The interaction between vRNPs and Rab11a is mediated by the RSV polymerase L

Previous studies have demonstrated that viral ribonucleoproteins (vRNPs) associate with Rab11a during transport. This interaction was confirmed by co-immunoprecipitation experiments in RSV infected cells [20]. To identify the viral protein that mediates vRNPs-Rab11a binding, we performed co-IP experiments using HEK-293T cells transiently co-expressing HA-Rab11a and one of the vRNPs proteins (N, P, L co-expressed with P or M2-1) fused to GFP. We also co-expressed the matrix protein M - known to associate with vRNPs [23]. Prior to lysis, cells were treated with DSP (dithiobis(succinimidyl propionate)), a cross-linker that stabilizes protein-protein interactions. Cell lysates were then incubated with agarose beads covalently linked to an anti-GFP nanobody (VHH) and bound proteins were revealed by western blotting (Fig 1A and 1B). Although GFP-L was co-expressed with P to ensure its stability, HA-Rab11a specifically co-immunoprecipitated with the L-P complex but not with P alone, indicating that L mediates the interaction with Rab11a. None of the other viral proteins tested (N, M2-1, or M) co-precipitated HA-Rab11a. Noteworthy immunoprecipitation experiment performed under native conditions also revealed an association between L and Rab11a, but the signal was weak and inconsistent (example on S1 Fig). These results demonstrate that the viral polymerase L interacts with Rab11a and suggest that the association between RSV vRNPs and Rab11a is mediated by L.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Viral polymerase L is the sole viral protein that interacts with Rab11a.

HEK 293T cells were transfected with wild-type HA-Rab11a together with individual viral proteins (N, P, L, M2-1, or M) for 24h. (A, B) Co-immunoprecipitation analysis. Cells were treated with DSP cross-linking agent for 30 min, lysed and GFP-Trap beads were used to immunoprecipitate viral proteins as described in methods section. (A) Analysis of input lysates and precipitated fractions by Western blot using anti-GFP, anti-P, and anti-HA antibodies is shown. Representative of three independent experiments. (B) HA-Rab11a band intensities were quantified using ImageLab software and normalized to HA-Rab11a co-precipitated with L-GFP (ratio = 1). Bar graph shows mean ± SD from three independent experiments. (C, D, E) Immunofluorescence colocalization analysis. The indicated viral protein was expressed alone (left column) or together with HA-Rab11a. Cells were fixed, immunolabelled and imaged using confocal microscopy to reveal viral proteins (green), HA-Rab11a (red) and nuclei (blue, Hoechst). (C) Images are representative of two independent experiments. Scale bars: 10 μm (main), 5 μm (zoom). Arrows indicate line scan vectors for intensity profiles. Fluorescence intensity is expressed as percentage of maximum signal detected for each condition. (D) Quantification of viral aggregate number per cell in absence or presence of HA-Rab11a co-expression. Bar graph shows mean ± SD from two independent experiments (n = 20 cells). **** p < 0.0001 by t-test with Welch’s correction. (E) HA-Rab11a enrichment in viral aggregates, calculated as ratio of HA-Rab11a intensity within aggregates to median cellular intensity. Dot plot shows mean ± SD, each point representing the mean enrichment ratio of aggregates for one cell. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 by Kruskal-Wallis ANOVA test followed by Kruskal-Wallis multiple comparisons tests for L-GFP sample. Data from n = 20 cells, two independent experiments. No values are included for cells without aggregates.

https://doi.org/10.1371/journal.ppat.1013671.g001

We further explored this interaction by analyzing the relative locations of Rab11a and the viral proteins using immunofluorescence microscopy. HEK-293T cells were transiently co-transfected with one expression plasmid encoding either N, P, L co-expressed with P, M2-1 or M, alongside either an empty plasmid or a HA-Rab11a expression plasmid (Fig 1C). In the absence of Rab11a overexpression, all viral proteins as detected by specific antibody staining (N, P) or GFP fluorescence (L-GFP, M2-1-GFP, M-GFP) displayed diffuse cytoplasmic distribution. By contrast, upon Rab11a overexpression, the polymerase L underwent significant relocation, forming distinct cytoplasmic aggregates that colocalized with HA-Rab11a. While this aggregation represents an overexpression artifact not observed in infected cells, it provides a reliable readout for protein-protein interactions in subsequent experiments. The other viral proteins (N, P, M2-1-GFP and M-GFP) retained their diffuse cytoplasmic localization, regardless of the levels of Rab11a expression. Intensity profile analysis along defined cellular segments confirmed specific colocalization between L-GFP and HA-Rab11a, with overlapping fluorescence peaks not observed for other viral proteins. Quantitative analysis of aggregate formation (Fig 1D) revealed a significant increase in L-GFP-positive aggregates upon Rab11a overexpression compared to control conditions, while other viral proteins showed no significant changes in aggregation patterns. To assess the specificity of this interaction, we calculated Rab11a enrichment within viral protein aggregates (Fig 1E) as the ratio of HA-Rab11a fluorescence intensity within aggregates to the median cellular HA-Rab11a intensity, averaged per cell. This analysis demonstrated specific enrichment of HA-Rab11a within L-GFP aggregates compared to the few aggregates formed by other viral proteins, confirming that L-GFP aggregates correspond to sites of L-Rab11a concentration. In infected cells, it has been shown that N and P undergo phase separation to form liquid condensates that recruit L and M2-1 [15,16]. We wondered whether overexpressed Rab11a would also be recruited to the N + P condensates. We transiently expressed N, P, and HA-Rab11a alone or together with L, M2-1, or M, but Rab11 was not recruited to the N + P condensates (see S2 Fig). However, when L is co-expressed, Rab11a is recruited to the periphery of the N + P condensates (see the plot profile in condition N + P + L in S2 Fig), but not when M2-1 or M are co-expressed.Taken together, these results indicate that the viral polymerase L interacts with Rab11a, suggesting that it is likely the viral protein responsible for mediating the association between RSV vRNPs and this cellular trafficking factor.

The polymerase L preferably interacts with the GTP-bound form of Rab11a

Rab11a is a small GTPase that cycles between two conformational states: an active, membrane-associated GTP-bound form, and an inactive, cytosolic GDP-bound form. To test whether the RSV polymerase L interacts preferentially with one of these states, we used two well-characterized Rab11a mutants: a constitutively active mutant Q70L (“CA”), locked in the GTP-bound state, and a dominant-negative mutant S25N (“DN”), locked in the GDP-bound state and unable to bind GTP [24]. Plasmids encoding HA-tagged Rab11a WT, CA, or DN were co-transfected with L-GFP in the presence of P. Co-immunoprecipitation assays were performed under cross-linking conditions, as described above (Fig 2A). Analysis revealed that L-GFP efficiently co-precipitated HA-Rab11a WT and HA-Rab11a CA, whereas HA-Rab11a DN was not detected in bound fraction (Fig 2A and 2B). These data demonstrate that L specifically interacts with the active, GTP-bound form of Rab11a.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Polymerase L interacts preferentially to the GTP-bound Rab11a.

HEK 293T cells were transfected with wild-type (WT), constitutively active (CA, Q70L) or dominant negative (DN, S25N) HA-Rab11a together L-GFP and P for 24h. (A, B) Co-immunoprecipitation analysis. Cells were treated with DSP cross-linking agent for 30 min, lysed and GFP-Trap beads were used to immunoprecipitate L-GFP. (A) Analysis of input lysates and precipitated fractions by Western blot using anti-GFP, and anti-HA antibodies is shown. Representative of three independent experiments. (B) HA-Rab11a band intensities were quantified using ImageLab software, normalized to the amount of L-GFP bound to the beads, and then expressed relative to HA-Rab11a CA co-precipitated with L-GFP (ratio = 1). Bar graph shows mean ± SD from three independent experiments. (C, D) Immunofluorescence colocalization analysis. Cells were fixed, immunolabelled and imaged using confocal microscopy to reveal L-GFP (green), HA-Rab11a (red) and nuclei (blue, Hoechst). Scale bars: 10 μm (main), 5 μm (zoom). (C) Images are representative of three independent experiments. Quantification of viral aggregates per cell is shown in S3A Fig (D) HA-Rab11a enrichment in L aggregates, calculated as ratio of HA-Rab11a intensity within aggregates to median cellular intensity. Dot plot shows mean ± SD, each point representing the mean enrichment ratio of aggregates for one cell. *** p < 0.001; **** p < 0.0001 by Brown-Forsythe and Welch ANOVA test followed by Dunnett’s multiple comparisons test. Data from n = 30 cells, three independent experiments.

https://doi.org/10.1371/journal.ppat.1013671.g002

We next assessed the subcellular distribution of Rab11a mutants relative to L using immunofluorescence microscopy. As observed previously, L-GFP expressed alone was diffusely distributed throughout the cytoplasm. When co-expressed with HA-Rab11a WT or HA-Rab11a CA, however, Rab11a and L relocalized into cytoplasmic aggregates (Fig 2C). In contrast, in the presence of HA-Rab11a DN, both proteins remained diffusely distributed without colocalization. To quantify aggregate formation, we measured the percentage of the total cell area occupied by L-GFP aggregates in each condition (S3A Fig). Cells co-expressing HA-Rab11a WT or CA displayed comparable fractions of their cytoplasmic area occupied by L-GFP aggregates, which was significantly larger than in cells expressing Rab11a DN. Consistently, Rab11a enrichment analyses (Fig 2D) demonstrated that HA-Rab11a WT and CA were specifically accumulated within L-GFP aggregates, whereas HA-Rab11a DN showed no enrichment. To determine whether Rab11a-DN failed to colocalise with vRNPs in infected cells, we analysed the distribution of vRNPs relative to Rab11a-DN in RSV-infected cells. In RSV-infected A549 cells expressing HA-Rab11a WT (A549-Rab11aWT), Rab11a displayed a vesicular distribution and showed partial colocalisation with vRNPs, in line with previous observations [20]. In contrast, HA-Rab11a DN exhibited diffuse cytoplasmic localisation and no longer formed discrete vesicular structures. Under these conditions, no apparent association between Rab11a DN and vRNPs was observed (S3B Fig). Taken together, these results demonstrate that RSV polymerase L interacts preferentially with the GTP-bound, active form of Rab11a, consistent with the role of active Rab11a in transport of vRNPs.

The polymerase L directly interacts with GTP-bound Rab11a

We next investigated whether the L-Rab11a interaction requires additional viral or cellular components or occurs through direct binding. To address this question, we performed GST pull-down assays using recombinant proteins. L-P^CTD complexes (P C-terminal Domain (CTD) is co-expressed to stabilize L) were expressed using the baculovirus system, while GST-tagged Rab11a WT, CA, and DN variants were produced in E. coli (S4A–S4C Fig). Purified proteins were incubated together in the presence of Glutathione Sepharose 4B beads to capture GST-fusion proteins. Pull-down analysis revealed direct interaction between L-P^CTD and Rab11a CA or Rab11a WT in the presence of GTP. By contrast, Rab11a DN and Rab11a WT in the absence of GTP showed no detectable binding to L-P^CTD (Fig 3A). These results demonstrate that the L-Rab11a interaction is direct and requires Rab11a to be in its GTP-bound conformation.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. The polymerase L directly interacts with GTP-bound Rab11a.

(A) Affinity coprecipitation with Glutathione Sepharose beads. Purified RSV L-Twin-Strep-tag + P[Δ1-120] was incubated with purified GST-Rab11a WT, GST-Rab11a(Q70L) CA, GST-Rab11a(S25N) DN and GST. After extensive washes, protein bounds were analyzed through Coomassie blue after SDS–polyacrylamide gel electrophoresis. Representative picture of five independent experiments (N = 5). (B,C) Real-time biolayer interferometry profiles showing association and dissociation of GST-Rab11a (B) and GST-Rab11a CA (C) with RSV L-Twin-Strep-tag + P-his loaded on Ni-NTA sensors.

https://doi.org/10.1371/journal.ppat.1013671.g003

To further characterize this interaction, we employed biolayer interferometry (BLI). Purified L-P^-his complexes were immobilized on Ni-NTA biosensors via the P^-his C-terminal His₆ tag and exposed to Rab11a variants. Initial experiments using single concentrations confirmed specific association of Rab11a WT in the presence of GTP and the constitutively active mutant, while no interaction was detected in the absence of GTP or with the DN mutant (S4D Fig). To determine the binding affinity, we exposed immobilized L-P complexes to increasing concentrations of either Rab11a (50–3200 nM) in the presence of GTP, or Rab11a CA (Fig 3B and 3C). Association kinetics were slow with equilibrium reached for all concentrations, enabling steady-state analysis. Fitting of equilibrium responses revealed specific interaction with an apparent KD in the sub-micromolar range for GST-Rab11a in presence of 1mM GTP (0.3 μM). Of note, minimal dissociation was observed during the buffer wash phase, indicating particularly stable binding under these in vitro conditions. However, this unusually high stability likely reflects the absence of physiological cofactors.

Rab11a does not abrogate RSV polymerase activity

Having established that Rab11a directly interacts with the viral polymerase, we hypothesized that Rab11a binding might modulate L function, potentially inactivating the polymerase during intracellular transport. To test this possibility, we employed three complementary assays measuring distinct polymerase activities in the presence or absence of Rab11a variants.

To directly evaluate the RNA-dependent RNA polymerase (RdRp) activity of L, we performed in vitro primer extension assays. In this system, purified L-P^-his complexes were provided with a synthetic RNA template annealed to a short fluorescently-labeled RNA primer, which is extended from the 3’ end by the polymerase using the template as a guide. Transcription products were visualized by denaturing gel electrophoresis. Gel analysis revealed distinct bands corresponding to synthesized RNA products with similar sizes and intensities across all conditions tested: GST control and GST-Rab11a fusion proteins (WT, CA, DN) (Fig 4A). Quantification of relative RNA transcript accumulation, expressed as percentage relative to the vehicle-treated control condition (set to 100%), confirmed that Rab11a variants had no detectable effect on L RdRp activity in vitro (Fig 4A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Rab11a interaction does not abrogate RSV L activity.

(A) 3’ primer-extension assay. RSV L-Twin-Strep-tag + P-his (0.2 µM) was incubated with a fluorescently-labeled RNA primer annealed to a 16nt RNA template. The RdRp activity leads to accumulation of fluorescent RNA primers extended with 1-12 nucleotides. The RdRp activity was assayed in presence of GST-Rab11a WT, CA or DN (7 µM). The activity in presence of PBS (vehicle) or GST was included as negative controls for inhibition, while an inhibitor (PC786, 1µM) of the RSV L RdRp activity was included as a positive control for inhibition. Fluorescent primers extended by the RSV polymerase were separated by Urea-PAGE and their fluorescence detected by a fluorescence imaging system. Top: Representative gel from 3 independent experiments; Bottom: Densitometry analysis of the intensity of fully +12nt extended primers (n = 3). (B) in vitro methyltransferase (MTase) activity assay. The RSV L-Twin-Strep-tag + P^CTD Methyltransferase activity was assayed in the presence of GST-Rab11a WT, CA or DN in vitro using an established protocol [26]. As a control, a known inhibitor of RSV methyltransferase activity (Sinefungin) was included (n = 3). (C) Polymerase activity assay in presence of Rab11a WT, CA or DN mutants. BSRT7/5 cells were transfected with plasmids encoding N, P, L, M2-1, and M/Luc subgenomic minireplicon together with control GFP or HA- Rab11a WT, HA-Rab11a(Q70L) CA, HA-Rab11a(S25N) DN. Polymerase activity was measured by luciferase luminescence normalized to β-galactosidase activity. Data represent individual data points and mean ± SD (n = 3).

https://doi.org/10.1371/journal.ppat.1013671.g004

Next, we assessed the methyltransferase (MTase) activity embedded in the RSV L-P complex using a radioactive filter-binding assay. The RSV L protein contains MTase activity within its C-terminal domain, which catalyzes methylation of viral mRNA caps (both 2′-O and N7 methylation), essential for transcript stability and translation efficiency. This assay monitors the transfer of methyl groups from a radioactive donor, tritiated S-adenosyl methionine ([³H]-SAM), to a synthetic RNA substrate containing a cap structure analog (GpppRSV9) by purified L-P^-his complexes. Following the methylation reaction, RNA products are captured on DEAE filter mats while free donor molecules are removed by washing. The incorporated radioactivity directly reflects L MTase activity. Rab11a was used in excess at 0.5, 1.6, or 4.7 μM compared to the RSV L-P complex at 50 nM. Increasing concentrations of purified Rab11a-CA did not statistically significantly affect MTase activity compared to control conditions (Fig 4B). A mild, dose-dependent reduction in RSV L polymerase MTase activity was observed in the presence of purified Rab11a WT and DN, but this was likely not related to L binding, as only Rab11a-CA binds to L in these experimental conditions (Fig 3). Altogether this suggests that Rab11a binding to RSV L does not abrogate the cap-modifying function of L in vitro.

Lastly, we assessed the impact of Rab11a on viral transcription and replication using an RSV minireplicon system [25]. BSRT7/5 cells were transiently transfected with plasmids expressing components of the viral polymerase complex (N, P, L, M2-1) along with a pseudoviral RNA construct containing the firefly luciferase reporter gene. In this setup, the level of luciferase activity reflects the ability of the viral proteins to replicate and transcribe the pseudoviral RNA. Co-transfection with Rab11a variants (WT, CA, or DN) enabled us to evaluate polymerase activity in the presence of different Rab11a conformational states. Luciferase activity measurements revealed no significant differences between control conditions and cells expressing either Rab11a WT, CA, or DN (Fig 4C). These results indicate that Rab11a does not substantially affect the transcription and replication functions of the viral polymerase complex in this cellular context.

Neither transcription, replication, RNA synthesis, nor cap methylation activities were significantly affected by the presence of Rab11a in any of its conformational states. Collectively, these results strongly suggest that Rab11a interaction does not result in inactivation of RSV polymerase enzymatic functions.

The Switch I region of Rab11a is required for interaction with L

Having established that L specifically interacts with the GTP-bound form of Rab11a, we sought to identify the molecular determinants of this interaction. The Switch I region of small GTPases is a critical interaction surface for effector proteins and has been implicated in viral protein binding. Previous studies demonstrated that the Switch I motif of Rab11a, specifically the conserved IGV sequence (residues 44–46), is a docking site for influenza A virus polymerase PB2 [27]. As both the influenza A and RSV polymerases interact with Rab11a, we hypothesized that RSV polymerase L might utilize a similar binding interface on Rab11a. To test this hypothesis, we generated a Rab11a mutant in which the conserved IGV sequence was replaced with three alanines, creating the Rab11a_AAA mutant (Fig 5A). This triple alanine substitution is expected to disrupt the native structure of the Switch I region, while likely preserving the overall protein fold, allowing us to specifically assess the contribution of this motif to L binding.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. The Switch I region of Rab11a is essential for interaction with RSV L polymerase.

(A) Schematic representation of Rab11a Switch I region mutation. Amino acids 44-46 were substituted from IGV to AAA. (B-F) HEK 293T cells were transfected with mutated (AAA) or unmutated (IGV) HA-Rab11a(Q70L) CA together with L-GFP and P for 24h. (B) Co-immunoprecipitation analysis. Cells were treated with DSP cross-linking agent for 30 min, lysed and GFP-Trap beads were used to immunoprecipitate L-GFP. Analysis of input lysates and precipitated fractions by Western blot using anti-GFP and anti-HA antibodies is shown. Representative of three independent experiments. (C) Quantification of co-immunoprecipitation. HA-Rab11a band intensities were quantified using ImageLab software, normalized to the amount of L-GFP bound to the beads, and then expressed relative to unmutated (IGV) HA-Rab11aCA (ratio = 1). Bar graph shows mean ± SD from three independent experiments. (D-F) Immunofluorescence colocalization analysis. Cells were fixed, immunolabelled and imaged using confocal microscopy to reveal L-GFP (green), HA-Rab11a (red) and nuclei (blue, Hoechst). (D) Images are representative of three independent experiments. Scale bars: 10 μm (main), 5 μm (zoom). Arrows indicate line scan vectors for intensity profiles. Fluorescence intensity is expressed as percentage of maximum signal detected for each condition. (E) Quantification of cellular area occupied by L aggregates. Bar graph shows mean ± SD from three independent experiments (n = 26 cells). ***p < 0.0001 by unpaired t-test with Welch’s correction. (F) HA-Rab11a enrichment in L aggregates, calculated as ratio of HA-Rab11a intensity within aggregates to median cellular intensity. Dot plot shows mean ± SD, each point representing the mean enrichment ratio of aggregates for one cell. **** p < 0.0001 by unpaired t-test with Welch’s correction. Data from n = 26 cells, three independent experiments.

https://doi.org/10.1371/journal.ppat.1013671.g005

Co-immunoprecipitation assays revealed a complete reduction of interaction between L-GFP and HA-Rab11aCA_AAA under cross-linking conditions (Fig 5B and 5C). While the HA-Rab11a CA signal in western blot analysis of the bound fraction was clear, only a very faint signal corresponding to background level was observed for HA-Rab11aCA_AAA, despite equivalent input protein levels. This result demonstrates that the IGV motif within the Switch I region is essential for the L-Rab11a interaction, even when Rab11a is locked in the active, GTP-bound conformation.

To further validate these findings, we examined the subcellular distribution of the HA-Rab11aCA_AAA mutant relative to L using immunofluorescence microscopy. Consistent with our co-immunoprecipitation results, HA-Rab11aCA_AAA failed to induce L-GFP aggregation as L-GFP remained diffusely distributed throughout the cytoplasm upon HA-Rab11aCA_AAA overexpression (Fig 5D). Quantification of L-GFP aggregate formation revealed that cells expressing Rab11aCA_AAA displayed significantly reduced aggregate area compared to those expressing Rab11a CA (Fig 5E). Furthermore, Rab11a enrichment analysis showed no significant accumulation of HA-Rab11aCA_AAA within the few L-GFP aggregates that formed, confirming the loss of specific interaction (Fig 5F).

These results demonstrate that the Switch I region of Rab11a, specifically the conserved IGV motif, is absolutely required for interaction with RSV polymerase L.

The two C-terminal domains of L are sufficient and essential for interaction with Rab11a

To further dissect the interaction from the viral side, we mapped the regions of L required for Rab11a binding. The RSV polymerase L is organized into five functional domains: the RNA-dependent RNA polymerase (RdRp), polyribonucleotidyl transferase (PRNTase), connecting domain (CD), methyltransferase (MTase), and C-terminal domain (CTD) (S5A Fig). Based on previous work characterizing the activity and structure of RSV L domains [28,29], we generated four truncated constructs to systematically test domain requirements: L_1-1460-GFP, containing the RdRp and PRNTase domains, L_1-1755-GFP, containing RdRp, PRNTase, and CD, GFP-L_1461-2165, containing CD, MTase, and CTD, GFP-L_1756-2165, containing MTase and CTD domains. It is noteworthy that the MTase activity of the GFP-L_1756-2165 was inactivated by mutating two residues of the catalytic site (R1820A and G1855S) to avoid cellular toxicity, while maintaining structural integrity [29].

We first screened these constructs for colocalization with Rab11a CA by immunofluorescence microscopy (S5B–S5D Fig). HA-Rab11aCA colocalized with the C-terminal fragments GFP-L_1461-2165 and GFP-L_1756-2165, but not with the N-terminal fragments L_1-1460-GFP or L_1-1755-GFP. To confirm these observations, we performed co-immunoprecipitation assays targeting GFP in cross-linked lysates of cells transiently expressing the truncated L proteins fused to GFP, together with P and HA-Rab11a variants (Fig 6A and 6B). Consistent with our immunofluorescence results the N-terminal fragment L_1-1755-GFP failed to co-precipitate HA-Rab11a in any conformational state when the full-length L (L_1-2165) and the C-terminal fragment GFP-L_1756-2165 specifically co-precipitated HA-Rab11a WT and CA but not HA-Rab11a DN. We further subdivided the C-terminal fragment GFP-L_1756-2165 into the MTase domain (L_1756-2039) and the CTD (L_2040-2165). Neither of these domains was sufficient to mediate co-precipitation of HA-Rab11a CA when expressed individually (S6A Fig). This suggests that either both regions contribute to the interaction, indicating that the binding site lies at the interface, or that subdividing these domains further disrupts their proper folding and structural integrity, thereby preventing any solid conclusion.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. The C-terminal domains of RSV L are essential and sufficient for Rab11a interaction.

(A, B) Co-immunoprecipitation analysis. HEK 293T cells transiently expressing HA-Rab11a WT, HA-Rab11a(Q70L) CA, HA-Rab11a(S25N) DN, together with P and L-GFP full lenght (1-2165) or fragments N-ter (1-1755) and C-ter (1756-2165) for 24h, were treated with DSP for 30 min, lysed and incubated with GFP-Trap beads. The GFP tag was inserted into the connecting domain of the L-GFP and the L_1-1755 fragment, and fused to the N-terminus of the L_1756-2165 fragment. (A) Analysis of input lysates and precipitated fractions by Western blot using anti-GFP, and anti-HA antibodies is shown. Representative of three independent experiments. (B) Quantification of co-immunoprecipitation. HA-Rab11a band intensities were quantified using ImageLab software and normalized to HA-Rab11a WT co-precipitated with full-length L-GFP (ratio = 1). Bar graph shows mean ± SD from three independent experiments. (C) GST pull-down assay. Sepharose beads coupled with GST or GST-Rab11a WT, CA or DN were incubated with lysates from HEK 293T cells transiently expressing L-GFP truncation constructs. Analysis of the precipitated fractions by Western blot using anti-GFP and anti-GST antibodies is shown. * indicates GFP-L_1756-2165 band. The lower band likely corresponds to a degradation product. Representative of two independent experiments. Quantification is presented on S6B Fig.

https://doi.org/10.1371/journal.ppat.1013671.g006

To further validate these findings, we performed reciprocal GST pull-down assays using purified GST-Rab11a variants immobilized on glutathione beads incubated with lysates from cells expressing GFP-tagged L constructs (Fig 6C and S6B). Both full-length L-GFP and C-terminal fragment GFP- L_1756-2165 were efficiently pulled down by GST-Rab11a CA but not by GST control or GST-Rab11a DN. No precipitation was observed with GST-Rab11a WT, possibly due to the absence of GTP in this experiment. Conversely, N-terminal fragments showed no detectable binding to any GST-Rab11a variant, confirming the specificity of the interaction for the C-terminal region of L. Based on our previous work showing that L expressed in the absence of P, although less soluble, remains properly folded and capable of interacting with its partners [30], we tested whether P is required for the L-Rab11a interaction. Similar GST pull-down results were obtained with full-length L-GFP expressed without P (S6C Fig), further confirming that P is dispensable for the L-Rab11a interaction.

These mapping studies demonstrate that the MTase and CTD domains of L (residues 1756–2165) are both necessary and sufficient for binding to Rab11a. The N-terminal domains (RdRp, PRNTase, and CD) spanning residues 1–1755 are dispensable for this interaction. Thus, the L-Rab11a interaction is mediated through the C-terminal portion of the polymerase.

Residue L1860 of polymerase L is essential for Rab11a interaction

To further define the Rab11a-binding site within the L C-terminal region, we sought to identify specific residues that might be critical for this interaction. Given that both RSV L and the cellular effector Rab11FIP2 interact with the Switch I region of Rab11a, we hypothesized that these proteins might share similarities in their Rab11a-binding interfaces. Alignment of the Rab11a-binding domain of Rab11FIP2 with the minimal L fragment capable of Rab11a interaction (L_1756-2165) revealed a region of sequence similarity located immediately downstream of the conserved GxGxG motif, a nucleotide-binding site essential for MTase activity conserved among Mononegavirales polymerases [29] (S7A Fig). We introduced a point mutation to replace leucine 1860 with alanine (L_L1860A-GFP). This leucine is located at a central position within the Rab11a binding domain of Rab11FIP2 and is highly conserved. Moreover, it is the only residue conserved between RSV L and Rab11-FIP2 that has been shown to be directly involved in Rab11-FIP2 interaction with Rab11a (highlighted in yellow in the sequence alignment, S7A Fig) [31]. To ensure that the L1860A mutation did not induce a complete misfolding of the L, we first assessed the catalytic activity of the L_L1860A mutant using the minigenome assay. It has been previously shown that the MTase domain is absolutely required to produce any transcriptional activity in this assay [32], and the introduction of point mutations directly in the MTase catalytic site strongly decrease transcriptional activity [29]. Here, we confirmed that the (L_{R1820A,G1855S}) mutation in catalytic residues drastically reduced the RSV transcriptional activity to ~10%, whereas the neighboring L1860A mutant retained approximately 25% of wild-type polymerase activity (S7B Fig). While the MTase function is altered with this mutation, it is not abrogated, we therefore evaluated the ability of the L1860A mutant to interact with Rab11a. The L_L1860A mutant showed a complete loss of Rab11a interaction in co-immunoprecipitation assays (Fig 7A and S7C). While wild-type L-GFP efficiently co-precipitated both HA-Rab11a WT and CA under cross-linking conditions, L_L1860A-GFP failed to recover detectable levels of Rab11a in any conformational state, despite equivalent expression levels. Immunofluorescence analysis confirmed these biochemical observations (Fig 7B). In contrast to wild-type L-GFP, which formed distinct cytoplasmic aggregates upon co-expression with HA-Rab11a CA, the L_L1860A mutant remained diffusely distributed throughout the cytoplasm with no evidence of colocalization or aggregate formation with Rab11a (S7D and S7E Fig). These results demonstrate that leucine 1860 is essential for the L-Rab11a interaction and support the conclusion that the MTase domain of L, particularly the area surrounding L1860, is involved in specific binding of the viral polymerase.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. Leucine 1860 in polymerase L is important for Rab11a interaction.

(A) Co-immunoprecipitation analysis. HEK 293T cells expressing HA-Rab11a WT, HA-Rab11a(Q70L) CA, HA-Rab11a(S25N) DN, together with P and L-GFP WT or L1860A mutant for 24h, were treated with DSP for 30 min 24 h post-transfection, lysed and incubated with GFP-Trap beads. Analysis of input lysates and precipitated fractions by Western blot using anti-GFP, and anti-HA antibodies is shown. Representative of two independent experiments. Quantification is presented on S7C Fig. (B) Immunofluorescence colocalization analysis. HEK 293T cells expressing HA-Rab11a CA together with P and L-GFP WT or L1860A mutant for 24h, were fixed, immunolabelled and imaged using confocal microscopy to reveal L-GFP WT or L1860A mutant (green), HA-Rab11a (red) and nuclei (blue, Hoechst). Scale bars: 10 μm (main), 5 μm (zoom). Images are representative of two independent experiments. Quantifications are presented on S7D and S7E Fig.

https://doi.org/10.1371/journal.ppat.1013671.g007

The L-Rab11a interaction is required for efficient vRNPs transport in infected cells

Having identified and characterized the L-Rab11a interaction biochemically, we sought to determine its functional relevance in the context of RSV infection. To test whether this interaction is necessary for vRNPs transport, we designed a competitive inhibition approach using the minimal L fragment (L_1756-2165) that binds to Rab11a as a dominant-negative inhibitor. We hypothesized that transient expression of mCherry-L_1756-2165 would compete with the full-length viral L for binding to Rab11a, thereby disrupting (at least partially) vRNP-Rab11a association and possibly impairing vRNPs transport. To prevent potential toxicity from methyltransferase activity, we mutated the catalytic site within the MTase domain (R1820A and G1855S mutations [29]) to eliminate enzymatic function while preserving the Rab11a-binding capability. HEp-2 cells were transfected with either mCherry-L_1756-2165 (competitive inhibitor) or mCherry alone (control), then infected with RSV-GFP-N to visualize vRNPs particles. At 18 hours post-infection, live-cell imaging was performed to track individual vRNPs movements over time as previously described in Consentino et al. [20] (See S1 and S2 Movies). Particle tracking analysis allowed quantitative assessment of multiple transport parameters including maximum speed, velocity and displacement.

Quantitative analysis of vRNPs tracking data revealed significant effects of L_1756-2165 expression on vRNPs dynamics. While maximum speed (Fig 8A) showed only a modest, non-significant decrease, both track displacement (Fig 8B) and track velocity (Fig 8C) were significantly reduced in cells expressing the competitive inhibitor compared to mCherry controls. Visualization of centered track projections showed markedly reduced spatial coverage of vRNPs tracks in cells expressing mCherry-L_1756-2165 compared to control cells, indicating less efficient long-range transport (Fig 8D and 8E). This spatial confinement suggests that vRNPs retain some motility but lose their ability to undergo sustained, directional transport.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 8. Long-range RNP transports depends on the L-Rab11a Interaction.

HEp-2 cells were transfected with control mCherry or mCherry-L_1756-2165 and infected with RSV-GFP-N. Live-cell imaging was performed at 18 h post-infection for vRNP tracking analysis (see method). (A-C) Quantification of RNP tracks parameters. Dot plots show mean ± SD and each data point represents the median values of track maximum speed (A), track displacement (B), and track velocity (C) from individual cells. *p < 0.05 by unpaired t-test with Welch’s correction. Data from n = 12 cells across two independent experiments. (D-E) Representative centered projection of RNPs tracks in cells transfected with control mCherry (D) or mCherry-L_1756-2165 (E) over 60 seconds. The corresponding time-lapse sequences are shown in S1 and S2 Movies. (F, G) Effect of Rab11a DN on viral growth. A549-HA-Rab11a-WT and A549-HA-Rab11a-DN cells were infected with RSV-GFP at an MOI of 0.005 and GFP fluorescence level was measured every 4h for 72h. (F) Scatter plot of GFP signal at different time post-infection in A549-HA-Rab11a-WT (green) or -DN (red). The data shown are the mean ± SD of four independent experiments performed in quadruplicate. (G) GFP intensity in infected A549-HA-Rab11a-DN cells at 72 hpi relative to WT. *** p < 0.0001 by unpaired t-test with Welch’s correction.

https://doi.org/10.1371/journal.ppat.1013671.g008

These results provide functional evidence that the L-Rab11a interaction plays a critical role in vRNPs transport within infected cells. The moderate effects observed may be due to the competitive nature of our inhibition approach. In this approach, the endogenous full-length L protein can still compete with the L_1756-2165 fragment for binding to Rab11a, which could enable residual vRNPs transport.

To further evaluate the significance of this transport step, and of Rab11a in general, for RSV replication and spread, we examined viral replication kinetics in A549 cells stably expressing either HA-Rab11a WT or DN infected with RSV-GFP at a low multiplicity of infection. Viral multiplication was quantified by measuring the GFP signal every four hours over 72 hours post infection [33]. At each time point, GFP levels in cells expressing Rab11a DN were lower than in cells expressing Rab11a WT (Fig 8F). At 72 hours post-infection, the GFP signal was ~ 50% lower in cells expressing HA-Rab11a DN than in cells expressing HA-Rab11a WT, indicating that the DN mutant impairs RSV replication (Fig 8G). This confirms the involvement of Rab11a in RSV multiplication.

Cells

Human embryonic kidney 293T (HEK-293T) cells, HEK-293T T7 cells and BHK-21 cells (clone BSRT7/5) stably expressing T7 RNA polymerase, were maintained in Dulbecco’s Modified Eagle Medium (DMEM), and human epidermoid larynx carcinoma HEp-2 cells (ATCC: CCL-23) were cultured in Eagle’s minimum essential medium (MEM). A549 cells constitutively expressing HA-Rab11a WT or HA-Rab11a DN are a kind gift from Dr. Nadia Naffakh [53] and were grown in DMEM supplemented with 1μg/mL puromycin. All media were supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS) and 1% (v/v) penicillin-streptomycin solution (5,000 U/mL). Cells were grown in an incubator at 37°C under 5% CO₂. Spodoptera frugiperda (SF9; ATCC, CRL-1711 Gibco cat# 11496015) were propagated in suspension using Sf-900 II serum-free medium (Gibco cat# 10902096) at 28°C without CO₂.

Virus

The recombinant wild-type RSV, RSV-GFP and RSV-GFP-N are derived from the human RSV Long strain sequence and were previously described [20,33]. They were amplified on HEp-2 cells at 37°C (3 passages) and titrated by plaque assay on HEp-2 [33].

Plasmids

The pM/Luc subgenomic replicon containing the firefly luciferase (Luc) gene and expression plasmids pCI-N, pCI-GFP-N, pCI-P, pCI-P-BFP and pCI-M2-1-GFP have been previously described [15,54,55]. The pCI-P-GFP was generated by replacing the BFP tag in pCI-P-BFP. The vector was digested with MluI and Eco81I, and the GFP insert obtained by gene synthesis (Twist Bioscience) was ligated using T4 DNA ligase. Expression plasmids pCI-L and pCI-M were constructed by cloning mammalian codon-optimized RSV L and M coding sequences into the pCI vector (GenBank U47119). Briefly, coding sequences were amplified by PCR using Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific) with specific primers and cloned into pCI using MluI/XbaI restriction sites for L and MluI/NotI for M. To generate fluorescent fusion constructs, the GFP coding sequence was inserted into the Hinge 2 region of L as previously described [34] or at the C-terminus of M sequence, yielding pCI-L-GFP and pCI-M-GFP expression vectors, respectively. Truncated L constructs were generated from pCI-L-GFP by restriction digestion followed by Gibson assembly (New England Biolabs) as detailed in S1 Table. Briefly, pCI-L-GFP was digested with appropriate restriction enzymes. The GFP tag and deleted L coding sequences were PCR-amplified using Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific) with primers designed to create overlapping ends. For constructs retaining the connector domain (pCI-L_1-1755-GFP and pCI-GFP-L_1461-2165), the eGFP tag remained in its original Hinge 2 position. For constructs lacking the connector domain (pCI-L_1-1460-GFP and pCI-GFP-L_1756-2165), the eGFP coding sequence was repositioned respectively at the C-terminus or N-terminus. PCR products were assembled with digested vectors using the Gibson Assembly Master Mix (New England Biolabs). The pCI-GFP-L_1756-2039 and pCI-GFP-L_2040-2165 constructs were generated by PCR-mediated deletion from the pCI-GFP-L_1756-2165 plasmid, followed by In-Fusion cloning (Takara Bio). MTase-inactivating mutations (R1820A and G1855S) were introduced into pCI-GFP-L_1756-2165 and pCI-GFP-L_1756-2039 by site-directed mutagenesis using the In-Fusion HD Cloning Kit (Takara Bio) [29]. The L1860A point mutation was introduced into pCI-L-GFP using the same approach. Mutagenic primers were designed to incorporate the desired substitutions and included 15 bp regions of homology flanking the mutation site. The entire plasmid was amplified using Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific). The parental template plasmid was digested with DpnI. The linearized plasmid was recircularized using the In-Fusion HD Cloning Kit according to the manufacturer’s instructions. All constructs were verified by Sanger sequencing.

The pHA-Rab11a WT, pHA-Rab11a CA (Q70L), and pHA-Rab11a DN (S25N) expression plasmids were generated by replacing the N-terminal GFP tag with an HA epitope tag in peGFP-Rab11a constructs (kindly provided by Dr. Nathalie Sauvonnet). The parental peGFP-Rab11a plasmids (WT, CA, or DN) were digested with NheI/XhoI (WT) or NheI/SalI (CA and DN) to excise the eGFP coding sequence. Complementary oligonucleotides encoding the HA epitope tag (YPYDVPDYA) with compatible overhangs were synthesized, annealed, and ligated into the linearized vectors. The Rab11a Switch I mutant (I44A/G45A/V46A, designated Rab11a_AAA) was generated from pHA-Rab11a CA by site-directed mutagenesis (In-Fusion HD Cloning Kit (Takara Bio)) to replace the conserved IGV sequence with three alanines as described above. All constructs were verified by Sanger sequencing. All primer sequences used in this study are available upon request.

For the purification of GST-tagged Rab11a proteins, the reference sequence for human Rab11a (uniprot identifier P62491) was codon-optimized for expression in Escherichia coli and a DNA fragment containing this sequence was synthesized (Twist Bioscience), alongside sequences variants encoding the CA (Constitutively Active) variant Q70L or DN (Dominant Negative) variant S25N. These sequences were inserted in pGEX 4T-3 expression plasmids (Invitrogen) multiple cloning sites (between EcoRI and XhoI restriction sites) downstream of a GST sequence, a thrombin-cleavage site sequence (LVPRGS), and short flexible linker sequence (PNSGGGGSGGGGS) using the In-Fusion HD Cloning Kit (Takara cat# 639650) to form a pGEX-Rab11a, pGEX-Rab11a CA and pGEX-Rab11a DN plasmids. All plasmids were amplified and then validated by Sanger sequencing.

For the purification of RSV P-L complexes, we modified a previously described pFastBacDual plasmid allowing the co-expression of both a codon-optimized RSV L protein with an HA-tag (YPYDVPDYASLGGP) at position 1740 under the control of the polyhedrin promoters, and a codon-optimized P protein with an N-terminal GST-tag under the control of the p10 promoters [30]. Using the In-Fusion HD Cloning Kit (Takara cat# 639650), the L protein sequence was extended at the C-terminus with a precision protease cleavage site (LLEVLFQGP) and a Twin-Strep-tag (SAWSHPQFEKGGGSGGGSGGSAWSHPQFEK), while the P protein N-terminal GST-tag was excised and its C-terminus was extended with a tobacco etch virus cleavage site (GSENLYFQ) and a 6xhis tag (HHHHHH), resulting in a pFBD RSV L-Twin-Strep-tag + P-his-tag plasmid. In parallel, we used this method to construct a pFBD RSV L-Twin-Strep-tag + P[Δ1–120] plasmid that allows the expression of a truncated P sequence that only contain the minimum P sequence that is required for L protein stabilization, as previously reported [56].

Cell transfection and infection

Cells were seeded 24 hours prior to transfection to reach 90% confluence at the time of transfection in either 24-well plates (immunofluorescence), Ibidi μ-Slide 4-well chambers (live imaging), or 10 cm dishes (co-immunoprecipitation). Transfections were performed using Lipofectamine 2000 (Thermo Fisher Scientific) at a ratio of 2.5 μL per 1 μg of DNA, with DNA amounts adjusted to culture vessel size (0.5 μg for 24-well plates/μ-Slides; 10 μg for 10 cm dishes), in antibiotic-free growth medium, according to the manufacturer’s instructions. Cells were fixed or collected at 24 hours post-transfection (hpt) for subsequent analyses. For transfection-infection assays, cells were transfected as described above in MEM supplemented with 2% FBS without antibiotics. At 1 hpt, RSV viral stocks diluted in the same medium were added to the cells at multiplicity of infection (MOI) of 1. Cells were incubated at 37°C under 5% CO₂ and processed at 18 hours post-infection.

Immunofluorescence and colocalization analysis

At 24 hours post-transfection, cells were fixed with 4% paraformaldehyde in PBS for 15 minutes, permeabilized with PBS containing 1% BSA and 0.1% Triton X-100 for 10 minutes, and incubated for 1 hour at room temperature with primary antibodies diluted in PBS-BSA 1%: rabbit anti-RSV N (VIM Unit, INRAE, 1:40,000), rabbit anti-RSV P (VIM Unit, INRAE, 1:1,000) [57], or mouse anti-HA (BioLegend, clone 16B12, 1:800). After two washes with PBS, cells were incubated for 1 hour with Alexa Fluor 488- or 647-conjugated secondary antibodies against rabbit or mouse IgG (H + L) (Thermo Fisher Scientific, 1:1000), phalloidin-Atto-590 (Sigma-Aldrich, 1/800), and Hoechst 33342 (1 μg/mL). Coverslips were mounted with ProLong Diamond antifade reagent (Thermo Fisher Scientific). Representative cells were captured by confocal microscopy using an inverted spinning-disk confocal microscope (IXplore SpinSR, Evident, Tokyo, Japan) equipped with a 60 × oil-immersion objective lens (Plan-Apochromat, Evident) and an ORCA-Fusion BT camera (C15440, Hamamatsu Photonics, Japan). Z-stacks were acquired and maximum intensity projections (1 μm) were generated using ImageJ (NIH).

For colocalization analysis, viral protein-positive aggregates were automatically detected in co-transfected cells using threshold-based segmentation in ImageJ, allowing quantification of aggregate number per cell and surface area. Rab11a enrichment within aggregates was calculated as the ratio of mean Rab11a fluorescence intensity in viral protein-positive regions to the median Rab11a intensity in the entire cell. All image analyses were performed using custom ImageJ macros. Statistical analyses and data visualization were performed using Prism (GraphPad).

Cross-linking and GFP Trap co-immunoprecipitation

Transfected cells were incubated in 300 µM DSP (dithiobis(succinimidyl propionate), Thermo Fisher Scientific) in PBS for 30 minutes at room temperature. The amine-crosslinking reaction was quenched with 20 mM Tris-HCl (pH 7.5) in PBS for 15 minutes at room temperature. Cells were then collected and lysed in 500 µL/10⁷ cells of buffer containing 25 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA, 0.5% NP40, 1 mM PMSF, and a cocktail of protease inhibitors (Thermo Scientific Pierce) on a rotary wheel at 4°C for 1 hour. Lysates were cleared by centrifugation (10,000 g, 10 minutes at 4°C) and incubated with GFP-Trap beads overnight on a rotary wheel at 4°C according to the manufacturer’s protocol. Beads were washed three times with lysis buffer before being resuspended in NuPAGE LDS sample buffer containing NuPAGE sample reducing agent (Thermo Fisher Scientific).

Co-immunoprecipitation assay using GST-tagged proteins

GST-Rab11a conjugated beads were prepared before use: Glutathione Sepharose 4B beads (Cytiva) were incubated with purified GST-Rab11a fusion proteins (purification described below) for 1 hour at room temperature with constant rotation. Briefly, 5 mg of purified GST-fusion proteins were mixed with 1 mL of pre-equilibrated beads in binding buffer (50 mM Tris-HCl pH 7.5, 300 mM NaCl, protease inhibitor cocktail). Efficient protein loading onto beads was verified by SDS-PAGE analysis followed by Coomassie blue staining. Transfected cells were harvested and lysed in ice-cold lysis buffer containing 50 mM Tris-HCl pH 7.5, 300 mM NaCl, protease inhibitor cocktail, 0.1 mg/mL BSA, 1 mM DTT, 1 mM PMSF, and 0.6% NP-40. Cells were incubated in lysis buffer for 1 hour at 4°C with constant rotation to ensure complete lysis. Following centrifugation at 12,000 × g for 10 minutes at 4°C to remove cellular debris, clarified lysates were incubated with GST-Rab11a conjugated beads for 3 hours at 4°C under constant rotation. After binding, beads were washed three times with wash buffer (50 mM Tris-HCl pH 7.5, 300 mM NaCl, protease inhibitor cocktail) to remove non-specifically bound proteins. Bound protein complexes were eluted by resuspending washed beads in NuPAGE LDS sample buffer containing NuPAGE sample reducing agent (Thermo Fisher Scientific).

Western blot

Samples were denatured at 80°C for 10 minutes, separated by SDS-PAGE, and transferred to PVDF membranes using liquid transfer systems according to standard protocols. Membranes were blocked for 1 hour at room temperature in PBS containing 0.25% Tween-20 (PBS-T) and 5% non-fat dry milk, then incubated overnight at 4°C with primary antibodies diluted in PBS-T containing 1% BSA: rabbit anti-RSV P (VIM Unit, INRAE, 1:1,000), rat anti-HA (Roche, 1:800), mouse anti-GFP (Roche, 1:2,000), or HRP-conjugated mouse anti-GST (BioLegend, 1:1,000). After washing with PBS-T, membranes were incubated for 1 hour at room temperature with appropriate HRP-conjugated secondary antibodies for mouse, rabbit, or rat IgG (H + L) (1:20,000; Promega or Santa Cruz Biotechnology). Proteins were visualized using SuperSignal West Pico PLUS or Atto Ultimate chemiluminescence substrates (Thermo Fisher Scientific) on a ChemiDoc imaging system (Bio-Rad). Densitometric quantification was performed using Image Lab software (Bio-Rad).

Protein expression and purification

The RSV L-P polymerase complex was expressed and purified using a protocol based on a previously described Sf9/baculovirus system [30] with the following modifications. Bacmids allowing for the expression of a yellow fluorescent protein and carrying either the RSV L-Twin-Strep-tag + P-his-tag sequences (shortened as L-P^-his in the main text) or the RSV L-Twin-Strep-tag + P[Δ1–120] sequences (shortened as L-P^CTD in the main text) were recovered after transformation of the corresponding pFBD plasmids in the DH10EMBacY bacteria (Geneva Biotech). The bac-to-bac baculovirus expression system (Invitrogen) was used to generate recombinant baculoviruses in SF9 cells by transfection with Cellfectin II (Gibco) reagent following the manufacturer’s instructions. Baculovirus stocks were amplified twice at MOI of 0.01 in SF9 cells to generate high titer baculovirus stock (1–3*10⁸ TCID50/ml). Titration was performed by 50% tissue culture infectious dose (TCID50) using yellow fluorescence to monitor infection.

For L-P protein production, SF9 cells in suspension were infected at a MOI of 1 for 72h. Infected cells were harvested by centrifugation at 300 g for 10 mins and lysed in a lysis buffer containing 50 mM Tris-HCl pH 8.0, 300 mM NaCl, 0.5% Igepal, 1 mM DTT, 2.5 U/ml Benzonase nuclease and an antiproteases and antiphosphatases cocktail (Pierce) for 1h on ice. Lysates were clarified by centrifugation 20 mins at 20,000xg at 4°C. The L-P complex was purified using a two-step chromatographic approach. Cell lysate was first applied to a Strep-Trap XT column (Cytiva) and the L-Twin-Strep-tag was eluted with biotin-containing buffer (50 mM Tris-HCl pH 8.0, 300 mM NaCl, 1 mM DTT, 50 mM Biotin). The eluate was then concentrated on Vivaspin 6 100kDa (Sartorius) and subjected to size exclusion chromatography on a Superose 6 increase 10/300 GL column (Cytiva) pre-equilibrated with 50 mM Tris-HCl pH 8.0, 300 mM NaCl, 1 mM DTT. Fractions containing the L and P proteins were collected and further concentrated on Vivaspin 6 100kDa (Sartorius) to ~0.5-1 mg/ml, flash-frozen and stored at -80°C. Proteins were stored in 50 mM Tris-HCl pH 8.0, 300 mM NaCl, 1 mM DTT buffer at 0.6 mg/ml at -80°C.

Wild-type and mutant GST-Rab11a fusion proteins (GST-Rab11a, GST-Rab11a CA, and GST-Rab11a DN) along with GST control were expressed after transformation of corresponding pGEX 4T-3 plasmids in Escherichia coli BL21 cells. Bacterial cultures were selected by ampicillin (100 µg/ml) and grown at 37°C to an OD₆₀₀ of 0.6, then induced with 0.33 mM IPTG for 16 hours at 28°C.

Cells were harvested by centrifugation and resuspended in lysis buffer containing 50 mM HEPES pH 7.4, 500 mM NaCl, 10% glycerol, 0.1% octyl β-D-glucopyranoside, 1 mM DTT, 1 mg/mL lysozyme, protease inhibitor cocktail (Roche), and 2.5 U/mL Benzonase nuclease following a previously described protocol [27]. Cell debris after sonication were removed by centrifugation at 15,000 × g for 30 minutes at 4°C. The soluble fraction was applied to Glutathione Sepharose 4B FastFlow resin (Cytiva). After extensive washing, bound proteins were eluted with 50 mM reduced glutathione in 50 mM HEPES pH 7.4, 500 mM NaCl. Eluted fractions were further purified by size exclusion chromatography using a Superdex 200 increase 10/300 GL column pre-equilibrated with PBS. Peak fractions were pooled, and protein concentrations were determined using the Bradford assay with bovine serum albumin as standard. Final protein concentrations ranged from 2-5 mg/ml, and aliquots were flash-frozen and stored at -80°C. A quality control of the protein samples was conducted in accordance with the ARBRE-MOBIEU guidelines [58].

GST pull-down assay

Recombinant GST, GST-Rab11a wild-type, GST-Rab11a CA, and GST-Rab11a DN fusion proteins were normalized to equimolar concentrations (35 μM) in PBS. Glutathione Sepharose 4B beads (20 μl settled volume per reaction) were pre-equilibrated in binding buffer (50 mM Tris-HCl pH 8.0, 300 mM NaCl) and incubated with 4 µM of each GST fusion protein and 0,6 µM of RSV L-Twin-Strep-tag + P[Δ1–120] supplemented with 0.1 mg/ml BSA and 0.02% Tween 20 for 1 hour at room temperature with constant shaking at 1300 rpm. For nucleotide-dependent binding studies, GTP was included at a final concentration of 1 mM. Following binding, beads were washed twice with binding buffer containing 0.1 mg/ml BSA and 0.02% Tween 20, then twice with binding buffer alone. When applicable, GTP (1 mM) was maintained throughout all wash steps. Bound proteins were eluted in 2 × SDS sample buffer and resolved by SDS-PAGE and visualized by Coomassie blue staining. Input controls consisting of individual GST constructs and L-P complex were included in all experiments.

Biolayer interferometry

Protein-protein interactions and binding affinities between RSV L-Twin-Strep-tag + P-his complexes and GST-Rab11a variants were analyzed by biolayer interferometry (BLI) using an Octet RED 384 system at 25°C with orbital shaking at 1,000 rpm. Purified RSV L-P complexes and GST-Rab11a variants were buffer-exchanged into assay buffer containing 50 mM Tris-HCl pH 8.0, 300 mM NaCl, 0.02% Tween-20, 100 μg/mL or 1 mg/mL BSA and 1 mM GTP when specified, using PD-10 Desalting Columns (Cytiva). For Protein-protein interaction assays, RSV L-Twin-Strep-tag + P-his complexes (20 μg/mL) were loaded onto Ni² ⁺ -NTA biosensors via the C-terminal 6 × His tag on the P protein for 900 seconds. Streptavidin-His6 coated sensors were used as controls to assess non-specific sensor binding. After baseline establishment in assay buffer (60 seconds), loaded biosensors were exposed to a 2 µM concentration of GST-Rab11a variants for 600s (association) and to assay buffer for 300s (dissociation step). As control for signal drift, loaded sensors were incubated in assay buffers in parallel. For binding affinity measurements, loaded sensors were incubated in two-fold serial dilutions of GST-Rab11a or GST-Rab11a CA fusion proteins ranging from 50 nM to 3.2 μM (two-fold increments). Association was monitored for 60 minutes until binding reached steady state. Then biosensors were incubated in assay buffer for 5 min (dissociation). BLI data were processed using Octet Data Analysis software. Since minimal dissociation was observed, equilibrium dissociation constants (KD) were determined by fitting steady-state binding responses to a one-site binding model using GraphPad Prism software, after double referencing of the BLI response by subtracting non-specific sensor binding and baseline drift signals.

Minigenome assay

Cells in 24-well plates were transfected as described above with a plasmid mixture containing the following RSV minigenome system components per well: 200 ng pCI-N, 200 ng pCI-P, 100 ng pCI-L or variant constructs (pCI-L-GFP, pCI-L L_1860A-GFP), 50 ng pM2-1, 200 ng pCI-GFP or pHA-Rab11a variants (WT, CA or DN), and 200 ng pGEM-M/Luc containing the RSV minigenome cassette with firefly luciferase reporter gene. To normalize for transfection efficiency, 10 ng of pβ-Gal (expressing β-galactosidase under CMV promoter control) was included in each transfection. Negative control transfections were performed by replacing pCI-L with empty pCI vector. All experimental conditions were performed in biological triplicate, with each independent experiment repeated three times. At 24 h post-transfection cells were lysed in luciferase lysis buffer (25mM Tris pH 7.9, 8mM MgCl₂, 1mM DTT, 1% Triton X-100 and 15% glycerol) for 5 minutes at 37°C. Luciferase activity was measured using a Centro XS³ LB 960 luminometer (Berthold Technologies) and normalized on β-galactosidase activity.

RNA synthesis

RNA primer (5′-[6-FAM]ACCA-3′) and RNA template (5’-CUUUGUUUUUUCUGGU-3’) were chemically synthesized on a solid support at 1 µmol scale using an ABI 394 automated synthesizer with commercial 2’-O-pivaloyloxymethyl 3’-O-phosphoramidite ribonucleosides and 6-FAM phosphoramidite (Chemgenes, USA). After assembly of RNA sequences, the solid support was directly subjected to 1M DBU in dry CH₃CN for 3 minutes then 30% aqueous ammonia solution for 3h at 40 °C to recover the crude RNA material. After filtration, evaporation in the presence of isopropylamine, and coevaporation with water, the two short RNA substrates were purified by IEX-HPLC and their identity was confirmed by MALDI-TOF mass spectrometry. 191 nmoles of pure RNA template and 442 nmoles of pure RNA primer were obtained.

In vitro polymerase assay

RNA polymerase activity was assessed using a primer extension assay based on a previously described assay, with a modification: radiolabeled nucleotides were omitted and instead the RNA primer was fluorescently labeled in order to visualize transcription products by fluorescence [28]. Each reaction (5 μL final volume) contained 0.2 μM recombinant RSV L-Twin-Strep-tag + P-his complex, 7 μM GST fusion proteins (GST control, GST-Rab11a WT, GST-Rab11a CA, or GST-Rab11a DN), 2 μM RNA template (3′-UGGUCUUUUUUGUUUC-5′), and 100 μM fluorescent RNA primer (5′-[6-FAM]ACCA-3′). Reactions were performed in buffer containing 50 mM Tris-HCl pH 8.0, 60 mM NaCl, 8 mM MgCl₂, 1 mM TCEP, 10 μM each NTP, and 10% glycerol. As a control, a reaction containing the RSV L inhibitor PC786 at 1 µM was included in a separate reaction [59]. Following incubation for 3 hours at 30°C, reactions were terminated by adding one volume (5 µl) of dionized formamide with 25 mM EDTA and denatured at 95°C for 3 minutes. RNA products were resolved on 20% polyacrylamide gels containing 7 M urea and visualized using a GelDoc Go imaging system (Bio-Rad). After imaging, the intensity of the fully extended transcripts (+12nt) was measured using ImageJ software, and the signals were normalized to the vehicle control in each independent experiment (n = 3).

MTase activity assay

Methyltransferase activity was measured using a filter-binding assay performed according to the method described previously [26]. Reactions contained 50 nM of RSV L-P complex, GST fusion proteins at varying concentrations (0.5, 1.6, or 4.7 μM), 1.8 μM synthetic RNA substrate (GpppRSV₉: 5′-GpppGGGACAAAA-3′), 0.17 μM unlabeled S-adenosyl methionine (SAM), and 0.8 μM [³H-methyl]-SAM (PerkinElmer) in reaction buffer (50 mM Tris-HCl pH 8.0). After 3 hours incubation at 30°C, reactions were quenched by a 20-fold dilution in cold water. Samples were transferred to DEAE filtermats (PerkinElmer) using a Filtermat Harvester (Packard Instruments) to separate methylated RNA products from radiolabelled SAM. The RNA-retaining mats were washed twice with 10 mM ammonium formate pH 8.0, twice with water and once with ethanol. They were then soaked with scintillation fluid (PerkinElmer), and ³H-methyl transfer to the RNA substrates was determined using a Wallac MicroBeta TriLux Liquid Scintillation Counter (PerkinElmer).

Live imaging and vRNPs tracking

Time-lapse images were acquired for 1 min at 37 °C and 5% CO₂ using an inverted spinning-disk confocal microscope (IXplore SpinSR, Evident, Tokyo, Japan) equipped with a 60 × oil-immersion objective lens (Plan-Apochromat, Evident) and an ORCA-Fusion BT camera (C15440, Hamamatsu Photonics, Japan). The system was fitted with a CSU-X1 confocal scanning unit and a SORA disk module (Yokogawa, Tokyo, Japan), coupled to a 3.1 × relay lens. System control and image acquisition were performed using CellSens software (Molecular Devices, Sunnyvale, CA, USA). Live-cell imaging and particle tracking analyses were performed as previously described [20]. Briefly, Z-stacks were acquired at intervals of less than 200 milliseconds and processed using ImageJ for maximum projections and background subtraction. Particle tracking was conducted using Imaris software 10.2 (Bitplane Inc.) with the autoregressive motion algorithm. The data were further filtered and analyzed using a custom Python script, which is available at https://github.com/mawelti/RSV-RNP-TrackAnalysis. Statistical comparisons between medians were performed using Prism software (GraphPad Inc.).

Viral multiplication assays

A549 cells stably expressing HA-Rab11a WT or DN were seeded in 24-well plates (1.25 × 10⁵ cells per well) in quadruplicate, 24 h prior to infection with RSV-GFP at a multiplicity of infection (MOI) of 0.005. GFP fluorescence was monitored every 4 h for 72 h post-infection using an IncuCyte S3 live-cell imaging system (Essen Bioscience), with images acquired using a 10 × objective (36 fields per well). Fluorescence signals were analyzed using the IncuCyte software by quantifying the total green object integrated intensity (GCU × µm² per image) for each well at each time point. The percentage of integrated GFP intensity in HA-Rab11a DN-expressing cells relative to HA-Rab11a WT controls was calculated from the mean integrated GFP intensity across quadruplicate wells at 72 h post-infection. Statistical analyses and data visualization were performed using GraphPad Prism (GraphPad Software).