame of an immediately downstream ORF, as described by Mike Powell [13]. A particularly novel instance of ribosomal frameshifting has recently been uncovered in the human mitochondrial genome, which deviates from the universal genetic code in a number of ways, including the presence of AGG and AGA termination codons. Several hypotheses for this phenomenon, including the presence of non-canonical translation release factors have been postulated, but Richter et al. [14] explain how recent work has shown that the terminating ribosome undergoes a −1 frameshift, meaning a standard UAG termination codon is utilized. This finding demonstrates just how much basic knowledge pertaining to basal translational components and mechanisms is still surprisingly lacking, and the importance of continually updating undergraduate textbooks as our knowledge of this field develops exponentially!

Signals and factors that suppress mRNA translation

As the assembly of the initiation factors and ribosomes at the 5′-end of the mRNAs is a crucial part of initiation, and often considered to be the rate-limiting step in translation, it is not surprising that disruption of their assembly can have a drastic effect on translation of all or specific subsets of mRNAs. One emerging theme is the role of repressor proteins which act by preventing eIF4E from binding to eIF4G, thereby blocking eIF4F complex formation. These include FMRP (fragile X mental retardation protein), as discussed by Joel Richter [6], although this protein may also influence elongation. Understanding the complex and multifaceted mechanisms of translational regulation of specific neuronal mRNAs by FMRP will be important in understanding the disease process in patients with fragile X syndrome.

The removal or sequestration of factors binding to the respective ends of the mRNA can be used not only as a mechanism to repress translation, but also to initiate the destruction of the mRNA. There are numerous eukaryotic deadenylases which catalyse the shortening of the poly(A) tail, as discussed in this issue by Wiederhold and Passmore [7], including the highly conserved Ccr4–Not complex. Deadenylation may occur before an mRNA has been translated to store it in a translationally inactive state (as is often the case during development), or as a way of removing an mRNA from the pool of those undergoing decoding, as outlined by Marnef and Standart [15], who describe the role of Pat proteins in this process. This sequestration is often accompanied by the binding of proteins to specific sequences/structures within the 3′-UTR of the mRNA and may result in the aggregation of mRNAs in cytoplasmic P-bodies (processing bodies) (see below).

Deadenylation is also linked to other post-transcriptional control points, including translational repression via miRNAs (microRNAs), with components of the RISC (RNA-induced silencing complex) capable of recruiting deadenylases to an mRNA [4,7]. This is yet another mechanism ascribed to miRNAs, alongside work showing that they interfere with multiple steps in initiation and also post-initiation events. miRNAs have been the focus of a huge amount of research effort in recent years, and were discussed extensively at the meeting. Indeed, the oral presentation prize awarded to David Ferland-McCollough [16] concerned the role of miRNAs in diabetes using both animal models and human studies. This work adds to our understanding that the expression patterns of particular miRNAs may be useful as a diagnostic marker in many disease states. This theme is also touched upon in the discussion by Horvilleur et al. [17] of how different miRNAs may be essential in the deregulation of translation observed during the development of B-cell lymphomas. Interestingly, not all miRNAs regulate translation: Lewis and Jopling [18] discuss the role of miR-122 in up-regulating synthesis of hepatitis C viral RNA during infection, alongside the physiological targets of this miRNA.

Signals and factors that promote degradation of the mRNA

Deadenylation of an mRNA not only reduces the translatability of the mRNA, but is often the preliminary step in inducing the decay of an mRNA, as discussed by Wiederhold and Passmore [7] and Reznik and Lykke-Andersen [19]. With ever-increasing numbers of types of functional RNA being identified, there is a corresponding increase in the numbers of RNA-surveillance and turnover pathways. Reznik and Lykke-Andersen [19] present a contemporary overview of the mechanisms of quality control and, where appropriate, degradation, applied to both coding and non-coding RNAs. Furthermore, whereas many experimental techniques look at the individual steps of post-transcriptional control in isolation, Laloo et al. [20] describe a method which aims to assess them all via a single assay.

As a single mRNA may be subjected to multiple quality-control and regulatory processes in its lifetime, an imbalance of the pathways involved in ensuring the appropriate destruction of mRNAs can either remove mRNAs prematurely or allow some mRNAs to survive and be translated beyond their intended lifespan. A review of this phenomenon in chondrocytes by Tew and Clegg [21] describes how deregulation of mRNA decay, as well as miRNA-mediated translational repression, may underlie the aetiology of some degenerative joint diseases. Furthermore, Tiedje et al. [22] review emerging evidence for the multiple interactions between a regulator of mRNA decay, tristetraprolin, and a number of proteins involved in regulating both mRNA stability and translation, again showing how related mRNP complexes can dictate alternative fates for an mRNA. Interestingly, the ubiquitination pathway commonly implicated in protein turnover may also be directly involved in mRNA decay, as ubiquitin ligases can directly bind mRNAs (discussed by Cano et al. [23]), demonstrating that post-transcriptional and post-translational control mechanisms may also be closely integrated. Similarly, studies of one of the central regulators of apoptosis, Bcl2 (reviewed by Willimott and Wagner [24]), add even more layers of complexity with translation, mRNA decay and post-translational activity each being regulated.

NMD (nonsense-mediated decay) rids the cell of mRNAs containing PTCs (premature-termination codons) before they are able to undergo multiple rounds of translation and produce truncated proteins with deleterious effects on the cell. NMD is a central method of intracellular quality control of gene expression and two reviews detail the different pathways [19] and many factors that comprise the NMD machinery [25]. Interestingly, recent work has revealed that the factors that mediate NMD also regulate mRNAs without PTCs, and Vicente-Crespo and Palacios [26] review the importance of these factors during development.

Links between mRNA localization and translation/decay

The signals present within an mRNA not only define the translational efficiency or stability of the transcript, but may also aid in the transport of the mRNA to defined sites in the cell, where these processes can be locally co-ordinated. In the case of localizing a single mRNA in order for multiple protein molecules to be synthesized at a specific locale, this circumvents the need for transport of each resulting product. These mRNAs are normally transported to the site of localized translation in a translationally repressed state. In this issue, Joel Richter [6] reviews the activation and repression of translation of specific mRNAs by FMRP and CPEB1 (cytoplasmic polyadenylation element-binding protein 1) in neurons, cells with extended morphologies that utilize localized translation at synapses.

The aggregation of repressed mRNAs to cytoplasmic P-bodies has been observed in all metazoan cells, as discussed by Marnef and Standart [15], and P-bodies are dynamically related to other microscopically observable foci such as stress granules, which are discussed by Yagüe and Raguz [27]. Together, these represent sites where translationally repressed mRNAs are sequestered before being either returned to the translating pool or degraded.

In the cases of both localized translation and translational repression/mRNA turnover at discrete foci, the role of subsets of dynamically regulated mRNPs is the subject of intense study which will probably yield exciting new discoveries in the very near future.

Future perspectives

When organizing this meeting, we hoped to bring together researchers from the fields of mRNA translation, localization and turnover to provide an opportunity to share knowledge and ideas and encourage collaboration. The work presented in this issue of Biochemical Society Transactions demonstrates the complex interplay between these processes. This meeting highlighted that paramount to elucidating the cytoplasmic life story of an mRNA is an understanding of the multifunctionality of key regulators and the dynamics of mRNP remodelling. A given mRNA may be subject to numerous facets of the processes outlined in this review during its lifetime and, degradation excepted, these processes can be repetitive and constantly under review according to the exact cellular context at any given time. While the complete life story of an mRNA may not yet be written in full, many of the chapters in the biography are becoming clearer and will be the focus of much future research.

Footnotes

  • Post-Transcriptional Control: mRNA Translation, Localization and Turnover: A Biochemical Society Focused Meeting held at University of Edinburgh, U.K., 8–10 June 2010. Organized and Edited by Matthew Brook (Edinburgh, U.K.), Mark Coldwell (Southampton, U.K.), Simon Morley (Sussex, U.K.) and Nicola Gray (Edinburgh, U.K.).

Abbreviations: eIF, eukaryotic initiation factor; FMRP, fragile X mental retardation protein; IRES, internal ribosome entry site; ITAF, IRES-specific trans-acting factor; miRNA, microRNA; mRNP, messenger ribonucleoprotein; NMD, nonsense-mediated decay; ORF, open reading frame; PABP, poly(A)-binding protein; P-body, processing body; PTC, premature-termination codon; UTR, untranslated region

References

View Abstract