RAW 264.7 Macrophage Polarization by Pancreatic Cancer Cells – A Model for Studying Tumour-promoting Macrophages

Abstract. Background/Aim: Tumour-associated macrophages

(TAMs) are highjacked M2-polarized macrophages that

especially promote pancreatic cancer growth. The aim of this

study was to identify an easy-to-use cell culture model suitable

for studying this interaction and macrophage polarization.

Materials and Methods: Co-cultures of two cell lines,

PDA6606 cells with RAW macrophages cells were used in

vitro and in ovo. Macrophages were analyzed by microscopy,

magnetic resonance imaging (MRI), and flow cytometry.

Results: By comparing chemically-induced M1 and M2

macrophages, a clear induction of the M2 phenotype of RAW

macrophages by PDA6606 pancreatic cancer cells was

observed in vitro. In ovo, PDA6606 cells and conditioned

media polarized macrophages to the M2 phenotype, which in

turn promoted tumour growth and angiogenesis via their

surface marker profiles and VEGF production. Conclusion:

PDA6606 pancreatic cancer cells expectantly and potently

induced M2 polarization of RAW264.7 macrophages. This

model may be used to study pancreatic cancer-macrophage

plasticity in e.g. drug research in vitro and in ovo.

Macrophages are cells with high plasticity that can adapt

their profile according to specific environmental stimuli (1-

3). Two extremes of macrophage activation continuum have

been designated: M1 and M2. The M1 phenotype can be

induced by IFNγ alone or combined with microbial products,

such as lipopolysaccharide (LPS). M1 cells release pro-

inflammatory cytokines, such as tumour necrosis factor alpha

(TNF-α), interleukin (IL)-12, and IL-23. They express high

levels of major histocompatibility complex (MHC)

molecules and inducible nitric oxide synthase (iNOS), and

can be cytotoxic against neoplastic cells (4). M2 or

‘alternatively activated’ macrophages are induced by e.g.

TH-2 type cytokines, such as IL-4, IL-10, and IL-13 (5).

These cells release anti-inflammatory cytokines, such as

IL-10, express high levels of arginase and scavenger receptor

A (including mannose receptor, CD206), have poor antigen-

presenting capability but enhanced debris clearance ability,

which is important for promotion of wound-healing and

angiogenesis (6).

Tumour-associated macrophages (TAMs) and M2

macrophages share tumour-promoting functions and surface

markers (7). Their presence is associated with poor prognosis

and survival of patients with several types of cancer (8). The

current understanding is that tumour cells highjack

macrophages and utilize their M2-associated properties to

fuel tumour growth. Hence, studying macrophage polarization

by tumour cells has been a heavily investigated research field

in the last decade (9-11). Two major points of investigation

are i) whether macrophage polarization can be reverted to e.g.

an M1 phenotype, and ii) how tumour cells are able to induce

the M2 phenotype in TAMs and use them for their benefit.

TAMs are especially important for fostering tumour

metastasis, which is responsible for the poor prognosis of

pancreatic cancer patients (12). Pancreatic cancer is a

malignancy that ranks 4th in terms of fatality among all

cancer types. Our previous research has shown the

importance of M2 macrophages in supporting tumour growth

in vitro and in vivo (13). In vivo depletion (efficacy of about

80%) of these macrophages resulted in abrogated tumour

growth and decreased blood vessel density within the tumour

microenvironment. The remaining tumour-resident

macrophages in both the non-depleted and depleted animals

were of the M2-phenotype. In vitro, a polarization towards

an M2 phenotype was observed in primary murine

2871

*These Authors contributed equally to this work.

Correspondence to: Sander Bekeschus, ZIK plasmatis, Leibniz

Institute for Plasma Science and Technology (INP Greifswald),

Felix Hausdorff Str. 2, 17489 Greifswald, Germany. Tel: +49

38345543948, e-mail: sander.bekeschus@inp-greifswald.de

Key Words: Angiogenesis, M2 macrophages, tumour-associated

macrophages, VEGF.

ANTICANCER RESEARCH 39: 2871-2882 (2019)

doi:10.21873/anticanres.13416

RAW 264.7 Macrophage Polarization by Pancreatic Cancer

Cells – A Model for Studying Tumour-promoting Macrophages

AYDAR KHABIPOV

, ANDRE KÄDING

, KIM ROUVEN LIEDTKE

ERIC FREUND

, LARS-IVO PARTECKE

and SANDER BEKESCHUS

Department of General, Visceral, Thoracic and Vascular Surgery,

Greifswald University Medical Centre, Greifswald, Germany;

ZIK plasmatis, Leibniz Institute for Plasma Science and Technology (INP Greifswald), Greifswald, Germany

macrophages upon co-culture with murine PDA6606

pancreatic cancer cells. Experiments using primary cells are

of high importance, but recent developments in biomedical

research calling for implementation of the 3-R principles

(animal replacement, refinement, and reduction) require

alternative models. Accordingly, the acquisition of the M2

phenotype by the murine macrophage cell line RAW264.7

upon co-culture with murine PDA6606 pancreatic cancer

cells was examined as a novel research tool for investigating

tumour-promoting macrophages.

Materials and Methods

Chemically-induced macrophage differentiation. Murine RAW264.7

(RAW) macrophages (ATCC TBI-71) were cultured in Roswell-

Park-Memorial-Institute medium (RPMI1640; Pan BioTech,

Aidenbach, Germany) supplemented with 10% fetal bovine serum

(Sigma, Taufkirchen, Germany) and 2% penicillin/streptomycin

(Pan-BioTech, Aidenbach, Germany). For differentiation, cells were

stimulated with M-CSF (20 ng/ml) for 48 h, followed by a 72-h

stimulation with either interferon (IFN) γ (0.3 μg/ml; Sigma,

Taufkirchen, Germany) to obtain M1 macrophages or IL4 (40

ng/ml) and IL13 (40 ng/ml; both from Sigma, Taufkirchen,

Germany) for M2 polarization. Together with non-stimulated (M0)

macrophages, these cells served as controls in several experiments.

PDA6606-induced macrophage differentiation

The murine pancreatic ductal adenocarcinoma cell line PDA6606

(source: Prof. Tuveson, Cambridge, UK) was cultured in fully

supplemented RPMI1640. For co-culture with macrophages, varying

concentrations of PDA6606 were seeded in 6-well plates. After

overnight incubation, 1×10

macrophages were added, and assayed

at the indicated time points. For transwell-assays, macrophages were

not added directly to PDA6606 cultures but instead into a transwell

(Nunc, Roskilde, Denmark) for up to 72 h, physically separating the

two cell types from each other while allowing the exchange of

soluble mediators through the cell culture medium. In a third

setting, RAW264.7 macrophages were cultured for up to 72 h with

varying concentrations of supernatants of PDA6606 cells that were

obtained only after overnight culture, to minimize nutrient

depletion. The medium contained phenol-red and was still red-pink

after collection (data not shown). Cells were characterized by

microscopy and flow cytometry.

Microscopy. Brightfield images of 6-well plates were acquired with

a 5× objective using an inverted microscope (Observer Z1; Zeiss,

Jena, Germany). Quantitative image analysis was performed using

Image J 1.8 to assess the size of macrophage cell clusters at 3 days

of culture (3d). For co-cultures with fluorescently-labelled

macrophages (stained with 1 μM cell trace red for 30 min prior

(ThermoFisher, Dreieich, Germany)), 1×10

RAW264.7 cells were

added to 1x104 PDA6606 pancreatic cancer cells grown in 96-well

plates and incubated overnight. The plate was imaged at different time

points using an Operetta CLS high content imaging device

(PerkinElmer, Hamburg, Germany), and overlay images were

retrieved using Harmony 4.8 software (PerkinElmer, Hamburg,

Germany). This software was used to quantitatively assess cluster size

in an algorithm-based manner at several incubation time points. In

principle, cell trace red-positive macrophages were segmented and

events <500 μm

(representing single cells or small clusters) were

excluded before the sum of the remaining area was calculated.

Several hundred images were used for analysis. For

immunofluorescence, tumours were explanted and embedded in OCT

prior to conservation in liquid nitrogen. Tissue sections were

generated using a cryotome (ThermoFisher, Dreieich, Germany),

fixed (PFA 4%) and permeabilized (0.1% Triton X100), and nuclei

were counterstained with 4’,6-Diamidin-2-phenylindol (DAPI; Sigma,

Taufkirchen, Germany). Additionally, macrophages were labelled

either using an anti-CD11b antibody conjugated to Alexa Fluor 488

or using antibodies against CD206 (Alexa Fluor 488) or VEGF

(Alexa Fluor 647; all BioLegend, London, UK). Tissue sections were

mounted using vectashield mounting medium (VWR, Darmstadt,

ermany), and images were acquired by a fluorescence microscope

(BZ-9000; Keyance, Neu-Isenburg, Germany) and evaluated with BZ-

II-Analyser 4.6.2.2 software (Keyence, Neu-Isenburg, Germany).

Flow cytometry. For determining the RAW264.7 polarization type,

multicolour flow cytometry was utilized. Cells were collected into

12×75 mm tubes (Sarstedt, Nuembrecht, Germany), washed, and

stained with monoclonal antibodies (depending on the assay) against

the following macrophage markers: Ly6C, CD206, F4/80, CD68,

CD11b, FcεR, CD115 (all from BioLegend, London, UK) or EGR2

(BioTechne, Wiesbaden, Germany). In some assays, intracellular

staining was performed against iNOS or arginase-1 (all from

BioLegend, London, UK) in cells that were fixed (fixation buffer,

BioLegend, London, UK) and permeabilized (permeabilization wash

buffer, BioLegend, London, UK). To distinguish RAW264.7

macrophages from PDA6606 pancreatic cancer cells in co-culture

samples or mixed-population tumours (see HET-CAM), CD11b was

used as a discriminating marker (data not shown). Data were

analysed on a Gallios or CytoFLEX S cytometer (Beckman-Coulter,

Krefeld, Germany). Data analysis was performed using Kaluza 2.1

software (Beckman-Coulter, Krefeld, Germany).

Chemokine/Cytokine analysis. For quantification of vascular

endothelial growth factor (VEGF), a commercially available ELISA

kit was utilized according to the manufacturer’s instructions

(BioLegend, London, UK), and absorption was measured on a

M200 multimode plate reader (Tecan, Maennedorf, Switzerland)

against a known standard. For screening of PDA6606 supernatants

for 13 different chemokines/cytokines, Legendplex assay

(BioLegend, London, UK) was utilized. This assay quantifies

analytes by flow cytometric quantification of beads carrying

antibodies targeted against a single analyte. A secondary,

fluorescently-labelled antibody visualizes the analyte if bound on

the primary antibody linked to the bead. Absolute quantification was

achieved using a standard curve of each analyte.

Hen’s egg test Chorion-Allantois-Membrane (HET-CAM-Model).

The HET-CAM is a model to investigate toxicity of drugs and their

effect on vascularization (14). The nervous system of chicken

embryos lacks pain perception until day 11 of ontogenesis. Hence,

experiments carried out on the carefully-opened CAM up to day 11

do not require permission of ethical committees, making this assay

a good alternative to traditional animal experiments. Previously, we

have adopted this assay to grow tumour cells on CAM called TUM-

CAM (15), as was also done in the present study. Briefly, eggs

(pathogen-free; Vakzine Lohmann, Cuxhaven, Germany) were

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2872

positioned in a horizontally in a hatching incubator (37˚C, 65%

humidity; Ehret, Mahlberg, Germany) where they were

automatically rotated hourly. After 6 days, the top of the eggs was

punctured with a sterile needle, sealed with leukosilk tape (BSN

medical, Hamburg, Germany), and put in a vertical position for

24 h. Then, eggs were carefully opened and a silicon ring (diameter

.5 cm) was placed on the CAM. Subsequently, either 2×10

RAW264.7 or PDA6606 cells were suspended in Matrigel (Corning,

Wiesbaden, Germany) and added into silicon rings embedded onto

the CAM. The RAW264.7 “tumours” served as a control to rule out

weight differences, because of the individual growth behaviour of

mixed-population (co-culture) tumours. For co-culture, 1×10

RAW264.7 and 1×10

PDA6606 cells were mixed at 1:1 ratio and

implanted into the silicon ring on CAM. Either non-stimulated

(naïve, M0) or pre-stimulated (M2 polarized) RAW264.7 cells were

used. To generate pre-stimulated macrophages, the cells were

cultured in vitro for 72 h in PDA6606 cell culture supernatants that

was diluted 1:2 with fresh culture medium. This procedure induced

a M2 macrophage phenotype. The defect in the eggshell was closed

with adhesive tape, and cells were grown for 72 h on the CAM.

Microphotographs, magnetic resonance imaging (MRI), and single

cell analysis of tumours. At day 3 after implantation of cells, eggs

were put in egg-shaped, pre-warmed metal boxes and photographed

using a 10x magnification objective for resolution of vessel

structures. For MRI, eggs were cooled on ice for 5 min before data

acquisition to inhibit chicken movement. The CAM was scanned in

a high field 7.0 T MRI scanner for small animals (ClinScan, 7.0 T,

290 mT/m gradient strength; Bruker, Ettlingen, Germany) as

previously described (16). MRI scan analyses were performed in an

egg coil filled with ice (Bruker, Ettlingen, Germany) using a T2-

TSE (turbo spin echo) sequence. For tumour size assessment, high

resolution T2-weighted images of the horizontal plane were used.

Generated images were analysed using MIPAV (medical imaging

processing and visualisation; National Institutes of Health,

Bethesda, MD, USA). For single cell analysis via flow cytometry,

the tumours were explanted, weighted, disintegrated with surgical

scissors and digested in 95% accutase and 5% trypsin for 30 min at

37˚C. Afterwards, suspensions were passed through a 30 μm filter

and prepared for multicolour flow cytometry.

Statistical analyses. All experiments were independently repeated

2-4 times with several replicates each. Calculation of mean and

standard errors as well as graphing was performed using Prism 8.01

(Graphpad software, San Diego, CA, USA). Statistical comparison

between several groups was performed with one-way analysis of

variances (Anova). If all groups were tested against a single control

group, Dunnett’s post-hoc test was performed.

Results

RAW264.7 M0 macrophages can be polarized into M1 or M2

macrophages. It is known that RAW264.7 (RAW)

macrophages can be polarized into M1 or M2 macrophages

(17). To establish the protocol and positive controls for

subsequent assays, RAW cells were differentiated using

various factors, as described in materials and methods. Three

days after incubation, flow cytometry revealed differences in

the expression of CD68, iNOS, and FcεR surface markers

among M0, M1, and M2 macrophages (Figure. 1a-c).

Specifically, M1 macrophages were positive for iNOS

whereas M2 macrophages were positive for FcεR compared

o M0 and M2 or M1 macrophages, respectively (Figure 1d-

f). CD68 was mainly expressed in M1 macrophages, whereas

its expression in M2 macrophages varied among different

assays in our study. Hence, CD68 did not serve as an ideal

marker to distinguish M1 and M2 from M0 macrophages.

Microscopy of all macrophage phenotypes revealed a

significant tendency of M2 macrophages to cluster together

(Figure 1g-i) as shown in the quantitative image analysis

(Figure 1j). Altogether, our results showed that different

macrophage types can be discriminated based on

morphological and marker expression patterns.

Co-culture of RAW macrophages with pancreatic cancer

cells induces the M2 phenotype. Next, co-cultures were setup

between RAW macrophages and PDA6606 (PDA) pancreatic

cancer cells. By using fluorescently-labelled macrophages,

the clustering phenotype (Figure 2a) of M2 macrophages was

reproduced (Figure 1j). The extent of clustering was

dependent on both the ratio of PDA:RAW cells (Figure 2b)

as well as the incubation time (Figure 2c). Clustering was

proportional to the PDA:RAW ratio and incubation time (up

to 72 h). Clustering was also observed in chemically-induced

M2 macrophages but not in non-stimulated M0 macrophages

(Figure 2c), serving as a positive and negative control,

respectively. To further confirm that PDA cells induced a M2

phenotype in macrophages, co-cultures (PDA:RAW ratio

3:1) were analyzed by flow cytometry at 72 h. Staining for

CD206 and arginase showed a subtle and prominent

increase, respectively, of RAW cells in the co-cultures

compared to RAW monocultures (Figure 2c, d). This was

concomitant with elevated numbers of macrophages upon co-

culture (Figure 2e).

Secretory products of PDA6606 cells are sufficient to induce

M2 polarization in RAW macrophages. The next question

was whether direct cell-cell contact is necessary for PDA

cells to induce polarization of RAW cells into M2

macrophages. To this end, a transwell assay was setup where

both cell types were co-cultured inside one well but were

separated by a membrane that allows free diffusion of

soluble molecules but disallows physical interaction between

the cell types (Figure 3a). The expression levels of several

polarization markers were investigated on RAW cells

retrieved from co-cultures at 72 h by flow cytometry.

Increased expression of CD206 and FcεR, similar to that in

chemically-induced M2 macrophages, was observed. iNOS

expression was somewhat enhanced upon co-culture but this

was negligible on a relative scale considering the 200 fold

increase seen with chemically induced M1 macrophages.

These results suggested that direct cell-cell contact between

Khabipov et al: RAW264.7 Macrophage Polarization and Pancreatic Cancer

2873

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2874

Figure 1. Differentiation of RAW macrophages. (a-c) Flow cytometric overlay charts of macrophage differentiation markers in non-stimulated and

stimulated cells at day 3 of culture; (d-f) quantitative analysis of surface marker expression in M0, M1, and M2 macrophages 3d; (g-i) microscopic

images (5×) of macrophage cultures at day 3 of culture, note the cluster formation in M2 macrophages; (j) quantitative analysis of cluster formation

at day 3 of culture. Data are from 2-3 independent experiments with several replicates. Scale bar is 100 μm.

Khabipov et al: RAW264.7 Macrophage Polarization and Pancreatic Cancer

2875

Figure 2. Co-culture of RAW macrophages with pancreatic cancer cells reproduces the M2 phenotype. (a) Fluorescently-labelled macrophages in co-

culture with non-labelled PDA6606 cells imaged at various time points, note the cluster formation over time; (b) total cluster size at day 3 of culture

with different PDA-to-RAW ratios in co-culture; (c) cluster size of co-cultures (RAW+PDA at 1:3) over several days and cluster size of macrophages

and M2-stimulated macrophages (RAW) without PDA; (d) flow cytometry analysis of M2 surface markers CD206 and arginase in RAW mono and

RAW-PDA-co-cultures (at ratio 1:3) at day 3 of culture; (e) percentage of F4/80+ macrophages from all cells retrieved from co-cultures of RAW and

PDA (at ratio 1:3) on day 3 of culture. Data are from 2-3 independent experiments with several replicates. Scale bar is 50 μm.

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2876

Figure 3. Secretory products of PDA6606 cells are sufficient to induce M2 polarization in RAW macrophages. (a) Transwell co-culture of RAW

(sitting inside the transwell insert) and PDA (at well bottom) cells as well as M1 and M2 macrophages as control, and flow cytometry analysis of

RAW surface markers at day 3 of culture; (b) brightfield images (5×) of RAW macrophages cultured with PDA6606 supernatants (70%) at different

time points; (c) flow cytometry of M2 macrophage surface markers (normalized to unstimulated macrophages) of RAW cells incubated with

supernatants of PDA6606 cells for 3 days; (d) quantification of VEGF in supernatants of mono and co-cultures of RAW and PDA cells via ELISA;

(e) absolute concentrations of cytokines in PDA6606 supernatants at 18 h using multiplex bead-array. Data are from 2-3 independent experiments

with several replicates each. Scale bar is 100 μm. n.d.=None determined meaning the signal was below the kit-specific limit of detection.

Khabipov et al: RAW264.7 Macrophage Polarization and Pancreatic Cancer

2877

Figure 4. RAW macrophages spur angiogenesis and growth of tumours implanted on the CAM of chicken eggs. (a) Angiogenesis and matrix

remodelling macroscopically observed in PDA and PDA-RAW tumours; (b) macroscopic size of PDA tumours increased with addition of RAW

macrophages, especially if pre-stimulated with supernatants of PDA cells (upper panel), quantification of tumour volume via MRI (lower panel);

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2878

Figure 5. RAW macrophages co-cultured with PDA cells on chicken embryos display an M2

phenotype. (a) Flow cytometry dot plots of macrophages, macrophages and PDA cells, PDA-

supernatant-stimulated macrophages, and PDA cells grown on chicken embryos; (b) ratio of

F4/80

/CD11b

low

and F4/80

int

/CD11b

cells in different tissues; (c-g) flow cytometry analysis

of F4/80hi/CD11blow cells for several markers of macrophage polarization. Data are pooled from

tumours retrieved from three eggs per group. (h) and (i) are representative immunofluorescence

images of tissue sections retrieved from tumours (macrophages+PDA cells), scale bar is 100 μm.

RAW and PDA cells was not necessary for polarization of

M2 macrophages. To further support this notion, RAW

macrophages were incubated with cell culture supernatants

btained from PDA6606 cells. Again, cluster formation was

observed (Figure 3b), and flow cytometry confirmed the

upregulation of M2 markers on RAW macrophages (Figure

3c) in a concentration-dependent manner when cultured with

supernatants from PDA6606 cells. This effect was not

observed when RAW macrophages were incubated with

normal culture medium (served as normalization control=1).

To elucidate, which chemokines and/or cytokines and/or

growth factors were present in PDA6606 supernatants that

could potentially induce the phenotypic switch from M0 to

M2 macrophages, ELISA and multiplex bead array was

performed for quantification of 14 analytes in the cell culture

supernatant. High concentrations of VEGF (Figure 3d) were

observed in the supernatants of PDA monocultures and co-

cultures but not (or only at low levels) in RAW culture

supernatants. The fact that VEGF is a prominent anti-

inflammatory molecule, prompted us to screen for relevant

molecules that could potentially be responsible for

polarization of macrophages (Figure 3e). Significant levels

of MCP-1, KC, and anti-inflammatory TGFβ were identified

together with low levels of anti-inflammatory IL-10. In

summary, our results indicated that M0 to M2 polarization

of RAW264.7 macrophages was facilitated by secretory

products of PDA6606 pancreatic cancer cells, presumably

via a mix of anti-inflammatory chemokines, cytokines, and

growth factors.

PDA6606 cells educate RAW264.7 macrophages to spur

tumour growth and angiogenesis via M2 polarization. To

link our in vitro findings with functional consequences in a

living host, PDA6606 and/or RAW264.7 macrophages were

implanted onto CAMs of chicken embryos. Compared to

PDA6606 tumours implanted as monocultures, implantation

of co-cultures of PDA6606 cancer cells with RAW264.7

macrophages significantly spurred angiogenesis (Figure 4a).

This was macroscopically visible as an increased number of

thick blood vessels radially directed towards the co-culture

but not the monoculture tumour implants. In addition,

magnified images showed a denser vessel network at the

edge of co-culture implants compared to monoculture ones.

Moreover, matrix re-modelling, visible as grey areas, was

observed at the edges of tumours containing macrophages at

the tumour edges. This was also visible in RAW264.7 (data

not shown) but not PDA6606 monoculture tissue on the

CAM. In the latter case, the edges were abrupt and not

interacting intensively with the surrounding mesh, attributing

the finding of matrix remodelling to RAW cells. Comparing

the overall macroscopic size, co-culture tumours were also

larger although both mono- and co-culture tumours were

implanted at equally cell counts initially. Interestingly,

PDA6606 that were mixed with stimulated macrophages

(RAW264.7 cells cultured in 50% PDA-supernatant medium

to retrieve M2 macrophages) were even larger than co-

ulture tumours containing non-stimulated macrophages

(Figure 4b). This was underlined by the MRI results.

Absolute quantification of tumour volume calculated by MRI

revealed a significant increase in co-culture tumours with

stimulated macrophages compared to all other groups (Figure

4c). This finding was supported by the measurement of the

absolute weight of the tumours (Figure 4d). To characterize

the phenotype of macrophages within these different

conditions, some of the tumours were digested to retrieve

single cell suspensions for subsequent flow cytometric

analysis. Because this protocol is prone to loss of cells, we

had to pool the three tumours (replicates) into a single

sample (containing all events of the three replicates) to

retrieve sufficient cell numbers per condition and provide

robust statistics. Nonetheless, we were able to clearly

identify RAW264.7 cells via the expression of CD11b and

also of F4/80 (Figure 5a). For subsequent quantification of

phenotypic marker molecules, the geometric mean

fluorescent intensity (gMFI) was used (Figure 5c g) as it is

less prone to outliers (that occur more likely in samples with

lower cell numbers) compared to the MFI. Comparing the

ratios of F4/80

over F4/80

int

RAW264.7 cells, the co-

culture tumours with non-stimulated tumours were found to

be increased (Figure 5b). However, the F4/80 expression in

the F4/80

population was much lower in this sample

compared to the other two (Figure 5a), possibly pointing to

only a number of cells maturing into F4/80

macrophages.

For the subsequent marker analysis, only the F4/80

population was considered to be true macrophages, and gated

on. The results regarding CD206 (Figure 5d) and FcεR

(Figure 5e) corroborated our in vitro findings indicating

increased expression of these markers in RAW macrophages

upon co-culture with PDA6606 cells or incubation with their

supernatants. The expression pattern of CD115 was distinct

in each treatment group (Figure 5c). iNOS expression was

also slightly increased (Figure 5f); however, we feel that this

is insignificant keeping in mind that M1 macrophages

express iNOS up to 200-fold above that of M0 macrophages

(Figure 3a). A small but consistent increase in the expression

of Ly6C was seen in the co-culture tumours compared to the

mono-cultures (Figure 5g). These data illustrated not only

the ability of (polarized) macrophages to spur angiogenesis

but also the potency of tumour-conditioned macrophages

(here referred to as M2, but generally also referred to as

tumour-promoting or tumour-associated macrophages) to

significantly spur tumour growth. This was supported by

immunofluorescence indicating that CD11b+ macrophages

not only induced formation of blood vessels (Figure 5h) but

also seemed to cluster in ovo while releasing pro-angiogenic

VEGF (Figure 5i).

Khabipov et al: RAW264.7 Macrophage Polarization and Pancreatic Cancer

2879

Discussion

Macrophages are described as cells with high plasticity and

ifferentiation potential (18). As tumour associated

macrophages with M2 properties (19), they play an

important role in tumour progression and metastasis

development in several cancerous diseases (20-22), such as

pancreatic carcinoma (23-25). Our previous findings (13) as

well as other studies (26, 27) have shown the ability of

PDA6606 pancreatic carcinoma cells to alter macrophage

differentiation to their favour. Differentiation into the pro-

tumour M2 subtype has been shown both in vitro and in vivo.

Furthermore, M2 macrophages have been shown to influence

tumour progression by promoting angiogenesis (28-30) and

matrix remodelling (31). Hence, there is a need to study

induction of tumour-promoting macrophage phenotype and

their plasticity. In this study, we have used for the first time

an in vitro and in ovo model system to study the interaction

of macrophages and pancreatic cancer cells that may present

a suitable addition to traditional rodent model systems or

primary cells obtained from rodents.

Flow cytometry was used to analyse intra-tumour

RAW264.7 macrophages in ovo following their isolation from

PDA6606 tumours and staining for the macrophage-specific

antigen CD11b. This method also allowed us to compare the

differentiation status of in ovo-stimulated RAW264.7 cells with

that of RAW264.7 cells stimulated in vitro using established

methods such as supernatant assay, trans-well assay, and co-

cultivation with cancer cells. Our results indicated that

differentiation to M2 macrophages was positively correlated

with the intensity of their interaction with PDA6606 cells.

However, as in previous studies (32-34), cell-to-cell contact

was not found to be a prerequisite for the induction of the M2

phenotype. In this study, high levels of VEGF were found in

PDA cell culture supernatants. As a functional consequence,

both pre-stimulated as well as non-stimulated macrophages,

increased angiogenesis and matrix re-modelling in co-culture

tumours in ovo, which is a typical trait of TAMs of the M2

subtype (30, 35, 36). Yet, angiogenesis in PDA tumours was

considerably less prominent, ruling out the possibility that

VEGF released by PDA6606 is responsible for the increased

angiogenesis. It is possible that VEGF released by PDA cells

is part of a feed-back loop that exists between PDA and

macrophages. Moreover, only pre-stimulated macrophages

significantly increased tumour weight in comparison to PDA

monoculture tumours, while naive macrophages did not.

Similar results have been obtained by a group using tumour

educated macrophages in a murine in vivo model (37).

Unfortunately, polarization of macrophages was not tested in

that study. It is known that tumour growth is limited by a finite

supply of oxygen and nutrients caused by insufficient

vascularization. Our hypothesis is that pre-stimulated (M2)

RAW macrophages were able to accelerate angiogenesis and

therefore nutrient supply, leading to increased tumour growth.

By contrast, tumour cells implanted as co-cultures with non-

stimulated (M0) macrophages first needed to polarize

macrophages, which then in a second step would accelerate

tumour growth via increasing vascularization and inducing

matrix re-modelling. However, this idea could not be tested in

our model as the assay lasts only for few days on CAM before

the embryos need to be euthanized. The importance of TAMs

is underlined by current efforts to target M2 into M1

macrophage polarization in a neo-adjuvant clinical setting (38-

40). This may even help preventing extensive peritoneal

metastasis, which is also supported by M2 macrophages (41).

Next to numerous potential targets reported throughout the

literature, chemokine secretion by tumour cells including

CCL3, CCL4, CCL22 (42), and potentially CCL2 as observed

in this and other studies (43-45), may serve as an additional

therapeutic avenue.

A more technical finding of our study was that the

clustering of RAW264.7 cells is a morphological correlate of

M2 differentiation. Clustering has not been described as an

M2 “trait”, yet and may be an additional and simple method

to recognize differentiation processes in RAW264.7

populations. The universality of this finding needs to be

confirmed in further morphological and molecular studies

using other monocyte/macrophage cell lines and cancer cells,

as well as primary cells. Our findings may be useful for

groups frequently using RAW264.7 cells to study

macrophage plasticity and differentiation (46-48).

Conclusion

Our study showed the suitability of the PDA6606 pancreatic

cancer cells and RAW264.7 macrophages co-culture

approach to study the differentiation of macrophages towards

the M2 phenotype and their role in tumour promotion.

Furthermore, the presented in ovo HET-CAM/TUM-CAM

assay is a rapid approach, that is not limited by ethical

concerns, used to test the functional consequences of

macrophage polarization in the context of living systems.

Conflicts of Interest

There are no conflicts of interest to declare regarding this study.

Authors’ Contributions

LIP and SB designed the study; AK and AK performed the

experiments; AK, AK, KRL, EF and SB analysed the data; SB

prepared the draft; all Authors reviewed the manuscript.

Acknowledgements

Technical support by Christine Hackbarth, Antje Janetzko, and Felix

Nießner is gratefully acknowledged. SB received financial support

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2880

by the Federal German Ministry of Education and Research, grant

number 03Z22DN11.

References

1 Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A and

Locati M: The chemokine system in diverse forms of

macrophage activation and polarization. Trends Immunol 25:

677-686, 2004. PMID: 15530839. DOI: 10.1016/j.it.2004.09.015

2 Martinez FO, Sica A, Mantovani A and Locati M: Macrophage

activation and polarization. Front Biosci 13: 453-461, 2008.

PMID: 17981560.

3 Sica A, Larghi P, Mancino A, Rubino L, Porta C, Totaro MG,

Rimoldi M, Biswas SK, Allavena P and Mantovani A:

Macrophage polarization in tumour progression. Semin Cancer

Biol 18: 349-355, 2008. PMID: 18467122. DOI: 10.1016/

j.semcancer.2008.03.004

4 Barkal AA, Weiskopf K, Kao KS, Gordon SR, Rosental B, Yiu

YY, George BM, Markovic M, Ring NG, Tsai JM, McKenna

KM, Ho PY, Cheng RZ, Chen JY, Barkal LJ, Ring AM,

Weissman IL and Maute RL: Engagement of MHC class I by the

inhibitory receptor LILRB1 suppresses macrophages and is a

target of cancer immunotherapy. Nat Immunol 19: 76-84, 2018.

PMID: 29180808. DOI: 10.1038/s41590-017-0004-z

5 Ashley JW, Hancock WD, Nelson AJ, Bone RN, Tse HM,

Wohltmann M, Turk J and Ramanadham S: Polarization of

macrophages toward M2 phenotype is favored by reduction in

iPLA2beta (Group VIA Phospholipase A2). J Biol Chem 291:

23268-23281, 2016. PMID: 27650501. DOI: 10.1074/

jbc.M116.754945

6 Mills CD: M1 and M2 Macrophages: Oracles of health and

disease. Crit Rev Immunol 32: 463-488, 2012. PMID: 23428224.

DOI: 10.1615/CritRevImmunol.v32.i6.10

7 Mantovani A, Schioppa T, Porta C, Allavena P and Sica A: Role

of tumor-associated macrophages in tumor progression and

invasion. Cancer Metastasis Rev 25: 315-322, 2006. PMID:

16967326. DOI: 10.1007/s10555-006-9001-7

8 Cook J and Hagemann T: Tumour-associated macrophages and

cancer. Curr Opin Pharmacol 13: 595-601, 2013. PMID:

23773801. DOI: 10.1016/j.coph.2013.05.017

9 Cassetta L and Kitamura T: Macrophage targeting: opening new

possibilities for cancer immunotherapy. Immunology 155: 285-

293, 2018. PMID: 29963704. DOI: 10.1111/imm.12976

10 Najafi M, Hashemi Goradel N, Farhood B, Salehi E, Nashtaei

MS, Khanlarkhani N, Khezri Z, Majidpoor J, Abouzaripour M,

Habibi M, Kashani IR and Mortezaee K: Macrophage polarity

in cancer: A review. J Cell Biochem 120: 2756-2765, 2019.

PMID: 30270458. DOI: 10.1002/jcb.27646

11 Bolli E, Movahedi K, Laoui D and Van Ginderachter JA: Novel

insights in the regulation and function of macrophages in the

tumor microenvironment. Curr Opin Oncol 29: 55-61, 2017.

PMID: 27792052. DOI: 10.1097/CCO.0000000000000344

12 Kurahara H, Shinchi H, Mataki Y, Maemura K, Noma H, Kubo

F, Sakoda M, Ueno S, Natsugoe S and Takao S: Significance of

M2-polarized tumor-associated macrophage in pancreatic cancer.

J Surg Res 167: e211-219, 2011. PMID: 19765725. DOI:

10.1016/j.jss.2009.05.026.

13 Partecke LI, Gunther C, Hagemann S, Jacobi C, Merkel M,

Sendler M, van Rooijen N, Kading A, Nguyen Trung D, Lorenz

E, Diedrich S, Weiss FU, Heidecke CD and von Bernstorff W:

Induction of M2-macrophages by tumour cells and tumour

growth promotion by M2-macrophages: a quid pro quo in

pancreatic cancer. Pancreatology 13: 508-516, 2013. PMID:

24075516. DOI: 10.1016/j.pan.2013.06.010

14 Steiling W, Bracher M, Courtellemont P and de Silva O: The

HET–CAM, a useful in vitro assay for assessing the eye

irritation properties of cosmetic formulations and ingredients.

Toxicol In Vitro 13: 375-384, 1999. PMID: 20654494. DOI:

10.1016/s0887-2333(98)00091-5

15 Partecke LI, Evert K, Haugk J, Doering F, Normann L, Diedrich

S, Weiss FU, Evert M, Huebner NO, Guenther C, Heidecke CD,

Kramer A, Bussiahn R, Weltmann KD, Pati O, Bender C and von

ernstorff W: Tissue tolerable plasma (TTP) induces apoptosis in

pancreatic cancer cells in vitro and in vivo. BMC Cancer 12: 473,

2012. PMID: 23066891. DOI: 10.1186/1471-2407-12-473

16 Partecke IL, Kaeding A, Sendler M, Albers N, Kuhn JP,

Speerforck S, Roese S, Seubert F, Diedrich S, Kuehn S, Weiss

UF, Mayerle J, Lerch MM, Hadlich S, Hosten N, Heidecke CD,

Puls R and von Bernstorff W: In vivo imaging of pancreatic

tumours and liver metastases using 7 Tesla MRI in a murine

orthotopic pancreatic cancer model and a liver metastases model.

BMC Cancer 11: 40, 2011. PMID: 21276229. DOI: 10.1186/

1471-2407-11-40

17 Lv R, Bao Q and Li Y: Regulation of M1type and M2type

macrophage polarization in RAW264.7 cells by Galectin9. Mol

Med Rep 16: 9111-9119, 2017. PMID: 28990062. DOI:

10.3892/mmr.2017.7719

18 Italiani P and Boraschi D: From Monocytes to M1/M2

Macrophages: Phenotypical vs. functional differentiation. Front

Immunol 5: 514, 2014. PMID: 25368618. DOI: 10.3389/

fimmu.2014.00514

19 Sica A, Schioppa T, Mantovani A and Allavena P: Tumour-

associated macrophages are a distinct M2 polarised population

promoting tumour progression: Potential targets of anti-cancer

therapy. Eur J Cancer 42: 717-727, 2006. PMID: 16520032.

DOI: 10.1016/j.ejca.2006.01.003

20 Kruse J, von Bernstorff W, Evert K, Albers N, Hadlich S,

Hagemann S, Gunther C, van Rooijen N, Heidecke CD and

Partecke LI: Macrophages promote tumour growth and liver

metastasis in an orthotopic syngeneic mouse model of colon

cancer. Int J Colorectal Dis 28: 1337-1349, 2013. PMID:

23657400. DOI: 10.1007/s00384-013-1703-z

21 Chen Y, Zhang S, Wang Q and Zhang X: Tumor-recruited M2

macrophages promote gastric and breast cancer metastasis via

M2 macrophage-secreted CHI3L1 protein. J Hematol Oncol 10:

36, 2017. PMID: 28143526. DOI: 10.1186/s13045-017-0408-0

22 Steenbrugge J, Breyne K, Demeyere K, De Wever O, Sanders

NN, Van Den Broeck W, Colpaert C, Vermeulen P, Van Laere S

and Meyer E: Anti-inflammatory signaling by mammary tumor

cells mediates prometastatic macrophage polarization in an

innovative intraductal mouse model for triple-negative breast

cancer. J Exp Clin Cancer Res 37: 191, 2018. PMID: 30111338.

DOI: 10.1186/s13046-018-0860-x

23 Cui R, Yue W, Lattime EC, Stein MN, Xu Q and Tan XL:

Targeting tumor-associated macrophages to combat pancreatic

cancer. Oncotarget 7: 50735-50754, 2016. PMID: 27191744.

DOI: 10.18632/oncotarget.9383

24 Caso R and Miller G: Role of tumor associated macrophages in

regulating pancreatic cancer progression. World J Immunol 6,

2016. DOI: 10.5411/wji.v6.i1.9

Khabipov et al: RAW264.7 Macrophage Polarization and Pancreatic Cancer

2881

25 Zhang A, Qian Y, Ye Z, Chen H, Xie H, Zhou L, Shen Y and

Zheng S: Cancer-associated fibroblasts promote M2 polarization

of macrophages in pancreatic ductal adenocarcinoma. Cancer

Med 6: 463-470, 2017. PMID: 28097809. DOI: 10.1002/

cam4.993

26 Chen PC, Cheng HC, Wang J, Wang SW, Tai HC, Lin CW and

ang CH: Prostate cancer-derived CCN3 induces M2

macrophage infiltration and contributes to angiogenesis in

prostate cancer microenvironment. Oncotarget 5: 1595-1608,

2014. PMID: 24721786. DOI: 10.18632/oncotarget.1570

27 Ruffell B, Affara NI and Coussens LM: Differential macrophage

programming in the tumor microenvironment. Trends Immunol

33: 119-126, 2012. PMID: 22277903. DOI: 10.1016/

j.it.2011.12.001

28 Tjiu JW, Chen JS, Shun CT, Lin SJ, Liao YH, Chu CY, Tsai TF,

Chiu HC, Dai YS, Inoue H, Yang PC, Kuo ML and Jee SH:

Tumor-associated macrophage-induced invasion and

angiogenesis of human basal cell carcinoma cells by

cyclooxygenase-2 induction. J Invest Dermatol 129: 1016-1025,

2009. PMID: 18843292. DOI: 10.1038/jid.2008.310

29 Linde N, Gutschalk CM, Hoffmann C, Yilmaz D and Mueller

MM: Integrating macrophages into organotypic co-cultures: a

3D in vitro model to study tumor-associated macrophages. PLoS

One 7: e40058, 2012. PMID: 22792213. DOI: 10.1371/journal.

pone.0040058

30 Squadrito ML and De Palma M: Macrophage regulation of

tumor angiogenesis: implications for cancer therapy. Mol

Aspects Med 32: 123-145, 2011. PMID: 21565215. DOI:

10.1016/j.mam.2011.04.005

31 Alonso-Nocelo M, Raimondo TM, Vining KH, Lopez-Lopez R,

de la Fuente M and Mooney DJ: Matrix stiffness and tumor-

associated macrophages modulate epithelial to mesenchymal

transition of human adenocarcinoma cells. Biofabrication 10:

035004, 2018. PMID: 29595143. DOI: 10.1088/1758-5090/aaafbc

32 Caras I, Tucureanu C, Lerescu L, Pitica R, Melinceanu L, Neagu

S and Salageanu A: Influence of tumor cell culture supernatants

on macrophage functional polarization: in vitro models of

macrophage-tumor environment interaction. Tumori 97: 647-

654, 2011. PMID: 22158498. DOI: 10.1700/989.10726

33 Sawa-Wejksza K, Dudek A, Lemieszek M, Kalawaj K and

Kandefer-Szerszen M: Colon cancer-derived conditioned

medium induces differentiation of THP-1 monocytes into a

mixed population of M1/M2 cells. Tumour Biol 40:

1010428318797880, 2018. PMID: 30183516. DOI:

10.1177/1010428318797880

34 Juusola M, Mustonen H, Vainionpaa S, Vaha-Koskela M,

Puolakkainen P and Seppanen H: The effect of pancreatic cancer

patient derived serum on macrophage M1/M2 polarization.

Pancreas 47: 1397-1397, 2018. PMID: WOS:000449304600137.

DOI: 10.1016/j.pan.2018.05.396

35 Jetten N, Verbruggen S, Gijbels MJ, Post MJ, De Winther MP

and Donners MM: Anti-inflammatory M2, but not pro-

inflammatory M1 macrophages promote angiogenesis in vivo.

Angiogenesis 17: 109-118, 2014. PMID: 24013945. DOI:

10.1007/s10456-013-9381-6

36 Bielenberg DR and Zetter BR: The contribution of angiogenesis

to the process of metastasis. Cancer J 21: 267-273, 2015. PMID:

26222078. DOI: 10.1097/PPO.0000000000000138

37 Menen RS, Hassanein MK, Momiyama M, Suetsugu A, Moossa

AR, Hoffman RM and Bouvet M: Tumor-educated macrophages

promote tumor growth and peritoneal metastasis in an orthotopic

nude mouse model of human pancreatic cancer. In Vivo 26: 565-

569, 2012. PMID: 22773569.

38 de Groot AE and Pienta KJ: Epigenetic control of macrophage

polarization: implications for targeting tumor-associated

macrophages. Oncotarget 9: 20908-20927, 2018. PMID:

9755698. DOI: 10.18632/oncotarget.24556

39 Rodell CB, Arlauckas SP, Cuccarese MF, Garris CS, Li R,

Ahmed MS, Kohler RH, Pittet MJ and Weissleder R: TLR7/8-

agonist-loaded nanoparticles promote the polarization of tumour-

associated macrophages to enhance cancer immunotherapy. Nat

Biomed Eng 2: 578-588, 2018. PMID: 31015631. DOI:

10.1038/s41551-018-0236-8

40 Liu L, He H, Liang R, Yi H, Meng X, Chen Z, Pan H, Ma Y and

Cai L: ROS-Inducing Micelles Sensitize Tumor-Associated

Macrophages to TLR3 Stimulation for Potent Immunotherapy.

Biomacromolecules 19: 2146-2155, 2018. PMID: 29669207.

DOI: 10.1021/acs.biomac.8b00239

41 Qian BZ and Pollard JW: Macrophage diversity enhances tumor

progression and metastasis. Cell 141: 39-51, 2010. PMID:

20371344. DOI: 10.1016/j.cell.2010.03.014

42 Murdoch C, Giannoudis A and Lewis CE: Mechanisms

regulating the recruitment of macrophages into hypoxic areas of

tumors and other ischemic tissues. Blood 104: 2224-2234, 2004.

PMID: 15231578. DOI: 10.1182/blood-2004-03-1109

43 Li X, Yao W, Yuan Y, Chen P, Li B, Li J, Chu R, Song H, Xie

D, Jiang X and Wang H: Targeting of tumour-infiltrating

macrophages via CCL2/CCR2 signalling as a therapeutic

strategy against hepatocellular carcinoma. Gut 66: 157-167,

2017. PMID: 26452628. DOI: 10.1136/gutjnl-2015-310514

44 Mantovani A, Allavena P, Sozzani S, Vecchi A, Locati M and

Sica A: Chemokines in the recruitment and shaping of the

leukocyte infiltrate of tumors. Semin Cancer Biol 14: 155-160,

2004. PMID: 15246050. DOI: 10.1016/j.semcancer.2003.10.001

45 Fridlender ZG, Kapoor V, Buchlis G, Cheng G, Sun J, Wang LC,

Singhal S, Snyder LA and Albelda SM: Monocyte

chemoattractant protein-1 blockade inhibits lung cancer tumor

growth by altering macrophage phenotype and activating CD8+

cells. Am J Respir Cell Mol Biol 44: 230-237, 2011. PMID:

20395632. DOI: 10.1165/rcmb.2010-0080OC

46 Liu CY, Xu JY, Shi XY, Huang W, Ruan TY, Xie P and Ding JL:

M2-polarized tumor-associated macrophages promoted

epithelial-mesenchymal transition in pancreatic cancer cells,

partially through TLR4/IL-10 signaling pathway. Lab Invest 93:

844-854, 2013. PMID: 23752129. DOI: 10.1038/labinvest.

2013.69

47 Gironella M, Calvo C, Fernandez A, Closa D, Iovanna JL,

Rosello-Catafau J and Folch-Puy E: Reg3beta deficiency impairs

pancreatic tumor growth by skewing macrophage polarization.

Cancer Res 73: 5682-5694, 2013. PMID: 23867474. DOI:

10.1158/0008-5472.CAN-12-3057

48 Prakash H, Nadella V, Singh S and Schmitz-Winnenthal H:

CD14/TLR4 priming potentially recalibrates and exerts anti-

tumor efficacy in tumor associated macrophages in a mouse

model of pancreatic carcinoma. Sci Rep 6: 31490, 2016. PMID:

27511884. DOI: 10.1038/srep31490

Received April 9, 2019

Revised May 6, 2019

Accepted May 15, 2019

ANTICANCER RESEARCH 39: 2871-2882 (2019)

2882