Supplemental Materials and Methods

Chemicals and peptides

Media and cell culture supplements were purchased from Invitrogen. Paclitaxel, vinblastine, vincristine, adriamycin, ectoposide, cyclosporin A were purchased from Sigma. Epothilone B, Pertussis toxin and TSP-1 purified from human platelets were purchased from EMD Biosciences. Docetaxel(Adventis) was a gift from Dr. Branimir Sikic. TSP-1 mimetic peptides, 4N1K (KRFYVVMWKK, amino acids 1016 to 1023) and its control peptide 4NGG (KRFYGGMWKK) were synthesized by GeneMed (South San Francisco, CA); ABT-510 (GVITRIR, amino acids 553 to 559) was a gift from Abbott Laboratories.

Taxol uptake and polymerized tubulin assay

1104cells were seededin each well of 24-well plates one day prior to drug treatment. Media were removed and replaced with media containing50 nM of [3H]paclitaxel (3.2 Ci/mmol, Moravek Biochemicals). Cells were incubatedwith [3H]paclitaxel for 3 hr, cooled on ice, washedthree times with ice-cold PBS, and lysed by addition of 0.25 mlof 1% SDS. The radioactivity in each sample was determined byscintillation counting.For tubulin polymerization assays, experiments were carried out as described (Minotti et al., 1991; Ng et al., 2006). Briefly, 1106cells were treated with taxol at the indicated concentration for 48 hr, lysed in tubulin stabilized buffer (0.1 M PIPES, pH 6.9, 2 M glycerol, 5 mM MgCl2, 2 mM EGTA, 0.5% Triton X-100 and protease inhibitors) supplemented with 4 M taxol to maintain microtubule stability during isolation. Supernatants containing solubilized tubulin were clarified by centrifugation (20,000 g for 45 min) and separated from pellets containing sedimented polymerized tubulin. Pellets were washed twice with in tubulin stabilizing buffer before immunoblotting analysis and probed with anti--tubulin.

Northern blot, RT-PCR and microarray analysis

Poly(A) RNA was extracted directly by FastTrack kit (Invitrogen) andseparated by formaldehyde/MOPS 1% agarose gel, transferred onto Hybond-N nylon membrane (Amersham Biosciences), and hybridized with radioactively labeled cDNA fragments. Multiple-tissue blots were purchased from BD Biosciences Clontech. For the chained reverse transcription and Taq polymerase reactions were performed as describedin the vendor’smanuals (Invitrogen), sequences of oligonucleotides, cycle numbers used for amplification, and product size of each gene were:

Tubulin classI:5’-GTGGCAAATATGTTCCTCG-3’and 5’-AGCCCTGCAGGCAGTCACAG-3’, 20 cycles, 230 bp

Tubulin class III:5’-ACATCTCTTCAGGCCTGACAATTTCATC-3’and 5’- TGCTGATGAGCAACGTGCCCATGCCGGAGC-3’, 20 cycles, 215 bp

Actin: 5’-GAGGCACTCTTCCAGCCTTCCTT-3’ and 5’-GCGTACAGGTCTTTGCGGATGT-3, 20 cycles, 110bp

Txr1: 5’-ATGGTTGGACCAGCAGTGATA-3’and 5’-GGAAGGGTCCAGGGCCTGTAT-3’, 20 cycles, 156 bp

Promoter activity analysis

2105 cells were plated in 6-well dishes for 24 hr prior to transfection. 2g of TSP-1 promoter luciferase report construct was transfected into cells with 0.2g of pCMVlacZ by FuGENE 6 (RocheApplied Science). 48 hr after transfection, luciferase activity was measured in cell lysates using a luciferase assay kit (Promega) according to vender’s instruction and -galactosidase was determined bya gal chemiluminescence assay (RocheApplied Science). Relative luciferase units were calculated by normalizing the luciferase activity against -galactosidase activity.

Flow cytometry analysis

Cells were stained 30min on ice with either FITC conjugated anti-CD36 antibody (1:20 dilution, clone FA6.152, Beckman), phycoerythrin conjugated anti-CD47 antibody (1:20 dilution, clone B6H12, BDPharmingen) or corresponding isotype antibody (mouse IgM for anti-CD36, mouse IgG for anti-CD47) at the same dilution. Flow cytometry was performed using FACScan (Becton Dickinson) and the results were analyzed using FlowJo software.