1. BCR-ABL–induced adhesion defects are tyrosine kinase–independent
by Jason A. Wertheim, Kevin Forsythe, Brian J. Druker, Daniel Hammer, David Boettiger, and Warren S. Pear
Blood
Volume 99(11):
June 1, 2002
©2002 by American Society of Hematology

2. Diagram of the cell detachment device and characteristic detachment profile.(A) Cells are allowed to settle onto a fibronectin-coated matrix atop the rotating stage.
Diagram of the cell detachment device and characteristic detachment profile.(A) Cells are allowed to settle onto a fibronectin-coated matrix atop the rotating stage. Motion of the stage and matrix induces fluid flow that generates a range of detachment forces that dislodges cells. (B) The surface density of pK1 after a 10-minute spin is plotted versus radial position showing that fewer cells remain attached near the circumference where the detachment force, τ, is high compared to the center where the detachment force approaches zero. (C) Fraction of adherent pK1 cells in panel B, normalized to the cell count at the center after a 10-minute spin is plotted versus surface shear stress, τ. Cells were incubated on BSA alone (▾) or in the presence of fibronectin ( ). (–○–) represents a curve fitted to the experimental points. The dashed line represents an adherent fraction of 0.5. The critical shear stress (τ50) is defined as the point on the abscissa (shear stress) that corresponds to an adherent fraction of 0.5 along the fitted curve.
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

3. Expression of P210BCR-ABL increases adhesion to fibronectin compared to control pK1 cells.(A) Fraction of adherent cells after a 10-minute spin is plotted versus surface shear stress, τ, for P210BCR-ABL( ) or pK1 control cells (●).
Expression of P210BCR-ABL increases adhesion to fibronectin compared to control pK1 cells.(A) Fraction of adherent cells after a 10-minute spin is plotted versus surface shear stress, τ, for P210BCR-ABL( ) or pK1 control cells (●). A higher proportion of P210BCR-ABLcells remain attached at nearly every given shear stress compared to pK1 cells suggesting that P210BCR-ABL leads to increased binding to fibronectin. (–○–) represents a curve fitted to the experimental points for P210BCR-ABL or vector control pK1. The dashed line represents an adherent fraction of 0.5. The critical shear stress (τ50) is defined as the point on the abscissa (shear stress) that corresponds to an adherent fraction of 0.5 along the fitted curve. (B) τ50 was determined for spins with cells expressing P210BCR-ABL (n = 5) or pK1 control (n = 12), indicating that an average of 1.7 times as much force is needed to detach 50% of P210BCR-ABL–expressing cells from fibronectin compared to pK1 cells (P < .001). Nonspecific binding between P210BCR-ABL 32D cells to BSA matrices was minimal and similar to pK1 32D control cells (data not shown).
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

4. The kinase dead mutant retains the ability to induce P210BCR-ABL–mediated elevated adhesion.(A) The profile of a typical spin using K1176R ( ) expressing 32D cells is depicted along with pK1 control cells for comparison (●).
The kinase dead mutant retains the ability to induce P210BCR-ABL–mediated elevated adhesion.(A) The profile of a typical spin using K1176R ( ) expressing 32D cells is depicted along with pK1 control cells for comparison (●). (–●–) represents a curve fitted to the experimental points for K1176R while (–○–) represent a curve fitted to the experimental points for pK1 cells. (B) The critical shear stress was calculated for a series of spins using 2 independent populations of cells (left, right) expressing pK1 (n = 12 from Figure 2B, n = 8) or K1176R (n = 18, n = 9) and shows that adhesion is increased in the absence of kinase activity (P < .001,P = .003). Nonspecific binding between K1176R 32D cells to BSA matrices was minimal and similar to pK1 32D control cells (data not shown). (C) Measurement of α4β1- and α5β1-integrin expression in transduced cells 32D. Surface expression of α4- and α5-integrin subunits (corresponding to the α4β1 and α5β1integrins that bind fibronectin) in pK1-, P210BCR-ABL–, or K1176R-transduced cells 32D are displayed. Cells stained with the secondary phycoerythrin-conjugated antibody alone are shown as negative controls. The studies were performed in triplicate, and the MFI and SDs are shown. Representative histograms showing the level of integrin expression are presented.
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

5. Treatment of wild-type P210BCR-ABL–expressing cells with 10 μM STI571 overnight reverses the IL-3 independence and decreases phosphotyrosine levels but does not affect P210BCR-ABL expression.(A) Western blots of whole cell lysates showing expression of P210...
Treatment of wild-type P210BCR-ABL–expressing cells with 10 μM STI571 overnight reverses the IL-3 independence and decreases phosphotyrosine levels but does not affect P210BCR-ABL expression.(A) Western blots of whole cell lysates showing expression of P210BCR-ABL and phosphotyrosine expression in pK1 control, untreated P210BCR-ABL, K1176R, or P210BCR-ABLcells treated overnight with 10, 5, 1, 0.5, or 0 μM STI571. The BCR-ABL–specific band is indicated by an asterisk. Expression of grb2 was used as a loading control. (B) P210BCR-ABL 32D cells were allowed to proliferate in triplicate in media containing (▴, ▵) or free (●, ○) from WEHI as a source of IL-3 and were treated with (▴, ●) or without (▵, ○) 10 μM STI571, as indicated. Cell viability was determined by trypan blue exclusion. The 32D cells expressing P210BCR-ABL proliferate in the absence of IL-3, but die if treated with STI571. However, cells treated with STI571 remain viable if supplemented with IL-3 showing that STI571 is not toxic at the levels used.
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

6. STI571 treatment of P210BCR-ABL 32D cells fails to decrease cell adhesion.(A) P210BCR-ABL 32D cells were incubated with (●) or without ( ) 10 μM STI571 overnight prior to assaying adhesion.
STI571 treatment of P210BCR-ABL 32D cells fails to decrease cell adhesion.(A) P210BCR-ABL 32D cells were incubated with (●) or without ( ) 10 μM STI571 overnight prior to assaying adhesion. Typical profiles are shown and indicate that adhesion was not reduced, and may have slightly increased, as a result of tyrosine kinase attenuation by STI571. (–○–) represents a curve fitted to the experimental points. The critical shear stress (τ50) is defined as the point on the abscissa (shear stress) that corresponds to an adherent fraction of 0.5 along the fitted curve. (B) The critical shear stress was calculated for a series of spins and is represented as the fold change in τ50 relative to untreated P210BCR-ABLcells. We tested 2 separate populations of P210BCR-ABLcells (left, right) treated with (n = 5, n = 4), or without (n = 3, n = 4) 10 μM STI571 overnight (P = .008,P = .016). (C) pK1 control cells were incubated with (n = 4) or without (n = 10) 10 μM STI571 as in (B) (P = .06) and the data are represented as the fold change in τ50 relative to untreated pK1 cells. Treatment of transduced 32D cells with STI571 did not affect nonspecific binding to BSA matrices (data not shown).
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

7. Adhesion of human Meg-01 leukemia cells to fibronectin is unresponsive to tyrosine kinase attenuation by STI571.(A) Meg-01 cells were treated in triplicate with 10 μM STI571 (▴) or an equal volume of PBS used as the delivery vehicle (○) and were allowed to ...
Adhesion of human Meg-01 leukemia cells to fibronectin is unresponsive to tyrosine kinase attenuation by STI571.(A) Meg-01 cells were treated in triplicate with 10 μM STI571 (▴) or an equal volume of PBS used as the delivery vehicle (○) and were allowed to proliferate in Meg-01 media. Cell viability was determined by trypan blue exclusion. (B) Western blots of whole cell lysates were run to show expression of P210BCR-ABL and tyrosine-phosphorylated proteins in Meg-01 cells treated with 10 μM STI571 or PBS for 3 hours. The P210BCR-ABL–specific band is indicated by an asterisk. Expression of grb2 was used as a loading control. (C) τ50 was determined for spins with Meg-01 cells treated with 10 μM STI571 for 3.5 hours (n = 11) or an equal volume of PBS for 1.5 hours or 3.5 hours (n = 12). No difference in adhesion was observed between 1.5 hours and 3.5 hours for control Meg-01 cells treated with only PBS (data not shown). The magnitude of binding was not significantly affected when the kinase activity was arrested by STI571 (P = .81); however, decreased adhesion could be measured by blocking α5β1integrins with a soluble RGD fragment (n = 4) that prevents binding to fibronectin. Meg-01 cells treated with STI571 or PBS were assayed on BSA only (no FN), which represents nonspecific binding and is very small, at or below the minimum detection threshold of the device (≤ 5 dyne/cm2). At times, profiles could not be determined because of the small number of cells remaining after a spin in the absence of fibronectin.
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology

8. The subcellular localization of P210BCR-ABL is independent of tyrosine kinase activity.(A) NIH3T3 cells were transduced with pK1, P210BCR-ABL, or K1176R retroviral vectors and selected with puromycin to determine if F-actin colocalization is also independen...
The subcellular localization of P210BCR-ABL is independent of tyrosine kinase activity.(A) NIH3T3 cells were transduced with pK1, P210BCR-ABL, or K1176R retroviral vectors and selected with puromycin to determine if F-actin colocalization is also independent of tyrosine kinase activity. Cells were attached to fibronectin-coated coverslips overnight in the presence or absence of 10 μM STI571 as indicated. (B) 32D cells expressing P210BCR-ABL or control pK1 (data not shown) were treated with 10 μM STI571 or an equal volume of PBS overnight in the presence of WEHI media containing IL-3. Cells were allowed to bind fibronectin-coated coverslips for 15 minutes prior to fixing and staining cells according to the same protocol used to prepare 32D cells for adhesion assays. 32D and NIH3T3 cells were imaged using confocal microscopy. Colocalization of BCR-ABL with F-actin appears yellow or orange when BCR-ABL (red) and actin (green) are merged. All cells were stained with DAPI to visualize the nucleus (blue). Positive nuclear staining observed in pK1-transduced control NIH3T3 cells is likely due to endogenous c-abl expression. Total magnification, × 120.
Jason A. Wertheim et al. Blood 2002;99:
©2002 by American Society of Hematology