Proteasome Inhibitor Immunotherapy for the Epithelial to Mesenchymal Transition: Assessing the A549 Lung Cancer Cell Microenvironment and the Role of M1, M2a and M2c ‘Hydrocortisone-Polarised’ Macrophages

Lung cancer is a leading cause of cancer-related deaths, primarily as a result of metastases. In this metastasis, the epithelial-to-mesenchymal transition (EMT) is essential. Interaction with the cancer cell microenvironment is primarily dependent on M1- and M2-polarized macrophage. In this study, we revealed the EMT-associated activity of M1, M2a and M2c macrophages in A549 lung cancer cells. We established a co-culture model of A549 lung cancer cells utilizing THP-1-derived M1/M2 polarised macrophages to explore the involvement of macrophages in the immune response, apoptosis, and EMT in lung cancer. Although multiple polarising agents are routinely used for M1 and M2 conversion, we assessed a new possible polarising agent, hydrocortisone. M1 increased A549 cell sensitivity to proteasome inhibitors and decreased A549 cell viability by inducing apoptosis. EMT was induced in the presence of M2c macrophages in A549 cells by the levels of vimentin, bronectin, E-cadherin, NF-kB, CCL-17. We also revealed the antiproliferative effects of bortezomib and ixazomib on A549 cells in both 2D and 3D cultures. Our ndings could help develop an immunotherapeutic strategy by shedding light on a previously undiscovered part of the EMT pathway. Furthermore, additional investigation may reveal that polarising tumour-associated macrophages to M1 and eliminating the M2a or particularly the M2c subtype are effective anti-cancer strategies.


Introduction
Cancer is de ned as uncontrolled cell proliferation caused by genetic and epigenetic disorders, and it is the world's second most common cause of death after cardiovascular system disorder. The 'hallmarks of cancer' include cell death resistance, the ability to evade immunological surveillance, triggered angiogenesis, invasion and metastasis activation, and others [1,2]. The primary goal of cancer research today is to discover different early detection technologies, as well as to investigate the molecular pathways of lung tumour progression, particularly speci c mechanisms of metastasis [3].
Non-cancer cells and tumor tissue structures are de ned by the microenvironment, and tumor-associated macrophages make up 5-40% of the tumor mass in solid tumors. [4,5]. Tumour-associated macrophages are central in the relationship between chronic in ammation conditions in cancer [6,7] and play an vital role in the progression of tumors, epithelial-mesenchymal transition (EMT), angiogenesis, cancer metastases, and drug resistance [4].
EMT is a morphological transformation of cells from an epithelial-polarized phenotype to a mesenchymal broblastoid phenotype that plays a critical role in cancer metastasis [8]. In the development process, a balanced epithelial-mesenchymal transition is essential, while abnormal EMT contributes to pathogenesis by disrupting epithelial homeostasis for diseases such as cancer [9]. The nuclear factor kappa B (NF-kB) is a signi cant marker of cancer invasion and EMT [10,11]. NF-kB is a transcription factor that regulates gene-encoding molecules such as chemokines, apoptosis, surface receptors in the immune response, cytokines, cell adhesion molecules, and in ammation [12]. Several therapeutic approaches have been proposed in recent years, focusing on the NF-kB axis targeting tumour invasion.
We utilised 2D and 3D cell culture approaches to evaluate the effects of bortezomib and ixazomib, recently developed therapeutic strategies, on antiproliferative, apoptotic and EMT pathways in A549 lung cancer co-cultures that included THP-1-derived M1, M2a and M2c polarised macrophages. This is the rst article of M2c macrophage polarization by hydrocortisone that we are aware of. Polarisation of macrophages into M1, M2a and M2c subtypes THP-1 cells were treated with 100 ng/mL PMA in serum-free medium (SFM) for 24 hours to induce differentiation into a macrophage-like phenotype. To obtain the macrophage (M0) resting state, differentiated adherent cells were rested in the culture medium (without the addition of PMA, but with 10% FBS and 1% penicillin-streptomycin). for another 48 hours. It was discovered that macrophages polarized into M1, M2a, or M2c after being exposed to 20 ng/mL LPS plus IFN-for 24 hours, 25 ng/mL IL-4 plus IL-13 for 72 hours, and 1 M hydrocortisone for 72 hours. To remove carry-over cytokines, LPS, and hydrocortisone, polarised macrophages were thoroughly washed and incubated in fresh media for 24 hours. The expression of cell surface markers CD80, CD-206, CD23 and CD163 (BioLegend, San Diego, CA, USA) were used to decide the macrophage subspecies using a BD Accuri C6 ow cytometry. The mRNA expression levels of some genes (IL-1β, IL-8, IL-18, CXCL-1, CXCL-3, CCL-22, CCL-24, TGF-β) were determined using a real-time polymerase chain reaction (RT-PCR) method to show the polarisation of M0, M1, M2a and M2c subtypes. Finally, macrophages M0, M1, M2a and M2c were photographed with the Leica DM inverted light microscope to assess morphology.

Chemicals
WST-1 cytotoxicity test WST-1 assay (Roche) was used to determine cell viability. M0 macrophages were cultured at a density of 1 x 10 4 cells per well into 96-well plates. Cells were treated with bortezomib and ixazomib at different concentrations for 24 and 48 hours. After incubation, wells were lled with WST-1 cell proliferation reagent (10 µl/well) and absorbances (420 nm) were measured after 3 hours. The absorbance of the samples was measured using a Cytation 3 Multi-mode reader. (BioTek, Winooski, VT, USA). The number of living cells was directly proportional to the absorbance, and cell viability was expressed as a percentage of controls [13,14].
Proliferation assay by real-time cell analysis system (RTCA DP) The RTCA DP system monitors cell behavior in real-time by measuring changes in impedance of attached cells and providing a cell index (CI) value. During the experiment days, the impedance of the wells for each group was measured every hour and the impedance of the wells for each concentration was measured every hour during the experiment days and de ned as a CI value. We searched at cell proliferation as well as the IC 50 concentrations of bortezomib and ixazomib. First, the background of the 16-well E-plates (Roche Applied Sciences, Indianapolis, IN, USA) was measured in 100 µl RPMI 1640 medium by the RTCA DP. The cells were then seeded at a density of 1 × 10 4 per well in 16-well E-plates in 100 µl medium. After 24 hours of incubation, the impedance of each well was measured at 1-hour intervals using the xCELLigence RTCA DP Instrument (ACEA Biosciences, San Diego, CA, USA). Lastly, the instrument was stopped, and 100 µl of medium was removed from the wells containing bortezomib and ixazomib at several concentrations. All assays were performed in 8 wells, and the average values were calculated. Cell proliferation and IC 50 concentrations of bortezomib and ixazomib were monitored and evaluated for 48 h according to the CI values by RTCA DP Software 1.2.1 [15,16] Co-culture procedures THP-1 polarised macrophages were co-cultured with A549 lung cancer cells in indirect contact via transwell inserts (Corning, NY, USA) to investigate the impact of M1, M2a and M2c on cancer cell sensitivity to proteasome inhibitors. The A549 cells were co-cultured with M1, M2a, and M2c macrophages in six-well plate cell culture inserts with a 0.4-m porous membrane separating the upper and lower chambers, allowing for the exchange of soluble factors but not cell transmigration. To ensure consistency between experiments, THP-1 cells were plated in the upper chamber at a 5:1 ratio to the number of A549 cells plated in the lower chamber. The THP-1 cells were seeded into the transwell apparatus's upper chamber, stimulated to differentiate into M0 macrophages by adding 100 ng/mL PMA for 24 h, washed with PBS, and incubated for another 48 h to eliminate the effect of PMA. The attached cells, which corresponded to M0 macrophages, were polarised into M1, M2a, or M2c macrophages by adding 20 ng/mL LPS plus IFN-γ for 24 h, 25 ng/mL IL-4 plus IL-13 for 72 h, and 1 µM hydrocortisone for 72 h. THP-1 monocyte differentiation for M1, M2a, and M2c macrophages was started on different days to obtain differentiated and polarised macrophages on the same day. A549 cells were seeded in the lower chamber 24 hours before macrophage polarisation ended. The upper chambers containing the M1, M2a, or M2c macrophages were then placed directly on top of the A549 cells in the six-well plates. The two cell populations were incubated for 48 hours with bortezomib and ixazomib IC 50 concentrations added directly into the wells. Apoptosis detection using Annexin V-FITC and propidium iodide (PI) staining A549 cells (2x10 5 /well) were seeded in 6-well plates 24 hours before the end of macrophage polarisation process. A549 cells and macrophages were co-cultured in SFM at a cell density of 1:5. A549 cells and M1, M2a or M2c co-culture groups were incubated with bortezomib and ixazomib IC 50 concentrations for 48 hours. A549 cell analysis was conducted according to the Annexin V-FITC Apoptosis Detection Kit procedure, used to detect apoptosis as described previously [13]. Finally, the samples were diluted with 250 mL of the binding buffer, processed for data acquisition, and analysed on a BD Accuri™ C6 Flow Cytometry; at least 1 x 10 4 cells were analysed per sample.

Detection of caspase-3 activity
Caspase-3 is an essential protease that initiates apoptosis [17]. Alterations of the caspase-3 level of cells were determined using the PE Active Caspase-3 Apoptosis Kit (BD Pharmingen, Catalogue Number: 550914). A549 cells (2x10 5 ) were seeded in 6-well plates 24 hours before the end of polarisation polarisation process. A549 cells and M1, M2a or M2c co-culture groups were incubated with bortezomib and ixazomib IC 50 concentrations for 48 hours. After incubation, the samples were analyzed using BD Accuri™ C6 Flow Cytometry. At least 1 x 10 4 cells were analysed per sample.
Determination of loss of mitochondrial membrane potential by JC-1 dye using ow cytometry We examined the loss of mitochondrial membrane potential in response to ixazomib and bortezomib in cells by the Flow Cytometry Mitochondrial Membrane Potential Detection Kit (BD Pharmingen, Catalogue Number: 551302). A549 cells (2x10 5 ) were seeded in 6-well plates 24 h before the end of polarisation. A549 cells and M1, M2a or M2c co-culture groups were incubated with bortezomib and ixazomib IC 50 concentrations for 48 h. After incubation, the samples were analyzed using BD Accuri™ C6 Flow Cytometry. At least 1 x 10 4 cells were analysed per sample [13,17].
Cytokine detection A549 cells (2 x10 5 ) were seeded in 6-well plates 24 hours before the end of polarisation. A549 cells and M1, M2a or M2c co-culture groups were incubated with bortezomib and ixazomib IC 50 concentrations for 48 h. At the end of the incubation, supernatants were collected from co-culture. The BD human in ammatory kit procedure was used to measure TNF-and IL-10 cytokine levels (BD™ Cytometric Bead Array [CBA] Human In ammatory Cytokines Kit, Catalogue Number: 551811). An FCAP array was used in a BD Accuri C6 ow cytometer for CBA analysis.
Quantitative RT-PCR Total RNA was extracted from M1, M2a, or M2c macrophages to determine the expression of M1 and M2 macrophage markers, from A549 lung cancer cells to assess single-cell control, and from co-cultured A549 cells and macrophages to determine the change in genes in the EMT pathway. RNA was extracted using a MagNA Pure Compact RNA Isolation Kit for the MagNA Pure Compact Instrument LC 2.0 system (Roche Diagnostics, Mannheim, Germany). Reverse transcription was performed using a High-Fidelity cDNA Synthesis Kit (Roche, Germany). A NanoDrop spectrophotometer was used to verify the RNA samples' purity. For cDNA synthesis, 500 ng of RNA were taken from each RNA population. Real-time PCR was performed using the SYBR Green Master Mix (Roche Applied Science, Germany) on a LightCycler® 480 instrument (Roche Applied Science, Germany). The internal control was glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Table 1 shows the primer sequences that were used. The results were normalized to GAPDH expression levels and presented as averages of three experiments.
Transforming Growth Factor Beta TGF-Beta

Immuno uorescence staining of vimentin and E-cadherin
Vimentin and E-cadherin levels in the A549 control and M2a and M2c co-culture groups were analysed by ow cytometry (Accuri C6, BD). The M2c macrophage co-culture groups found to be more signi cant, vimentin and E-cadherin levels were investigated by immuno uorescence staining in the Cytation 3 Cell Imaging Multimode Reader. An Immuno uorescence Application Solutions Kit (Cell Signaling) was used for the xation, permeabilisation and immunostaining of antibodies to determine the levels of E-cadherin and vimentin in the cells. The vimentin antibodies (vimentin [D21H3] XP® rabbit mAb, Cell Signaling) were diluted 1:100 and E-cadherin monoclonal antibody (E-cadherin [4A2] mouse mAb, Cell Signaling) was diluted 1:50 in the same solution. The cells were incubated with 100 µl of antibody overnight at 4°C. After incubation, cells were rinsed twice with PBS and secondary antibody Anti-rabbit Alexa Fluor 594 (anti-rabbit IgG [H+L], F(ab')2 Fragment, Cell Signaling) at a 1:500 dilution. Secondary antibodies (antimouse Alexa Fluor 488, ab150113, Abcam) were added to the wells and incubated for 1 hour at room temperature. Hoechst 33258 (10 µg/ml) was used to stain the cell nuclei (blue) 10 minutes before imaging with Cytation 3 Cell Imaging Multi-Mode Reader (Bio-Tek, USA) [18].
3D culture studies

Culturing cells with AlgiMatrix
In 96-well 3D alginate scaffold plates, cells (2x10 4 /well) were incorporated into a 3D alginate scaffold in 100 mL of media to determine the optimal time for the formation of spheroids. After 20 minutes, a further 100 mL of media was added, and the cells were grown in an incubator at 37°C with 5% CO 2 for another 20 minutes. Every 48 hours, the media was changed. The formation of tumor spheroids in 3D alginate scaffold wells was observed at 3, 5, and 7 days after cell seeding to assess growth. Images were taken using an inverted microscope (Leica DM 300), and the size of spheroids was determined by the Leica LAS Analysis program. Also, cell spheroids were imaged by Hoechst uorescent dye using BioTek Cytation 3 Imaging Multi-mode Reader.

Comparison of 2D and 3D cytotoxicity data of bortezomib and ixazomib
Based on the preliminary experiments, the more abundant spheroid formation was observed in the 7-day culture. As a result, 7 day was chosen as the starting point for treatment. Bortezomib and ixazomib at concentrations ranges of 0.1-400 µM were used for the treatment. After 48 h incubation, for WST-1 viability determination, sponges were dissolved using an AlgiMatrix Dissolving Buffer, which provides a quick and straightforward way to recover cells from the AlgiMatrix bioscaffold. After 3 hours, wells were lled with the cell proliferation reagent WST-1 and absorbances were measured. 2D culture studies were also conducted in parallel with this experiment. IC 50 values were then calculated for both 2D and 3D cultures using MS Excel.

Statistical Analysis
The results were analyzed using one-way ANOVA and Tukey's post hoc test in Graphpad Prism 6.0. The means of three independent experiments are expressed as mean standard deviation (±). The signi cance of the results in comparison to the control group is represented by P values (P>0.05 n.s, P<0.05*, P<0.01**, P<0.001***, and P<0.0001****).

Results
Monocytic THP-1 cells differentiation into M0 macrophages THP-1 cells grew in suspension as single cells or partially in clusters and had a wide and round morphology ( Figure A1). THP-1 cells were differentiated into macrophages after 24 hours of stimulation with 100 ng/mL PMA. During differentiation, the cells attached to the ask bottom changed shape from round to almost spindle-shaped, and pseudopods appeared in some cells. Due to the fact that all macrophage populations express CD68, different PMA concentrations and incubation times were used to determine differentiation. Based on our ow cytometry results, we chose a 24-hour incubation with 100 ng/mL PMA, followed by a 48-hour rest period (CD-14 [5.5%] and the CD-68 [58.2%] levels were determined, data not shown).

THP-1 polarisation into pro-in ammatory M1 macrophages
In the involvement of IFN-and in combination with in ammatory stimuli such as LPS, macrophages polarize to the M1 phenotype, according to numerous studies [19][20][21][22][23]. Based on these data, we tested various concentrations of LPS (20 ng/mL or 1µg/mL), combined with 20 ng/ml of IFN-γ, and incubated THP-1 macrophages for 24 h. Macrophage M1 polarisation was assessed by measuring the mRNA expression of M1 markers, pro-in ammatory cytokines IL-1β, IL-8, IL-18, CXCL1 and CXCL3, using RT-qPCR and CD80 membrane receptors level by ow cytometry. Based on our CD80 results, 20 ng/mL IFN-γ + 20 ng/mL LPS was selected for M1 polarisation to be applied during the 24-hour incubation time ( Figure B1). An increased pro-in ammatory marker (IL-1β, IL-8, IL-18, CXCL1 and CXCL3) expression pro le was obtained by incubation with IFN-γ combined with 20 ng/mL LPS in comparison to M0 macrophage ( Figure C1). Additionally, we examined the mRNA expression of some M2 markers (TGF-Beta, CCL22, and CCL24) in M1 macrophages and found that these genes were not signi cantly expressed. Finally, incubating THP-1 macrophages (M0) for 24 hours with IFN-20 ng/ml and LPS 20 ng/mL causes them to polarize into M1 macrophages.

Antiproliferative effects of bortezomib and ixazomib determined by RTCA DP
RTCA DP detects the cell number indirectly by measuring electrical impedance and generates real-time data by collecting measurements at predetermined intervals over a period of days [14][15][16]. In our study, the RTCA DP system was used for the real-time and time-dependent analysis of IC 50 concentrations of A549 and CCD-19Lu cells. The WST-1 method was used to select the appropriate concentrations of bortezomib and ixazomib for evaluation in the xCELLigence system (data not shown). Figures A2, B2, C2 and D2 show that bortezomib and ixazomib decreased the viability of A549 and CCD-19Lu cells in a dose-dependent manner. Also, to compare the cytotoxic effects of the bortezomib and ixazomib on A549 lung cancer and CCD-19Lu healthy lung broblast cells, their IC 50 values were calculated after 48 h exposure. The IC 50 value determined for bortezomib and ixazomib in A549 lung cancer cells was found to be lower than the IC 50 value determined for CCD-19Lu cells. IC 50 concentrations of bortezomib and ixazomib (45 nM and 2 µM, respectively) were determined and applied to macrophages during the co-culture process, and macrophage viability was determined by the WST-1 method. Bortezomib and ixazomib had no signi cant cytotoxic effect on M0 macrophages at 24 and 48 h incubation ( Figure E2).

M1 macrophages reduced A549 cell viability and proliferation while increasing drug sensitivity to bortezomib and ixazomib.
Numerous studies demonstrate that interactions with tumor-associated macrophages can in uence cancer cell proliferation and chemosensitivity [23]. There is no other study using macrophage polarization co-culture and a real-time cell analysis system in the literature. The E-plate insert allows for real-time cellcell interactions while keeping cells apart. Proliferation results were determined more precisely because A549 cells were located in the same culture medium as the E-plate insert used in co-cultures with M1, M2a or M2c polarised macrophages and their proliferation was measured in real-time ( Figure A3, B3, C3, D3). A549 and M1 macrophage co-culture groups signi cantly decreased A549 cell viability and proliferation relative to the A549 control ( Figure A3). Also, the results showed that the antiproliferative effects of bortezomib and ixazomib were higher in the A549 and M1 macrophage co-culture groups ( Figure B3). We used annexin V/PI staining, caspase-3 activation, and mitochondrial membrane potential assays to determine whether the reduced cell number of A549 cells exposed to M1 was due to apoptosis. M1 co-culturing for 48 hours induced apoptosis in A549 cells, according to annexin V/PI staining. When A549 cells were incubated with M1 macrophages in the lack of drugs, there was an increase in apoptosis. Notably, bortezomib and ixazomib induced apoptosis most frequently in the M1 macrophage co-culture group ( Figure E3). When A549 cells were incubated with M1 macrophages, the level of caspase-3 was higher than when the cells were incubated alone. M1 macrophages also signi cantly increased caspase-3 activity relative to the control group incubated with bortezomib and ixazomib without macrophages. M1 macrophages also increased caspase-3 activity in A549 cells treated with bortezomib and ixazomib ( Figure F3). The observed loss of mitochondrial depolarisation also supports these ndings ( Figure G3). When A549 cells were incubated with M2a or M2c macrophages, the apoptosis of cancer cells did not change compared to the control group. Furthermore, JC-1 depolarisation was much less common in cells incubated with M2, particularly M2a. However, M2a and M2c macrophages increased the A549 cell apoptosis induced by bortezomib and ixazomib for all three apoptosis-related experiment results ( Figure  E3, F3 and G3). Overall, these ndings show that M1, M2a, and M2c macrophages in A549 lung cancer cells modulated the apoptotic response to proteasome inhibitors bortezomib and ixazomib. M1 macrophages alone inhibited cell proliferation and increased drug-induced apoptosis. In the absence of drugs, however, M2 macrophages were ineffective or reduced apoptosis in A549 cells.

M2c macrophages induced the EMT of lung A549 cancer cells
To determine whether the EMT pathway was activated, TNF-α and IL-10 cytokine levels were determined from A549 cell supernatants and M1, M2a and M2c polarised macrophage co-culture groups using ow cytometry. IL-10 cytokines increased in co-culture groups formed by both M2a and M2c polarised macrophages and A549 cells, and there was a signi cant increase in the M2c macrophage co-culture group (p<0.01**) ( Figure A4). mRNA expression levels of EMT-related factors (vimentin, bronectin, Ecadherin, NF-kB, VEGF and CCL-17) in A549 cells were evaluated in a co-culture model of M1, M2a and M2c polarised macrophages. The most important mesenchymal factors, vimentin and bronectin, increased, and E-cadherin decreased in both A549 M2a and M2c macrophage co-culture groups. Also, no signi cant changes in vimentin, bronectin, E-cadherin, NF-kB, VEGF and CCL-17 mRNA expression were found in the M1 macrophage groups. mRNA expression levels of CCL-17 did not change signi cantly in the M1 polarised macrophage co-culture group, but they decreased in the M2a macrophage group compared to the A549 control group (p<0.05*). A signi cant increase (p<0.0001****) in the M2c macrophage co-culture group was distinct as it indicated that EMT increased the most in the M2c polarised macrophage co-culture group ( Figure B4). Based on these ndings, it was determined that EMT was induced, in particular, in M2c macrophage A549 co-culture groups. This result is supported by the increased vimentin and decreased E-cadherin levels determined by ow cytometry ( Figure C4).
Proteasome inhibitors have also been found to reduce EMT by decreasing vimentin, NFkB, VEGF and CCL-17 expression levels. Especially for ixazomib, the reduction of EMT was statistically more signi cant than bortezomib ( Figure D4). In addition, vimentin and E-cadherin in M2c co-culture groups, as imaged by immuno uorescence staining images ( Figure E4), show an increase in vimentin (red) and a decrease in Ecadherin (green) monoclonal antibody levels in the A549 and M2c co-culture group compared to the A549 control group. Finally, it was determined that bortezomib and, more so, ixazomib decreased M2c-induced EMT in A549 lung cancer cells.

Optimisation of In Vitro 3D A549 Lung Cancer Tumor Model
A 3D culture environment was created using the AlgiMatrix® 3D Culture System. Based on the type of cells incorporated, the spongy features of the AlgiMatrix® system provides the needed place for cells to grow as models of in vitro solid tumors include lung cancer [28]. Figure A5 shows spheroid areas and morphological images of randomly selected A549 cells on days 3, 5 and 7. The graph generated by the average spheroid area (µm 2 ) was obtained from A549 cell images taken on different days ( Figure B5). The spheroid area increased statistically on day 7 compared to days 3 and 5 (p <0.0001****).
Comparative analysis of IC 50 values of bortezomib and ixazomib in 2D and 3D systems 3D cultures are known for accurately replicating the physiologic microenvironment and having a high level of harmony with in vivo conditions. In our study, we aimed to determine IC 50 concentrations for proteasome inhibitors to shed light on in vivo models by taking advantage of 3D culture. The IC 50 values for bortezomib and ixazomib were signi cantly higher in AlgiMatrix systems than in 2D culture models.
After 48 hours of A549 incubation with bortezomib and ixazomib in 2D and 3D culture, the IC 50 value in 3D culture was approximately 15 times higher for bortezomib and 9 times higher for ixazomib than in 2D culture ( Table 2).

Discussion
The initiation and metastasis of lung cancers depend not only on the genetic and molecular characteristics of the cancer cells but also on their interaction with the immune system and the tumour microenvironment [29]. In solid tumours, TAMs comprise 5-40 percent of the total tumor mass in most cases. Furthermore, there is usually a link between the number of TAMs and the speci c condition, depending on the type of tumor [5]. In response to a stimulus in their environment, macrophages can be differentiated into M1 or M2 macrophages. Recent research suggests that macrophage polarization may be used as a therapeutic strategy in the treatment of cancer and angiogenesis. [4,23]. Previous studies have mostly focused on M1 macrophages, and no subtype distinction has been observed in M2 macrophages. There are limited data available on M2c macrophages. In this study, we found that macrophages with M1, M2a, and especially M2c polarization had different effects on lung cancer cell apoptosis, the EMT pathway, and gene expression pro les.
Notably, we used hydrocortisone for the rst time in M2c polarisation by focusing more on M2 macrophages. M2 macrophages are polarised to M2c macrophages by different stimuli. Although M2c macrophages are reported to be polarised with IL-10, TGF-and glucocorticoids [30,31], no studies to date have reported polarisation using the glucocorticoid hydrocortisone. We optimised hydrocortisone polarisation of M2c for the rst time in this study, identifying many surface markers (CD206 and CD163) and cytokines and observing chemokine gene expression (IL-1 β, IL-8, IL-18, CXCL-1, CXCL-3, CCL-22, CCL-24, TGF-β).
Studies have shown that M1 macrophages promote to the inhibition of tumour growth and increase the sensitivity of chemotherapy agents [23,27]. We found that M1 polarised macrophages alone decreased A549 cell viability using the real-time cell analysis system. In addition, M1 macrophages alone affected and increased the bortezomib and ixazomib-induced apoptosis. Proteasome inhibitors have previously been reported to cause apoptosis in cancer cells [13,17,32,33]. Although the apoptotic effect of ixazomib on lung cancer cells has not been previously investigated, its apoptotic effects have been observed on many other solid tumour cell lines (colon, prostate, hepatocellular carcinoma cells) [17,34,35].
EMT is a process in which cells morphologically transition from the epithelial phenotype to the mesenchymal broblastoid phenotype and plays an important role in the cancer metastasis process [8].
TNF-α, released by M1 macrophages, is an important cytokine that modulates in ammatory responses in the microenvironment of many tumours [36]. Although studies have reported many factors, such as EGF, TGF-β, TNF-α, generally found in the tumour microenvironment, induce EMT [37], more recent studies have also reported that TNF-α inhibits EMT by suppressing TGF-β levels [38]. Furthermore, the TLR4/IL-10 axis has been implicated in the EMT signaling pathway, and inhibition of effectively suppresses EMT induction. [39]. In our results, decreased TNF-α and increased IL-10 levels in A549 cells and M2a and M2c polarised macrophage co-culture are responsible for increased EMT in A549 cells. Also, EMT was induced in A549 cancer cells in co-culture groups formed with M2a and M2c macrophages, depending on the expression levels of vimentin, E-cadherin, bronectin, NF-kB, VEGF and CCl-17 mRNA. An increase in TGFβ mRNA expression is a particularly relevant result because studies have shown that TGF-β is an important factor for EMT and can be an EMT-inducing agent [38,40]. According to our results, TGF-β may play a role in the induction of EMT in A549 cells with co-culture of M2a and M2c macrophages.
Factors such as CCL-17 derived from tumour-associated macrophages are important contributors to immunosuppression. Studies have shown that this causes and intensi es various stages of cancer progression [41]. The high levels of CCL-17 mRNA expression levels found only in the M2c macrophage and A549 co-culture group are informative.
NF-kB is a key transcriptional factor responsible for regulating the in ammatory signal and is important in cellular proliferation and differentiation. It's a key marker for cancer invasion and EMT, and it's involved in both the induction and support of EMT. As a result, it has been proposed that NF-kB activity decreases the expression of genes associated with EMT [10,11,42]. We observed, bortezomib and ixazomib 'wellknown NF-KB inhibitors' decreased M2c-induced EMT.
E-cadherin is recognised as the main marker of epithelial morphology [43], and E-cadherin expression usually decreases during EMT, while the levels of mesenchymal speci c marker vimentin increase [44]. We determined vimentin and E-cadherin mRNA expression and monoclonal antibody levels correlated to the increase in EMT in the A549 and M2c macrophage co-cultures were more signi cant than M2a macrophage groups.

Conclusion
M2c macrophages promote EMT in lung cancer cells, according to our ndings. M1 macrophages, on the other hand, reduce the proliferation and viability of A549 cancer cells and induce apoptosis in those cells.
The observed inhibition of M2c-induced EMT in A549 cells exposed to bortezomib and ixazomib has provided insight into an unknown aspect of proteasome inhibition of the EMT pathway, which is the current therapeutic approach. This is the rst study to show that hydrocortisone polarizes M2c macrophages, to our knowledge. Our ndings suggest that in lung cancer patients taking glucocorticoidrelated drugs, the M1/M2 macrophage amount in the tumor microenvironment should be investigated. Finally, polarising tumour-associated macrophages to M1 and eliminating M2a or particularly M2c macrophages, after additional investigation, might represent effective future anti-cancer therapy strategies.

Declarations
Data availability: All data generated or analysed during this study are included in this published article (and its supplementary information les).

Compliance with Ethical Standards
Funding: This study was funded by Anadolu University Scienti c Research Projects (grant number 1704S095.).
Con ict of interest: Author Selin Engür-Öztürk declares that she has no con ict of interest. Author Miriş Dikmen declares that she has no con ict of interest.
Ethical approval: This article does not contain any studies with human participants or animals performed by any of the authors. M2c macrophages, respectively, and the marker CD206 is common to both M2 subtypes. Experiments were performed in triplicate (C1) Expressions of some M1/M2 markers in differentiated macrophage subtypes, determined by RT-PCR. The gene expression of differentiated macrophage subtypes was normalized to that of untreated M0 macrophages. p<0.05*, p<0.01**, p<0.0001****, (mean ± SD, n = 3).    hoechst 33258 on day 7 (10X). (B5) Spheroid sizes obtained from cell images of A549 cells taken on different days (n = 25, mean ± standard deviation, p<0.0001 ****) Random photos were taken for the analysis of the areas of the spheroids. A total of 25 photographs were selected among these photographs for each day, including 5 photographs with at least 5 spheroids.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. certi cate1309644.pdf