Translation elongation factor 1A facilitates the assembly of the tombusvirus replicase and stimulates minus-strand synthesis

Abstract

Replication of plus-strand RNA viruses depends on host factors that are recruited into viral replicase complexes. Previous studies showed that eukaryotic translation elongation factor (eEF1A) is one of the resident host proteins in the highly purified tombusvirus replicase complex. Using a random library of eEF1A mutants, we identified one mutant that decreased and three mutants that increased Tomato bushy stunt virus (TBSV) replication in a yeast model host. Additional in vitro assays with whole cell extracts prepared from yeast strains expressing the eEF1A mutants demonstrated several functions for eEF1A in TBSV replication: facilitating the recruitment of the viral RNA template into the replicase complex; the assembly of the viral replicase complex; and enhancement of the minus-strand synthesis by promoting the initiation step. These roles for eEF1A are separate from its canonical role in host and viral protein translation, emphasizing critical functions for this abundant cellular protein during TBSV replication.

Figure 1. The effect of eEF1A mutations on TBSV repRNA accumulation in yeast.

(A) The yeast strains expressed only one form of eEF1A, as indicated. Top panel: Replication of the TBSV repRNA was measured by Northern blotting 24 h after initiation of TBSV replication. The accumulation level of repRNA was normalized based on the rRNA (middle panel, the 18S ribosomal RNA levels were estimated by Northern blotting). Bottom two panels: Accumulation of p33/p92^pol and eEF1A was estimated by Western blotting using anti-His and anti-eEF1A antibody, respectively. Note that * marks an SDS-resistant p33 homodimer band. (B) An in vitro replicase assay to test the relative activity of the tombusvirus replicase obtained from yeast expressing various mutants of eEF1A. Top panel: We tested the in vitro replicase activity using comparable amounts of affinity-purified replicase with added DI-72 RI(−) RNA template. Bottom panels: Western blot analysis showing p33 viral replication protein and the co-purified eEF1A in the above purified replicase preparations. (C) Critical eEF1A residues for tombusvirus replication. Three novel mutants of eEF1A were identified, which exhibited increased tombusvirus replication (V301D, L374V/N377K, and F413L; yellow balls) while the new A76V and the previously identified T22S exhibited decreased tombusvirus replication (green balls). The structure of eEF1A was generated using Jmol with PDB coordinates 1IJE.

Figure 2. Cell-free TBSV replicase assay supports a role for eEF1A in minus-strand synthesis.

(A) Purified recombinant p33 and p92^pol replication proteins of TBSV in combination with DI-72 (+)repRNA were added to the whole cell extract prepared from eEF1A mutant or WT yeast strains as shown (lanes 1–5). Top panel: The denaturing PAGE analysis of the ³²P-labeled repRNA products obtained is shown. The full-length repRNA is pointed at by an arrow. Panels below show Western blot analysis of the whole cell extracts for the indicated yeast proteins based on specific antibodies. Bottom panel shows the coomassie-blue stained SDS-PAGE gel to visualize total protein levels in the whole cell extracts. (B) Detection of single- and double-stranded RNA products produced in the cell-free TBSV replicase assay. Odd numbered lanes represent replicase products, which were not heat treated (thus both ssRNA and dsRNA products are present), while the even numbered lanes show the heat-treated replicase products (mostly ssRNA is present). The amount of dsRNA and the ratio of ssRNA/dsRNA in the samples are shown. Note that, in the nondenatured samples, the dsRNA product represents the annealed (−)RNA and the (+)RNA, while the ssRNA products represents the newly made (+)RNA products. (C) Denaturing PAGE analysis of the TBSV replicase products obtained in the cell-free replicase assay after S1 nuclease treatment, which cleaves the ssRNA, but not the dsRNA product. (D) The denaturing PAGE analysis of the ³²P-labeled repRNA products obtained in the in vitro reconstitution assay is shown. The membrane fraction of the whole cell extracts prepared from eEF1A mutant strains were mixed with the supernatant fraction of CFE prepared from WT eEF1A (lanes 6–10) or the supernatant fraction of CFE from the mutant strains were added to the membrane fraction from the wt strain (lanes 11–15). The reconstituted extracts were programmed with purified recombinant TBSV p33/p92^pol and (+)repRNA. Bottom panel: Western blot analysis shows the amount of endogenous eEF1A in various fractions (see above) prepared from yeast expressing various mutants of eEF1A.

Figure 3. eEF1A promotes the initiation by the TCV RdRp during minus-strand synthesis.

(A) Purified eEF1A was added to the TCV RdRp assay as shown. The TBSV (+)RNA template was the short 3′ end region (SL1/SL2/SL3), which contain the promoter region (SL1) for initiation and the replication silencer element (within SL3) that down-regulates initiation. The second template was SL1m with a point mutation within the promoter sequence, which is being used more efficiently by the TCV RdRp in vitro. Note that eEF1A has been shown to bind to the replication silencer element. The RdRp assay had two steps: first, the shown components were incubated at room temperature to facilitate their interaction, followed 5 min latter the addition of the shown component and the ribonucleotides to start RNA synthesis. The RdRp activity in samples containing the template RNA and the RdRp were chosen as 100% (lanes 3–4). (B) Detection of abortive RdRp products in the in vitro assay. 15% PAGE/UREA gel was used to resolve the 4–10 nt long products produced during initiation followed by rapid termination. Note that abortive RNA products are characteristic products for RNA polymerases that initiate de novo (in the absence of a traditional primer). (C) Lack of stimulation of 3′-terminal extension by eEF1A in vitro. The template RNAs (shown schematically) contain a common artificial hairpin structure at the 3′ end that facilitates 3′-TEX by the TCV RdRp. The black bar represents 3 different sequences in the three constructs, derived from RIV(+)(includes SL1/SL2/SL3 sequences), RIII(−) and RIII(+) of DI-72 RNA, respectively. The gel image shows the results of 3′-TEX in the presence of 0 or 1 µg eEF1A as shown in a TCV RdRp assay.

Figure 4. Inhibition of TBSV repRNA replication by Didemnin B and Gamendazole in a cell-free TBSV replicase assay.

(A) The cell-free TBSV replicase assay was performed as described in Fig. 2. DB and GM were added in the following amounts: 0, 50, 100, 150, 200, 250 µM for DB and 0, 5, 25, 50, 100, and 200 µM for GM, respectively. The replicase activity in the samples containing the DMSO solvent instead of DB or GM was chosen as 100%. (B) Time course analysis was performed in a cell-free TBSV replicase assay as described in Fig. 2. DB (150 µM) and GM (100 µM) were added at various time points and the replicase assay was stopped after 3 hours for each treatment, followed by RNA analysis in a denaturing PAGE gel. The replicase activity in the samples containing DMSO added at the 0 time point was chosen as 100%. (C) A step-wise approach was used to separate the possible effect of DB and GM during either the assembly of the TBSV replicase or RNA synthesis steps. In step 1, the purified recombinant TBSV p33, p92^pol and (+)repRNA were added to the whole cell extract in the presence of ATP and GTP, which only supports the assembly of the TBSV replicase, but prevents RNA synthesis. This was followed by removal of the extra amount of p33, p92^pol and repRNA, which were not bound to the membranes of cell-free extract, and then by the standard replicase assay in a buffer containing ³²P-UTP and ATP, CTP and GTP (step “RNA synthesis”). The denaturing PAGE analysis of the ³²P-labeled repRNA products obtained is shown. Note that DB (150 µM) and GM (100 µM) were added to the assay either at the beginning (prior to replicase assembly) or after the replicase assembly. See further details in panel B. (D) The effect of DB and GM on binding between the purified eEF1A and ³²P-labeled template RNA (SL1/SL2/SL3) based on EMSA. The bound and unbound RNAs are pointed at by arrowheads. GM and DB were applied in the following amounts: 0, 5, 50, 250, and 1000 µM. Note that the amount of unbound RNA in the absence of eEF1A (lane 1) was chosen as 100%. (E) The inhibitory effect of DB on co-purification of eEF1A with the viral repRNA. WT ³⁵S-labeled eEF1A was produced in a translation assay using rabbit reticulocyte lysate, followed by incubation with biotin-labeled DI-72(+) repRNA in the presence of 0, 50, 150, 500 and 1000 µM DB. Then the repRNA was captured with streptavidin-coated magnetic beads, followed by elution of the co-purified proteins from the beads. SDS-PAGE analysis shows the amount of co-purified ³⁵S-labeled eEF1A. (F) The inhibitory effect of DB and GM on the template recruitment step in vitro. Purified recombinant p33/p92 and ³²P-labeled DI-72 (+)repRNA (indicated as W, lanes 1 and 3–8) or C₉₉-G mutant (+)repRNA (indicated as M, lane 2) were added to a whole cell extract (CFE) in the presence of DMSO (control), 100 µM GM or 150 µM DB, followed by centrifugation/washing to remove the ³²P-labeled repRNA that is not bound to the membrane. Then the membrane-bound RNA was analyzed in a denaturing PAGE gel. Note that the recruitment deficient C₉₉-G mutant repRNA bound to the membrane nonspecifically (∼20% level).

Figure 5. A model describing the functions of eEF1A during tombusvirus replication.

eEF1A not only affects the stability of p33 in cells, but it binds to both the p92^pol replication protein and the viral RNA, facilitating RNA recruitment into replication and the assembly of the viral replicase complex (shown as a membrane-bound complex with multiple p33, p92^pol and additional host factors, HF, such as Hsp70 and Cdc34p). Subsequently, eEF1A promotes minus-strand synthesis by facilitating initiation on the viral template RNA by the viral replicase.