br Cryopreserved MCF and T D cells were thawed washed
Cryopreserved MCF7 and T47D 6-NBDG were thawed, washed once in 1 ml Phosphate Buﬀered Saline (PBS) and resuspended in 1 ml re-spective media. The trypan blue viability assay coupled with a TC20 Automated Cell Counter (Bio-Rad, Hercules, CA USA) was used to de-termine cell viability. Cells were seeded in duplicate at a concentration of 1 × 105 cells/well into 24-well plates (Nunc, Roskilde, Denmark) and incubated in 200 μl respective media for 24 h at 37 °C and 5% CO2 for cell adherence and formation of a monolayer.
2.3. Hormone-therapy treatment
The maximum eﬀective concentrations of Anastrozole (Sigma Aldrich, Saint Louis, USA; A273) and Tamoxifen (Sigma Aldrich; T5648), were determined using the LDH cytotoxicity assay (Thermofisher Scientific, Waltham, USA) and Neutral Red cell viability assay . Tamoxifen and Anastrozole were tested at ranges from 0.5 to 2.5 μM and 1–10 μM, respectively. This correlates with other cellular studies conducted in which concentrations mimicked that of treatment regimens given to women with breast cancer [38–40]. Based on ensuing results and corresponding with other studies, the maximum eﬀective concentration of Tamoxifen was determined to be 2 μM and Anastro-zole, 1 μM [38–40].
Cells were thus rinsed with PBS and incubated in 200 μl 1 μM Anastrozole or 200 μl 2 μM Tamoxifen in respective media for 24 h at 37 °C. Controls included normal media and the diluent (0.1% DMSO).
2.4. Blood preparation and co-culture
Healthy female volunteers between 19 and 30 years old, and be-tween days 1–10 of their menstrual cycle (low levels of circulating oestrogen and progesterone, lessening hormonal eﬀects on breast cancer cells) participated in this study. Exclusion criteria included: pregnancy; contraceptive use; autoimmune diseases or im-munodeficiency; history of cancer; cancer; smoking; and consumption of anti-platelet and anti-coagulation medication (e.g. aspirin and war-farin) in the previous 72 h. Ethics was approved by University of the Witwatersrand (#M160826).
Peripheral whole blood (n = 5) was collected in two 3.2% sodium citrate vacuette coagulation tubes. The first 2 ml of blood drawn was discarded to exclude mechanically activated platelets. Following drug treatment, breast cancer cells were rinsed twice with PBS and incubated with 200 μl whole blood (WB) for 2.5 min.
WB samples underwent erythrocyte lysis with 3 ml ammonium chloride (NH4Cl) buﬀer for 10 min at room temperature for subsequent analysis . For the positive control WB was incubated with 0.1 U/ml human thrombin-α (SANBS, South Africa), for 5 min to induce platelet activation; the negative control included WB not exposed to breast cancer cells.
2.5. Scanning electron microscopy (SEM)
Twenty microliters of sample was placed on glass coverslips (10 mm round, Lasec, Johannesburg, South Africa) in 24-well plates and was incubated at 37 °C, 5% CO2 for 5 min to facilitate adhesion to the coverslip . Samples were washed in 0.1 M PBS on a microplate
shaker for 20 min, fixed in 2.5% formaldehyde/glutaraldehyde for 15 min, rinsed thrice with 0.1 M PBS, secondarily fixed in 1% osmium tetroxide and rinsed in PBS. Cells were dehydrated through a series of ethanol (30%, 50%, 70%, 90%, and three times absolute ethanol), and dried using hexamethyldisilazane. Samples were mounted onto alumi-nium stubs, coated by carbon evaporation and examined using a FEI Nova 600 Scanning Electron Microscope (acceleration voltage 30 kV) at the Microscopy and Microanalysis Unit (University of the Witwa-tersrand) for qualitative assessment.
Samples were centrifuged at 200 ×g for 5 min and the resulting pellet resuspended in 150 μl Tyrode's buﬀer for double-labelling with APC-conjugated mouse anti-human CD41a (BD Pharmingen, 559777) and FITC-conjugated mouse anti-human CD62P (BD Pharmingen, 555523) at 1:20, for each antibody, as determined via titration. Samples were fixed in 1% paraformaldehyde at 1:1 for 10 min, washed with 1 ml Tyrode's buﬀer, and centrifuged at 200 ×g for 5 min. The resulting pellet was resuspended in 500 μl Tyrode's buﬀer, kept at 4 °C on ice overnight prior to data acquisition on the LSR Fortessa (BD Biosciences) at the Department of Surgery, University of the Witwatersrand. Compensation and technical controls included an unlabelled sample and single antibody labelled samples to determine auto-fluorescence and fluorescence overlap.
Data was acquired as 100 000 events per sample using FACSDiva Software (BD Biosciences). The platelet population initially gated on size and granularity (forward scatter voltage 300 V, side scatter voltage 275 V), was further gated based on expression of CD41a platelet marker (APC, 565 V), denoting the parent population. Interval gates were drawn to classify a graded level of platelet activation determined using CD62P (FITC 469 V) geometric Mean Fluorescence Intensity (gMFI) of the platelet population (Fig. 1).