Prostaglandin E3 attenuates macrophage‐associated inflammation and prostate tumour growth by modulating polarization

Abstract Alternative polarization of macrophages regulates multiple biological processes. While M1‐polarized macrophages generally mediate rapid immune responses, M2‐polarized macrophages induce chronic and mild immune responses. In either case, polyunsaturated fatty acid (PUFA)‐derived lipid mediators act as both products and regulators of macrophages. Prostaglandin E3 (PGE3) is an eicosanoid derived from eicosapentaenoic acid, which is converted by cyclooxygenase, followed by prostaglandin E synthase successively. We found that PGE3 played an anti‐inflammatory role by inhibiting LPS and interferon‐γ‐induced M1 polarization and promoting interleukin‐4‐mediated M2 polarization (M2a). Further, we found that although PGE3 had no direct effect on the growth of prostate cancer cells in vitro, PGE3 could inhibit prostate cancer in vivo in a nude mouse model of neoplasia. Notably, we found that PGE3 significantly inhibited prostate cancer cell growth in a cancer cell‐macrophage co‐culture system. Experimental results showed that PGE3 inhibited the polarization of tumour‐associated M2 macrophages (TAM), consequently producing indirect anti‐tumour activity. Mechanistically, we identified that PGE3 regulated the expression and activation of protein kinase A, which is critical for macrophage polarization. In summary, this study indicates that PGE3 can selectively promote M2a polarization, while inhibiting M1 and TAM polarization, thus exerting an anti‐inflammatory effect and anti‐tumour effect in prostate cancer.

In the synthesis of PGE 2 , arachidonic acid (AA) is first converted to prostaglandin H 2 by the enzymes COX-1 and COX-2, and then cellspecific prostaglandin synthases convert prostaglandin H 2 into various prostaglandins, including PGI 2 , PGF 2α , PGD 2 and PGE 2 . Prostaglandins E 2 and E 3 (PGE 2 and PGE 3 ) have received the most attention for their roles in modulating inflammation. 4-6 PGE 2 has been well-characterized and is known to play important roles in the regulation and activity of T lymphocytes. [7][8][9] It primarily acts through four G protein-coupled receptor subtypes of EP receptors, EP1, EP2, EP3 and EP4. 10,11 In addition, recent studies indicated that PGE 2 can promote M2 polarization of macrophages in vivo through the EP4 receptor and its downstream cyclic AMP (cAMP)-protein kinase A (PKA) pathway, which activates cAMP-responsive element-binding protein (CREB) and Kruppel-like factor 4. 12 PGE 2 and PGE 3 share the same metabolic pathway and have high structural similarity. However, unlike AA-derived PGE 2 , eicosapentaenoic acid (EPA)-derived PGE 3 has not been widely studied. PGE 3 is widely presumed to be anti-inflammatory and anti-neoplastic.
It is thought that PGE 3 can antagonize PGE 2 to yield anti-tumour or anti-inflammatory effects, but there is a lack of supporting evidence. 13 Moreover, there are few reports of PGE 3 modulation of macrophage phenotypes. Existing studies suggest that PGE 3 shares the same EP receptors with PGE 2 , although with different potencies. 10,14,15 Based on these studies, we investigated whether PGE 3 could engage this pathway to alternatively activate macrophages.
We explored the role of PGE 3 in macrophage polarization and found that PGE 3 selectively promoted M2a polarization, while inhibiting M1 and TAM polarization. Therefore, it exerted an inhibitory effect on prostate cancer inflammation. This process depended on PGE 3 -induced up-regulation of the PKA pathway. and LNCaP cells were cultured in RPMI 1640 medium (Gibco) sup-
Once the cells were adherent, they were transferred to PMA-free media for 24 hours (M0). These cells were then polarized to M1-like macrophages by incubation with 100 ng/mL LPS (Sigma-Aldrich) and 20 ng/mL IFNγ (R&D Systems) for 72 hours or M2a-like macrophages by incubation with 20 ng/mL IL-4 (R&D Systems) for 72 hours. At the same time, the cells were treated with or without PGE 3 (100 nmol/L, Cayman Chemical).
Bone marrow-derived macrophage (BED) isolation and cultivation were performed as described by the Cold Spring Harbor Protocols. 17 Briefly, femur and tibia bones were collected from 6to 8-week-old C57BL6/J mice, and then bone marrow cells were flushed out using phosphate-buffered saline (PBS) supplemented with 2% heat-inactivated FBS. After the red blood cells were lysed with red blood cell lysis buffer, the cells were cultured in BMDM growth medium for 7 days, followed by analysis of the purity of the cell population. BMDM growth medium contains 30% supernatant of L929 cells, which provides macrophage colony-stimulating factor.

| Co-culture and differentiation of TAMs and conditioned medium preparation
To obtain M0 macrophages, THP-1 cells were seeded at a density of 1 × 10 5 cells per well in a six-well culture plate and treated with 100 nmol/L PMA for 24 hours as previously described. Next, M0 cells in the six-well plates were co-cultured with PC3 cells that had been left to attach to the cell culture inserts for 12 hours before coculture. The cells were co-cultured in RPMI 1640 supplemented with 10% FBS for 7 days and treated with or without PGE 3 (100 nmol/L).
The medium was replaced every 2 days. After 7 days, the medium was collected as the conditioned medium (CM-control).

| RNA extraction and quantitative real-time PCR (qPCR)
Total RNA was extracted using TRIzol (Cat#15596026, Invitrogen) according to the manufacturer's protocol. Reverse transcription was performed according to the manufacturer's protocol with the Prime Script ® RT reagent Kit with gDNA Eraser (Cat#PR047A, Takara). Real-time PCR was performed on a CFX96 Real-Time System (Bio-Rad Laboratories) using SYBR Green PCR master mix (Applied Biosystems) and the respective primer pairs for each gene (sequences presented in Table 1). All experiments were performed in triplicate, and data were normalized against beta-actin.
The relative mRNA expression of various genes is presented as the fold-change, which was determined using the 2 -ΔΔCT method as previously described. 18

| Flow cytometry
Cells were detached using 0.05% trypsin (Gibco) and re-suspended in PBS. Cell suspensions were then stained with anti-CD206 or CCR7 monoclonal antibodies for 30 minutes at 4°C after FcγRII/III blocking.
Flow staining buffers and antibodies were purchased from different companies, with detailed information listed in Table 2. Isotype-matched controls were included in all experiments. Stained cells were analysed on an Attune NxT flow cytometer (Thermo Fisher Scientific).

| Enzyme-linked immunosorbent assay (ELISA)
Sandwich ELISAs were performed to detect cytokines in the serum and MPMs of LPS-stimulated mice. ELISA kits for IL-6 and IFNγ were obtained from R&D Systems. The serum was directly used for ELISAs in 96-well plates without dilution. MRMs were diluted to 1 × 10 7 cells/mL in PBS, sonicated, and centrifuged; the supernatant was used for ELISA. All procedures were performed according to the manufacturer's protocol. Cytokine concentrations were calculated using to standard curves.  Data were calculated as the OD570-OD630, and cell numbers were reported as percentages compared to the control.

| Western blotting
Protein samples were separated by 10% SDS-PAGE and then transferred to polyvinylidene difluoride membranes (MA01821, Millipore).
The membranes were blocked for 2 hours in 5% skim milk and Horseradish peroxidase-labelled goat anti-mouse or anti-rabbit IgG secondary antibodies were used for ECL detection (WBKLS0500, Sigma-Aldrich).

| Statistics
All data were analysed using GraphPad Prism 6 software. All data are shown as the means ± SD. One-way analysis of variance was performed to determine the significance among three or more groups followed by the indicated post hoc tests. t test was used to analyse two independent samples. P < .05 was considered as statistically significant.

| Prostaglandin E 3 suppresses M1 markers and induces M2a marker expression in polarized macrophages
Prostaglandins are potent lipid molecules affecting vital aspects of immunity. 19 We predicted that PGE 3 affects the immune microenvironment. We used THP-1 cells as an in vitro macrophage model. 16

| PGE 3 has anti-inflammatory effects in vivo
We next evaluated the role of PGE 3 during acute inflammation in vivo.
Mice 6-8 weeks old were divided into 3 different groups (6 in each group) and injected intraperitoneally with saline, LPS or LPS + PGE 3 .
After 24 hours, PGE 3 significantly reduced MPMs with an M1-like phenotype and increased M2-like MPMs ( Figure 4A). Additionally, PGE 3 inhibited inflammatory cytokine secretion in both the serum and MPM cell lysates ( Figure 4B). These results indicate that PGE 3 can significantly (P < .05) modulate macrophage polarization towards an anti-inflammatory function.

| Prostaglandin E 3 suppresses TAM polarization and enhances the phagocytosis of TAM
TAMs (also known as M2d) reside in the tumour microenvironment promote tumour cell migration and invasion; these cells are a unique subtype of M2 macrophages. 21 Although PGE 3 promoted M2a polarization, the role of PGE 3 in TAM polarization is unclear.
TAMs were induced by co-culturing THP-1 cells with PC3 prostate cancer cells. The canonical M2 macrophage marker CD206 was upregulated in THP-1-derived TAMs ( Figure 5A). qPCR also revealed down-regulation of COX2 and up-regulation of VEGF and EGF in THP-1 TAM cells compared to in M0 cells ( Figure 5C). In contrast, PGE 3 promoted the expression of COX2 and down-regulated the expression of CD206, VEGF and EGF ( Figure 5B, D, and E), suggesting that PGE 3 suppresses TAM polarization.
As inducing phagocytosis of macrophages is a therapeutic strategy in clinic, 22 we also performed the phagocytosis assay to better corroborate the effect of PGE 3 on TAM. Results showed that PGE 3 treatment enhanced the phagocytosis ( Figure 5F and G), suggesting that PGE 3 might reduce tumour growth by increasing TAM phagocytosis.

| Prostaglandin E 3 suppresses prostate tumour cell proliferation
To investigate the role of PGE 3 in tumorigenesis, we performed tumour transplantation experiments and found that PGE 3 decreased the weights of transplanted tumours after 1 month of continuous PGE 3 injection ( Figure 6A). As TAMs always show M2-like phenotypes, we measured CD68 (a total macrophage marker) and CD206 (an M2-like macrophage marker) expression in the tumours and observed that both CD68 and CD206 expression can be inhibited by PGE 3 ( Figure 6B).
These results suggest that PGE 3 has antitumorigenic effects by regulating macrophage polarization in vivo. Next, we tested whether PGE 3 also inhibits the proliferation of cancer cells in vitro in a co-culture system with THP-1 cells. CM with or without PGE 3 (CM/PGE 3 ) was collected ( Figure 6C) and used in cell proliferation assays ( Figure 6D).

| Prostaglandin E 3 influences macrophage polarization through the PKA pathway
Normally, prostaglandins affect target cells by activating their cognate receptors EP1, EP2, EP3 or EP4. Expression of EP1, EP2 and EP4 was detectable by RT-PCR in THP-1 cells ( Figure S2A). EP1 is coupled to G q/p and induces PKC activation by mobilizing intracellular calcium. EP2 and EP4, however, are coupled with G s and induce PKA by up-regulating cAMP. 19 Previous studies showed that EP4 can activate protein kinase B, also known as AKT. 23 We detected the expression of PKA, PKC and AKT and their phosphorylation states by Western blotting. The results are shown in Figure S2B. To determine which EP subtype(s) was (were) involved in the polarization of macrophages, we did a series of experiments using receptor antagonists.
Results showed that EP4 mediated the PGE 3 -induced macrophage polarization (Figure 7 and Figure S3).
Further analyses showed that both AKT and p-AKT (Ser473 phosphorylation) did not differ in M0, M2a and TAMs, even when stimulated with PGE 3 (Figure 8A However, these results were not affected by PGE 3 treatment of TAMs ( Figure 8A, B).
As our results showed that PGE 3 up-regulated the expression of PKA in both M1 and M2a cells, we predicted that PGE 3 influences macrophage polarization via PKA. To test this mechanism, we added a potent and selective PKA inhibitor (H-89 dihydrochloride) during macrophage polarization and found that the immunomodulatory effects of PGE 3 on macrophages were inhibited by H-89 dihydrochloride ( Figure 8C, D).  31,32 It is reported that the 3-series prostaglandins tend to have anti-inflammatory and anti-tumour effects. 10,33,34 In our research, we found that Prostaglandin E 3, as a classic 3-series prostaglandins, can regulate the polarization of M1 and M2a macrophages by inhibiting M1 and promoting M2a macrophage polarization, supporting its proposed anti-inflammatory effects.
Macrophages are a major component of tumour-infiltrating inflammatory cells, 21,35 and TAMs contribute to tumour progression at different levels. 36 Previous studies reported that some anti-inflammatory and anti-cancer effects of EPA are mediated by PGE 3 production, 13 Given that prostaglandins are natural ligands of peroxisome proliferator-activated receptor (PPAR) γ, which plays a crucial role in the development of inflammation 46,47 and differentiation of macrophage, 48 we determined the effect of PPARγ activation on macrophage polarization. Our results showed that the activation of PPARγ (induced by Rosiglitazone) had little effect on the polarization of macrophage induced by PGE 3 ( Figure S4). However, PPARγ may have anti-inflammation effect by reducing inflammatory factors produced by macrophage.

| CON CLUS IONS
Our findings reveal a novel mechanism through which eicosanoids can influence tumorigenesis. Specifically, PGE 3 can influence the macrophage polarization to resolve inflammation and inhibit prostate tumour growth.

CO N FLI C T O F I NTE R E S T
The authors declared that they have no conflicts of interest to this work.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support this study are available from the corresponding author upon request.