Commandeering the Ribosome: Lessons Learned from Dicistroviruses about Translation

Abstract

To replicate, all viruses depend entirely on the enslavement of host cell ribosomes for their own advantage. To this end, viruses have evolved a multitude of translational strategies to usurp the ribosome. RNA-based structures known as internal ribosome entry sites (IRESs) are among the most notable mechanisms employed by viruses to seize host ribosomes. In this article, we spotlight the intergenic region IRES from the Dicistroviridae family of viruses and its importance as a model for IRES-dependent translation and in understanding fundamental properties of translation.

INTRODUCTION

All viruses are completely reliant on the host translational machinery to drive viral protein synthesis. Therefore, viruses have evolved numerous strategies to effectively recruit the host ribosome, from influencing translation factor activity and abundance to commandeering the ribosome via noncanonical mechanisms. Moreover, because of their compact genomes, viruses have found ways to maximize coding capacity by manipulating ribosomes to shift reading frames and translate overlapping open reading frames. One of the most prominent and well-studied translation mechanisms is the internal ribosome entry site (IRES), which is utilized by some RNA and DNA viruses (for a review of viral translation strategies, see reference 1). In contrast to most host mRNAs, which use a cap-dependent mechanism, IRESs are generally structured RNA elements that recruit the ribosome in a cap-independent manner using a subset of translation and auxiliary factors. Although there are many types of viral IRESs, the Dicistroviridae family boasts one of the most well studied and arguably simplest IRESs, the intergenic region (IGR) IRES. This IRES has provided a wealth of insights not only into fundamental viral translation strategies but also into general translation mechanisms such as ribosome-tRNA functions and ribosome reading frame selection (1). In this article, we provide background on this family of viruses and highlight the major findings of this IRES translation mechanism, as well as its use as a research tool.

Dicistroviruses are positive-sense single-stranded RNA viruses with genome sizes ranging from 8 to 10 kb that mainly infect arthropods. Notable members of this family include Cricket paralysis virus (CrPV) and Drosophila C virus, both of which infect genetically amenable Drosophila and have provided excellent models of virus-host interactions in insect cells. Additionally, other members of this family have been linked to honeybee health, namely, Israeli acute paralysis virus and Kashmir bee virus. The name of this viral family derives from its unique dicistronic genome arrangement consisting of two open reading frames (ORFs) (2). Translation of each ORF is directed by a distinct IRES; the 5′ untranslated region (UTR) and the IGR IRES drive the expression of viral nonstructural and structural proteins, respectively, thus enabling a viral strategy using two IRESs to regulate viral protein expression separately and temporally (3, 4). The dicistrovirus 5′ UTR IRESs vary considerably in sequence and size, and to date, there have been very few studies examining its mechanism. A recent report by Abaeva et al. suggested that the 5′ UTR IRES from Halastavi árva virus (HaIV; note that this virus is not currently a member of the family Dicistroviridae, but the ORF1 and ORF2 proteins are most closely related to those of dicistroviruses) does not rely on the structural integrity of the RNA, like a typical IRES element. Instead, the ribosome is recruited to an unstructured region flanking the AUG start codon of the IRES, where it undergoes retrograde scanning before initiating translation (5). Moreover, the 5′ UTR IRESs of both HaIV and the related Rhopalosiphum padi virus demonstrate cross-kingdom activity, suggesting that the mechanism of translation initiation is conserved. Whether all dicistrovirus 5′ UTR IRESs function similarly and how sequence diversity between the 5′ UTR IRESs affects translation in different host cells remain to be investigated.

In contrast to the 5′ UTR IRES, the IGR IRES is ∼190 nucleotides in length, is highly structured, and is, in general, conserved throughout the members of the family Dicistroviridae. Several biochemical and structural studies, via crystallography and cryo-electron microscopy analyses, have revealed that the IGR IRES is composed of three pseudoknots (termed pseudoknot I [PKI], PKII, and PKIII) (6–8); PKII and PKIII together form the core ribosome-binding domain that recruits both the 40S and 60S ribosomal subunits, while PKI acts to position the 80S ribosome at the correct initiation codon (Fig. 1A). The most remarkable aspect of the IGR IRES is that it requires no translation initiation factors or initiator Met-tRNA_i to recruit the ribosome and initiate translation from a non-AUG codon (Fig. 1C) (3). The IGR IRES binds with high affinity to the conserved core of the ribosome and spans all three tRNA binding sites (A, P, and E sites; Fig. 1B) (7). In an extraordinary case of molecular mimicry, the PKI domain adopts a tRNA-like fold to mimic the anticodon stem of a tRNA, thereby enabling access to the ribosome decoding center (A site; Fig. 1B) (7, 9–11). From here, the IRES undergoes a pseudotranslocation step that involves the movement of the PKI domain to the P site, thereby resulting in delivery of the first aminoacyl-tRNA to the non-AUG codon at the A site, essentially setting the ribosome into an elongation-competent state (Fig. 1C) (7, 12). By mimicking a fundamental step in the host translation cycle, the dicistrovirus IGR IRESs can entirely bypass the heavily regulated initiation step of translation. Indeed, the IGR IRES can direct translation in eukaryotes from yeasts to humans and, according to a recent report, in prokaryotes, suggesting a universal translation mechanism across kingdoms (13, 14).

FIG 1.

(A) Secondary structure of the dicistrovirus IGR IRES. (B) Cryo-electron microscopy structure of a human ribosome bound to the CrPV IGR IRES at 3.5 Å. The E, P, and A tRNA binding sites are indicated. (Republished from Cell [7].) (C) Model of the IGR IRES factorless translation mechanism. The IGR IRES first occupies the A site of the ribosome. After eukaryotic elongation factor 2 (eEF2)-assisted pseudotranslocation to the P site, an amino-acyl tRNA is delivered by eEF1A (thick black arrow). Without tRNA delivery, the IGR IRES may spontaneously back translocate to the P site (thin gray arrow).

How does the IGR IRES manipulate the ribosome? The IGR IRES makes extensive contacts with the 80S ribosome to recruit both the 40S and 60S subunits. For example, interactions between stem-loops SLIV and SLV of the IRES and uS7 and eS25 of the 40S subunit, respectively, primarily direct 40S binding, whereas the conserved L1.1 domain of the IRES interacts with the L1 stalk of the large ribosomal subunit (Fig. 1B) (7, 15). Besides these initial binary IRES-ribosome interactions, cryo-EM structures of the pseudotranslocation step reveal that the IGR IRES interacts with distinct regions of the ribosomal core by adopting conformational states that resemble the movement of tRNAs at the A, P, and E sites during translocation (12), indicating that the IRES actively manipulates the entire translocation process as the IRES moves through the ribosome (Fig. 1B). Indeed, a recent report showed that specific mutations in the variable loop region (VLR) of the IRES may drive the pseudotranslocation step specifically (16). Thus, how the IGR IRES manipulates the ribosome offers insights into the core mechanics of translation such as ribosome-tRNA dynamics during translocation.

While the IGR IRES serves as a powerful model IRES for other viral IRESs, it has also proven to be an important tool for studying other aspects of translation. Since the IGR IRES bypasses the need to reconstitute the translation initiation step, which requires at least nine core translation factors, the IGR IRES can be exploited in a simple purified system for studying translation elongation, and termination, and its initiation factor-independent nature permits investigations into whether a process is initiation factor dependent. For example, the use of the IGR IRES has provided significant insights into microRNA-dependent translational control and nonsense-mediated mRNA decay. It is anticipated that IGR IRES translation can be utilized to probe many aspects of translational regulation.

Studies on the IGR IRES mechanism have also provided new insights into ribosome heterogeneity. As mentioned above, ribosomal protein eS25 interacts with a specific IGR IRES domain to mediate function (15, 17). However, it has also been shown that depletion of ribosomal proteins (e.g., eS6) leads to a specific block in IGR IRES-dependent translation but not overall cap-dependent translation, thus hinting that the ribosome may have specificity for specific mRNAs (i.e., the ribosome filter hypothesis) (18). Indeed, this process extends past IGR IRES translation. Depletion of ribosomal proteins also affects other viral and cellular IRESs such as the more complex hepatitis C virus IRES (17). In addition to ribosomal proteins, the IGR IRES has also expanded our understanding of the role of pseudouridylation (pseudoU), an RNA modification within the rRNA (19). PseudoU of rRNA is highly specific and conserved; however, the exact biological impact was elusive. Jack et al. demonstrated that lack of pseudoU resulted in reduced ribosome binding to the IGR IRES and a decrease in IGR IRES-mediated but not cap-dependent translation. These experiments suggest that ribosomes may be heterogeneous in nature and may be regulated via RNA modification and/or posttranslational modification to modulate the translation of a specific mRNA. It is likely that the regulatory effects of ribosomal proteins and modifications extend beyond IRES-mediated translation; however, this has given an initial hint of their biological roles.

As viral genomes are extremely compact, with limited sequence space, viruses have evolved a variety of strategies to expand their coding capacity. Recently, a subset of IGR IRESs have been shown to facilitate the translation of a +1 frame overlapping ORF, termed ORFx, in addition to translation in the 0 frame (10, 20). Specific mutations within the tRNA-like PKI domain uncouple 0 and +1 frame translation, which is correlated with distinct, subtle conformations, which may drive reading frame selection (10). Unlike most viral RNA signals that shift reading frames of translating ribosomes such as −1 frameshift signals in retroviruses and coronaviruses, IGR IRES-mediated 0 and +1 frame translation represents a novel translation initiation strategy to increase the coding capacity of a viral genome. Normally, translation is highly accurate and the ribosomal reading frame is maintained. As the IGR IRES adopts a tRNA-like domain that recruits the ribosome directly and sets it into a specific reading frame, it provides a unique yet powerful model for elucidating not only RNA-driven recoding strategies in virus infections but also fundamental mechanisms of ribosome reading frame maintenance.

Overall, many different RNA structures have defined functions to facilitate viral infection. Remarkably, the ∼190-nucleotide IGR IRES is an unprecedented minimal RNA structure that can hijack a 3-MDa molecular machine to facilitate translation and does so without the need of any additional factors, all while selecting a reading frame. The IGR IRES continues to prove invaluable for understanding mechanisms of viral IRES-dependent translation initiation, as well as serving as a useful research tool. Finally, this IRES truly highlights the ingenuity and diverse means that viruses have evolved to exploit host processes for the expression of their genomes.