Supplementary Figure 1

Transcription

1 Supp. Fig. NL43R Metabolic labelling Cytoplasmic grn (% of control) grn translation (% of control) Methionine incorporation (% of control) C pcirenilla D Renilla luciferase mrn translation (% of control) % of control,2,8,6,4,2 HIVHIV SL % of control globin HIV HIV SL sictrl siddx3 Supplementary Figure () Control (sictrl) or DDX3depleted (siddx3) HeLa cells were transfected with.3 µg of the HIV pnl43r reporter proviral DN. Cytoplasmic HIV grn (right graph) and grn translation (Gag Renilla luciferase activity normalized by cytoplasmic mrn, left graph) were determined at 24 hpt. Results were normalized to sictrl (arbitrary set to %) and are presented as mean +/ SD of three duplicated experiments. () Control (sictrl) or DDX3depleted (siddx3) HeLa cells were transfected with.3 µg of the pci Renilla vector and translation (Renilla luciferase activity normalized by cytoplasmic mrn) was determined at 24 hpt. Results were normalized to sictrl (arbitrary set to %) and are presented as mean +/ SD of three duplicated experiments. (C) [ 35 S]methionine incorporation in control (sictrl) or DDX3depleted (siddx3) HeLa cells following a 3 min pulse. Results were normalized to sictrl (arbitrary set to %) and are presented as mean +/ SD of three duplicated experiments. (D) Control (sictrl) or DDX3depleted (siddx3) HeLa cells were transfected with.25 pmol of Renilla luciferase mrns carrying the indicated 5 UTR. Renilla luciferase activity was analyzed 3 hpt. Results are normalized to HIV (arbitrary set to, left graph) or to sictrl (arbitrary set to, right graph) and presented as the mean +/ SD of three duplicated independent experiments.

2 Supp. Fig. 2 HIV 5 UTR TR PS Poly DIS SD Ψ UG m 7 G deletion (ΔTR) 3 deletion (TRPS) TR ΔG= 29,6 kcal/mol Stem loop (SL) ΔG= 8,6 kcal/mol Supplementary Figure 2: () Schematic representation of the HIV 5 UTR indicating the regions deleted to generate the Renilla luciferase mrns used in Fig. 2C. The 5 deletion correspond to TR (position 57) while the 3 deletion correspond to DIS, SD and ψ (positions ) and thus, keeps the region encompassing TR to PS (positions 243). () mfold predicted secondary structure of HIV TR and the artificial stem loop (SL) inserted in the 5 end of the C 6 construct. Structure stability (indicated as ΔG) is indicated at the bottom.

3 Supp. Fig. 3 Toeprint # Toeprint #2 C6 Competitor C Poly I:C Competitor 2 rddx3/5ʼutr complexes rddx3/5ʼutr complexes Free 5ʼUTR Free 5ʼUTR Supplementary Figure 3: () DDX3 binding sites were mapped by toeprint in Fig. 3C and extrapolated to the HIV 5 UTR structure previously determined by several groups3335. oth toeprints falls close to regions predicted to be single stranded. ()(C).5% agarose gel showing results from binding competitions assays..44 pmol ( ng) of in vitro transcribed, [32P]labelled HIV 5 UTR was incubated with buffer (lanes ), with rddx3 (lanes 2) or with rddx3 in the presence of, 25 or 5 ng of cold ssrn (6 repeats of C, lanes 35 in ) or 25, 5 or 25 ng of dsrn (poly I:C; lanes 26 in C). The free 5 UTR and the 5 UTR/DDX3 complexes are indicated on the left. Results showed are representative of at least three independent experiments.

4 Supp. Fig. 4 GG U G C TR wt m 7 G Renilla () C29 GG U G m 7 G TRUG Renilla () GG U G CUGTR Renilla () Supplementary Figure 4: Schematic representation of the reporter Renilla luciferase mrns used in the indirect ex vivo TR unwinding assay presented in Fig. 4. The TR wt mrn contains the HIV TR structure immediately appended to the Renilla ORF without any initiation codon. In the TRUG mrn, a single nucleotide mutation at position 29 (C29, indicated in blue) located in the loop generates an in frame UG codon that could only be recognized upon TR unwinding. s a control, the 5 proximal 28 nucleotides of the TRUG mrn were replaced by 9 C repeats to generate the CUGTR.

5 Supp. Fig. 5 68/ PP eif4e eif4 eif3 eif4 eif4gi (68) eif4gi (68286) PP I G D I G D I G D Supplementary Figure 5: Schematic representation not to scale of human eif4gi and the binding sites for translation initiation factors (upper cartoon). GSTPull down assays using recombinant GST or GSTDDX3 and the indicated [ 35 S]labelled in vitro translated fragment of eif4gi (PP was used as a positive control). I=input (/); G=Pull down with GST; D=Pull down with GSTDDX3.

6 Supp. Fig. 6 Viral 5ʼUTR Length (nt) %GC DDX3 Cellular 5ʼUTR Length (nt) %GC DDX3 HIV yes HIV yes SIV yes FIV no HTLVI no MLV no βglobin 5 44 no CL yes GPDH 2 6 yes NDUF 4 65 yes IL yes IL yes rtificial 5ʼUTR Length (nt) %GC DDX3 Hsp yes PP no C (6) no Line yes 2 Renilla luciferase activity (RLU) Supplementary Figure 6: () Table summarizing the structural and functional properties of the mrns used in this study. In the DDX3 column, yes indicates DDX3dependent translation and no indicates DDX3independent translation. () Hela cells were transfected with.25 pmol of Renilla luciferase mrns carrying the indicated 5 UTR in order to compare translation efficiency. Renilla luciferase activity was analyzed 3 hpt. Results are presented as the mean +/ SD of the Relative Luminiscence Units (RLU) obtained in three duplicated independent experiments.