ER stress and autophagy: new players in the mechanism of action and drug resistance of SUPPLEMENTAL DATA

Transcription

1 ER stress and autophagy: new players in the mechanism of action and drug resistance of the cyclin-dependent kinase inhibitor SUPPLEMENTAL DATA METHODS Confocal immunofluorescence microscopy of fixed cells: After hours exposure to agents, cells were washed in phosphate-buffered saline (PBS) and made adherent on a microscope slide by centrifugation in a Cytospin (Shandon) centrifuge. Immediately after cytospin, cells were fixed in cold acetone. Next, the cells were incubated in blocking solution (% bovine serum albumin in PBS) and stained for LC (Cell Signaling, Boston, MA) and Lamp (Santa Cruz Biotechnology, Santa Cruz, CA) by incubating separately with the respective primary antibodies overnight, in ºC, followed by incubation with fluorescent secondary antibodies (Invitrogen, Carlsbad, CA): Alexa fluor 9 (red) anti-rabbit for LC and Alexa fluor 88 (green) anti-mouse for Lamp. The LC-II form is usually very bright due to the dotted pattern, completely overpowering the intensity of uncleaved and LC-I forms, which are virtually comparable to the background due to the homogenous spread in the cell. Therefore, when we refer to LC fluorescence intensity, we are in fact presenting the LC-II form, which localizes to the walls of the autophagosomes. Nuclei were stained blue with DAPI (Vector Laboratories, Burlingame, CA). Images were taken using a x objective and x digital zoom with Olympus Fluoview Laser Scanning Confocal microscope at the Ohio State University Campus Microscopy and Imaging Facility. ATF antibody was purchased from LifeSpan Biosciences (Seattle, WA) and Nrf was from Santa Cruz Biotechnology (Santa Cruz, CA). Nuclei were evaluated for ATF and Nrf using confocal microscopy to collect Z stacks, approximately slices per cell,. µm each.

2 Quantification of immunofluorescence data Integrated intensity was measured using Metamorph v.7.. To calculate the number of dots (local extreme point in pixel-intensity profile at which a number of high-intensity pixels form a cluster), a Gaussian filter was first applied to reduce noise. Cells were enumerated electronically by an improved watershed-based algorithm. Quantification of co-localization was done by measuring the correlation index, which reflects the degree of overlap between red (LC) and green (Lamp) signals; if the index is, it indicates complete overlap, is no overlap. only and plus chloroquine conditions were not included in the quantification due to the fact that when autophagy is not stimulated, the signal for Lamp overpowers the signal for LC and co-localization between the two is not assessable. To quantify co-localization, two binary nuclei images - red (R) and green (G) - are obtained. The correlation value is calculated as the number of white pixels of (R&G) divided by the number of pixels of (R G), where (R&G) is the intersection of R and G, and (R G) is the union of R and G. Caspase activity assays Briefly, lysates containing approximately x cells were incubated with µm LEVD-AFC (caspase substrate) or LETD-AFC (caspase 8 substrate) from MP Biomedicals (Solon, OH) in X cytobuffer from BioVision (Exton, PA) containing mm dithiothretiol (DTT). Caspase activity was measured one hour after addition of substrate and caspase 8 activity was measured minutes after addition of substrate. Release of free AFC was measured with a Beckman- Coulter DTX 88 multimode detector (Filters: excitation, / nm; emission, / nm). Caspase inhibitor Z-YVAD-FMK was purchased from R&D Systems. Cell viability

3 Percent of live cells was determined by staining with annexin V-fluorescein isothiocyanate and propidium iodide (PI). After exposure to agents, cells were washed with PBS and stained in X binding buffer (BD Biosciences, Franklin Lakes, NJ). Cell viability was assessed by flow cytometry using a Beckman-Coulter Cytomics FC cytometer. Data were analyzed with the CXP Cytometer software package (Beckman-Coulter). A total of, cells were counted for each sample. The cells treated with were washed after hours with PBS and resuspended in regular growth medium (RPMI ) supplemented with % human serum and antibiotics for the remainder of the incubation time. In the case of /chloroquine samples, chloroquine was re-added in the fresh media after was washed at hours. For all the other conditions, cells were incubated with the respective drugs for hours continuously. Statistical analyses: Data from experiments involving quantification of fluorescence in vitro data, RT-PCR, viability measured by flow cytometry and caspase assays were first log-transformed to stabilize variances. Next, linear mixed effects models were applied to the log-transformed data in order to account for the correlations among observations from the same patients. Holm s procedure was used to adjust for multiple comparisons and control overall Type I error. For the quantification of LC fluorescence in serial samples (in vivo data), CLL patients were classified as either no response (n = ) or partial response (n = ). Similar to above, data from both experiments were log-transformed, and a linear mixed effects model was applied to the log-transformed data. The significance of the interaction between response category and time was tested.

4 SUPPLEMENTAL FIGURE LEGENDS S FIG : Autophagy markers in CLL cells. Immunoblot showing autophagy regulatory protein expression in CLL cells and normal B cells; Gapdh was used as loading control. S FIG : Chloroquine inhibits LC-Lamp co-localization. CLL cells were incubated with µm rapamycin, with or without chloroquine at several concentrations. Confocal fluorescence microscopy shows LC (autophagosomes) in red, Lamp (lysosomes) in green and DAPI (nuclei) in blue. Images were collected with x objective and x optical zoom. S FIG : Expression of ATG and ATG7 with sirna treatment. CLL cells were transfected with scrambled or combination of ATG and ATG7 sirna. hours after transfection, cells were incubated hours without or with (A) ( µm) or ( µm), or (B) ( µm). RNA was collected at the end of the treatment and analyzed by real-time RT-PCR. ATG and ATG7 expression was significantly reduced by sirna (P <.). For (A), n =, for (B), n =. CT was calculated by subtracting CT values for CD (our housekeeping gene) from CT values for ATG and ATG7. Higher values indicate lower gene expression. S FIG : Effects of CDK and CDK knockdown on autophagy. (A) CLL cells (n = ) were transfected with scrambled or CDK-specific sirna, and expression level of CDK was assessed by real-time RT-PCR. CT was calculated by subtracting CT values for CD (our housekeeping gene) from CT values for CDK. Higher values indicate lower gene expression. CDK sirna significantly reduced CDK expression (P =. for CDK sirna and P =.9 for CDK/CDK sirna). (B). Immunoblot for CDK and CDK in CLL cells incubated with different agents or transfected with CDK sirna. CDK protein expression was undetectable in all CLL samples; lysate from Jurkat cells was used as positive control for the antibody. (C). Confocal fluorescence microscopy shows increased LC fluorescence in the

5 presence of CDK sirna compared to the scrambled control sirna. Results shown are representative of experiments. S FIG : ER stress markers in CLL cells treated with. Real-time RT-PCR was used to analyze expression of (A) XBP, (B) GRP78 and (C) IRE in CLL cells (n = ) incubated with µm, µm and tunicamycin µg/ml. P values were calculated for versus each reagent used and they were as follows: for IRE: P =.8 for, P =. for, P =. for tunicamycin; for XBP: P =. for, P =.77 for, P =.99 for tunicamycin; for GRP78: P =.8 for, P =. for, P =. for tunicamycin. (D) Real-time RT-PCR for XBP, P =.7, (E) IRE, P =. and (F) GRP78, P =. in CLL cells (n = ) transfected with CDK sirna. S FIG : PERK and IRE activity in CLL cells. (A). Immunoblot for phosphorylation of eifα in the Jurkat cell line and CLL patient cells. (B). XBP splicing in cell lines, CLL cells and normal B cells. cdna from various cells was subjected to XBP-specific PCR, and products were separated on a % acrylamide gel. XBP splicing is detected with and in the cell lines 9T, RAW and 97, as well as in normal B cells, but was not detectable in CLL cells. (C) Immunoblot for GRP78 in CLL patient cells (representative of immunoblots). Gapdh was used as loading control. S FIG 7: Markers of cell death by ER stress. Activity of caspases (A) and 8 (B) in CLL cells (n = ) treated in vitro with ( µm), ( µm), tunicamycin ( µg/ml) or F- ara-a ( µm) alone or in combination with caspase inhibitor Z-YVAD-FMK ( µm). Cells were incubated with hours, washed, and re-plated in fresh media without drug; all the other reagents were left with the cells for hours. Activity is shown as free AFC/min/mg protein (fluorescence released by cleaved substrate). Caspase and 8 activity was significantly

6 increased by each agent in the absence of Z-YVAD-FMK (P =.). (C). Viability of CLL cells from (A) and (B), (n = ). Viability is shown as percent cells negative for both annexin and PI by flow cytometry at hours. Z-YVAD-FMK significantly blocked -mediated cell death (P =.). (D) Immunoblot for ASK after sirna transfection: lysates were collected after hours incubation of cells with reagents described in (B). (E). Activity of caspases and 8 was measured in CLL cells obtained from patients before treatment, at the end of infusion (. hours) and hours after the beginning of the infusion. (F). Real-time RT-PCR for ASK expression in CLL cells (n = ) transfected with scrambled or ASK sirna, then incubated hours with µm, µm, µm, or tunicamycin µg/ml (treatments were done 8 hours post-transfection). CT was calculated by subtracting CT values for CD (our housekeeping gene) from CT values for ASK. Higher values indicate lower gene expression. ASK decreased significantly (P =.). (G) Caspase assay in CLL cells (n = ) untransfected or transfected with either scrambled or sirna against ASK. S FIG 8: p/sqstm and LC gene expression does not change in the presence of autophagy inducers. Real-time RT-PCR tested p/sqstm (A, B) and LC (C, D) expression in CLL cells treated with autophagy inducers in vitro (n = ) and in vivo in CLL cells harvested from patients treated in clinic with, before infusion and at. h into the infusion.

7 Supplemental Figure Normal B ctrl cells CLL patient cells ATG (Barkor) ( kda) GAPDH ( kda) ctrl Normal B cells CLL patient cells Beclin- ( kda) ATGC ( kda) ATG- ( kda) ULK ( kda) GAPDH ( kda) UVRAG (9 kda) Bif- ( kda) PIK class III ( kda) GAPDH ( kda) 7 8 ctrl Normal B cells CLL patient cells

8 Supplemental Figure Rapamycin Rapamycin CQ (. µm) Rapamycin CQ (. µm) Rapamycin CQ (. µm) DAPI, LC, Lamp Lamp LC

9 Supplemental Figure A ATG ATG7 average CT 8 8 Untransfected Scrambled sirna sirna ATG/7 B average CT ATG ATG7 Untransfected Scrambled sirna sirna ATG/7

10 Supplemental Figure A average CT 8 Scrambled sirna CDK sirna CDK sirna CDK/CDK sirna B untransfected Scrambled sirna CDK sirna CDK sirna CDK + CDK sirna Jurkat lysate untransfected Scrambled sirna CDK sirna CDK sirna CDK + CDK sirna Jurkat lysate CDK ( kda) CDK ( kda) Tubulin ( kda) CLL PT CLL PT C Rapamycin CDKsiRNA scrrna DAPI, LC, Lamp Lamp LC

11 average fold change.. h. h h h h h E Tunicamicin h h h C F Tunicamicin... h.. h h B D. Tunicamicin A. average fold change. average fold change Supplemental Figure..

12 Supplemental Figure A B 9T RAW 97 h h Jurkat cell line h h Tunicamycin h Tunicamycin h Ladder h h h Tunicamycin h h h h i h h h h h Ladder P-eIFα (8 kda) Total eifα (8 kda) CLL cells h h F -ara -A 8h F -ara -A Jurkat Ladder h Control agent h h h h Tunicamycin h 8h Control agent 8h 8h 8h 8h Tunicamycin 8h 8h 8h P -eif α PT Total eif α Normal B cells HeLa P -eif α Total eif α PT Ladder h h h h 8h 8h 8h 8h h h P -eif α PT Total eif α C PT PT P -eif α Total eif α PT Ctrl F -ara-a F-ara -A GRP78 (78 kda) GAPDH ( kda) GRP78/GAPDH

13 P R E. h P R E. h PRE Supplemental figure 7 A B E fold change of average free AFC/min/mg protein Z-YVAD-FMK + Z-YVAD-FMK + Z-YVAD-FMK Tunicamycin Tunicamycin + Z-YVAD-FMK + Z-YVAD-FMK fold change of average free AFC/min/mg protein Z-YVAD-FMK + Z-YVAD-FMK + Z-YVAD-FMK Tunicamycin Tunicamycin + Z-YVAD-FMK + Z-YVAD-FMK fold change of average free AFC/min/mg protein. h PRE Caspase Caspase 8. h F C 9 8 percent ann-/pi- cells 7 average CT 8 Z-YVAD-FMK flavo + Z-YVAD thapsi + Z-YVAD tunicamycin tunica + Z-YVAD + Z-YVAD tunicamycin tunicamycin untransfected scrambled sirna ASK sirna tunicamycin D G ASK ( kda) GAPDH ( kda) scrrna sirna ASK scrrna sirna ASK scrrna F -ara -a sirna ASK F -ara -A scrrna sirna ASK thapsigagin Ctrl Jurkat cells average free AFC/min/mg protein scr scr scr scr sirna ASK sirna ASK sirna ASK sirna ASK

14 Supplemental Figure 8 A C fold change relative to fold change relative to B D.8.8 fold change relative to PRE PRE.h fold cahnge relative to PRE PRE. h.. PT PT PT PT PT PT PT PT PT PT