A conserved viral amphipathic helix governs the replication site-specific membrane association

Abstract

Positive-strand RNA viruses assemble their viral replication complexes (VRCs) on specific host organelle membranes, yet it is unclear how viral replication proteins recognize and what motifs or domains in viral replication proteins determine their destinations. We show here that an amphipathic helix, helix B in replication protein 1a of brome mosaic virus (BMV), is necessary for 1a's localization to the nuclear endoplasmic reticulum (ER) membrane where BMV assembles its VRCs. Helix B is also sufficient to target soluble proteins to the nuclear ER membrane in yeast and plant cells. We further show that an equivalent helix in several plant- and human-infecting viruses of the Alsuviricetes class targets fluorescent proteins to the organelle membranes where they form their VRCs, including ER, vacuole, and Golgi membranes. Our work reveals a conserved helix that governs the localization of VRCs among a group of viruses and points to a possible target for developing broad-spectrum antiviral strategies.

Fig 1. Mutations in BMV 1a helix B inhibit viral genome replication.

(A) Helix A and helix B are highlighted in blue and red respectively and their amino acid sequences are shown. Note region E is indicated as a 114 amino acid-long sequence consisting of helix A and helix B. (B) Helical wheel projection generated using an 18 aa analysis window in Heliquest (https://heliquest.ipmc.cnrs.fr/). The predicted hydrophobic face is formed by W419, G430, F423, W416, F427, and C420, as indicated by the dotted lines. The size of each residue in the helical wheel represents the bulkiness of their sidechain and the arrow in the helical wheel corresponds to the hydrophobic moment. (C) BMV genomic replication was measured by using Northern blot for WT and various 1a mutants. Positive- and negative-strand viral RNAs were detected using BMV strand-specific probes. 18S rRNA served as a loading control. Band intensities were measured using Adobe photoshop and the standard deviations are indicated. All experiments were repeated at least three times in duplicates each time. Note: L3A has 3 Leu residues replaced by Ala in helix A and serves as a negative control.

Fig 2. Mutations in BMV 1a helix B inhibit the nuclear ER membrane association of BMV 1a.

(A) Membrane association of BMV 1a mutants was determined via a membrane flotation assay. Histograms represent the percentage of membrane-associated proteins as determined by the signal intensity of each protein segregated into the top three membrane fractions compared to the sum of all six fractions. Lysates of yeast cells expressing WT or mutant 1a were subjected to an iodixanol density gradient and analyzed by western blotting using antibodies against ER membrane protein Dpm1, soluble protein Pgk1, or His6 for His6-tagged BMV 1a derivatives. The values shown are from at least two independent experiments. The adjusted p-values for all samples except Dpm1 compared to WT 1a were <0.05. (B) Localization of mC-tagged 1a mutants in yeast cells co-expressing Sec63-GFP as examined using confocal microscopy. Nuclear ER and peripheral ER associations are indicated by solid and dotted arrows, respectively. The yellow signal in the merge panels is representative of colocalization of mC-tagged WT or mutant 1a with Sec63-GFP. Scale bars: 2μm.

Fig 3. Helix B is capable of targeting soluble proteins to the nuclear ER membrane.

(A) Membrane flotation analysis of GFP-fused region E, helices A+B, helix A, and helix B. Histograms represent the percentage of proteins detected in the top three fractions compared to the sum of all six fractions. A representative western blot is shown, with bands corresponding to respective 1a fragments detected using anti-GFP antibody. The values indicated were obtained from at least two independent experiments. (B) Localization of GFP-tagged region E, helices A+B, helix A, or helix B co-expressed with an mC-tagged ER marker protein Scs2. Dotted arrows point to punctate structures and solid arrows point to the nER membrane. Images were obtained using a confocal microscope. (C) Membrane flotation analysis for Pgk1-GFP and HB-Pgk1-GFP, using 1a-His6 as a positive control. The values are indicative of two independent experiments. (D) Colocalization of HB-PGK-GFP at the nER membrane with Scs2-mC. Solid arrows point to the nER membrane. Images are obtained using a fluorescence microscope. Scale bars: 2μm.

Fig 4. Helix B-mediated organelle targeting is conserved across members of the Alsuviricetes class.

(A) Colocalization of GFP-tagged HB of CCMV, CMV, or RuV with an ER marker Scs2-mC, and colocalization of the mC-tagged HEV HB and Sec63-GFP at the nER membrane. Solid arrows point to the nER membrane and dotted arrows point to the putative vacuole membrane. (B) Colocalization observed for CMV HB and RuV HB with vacuole membranes stained by a lipophilic dye, FM4-64 [34]. The solid arrows point to vacuolar membranes, the arrowheads point to the putative nER membrane, and the dotted arrow in the RuV-HB panel shows punctate structures. (C) Colocalization of RuV HB-mC with GFP-tagged Sed5, a marker for vesicles that traffic from the ER to the Golgi. Images were obtained using a confocal microscope. Scale bars: 2μm.

Fig 5. Localizations of BMV and CMV helix B in Nicotiana benthamiana cells parallel those in yeast cells.

(A) YFP-tagged helix B from BMV or CMV was expressed in histone 2B-RFP transgenic N. benthamiana plants via agroinfiltration. Solid and dotted arrows point to the nER membrane and the nucleus, respectively. (B) BMV or CMV HB-YFP was co-expressed with an mC-tagged ER marker. Arrows point to the nER membrane and arrowheads point to small vesicles that are colocalized with the ER marker. (C) Colocalization of CMV HB-YFP with an mC-tagged tonoplast marker as indicated by dotted arrows. The fluorescence signal was observed under laser confocal microscopy at 40 hpai. Scale bars: 50μm.