The Integrity of the YxxL Motif of Ebola Virus VP24 Is Important for the Transport of Nucleocapsid-Like Structures and for the Regulation of Viral RNA Synthesis

Abstract

While it is well appreciated that late domains in the viral matrix proteins are crucial to mediate efficient virus budding, little is known about roles of late domains in the viral nucleocapsid proteins. Here, we characterized the functional relevance of a YxxL motif with potential late-domain function in the Ebola virus nucleocapsid protein VP24. Mutations in the YxxL motif had two opposing effects on the functions of VP24. On the one hand, the mutation affected the regulatory function of VP24 in viral RNA transcription and replication, which correlated with an increased incorporation of minigenomes into released transcription- and replication-competent virus-like particles (trVLPs). Consequently, cells infected with those trVLPs showed higher levels of viral transcription. On the other hand, mutations of the YxxL motif greatly impaired the intracellular transport of nucleocapsid-like structures (NCLSs) composed of the viral proteins NP, VP35, and VP24 and the length of released trVLPs. Attempts to rescue recombinant Ebola virus expressing YxxL-deficient VP24 failed, underlining the importance of this motif for the viral life cycle.IMPORTANCE Ebola virus (EBOV) causes a severe fever with high case fatality rates and, so far, no available specific therapy. Understanding the interplay between viral and host proteins is important to identify new therapeutic approaches. VP24 is one of the essential nucleocapsid components and is necessary to regulate viral RNA synthesis and condense viral nucleocapsids before their transport to the plasma membrane. Our functional analyses of the YxxL motif in VP24 suggested that it serves as an interface between nucleocapsid-like structures (NCLSs) and cellular proteins, promoting intracellular transport of NCLSs in an Alix-independent manner. Moreover, the YxxL motif is necessary for the inhibitory function of VP24 in viral RNA synthesis. A failure to rescue EBOV encoding VP24 with a mutated YxxL motif indicated that the integrity of the YxxL motif is essential for EBOV growth. Thus, this motif might represent a potential target for antiviral interference.

Keywords: Alix; Ebola virus; VP24; late-domain YxxL; nucleocapsid-like structure; transport; viral transcription and replication.

FIG 1

Expression of VP24 constructs containing mutations in the YxxL motif. (A) Schematic representation of VP24 amino acid sequences of the various Ebola viruses and VP24 mutants from positions 165 to 184. The putative YxxL motif is indicated in blue, and the mutated amino acids are shown in red. (B) Huh-7 cells were transfected with plasmids encoding either VP24^wt or VP24 mutants. Cells were harvested at 24 h p.t., lysed, and subsequently subjected to Western blot analysis using an anti-VP24 antibody and alpha-tubulin as a control. (C) Huh-7 cells were transfected with plasmids encoding each VP24 construct. Anti-VP24 antibody signals were analyzed by confocal microscopy. DAPI was used to visualize the nucleus.

FIG 2

Influence of mutations in the YxxL motif of VP24 on minigenome transcription/replication and NCLS assembly. (A) HEK293 cells were transfected with either of the VP24 constructs together with the minigenome assay components. Protein expression of VP24 and alpha-tubulin was analyzed by Western blotting. At 48 h p.t., cells were lysed, and the reporter gene activity was measured. The relative reporter activity without expression of VP24^wt was set to 100%. The negative control was without expression of VP30 (Ctrl) and displayed the background of the assay. The mean and SD values of the results from three independent experiments are shown. The asterisks indicate statistical significance, as follows: ***, P < 0.001; ****, P < 0.0001. (B and C) Coimmunoprecipitation analysis. HEK293 cells were transfected with plasmids encoding Flag-tagged NP and one of the VP24 constructs. At 48 h p.t., the cells were lysed, and protein complexes were precipitated using anti-Flag M2 agarose. An aliquot of cell lysate (input) was collected before precipitation. (B) Western blot analysis was performed using Flag-, VP24-, and alpha-tubulin-specific antibodies. IP, immunoprecipitation. (C) The relative amounts of precipitated VP24 are displayed. The intensities of the bands of precipitated VP24 were normalized to the intensities of the respective bands in the cell lysates. The precipitants of VP24^wt were set to 100%. The mean and SD values of the results from three independent experiments are shown. (D) Huh-7 cells expressing NP, VP35, and with the addition of one of the VP24 constructs were fixed at 24 h p.t. and processed for transmission electron microscopy analyses. The absence or presence of electron-dense walls in the tubular-like NCLSs is indicated by arrows on the transversal or longitudinal sections of NCLSs. The boxed areas in the upper images were enlarged and are shown in the lower images. Scale bars are as follows: upper images = 500 nm, lower images = 200 nm.

FIG 3

Mutation of the YxxL motif in VP24 does not influence the incorporation of viral proteins into trVLPs. (A) Schematic representation of the trVLP assay. The producer (P0) cells were transfected with plasmids that encode the trVLP components. In the P0 cells, minigenome RNP transcription/replication, assembly, and budding of trVLPs were performed. The purified trVLPs were used for infection of indicator (P1) cells. (B to F) HEK293 cells were transfected with plasmids encoding the components of the trVLP assay (NP, VP35, VP30, VP40, GP, L, EBOV-specific minigenome, T7 polymerase, and either of the VP24 constructs). Cells were lysed at 72 h p.t., and trVLPs were purified from the supernatants of transfected cells by ultracentrifugation through a 20% sucrose cushion. (B) Cell lysates and purified trVLPs were processed for SDS-PAGE and analyzed by Western blotting using NP-, VP40-, VP35-, VP24-, and alpha-tubulin-specific antibodies. (C to F) The relative amounts of viral proteins in trVLPs are displayed by plotting the ratios between the band intensities of NP (C), VP35 (D), VP24 (E), and VP40 (F) in trVLPs and cell lysates. The band intensity of VP24^wt was set to 100%. The mean and SD values of the results from three independent experiments are shown.

FIG 4

Mutations of the YxxL motif in VP24 increased the infectivity of trVLPs and increased the percentage of shorter trVLPs. (A) The producer (P0) cells were transfected as described in Fig. 3B, and purified trVLPs were used for infection of indicator cells. The reporter gene activity in the producer cells was measured at 72 h p.t., and in indicator cells at 48 h p.i. The value of reporter activity in VP24^wt-expressing cells was set to 100%. The negative control was without the expression of VP30 (Ctrl) and displayed the background of the assay. The mean and SD values of the results from three independent experiments are shown. (B) RNA was isolated from purified trVLPs, and the amount of viral genomic RNA was analyzed by 2-step quantitative reverse transcription-PCR (qRT-PCR). The negative control was without the expression of VP30 (Ctrl) and displayed the background of the assay. The genome amount of VP24^wt was set to 1, and the fold change was analyzed. The mean and SD values of the results from three independent experiments are shown. The asterisks indicate statistical significance, as follows: **, P < 0.01; ***, P < 0.001; ***, P < 0.0001. (C and D) Electron microscopy (EM) analysis of the length and number of trVLPs produced in the presence of either the VP24^wt or VP24 mutants. (C) The length of over 100 particles was measured for every sample in three independent experiments. The percentages of trVLPs ranging from 200 to 500 nm in length and particles above 500 nm were calculated. The asterisks indicate statistical significance in comparison to VP24^wt, as follows: *, P < 0.05; **, P < 0.005. (D) The relative numbers of trVLPs produced in the presence of either VP24^wt or VP24 mutants. The number of trVLPs in the VP24^wt sample was set to 100%. The mean and SD values of the results from three independent experiments are shown. (E) EM images of trVLPs with condensed NCLSs formed in the presence of the VP24^wt or VP24 mutants. Scale bars = 100 nm.

FIG 5

Mutations in the YxxL motif of VP24 impaired intracellular transport of NCLSs. Huh-7 cells were transfected with plasmids encoding VP30-GFP, NP, VP35, and one of the VP24 constructs. Time-lapse images were captured from 20 h p.t. The pictures show the maximum intensity projection of time-lapse images of cells recorded for 3 min, and images were captured every 2 to 3 s. (B) The velocities of the NCLSs (n = 20). The mean and SD values are shown at the top of each graph. (C) The length of trajectories of the NCLSs (n = 20). The mean and SD values are shown at the top of each graph. The asterisks indicate statistical significance; ***, P < 0.001; ****, P < 0.0001. (D) The length of motile and immotile NCLSs was measured for each group (n = 50). The asterisks indicate statistical significance in comparison to VP24^wt; ***, P < 0.001; ****, P < 0.0001.

FIG 6

Analysis of the interaction between VP24 and Alix and live-cell imaging study of NCLS transport in Alix-knockdown cells. (A and B) Coimmunoprecipitation analysis of Flag-tagged Alix with each VP24 construct. HEK293 cells transiently expressing Flag-tagged Alix and one of the VP24 constructs were lysed at 48 h p.t., and protein complexes were precipitated using mouse anti-Flag M2 agarose. An aliquot of cell lysate (input) was collected before precipitation. (A) Western blot analysis was performed using Flag-, VP24-, and alpha-tubulin-specific antibodies. (B) The quantification of three independent experiments is shown in the graphs. The immunoprecipitated amount of VP24^wt was set to 100%. The mean and SD values of the results from three independent experiments are shown. The asterisk indicates statistical significance; *, P < 0.05. (C) Treatment of Huh-7 cells with siRNA targeting Alix (si-Alix) or nonspecific siRNA (si-Ctrl) was applied, and cell lysates were collected at 48 h p.t. Western blot analysis was performed using Alix- and alpha-tubulin-specific antibodies. (D) Functional control of siRNA-mediated downregulation of Alix. Huh-7 cells were transfected with 500 ng of pCAGGS-VP40. One day prior to transfection, cells were treated with si-Alix or si-Ctrl. Cell lysates and VP40-induced VLPs were collected at 48 h p.t. Western blot analysis was performed using Alix-, VP40-, and alpha-tubulin-specific antibodies. (E and F) Live-cell imaging analysis of NCLS transport in Huh-7 cells, which were pretreated with si-Alix or si-Ctrl 1 day prior to transfection. These cells were transfected either with plasmids encoding VP30-GFP, NP, VP35, and VP24^wt (E) or plasmids encoding VP30-GFP, NP, VP35, and VP24^AxxA (F). Time-lapse images were captured from 20 h p.t. The pictures show the maximum intensity projection of time-lapse images recorded for 2 min, and images were captured every 1 s. The graphics show the velocities of the NCLSs (n = 20). Mean and SD are shown as numbers.

FIG 7

Rescue of recombinant EBOV encoding double mutations in the YxxL motif of VP24. (A) Schematic representation of the procedure for full-length recombinant EBOV rescue experiments. Huh-7 cells were transfected with a full-length EBOV cDNA clone encoding VP24^wt or VP24^AxxA (FL-VP24^AxxA). The helper plasmids encoding NP, VP35, VP30, L, T7 polymerase, and, optionally, either VP24^wt or VP24^AxxA were transfected as well (passage 0). At 7 days p.t., the cell supernatants were collected and applied to fresh cells (passage 1). The cytopathic effect was monitored up to 7 days p.i., and the cell supernatants were collected. (B) Light microscopy of passage 1 cells at day 7 p.i. (C) Western blot analysis of purified particles. The cell supernatants were collected at 7 days p.i. in passage 1 cells and subjected to ultracentrifugation through a 20% sucrose cushion. Western blot analysis was performed using anti-EBOV serum. (D) Summarized results of three individual rescue experiments with a full-length EBOV cDNA clone encoding VP24^wt or VP24^AxxA. Additionally applied VP24 plasmids are indicated in the left column. FL-wt, full-length EBOV genome encoding VP24^wt; FL-AxxA, full-length EBOV genome encoding VP24^AxxA. The number of successful rescue experiments and total number of performed rescue experiments are indicated.

FIG 8

Influence of mutations in YxxL motif of VP24 on its IFN-antagonistic function. HEK293 cells were transfected with increasing amounts of plasmids (10 or 100 ng) encoding one of the VP24 constructs or an empty vector (control) and a luciferase-reporter plasmid under the control of an ISRE. Twenty-four hours later, the cell medium was replaced with either fresh medium (unstimulated) or medium containing 1,000 IU human IFN-β1a (stimulated). After a further 8 h, cells were lysed, and the reporter activity was measured. Fold induction was calculated as the ratio of reporter activities in stimulated to unstimulated cells. The mean and SD values of the results from three independent experiments are shown. Protein expression of VP24 (upon transfection with 100 ng of plasmids) and alpha-tubulin was analyzed by Western blotting.

FIG 9

Localization of the YxxL motif in VP24 within the NCLS structure. (A) The location of the VP24 YxxL motif in the subtomogram averaging structure of the EBOV nucleocapsid subunit (PDB identifier [ID] 6EHM) reported by Wan et al. (32). The suggested nucleoprotein is shown in blue. The putative RNA encapsidation site is highlighted in orange. The locations of the YxxL motifs in the two VP24 molecules (in deep and light green) associated with the nucleocapsid subunit are indicated in red. (B and C) The YxxL motif location in the V24 molecule structure (PDB ID 4M0Q). (B) The YxxL motif (in red) located at the tip of a small protrusion-like structure. (C) The YxxL motif (in red) located in the large protrusion.