Figure 1.
A toolkit of transgenic Sindbis replicons.
A. Schematic of Sindbis genome, a bicistronic single-stranded RNA with positive polarity: the 5′ end contains a ‘packaging signal’ (PS) for incorporation into the particle. An ‘internal Promoter’ (iP) can be found (on the ‘antigenome’; see Figure S1A) in between the viral ORF's. Abbreviations: nsp = ‘non-structural proteins’; sp = ‘structural proteins’. B. Four transgenic fly strains containing different Sindbis replicons (SinR) stably inserted into the genome. Each transgenic replicon is harboring different reporter genes, or mutations. Abbreviations: UAS = GAL4 ‘GAL4 Upstream activating sequence’; TATA = hsp70 TATA box; RBZ: Hepatitis Delta Ribozyme; GFP = membrane tagged mCD8:eGFP fusion protein; TOM = myristoylated Tomato; Luc = firefly luciferase; nsp[GVD] = point-mutated RNA-dependent RNA Polymerase. C. Four replication-incompetent replicons (SinR), all lacking ORF1 due to deletions in the Sindbis genomic DNA sequence, and harboring different sequences in ORF2. Note that DH-EB harbors a smaller deletion, thus retaining a ‘packaging signal’.
Figure 2.
Qualitative and quantitative reporters of in vivo viral replication.
A. Adult, anesthetized flies expressing SinR-GFP using eye-specific driver GMR-GAL4. Virtually no GFP expression is detectable. B. Quantification of viral replication in vivo, using Luciferase-expressing, replication-competent replicon SinR-Luc. Mutations in imd resulted in significantly higher activity. In contrast, knock-down of Akt and Pi3K using UAS-RNAi constructs resulted in a significant decrease. (Luminometer counts in relative units per fly, per µL of homogenate). C. Strong GFP expression can be observed in the entire eye, when RNAi is inhibited using homozygous Dcr2 mutants. D. Quantification of RNAi effects on viral replication in vivo, using SinR-Luc in combination with different ways of inhibiting RNAi (UAS-B2 co-expression, homozygous Drc2 mutants). Inhibition of RNAi greatly increased SinR-Luc activity. Note that UAS-Luciferase control levels are unaffected by suppression of RNAi. E. SinR-GFP[GVD] with a point mutated RNA-dependent RNA Polymerase never results in GFP expression as detected by in vivo fluorescence. F. Rescue of GFP expression from SinR-GFP[GVD] in trans, using GMR-GAL4, through co-expression of non-fluorescent replicon SinR-Luc, providing an active replicase.
Figure 3.
Quantification of Sindbis replicon expression in different tissues.
A,C. Examples of SinR-GFP expression in different tissues. Labeled are adult neurons (NSyb-GAL4; A) or adult fat body (r4-GAL4; C). For both tissues, RNAi was inhibited using UAS-B2. B. Quantification of SinR-Luc Luciferase activity in neurons. Significantly higher levels of Luciferase activity were obtained in homozygous Dcr2 mutants, comparable to those obtained with UAS-Luciferase controls. Inactivation of RNAi using UAS-B2 had weaker yet comparable effects while Drc2 heterozygotes show little to no effect. D. Similar effects were obtained in other tissues, like the adult fat body). All luminometer counts in relative units per fly, per uL of homogenate.
Figure 4.
Trans-activation of defective helper replicons.
A. Expression of replication-defective replicon DH-TOM in the adult eye, using GMR-GAL4 in Dcr2 homozygotes. Low levels were visible as ‘pseudopupil’, in the center of the eye, most likely due low-level ribosomal read-through (despite numerous nonsense ATG's). B. Strong expression of DH-TOM activated in trans from a 2nd replicon (SinR-GFP), contributing an intact RdRP. However, red fluorescence is sparse, and co-expression of GFP and Tomato is rare. C. The point-mutated replicon SinR-GFP[GVD] always failed to trans-activate defective replicon DH-TOM. D. Third instar larval eye discs dissected from flies co-expressing UAS-B2, SinR-GFP, and DH-TOM reveal a low level of myr:Tomato trans-activation (C′). E. Luciferase activity (in relative units per fly, per µL of homogenate) of defective DH-Luc in different genetic backgrounds. Significant levels of trans-activation by an intact replicon (SinR-GFP) were observed when RNAi was inactivated (+UAS-B2, or Dcr2 homozygotes). However, the absolute levels were very low when compared to SinR-Luc activity in the same backgrounds. F. Co-expression of replicons with deleted replicase ORF (DH-Tom), or a point-mutation (SinR-GFP[GVD]) never trans-activated Luciferase activity of DH-Luc.
Figure 5.
Stochastic exclusion between transgenic replicons.
A. Expression of red fluorescent, replication-competent replicon SinR-Tom was undetectable when driven in the adult eye using GMR-GAL4, with the RNAi pathway intact. B. Strong levels of Tomato expression from SinR-TOM observed in homozygous Dcr2 mutants. C. Strong red and green fluorescence in flies co-expressing both SinR-GFP and SinR-TOM in Drc2 mutants. Note that not all ommatidia co-expressed the two replicons. D. Third instar larval eye discs dissected from flies co-expressing SinR-GFP, SinR-TOM, and UAS-B2 under GMR-GAL4 control. Strong co-expression was observed only in ∼10% of ommatidia. Apparently, the two replicons exclude each other's expression. E. Third instar larval eye discs co-expressing replication-deficient SinR-GFP[GVD] and SinR-TOM revealing a similarly low degree of GFP trans-activation. In these cases, the two proteins always co-localized, as expected. F. Quantification of in vivo replicon exclusion using SinR-Luc. In homozygous Dcr2 mutants, SinR-Luc activity was not affected by co-expression of either UAS-mCD8GFP, or UAS-myr:Tomato. However, co-expression of different insertions of SinR-GFP (suffix 1 and 2) or SinR-Tom all reduced Luciferase levels by approximately half (−46.7%). Moreover, expression of all three replicons in Dcr2 mutants lowered the Luciferase counts by approximately two thirds (−66.8%). Interestingly, co-expression of replication-deficient SinR[GVD]-GFP, also reduced Luciferase activity (−30%).
Figure 6.
ORF2 does not induce exclusion and Model.
A. Quantitative test whether defective helper transgenes excluded replicon expression, using SinR-Luc. In homozygous Dcr2 mutants, expression levels of SinR-Luc were unaffected by co-expression of either DH-TOM, DH-BB or DH-EB defective replicons, all harboring deletions of ORF1. Co-expression of foreign glycoproteins (UAS-G[VSV]) also had no significant effect on Luciferase expression. B. Model summarizing factors regulating in vivo replicon expression. While replicon expression is inhibited by cellular pathways (RNAi, NMD, innate immunity), a strong preference of the viral RdRP for the internal promoter on the ‘subgenomic RNA’ originating from the same transcript exists. As a result, trans-activation is weak, even from transgenes with a deleted RdRP.