Macrophage traits in cancer cells are induced by macrophage-cancer cell fusion and cannot be explained by cellular interaction

Cell culture

MCF-7/GFP breast cancer cell line (Cell Biolabs, INC. San Diego, USA) was cultured
in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 1 % PEST,
10 % FBS, 2.5 % HEPES and 1 % L-glutamine (Gibco®, Life Technologies, USA) in a T-75
tissue culture flasks (Sigma-Aldrich Co, ST. Louis, USA) and incubated at 37 °C in
humidified air 5 % CO
₂
atmosphere. Cell medium was changed every 2–3 days, and the cells were passaged at
95 % confluence.

Monocyte isolation

Monocytes were isolated from buffy coat obtained from male healthy blood donors at
the department of Transfusion Medicine, County Council of Östergötland, in Linköping,
Sweden. All the blood donors had given their informed consent according to the local
guidelines (University Hospital in Linköping) and the Swedish National Law on ethical
review of research involving humans (2003:460: 3–4 §). The buffy coat was mixed with
70 ml NaCl, layered onto Lymphoprep (Axis-Shield, Oslo) and centrifuged at 480 g in
room temperature for 40 minutes. The mononuclear cell layer was collected into new
tubes and washed twice with PBS-Heparin for 5 min and centrifuged at 220 g in 4 °C.
The white blood cells were seeded to T-75 tissue culture flasks with RPMI 1640 medium,
supplemented with 1 U/ml penicillin, 10 ?g/ml streptomycin and incubated for 1–2 h
to allow monocyte adhesion. The non-adherent cells were eliminated by washing 2–3
times using PBS. The adherent monocytes were allowed to differentiate to macrophages
with 40 ng/ml of macrophage colony-stimulating factor, M-CSF (Nordic Biosite, Sweden)
for 5–7 days. To induce M2 macrophages, the M-CSF differentiated macrophages were
stimulated with 20 ng/ml human interleukin-4, IL-4 (Nordic Biosite, Sweden) for 18–24 h.

Cell fusion and cellular interaction model

Green fluorescent protein (GFP) labeled MCF-7 cancer cells and macrophages were co-cultured
in ThinCert
^TM
cell culture inserts (Greiner Bio One, Kremsmünster, Austeria), where both cell types
were allowed to have cellular interaction without physical contact to prevent cell
fusion. The macrophages (5×10
⁵
) were seeded in the upper chamber on a polyethylene terephthalate (PET) membrane
with 0.4 ?m pores and physically separated from MCF-7/GFP cancer cells that cultured
in bottom of the lower chamber. This cell culture model allows intercellular signaling
via e.g. cytokines and exosomes, which can freely pass through the PET-membrane pores
between the cells (Fig. 1a). It does not allow cell fusion.

Fig. 1. Transwell culture system. The porous bottom of the insert provides independent access
to both sides of a cell monolayer, allowing in vitro cellular interactions (a). Spontaneous cell fusion was allowed by culturing MCF-7 cancer cells and M2 macrophages
along the bottom of the same chamber (b)

To induce spontaneous cell fusion, macrophages and GFP-labeled MCF-7 cancer cells
were co-cultured in the same cell culture vial in RPMI 1640 medium during 2–3 days.
The cells were seeded at a ratio of about 3–5:1 (macrophages: MCF-7). Cell fusion
experiments were repeated several times, and approximately 5×10
⁵
macrophages were used in each trial. We estimated the size of the population of hybrids
on the basis of the number of macrophages cultured with MCF-7 cancer cells. We did
so for a number of reasons, viz. MCF-7 cancer cells proliferate rapidly, macrophages
do not undergo cell division, and we assumed that a hybrid cell is generated by fusion
between a macrophage and a cancer cell (Fig. 1b).

Fluorescence-activated cell sorting (FACS)

Cells were washed once with PBS and harvested with a 0.05 % trypsin-EDTA solution.
Detached cells were washed with PBS and resuspended in 95 ?l Cell Staining Buffer
(Biolegend, San Diego, USA) at a concentration of about 5×10
⁶
cells/ml. The cell suspension was incubated on ice for 10 min with 5 ?l TrueStain
FcX solution (BioLegend, San Diego, USA) per 1×10
⁶
cells. Combinations of direct conjugated monoclonal anti-human CD163 (APC Anti-human
CD163 (IgG1 k), clone GHI/61, con 100 ?g/ml) and anti-human CD45 (PerCP/Cy5.5 anti-human
CD45 (IgG1 k), clone HI30, 50 ?g/ml) antibodies or their respective isotype controls
(APC and PerCP/Cy5.5 mouse IgG1 k, clone MOPC-21, con 200 ?g/ml) (Biolegend, San Diego,
USA) were added to the cell suspension at concentrations recommended by the manufacturer
and incubated at 4 °C in the dark for 30 min. The labeled cells were washed twice
and diluted in 1 ml PBS and filtrated in pre-separation filter 30 ?m (Miltenyi Biotech,
Lund, Sweden) before flow cytometry analysis. Cells in both the ThinCert culture system
and co-culture were examined initially with a Gallios flow cytometer (Beckman Coulter,
Inc.) and cells were sorted with BD FACSAriaâ„¢ III (BD Bioscience, USA). The cells
were examined in relation to GFP, CD163 and CD45 expression. Cells were initially
sorted by GFP expression (positive selection of MCF-7/GFP origin) and subsequently
by CD163 and CD45 expression (positive selection of cancer cells with macrophage phenotype).

Immunofluorescence microscopy

Macrophages, MCF-7 cells and hybrids (1×10
⁵
cells) were seeded on coverslips and incubated 24 h in RPMI?+?10 % FBS. Cells were
fixed with 4 % paraformaldehyde for 30 min at 37 °C, washed once in PBS followed by
permeabilization/blocking for 30 min in 2 % BSA/0.1 % Saponin in PBS. Cells were then
incubated with a mouse monoclonal ?-CD163 antibody (Abcam) in PBS/0.5 % BSA for 2 h
at room temperature and washed three times with PBS. A secondary antibody goat anti-mouse
IgG Alexa Fluor 546 (Invitrogen) was added in PBS/0.5 % BSA for 45 min, followed by
three washes with PBS. The cover slips were mounted on microscope slides in Dako fluorescence
mount media containg DAPI. Fluorescence images were taken with a Zeiss Axiovert 200 M
fluorescence microscope with a Zeiss Plan-APOCHROMAT 63x/1.4 oil DIC objective.

Immunostaining and expression levels of CD163 in relation to survival data

To investigate whether the proportions of CD163 positive breast cancer cells have
been correlated to clinical data, we re-evaluated breast cancer specimens from 127
women, a well controlled patient material that was reported in previous studies 11], 14], 15]. Written informed consent for participation in research was obtained from participants
in connection with previous studies. Ethical approval from the Regional Ethics Committee
in Linköping obtained according to Swedish Biobank Law (Reference number: 2010/311–31).
The patients were diagnosed and treated using conventional methods at surgical departments
in southeastern Sweden. All patients were in Stage II according to the UICC, and all
received adjuvant tamoxifen therapy. These specimens had previously been collected
in a tissue microarray and originated from a Swedish randomized trial of 2 versus
5 years of tamoxifen treatment. Serial sections of 5 mm were cut from tissue array
blocks, deparaffinized in xylene, and hydrated in a series of graded alcohols (100 %,
95 %, and 70 %). Heat-induced antigen retrieval was carried out using a water bath
pretreatment in Tris Ethylenediaminetetraacetic acid (1 mM, pH 9) for 50 min before
staining for CD163. Detection was carried out using the DAKO Envision system. The
immunoreactivity of CD163 was characterized by granular cytoplasmic, or cytoplasmatic
and membrane staining patterns. In negative control samples, the primary antibody
was replaced by an isotype-antimouse immunoglobulin G1 antibody. All immunostaining
was evaluated by two of the authors (HO and IS) and scored on a 5-tiered score as
follows: 0 %, 1–25 %, 26–50 %, 51–75 %, and 76–100 % of the cancer cells. Macrophages
and cancer cells could be distinguished on morphological basis. Macrophage nuclei
were small and regular, whereas the cancer cells were enlarged and atypical with pleomorphic
hypertrophic and darker nuclei. Moreover, cancer cells show a decreased cytoplasmic
– nuclear ratio.

To investigate the significance of CD163 expression levels in relation to survival
data, we used four different cut-off points 1–25 %, 25–50 %, 50–75 % and 57–100 %
of CD163 positive cancer cells in tumor sections. The correlation of CD163 expression
levels and survival rates, both disease specific survival (DSS) and distant recurrence
free survival (DRFS), was estimated using Kaplan-Meier analyses and the log rank test.

STR analysis/Quantitative fluorescent PCR

DNA was extracted from macrophage, MCF-7 breast cancer cells and MCF-7/macrophage
hybrid cell suspensions in a biorobot (EZ1, Qiagen) with DNA Tissue kit (Qiagen) following
the manufacturer’s instructions. Each sample was then subjected to multiplex amplification
of 24 STR markers on chromosomes 13, 18, 21, X and Y in two sets of tubes (Table 1) using ChromoQuant® QF PCR kit (Cybergene AB). The PCR was carried out in 25 ?l reactions
containing 14.6 ?l mastermix, 0.4 ?l GoTaq polymerase (Promega Inc) and 10 ?l DNA
(1.5 ng/?l). An initial denaturation at 94 °C for 3 min was followed by 26 cycles
of 30 seconds at 94 °C, 1 min of annealing at 57 °C, and 2 min of extension at 71 °C.
An extension period of 5 min at 71 °C followed the final cycle.

Table 1. Size and number of alleles from short tandem repeat (STR) analysis

PCR products were separated by capillary electrophoresis on an ABI 3130xl Genetic
Analyzer (Applied Biosystems). For each well, 1 ?l PCR product was mixed with 12 ?l
HiDi formamide and 0.3 ?l GeneScan-500ROX size marker, followed by denaturation at
95 °C for 2 min before loading. The POP7 polymer was used in the electrophoresis,
and results were analyzed using GeneMapper software version 4 (Applied Biosystems).