SUPPLEMENTARY DATA

Supplementary Figures

Figure S1

Figure S1. YB-1 binds to the MKNK1 promoter. The first 1Kb region of the MKNK1 promoter with putative YB-1 binding sites. Chromatin immunoprecipitation of HR5 cells using YB-1 antibody demonstrates YB-1 binding in region ChIP 2, but not to regions ChIP 1, 3, or 4.

Figure S2

Figure S2. Silencing YB-1 decreases MNK1.BT474, HR5, and HR6 cells transfected with scrambled control or YB-1 siRNA (5nM) for 96 hours were analyzed by immunoblotting for changes in MNK expression.

Figure S3

Figure S3. YB-1 induction increases MNK1 in non-tumorigenic cells.Non-tumorigenic immortalized mammary epithelial cells stably transfected with a tetracycline inducible YB-1 plasmid were analyzed by quantitative real time PCR before and after 72 hours of induction. * p-value < 0.05 compared to control.

Figure S4

Figure S4. Silencing MNK1but not MNK2 induces apoptosis. BT474, HR5, and HR6 cells transfected with scrambled control, MNK1, MNK2, or MNK1 and MNK2 siRNA for 96 hours were assessed by MTT assays for cell viability and by immunoblotting for PH2A.XS139 and cleaved PARP for apoptosis. * p-value < 0.05 compared to scrambled control.

Figure S5

Figure S5. Silencing MNK1 induces apoptosis. (a) AU565, (b) MDA-MB-453, and JIMT-1 cells transfected with scrambled control or MNK1 siRNA were treated with trastuzumab (20g/ml) for 72 hours or left untreated. Cell viability was assessed by MTT assays. * p-value < 0.05 compared to each treatment alone.

Figure S6

Figure S6. MNK1 overexpression renders BT474 cells less sensitive to trastuzumab.BT474 cells transfected with EV or MNK1 plasmids were treated with trastuzumab for 72 hours or left untreated. Apoptosis was analyzed by immunoblotting for cleaved PARP. The expression of recombinant GST-MNK1 was confirmed by blotting for MNK1.

Figure S7

Figure S7. Inhibiting RSK in combination with trastuzumab induces apoptosis. (a) BT474 cells transfected with EV or MNK1 plasmids were treated with BI-D1870 (10M) or DMSO either in combination or without trastuzumab for 72 hours and apoptosis was analyzed by immunoblotting for P-H2A.XS139. The expression of recombinant GST-MNK1 was confirmed by blotting for MNK1. (b) BT474, HR5, and HR6 cells transfected with scrambled control or RSK1/2 siRNA were treated with trastuzumab (20g/ml) for 72 hours or left untreated. Cell viability and apoptosis were assessed by MTT assays and immunoblotting for P-H2A.XS139 respectively. (c-d) BT474, HR5, and HR6 cells were treated with DMSO control, BI-D1870 (10M), trastuzumab (20g/ml), or combination of BI-D1870 and trastuzumab for 7 days. (c) Cell viability was analyzed by MTT assays. (d) Pictures of representative wells are shown. * p-value < 0.05 compared to each treatment alone.

Supplementary Materials And Methods

siRNA/Plasmid Transfections

Cells were transfected with 20nM of siRNA (Qiagen) using Lipofectamine RNAiMAX (Invitrogen) on day 0, treated with trastuzumab (20μg/ml) on day 1, and end point measurements (MTT assay and immunoblotting) were taken on day 4, after 96 hours of knockdown and 72 hours of trastuzumab treatment. The MNK transfections were performed with 2mg of DNA and Lipofectamine 2000 (Invitrogen) using MNK1-pEBG6P (GST-tagged) that was generously provided by Dr. Christopher G. Proud (University of Manchester).For MNK1 over-expression studies, BT474 cells were seeded on day 0, transfected with empty vector (EV) or MNK1 on day 1, re-seeded in replicates on day 2, treated with drug (trastuzumab, BI-D1870, or both) on day 3, and end point measurements were taken on day 6, after 5 days of over-expression and 72 hours of drug treatment.

Chromatin Immunoprecipitation (ChIP)-PCR, ChIP-sequencing, and Bioinformatics

YB-1 promoter complexes were isolated as previously described (Astanehe et al., 2009). QIAquick PCR purification kit (Qiagen) was used to extract DNA fragments. For ChIP-PCR, the four sets of oligonucleotides designed around the putative YB-1 binding sites on the MKNK1 promoter were: ChIP1 forward 5’-GGTCCTGAATCTCTTGGGCTA-3’, ChIP1 reverse 5’-TGTGTTCATTCCCTTCTGGA-3’, ChIP2 forward 5’-AAGCTTGATCTTCCGCCTCT-3’, ChIP2 reverse 5’-ACTTGCTTCCCATCATCCAC-3’ ChIP3 forward 5’-TGTTTTTCCCCAAGAAATGAA-3’, ChIP3 reverse 5’-TGGGGAAATGCTGCTAGTC-3’, ChIP4 forward 5’-ACCCTCCATTCGGAAAAGAA-3’, ChIP4 reverse 5’-CTGCGCCCTGATGGCTAC-3’.

ChIP-sequencing was performed at the BC Cancer Agency Genome Sciences Centre using the second generation Illumina sequencing platform. For each sample, a 200bp fragment library was created and single-end 50bp reads were collected from one flowcell. Approximately 15.4M, 16.0M and 15.3M reads were collected for BT474, HR5 and HR6, respectively, and aligned to the hg18 human reference sequence using MAQ. Read alignments were converted into a coverage profile by first extending the read alignment to library fragment length (200bp) and subsequently, for each base position, counting the number of extended reads that overlapped the position. The coverage profile was then divided into contiguous regions of non-zero coverage. Each region was categorized as a peak if its false discovery rate (FDR) was less than 0.05. To determine the FDR for a region, a null hypothesis distribution of reads was generated by repeatedly shuffling the reads within the region and measuring the maximum coverage generated by the simulation. The fraction of simulated coverage profiles that resulted in maximum coverage equal or greater than what was observed was taken as the FDR for region. The coverage cutoffs were determined to be 8, 7, and 7, respectively, for BT474, HR5 and HR6. Genes with regions categorized as peaks within 20kb upstream and 1kb downstream of the gene were identified.

ChIP-sequence coverage profiles are available for each library at:

for N = 4, 5, 6. Data in these files can be browsed using UCSC Genome Browser by registering the file as a custom track.The UCSC assembly version used for data analysis was hg18 (Ensembl v53 Mar 2006).

1