Supplemental Data. A Hypoxia-Controlled. Cap-Dependent to Cap-Independent. Translation Switch in Breast Cancer

Molecular Cell, Volume 28 Supplemental Data A Hypoxia-Controlled Cap-Dependent to Cap-Independent Translation Switch in Breast Cancer Steve Braunstei...
1 downloads 1 Views 174KB Size
Molecular Cell, Volume 28

Supplemental Data A Hypoxia-Controlled Cap-Dependent to Cap-Independent Translation Switch in Breast Cancer Steve Braunstein, Ksenia Karpisheva, Carolina Pola, Judith Goldberg, Tsivia Hochman, Herman Yee, Joan Cangiarella, Rezina Arju, Silvia C. Formenti, and Robert J. Schneider Table S1

Tumor Characteristics by Markers (N=94)

I. Tumor Category by 4E-BP1 level Tumor 4E-BP1 Total Category Normal +1 +2 # 11 10 6 27 T1 % 41 37 22 9 14 17 40 T2 23 35 42 3 7 17 27 T3 11 26 63 23 31 40 94 Total p= 0.0011 III. LABC by 4E-BP1 level LABC 4E-BP1 Total Normal +1 +2 26 23 70 No # 21 % 30 37 33 2 5 17 24 Yes 8 21 71 23 31 40 94 Total p= 0.0013 V. Tumor Category by eIF4E level Total Tumor 4E-BP1 Category Normal +1 +2 # 6 16 5 27 T1 % 22 59 19 5 25 10 40 T2 12 63 25 2 17 8 27 T3 7 63 30 13 58 23 94 Total p= 0.0017

II. Tumor Category by eIF4G Tumor eIF4G Category Normal +1 +2 # 16 8 3 T1 % 59 30 11 23 8 9 T2 58 20 22 7 6 14 T3 26 22 52 46 22 26 Total p= 0.0023

level Total 27 40 27 94

IV. LABC by eIF4G level LABC eIF4G Total Normal +1 +2 16 14 70 No # 40 % 57 23 20 6 6 12 24 Yes 25 25 50 46 22 26 94 Total p= 0.0023

Supplemental Experimental Procedures

IHC, immunoblot and VEGF ELISA analysis. Arrays were dewaxed, hydrated, blocked with preimmune sera and immunohistochemistry performed using optimal concentrations of primary antibodies which failed to stain tissues (details available upon request), then developed with secondary antibody and the ABC kit (Vector). Specimens were counterstained with hematoxylin. Antibody controls consisted of the absence of primary antibodies and inclusion of irrelevant concentration-matched isotype antibodies. Results were scored by one pathologist in a blinded manner and confirmed by another. The percentage of stained cancer cells was recorded from three sections of each specimen at x200 magnification. Staining was evaluated by a 3 point scale requiring 50% or more of specimen involvement: normal tissue levels; +1, moderate increased expression; +2, intense staining. Photographs were recorded on a Zeiss Axioplan 2 microscope. Immunoblot analysis was performed on total cell lysates after SDS-PAGE and electrophoretic transfer to membrane, and developed using the Enhanced chemiluminescence (ECL) system (Amersham). ELISA analysis for VEGF levels secreted into medium was performed by replacing medium 24 hours prior to testing under normoxic or hypoxic conditions, using equal volumes of media. VEGF levels were determined by triplicate ELISA (Quantikine, R&D Systems) and standard deviations calculated.

Cell lines, cell culture, antibodies, plasmids and transfection.

Rabbit polyclonal antiserum to

eIF4G was previously described (Cuesta et al., 2000). Mouse monoclonal anti-eIF4A antibody was provided by W. Merrick (Case Western Reserve University, Cleveland, OH). Other antibodies were from commercial sources and included mouse monoclonal anti-FLAG or anti-HA antibody (Sigma), horseradish peroxidase (HRP)-conjugated donkey anti-rabbit or sheep anti-mouse secondary antibodies (Amersham). All other antibodies were from Cell Signaling Technology. Cell lines MCF10A, BT474, MDA-MB435 and CRL1902 used for these studies were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and were grown under ATCC guidelines in recommended media. Hypoxia studies were performed at 0.5% O2/95% N2/5% CO2 for 24 hours unless otherwise noted, in a BioSphere hypoxia incubator. All experiments were performed with exponentially growing cells plated at approximately 40% cell density, then made hypoxic 18-24 hours later for 24 hours. Hypoxic media was pre-equilibrated overnight in the hypoxia chamber.

Normoxic cells used for

comparison were grown and treated under atmospheric oxygen in parallel. Transfection of plasmid DNA was carried out using standard conditions as described by the manufacturer (Fugene, Roche) with

1-2µg of DNA. The bicistronic mRNA constructs pRF, pRhifF, and pRvegfF were obtained from the Goodall lab (PMID12006670). The pRemcvF reporter was generated by insertion of the EMCV IRES from pBMN-IRES-EGFP (gift of G. Nolan, Stanford Univ.) into the pRF plasmid (EcoRI-NcoI). The pPRIG-VEGF-HARed reporter plasmid was generated by replacement of the EMCV IRES in pPRIGHd-HARed (Pognonec, PMID16409632) with the VEGF IRES from pRvegfF (SalI-HindIII).

The

bicistronic Bcl2 construct was similarly derived. Methionine incorporation assay. Cells were labeled with 50 µCi of [35S]methionine per ml (Easytag Express Protein Labeling Mix, Dupont/NEN) in DMEM without cold methionine for 1 h, washed twice with PBS and lysed in 0.5% NP-40 lysis buffer (0.5% NP-40, 50 mM HEPES, pH 7.0, 250 mM NaCl, 2 mM EDTA, 2 mM sodium orthovanadate, 25 mM glycerophosphate, and 1 tablet of protease inhibitor (Roche) per 10 ml) at 4°C. Lysates were clarified by centrifugation for 10 min at 13,000×g. Specific

Subcutaneous xenotransplant tumor model. Tumors were established in 6 week old female NCRnu CD1 nude mice (8 mice per group) by subcutaneous injection of 2 x 106 tumor cells in the mid-dorsal region in both flanks. Exogenous estrogen was supplied by implantation of 60 day estrogen release pellets containing 0.72 mg β-estradiol (Innovative Research of America, Sarasota, FL.). Tumor volumes were measured by short and long tumor axis with calipers (length x width2)x0.5. Mice were sacrificed at indicated times, tumors excised, weighed, and embedded in paraffin for IHC or in OTC for frozen section analysis. All mouse protocols were approved by the NYUSM Animal Care and Use Committee. For long-term tumor studies, a second set of estrogen pellets were implanted after 2 months in mice (5 animals per group).

Chick chorioallantoic membrane (CAM) analysis of tumor growth and angiogenesis. Studies were carried out as described (Brooks et al., 1994; Brooks et al., 2002) with 10 day old fertilized hen eggs and tumors grown for 8 days at a time when CAM vascularization has largely ceased. Eggs were maintained at 37°C with 60–70% humidity, a small window was drilled in the region of air sac to deflate the CAM, 106 cells were injected into the CAM of the chick embryo and sealed with 3M Tegaderm. At 8 days, tumors were photographed at high resolution, representative tumors were dissected, weighed, photographed and fixed in 4% paraformaldehyde and paraffin for IHC, or OCT for frozen section analysis.

Retroviral and lentiviral expression studies. 4E-BP1 wild type and mutant (Gift of P. Bitterman) were cloned into a retrovirus expression vector (pBABE-puro). HEK293 Gag-Pol packaging cells (from

G. Nolan) cells at 80% confluency in 10 cm diameter dishes were transfected with 5 µg each of pVPack-VSV-G and pBABE-puro vectors with the use of Lipofectamine plus (Invitrogen) or Fugene 6 (Roche). Retrovirus-containing supernatants were harvested after 48 hours, passed through a 0.45-µm filter, and frozen at -80°C until use. Target cells were subsequently infected by addition of retroviruscontaining supernatants to the medium along with polybrene (4 µg/ml). Target cells were selected 24 hours following infection by the addition of puromycin (1-1.5 µg/ml). Cells shown by immunoblot to express high levels of transgene following puromycin selection were released from selection for 48 hours before treatment with hypoxia. Interfering RNAs were delivered by transduction of cells with lentivirus shRNA expression vectors. Double-stranded shRNAs for cloning into lentivirus vectors were designed directed to either the 5’ or 3’ untranslated regions. To produce virus containing the shRNA generating cassette, 293GP cells were transfected with 4 µg each of pCI-VSV-G, pCMVD8.2R’, and pLK0.1 with the use of Fugene (Roche). Target cells were infected in the presence of polybrene (4 mg/ml) and selected with puromycin at 3.0 mg/ml for 48 hours.

Single-step real-time quantitative RT-PCR. qRT-PCR analysis of mRNAs was performed using the SYBR Green QRT–PCR Kit (Sigma) in a Lightcycler instrument (Roche). Forward and reverse primers used in this analysis for VEGF mRNA were: forward primer-CGAAACCATGAACTTTCTGC; reverse primer- CCTCAGTGGGCACACACTCC. These primers will amplify all splice variants. Reverse transcription was carried out at 61ºC for 20 min, denaturation at 95ºC for 30 sec, followed by 45 cycles of amplification: 95ºC for 2 sec for denaturation, 59ºC for 5 sec for annealing, and 72ºC for 10 sec for amplification and acquisition.

Statistical analysis. Nonparametric exact Jonckheere-Terpstra tests (2-sided) were used for doubly ordered contingency tables to test the associations between expression levels and each of the tumor characteristics individually. Distributions of patient age and tumor size were examined using boxplots and in contingency tables with categories defined by quartiles. Modeling of tumor size within expression levels of eIF4E, 4E-BP1 and eIF4G utilized Kruskal Wallis p-values. Fisher’s exact tests (2sided) were performed to assess the association of the presence or absence of data for tumor characteristics and markers. Logistic regression models were used to examine the associations between tumor markers and other variables with stage or LABC status. Odds ratios for LABC and 95% confidence intervals were estimated for increased levels of 4E-BP1 and eIF4G compared to normal levels.

References: Brooks, P. C., Montgomery, A. M., Rosenfeld, M., Reisfeld, R. A., Hu, T., Klier, G., and Cheresh, D. A. (1994). Integrin alpha v beta 3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 1157-1164. Brooks, P. C., Roth, J. M., Lymberis, S. C., DeWyngaert, K., Broek, D., and Formenti, S. C. (2002). Ionizing radiation modulates the exposure of the HUIV26 cryptic epitope within collagen type IV during angiogenesis. Int. J. Radiat. Oncol. Biol. Phys. 54, 1194-1201. Cuesta, R., Laroia, G., and Schneider, R. J. (2000). Chaperone Hsp27 Promotes Disassembly of Translation Cap Initiation Complex During Heat Shock. Genes & Devel. 14, 1460-1470.

HIF1α normal T3 LABC

Figure S1A. Represented are results typical of normal breast epithelium and a tumor core biopsy from a T3 tumor stained by IHC with antibodies to HIF1α.

MCF10A cells Vector N

Hx

BT-474 cells

4E-BP1 N

Hx

Vector N

Hx

4E-BP1 N

Hx HIF1α β-tubulin Bcl2

Figure S1B. MCF10A and BT-474 cells transformed with vector or overexpressing 4E-BP1 were subjected to 1% 02 for up to 12 hrs and lysates prepared. Equal amounts of protein lysates were resolved by SDS-PAGE and proteins identified above detected by immunoblot analysis. Autoradiograms were quantified by densitometry of typical results representative of at least three independent experiments. HIF1α and Bcl2 mRNA levels were quantified by qRT-PCR (not shown), and used to normalize protein levels.

Cap-dep 5’NCR 5’cap Lu c 5’cap

Lu c

VEGF IRES (polyA)

5’cap

(polyA)

5’cap

Lu c

(polyA)

Luc (polyA)

3.0 2.5

Relative Luc activity 2.0 (normalized to vector) 1.5 1.0 0.5 0

V BP ∆ V BP ∆

V BP ∆ V BP ∆

Nx Hx Cap-dep

Nx Hx Cap-dep+stem

V BP ∆

V BP ∆

V BP ∆

V BP ∆

Nx Hx Nx Hx VEGF IRES VEGF IRES+stem

Figure S1C. BT474 cells transformed with vector (V), 4E-BP1 (BP) or ∆4E-BP1 mutant (∆) as above were transfected with monocistronic vectors shown that express an mRNA encoding luciferase controlled by the cap-dependent 5’NCR, or VEGF IRES, used in Figure 2, with and without insertion of a stable (-50 kcal/mol) hairpin (stem) upstream of these elements. Cells were maintained under normoxic (Nx) or hypoxic (Hx) conditions as above. Equal amounts of cell lysates were prepared and luciferase activity determined. Results were normalized to mRNA levels and derived from three independent experiments and luciferase activity was normalized to the vector control in each.

Bcl2 bicistronic translation activity

A

Relative Luc activity (hypoxia/normoxia)

1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0

BT-474 cells

MCF10A cells

vector 4E-BP1 vector 4EBP1 Cap-dependent translation IRES-mediated translation

B

vector Normoxia Hypoxia

+ rapamycin

+

4E-BP1 Normoxia Hypoxia

-

+

-

+ HIF1α β-tubulin Bcl2

Figure S2B. BT-474 cells transformed with vector or overexpressing 4E-BP1 were subjected to 1% 02 for up to 24 hrs with or without treatment with 20 µg/ml rapamycin for 4 hrs. Equal amounts of protein lysates were resolved by SDS-PAGE and proteins identified above detected by immunoblot analysis. Autoradiograms were quantified by densitometry of typical results representative of at least three independent experiments.

Suggest Documents