Cap-independent co-expression of dsRNA-sensing and NF-κB pathway inhibitors enables controllable self-amplifying RNA expression with reduced immunotoxicity
- Department of Pharmacology, University of Cambridge, United Kingdom
- MRC Laboratory of Molecular Biology, United Kingdom
Figures
Figure 1 with 1 supplement
Differential effects of moderate and strong double-stranded RNA (dsRNA)-sensing pathway inhibition on self-amplifying RNA (saRNA) transgene expression and cell number.
(a) Schematic of saRNA constructs co-expressing fluorescent reporters and dsRNA-sensing pathway inhibitors. nSP1–4 encodes the saRNA replicase. mScarlet3 indicates cap-dependent transgene expression and is expressed from subgenomic RNA (angled arrow denotes subgenomic promoter). EGFP indicates cap-independent transgene expression and is expressed from an IRES. A second IRES expresses varying levels of dsRNA-sensing pathway inhibition: ‘Conventional saRNA’ expresses moxBFP (control). ‘E3’ expresses vaccinia virus E3 (dsRNA-binding protein). ‘E3-NSs-L*’ expresses E3 plus Toscana virus NSs (ubiquitin ligase targeting PKR [protein kinase R]) and Theiler’s virus L* (RNase L inhibitor). saRNA was transfected into mouse primary fibroblast-like synoviocytes (FLS), labeled with BioTracker to indicate cell number. (b) Representative composite images of microplate wells showing EGFP (green) and mScarlet3 (red). (c) Representative images of microplate wells showing BioTracker. (d) Longitudinal quantification of EGFP (n=11). E3 provided the highest expression, while E3-NSs-L* expressed at intermediate levels. (e) Longitudinal quantification of mScarlet3 (n=11). E3 provided the highest expression, while E3-NSs-L* expressed at intermediate levels. (f) Longitudinal quantification of BioTracker (n=11). Conventional saRNA led to immediate and long-term reductions in signal. E3 initially maintained signal, but it gradually decreased over time. E3-NSs-L* preserved signal throughout the time course. For panels (b–c): Scale bar = 5 mm. For panels (d–f): Data are normalized to starting cell number (pre-transfection BioTracker signal). Statistical significance relative to mock transfection was assessed using two-way repeated-measures (RM) ANOVA with Greenhouse-Geisser correction and Dunnett’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Data are presented as mean ± standard error of the mean (SEM). The mock transfection control data is also presented in Figure 5c–e.
-
Figure 1—source data 1
Whole-plasmid sequencing results for self-amplifying RNA (saRNA) constructs depicted in Figure 1a.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig1-data1-v1.zip
-
Figure 1—source data 2
Numerical data used to generate the plots in Figure 1.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig1-data2-v1.xlsx
Figure 1—figure supplement 1
Inhibiting dsRNA-sensing pathways or saRNA replication mitigates saRNA-induced cell loss.
(a) Area under the curve (AUC) analysis of BioTracker data shown in Figure 1f, integrating effects over 3 weeks (n=11). Signal preservation increased stepwise with greater dsRNA-sensing pathway inhibition. Statistical analysis was performed on AUC values using one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test to compare all groups. Mock transfection data is also presented in Figure 5—figure supplement 1a. (b) CellTag, a cell number normalization dye, quantified day 2 post-transfection (n=29). Signal preservation increased stepwise with greater dsRNA-sensing pathway inhibition. Data were normalized to mock transfection. Statistical analysis was performed using one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test to compare all groups. Data in this panel were pooled from in-cell western assays, incorporating data presented in Figure 3, Figure 3—figure supplement 1c, Figure 6, and additional data not shown elsewhere. (c) Calcein AM, a viability dye, quantified one day after mock or conventional saRNA transfection, with or without the replicase inhibitor ML336 (n=4). Conventional saRNA reduced cell viability, an effect prevented by ML336. Data were normalized to pre-transfection levels. Statistical analysis was performed using two-way RM ANOVA with Greenhouse-Geisser correction, followed by Dunnett’s multiple comparisons test relative to the mock-transfected, vehicle-treated group. For all statistical reporting, *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Data are presented as mean ± SEM.
-
Figure 1—figure supplement 1—source data 1
Numerical data used to generate the plots in Figure 1—figure supplement 1.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig1-figsupp1-data1-v1.xlsx
Figure 2
E3-NSs-L* prevents saRNA-induced elevations in annexin V staining and reductions in cell viability.
(a) Representative cropped microplate well images of Annexin V-CF800, indicating phosphatidylserine exposure or loss of membrane integrity. FLS were transfected with saRNA or treated with staurosporine, an apoptosis inducer. (b) Representative cropped microplate well images of calcein AM. Cultures are the same as in panel (a). (c) Quantification of Annexin V positive area, determined using Li thresholding and normalized to average mock transfection values (n=6). Annexin V positive area was increased by staurosporine, conventional saRNA, and E3—but not by E3-NSs-L*. Statistical significance relative to mock transfection was determined using two-way RM ANOVA with Bonferroni’s multiple comparisons test. Data are presented as mean ± SEM. (d) Quantification of calcein AM (n=6). Conventional saRNA and E3 reduced signal, an effect not observed with E3-NSs-L*. Data are normalized to starting cell number (pre-transfection BioTracker signal). Statistical significance was determined by one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test comparing all groups. All groups differed significantly from staurosporine (significance indicators omitted for clarity). Connecting lines indicate responses from the same biological replicate. For panels (a–b): Scale bar = 1.5 mm. For panels (c–d): *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.
-
Figure 2—source data 1
Numerical data used to generate the plots in Figure 2.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig2-data1-v1.xlsx
Figure 3 with 1 supplement
dsRNA-sensing pathway inhibitors suppress saRNA-induced eIF2α phosphorylation but not the reduction in eIF4E phosphorylation.
(a) Phosphorylated eIF2α levels examined by in-cell western (n=6). Conventional saRNA increased eIF2α phosphorylation, while E3 and E3-NSs-L* did not. Data are normalized to total eIF2α and presented as fold-change relative to mock transfection. (b) Total eIF2α levels examined by in-cell western (n=5). E3-NSs-L* increased total eIF2α. Data are normalized to CellTag and presented as fold-change relative to mock transfection. (c) Phosphorylated eIF4E levels, normalized to total eIF4E, examined by in-cell western (n=6). All constructs reduced eIF4E phosphorylation. (d) Total eIF4E levels, normalized to CellTag, examined by in-cell western (n=6). No significant differences were revealed (F(3,15) = 1.207, p=0.3410). For all panels: Statistical significance was determined by one-way RM ANOVA and Tukey’s multiple comparison test. *p<0.05, **p<0.01, ***p<0.001. The mock transfection control data are also used in Figure 6a–d. Connecting lines indicate responses from the same biological replicate.
-
Figure 3—source data 1
Numerical data used to generate the plots in Figure 3.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig3-data1-v1.xlsx
Figure 3—figure supplement 1
E3-NSs-L* protects against saRNA-induced PKR upregulation, RNA degradation, and decreased protein synthesis rate.
(a) PKR levels examined by in-cell western (n=4). E3-NSs-L* inhibits PKR upregulation. Data are normalized to CellTag and presented as fold-change relative to mock transfection. (b) rRNA integrity assessed from total RNA using the RNA Integrity Number (RIN) algorithm (n=5). E3-transfected cells demonstrate lower RIN than E3-NSs-L*–transfected cells. Data are shown as a Gardner-Altman comparison plot, with the mean effect size ± 95% confidence interval (CI) illustrated on the right subpanel. Dotted lines indicate group means. Statistical significance was determined using a paired t-test (p=0.0054). (c) Protein translation rates measured by puromycin incorporation, normalized to CellTag (n=5). Conventional saRNA or E3 reduced protein synthesis rates, an effect not observed with E3-NSs-L*. For panels (a,c): Statistical significance was determined by one-way RM ANOVA with Tukey’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Connecting lines indicate responses from the same biological replicate.
-
Figure 3—figure supplement 1—source data 1
Numerical data used to generate the plots in Figure 3—figure supplement 1.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig3-figsupp1-data1-v1.xlsx
Figure 4 with 3 supplements
saRNA-triggered cytokine responses are variably affected by dsRNA sensing inhibition, but broadly suppressed by NF-κB inhibition.
(a) Antiviral cytokine response quantified by multiplex bead-based immunoassay (n=6). Inhibition of dsRNA-sensing pathways reduced some cytokine responses while enhancing others. Cytokine levels were normalized to pre-transfection cell number (indicated by BioTracker), scaled within each biological replicate with the highest value set to 100%, and shown as a heatmap of group means. Plots of unscaled data are presented in Figure 4—figure supplement 1. Statistical significance was assessed on unscaled data by one-way repeated-measures ANOVA for each cytokine. Treatment effect was significant for all cytokines after controlling for multiple comparisons using the two-stage linear step-up procedure of Benjamini, Krieger, and Yekutieli (FDR = 5%). Tukey’s multiple comparisons test identified between-group differences. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 vs. mock transfection; #p<0.05, ##p<0.01, ###p<0.001 vs. conventional saRNA. The mock transfection control is shared with panel (c). (b) Schematic of saRNA constructs designed for inhibiting inflammatory signaling. dsRNA-sensing pathway inhibitors and EGFP are expressed via an IRES. A second IRES expresses variable amounts of inflammatory signaling pathway inhibition: ‘moxBFP’ expresses moxBFP (control). ‘srIκBα’ expresses srIκBα (inhibiting NF-κB). ‘srIκBα-Smad7-SOCS1’ expresses srIκBα plus Smad7 (inhibiting TGF-β) and SOCS1 (inhibiting IFN). (c) Antiviral cytokine response quantified by multiplex bead-based immunoassay (n=6). srIκBα and srIκBα-Smad7-SOCS1 broadly suppressed cytokine secretion. Data normalization, visualization, and statistical analysis were performed as described in panel (a), but scaling was applied independently to account for the different constructs tested. Plots of unscaled data are presented in Figure 4—figure supplement 2. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 vs. mock transfection; #p<0.05, ##p<0.01, ###p<0.001, ####p<0.0001 vs. moxBFP. The mock transfection control is shared with panel (a). Abbreviations: srIκBα, super repressor inhibitor of κBα; Smad7, mothers against decapentaplegic homolog 7; SOCS1, suppressor of cytokine signaling 1; IFN, interferon; CXCL, C-X-C motif chemokine ligand; TNF, tumor necrosis factor; MCP-1, monocyte chemoattractant protein-1; IL, interleukin; CCL5, chemokine ligand 5; GM-CSF, granulocyte-macrophage colony-stimulating factor.
-
Figure 4—source data 1
Whole-plasmid sequencing results for plasmids depicted in Figure 4b.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-data1-v1.zip
Figure 4—figure supplement 1
Detailed cytokine responses to saRNA constructs inhibiting dsRNA-sensing pathways.
(a–m) Unscaled data underlying the heatmap in Figure 4a, showing individual cytokine responses (n=6). Conventional saRNA triggered robust secretion of antiviral cytokines. E3 and E3-NSs-L*reduced IFN-α and IFN-β. E3-NSs-L* additionally suppressed TNF. However, E3 increased MCP-1, and E3-NSs-L* increased GM-CSF relative to conventional saRNA. Data were normalized to starting cell number (indicated by pre-transfection BioTracker signal). Statistical significance was assessed using one-way repeated-measures ANOVA for each cytokine. Multiple comparisons were controlled using the Benjamini-Krieger-Yekutieli false discovery rate (FDR) procedure (Q=5%). All cytokines passed the discovery threshold. Treatment effects were analyzed using Tukey’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Connecting line indicates responses from the same biological replicate. Mock transfection controls are shared with Figure 4—figure supplement 2.
-
Figure 4—figure supplement 1—source data 1
Numerical data used to generate the plots in Figure 4—figure supplement 1.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-figsupp1-data1-v1.xlsx
Figure 4—figure supplement 2
Detailed cytokine responses to saRNA constructs inhibiting inflammatory signaling pathways.
(a–m) Unscaled data underlying the heatmap in Figure 4c, showing individual cytokine responses (n=6). moxBFP significantly increased all cytokines relative to mock transfection. Cytokine responses were broadly suppressed by both srIκBα and srIκBα-Smad7-SOCS1. Data were normalized to starting cell number (indicated by pre-transfection BioTracker signal). Statistical significance was assessed using one-way repeated-measures ANOVA for each cytokine. Multiple comparisons were controlled using the Benjamini-Krieger-Yekutieli false discovery rate (FDR) procedure (Q=5%). All cytokines passed the discovery threshold. Treatment effects were analyzed using Tukey’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Connecting line indicates responses from the same biological replicate. Mock transfection controls are shared with Figure 4—figure supplement 1.
-
Figure 4—figure supplement 2—source data 1
Numerical data used to generate the plots in Figure 4—figure supplement 2.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-figsupp2-data1-v1.xlsx
Figure 4—figure supplement 3
Reduced antiviral gene expression and replicon activity observed with co-expression of inflammatory signaling inhibitors.
(a) Antiviral and proinflammatory transcripts quantified by qPCR (n=3). Conventional saRNA upregulated antiviral and proinflammatory transcripts. E3 and srIκBα-Smad7-SOCS1 elicited less upregulation, with the latter showing the greatest suppression. Statistical comparisons were performed on ΔCT values normalized to 18S rRNA using one-way ANOVA with multiple comparisons controlled using the two-stage linear step-up procedure of Benjamini, Krieger, and Yekutieli (FDR = 5%). All investigated transcripts passed significance thresholds for discovery. Treatment effects were analyzed using Holm-Šídák’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001 vs. mock transfection; #p<0.05 vs. conventional saRNA. Mean –ΔΔCT values normalized to mock transfection are shown on the heatmap (larger values indicate higher expression). (b) EGFP, encoded on all saRNA constructs, quantified by qPCR (n=3). srIκBα-Smad7-SOCS1 demonstrated higher ΔCT values (indicating lower EGFP transcript levels) compared to conventional saRNA or E3. Expression was normalized to 18S rRNA, and data are shown as ΔCT values. Statistical comparisons were performed on ΔCT values using one-way ANOVA with Tukey’s multiple comparisons test. (c) In vitro transcribed saRNA constructs were resolved by denaturing gel electrophoresis. Conventional saRNA (11,181 nt), E3 (11,030 nt), and E3-NSs-L* (12,562 nt) migrated as single bands corresponding to their expected sizes. In contrast, the moxBFP (13,282 nt), srIκBα (13,614 nt), and srIκBα-Smad7-SOCS1 (15,664 nt) constructs each exhibited two bands: one at the expected size and a smaller lower intensity band of consistent size across all three constructs, suggestive of a common truncated transcript. (d) Band intensity plots of moxBFP, srIκBα, and srIκBα-Smad7-SOCS1 derived from the gel shown in panel (c).
-
Figure 4—figure supplement 3—source data 1
Numerical data used to generate the plots in Figure 4—figure supplement 3.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-figsupp3-data1-v1.xlsx
-
Figure 4—figure supplement 3—source data 2
PDF file containing the original gel image for Figure 4—figure supplement 3c, with relevant bands labeled.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-figsupp3-data2-v1.pdf
-
Figure 4—figure supplement 3—source data 3
Unmodified/uncropped original gel image as shown in Figure 4—figure supplement 3c.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig4-figsupp3-data3-v1.zip
Figure 5 with 1 supplement
Differential effects of srIκBα and srIκBα-Smad7-SOCS1 on cell number and transgene expression.
(a) Representative images of microplate wells showing BioTracker. (b) Representative composite images of microplate wells showing EGFP (green) and mScarlet3 (red). (c) Longitudinal quantification of BioTracker (n=11). srIκBα reduces signal, an effect not observed with srIκBα-Smad7-SOCS1.(d) Longitudinal quantification of EGFP (n=11). Expression was low across constructs. (e) Longitudinal quantification of mScarlet3 (n=11). srIκBα-Smad7-SOCS1 produced 2–3 times greater fluorescence than other constructs. For panels (a–b): Scale bar = 5 mm. For panels (c–e): Data are normalized to starting cell number (pre-transfection BioTracker signal). A dotted line shows the response to conventional saRNA (for reference). Statistical significance of treatment effects at each time point compared to mock transfection was determined by two-way RM ANOVA with Greenhouse-Geisser correction and Dunnett’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Data are presented as mean ± SEM. The mock transfection control data used in this figure is also presented in Figure 1d–f.
-
Figure 5—source data 1
Numerical data used to generate the plots in Figure 5.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig5-data1-v1.xlsx
Figure 5—figure supplement 1
srIκBα induces cell loss, an effect not observed with srIκBα-Smad7-SOCS1.
(a) AUC analysis of BioTracker fluorescence data shown in Figure 5c, integrating effects over 3 weeks (n=11). srIκBα reduces integrated signal, while srIκBα-Smad7-SOCS1 preserves it. Statistical analysis was performed on AUC values using one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test. Mock transfection data is also presented in Figure 1—figure supplement 1a. (b) CellTag signal on day 2 post-transfection, normalized to mock transfection (n=20). srIκBα reduces signal, an effect not observed with srIκBα-Smad7-SOCS1. Statistical significance was determined by one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test. Data in this panel were pooled from in-cell western assays, incorporating data presented in Figures 3 and 6, as well as additional data not shown elsewhere. ***p<0.001. (c) Calcein AM signal, 3 days post-transfection, normalized to moxBFP (n=4). The signal reduction observed with srIκBα is not seen with srIκBα-Smad7-SOCS1. Statistical significance was determined by a paired t-test. For all statistical reporting, *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001. Data are presented as mean ± SEM or individual values with connecting lines indicating responses from the same biological replicate.
-
Figure 5—figure supplement 1—source data 1
Numerical data used to generate the plots in Figure 5—figure supplement 1.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig5-figsupp1-data1-v1.xlsx
Figure 6
srIκBα reduces eIF2α phosphorylation and total eIF4E levels, effects not observed with srIκBα-Smad7-SOCS1.
(a) Phosphorylated eIF2α levels examined by in-cell western, normalized to total eIF2α levels (n=6). Phosphorylation was reduced by srIκBα but unchanged by srIκBα-Smad7-SOCS1. (b) Total eIF2α levels examined by in-cell western, normalized to CellTag (n=5). No significant differences were revealed (F(2,8) = 3.683, p=0.0735). (c) Phosphorylated eIF4E levels examined by in-cell western, normalized to total eIF4E levels (n=6). No significant differences were revealed (F(2,10) = 1.336, p=0.3059). (d) Total eIF4E levels examined by in-cell western, normalized to CellTag (n=6). srIκBα reduced total eIF4E, while srIκBα-Smad7-SOCS1 had no effect. For all panels: Data are presented as fold change relative to mock transfection. Statistical significance was determined by one-way RM ANOVA and Holm-Šídák’s multiple comparisons test to compare all groups. *p<0.05, **p<0.01, and ***p<0.001. Connecting lines indicate responses from the same biological replicate. Mock transfection data used for normalization are the same as in Figure 3.
-
Figure 6—source data 1
Numerical data used to generate the plots in Figure 6.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig6-data1-v1.xlsx
Figure 7
Long-term expression of saRNA constructs inhibiting inflammatory signaling reduces basal fibroblast activation factor-α (FAP-α) levels.
(a) Representative in-cell western images showing FAP-α, a marker of fibroblast activation, on day 11 post-transfection. Columns show different biological replicates, and rows show different treatments. The tiled composite on the right shows FAP-α signal normalized to CellTag signal (FAP-α/CellTag). (b) Quantification of FAP-α, normalized to CellTag (n=8). Both srIκBα and srIκBα-Smad7-SOCS1 significantly reduce FAP-α levels versus mock transfection, while moxBFP shows no significant difference. Statistical significance was determined by one-way RM ANOVA with Greenhouse-Geisser correction and Dunnett’s multiple comparisons test to compare groups to mock transfection. Connecting lines indicate responses from the same biological replicate. **p<0.01.
-
Figure 7—source data 1
Numerical data used to generate the plots in Figure 7.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig7-data1-v1.xlsx
Figure 8
Reversible and irreversible control of saRNA replicon activity with ML336.
Panels (a–b) show the effect of ML336, a small-molecule replicase inhibitor, applied at the time of transfection with srIκBα-Smad7-SOCS1. (a) ML336 concentration–response curve for inhibiting mScarlet3 expression, measured 2 days post-transfection (n=5). IC50 = 8.5 nM (95% CI: 6.1–11.9 nM). Data are normalized to vehicle-treated controls and fit to a variable-slope sigmoidal curve. (b) Reversibility of ML336-mediated inhibition of mScarlet3 expression determined by measuring fluorescence recovery, one day after compound washout from cultures in panel (a). mScarlet3 inhibition is reversible at intermediate concentrations (10 nM–1 μM) but irreversible at high concentrations (>1 μM). Data are normalized to each biological replicate’s maximum recovery and fit to a bell-shaped concentration–response curve.
Panels (c–g) show the effect of ML336 on an established srIκBα-Smad7-SOCS1 replicon. FLS were continously treated (indicated by shading) with vehicle or 1 μM ML336, starting 1 day post-transfection. (c) Representative composite images of microplate wells showing EGFP (green) and mScarlet3 (red). Scale bar = 5 mm. (d) Representative images of microplate wells showing calcein AM at end of time course. Scale bar = 5 mm. (e) Quantification of EGFP (n=6). By day 7, EGFP signal is reduced by ML336. (f) Quantification of mScarlet3 (n=6). By day 3, mScarlet3 signal is reduced by ML336. (g) Quantification of calcein AM (n=6). Vehicle, but not ML336 treatment, reduced calcein AM signal relative to mock transfection. Statistical significance was assessed by one-way RM ANOVA with Greenhouse-Geisser correction and Tukey’s multiple comparisons test. **p<0.01, ***p<0.001. All fluorescence data were normalized to pre-transfection cell number indicated by BioTracker signal. Connecting lines indicate responses from the same biological replicate. For panels (e−f): Statistical significance relative to vehicle-treated cells was determined by two-way RM ANOVA with Greenhouse-Geisser correction and Dunnett’s multiple comparisons test. #p<0.05, ##p<0.01, ###p<0.001. Data are presented as mean ± SEM.
-
Figure 8—source data 1
Numerical data used to generate the plots in Figure 8.
- https://cdn.elifesciences.org/articles/105978/elife-105978-fig8-data1-v1.xlsx
Figure 9
Diagrammatic comparison of conventional and immune-evasive self-amplifying RNA (saRNA).
Conventional saRNA activates multiple innate immune responses upon replication, leading to translation shutdown, host mRNA degradation, cytotoxicity, and cytokine secretion. These immunotoxic effects restrict its therapeutic utility to vaccine applications and often necessitate co-administration of exogenous immunosuppressants to enable effective transgene expression. In contrast, immune-evasive saRNA co-expresses inhibitors of key dsRNA-sensing and inflammatory signaling pathways via cap-independent translation, intrinsically suppressing these immune responses. This enables sustained transgene expression while reducing immunotoxicity and eliminating the need for exogenous immunosuppressive agents. Transgene expression can be externally regulated or terminated using small-molecule inhibitors of the saRNA replicase, such as ML336, providing control over the duration of therapy and a means to mitigate potential adverse effects.
Appendix 1—figure 1
Confocal microscopy images confirming fibroblast-like synoviocyte (FLS) identity of primary cells isolated from mouse patellar explants.
Cells were labelled with BioTracker NIR680 (lipophilic membrane dye) and Hoechst 33342 (nuclear stain). The top row shows cells treated with rabbit anti-cadherin-11 (CDH11) primary antibody, while the bottom row presents the no-primary control to verify specificity. Isolated cells exhibit positive CDH11 immunostaining, an established FLS marker. Scale bar = 200 μm.
Appendix 1—figure 2
BioTracker fluorescence indicates cell number following staurosporine-induced apoptosis, despite an increase after mock transfection.
(a) Representative microplate images of FLS stained with BioTracker under three conditions: no transfection control, mock transfection, and mock transfection with 1 μM staurosporine. Scale bar = 5 mm. (b) Quantification of BioTracker fluorescence (n=3). BioTracker signal increases after mock transfection but decreases following staurosporine treatment. (c) Representative images of the same FLS in panel (a) stained with calcein AM, a viability dye. Scale bar = 5 mm. (d) Quantification of calcein AM fluorescence (n=3). Calcein AM fluorescence does not increase after mock transfection but decreases after staurosporine treatment. For panels (b,d): Statistical significance was determined by two-way repeated-measures ANOVA with Dunnett’s multiple comparisons test comparing groups to the no treatment control. **p<0.01 and ***p<0.001. Data are presented as mean ± standard error of the mean.
-
Appendix 1—figure 2—source data 1
Numerical data used to generate the plots in Appendix 1—figure 2.
- https://cdn.elifesciences.org/articles/105978/elife-105978-app1-fig2-data1-v1.xlsx
Appendix 1—figure 3
Linear unmixing corrects spectral overlap of EGFP and mScarlet3 fluorescent signals imaged using the Odyssey M laser scanner.
(a) Schematic of the self-amplifying RNA constructs used for linear unmixing, each expressing a different fluorescent protein: moxBFP, EGFP, and mScarlet3. Constructs were transfected into tSA201 cells. (b) Representative microplate images showing fluorescence in the 488 and 520 channels, 1 day after transfection. moxBFP-transfected cells were not detected in either channel. EGFP was primarily detected in the 488 channel, with 11.32% bleed-through into the 520 channel. mScarlet3 was primarily detected in the 520 channel, with 0.94% bleed-through into the 488 channel. Background signal was determined from mock-transfected cells. (c) Corrected images using linear unmixing to counteract spectral bleed-through from mScarlet3 and EGFP signals. Scale bars = 5 mm.
-
Appendix 1—figure 3—source data 1
Whole-plasmid sequencing results for self-amplifying RNA (saRNA) constructs depicted in Appendix 1—figure 3a.
- https://cdn.elifesciences.org/articles/105978/elife-105978-app1-fig3-data1-v1.zip
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Recombinant DNA reagent | TagGFP2 Simplicon Plasmid (E3L) | Merck-Millipore | SCR725 | |
| Recombinant DNA reagent | pIRES2-EGFP | Clontech | 6029-1 | |
| Recombinant DNA reagent | pDx_mScarlet3 | Addgene | 189754 (RRID:Addgene_189754) | A gift from Dorus Gadella |
| Gene (Engineered fluorescent protein) | moxBFP | FPbase | SSTDU | Amino acid sequence |
| Gene (Engineered fluorescent protein) | mEGFP | FPbase | QKFJN | Amino acid sequence |
| Gene (Orthopoxvirus vaccinia—Strain: Western Reserve) | Vaccinia Virus E3 protein | UniProt | P21605-1 [1991-05-01 v1] | Amino acid sequence |
| Gene (Phlebovirus toscanaense) | Toscana virus NSs protein | UniProt | P21699 [1991-05-01 v1] | Amino acid sequence |
| Gene (Cardiovirus theileri—Strain: DA) | Theiler’s virus L* protein | UniProt | P0DJX4 [2013-07-24 v1] | Amino acid sequence |
| Gene (Alphapermutotetravirus thoseae) | T2A | UniProt | Q9YK87 (139–156) [1999-05-01 v1] | Amino acid sequence |
| Gene (Teschovirus asilesi) | P2A | UniProt | A0A077CZU0 (977–995) [2014-10-29 v1] | Amino acid sequence |
| Gene (Aphthovirus burrowsi) | E2A | UniProt | K9MZ26 (991–1010) [2013-03-06 v1] | Amino acid sequence |
| Gene (Mus musculus) | IκBα | UniProt | Q9Z1E3 [2004-10-25 v2] | Amino acid sequence |
| Gene (Engineered mouse protein) | srIκBα | This paper | Substituted serines 32 and 36 with alanines | |
| Gene (Mus musculus) | Smad7 | UniProt | O35253-1 [1998-01-01 v1] | Amino acid sequence |
| Gene (Mus musculus) | SOCS1 | UniProt | O35716 [1998-01-01 v1] | |
| Software, algorithm | GenSmart | GenScript | RRID:SCR_026296 | Used to codon optimize amino acid sequences for mouse expression |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-EGFP-IRES-moxBFP | This paper | RRID:Addgene_242407 | Deposited with Addgene: 242407 |
| Sequence-based reagent | Conventional saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-EGFP-IRES-moxBFP | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-EGFP-IRES-E3 | This paper | RRID:Addgene_242408 | Deposited with Addgene: 242408 |
| Sequence-based reagent | E3 saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-EGFP-IRES-E3 | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-EGFP-IRES-E3-NSs-L* | This paper | RRID:Addgene_242409 | Deposited with Addgene: 242409 |
| Sequence-based reagent | E3-NSs-L* saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-EGFP-IRES-E3-NSs-L* | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES-moxBFP | This paper | RRID:Addgene_242410 | Deposited with Addgene: 242410 |
| Sequence-based reagent | moxBFP saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES-moxBFP | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES- srIκBα | This paper | RRID:Addgene_242411 | Deposited with Addgene: 242411 |
| Sequence-based reagent | srIκBα saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES- srIκBα | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES- srIκBα-Smad7-SOCS1 | This paper | RRID:Addgene_242412 | Deposited with Addgene: 242412 |
| Sequence-based reagent | srIκBα-Smad7-SOCS1 saRNA construct | This paper | saRNA transcribed from VEE-mScarlet3-IRES-E3-NSs-L*-EGFP-IRES- srIκBα-Smad7-SOCS1 | |
| Recombinant DNA reagent | VEE-mScarlet3-IRES-PuroR | This paper | RRID:Addgene_242415 | Deposited with Addgene: 242415 |
| Recombinant DNA reagent | VEE-EGFP-IRES-PuroR | This paper | RRID:Addgene_242414 | Deposited with Addgene: 242414 |
| Recombinant DNA reagent | VEE-moxBFP-IRES-PuroR | This paper | RRID:Addgene_242413 | Deposited with Addgene: 242413 |
| Commercial assay, kit | Phusion High-Fidelity DNA Polymerase | Thermo Scientific | F530S | |
| Commercial assay, kit | NEBuilder HiFi DNA Assembly Cloning Kit | NEB | E5520S | |
| Strain, strain background (Escherichia coli) | 5-alpha Competent E. coli (High Efficiency) | NEB | C2987H | |
| Commercial assay, kit | T7 RiboMAX Large Scale RNA Production System | Promega | P1300 | |
| Commercial assay, kit | PureLink HiPure Plasmid FP Maxiprep kit | Invitrogen | K210027 | |
| Commercial assay, kit | Vaccinia Capping System | NEB | M2080S | |
| Commercial assay, kit | mRNA Cap 2´-O-Methyltransferase | NEB | M0366S | |
| Commercial assay, kit | Antarctic Phosphatase | NEB | M0289S | |
| Other | NorthernMax-Gly Sample Loading Dye | Invitrogen | AM8551 | |
| Other | Lipofectamine MessengerMAX Transfection Reagent | Invitrogen | LMRNA001 | |
| Strain, strain background (M. musculus) | C57BL/6J | Envigo | ||
| Cell line (Homo sapiens) | tSA201 | ECACC | 96121229 (RRID:CVCL_2737) | |
| Other | BioTracker NIR680 | Merck | SCT112 | |
| Chemical compound, drug | ML336 | Cayman Chemical | 9001920 | |
| Chemical compound, drug | Puromycin | Thermo Scientific | J67236.XF | |
| Other | 6-well black glass bottom plate | CellVis | P06-1.5H-N | |
| Other | 24-well black glass bottom plate | Grenier Bio-One | 662892 | |
| Other | Live Cell Imaging Solution | Invitrogen | A59688DJ | |
| Other | Poly-D-lysine–coated glass-bottom 35 mm dishes | MatTek | P35GC-1.5-14-C | |
| Software, algorithm | Linear unmixing script for Odyssey M images | This paper; Ferguson, 2025 | Available at https://github.com/lariferg/spectral_unmixing | |
| Software, algorithm | ImageJ | NIH | RRID:SCR_003070 | |
| Other | Annexin V-CF800 | Biotium | 29078 | |
| Antibody | Anti-cadherin-11 rabbit polyclonal | Affinity Biosciences | DF3523 (RRID:AB_2835743) | |
| Antibody | Anti-Phospho-eIF2α (Ser51) rabbit monoclonal | Cell Signaling Technology | 3398 (RRID:AB_2096481) | |
| Antibody | Anti-eIF2α mouse monoclonal | Cell Signaling Technology | 2103 (RRID:AB_836874) | |
| Antibody | Anti-PKR rabbit polyclonal | Proteintech | 18244-1-AP (RRID:AB_2246451) | |
| Antibody | Anti-Phospho-eIF4E (S209) rabbit monoclonal | Abcam | ab76256 (RRID:AB_1523534) | |
| Antibody | Anti-eIF4E mouse monoclonal | Invitrogen | MA1-089 (RRID:AB_2536738) | |
| Antibody | Anti-fibroblast activation protein mouse monoclonal | InVivoMAb | BE0374 (RRID:AB_2927511) | |
| Antibody | Anti-puromycin mouse monoclonal | Absolute Antibody | 3RH11 (RRID:AB_2620162) | |
| Antibody | Goat anti-mouse IRDye 800CW secondary | LI-COR | 926-32210 (RRID:AB_621842) | |
| Antibody | Donkey anti-rabbit IRDye 800CW secondary | LI-COR | 926-32213 (RRID:AB_621848) | |
| Antibody | Donkey anti-mouse IRDye 680RD secondary | LI-COR | 926-68072 (RRID:AB_10953628) | |
| Other | CellTag 700 | LI-COR | 926-41090 | |
| Commercial assay, kit | DyLight Antibody Labeling Kit | Thermo Scientific | 53062 | |
| Software, algorithm | Empiria Studio | LI-COR | RRID:SCR_014281 | |
| Commercial assay, kit | LEGENDplex Mouse Anti-Virus Response Panel | BioLegend | 740621 | |
| Software, algorithm | Qognit | BioLegend | ||
| Other | 3D printed vacuum manifold for LEGENDplex filter plates | This paper | Available at the NIH 3D Print Exchange under accession number 3DPX-021388 | |
| Commercial assay, kit | RNeasy Mini Kit | QIAGEN | 74104 | |
| Commercial assay, kit | TaqMan Fast Advanced Cells-to-CT Kit | Invitrogen | A35377 | |
| Commercial assay, kit | TaqMan array plates | Invitrogen | 4413261 | |
| Software, algorithm | StepOne Software | Life Technologies | RRID:SCR_014281 | |
| Software, algorithm | Prism | GraphPad | RRID:SCR_002798 |
Additional files
-
MDAR checklist
- https://cdn.elifesciences.org/articles/105978/elife-105978-mdarchecklist1-v1.docx
-
Appendix 1—figure 2—source data 1
Numerical data used to generate the plots in Appendix 1—figure 2.
- https://cdn.elifesciences.org/articles/105978/elife-105978-app1-fig2-data1-v1.xlsx
-
Appendix 1—figure 3—source data 1
Whole-plasmid sequencing results for self-amplifying RNA (saRNA) constructs depicted in Appendix 1—figure 3a.
- https://cdn.elifesciences.org/articles/105978/elife-105978-app1-fig3-data1-v1.zip
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Cap-independent co-expression of dsRNA-sensing and NF-κB pathway inhibitors enables controllable self-amplifying RNA expression with reduced immunotoxicity
eLife 14:RP105978.
https://doi.org/10.7554/eLife.105978.3
