Figure 1.
TonB cells and isoform-specific anti-STAT5 shRNAs.
(A) TonB cells were cultured in IL-3 (IL-3), IL-3 plus 1 µM imatinib mesylate (IL-3+IM) IL-3+ Ruxolitinib, without IL-3 (Starvation), with IL-3 plus doxycycline (IL-3+BCR-ABL), with IL-3 and doxycycline and imatinib mesylate (IL-3+BCR-ABL+IM), with doxycycline (BCR-ABL) and with doxycycline plus 1 µM imatinib mesylate (BCR-ABL+IM) for 72 hours before viable cells were counted by Trypan-blue dye exclusion. Results are shown as a percentage of control (TonB cells in the presence of IL-3). (B) TonB cells were lentivirally transduced with shRNAs targeting control (shGL2), STAT5A/B (shS5A/B, targeting both isoforms), murine STAT5A (sh-muS5A), murine STAT5B (sh-muS5B), and a mixture of both. Four days after transduction protein expression of both STAT5 isoforms and Tubulin was analyzed by western blotting. The numbers indicate changes in protein expression in % as determined by densitometry.
Figure 2.
Isoform-specific STAT5 loss-of-function phenotypes in TonB cells.
(A) TonB cells were lentivirally transduced with shRNAs targeting control (shGL2), STAT5A/B (shS5A/B), murine STAT5A (sh-muS5A), murine STAT5B (sh-muS5B), and a mixture of both and cultured in the presence of IL-3. Four days after transduction cells were either cultured with IL-3 or switched to BCR-ABL, and the number of viable cells was determined after additional four days by Trypan-blue dye exclusion. The proliferation of non-transduced cells under IL-3 and BCR-ABL was set 100%, respectively. The data represent mean of ten independent experiments. (B) TonB cells transduced with isoform-specific anti-STAT5 shRNAs were cultured four days in the presence of IL-3 or BCR-ABL. Protein expression of BCL-XL and BCL-2 was analysed by western blotting. Expression of ERK2 served as loading control. The numbers indicate changes in protein expression in % as determined by densitometry. (C) TonB cells lentivirally transduced with isoform specific STAT5 shRNAs were cultured in the presence of IL-3 or doxycycline as described in (B). mRNA expression was quantified by qRT-PCR. BCL-XL expression was normalized to HPRT mRNA-expression, and GL2-ctrl was set as 100%. The data represent mean ± SE of four independent experiments.
Figure 3.
Over-expression of STAT5 isoforms in TonB cells.
(A) Expression of STAT5 isoforms was analysed by western blotting of TonB cells transduced with empty vector or murine cDNAs encoding STAT5A or STAT5B four days before. (B) TonB cells transduced with empty vector or STAT5 isoforms as shown in (A) were starved by removal of IL-3, and the number of viable cells was determined by Trypan-blue dye exclusion over time. The data represent mean of five independent experiments. (C) TonB cells over-expressing STAT5B as shown in (A) and (B) were cultured in the presence (+IM) or absence (−IM) of 1.5 µM imatinib mesylate and the number of viable cells was plotted over time. The data represent mean of 3 independent experiments. Due to the scale standard deviation for the –IL-3/+IM condition is not visible.
Figure 4.
STAT5A and STAT5B phosphorylation by IL-3 receptor and BCR-ABL signalling.
(A) TonB cells cultured with IL-3 or BCR-ABL-expressing cells were treated with different concentrations of Ruxolitinib (0.5 µM –25 µM) or Imatinib (0.15 µM –5 µM). Cells were incubated for 48 hours and proliferation was measured using MTS. Data represent mean ± SD of three independent experiments. (B) TonB cells were cultured in the presence of IL-3, IL-3+ doxycycline to induce BCR-ABL-expression or doxycycline alone. Aliquots were treated with Ruxolitinib (5 µM), starved by withdrawal of IL-3 (Starvation), AG490 (50 µM) or 1 µM imatinib mesylate for 24 hours. Cells were lysed and STAT5A (upper panel) and STAT5B (lower panel) were immunoprecipitated with isoform-specific anti-STAT5 antibodies. Tyrosine phosphorylation and the amount of STAT5 (loading control) were analyzed by western blotting. (C) TonB cells were cultured either in the presence of IL-3 (+IL-3) or BCR-ABL (+Dox). After starvation (20 hours in the absence of IL-3) cells were replated in the presence or absence of IL-3 and the JAK2 kinase inhibitor AG490 at final concentrations of 50 µM. BCR-ABL-expressing cells were split and supplied with either imatinib at final concentrations of 1.5 µM and/or AG490 at final concentrations of 50 µM. Twenty hours later equal numbers of viable cells were directly lysed in SDS-boiling buffer and subjected to SDS-PAGE. The phosphorylation levels of JAK2, GAB2, and STAT5 were analyzed by western blotting using antibodies recognizing the following phospho-epitops: Tyr1007/1008 (JAK2); Tyr452 (GAB2); and Y694 (STAT5). Expression of Actin served as loading control. One out of three representative experiments is shown.
Figure 5.
Heterodi-(oligo)merization of STAT5A and STAT5B in TonB cells.
TonB cells were cultured in the presence of IL-3, BCR-ABL, and IL-3 plus BCR-ABL. Aliquots were starved overnight by withdrawal of IL-3 (Starvation) and re-stimulated with IL-3 for 30 minutes (Restimulation) or were treated with 1 µM imatinib mesylate (BCR-ABL+IM). Cells were lysed and STAT5A (left) and STAT5B (right) were immunoprecipitated with isoform-specific anti-STAT5 antibodies. Overall tyrosine phosphorylation of precipitated proteins (upper panel), co-precipitation of the other STAT5 isoform (middle panel), and the amount of precipitated STAT5 as a loading control (lower panel) were analyzed by western blotting.
Figure 6.
Intracellular localization of STAT5 isoforms.
(A) TonB cells cultured in the presence of IL-3 or doxycycline were spun down on glass slides, fixed by ice-cold methanol and stained sequentially with anti-STAT5B, anti-rabbit-FITC and DAPI. Intracellular localization was analyzed using confocal microscopy. Nuclear localization was confirmed by a DNA-specific DAPI counterstain. The lower panel shows antibody (with and without enhancement) and DAPI control. (B) TonB cells expressing STAT5A-eGFP were cultured in the presence of IL-3, BCR-ABL or BCR-ABL plus imatinib mesylate (BCR-ABL+IM) and live-cell analysis was performed by confocal microscopy (upper panel). The middle panel shows STAT5A-eGFP in BCR-ABL-expressing TonB cells after removal of imatinib mesylate (−IM) for the indicated periods of time. The lower panel shows empty vector control (eGFP). (C) TonB-STAT5A-eGFP cells treated as in Figure 6B, middle panel, were fixed 45 minutes after imatinib-removal, stained with anti-CD131-PE antibody (IL-3/IL-5/GM-CSF-receptor common β-chain) and DAPI before analysis by confocal microscopy.
Figure 7.
Mass spectrometry-based analysis of aberrant STAT5 tyrosine phosphorylation.
For each part of the figure there are 2 panels, the left hand panel the parent ion transition is 477.77Th (unphosphorylated STAT5A) and the right hand panel the parent ion transition is 517.75Th (phosphorylated STAT5A). (A) STAT5 was isoform-specific immunoprecipitated from TonB cells in the presence of IL-3 or BCR-ABL. A peptide (YYTPVLAK) in STAT5A was initially found to be phosphorylated on Y682 in the presence of BCR-ABL but not in the presence of IL-3. Ions with the appropriate parent mass/charge ratios were selected for tandem mass spectrometric analysis. Product ions 527.3552Th (identifying the peptide), 871.4325Th (indicative for pY683), 791.4662Th (indicative for pY682), and 216.04Th (indicative for phosphorylated tyrosine) are labelled. A list of all product ions is shown in Figure S5. (B) STAT5A was precipitated from K562 cells and analyzed as described above.
Figure 8.
Isoform-specific inhibition of STAT5 in human cell lines.
(A) K562 cells were lentivirally transduced with shRNAs targeting control (shGL2), human STAT5A (sh-huS5A), human STAT5B (sh-huS5B), and a mixture of both. Four days after transduction protein expression of both STAT5 isoforms and ERK2 was analyzed by western blotting. The numbers indicate changes in protein expression in % as determined by densitometry. (B) The human BCR-ABL-positive cell lines EM-2, K562 and LAMA-84 were lentivirally transduced with control (shGL2), isoform-specific (sh-huS5A, sh-huS5B), and a mixture of both isoform-specific shRNAs. Four days after transduction, equal numbers of cells were plated and the number of viable cells was determined after additional four days by Trypan-blue dye exclusion. The number of viable non-transduced cells (native) was set 100%. The data represent mean of four independent experiments. (C) Impact of isoform-specific STAT5 shRNAs on imatinib mesylate (IM) response as determined by IC50. STAT5 shRNAs were compared to GL2 controls with * p<0.05 and ** p<0.01.