7.1. An Overview of Eukaryotic mRNA Translation and Diverse Strategies Employed by Different Viruses to Inhibit Host Translation
All viruses rely on their infected host cells for mRNA translation as they do not encode genes for translation machinery. In many cases, viral and cellular mRNA translation represents a significant conflict of interest to compete for translation machinery. Not surprisingly, many viruses target translation processes to gain a translational advantage for viral mRNAs over cellular mRNAs. More efficient utilization of the translation machinery by viral mRNAs would ultimately put host mRNA translation at a disadvantage and contribute to host protein synthesis shutoff.
Messenger RNA translation is the most energy-consuming process in a cell. Cap-dependent translation is the dominant mode of eukaryotic mRNA translation with three major steps: initiation, elongation, and termination. In eukaryotic cells, the initiation factor 2 (eIF2) forms a ternary complex (TC) with initiator-methionine tRNA (Met-tRNAi) and GTP. Consequently, TC binds to the 40S ribosome in complex with eIF1, eIF3, and eIF5 to form the 43S pre-initiation complex. A rate-limiting translation initiation step is followed by the recruitment of a hetero-trimeric complex called eIF4F on mRNA. The eIF4F complex consists of an m7
G cap-binding protein eIF4E, RNA helicase eIF4A, and scaffold protein eIF4G. Once formed, the eIF4F complex binds the m7
G cap of the mRNA, and the scaffold protein eIF4G interacts with the poly(A) binding protein (PABP) bound to the 3′ poly(A) tail to promote transient 5′-3′ communication, which is known as the closed-loop model. Subsequently, by binding to the eIF3 complex, eIF4G helps to recruit the 43S pre-initiation complex on the mRNA to form the 48S pre-initiation complex. The 48S pre-initiation complex then scans the mRNA in the 5′→3′ direction until it reaches the start codon, usually AUG. Once the start codon is recognized, eIF5B mediates the hydrolysis of eIF2-bound GTP. This change prompts the joining of the 60S ribosome to the 48S pre-initiation complex to form an elongation-competent 80S complex [95
]. This sophisticated process is followed via translation elongation, termination, and ribosome recycling. Since translation initiation is the rate-limiting step of translation, many viruses target different proteins involved in this process.
The multi-subunit complex eIF4F is a primary target for viruses to hijack host translation. All three proteins in the eIF4F complex are targeted by different viruses belonging to different families. FMDV protease 3C cleaves eIF4A to block host translation [97
]. HSV-1 VHS protein binds to eIF4A to gain proximity to cleave and degrade host mRNAs [98
]. Enterovirus (EV) 2A, various retrovirus proteases, FMDV leader protease, and feline calicivirus 3C protease cleave eIF4G [99
]. Viral proteins, such as rhinovirus 2A and rotavirus NSP3, bind eIF4G and displace PABP to prevent the closed-loop conformation required for translation in many host mRNAs [103
]. Cap-binding protein eIF4E is a common target for many viruses to reduce host translation. Some viral proteins bind and recruit eIF4E to viral mRNA to gain a translation advantage. Notable examples include the Adenovirus shutoff protein 100K and turnip mosaic virus (TMV) VPg protein, while Cripavirus
and influenza virus perform this function using unknown proteins [105
]. Enteroviruses (EV) use a distinctive approach to target eIF4E, by which virus-induced miR-141 suppresses eIF4E mRNA translation to limit the availability of this protein [109
]. Many viruses also target the phosphorylation state of translation repressor protein eIF4E binding protein 1 (4EBP1). 4EBP1 is a translation repressor that limits the availability of eIF4E to form the eIF4F complex [110
]. Hyperphosphorylated (four sites) 4EBP1 releases eIF4E for eIF4F complex formation. SV40 small t antigen carries out the PP2A-dependent dephosphorylation of 4EBP1, whereas VSV M protein and reovirus p17 dephosphorylate 4EBP1 through inactivating Akt-mTOR, which is required for 4EBP1 hyperphosphorylation [111
]. In doing so, viruses induce host shutoff since the vast majority of host mRNAs depend on eIF4F-complex-reliant cap-dependent translation.
Viruses are known to alter the function of multifactor initiation complexes other than eIF4F to induce host shutoff. Alphaviruses (Sindbis and Semliki Forest virus) induce phosphorylation of eIF2α to block global host translation [114
]. eIF3 is a multiprotein complex composed of 13 different subunits, several of which are known targets for viruses to decimate host mRNA translation. Measles N protein binds eIF3g, whereas rabies M protein binds eIF3h to impede host translation [115
]. Similarly, SARS-CoV and infectious bronchitis virus (IBV) spike protein binds eIF3f to impair host mRNA translation [117
]. Enteroviruses encode 3C proteases that cleave eIF5B, thus impairing translation by preventing eIF5B from interacting with eIF1A and the ribosome to accurately position met-tRNA on the start codon of an mRNA [118
]. Some viruses, such as FMDV, use multiple modes to induce host shutoff. In addition to cleaving eIF4A and eIF4G, FMDV infection induces the cleavage of eIF3a, eIF3b, and PABP [97
PABP binds to the 3′ poly(A) tail of an mRNA to enhance RNA stability, and this interaction is also vital for translation initiation complex formation in the cytoplasm [119
]. Many viruses target PABP to induce host shutoff. Calicivirus and enterovirus protease 3C or 3C-like protein cleaves PABP to inhibit its function [120
]. HIV-1 protease, Rubella capsid protein, and influenza NS1 bind PABP and suppress host translation [122
]. Rotavirus NSP3 displaces PABP from eIF4G and interacts with RoXaN to cause nuclear accumulation of PABP [104
]. The ORF57 (HSV-1 ICP27 homolog) and K10 proteins of KSHV and HSV-1 UL47 protein bind PABP to cause nuclear accumulation and abolish its function in the cytoplasm [126
To efficiently produce viral proteins during the shutoff, many viruses have evolutionarily acquired cis-elements in mRNA to induce selective viral protein synthesis. The IRES is used by many RNA viruses, as well as some DNA viruses. Notable viruses that were suggested to be able to use IRES to mediate translation initiation via cap-independent mode to produce proteins include coronavirus, HCV, CSFV, HIV, and CrPV [129
]. Influenza B virus induces the combined translation of both M1 and BM2 protein via the base pairing of mRNA with 18S ribosomal RNA to promote the re-initiation of translation [134
]. Adenovirus uses ribosome shunting to enhance viral mRNA translation, where the tripartite leader in the non-coding region of viral late mRNAs exhibits high complementarity with 18S rRNA, which promotes the ribosome shunting mechanism [135
]. Viruses are notorious for recruiting translation initiation factors to their mRNAs to promote viral mRNA translation. Calicivirus VPg protein binds and recruits eIF3 and eIF4E to viral mRNA [136
] and HSV-1 ICP27 and UL47 proteins bind PABP [126
]. As such, the recruitment of initiation factors to viral mRNAs stimulates their translation. During replication, viruses produce dsRNA that is sensed by innate immune molecule PKR, leading to PKR and downstream eIF2α phosphorylation [138
]. When eIF2α is phosphorylated, global inhibition of translation initiation occurs [139
]. To avert this situation, viruses use various strategies to hinder PKR-mediated eIF2α-phosphorylation. Influenza NS1 and HCMV’s two related proteins (TRS1 and IRS1) sequester dsRNA and prevent PKR phosphorylation [140
]. EBV SM and KSHV ORF57 directly sequester PKR to thwart PKR activation [142
]. HPV E6 and HSV ICP34.5 proteins regulate eIF2α phosphatase to dephosphorylate eiF2α [144
]. VACV encodes two PKR inhibitors, namely E3L and K3L. E3L binds to dsRNA to prevent PKR dimerization, whereas K3L, with its homology to eIF2α, acts as a pseudo-substrate for PKR. Many great details on how VACV, as well as other poxviruses, relaxes PKR-mediated translation inhibition have been revealed [146
7.2. Suppression of Host Cell Translation during VACV Infection
VACV uses multiple tactics to modulate cellular mRNA translation. Our transcriptome-wide analysis showed that VACV mRNAs have a higher translation efficiency than host mRNAs [149
]. During VACV infection, prompt inhibition of cellular protein synthesis occurs. Earlier findings have shown that VACV infection results in the inhibition of protein synthesis via a surface tubular element (STE) displayed on the VACV membrane. However, VACV STE did not affect either cellular RNA or DNA synthesis. When the authors exposed cells to purified STE, a decrease in the polyribosome occurred, accompanied by an increase in the free ribosome pool [150
A remarkable factor present in virions implicated in inducing host shutoff is the phosphorylated 11 kDa protein [151
], encoded by F17R [152
]. Purified 11 kDa/F17 protein from VACV virion and cell-free extract from VACV-infected cells prevents methionyl-tRNA_fMet-40S initiation complex formation, thereby inducing host shutoff [151
]. The importance of this small protein is further augmented by the finding that preventing the expression of F17 protein interrupts VACV morphogenesis [153
]. It was recently shown that F17 sequestering of Rictor and Raptor dysregulates mTOR to counter the antiviral response while retaining mTOR-mediated enhancement of viral protein synthesis [152
]. These findings indicate that F17 has multiple roles ranging from inducing host shutoff and countering the antiviral response to enhancing viral protein synthesis during infection.
VACV expresses a protein using its early gene VACWR169, which is confined in the host cell cytoplasm. In the cytoplasm, protein 169 impairs host protein synthesis, thereby facilitating the inhibition of host antiviral responses. Unlike other factors that induce host shutoff, protein 169 targets translation initiation by affecting both cap-dependent and cap-independent mechanisms, although how this protein manages to do it is yet to be elucidated. Protein 169 is not vital to VACV replication and spread in culture cells; however, it is required to regulate virulence as a VACV lacking protein 169 causes severe infections that induce stronger immune responses and are thus promptly cleared. By inducing host protein synthesis shutoff, VACV protein 169 suppresses host antiviral response and hence regulates virulence [155
]. Interestingly, full-length 169 is not encoded in all VACV strains; for example, a search of a VACWR169 homolog in the VACV Copenhagen strain reveals a premature stop-codon that would result in a truncated, possibly non-functional, protein product.
As discussed earlier, VACV-poly(A)-polymerase-induced POLADS may sequester PABP to make them inaccessible for host mRNA translation [69
]. In a cell-free protein-synthesizing system, the addition of POLADS inhibits cellular mRNA translation by up to 70%, while viral mRNA displayed minimal inhibition [156
]. The selective host protein synthesis inhibitory property of the POLADS is due to the poly(A) tail, as the increased length of the poly(A) tail in POLADS corresponds to increased inhibitory activity [69
]. Moreover, the addition of PABP reversed the host cell mRNA translation inhibition by POLADS, suggesting that PABP becomes the limiting factor that may be sequestered by these POLADS for host cell mRNA translation during VACV infection [69
]. These results also suggest that VACV mRNA translation is less dependent on PABP, which is in agreement with the finding that PABP1 is localized outside of viral factories, where viral mRNA translation is concentrated [159
]. However, the molecular mechanisms involved, as well as whether there is another poly(A)-binding protein that can substitute for PABP, are yet to be discovered.
Our genome-wide identification of VACV transcription start sites and polyadenylation sites revealed pervasive transcription initiation and termination [161
]. Most of them are not transcripts of the ≈200 annotated genes. The findings indicate a vast number of “dark” transcripts that are both capped and polyadenylated, which likely include some of the POLADS described above. These transcripts could be competitors of cap and poly(A)-tail-binding proteins necessary for cap-dependent mRNA translation. Because VACV post-replicative mRNA translation is less dependent on these factors, as shown by several groups, including us [149
], these “dark transcripts” can place VACV mRNA at a translational advantage. This hypothesis is under active investigation in our laboratory.
Recently, antiviral granules (AVGs) and RNA granules were suggested to be present in VACV-infected cells [160
]. In fact, AVGs contain translation initiation proteins (eIF3h, eIF4E, and PABP), leading to speculation that such redistribution of translation initiation factors could be a mechanism used to limit the availability of these factors for host mRNA translation [160
], although further investigation is needed.
7.3. Preferential Translation of VACV mRNAs
A recent review by Meade et al. very nicely summarized how VACV infection modulates host translation machinery to selectively translate viral mRNAs [166
]. We will only give a brief overview of the strategies used by VACV to preferentially translate VACV mRNAs.
Several studies showed that VACV’s post-replicative mRNA translation is enriched in or near virus replication sites called “viral factories," which are characterized by the intense staining of VACV viral DNA in the cytoplasm [159
]. It has been shown that translation initiation factors, such as eIF4E and eIF4G, could be recruited to the viral factories, likely enhancing cap-dependent translation initiation [168
]. Another study suggests that translation outside of the viral factories could also occur [169
]. VACV infection stimulates eIF4F complex formation. The eIF4E is repressed by hypo-phosphorylated eIF4E-binding proteins (4EBPs) [170
]. In the early stages of infection, VACV induces surface integrin-β1-mediated PI3K activation, leading to hyperphosphorylation of 4EBP1 and the subsequent release of cap-binding protein eIF4E [171
], which consequently augments the formation of the eIF4F complex enhancing VACV protein synthesis [172
]. Another poxvirus, namely myxoma virus (MYXV), activates AKT using the host range protein MT-5 [173
], although the role of AKT activation on mRNA translation during MYXV infection has not been studied yet. Additionally, VACV infection activates the MAPK/ERK pathway that stimulates the phosphorylation of eIF4E by MNK1 [159
]. Phosphorylation of eIF4E at the serine 209 residue may lead to an increase in translation initiation of VACV mRNAs [175
The poly(A) leader at the 5′ end of transcripts is a unique feature of all VACV post-replicative mRNAs, which was discovered three decades ago. Only recently, we and others found that the 5′-poly(A) leader confers a selective translational advantage to viral post-replicative mRNAs, specifically in poxvirus-infected cells [149
]. Although the mechanism is still largely unknown, it was suggested that post-translational phosphorylation of small ribosomal protein RACK1 by VACV kinase B1 is necessary for the poly(A)-leader-mediated translational advantage [179
]. We found that the poly(A)-headed mRNAs can be efficiently translated in cells with impaired cap-dependent translation, suggesting that it is a cap-independent translation-enhancing element [149
]. Moreover, the 5′-poly(A) leader is not an IRES [149
]. An A-tract, omega prime, found in the tobacco mosaic virus (TMV) 5′ untranslated region, could also enhance translation [180
]. The TMV omega prime enhances translation via promoting the recruitment of the eIF4F complex [149
]. Recently, it was found that mRNAs of yeast-virus-like elements contain a similar non-templated 5′-poly(A) leader, which also drives eIF4E-independent translation [183
]. It is of note that these VACV post-replicative mRNAs have 5′-caps [184
]. As discussed above, VACV modulates cap-dependent translation initiation factors and recruits them to viral replication sites for efficient viral mRNA translation. How the cap-independent translation element of the 5′-poly(A) leader coordinates with cap-dependent translation promotion to facilitate selective translation of viral mRNAs is an active area of investigation in our laboratory. We hypothesize that VACV can utilize the advantages of both cap-dependent and cap-independent translation modes to achieve a maximal translational advantage for viral mRNAs, which is facilitated by viral and cellular factors.