Ribosomes have the capacity to selectively control translation through changes in their composition that enable recognition of specific RNA elements1. However, beyond differential subunit expression during development2,3, evidence for regulated ribosome specification within individual cells has remained elusive1. Here we report that a poxvirus kinase phosphorylates serine/threonine residues in the human small ribosomal subunit protein, receptor for activated C kinase (RACK1), that are not phosphorylated in uninfected cells or cells infected by other viruses. These modified residues cluster in an extended loop in RACK1, phosphorylation of which selects for translation of viral or reporter mRNAs with 5′ untranslated regions that contain adenosine repeats, so-called polyA-leaders. Structural and phylogenetic analyses revealed that although RACK1 is highly conserved, this loop is variable and contains negatively charged amino acids in plants, in which these leaders act as translational enhancers. Phosphomimetics and inter-species chimaeras have shown that negative charge in the RACK1 loop dictates ribosome selectivity towards viral RNAs. By converting human RACK1 to a charged, plant-like state, poxviruses remodel host ribosomes so that adenosine repeats erroneously generated by slippage of the viral RNA polymerase4 confer a translational advantage. Our findings provide insight into ribosome customization through trans-kingdom mimicry and the mechanics of species-specific leader activity that underlie poxvirus polyA-leaders4.
Regulated mRNA translation allows precise temporal and spatial control of protein production. Changes in translation rates for individual mRNAs are often attributed to the controllable activity of eukaryotic initiation factors5. However, it is postulated that ribosomes can specialize through alterations in subunit composition or post-translational modifications to control mRNA-specific translation1. Curiously, some ribosomal subunits exhibit tissue-specific expression patterns and are associated with disease states termed ‘ribosomopathies’1. Recently, regulons in homeobox (HOX) mRNAs were found to bind RPL38 and mediate their selective translation2,3, revealing how tissue-specific ribosomal protein expression contributes to development. Similar to developmental systems, genetic screens have identified specific ribosomal proteins required for translation of certain viral but not host mRNAs6. This includes the small ribosomal protein RACK1 (refs 7, 8, 9, 10), which contributes to cap-independent internal ribosome entry site (IRES)-mediated translation by some RNA viruses11. By contrast, most DNA viruses produce capped mRNAs that are structurally similar to their host counterparts. To determine if RACK1 regulates translation of mRNAs encoded by DNA viruses, we infected wild-type or RACK1-knockout human Hap1 cells with the poxvirus vaccinia virus (VacV) or herpes simplex virus type 1 (HSV-1). Unlike VacV, in which viral protein synthesis was suppressed in RACK1-knockouts, HSV-1 protein synthesis was unaffected (Fig. 1a). Although RACK1 is implicated in regulating some poorly understood aspects of 80S ribosome assembly12 and translation of short open reading frames13, both viruses encode similar-sized capped mRNAs. Moreover, RACK1 was required for late but not early VacV protein production (Fig. 1b), implying a specific function for RACK1 in late VacV mRNA translation.
Unexpectedly, RNA interference (RNAi)-mediated depletion of RACK1 in primary normal human dermal fibroblasts (NHDFs) did not affect HSV-1 or VacV translation (Fig. 1c). However, RACK1 migrated as a close-running doublet in VacV-infected samples, and VacV appeared to convert residual RACK1 in RACK1-depleted cells to the slower-migrating form. Liquid chromatography–tandem mass spectrometry (LC–MS/MS) identified a cluster of serine and threonine residues (S276/T277/S278/S279) uniquely phosphorylated in VacV-infected cells (Fig. 1d, Extended Data Fig. 1a). Although LC–MS/MS pointed to potential modifications to any of the STSS residues, only one modification was likely per RACK1 peptide, with S278 exhibiting the highest confidence score. Phosphomimetic substitutions at S278 or the entire STSS motif caused increasing RACK1 mobility shifts (Fig. 1e). Alignment of these samples alongside infected samples supported the notion of single-site modifications by VacV, explaining the close migration of RACK1 doublets (Fig. 1f). In VacV-infected cells, RACK1 phosphorylation was prevented by depletion or inhibition of the RACK1-binding protein, PKCβII, while the remaining RACK1 in RACK1-depleted cells was phosphorylated (Fig. 1f, Extended Data Fig. 2). This suggests that VacV did not need to modify the entire RACK1 population under normal conditions, and generated enough modified RACK1 to support viral protein synthesis in cells treated with RACK1-specific siRNA. Only when RACK1 was completely unavailable to modify, namely in knockout cells, was VacV protein synthesis suppressed.
Although VacV required PKCβII to modify RACK1, PKCβII inhibition prevented both the formation of cytoplasmic replication compartments, called viral factories4, and host translational shut-off associated with late poxvirus infection (Extended Data Fig. 3). Moreover, phosphorylation of the STSS motif could not be detected in uninfected or HSV-1-infected cells, or cells infected with RNA viruses that employ cap-dependent or cap-independent initiation5, all of which contained phosphorylated PKCβII (Extended Data Fig. 4). S276/T277/S278/S279 phosphorylation has not been identified in RACK1 modification screens14, suggesting that this is unique to VacV and only indirectly affected by the effects of PKCβII inactivation on VacV replication. VacV encodes two kinases, B1 and F10 (ref. 4). Using temperature-sensitive mutants15,16, RACK1 doublets indicating STSS phosphorylation were detected in cells infected with wild-type or F10-mutant, but not B1-mutant, viruses despite PKCβII phosphorylation (Fig. 2a). ‘Temperature sensitive’ is a misnomer, as mutant B1 proteins are kinase deficient and unstable at many temperatures15. Although host kinases substitute for some B1 functions at permissive temperatures17,18, RACK1 doublets were undetectable with B1 mutants at either temperature, indicative of B1-specific substrates (Extended Data Fig. 4e). B1 also associated with RACK1 complexes (Fig. 2b). To test this further, a B1–mCherry fusion protein was expressed in NHDFs. As in previous reports15, B1 was expressed at low, heterogeneous levels, limiting detection of mobility shifts against the background of unmodified RACK1 (Fig. 2c). However, LC–MS/MS confirmed that B1 expression recapitulated the RACK1 STSS phosphorylation profile of VacV-infected cells, with S278 again exhibiting the highest confidence score (Fig. 2d, Extended Data Fig. 1b). B1–mCherry also localized to viral factories upon infection (Fig. 2e, Extended Data Fig. 5). Although broadly distributed throughout viral factories, a subpopulation of B1 co-localized with RACK1 within viral factories. Live cell imaging of NHDFs expressing RACK1–eGFP and Cro–mCherry (labelling viral factories19), further revealed dynamic RACK1 accumulation around viral factories and within cavities that formed as the viral factories matured (Fig. 2f, Extended Data Fig. 6, Supplementary Videos 1, 2, 3, 4, 5, 6, 7, 8, 9). Viral factories are the sites of late viral protein synthesis and viral mRNAs localize to viral-factory cavities and peripheries20, further implicating RACK1 in their translation.
Although RACK1 has extra-ribosomal functions21, VacV did not reduce RACK1 association with ribosomal subunits, and modified RACK1 was found in ribosomal but not free fractions (Extended Data Figs 7, 8). RACK1 does not appear to be extra-ribosomal in cells from several species22 (Fig. 3a), suggesting this only arises under specific conditions such as transformation8,21,22. Indeed, RACK1 could not be overexpressed in primary NHDFs, and endogenous RACK1 was downregulated upon exogenous RACK1–eGFP expression (Fig. 3b). This did not require RACK1 phosphorylation at canonical sites, but required RACK1 association with ribosomes (Fig. 3a, b). The RACK1 ribosome-binding mutant was poorly expressed owing to proteasomal degradation (Fig. 3c). This instability is not intrinsic to the mutation itself11,23, but phosphorylation at T143 stabilizes ribosome-binding mutants14. LC–MS/MS did not detect T143 phosphorylation in uninfected or infected NHDFs, suggesting that this only occurs in certain contexts. This revealed careful homeostatic control of RACK1 levels through coupling of its ribosome association with protection from proteasomal degradation, and suggests that RACK1 has limited potential for extra-ribosomal functions in normal human fibroblasts.
To explore the role of modified RACK1 in VacV translation, we noted that 5′ untranslated regions (UTRs) of late VacV mRNAs contain unusual polyA-leaders4. Luciferase reporters containing β-actin-, polyU- or polyA-leaders revealed that none of these elements naturally act as enhancers in mammalian cells, but upon VacV infection the polyA-leader became a specific enhancer in a B1-dependent manner (Fig. 3d, e, Extended Data Fig. 9). High luciferase levels in F10-mutant samples suggested that F10 may limit B1-dependent polyA-enhancer stimulation during wild-type infection (Extended Data Fig. 4e). F10 suppresses phosphorylation of several host and viral proteins16, and viruses often restrain stress-inducing processes to avoid killing cells too quickly.
The RACK1 STSS motif lies at the tip of an extended loop that contacts 18S ribosomal RNA and other ribosomal subunits at the mRNA exit channel7,8,9,10,23 (Fig. 4a). Phylogenetic comparisons revealed that although RACK1 is highly conserved, this loop region is variable23 (Fig. 4b, c, Extended Data Fig. 9). In mammals it is uncharged but in plants, where these leaders act as natural enhancers, the loop is extended and contains non-consecutive negatively charged amino acids. To test the effects of loop modifications, NHDFs stably expressing luciferase reporters together with wild-type RACK1, RACK1(S278E) or RACK1(S276E/T277E/S278E/S279E) (denoted STSS-EEEE) phosphomimetics or human RACK1 with its loop replaced with the plant loop were generated. Both phosphomimetic substitutions and the plant-loop chimaera produced higher levels of polyA-leader-luciferase compared to wild-type RACK1, whereas β-actin reporters were modestly suppressed (Fig. 4d). Although this suggests that RACK1 loop modifications confer an advantage to polyA-leader mRNAs, steady-state protein levels are not a direct measure of translation, nor do these assays discern whether transcripts are translated by ribosomes containing endogenous or exogenous RACK1 forms. When ribosomes were paused and those containing wild-type or mutant RACK1–eGFP were isolated from mock-infected or infected cells using GFP-TRAP sepharose, β-actin RNA recovery was significantly reduced by all three charged loops (Fig. 4e). In infected cells, virus-modified wild-type RACK1 selected against β-actin, whereas viral RNAs were readily detected (Fig. 4e, f). Similarly, 5′ polyA-luciferase RNA exhibited a selective advantage over β-actin RNA in the presence of RACK1 phosphomimetics or the plant-loop chimaera (Extended Data Fig. 10a). Notably, viral RNA was recovered at equivalent levels from complexes containing the S278E phosphomimetic or plant-loop chimaera. However, recovery was reduced in the STSS-EEEE phosphomimetic, suggesting that although selective against β-actin and enhancing polyA-luciferase accumulation, clustered loop charge afforded suboptimal selectivity. The plant-loop chimaera co-migrated with the STSS-EEEE phosphomimetic (Fig. 4d), but its charge was interspersed (Fig. 4b). As B1 phosphorylates single sites in the STSS motif of the smaller, human loop, this highlights the exquisite refinement with which VacV modifies RACK1 to mimic its plant counterpart. The effects of RACK1 loop modifications on viral RNA recovery mirrored effects on small subunit ribosomal protein (RPS) association (Fig. 4e). As the STSS-EEEE mutant is stable, it is either a poor proteasome substrate or these differences reflect translational changes, such as altered RACK1 cycling between ribosomes. In the extreme case of the STSS-EEEE phosphomimetic, polysomes were reduced and 40S ribosomes became predominant (Fig. 4g). Based on effects on RPS recovery, this would be more refined with S278E or plant loops. RACK1 can regulate ribosome movement on mRNAs, as the yeast RACK1 orthologue has previously been shown to suppress frameshifts24. Slower scanning rates could conceivably offset the ‘ribosome sliding’ that occurs on long polyA sequences25,26,27, simultaneously customizing ribosomes to a state optimal for viral mRNAs yet suboptimal for host mRNAs (Extended Data Fig. 10b). This represents an alternative mode of ribosome specification distinct from RPL38-mediated binding to HOX mRNAs2,3.
The biological significance of poxvirus polyA-leaders has remained unknown. Together with the tobacco mosaic virus ‘omega leader’, consisting of CAA repeats, these leaders exhibit little or no enhancer activity in mammalian systems, but function efficiently in plants27,28,29. Early experiments hinted that the enhancer-responsive activity in plants resided in ribosomal fractions30. Here we show that enhancer function is regulated by the RACK1 loop that differs in charge between species. Through precise phosphorylation events, a poxvirus kinase converts human ribosomes to a plant-like state to exploit the polyA-leaders generated by an erroneous viral RNA polymerase, providing insight into their species-specific leader activity.
No statistical methods were used to predetermine sample size. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment.
Cell culture and viruses
Validated and certified primary NHDFs were purchased from Lonza Walkersville, Inc. (CC-2509). BSC40, Vero and 293T cells used to propagate viruses were obtained from I. Mohr, NYU School of Medicine31. These cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; MT15013CV, Fisher Scientific) supplemented with 2 mM l-glutamine, 1× penicillin-streptomycin and 5% fetal bovine serum (FBS). Hap1 cells were purchased from Horizon (C859) and are fully characterized by the supplier. Hap1 cells were grown in Iscove’s Modified Dulbecco’s medium (IMDM; SH3022801, Fisher Scientific) supplemented with 2 mM l-glutamine, 1× penicillin-streptomycin and 5% FBS. All cultures were confirmed negative for mycoplasma using Hoechst staining. Wild-type and mutant strains of VacV were grown and titrated on BSC40 cells, as described previously15,32,33. HSV-1 was grown and titrated on Vero cells as described previously31. All VacV or HSV-1 infections were performed at multiplicity of infection (MOI) 3–10 for 20 h, unless otherwise stated. Vesicular stomatitis virus and encephalomyocarditis virus (EMCV) were grown and titrated as described previously34, and infections were performed at MOI 5 for 8 h. Lentivirus vectors were produced by co-transfection of 293T cells with pLVX-hygro-based plasmids, described below, together with p8.91 (gag-pol) and p-VSV-G (envelope). Supernatants containing lentivirus were then filtered and used to transduce NHDFs. Transduced cultures were then selected with 100 μg ml−1 hygromycin to generate pools of NHDFs stably expressing proteins of interest. Retroviruses were produced by transfection of pBABE-puro-based plasmids, described below, in Phoenix-Ampho cells (ATCC). Supernatants containing retrovirus were then filtered and used to transduce NHDFs, which were selected with 0.8 μg ml−1 puromycin to generate stably expressing pools.
To generate RACK1-knockout cells, DNA oligonucleotides (GNB2L1 exon 2, 5′-CACCGATTCCACAGCGTGCTCTTGCG-3′ and 5′-AAACCGCAAGAGCACGCTGTGGAATC-3′, and GNB2L1 exon 3, 5′-CACCGACCACCACGAGGCGATTTGT-3′ and 5′-AAACACAAATCGCCTCGTGGTGGTC-3′) representing single-guide RNA sequences targeting exons 2 and 3 were annealed and cloned into pX458 (Addgene), then transfected into Hap1 cells35,36. 48 h after transfection, GFP+ cells were FACS sorted, subcloned and clonal cell lines were screened by immunoblotting with RACK1 antibody. Potential CRISPR-knockout cell lines were then genotyped using the primers 5′-CCAGTGTGTTAAACGGGCTGC-3′ and 5′-GGAAGAGATCCTTGGAGATGG-3′ for amplification and sequencing (Supplementary Fig. 2).
RNAi, inhibitors and metabolic labelling
Pre-designed siRNAs were obtained from Life Technologies (Thermo Fisher Scientific): Control siRNAs (AM4635 and AM4637), RACK1 (siRNA1 ID 520341, siRNA2 ID 520342) or PKCβII (siRNA1 ID 103396, siRNA2 ID 261007, siRNA3 103309). Cells were transfected with 30 pmol ml−1 RACK1 siRNA or 150 pmol ml−1 PKCβII (PRKCB2) siRNA on two consecutive days using RNAiMax (Invitrogen). 72 h post-transfection cells were collected or infected as indicated. For inhibitors, cultures were treated with DMSO solvent control or 50 nM phorbol 12-myristate 13-acetate (PMA; Calbiochem) dissolved in DMSO 1 h after infection. Cells were treated with 30 μM MG115 dissolved in DMSO for 20 h. Where shown, cultures were metabolically labelled by incubation in methionine/cysteine-free DMEM (17-204-CL; Corning) supplemented with 77 μCi 35S-methionine/cysteine (NEG072; Amersham)37 for 30 min before cell lysis.
Plasmids, cloning and mutagenesis
Human receptor for activated C kinase 1 (RACK1) with a C-terminal eGFP tag was purchased from Addgene (pEGFP-N1-RACK1; plasmid 41088)38. RACK1–eGFP cDNA was amplified using forward primer with SpeI site: 5′-AAAAAACTAGTCTCAAGCTTATGACTGAGCAGATG-3′; and reverse primer with NotI site: 5′-AAAAAGCGGCCGCTTACTTGTACAG-3′. PCR-amplified cDNA was digested with SpeI and NotI (NEB Biolabs) and ligated into pLVX-IRES-Hygromycin plasmid (Takara Bio USA, Inc.) following standard cloning procedures. PCR and gel extraction kits used were from Qiagen, restriction enzymes and T4 DNA ligase were from NEB Biolabs, and subcloning efficiency DH5α competent cells were from Thermo Fisher Scientific. All constructs were verified by sequencing at the NUSeq Core Facility, Northwestern University. Site-directed mutagenesis of RACK1 was performed with two separate pairs of primers using QuickChange II XL Site-Directed Mutagenesis Kit (Agilent Technologies) or Q5 High-Fidelity DNA polymerase reaction followed by Dpn1 treatment (NEB Biolabs). Primers used for site directed mutagenesis were: (1) R36/K38 ribosome binding mutant: forward 5′-CTCCGCCTCTGATGATGAGACCA-3′, reverse 5′-TGGTCTCATCATCAGAGGCGGAG-3′ (2) Y52F phosphorylation site mutant: forward 5′-GATGAGACCAACTTTGGAATTCCACAGCGTG-3′, reverse 5′-CACGCTGTGGAATTCCAAAGTTGGTCTCATC-3′. (3) S146A phosphorylation site mutant: forward 5′-CAGGATGAGGGCCACTCAGAGTGG-3′, reverse 5′-CCACTCTGAGTGGCCCTCATCCTG-3′. (4) Untagged RACK1: forward 5′-GCACACGCTAGGGTACCGCGG-3′, reverse 5′-CCGCGGTACCCTAGCGTGTGC-3′. (5) S278E phosphomimetic mutant: forward 5′-GTTATCAGTACCGAAAGCAAGGCAG-3′, reverse 5′-GTTATCAGTACCGAAAGCAAGGCAG-3′. (6) S276E/T277E/S278E/S279E phosphomimetic mutant: forward 5′-CAAGAAGTTATCGAAGAGGAAGAAAAGGCAGAACCAC-3′, reverse 5′-GTGGTTCTGCCTTTTCTTCCTCTTCGATAACTTCTTG-3′. The plant-loop chimaera was created by Gibson cloning39 using Gibson assembly master mix (NEB Biolabs) and a gBlock DNA fragment (Integrated DNA Technologies), digested with BamHI and assembled with into pEGFP-N1-RACK1 treated with BamHI. Insertions were confirmed by sequencing before being used as templates for PCR amplification and sub-cloning in pLVX-IRES-Hygromycin (using SpeI-NotI digestion). All PCR-amplified mutants ligated into pLVX vectors were verified by sequencing at NUSeq Core Facility, Northwestern University. The gBLOCK sequence for the Arabidopsis thaliana loop was: (VISTSS>LKAEAEKADNSGPAAT): gtgactgtctctccagatggatccctctgtgcttctggaggcaaggatggccaggccatgttatgggatctcaacgaaggcaaacacctttacacgctagatggtggggacatcatcaacgccctgtgcttcagccctaaccgctactggctgtgtgctgcTacaggccccagcatcaagatctgggatttagagggaaagatcattgtagatgaactgaagcaagaaCTCAAGGCTGAGGCTGAAAAGGCTGACAACAGTGGTCCTGCTGCCACCaaggcagaaccaccccagtgcacctccctggcctggtctgctgatggccagactctgtttgctggctacacggacaacctggtgcgagtgtggcaggtgaccattggcacacgcGGGGTACCGCGGGCCCGGGATCCACCGGTCGCCACCatggt.
Luciferase reporters containing β-actin, polyU or polyA leaders were generated by PCR amplification of the luciferase gene using the following primers: (7) Poly-A-leader luciferase: forward 5′-AAAACCGGTAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACATATGGAAGACGCCAAAAAC-3′, reverse 5′-AAAAAGGATCCTTACAATTTGGACTTTCCGCCC-3′. (8) Poly-U-leader luciferase: forward 5′-AAAACCGGTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTCATATGGAAGACGCCAAAAAC-3′, reverse 5′-AAAAAGGATCCTTACAATTTGGACTTTCCGCCC-3′. (9) β-actin leader luciferase: forward 5′-AAAACCGGTTCCGCCCGTCCACACCCGCCGCCAGCTCACCATATGGAAGACGCCAAAAAC-3′, reverse 5′-AAAAAGGATCCTTACAATTTGGACTTTCCGCCC-3′. PCR products were ligated into the retroviral vector pBABE-puro using AgeI and BamHI restriction and T4 ligation (NEB Biolabs). B1–mCherry was generated by PCR amplification of the B1 gene from VacV genomic DNA using the following primers: forward 5′-AAAAAGCCGGCATGAACTTTCAAGGACTTGTG-3′, reverse 5′-AAAAACTCGAGCATATGATAATATACACCCTGCATTAATATG-3′. The PCR product was subcloned into the retroviral vector, pBABE-puro using NgoMIV and XhoI restriction digestion and T4 ligation. The B1–mCherry insert was confirmed by sequencing. Cro–pmCherry was a gift of D. Evans19. Cro–mCherry was subcloned into pBABE-puro by restriction digestion with Age1 and BamH1, followed by T4 ligation. The insert was confirmed by sequencing.
Western blotting and antibodies
Whole-cell lysates were prepared in lysis buffer (62.5 mM Tris-HCl at pH 6.8, 2% SDS, 10% glycerol, 0.7 M β-mercaptoethanol), followed by boiling for 3 min40. Samples were resolved using 10% Tris-glycine SDS PAGE performed under reducing conditions40. Proteins resolved by gel electrophoresis were transferred to a 0.2-μm pore-size nitrocellulose membrane (GE Healthcare Life Sciences) using a wet electro blotting system (Mini Trans-Blot, Bio-Rad Laboratories, Inc.) at 57 V for 70 min. After transfer, the membrane was blocked (5% non-fat dry milk, 0.1% Tween in TBS) for 1 h at room temperature. Blocking buffer was then removed and membranes were rinsed before incubation with primary antibody (diluted in 3% BSA, 0.1% Tween in TBS) overnight at 4 °C. Membranes were washed before incubation with HRP-conjugated secondary antibody (GE Healthcare Life Sciences) diluted 1:3,000 in 5% non-fat dry milk, 0.1% tween in TBS for 1 h at room temperature, followed by washing. For detection, membrane was incubated with Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific) for 1 min before exposure to X-ray film.
The following antibodies were used: RACK1, rabbit mAb 5432(D59D5); β-actin, mouse mAb 3700 (8H10D10); RPL11, rabbit polyclonal Ab (14382); GFP for western blotting, rabbit mAb 2956 (D5.1); eIF3c, rabbit polyclonal Ab (2068) from Cell Signaling Technology. α-Actin, rabbit polyclonal Ab A2103 from Sigma-Aldrich. Rabbit mAb (EPR8545) to ribosomal protein S10 was from Abcam (Ab151550). Rabbit anti-VacV polyclonal antibody (8101) from Virostat. Anti-GFP for immunofluorescence was from Novus Biologicals (NB100-62622). Rabbit anti-phospho-PKCβII (Thr641) polyclonal antibody was from EMD Millipore. Anti-mCherry/RFP antibodies were mouse mAb (6G6) for western blotting or rat mAb (5f8) for immunofluorescence, Chromotek. Total 4E-BP1 antibody A300-501A from Bethyl Laboratories. Rabbit PABP antiserum was from S. Morley (University of Sussex). Rabbit anti-B1 was from P. Traktman (Medical University of North Carolina). Mouse anti-I3 was from D. Evans and N. Favis (University of Alberta). Mouse anti-D8 and anti-A14 mAbs were from Y. Xiang (University of Texas Health Science Center).
PCR and RT–qPCR
RNA was isolated using TRIzol (Invitrogen, Thermo Fisher Scientific). 0.1 μg of RNA was reverse-transcribed using Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science Mannheim) followed by PCR using Q5 High-Fidelity DNA polymerase (NEB Biolabs). 1 μl cDNA was used as template and amplification products were resolved on 1% Agarose gels, staining using SYBR safe. Real-time PCR was performed using FastStart Universal SYBR Green Master (Rox) on a 96-well plate using 7500 Fast Real-Time PCR System (Applied Biosystems). RNA quantitation was performed using comparative Ct methods (ΔΔCt method). β-actin was used as endogenous control and mock sample was used as calibrator. Average Ct values were used to calculate fold change or 2−ΔΔCt. The following primers were used: luciferase forward primer 5′-TCAAAGAGGCGAACTGTGTG-3′, luciferase reverse primer 5′-TTTTCCGTCATCGTCTTTCC-3′. β-actin forward primer 5′-CATGTACGTTGCTATCCAGGC-3′, β-actin reverse primer 5′-CTCCTTAATGTCACGCACGAT-3′. POLR2L forward primer 5′-AGGAGAGCCTTCCATCTCG-3′, POLR2L reverse primer 5′-ATCTGGCTCTTCAGATTCCG-3′, MFGE8 forward primer 5′-CACTCTGCGCTTTGAGCTAC-3′, MFGE8 reverse primer 5′-TCCAGCTGAAGAGATGCAAG-3′. EB1 forward primer 5′-ctgtatgccacagatgaagg-3′, EB1 reverse primer 5′-ccagacacaatgtcaaacgc-3′. A14 forward primer 5′-GGACATGATGCTTATGATTGG-3′, A14 reverse primer 5′-CTTTCCATGTACGAGTGGGACTG-3′. A27 forward primer 5′-CTAAACGCGAAGCAATTGTTAAAG-3′, A27 reverse primer 5′-CTTCATCGTTGCGTTTACAACAC-3′. Amplicons were resolved on 1.5% agarose gel and stained with SYBR safe stain.
Luciferase and RNA selectivity assays
For infected cell luciferase assays, NHDFs were seeded onto 12-well plates and transfected with 200 ng plasmid DNA encoding luciferase reporters harbouring 5′ polyA, polyU or β-actin leaders, described above. 48 h post-transfection cells were mock infected or infected with VacV at MOI 10. 20 h post-infection cultures were washed with PBS and lysed with 200 μl Luciferase Cell Culture Lysis Reagent (Promega). Lysates were clarified by centrifugation at 10,000g for 2 min. 20 μl supernatant was added to 96-well plates and luciferase activity was measured using a Spectramax microplate reader.
For RNA selectivity assays to isolate RACK1-containing ribosomes and associated mRNAs, 6-cm dishes were seeded with NHDFs stably expressing wild-type, plant loop or phosphomimetic forms of RACK1–eGFP, described above. Cells were either mock-infected or infected with VacV at MOI 3 for 20 h. Cells were then pre-treated for 10 min with 100 μg ml−1 cycloheximide, washed with ice-cold PBS containing cycloheximide, and collected in lysis buffer (50 mM HEPES pH 7.4, 150 mM NaCl, 1.5 mM MgCl2, 2 mM EDTA, 1.6 mM Na3VO4, 25 mM glycerophosphate, 1.5% NP-40, 100 μg ml−1 cycloheximide, 25 U ml−1 Ribolock RNase inhibitor and complete mini EDTA-free protease inhibitor cocktail (Roche)). After 40 min rocking at 4 °C, lysates were clarified by centrifugation and incubated for 4 h with GFP-TrapA beads (ChromoTek). Beads were then washed with lysis buffer before elution of bound protein by boiling in Laemmli buffer, or isolation of RNA using TRIzol (Invitrogen, Thermo Fisher Scientific). Samples were then subject to western blotting analysis of proteins isolated, or reverse transcription PCR (RT–PCR) analysis of bound mRNA. For RT–PCR, 10 μl of RNA was reverse-transcribed using Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science Mannheim). 1.3 μg of cDNA was used as template for PCR using Q5 High-Fidelity DNA polymerase (NEB Biolabs) and primers outlined above.
Immunofluorescence and live cell imaging
For fixed imaging, NHDFs expressing RACK1–eGFP were infected at MOI 3 for 20 h. Cultures were then rinsed in PBS and fixed in 4% paraformaldehyde for 20 min. After washing, samples were blocked in PBS containing 10% FBS/0.25% saponin and then incubated at 4 °C overnight with anti-RSP10 antibody diluted in PBS containing 10% FBS/0.025% saponin. The following day, samples were washed in PBS containing 0.025% saponin and incubated with Alexa-Fluor-conjugated anti-mouse secondary antibody (Thermo Fisher Scientific) for 1 h at room temperature. After washing, samples were stained with Hoechst 33342, followed by washing and mounting in FluorSave Reagent (Calbiochem). For experiments involving B1–mCherry and RACK1–eGFP, cells were infected and processed as described above except to compensate for low-level B1 expression fixed samples were stained with anti-mCherry/RFP and anti-GFP antibodies to enhance signals (Chromotek). For experiments examining the effects of PMA on formation of viral factories, cells were infected in the presence of DMSO or 50 nM PMA and processed as described above, staining with anti-I3 antiserum. Images were acquired using a Leica DMI6000B microscope using a 100× objective (HC PL APO 100×/1.44 NA oil), ORCA FLASH 4.0 cMOS camera and Metamorph software. All sample sets were imaged using the same acquisition settings, and all post-acquisition processing was minimized and applied equally throughout samples using Metamorph software. For live cell imaging, NHDFs expressing RACK1–eGFP and Cro–mCherry were seeded on 35 mm MatTek glass-bottom dishes (P35G-1.5-14-C) and infected with VacV at MOI 10. For imaging, culture medium was changed to CO2-independent medium (Gibco, 18045-088) and cultures were transferred to a Leica DMI6000B-AFC microscope with an in vivo environmental chamber at 37 °C. Images were acquired approximately 20 h.p.i. at the indicated frame rate with intermittent fast filter switching, using a 100× objective (HC PL APO 100×/1.44 NA oil) and an ORCA FLASH 4.0 cMOS camera with Metamorph software.
Sucrose gradient centrifugation
For polysome analysis NHDFs were treated for 10 min with 100 μg ml−1 cycloheximide to freeze ribosomes, washed with ice-cold PBS containing cycloheximide and then scraped into ice-cold 1× lysis buffer (1% Triton 100 μ ml−1 RNase inhibitor, complete EDTA-free protease inhibitor tablet, 20 mM Tris-Cl pH 7.5, 10 mM MgCl2, 100 mM potassium acetate, 1 mM DTT, 100 μg ml−1 cycloheximide). Cells were lysed for 30 min. Lysates were clarified by spinning at 10,000g for 10 min before layering on top of 10 ml of 5–50% sucrose gradient made in 1× polysome buffer (20 mM Tris-Cl pH 7.5, 10 mM MgCl2, 100 mM potassium acetate, 1 mM DTT, 100 μg ml−1 cycloheximide) and centrifuged in Beckman Coulter SW 41-Ti rotor (Beckman Coulter, Inc.) at 41,000 r.p.m., 4 °C for 2 h. Following centrifugation, sucrose gradients were fractionated using an automated Density Gradient Fractionation System (Brandel Biomedical Research & Development Laboratories, Inc.) with continuous monitoring at 254 nm using an UA-6 absorbance detector and recorded using PeakChart Software. For western blot analysis, fractions were trichloroacetic acid (TCA)-precipitated as follows: samples were incubated at 4 °C overnight in TCA at a final concentration of 10%. Samples were then centrifuged at 10,000g for 15 min. Pellets were washed in a 1:4 solution of 1× polysome buffer:acetone followed by centrifugation at 10,000g for 15 min. Supernatants were removed and protein pellets were air dried. Pellets were suspended in 1× Laemmli buffer and boiled for 3 min.
Isolation of RACK1 complexes and mass spectrometry
For mass spectrometry analysis, eGFP or RACK1–eGFP were isolated from soluble cell lysates as follows. Four 30-cm dishes were seeded with NHDFs stably expressing RACK1–eGFP or eGFP control. Cells were mock infected or infected with VacV at MOI 10 for 20 h. Cells were then washed with ice-cold PBS and scraped into lysis buffer (50 mM HEPES pH 7.4, 150 mM NaCl. 0.5 mM MgCl2, 2 mM EDTA, 2 mM Na3VO4, 25 mM glycerophosphate, 1.5% NP-40 and complete mini EDTA-free protease inhibitor cocktail (Roche)). After incubation at 4 °C with continuous rocking for 40 min, lysates were clarified by centrifugation and incubated for 2–4 h with sepharose resin covalently conjugated to GFP-binding protein41. Samples were then extensively washed before boil-elution followed by TCA precipitation. Precipitated protein pellets were solubilized in 100 μl of 8 M urea for 30 min and then 100 μl of 0.2% ProteaseMAX (Promega) was added for 2 h. Protein extracts were reduced and alkylated as previously described42, followed by the addition of 300 μl of 50 mM ammonium bicarbonate, 5 μl 1% ProteaseMAX and 0.5 μg sequence-grade trypsin (Promega). Samples were digested overnight in a 37 °C thermomixer (Eppendorf).
For Orbitrap Fusion Tribrid MS analysis, the tryptic peptides were purified with Pierce C18 spin columns (Thermo Scientific). Three micrograms of peptide was auto-sampler loaded with a Thermo EASY nLC 1000 UPLC pump onto a vented Acclaim Pepmap 100, 75 μm × 2 cm, nanoViper trap column coupled to a nanoViper analytical column (Thermo-164570, 3 μm, 100 Å, C18, 0.075 mm, 500 mm) with stainless steel emitter tip assembled on the Nanospray Flex Ion Source with a spray voltage of 2,000 V. Buffer A contained 94.785% H2O with 5% acetonitrile and 0.125% formic acid, and buffer B contained 99.875% acetonitrile with 0.125% formic acid. The chromatographic run was for 4 h in total with the following profile: 0–7% over 7 min, ramp to 10% over 6 min, ramp to 25% over 160 min, ramp to 33% over 40 min, ramp to 50% over 7 min, ramp to 95% for 5 min and stay at 95% for 15 additional minutes. Additional MS parameters include: ion transfer tube temperature to 300 °C, Easy-IC internal mass calibration, the default charge state was set to 2 and cycle time was set to 3 s. Detector type set to Orbitrap, with 60,000 resolution, with wide quad isolation, mass range was set to normal, scan range was set to 300–1,500 (m/z), max injection time was set to 50 ms, AGC target was set to 200,000, microscans was set to 1, S-lens RF level was set to 60, without source fragmentation, and data type was set to positive and centroid. Monoisotopic precursor selection was set as on, included charge states equal to 2–6 (and reject unassigned). Dynamic exclusion enabled and set to 1 for 30 s and 45 s exclusion duration at 10 p.p.m. for high and low. Precursor selection decision was set to most intense, top 20, isolation window was set to 1.6, scan range was set to auto normal, first mass was set to 110, collision energy was set to 30%. For CID, we used the ion trap detector, ion trap resolution was set to 30 K, ion trap scan rate was set to rapid, maximum injection time was set to 75 ms, AGC target was set to 10,000, and Q was set to 0.25, finally we injected ions for all available parallelizable time.
Spectrum raw files were extracted into ms1 and ms2 files using the in-house program RawXtractor or RawConverter (http://fields.scripps.edu/downloads.php)43, and the tandem mass spectra were searched against UniProt human database (downloaded on 03-27-2015) (The UniProt Consortium 2015) and matched to sequences using the ProLuCID databse search program (ProLuCID ver. 3.1)44,45,46. ProLuCID searches were done on an Intel Xeon cluster running under the Linux operating system. The search space included all fully and half-tryptic peptide candidates that fell within the mass tolerance window with no miscleavage constraint. Carbamidomethylation (+57.02146 Da) of cysteine was considered as a static modification and a differential modification of 79.9663 on serine, threonine, or tyrosine. The validity of peptide/spectrum matches (PSMs) was assessed in DTASelect2 (using two SEQUEST-defined parameters, the cross-correlation score (XCorr), and normalized difference in cross-correlation scores (DeltaCN)44,47,48. The search results were grouped by charge state (+1, +2, +3, and greater than +3) and tryptic status (fully tryptic, half-tryptic, and nontryptic), resulting in 12 distinct subgroups. In each of these subgroups, the distribution of Xcorr, DeltaCN, and DeltaMass values for (1) direct and (2) decoy database PSMs was obtained; then the direct and decoy subsets were separated by discriminant analysis. Full separation of the direct and decoy PSM subsets is not generally possible; therefore, peptide match probabilities were calculated on the basis of a nonparametric fit of the direct and decoy score distributions. A peptide confidence of 0.95 was set as the minimum threshold. The false discovery rate (FDR) was calculated as the percentage of reverse decoy PSMs among all the PSMs that passed the confidence threshold. Each protein identified was required to have a minimum of one half-tryptic peptide; however, this peptide had to be an excellent match with an FDR less than 0.001 which represents at least one excellent peptide match. After this last filtering step, protein FDRs were below 1% for each sample analysis based on decoy hits. To quantitatively assess the confidence of the phosphorylated peptides we used the probability-based protein phosphorylation analysis and site localization algorithm Ascore49. All mass spectrometry files can be found in the public data repository MassIVE: http://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=70c7c763026d4eadb102b21a9afef755 and ftp://massive.ucsd.edu/MSV000080661.
Statistical analyses were performed using one-way analysis of variance (ANOVA) followed by Tukey HSD test for post-hoc comparisons of three or more experimental groups (only when ANOVA was significant). This was done using online tools available at http://astatsa.com/OneWay_Anova_with_TukeyHSD. The P value for all cases was set to <0.01 for significant difference. Two-tailed t-test was also used for testing significance between two groups. The P value for all cases was set to <0.05 for significant difference.
Structure modelling and phylogenetic analysis
RACK1 sequence alignment was performed using Clustal omega program and further processed using ESPript 3.0 (ref. 50). All conserved residues are highlighted in black and similar residues are bold. Human RACK1 (PDB: 4AOW) structure was used to generate structural data presented on top of the alignment. Alignment was done between RACK1 from Homo sapiens (UniProt ID: P63244), Saccharomyces cerevisiae (UniProt ID: P38011), Arabidopsis thaliana (UniProt ID: P68040), Nicotiana tabacum (UniProt ID: P49026), Caenorhabditis elegans (UniProt ID: Q21215), Mus musculus (UniProt ID: P68040), Drosophila melanogaster (UniProt ID: O18640).
The structure of A. thaliana RACK1A (PDB: 3DM0) was overlayed on the 80S mammalian ribosome (PDB: 4UG0). The loop between blade 6 and 7 was highlighted in figures. A. thaliana RACK1 was modelled using Modeller 9.16 (ref. 51) using human RACK1 (PDB: 4AOW) as template. Modelling was performed because the loop between blade 6 and 7 in the A. thaliana RACK1A (PDB: 3DM0) crystal structure was disordered. All coordinates were retrieved from protein data bank (PDB) (http://www.rcsb.org) and visualized by Swiss-Pdb Viewer v4.1 (ref. 52).
All data generated or analysed during this study are included in this published article and the Supplementary Information files.
This work was supported by grants from the National Institutes of Health (NIH) R01AI127456 and R21AI105330 to D.W., R00DC013805 to J.N.S., R01AI099506 to P.S., and Catalyst Award C-068 from the Chicago Biomedical Consortium to J.N.S. M.G.R. was supported by training grant T32GM008061. We thank G. McFadden, R. Condit, P. Traktman, D. Evans and Y. Xiang for reagents.
Extended data figures
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). RACK1 does not localize to smaller, immature viral factories (in the lower field of view) but accumulates within cavities that form in larger, mature factories (center-left). Cro-mCherry does not efficiently label nuclei of non-dividing primary cells or enlarged, disorganized older viral factories (adjacent to nucleus in center field) as rates of DNA synthesis are low. Timestamp is in minutes. Bar = 20µm. Acquisition rate = 2 frames per minute (fpm). Playback = 7 frames per second (fps).
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). RACK1 appears in smaller factories as cavities form, and RACK1 puncta enlarge to fill cavities as they grow. Timestamp is in minutes. Bar = 20µm. Acquisition rate = 4 frames per minute (fpm). Playback = 7 frames per second (fps).
Video 3: Dynamic reorganization and accumulation of RACK1 relative to cavity domains of viral factories.
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). Large RACK1 accumulations dynamically track with and mirror the behavior of large cavity domains in mature viral factories. RACK1 can also be seen to appear within new cavities as they form. Timestamp is in minutes. Bar = 20µm. Acquisition rate = 1 frames per minute (fpm). Playback = 7 frames per second (fps).
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). An independent example of the accumulation of RACK1 within cavities within mature viral factories that are actively synthesizing DNA (strongly staining with cro-mCherry), and the disorganized structure of older factories (poorly stained for cro-mCherry). Timestamp is in minutes. Bar = 20µm. Acquisition rate = 4 frames per minute (fpm). Playback = 7 frames per second (fps).
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). RACK1 accumulates at peripheral and cavity subdomains of viral factories. Timestamp is in minutes. Bar = 20µm. Acquisition rate = 4 frames per minute (fpm). Playback = 7 frames per second (fps).
Zoomed images from Video 3. Events are marked with green arrows (RACK1) or Red Arrows (VF Cavities): First Arrow Set highlights RACK1 concentrations within large cavities that mirror the dynamic behavior of VF cavities. Second and Third Arrow Sets highlight the appearance of new cavities within factories that contain RACK1. Timestamp is in minutes. Bar = 5µm. Acquisition rate = 1 frames per minute (fpm). Playback = 7 frames per second (fps).
Zoomed images from Video 2. Events are marked with green arrows (RACK1) or Red Arrows (VF Cavities): First Arrow Set highlights a small, immature VF with no RACK1 accumulation (upper arrow) and a mature VF cavity containing RACK1 (lower arrow). Second Arrow Set highlights the appearance of RACK1 within the smaller, maturing factory as a cavity forms (upper arrow) and the expansion of RACK1 in the mature VF as its cavity grows (lower arrow). Timestamp is in minutes. Bar = 5µm. Acquisition rate = 4 frames per minute (fpm). Playback = 7 frames per second (fps).
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). Timestamp is in minutes. Bar = 5µm. Acquisition rate = 1 frame per minute (fpm). Playback = 7 frames per second (fps).
Live cell imaging of VacV-infected NHDFs expressing RACK1-eGFP (Left panel; green in merge) and Cro-mCherry (middle panel; red in merge). Timestamp is in minutes. Bar = 5µm. Acquisition rate = 4 frames per minute (fpm). Playback = 7 frames per second (fps).