Figure 1.
Mutation in the HoxA9 eIF4E-binding site reduces eIF4E interaction and activity without modulating the DNA-binding activity of HoxA9. (A) Schematic depiction of the HoxA9 protein with consensus eIF4E-binding site (black bar) that is conserved from humans to sharks. Mutation of eIF4E-binding site was generated by alanine substitutions of the conserved Y and L and denoted HoxA9AA. HoxA9AA was generated using QuikChange site-directed mutagenesis (Agilent) and validated by sequencing. Also depicted is the W73A eIF4E mutation (red) that disrupts the eIF4E-HoxA9 interaction.9 Structure of cap-free eIF4E (pdb 2GPQ)16 was generated in Pymol. Both HoxA9AA and wildtype can bind DNA (1D). (B) Bacterially produced and purified GST- HoxA9wt, but not HoxA9AA or the GST control (GST-GFP), binds to endogenous eIF4E from K562 cell lysates. Right, quantification of 2 biological replicates. Equal loading of GST proteins is provided in supplemental Figure 1A. (C) Purified GST-eIF4E and GST were used in pulldowns from K562 cells stably expressing HoxA9AA and HoxA9wt from bicistronic vectors with GFP, with empty vector as control. Inputs are also shown. Right panel, quantification of 3 biological replicates. (D) Electrophoretic mobility assay showed HoxA9AA retains DNA-binding activity compared to HoxA9wt. DNA oligonucleotides containing a HoxA9 consensus binding motif described in Calvo et al.17 Super-shift reactions were performed with the anti-FLAG M2 antibody. “Comp” refers to cold competition reactions completed by using a 100-fold molar excess of nonradioactive labeled dsDNA probe. (E). HoxA9wt, but not HoxA9AA, promotes eIF4E-dependent mRNA export. Left panel: relative cyt/nuc ratio of MYC and MCL1 mRNAs in HoxA9 overexpressing cell lines or in GFP vector controls. Fractionation controls are shown in supplemental Figure 1C. Right panel: total RNAs levels do not change levels of target mRNAs or the housekeeping transcript glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Three biological replicates are shown. (F) Immunoblot and quantitation of 3 biological replicates to assess protein expression of eIF4E-dependent RNA export targets in HoxA9wt/HoxA9AA stably overexpressing K562 cells relative to vector controls. Antibodies used for immunoblotting: mouse monoclonal anti-eIF4E (BD PharMingen), mouse monoclonal anti–β-actin (A5441, Sigma Aldrich), rabbit polyclonal anti-Mcl-1 (S19, sc819, Santa Cruz), rabbit polyclonal anti-Myc (ab32072, Abcam), mouse monoclonal anti-Hsp90 (sc-69703, Santa Cruz), rabbit polyclonal anti-GAPDH (sc-25778, Santa Cruz), rabbit monoclonal eIF4E binding-protein 1 (53H11, Cell Signaling), and rabbit polyclonal anti-HoxA9 (ab140631, Abcam). For panels B-C,E-F, means are shown; error bars indicate standard deviations, P values (student t test). Cyt, cytoplasmic; Exp, exposure; HD, homeodomain; nuc, nuclear; nonspecific band is noted on the EMSA (panel D).

Mutation in the HoxA9 eIF4E-binding site reduces eIF4E interaction and activity without modulating the DNA-binding activity of HoxA9. (A) Schematic depiction of the HoxA9 protein with consensus eIF4E-binding site (black bar) that is conserved from humans to sharks. Mutation of eIF4E-binding site was generated by alanine substitutions of the conserved Y and L and denoted HoxA9AA. HoxA9AA was generated using QuikChange site-directed mutagenesis (Agilent) and validated by sequencing. Also depicted is the W73A eIF4E mutation (red) that disrupts the eIF4E-HoxA9 interaction.9 Structure of cap-free eIF4E (pdb 2GPQ)16 was generated in Pymol. Both HoxA9AA and wildtype can bind DNA (1D). (B) Bacterially produced and purified GST- HoxA9wt, but not HoxA9AA or the GST control (GST-GFP), binds to endogenous eIF4E from K562 cell lysates. Right, quantification of 2 biological replicates. Equal loading of GST proteins is provided in supplemental Figure 1A. (C) Purified GST-eIF4E and GST were used in pulldowns from K562 cells stably expressing HoxA9AA and HoxA9wt from bicistronic vectors with GFP, with empty vector as control. Inputs are also shown. Right panel, quantification of 3 biological replicates. (D) Electrophoretic mobility assay showed HoxA9AA retains DNA-binding activity compared to HoxA9wt. DNA oligonucleotides containing a HoxA9 consensus binding motif described in Calvo et al.17 Super-shift reactions were performed with the anti-FLAG M2 antibody. “Comp” refers to cold competition reactions completed by using a 100-fold molar excess of nonradioactive labeled dsDNA probe. (E). HoxA9wt, but not HoxA9AA, promotes eIF4E-dependent mRNA export. Left panel: relative cyt/nuc ratio of MYC and MCL1 mRNAs in HoxA9 overexpressing cell lines or in GFP vector controls. Fractionation controls are shown in supplemental Figure 1C. Right panel: total RNAs levels do not change levels of target mRNAs or the housekeeping transcript glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Three biological replicates are shown. (F) Immunoblot and quantitation of 3 biological replicates to assess protein expression of eIF4E-dependent RNA export targets in HoxA9wt/HoxA9AA stably overexpressing K562 cells relative to vector controls. Antibodies used for immunoblotting: mouse monoclonal anti-eIF4E (BD PharMingen), mouse monoclonal anti–β-actin (A5441, Sigma Aldrich), rabbit polyclonal anti-Mcl-1 (S19, sc819, Santa Cruz), rabbit polyclonal anti-Myc (ab32072, Abcam), mouse monoclonal anti-Hsp90 (sc-69703, Santa Cruz), rabbit polyclonal anti-GAPDH (sc-25778, Santa Cruz), rabbit monoclonal eIF4E binding-protein 1 (53H11, Cell Signaling), and rabbit polyclonal anti-HoxA9 (ab140631, Abcam). For panels B-C,E-F, means are shown; error bars indicate standard deviations, P values (student t test). Cyt, cytoplasmic; Exp, exposure; HD, homeodomain; nuc, nuclear; nonspecific band is noted on the EMSA (panel D).

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal