Downregulation of IL-8 and IL-10 by LRRC8A Inhibition through the NOX2–Nrf2–CEBPB Transcriptional Axis in THP-1-Derived M2 Macrophages

M2-polarized, tumor-associated macrophages (TAMs) produce pro-tumorigenic and angiogenic mediators, such as interleukin-8 (IL-8) and IL-10. Leucine-rich repeat-containing protein 8 members (LRRC8s) form volume-regulated anion channels and play an important role in macrophage functions by regulating cytokine and chemokine production. We herein examined the role of LRRC8A in IL-8 and IL-10 expression in THP-1-differentiated M2-like macrophages (M2-MACs), which are a useful tool for investigating TAMs. In M2-MACs, the pharmacological inhibition of LRRC8A led to hyperpolarizing responses after a transient depolarization phase, followed by a slight elevation in the intracellular concentration of Ca2+. Both the small interfering RNA-mediated and pharmacological inhibition of LRRC8A repressed the transcriptional expression of IL-8 and IL-10, resulting in a significant reduction in their secretion. The inhibition of LRRC8A decreased the nuclear translocation of phosphorylated nuclear factor-erythroid 2-related factor 2 (Nrf2), while the activation of Nrf2 reversed the LRRC8A inhibition-induced transcriptional repression of IL-8 and IL-10 in M2-MACs. We identified the CCAAT/enhancer-binding protein isoform B, CEBPB, as a downstream target of Nrf2 signaling in M2-MACs. Moreover, among several upstream candidates, the inhibition of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 2 (NOX2) suppressed the Nrf2–CEBPB transcriptional axis in M2-MACs. Collectively, the present results indicate that the inhibition of LRRC8A repressed IL-8 and IL-10 transcription in M2-MACs through the NOX2–Nrf2–CEBPB axis and suggest that LRRC8A inhibitors suppress the IL-10-mediated evasion of tumor immune surveillance and IL-8-mediated metastasis and neovascularization in TAMs.


Introduction
Leucine-rich repeat-containing 8A (LRRC8A) is a component of volume-regulated anion channels (VRACs) and contributes to regulatory volume decreases [1].VRACs are heteromeric chloride ion (Cl − ) channels comprising an essential LRRC8A and at least one other LRRC8 member (LRRC8B-E).The intracellular concentration of Cl − varies among cell types and affects the functional outcome of transmembrane Cl − movement through Cl − channels and transporters by regulating the driving force for Cl − conductance [2].In cells that accumulate Cl − intracellularly, the activation of Cl − channels results in membrane depolarization through Cl − efflux [2,3].In immune cells, including macrophages, 2 of 21 membrane depolarization has been shown to decrease the driving force for Ca 2+ entry through voltage-independent Ca 2+ channels, subsequently suppressing intracellular Ca 2+ signaling [4].
M 2 -polarized, tumor-associated macrophages (TAMs) are mainly derived from monocytes and abundantly infiltrate solid tumors [5].The localization of TAMs in the tumor microenvironment (TME) correlates with poor clinical outcomes in solid tumors [5].TAMs represent up to 50% of the tumor mass in solid tumors and support the survival of cancer stem cells by inhibiting immunosurveillance [6].TAMs, mostly with M 2 -like characteristics in the TME, highly produce the anti-inflammatory cytokine, interleukin 10 (IL-10), cyclooxygenase 2 (COX2), and mannose and scavenger receptors (CD206 and CD163, respectively) and increase the release of ornithine and polyamine through the arginase (Arg) pathway [7,8].Macrophages differentiated from human monocytic leukemia THP-1 cells are used as human in vitro models to examine M 1 -and M 2 -polarized macrophages [9].The manipulation of TAM functions is a useful therapeutic approach in clinical practice for the prevention of tumor metastasis [10].
IL-10 is highly expressed in TAMs and contributes to the differentiation and recruitment of regulatory T cells (T reg s) in tumors [11].Additionally, IL-8 plays a critical role in the recruitment and accumulation of TAMs and other immunosuppressive cells into the TME, as well as in neovascularization, accelerating tumor proliferation and metastasis in the TME [12].Elevated numbers of IL-8-expressing TAMs are associated with higher clinical stages and an increased risk of a poor prognosis in cancer patients [13].We previously demonstrated that the transcription of IL-8 and IL-10 was critically repressed by the activation of the Ca 2+ -activated K + channel, K Ca 3.1, through the extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) signaling pathways in M 2 -MACs [14].
Nuclear factor-erythroid 2-related factor 2 (Nrf2) is known as a key regulator of oxidative stress, and Nrf2 dissociates from Kelch-like ECH-associated protein 1 (Keap1) and subsequently translocates into the nucleus.The Nrf2-Keap1 signaling pathway plays an important role in the regulation of oxidative stress in cancerous and non-cancerous cells [15,16].In macrophages, Nrf2 inhibits M 1 polarization and promotes M 2 polarization [17,18].It also upregulates the expression of M 2 markers such as CD163 and Arg1 [19].Liu et al. indicated that the inhibition of LRRC8A promoted macrophage phagocytosis by promoting the nuclear translocation of Nrf2 [20].The Nrf2-mediated regulation of IL-8 and IL-10 expression and secretion remains controversial [21][22][23][24][25].We previously reported the transcriptional repression of IL-8 and IL-10 by the pharmacological inhibition of Nrf2 in M 2 -MACs [14]; however, the underlying mechanisms have yet to be elucidated.
LRRC8A co-localizes with NOX isoforms, such as NOX2 and NOX4, in the plasma membrane and promotes the production of reactive oxygen species (ROS) via these subtypes [31].NOX2 signaling plays a crucial role in oxidative stress during differentiation into macrophages and in their expression and secretion of pro-and anti-inflammatory mediators [32,33].Previous studies indicated that Nrf2 was activated in macrophages through the NOX2 signaling pathway [34].
We herein examined the involvement of LRRC8A in the expression and secretion of IL-8 and IL-10 in M 2 -MACs.The results obtained provide novel mechanistic insights into the therapeutic potential of LRRC8A inhibitors in cancer immunotherapy.

Functional Expression of LRRC8A in M 2 -MACs
As previously reported, high expression levels of the M 2 marker genes CD163, Arg1, and C-C motif chemokine 22 (CCL22) in M 2 -MACs were confirmed in the present study [14].

Functional Expression of LRRC8A in M2-MACs
As previously reported, high expression levels of the M2 marker genes CD163, Arg1, and C-C motif chemokine 22 (CCL22) in M2-MACs were confirmed in the present study [14].We previously demonstrated that the activation of the Ca 2+ -activated K + channel KCa3.1 downregulated the pro-tumorigenic cytokine IL-10 and angiogenic chemokine IL-8/CXCL8 in M2-MACs through the ERK-cAMP response element-binding protein (CREB) and JNK-c-Jun cascades [14].We also reported that the inhibition of Nrf2 downregulated IL-10 and IL-8 in M2-MACs; however, the KCa3.1-mediatedregulation of these cytokines was not related to the Nrf2 signaling pathway [14].We herein examined the role of LRRC8A in the regulation of IL-10 and IL-8 in M2-MACs.As shown in Figure 1A, LRRC8A transcripts, but not the Ca 2+ -activated Cl -channel (anoctamin 1, ANO1), were highly expressed in M2-MACs.Similar results were obtained at the protein level by Western blotting and immunocytochemical imaging (Figure 1B,C).Among the other LRRC8 isoforms, LRRC8B and 8C transcripts were expressed at high levels (Figure S1A).Among the Cl - channels/transporters present in intracellular organelles, voltage-dependent anion channel 1 (VDAC1), chloride intracellular channel protein 1 (CLIC1), H + /Cl -exchanger transporter 3 (CLCN3), and CLCN7 transcripts were expressed at high levels (Figure S1B,C).The transcriptional expression levels of the isoforms [VDAC2, 3, CLIC2 to 6, K + -Cl -cotransporter 1 (KCC1) to 4, cystic fibrosis transmembrane conductance regulator (CFTR), and osteopetrosis-associated transmembrane protein 1 (Ostm1)] were less than 0.01 in arbitrary units (a.u., ratio to β-actin, ACTB) (n = 4 for each).Among the Cl -channel modulators expressing at the plasma membrane, chloride channel accessory 1 (CLCA1), CLCA2, CLCA4, and transmembrane protein 206 (TMEM206), the expression levels of CLCA4 and TMEM206 were approximately 0.05 a.u., and those of CLCA1 and CLCA2 were less than 0.01 a.u.(Figure S1D).Among the ANO isoforms (ANO1, 2, and 6) that function as Cl -channels, ANO6 transcripts were expressed at high levels (Figure S1E).We then examined the effects of the pharmacological blockade of LRRC8A with endovion (EDV) on membrane potential and [Ca 2+ ]i.The application of 1 μM EDV induced large membrane hyperpolarization, following slight membrane depolarization (Figure 2A,C,D).Correspondingly, the application of EDV for 10 min (min) induced a significant elevation in [Ca 2+ ]i in M2-MACs (Figure 2B,E,F).EDV is also known as an ANO1 inhibitor We then examined the effects of the pharmacological blockade of LRRC8A with endovion (EDV) on membrane potential and [Ca 2+ ] i .The application of 1 µM EDV induced large membrane hyperpolarization, following slight membrane depolarization (Figure 2A,C,D).Correspondingly, the application of EDV for 10 min (min) induced a significant elevation in [Ca 2+ ] i in M 2 -MACs (Figure 2B,E,F).EDV is also known as an ANO1 inhibitor (50% inhibitory concentration (IC 50 ) = 0.5 µM for LRRC8A; IC 50 = 0.7 to 2.0 µM for ANO1).Therefore, we examined the effects of the ANO1-selective inhibitor, ANO1-IN-1, on membrane potential in M 2 -MACs.As shown in Figure 2G,H, no significant responses were observed in M 2 -MACs following the application of 1 µM ANO1-IN-1.The

Decreased Expression and Secretion of IL-8 and IL-10 by the Small Interfering RNA (siRNA)-Mediated and Pharmacological Inhibition of LRRC8A in M2-MACs
We examined the effects of the siRNA-mediated inhibition of LRRC8A on the expression and secretion of IL-8 and IL-10 in M2-MACs.IL-8 and IL-10 mRNA levels were significantly reduced by the inhibition of LRRC8A (n = 4 for each, p < 0.01) (Figure 3A,C).Correspondingly, their secretion was significantly decreased by the inhibition of LRRC8A (n = 4 for each, p < 0.01) (Figure 3B,D).The inhibition efficacy of LRRC8A siRNA was approximately 65% (Figure S2A).We also examined the effects of treatment with EDV for 12 h (hr) on their expression and secretion in M2-MACs.IL-8 and IL-10 mRNA levels were significantly suppressed by the treatment with 10 μM EDV in a concentration-dependent manner (n = 4 for each, p < 0.05, 0.01) (Figure 3E,G).Their secretion was also significantly decreased by approximately 50% by treatment with EDV for 24 h (n = 4 for each, p < 0.01)  We examined the effects of the siRNA-mediated inhibition of LRRC8A on the expression and secretion of IL-8 and IL-10 in M 2 -MACs.IL-8 and IL-10 mRNA levels were significantly reduced by the inhibition of LRRC8A (n = 4 for each, p < 0.01) (Figure 3A,C).Correspondingly, their secretion was significantly decreased by the inhibition of LRRC8A (n = 4 for each, p < 0.01) (Figure 3B,D).The inhibition efficacy of LRRC8A siRNA was approximately 65% (Figure S2A).We also examined the effects of treatment with EDV for 12 h (hr) on their expression and secretion in M 2 -MACs.IL-8 and IL-10 mRNA levels were significantly suppressed by the treatment with 10 µM EDV in a concentration-dependent manner (n = 4 for each, p < 0.05, 0.01) (Figure 3E,G).Their secretion was also significantly decreased by approximately 50% by treatment with EDV for 24 h (n = 4 for each, p < 0.01) (Figure 3F,H).No significant changes were observed in their mRNA levels following the pharmacological inhibition of ANO1 and ANO6 (Figure S3) or the siRNA-mediated inhi-bition of VDAC1, CLIC1, CLCN3, and ANO6 (Figure S4).These results suggest that the inhibition of LRRC8A suppressed the pro-tumorigenic and pro-angiogenic functions of TAMs by repressing the secretion of IL-8 and IL-10.
(Figure 3F,H).No significant changes were observed in their mRNA levels following the pharmacological inhibition of ANO1 and ANO6 (Figure S3) or the siRNA-mediated inhibition of VDAC1, CLIC1, CLCN3, and ANO6 (Figure S4).These results suggest that the inhibition of LRRC8A suppressed the pro-tumorigenic and pro-angiogenic functions of TAMs by repressing the secretion of IL-8 and IL-10.

LRRC8A Inhibition-Induced Transcriptional Repression of IL-8 and IL-10 through the Nrf2 Signaling Pathway in M2-MACs
We previously reported that the pharmacological inhibition of Nrf2 with a selective inhibitor, ML385 (5 μM), repressed the transcriptional expression of IL-8 and IL-10 in M2-MACs [14].Similarly, their expression and secretion were significantly suppressed by the siRNA-mediated inhibition of Nrf2 in M2-MACs (Figure 4A-D).The inhibitory efficacy of Nrf2 siRNA was approximately 65% (Figure S2B).To elucidate whether the LRRC8A inhibition-induced downregulation of IL-8 and IL-10 was mediated through the Nrf2 signaling pathway, we examined the recovery effects of the Nrf2 activator, NK252, on the downregulation of IL-8 and IL-10 in M2-MACs.Co-treatment with NK252 (100 μM) and  We previously reported that the pharmacological inhibition of Nrf2 with a selective inhibitor, ML385 (5 µM), repressed the transcriptional expression of IL-8 and IL-10 in M 2 -MACs [14].Similarly, their expression and secretion were significantly suppressed by the siRNA-mediated inhibition of Nrf2 in M 2 -MACs (Figure 4A-D).The inhibitory efficacy of Nrf2 siRNA was approximately 65% (Figure S2B).To elucidate whether the LRRC8A inhibition-induced downregulation of IL-8 and IL-10 was mediated through the Nrf2 signaling pathway, we examined the recovery effects of the Nrf2 activator, NK252, on the downregulation of IL-8 and IL-10 in M 2 -MACs.Co-treatment with NK252 (100 µM) and EDV (10 µM) for 12 h reversed the EDV-induced downregulation of these cytokines in M 2 -MACs (n = 4, p < 0.01 vs. +/−) (Figure 4E,F).A significant increase was observed in IL-8 mRNA levels in M 2 -MACs by a single treatment with NK252 (n = 4, p < 0.01 vs. −/−);  To investigate whether the inhibition of LRRC8A reduced the nuclear translocation of P-Nrf2 in M2-MACs, the cellular localization of P-Nrf2 was visualized 2 h after the treatment with 10 μM EDV by confocal laser scanning fluorescence microscopy.The anti-P-Nrf2 antibody and nuclei were labeled with the Alexa Fluor 488-conjugated secondary antibody and 4′,6-diamidino-2-phenylindole (DAPI), respectively (Figure 5A).Images from more than 30 cells for each case (n = 1) were obtained, and the summarized results are expressed as the percentage of cells with P-Nrf2-positive nuclei.The percentage of these cells was significantly reduced in M2-MACs treated with EDV (n = 6, p < 0.01) (Figure 5A(a,b),B).As a positive control, they were significantly increased by the treatment with NK252 for 2 h (n = 6, p < 0.01) (Figure 5A(c),C).Nrf2 staining was detected at similar levels in the cytosolic regions of vehicle-(Figure 5D  To investigate whether the inhibition of LRRC8A reduced the nuclear translocation of P-Nrf2 in M 2 -MACs, the cellular localization of P-Nrf2 was visualized 2 h after the treatment with 10 µM EDV by confocal laser scanning fluorescence microscopy.The anti-P-Nrf2 antibody and nuclei were labeled with the Alexa Fluor 488-conjugated secondary antibody and 4 ′ ,6-diamidino-2-phenylindole (DAPI), respectively (Figure 5A).Images from more than 30 cells for each case (n = 1) were obtained, and the summarized results are expressed as the percentage of cells with P-Nrf2-positive nuclei.The percentage of these cells was significantly reduced in M 2 -MACs treated with EDV (n = 6, p < 0.01) (Figure 5A(a,b),B).As a positive control, they were significantly increased by the treatment with NK252 for 2 h (n = 6, p < 0.01) (Figure 5A(c),C).Nrf2 staining was detected at similar levels in the cytosolic regions of vehicle-(Figure 5D

Involvement of the NOX2-ROS Axis in the LRRC8A Inhibition-Induced Transcription Repression of IL-8 and IL-10 in M2-MACs
Previous studies reported that AKT and AMP-activated protein kinase (AMPK vated the Nrf2 downstream signaling pathway in stimulated RAW264.7 macroph [35,36].In the present study, the level of Nrf2 phosphorylation was high in M2-M (Figure 7A, left panels); however, the level of AKT phosphorylation was very low (F 7A, right panels).Correspondingly, no significant changes were observed in the ex sion levels of IL-8 and IL-10 transcripts in M2-MACs following a treatment with the inhibitor AZD5363 (2 μM) for 12 h (Figure 7B,C).Additionally, a significant decreas not detected in the level of AMPK phosphorylation following LRRC8A inhibition w μM EDV for 2 h (Figure 7D,E).AMPK is activated in macrophages in a Ca 2+ -depen manner [37]; however, [Ca 2+ ]i only slightly increased in M2-MACs treated with EDV ure 2B).The pharmacological inhibition of AMPK by treatment with Bay3827 (BA μM) for 12 h suppressed the expression and secretion of IL-8 and IL-10 and expressi CEBPB in M2-MACs (Figure S6).These results suggest that the inhibition of LRRC8 activated the Nrf2 downstream signaling pathway through a pathway independe AMPK.Corresponding to the results in Figure 7D,E, P  Previous studies reported that AKT and AMP-activated protein kinase (AMPK) activated the Nrf2 downstream signaling pathway in stimulated RAW264.7 macrophages [35,36].In the present study, the level of Nrf2 phosphorylation was high in M 2 -MACs (Figure 7A, left panels); however, the level of AKT phosphorylation was very low (Figure 7A, right panels).Correspondingly, no significant changes were observed in the expression levels of IL-8 and IL-10 transcripts in M 2 -MACs following a treatment with the AKT inhibitor AZD5363 (2 µM) for 12 h (Figure 7B,C).Additionally, a significant decrease was not detected in the level of AMPK phosphorylation following LRRC8A inhibition with 10 µM EDV for 2 h (Figure 7D,E).AMPK is activated in macrophages in a Ca 2+ -dependent manner [37]; however, [Ca 2+ ] i only slightly increased in M 2 -MACs treated with EDV (Figure 2B).The pharmacological inhibition of AMPK by treatment with Bay3827 (BAY, 10 µM) for 12 h suppressed the expression and secretion of IL-8 and IL-10 and expression of CEBPB in M 2 -MACs (Figure S6).These results suggest that the inhibition of LRRC8A inactivated the Nrf2 downstream signaling pathway through a pathway independent of AMPK.Corresponding to the results in Figure 7D,E, P Apart from AKT and AMPK, ROS also activates the Nrf2 downstream signaling pathway in macrophages [38].LRRC8A has been shown to activate NOX-dependent ROS production and physically binds to NOX1, 2, and 4 [39,40].As shown in Figure 8A, among the seven NOX isoforms, NOX2 transcripts were predominantly expressed in M 2 -MACs.NOX2 protein expression was observed in M 2 -MACs (Figure 8B).Correspondingly, the fluorescence signals of the Alexa 488-labeled anti-NOX2 antibody were observed along the plasma membrane of M 2 -MACs (Figure S8A(b)).With the same acquisition settings, very weak signals were observed in native THP-1 cells stained with the Alexa 488-labeled anti-NOX2 antibody and in native THP-1 cells and M 2 -MACs stained with the Alexa 488-labeled anti-NOX4 antibody (Figure S8A(a),B).GLX351322 (GLX) is a NOX4 inhibitor with IC 50 = 5 µM and also inhibits NOX2 with IC 50 = 40 µM [40].On the other hand, GSK2796039 (GSK) is a selective NOX2 inhibitor with IC 50 = 1 µM [41].No significant decrease in the expression of IL-8 and IL-10 transcripts was found following the inhibition of NOX4 in M 2 -MACs with 10 µM GLX for 12 h (Figure 8C,D); however, it was significantly decreased by approximately 50% following the inhibition of NOX2 with 100 µM GLX and 10 µM GSK (Figure 8C,D,H,I) for 12 h.Correspondingly, IL-8 and IL-10 secretion levels were reduced following the inhibition of NOX2 (Figure 8F,G,K,L).The expression levels of CEBPB transcripts were also significantly decreased by the inhibition of NOX2 (Figure 8E,J).
inhibitor AZD5363 (2 μM) for 12 h (Figure 7B,C).Additionally, a significant decrease was not detected in the level of AMPK phosphorylation following LRRC8A inhibition with 10 μM EDV for 2 h (Figure 7D,E).AMPK is activated in macrophages in a Ca 2+ -dependent manner [37]; however, [Ca 2+ ]i only slightly increased in M2-MACs treated with EDV (Figure 2B).The pharmacological inhibition of AMPK by treatment with Bay3827 (BAY, 10 μM) for 12 h suppressed the expression and secretion of IL-8 and IL-10 and expression of CEBPB in M2-MACs (Figure S6).These results suggest that the inhibition of LRRC8A inactivated the Nrf2 downstream signaling pathway through a pathway independent of AMPK.Corresponding to the results in Figure 7D,E, P    Moreover, the cellular localization of P-Nrf2 was visualized 2 h after the inhibition of NOX2 with 100 µM GLX and 10 µM GSK.As shown in Figure 9A,B,D,E, the percentages of cells with P-Nrf2-positive nuclei in M 2 -MACs were significantly reduced by the GLX and GSK treatments (n = 6 for each, p < 0.01).Nrf2 staining was detected at similar levels in the cytosolic regions of vehicle-, GLX-, and GSK-treated M 2 -MACs (Figure 9C,F).Moreover, the cellular localization of P-Nrf2 was visualized 2 h after the inhibit NOX2 with 100 μM GLX and 10 μM GSK.As shown in Figure 9A,B,D,E, the percen of cells with P-Nrf2-positive nuclei in M2-MACs were significantly reduced by the and GSK treatments (n = 6 for each, p < 0.01).Nrf2 staining was detected at similar in the cytosolic regions of vehicle-, GLX-, and GSK-treated M2-MACs (Figure 9C,F)

Decreased ROS Production by LRRC8A Inhibition in M2-MACs
We examined the effects of the inhibition of LRRC8A with EDV on intracellular ROS levels in M2-MACs using an ROS assay kit.Fluorescence images of the fluorescent compound 2,7-dichlorofluorescein oxidized by ROS were assessed.As shown in Figure 10A,B, fluorescence intensity was significantly decreased by the treatments with 10 μM EDV and 10 μM GSK for 30 min (n = 4 for each, p < 0.01).Therefore, the inhibition of LRRC8A may downregulate the expression of IL-8 and IL-10 in M2-MACs through the NOX2-Nrf2-CE-BPB transcriptional axis.

Suppression of the High Extracellular K + -Induced Upregulation of IL-8 and IL-10 by LRRC8A and NOX2 Inhibition in M2-MACs
In a hypoxic TME, the death of cancer cells leads to an increase in K + in the extracellular compartment, and high intratumoral K + promotes immunosuppressive potency [42,43].Additionally, we previously reported the high extracellular K + ([K + ]e)-induced upregulation of IL-8 and IL-10 in M2-MACs [14].The concentration of K + in Roswell Park Memorial Institute (RPMI)-1640 medium was 5 mM, and media with high K + (final concentration: 35 mM) were prepared by supplementation with 30 mM KCl.As previously reported [14], real-time PCR and ELISA experiments showed increases in the transcriptional expression (n = 4, p < 0.01) (Figure 11A,B,F,G) and secretion (n = 4, p < 0.01) (Figure 11D,E,I,J) of IL-10 and IL-8 in M2-MACs exposed to high [K + ]e.Correspondingly, the expression levels of CEBPB were increased by high [K + ]e (Figure 11C,H).No significant changes in their expression and secretion were noted by supplementation with 30 mM NaCl instead of 30 mM KCl (n = 4, p > 0.05) (Figure S9).High [K + ]e-induced increases in IL-8 and IL-10 expression and secretion were significantly reduced by simultaneous treatments with 10 μM EDV (n = 4, p < 0.01) (Figure 11A,B,D,E) and 10 μM GSK (n = 4, p < 0.01) (Figure 11F,G,I,J).Correspondingly, high [K + ]e-induced increases in CEBPB expression were reduced by treatments with EDV (Figure 11C) and GSK (Figure 11H).These results demonstrated that the inhibition of LRRC8A in TAMs may suppress overexpressed IL-10 and IL-8 in a high [K + ]e-exposed TME.

Suppression of the High Extracellular K + -Induced Upregulation of IL-8 and IL-10 by LRRC8A and NOX2 Inhibition in M 2 -MACs
In a hypoxic TME, the death of cancer cells leads to an increase in K + in the extracellular compartment, and high intratumoral K + promotes immunosuppressive potency [42,43].Additionally, we previously reported the high extracellular K + ([K + ] e )-induced upregulation of IL-8 and IL-10 in M 2 -MACs [14].The concentration of K + in Roswell Park Memorial Institute (RPMI)-1640 medium was 5 mM, and media with high K + (final concentration: 35 mM) were prepared by supplementation with 30 mM KCl.As previously reported [14], real-time PCR and ELISA experiments showed increases in the transcriptional expression (n = 4, p < 0.01) (Figure 11A,B,F,G) and secretion (n = 4, p < 0.01) (Figure 11D,E,I,J) of IL-10 and IL-8 in M 2 -MACs exposed to high [K + ] e .Correspondingly, the expression levels of CEBPB were increased by high [K + ] e (Figure 11C,H).No significant changes in their expression and secretion were noted by supplementation with 30 mM NaCl instead of 30 mM KCl (n = 4, p > 0.05) (Figure S9).High [K + ] e -induced increases in IL-8 and IL-10 expression and secretion were significantly reduced by simultaneous treatments with 10 µM EDV (n = 4, p < 0.01) (Figure 11A,B,D,E) and 10 µM GSK (n = 4, p < 0.01) (Figure 11F,G,I,J).Correspondingly, high [K + ] e -induced increases in CEBPB expression were reduced by treatments with EDV (Figure 11C) and GSK (Figure 11H).These results demonstrated that the inhibition of LRRC8A in TAMs may suppress overexpressed IL-10 and IL-8 in a high [K + ] e -exposed TME.-/-and +/-, respectively.

Mechanisms Underlying the Transcriptional Upregulation of LRRC8A in M 2 -MACs
Chen et al. recently reported that LRRC8A was upregulated through mRNA modification by NOP2/Sun RNA methyltransferase 2 (NSUN2), an m 5 C methyltransferase, and the stability of the m 5 C-modified LRRC8A mRNA was further enhanced by binding to Y-box binding protein 1 (YBX1) [44].As shown in Figure 12A, LRRC8A was upregulated by M 2 differentiation in THP-1 cells.In addition, NSUN2 and YBX1 were highly expressed in M 2 -MACs, similar to native THP-1 and M 0 -MACs (Figure 12B,C).However, the siRNA-mediated inhibition of NSUN2 and pharmacological inhibition of YBX1 with SU056 (10 µM) did not induce LRRC8A mRNA destabilization in M 2 -MACs (Figure 12D,E).On the other hand, histone deacetylases (HDACs) have been shown to play crucial roles in macrophage functions [45] and are responsible for the post-transcriptional modification of ion channels such as ANO1 and K Ca 3.1 [46,47].As shown in Figure 12F,G, HDAC1, 2, and 3 and sirtuin 1 (SIRT1) transcripts were expressed in M 2 -MACs at high levels.Among them, HDAC3 levels were approximately four-fold higher in M 2 -MACs than in native THP-1 and M 0 -MACs (Figures 12F,G and S10).The expression levels of LRRC8A transcripts were reduced by approximately 50% following treatments with the pan-HDAC inhibitor, vorinostat (1 µM), and the selective HDAC3 inhibitor, T247 (1 µM), for 24 h; however, no significant changes were observed following treatments with the HADC1/2 inhibitor, AATB (10 µM), and the SIRT1 inhibitor, Ex527 (1 µM) (Figure 12H).Correspondingly, the siRNA-mediated inhibition of HDAC3 significantly reduced their levels (Figure 12I).These results suggest that HDAC3 is, at least in part, responsible for the upregulated expression of LRRC8A in M 2 -MACs.These results suggest that HDAC3 is, at least in part, responsible for the upregulated expression of LRRC8A in M2-MACs.

Discussion
TAMs are a major component of the TME and maintain an immunosuppressive microenvironment by secreting pro-tumorigenic and pro-angiogenic factors, including IL-10 and IL-8 [11][12][13].We previously reported that the activation of the Ca 2+ -activated K + channel KCa3.1 repressed the transcriptional expression of IL-10 and IL-8 through the JNK and ERK signaling pathways in M2-MACs [14].However, the mechanisms underlying the transcriptional regulation of IL-10 and IL-8 through the Nrf2 signaling pathway in M2-MACs remain unclear.Consistent with our previous findings [14], the pharmacological activation of KCa3.1 with SKA-121 did not suppress the nuclear translocation of P-Nrf2, and Nrf2 activation with NK252 did not reverse the KCa3.1 activation-induced transcriptional repression of IL-8 and IL-10 in M2-MACs (Figure S5).LRRC8A is associated with phenotype polarization and phagocytosis in macrophages [18] and is physically coupled with NOX isoforms [31,39,48].The NOX-ROS-Nrf2 axis is important for M2 macrophage activation [49].The present study elucidated the critical role of LRRC8A in the Nrf2-

Discussion
TAMs are a major component of the TME and maintain an immunosuppressive microenvironment by secreting pro-tumorigenic and pro-angiogenic factors, including IL-10 and IL-8 [11][12][13].We previously reported that the activation of the Ca 2+ -activated K + channel K Ca 3.1 repressed the transcriptional expression of IL-10 and IL-8 through the JNK and ERK signaling pathways in M 2 -MACs [14].However, the mechanisms underlying the transcriptional regulation of IL-10 and IL-8 through the Nrf2 signaling pathway in M 2 -MACs remain unclear.Consistent with our previous findings [14], the pharmacological activation of K Ca 3.1 with SKA-121 did not suppress the nuclear translocation of P-Nrf2, and Nrf2 activation with NK252 did not reverse the K Ca 3.1 activation-induced transcriptional repression of IL-8 and IL-10 in M 2 -MACs (Figure S5).LRRC8A is associated with phenotype polarization and phagocytosis in macrophages [18] and is physically coupled with NOX isoforms [31,39,48].The NOX-ROS-Nrf2 axis is important for M 2 macrophage activation [49].The present study elucidated the critical role of LRRC8A in the Nrf2-mediated production of IL-8 and IL-10 in M 2 -MACs.The main results obtained in the present study are as follows.(1) LRRC8A expression was higher in M 2 -MACs than in native THP-1 and M 0 -MACs.The pharmacological inhibition of LRRC8A induced prolonged hyperpolarization and a sustained elevation in basal [Ca 2+ ] i in M 2 -MACs (Figures 1, 2 and 12).The overexpression of HDAC3 may be associated with the upregulation of IL-8 and IL-10 in M 2 -MACs (Figure 12).( 2) The inhibition of LRRC8A repressed the expression and secretion of IL-8 and IL-10 through the Nrf2-CEBPB transcriptional axis in M 2 -MACs (Figures 3-6).
(3) The inhibition of LRRC8A suppressed the activation of Nrf2 through the NOX2-ROS signaling pathway in M 2 -MACs (Figures 8-10).( 4) The upregulation of IL-8 and IL-10 following the exposure of M 2 -MACs to high K + mimicking the TME was mostly reversed by the inhibition of LRRC8A (Figure 11).Cytosolic Cl − is increased by the inhibition of LRRC8A through a reduction in Cl − efflux.As shown in Figure 2A,D, the LRRC8A inhibition-induced accumulation of [Cl − ] i contributed to plasma membrane hyperpolarization in M 2 -MACs.In myoblasts expressing voltage-gated Ca 2+ channels, the inhibition of LRRC8A decreased basal [Ca 2+ ] i [50]; however, in M 2 -MACs, LRRC8A inhibition-induced hyperpolarization elevated [Ca 2+ ] i through voltage-independent Ca 2+ channels such as Orai/STIM and transient receptor potential (TRP) channels [51,52].
The activation of the PI3K-AKT signaling pathway is associated with M 2 macrophage polarization and activates Nrf2 downstream in macrophages [53,54].The inhibition of LRRC8A mitigated the activation/phosphorylation of AKT in T cells [17].Therefore, we predicted the possible contribution of AKT to the LRRC8A inhibition-induced inactivation of Nrf2 in M 2 -MACs.However, the expression level of P-AKT was very low in M 2 -MACs (Figure 7A), and no significant changes in the expression of IL-8 and IL-10 were found by AKT inhibition in M 2 -MACs (Figure 7B,C).Liu et al. showed that the inhibition of LRRC8A promoted macrophage phagocytosis by activating AMPK, resulting in the nuclear translocation of Nrf2 [55].However, AMPK is not distinguished as a Cl − -sensitive and volume-sensitive protein kinase.AMPK is activated by [Ca 2+ ] i elevation through the activation of Ca 2+ /calmodulin-dependent protein kinase kinase 2 (CaMKK2) [55].In the present study, LRRC8A inhibition-induced hyperpolarization slightly increased intracellular steady-state [Ca 2+ ] i in M 2 -MACs (Figure 2); however, the phosphorylation levels of AMPK were not changed by the inhibition of LRRC8A with EDV (Figures 7 and S6).
WNK lysine-deficient protein kinase 1 (WNK1) is a Cl − -sensitive protein kinase known as an upstream molecule of AMPK.The activation of WNK1 inhibits the AMPK signaling pathway [56,57].In M 2 -MACs, among four WNK isoforms (WNK1-4), WNK1 was highly expressed; however, no significant changes in the expression levels of IL-8 and IL-10 were found by the pharmacological and siRNA-mediated inhibition of WNK1 in M 2 -MACs (Figure S11).The expression levels of the other isoforms (WNK2-4) were less than 0.0001 a.u. in M 2 -MACs.As described above, the inhibition of LRRC8A did not affect the phosphorylation level of AMPK in M 2 -MACs.However, previous studies indicated that the AMPK-Nrf2 axis is an important upstream signaling pathway for cytokine expression such as IL-10 [58,59].As shown in Figure S6, the pharmacological inhibition of AMPK suppressed IL-8 and IL-10 transcription through the Nrf2-CEBPB transcriptional axis.These results suggest that AMPK inhibitors suppress these cytokines through mechanisms independent of LRRC8A.In the present study, we focused on the NOX-ROS-Nrf2 transcriptional axis, but not the WNK1-AMPK-Nrf2 one, as the downstream of LRRC8A in M 2 -MACs.
LRRC8A is physically associated with NOX1, 2, and 4, and is required for NOX activity [31,39,48].Macrophage-specific NOX2-derived ROS play an essential role in Nrf2 activation and M 2 polarization [49].In cancer cells, Nrf2 suppresses tumor promotion or conversely exerts pro-tumorigenic effects [60]; however, the Nrf2 signaling pathway in M 2 -MACs prevents cancer immunosurveillance by producing pro-tumorigenic cytokines.In the present study, NOX2 expression was higher in M 2 -MACs than NOX1 and NOX4 expression and was distributed along the plasma membrane (Figures 8A,B and S8A).The expression and secretion levels of IL-8 and IL-10 and the nuclear translocation of P-Nrf2 were significantly reduced by the inhibition of NOX2 (Figures 8 and 9).Additionally, ROS levels were reduced by the inhibition of LRRC8A and NOX2 (Figure 10).
CEBPB contributes to M 2 macrophage polarization by promoting the expression of M 2 marker genes such as Arg1 and immunosuppressive enzymes including NOX2 [5,27,61].In the present study, CEBPB was predominantly expressed in M 2 -MACs (Figure 6A), and the Nrf2-CEBPB transcriptional axis was involved in the LRRC8A inhibition-mediated downregulation of IL-8 and IL-10 (Figure 6B-E).The present study is the first to provide evidence for the LRRC8A-mediated regulation of the Nrf2-CEBPB transcriptional axis.Therefore, the LRRC8A-NOX2-Nrf2-CEBPB transcriptional axis may play a key role in M 2 polarization by regulating the gene expression of M 2 -related molecules.As shown in Figure S12, the pharmacological and/or siRNA-mediated inhibition of LRRC8A, Nrf2, and CEBPB in M 2 -MACs significantly reduced the expression levels of NOX2, CD169, and Arg1.
ANO6, VDAC1, and CLIC1 transcripts were highly expressed in M 2 -MACs (Figure S1B,D).However, their pharmacological and/or siRNA-mediated inhibition did not affect the expression levels of IL-8 and IL-10 in M 2 -MACs (Figures S3C,D and S4C,D,F,G).The involvement of ANO6 and VDAC1 in the Nrf2 signaling pathway has not yet been demonstrated.In vascular endothelial cells, CLIC1 inhibits the nuclear translocation and phosphorylation of Nrf2 [62,63].In macrophages, CLIC1 plays an essential role in phagosome acidification and superoxide production [64]; however, it remains unclear whether CLIC1 regulates the upstream signaling pathways of IL-8 and IL-10 transcription.
A recent study reported that LRRC8A was upregulated by NSUN2-mediated m 5 C modification in cancer cells, and m 5 C-modified LRRC8A mRNA increased RNA stability by binding to YBX1 [44].However, neither siRNA-mediated NSUN2 inhibition nor YBX1 inhibition with SU056 downregulated LRRC8A (Figure 12D,E).A previous study reported that the expression of NSUN2 negatively correlated with that of M 2 macrophage markers, such as CD163, and its methylation inhibited the polarization of M 2 macrophages [65].Therefore, it is reasonable to assume that NSUN2 is not related to the transcriptional regulation of LRRC8A in M 2 -MACs.On the other hand, HDAC3 was shown to play a significant role in macrophage development and homeostasis [45,66].We found that the gene expression of K Ca 3.1 and ANO1 was epigenetically regulated by HDAC3 in breast and prostate cancers [46,47].As shown in Figure 12F, HDAC3 was upregulated in M 2 -MACs, and the selective inhibition of HDAC3, but not HDAC1/2 or SIRT1, significantly reduced the expression level of LRRC8A transcripts in M 2 -MACs (Figure 12H,I).The present study is the first to provide evidence for the HDAC3-mediated epigenetic modification of LRRC8A.

Cell Culture and Differentiation into M 2 Macrophages
The differentiation of the human monocytic leukemia cell line, THP-1, into M 0 macrophages was induced by treatment with PMA (100 ng/mL) for 8 h.After removal of the medium, cells were incubated with RPMI-1640 medium supplemented with IL-4 and IL-13 (20 ng/mL each) for 72 h to induce the polarization of M 2 macrophages [14].Drug applications were performed 48 (for Western blotting) or 60 (for qPCR) hr after the incubation with IL-4/IL-13.

Measurements of Membrane Potential and Intracellular Concentration of Ca 2+ ([Ca 2+ ] i )
Membrane potential was measured using the fluorescent voltage-sensitive dye, DiBAC 4 (3) [14].In membrane potential imaging, cells loaded with DiBAC 4 (3) were illuminated at a wavelength of 490 nm.[Ca 2+ ] i was measured using the fluorescent Ca 2+ indicator dye, Fura 2-AM, and cells loaded with Fura 2 were alternatively illuminated at wavelengths of 340 and 380 nm.The fluorescent intensity of Fura 2 was expressed as measured 340/380 nm fluorescence ratios after background subtraction.After the adherent cells were treated with trypsin solution, isolated cells were seeded onto the glass bottom dish and incubated with the medium for 4-6 h.Before fluorescence measurements, cells were incubated in extracellular solution containing both 100 nM DiBAC 4 (3) and 10 µM Fura 2-AM for 30 min.The extracellular solution was composed of the following (in mM): 137 NaCl, 5.9 KCl, 2.2 CaCl 2 , 1.2 MgCl 2 , 14 glucose, and 10 4-(2-hydroxyethyl)piperazine-1ethanesulfonic acid (pH 7.4).Fluorescence images were recorded using the ORCA-Flash2.8digital camera (Hamamatsu Photonics, Hamamatsu, Japan).Data collection and analyses were performed using an HCImage system (Hamamatsu Photonics).Images were obtained every 5 s, and fluorescent intensity values were assessed using the average for 1 min (12 images).Drug solutions were applied by operating the three-way stopcock.

Real-Time PCR Assay
Total RNA extraction and cDNA synthesis were conducted as previously reported [14].Gene-specific primers for real-time PCR examinations were designed using Primer Express TM software (Ver 1.5, Thermo Fisher Scientific).Real-time PCR was performed using Luna Universal qPCR Master Mix (New England Biolabs Japan, Tokyo, Japan) on an ABI 7500 real-time PCR instrument (Thermo Fisher Scientific) [14].PCR primers of human origin are listed in Table S1.Unknown quantities relative to the standard curve for a particular set of primers were calculated [14], yielding the transcriptional quantitation of gene products relative to the ACTB.

Western Blots
Whole-cell lysates of M 2 -MACs were extracted by RIPA buffer with phosphatase inhibitor cocktail at a final concentration of 1%.Equal amounts of protein were subjected to SDS-PAGE and immunoblotting with the antibodies shown in Table S2 and were then incubated with an anti-rabbit or anti-mouse HRP-conjugated IgG secondary antibody.The ECL Advance Chemiluminescence Reagent Kit was used to identify the bound antibody.The resulting images were analyzed using an Amersham Imager 600 (GE Healthcare Japan) [14].The optical density of the protein band signal relative to that of the ACTB signal was calculated using ImageJ software (Ver.1.42, NIH, USA), and protein expression levels in the vehicle control are expressed as 1.0.

Measurement of Cytokine Production by ELISA and ROS by a ROS Assay Kit
Human IL-10 and IL-8 levels in supernatant samples were measured with the respective IL-10/IL-8 Human Uncoated ELISA Kits, according to the manufacturer's protocols.Standard curves were plotted using a series of cytokine/chemokine concentrations [14].Absorbance was measured using the microplate reader SpectraMax 384 (Molecular Devices Japan, Tokyo, Japan) at a test wavelength of 450 nm and a reference wavelength of 650 nm.To measure intracellular ROS levels, isolated cells with trypsin solution were seeded onto the glass bottom dish for 4-6 h and then were loaded with 10 µM photo-oxidation-resistant DCFH-DA dye at 37 • C for 30 min, according to the manufacturer's instruction.The dyes were washed out by the extracellular solution (see above) and illuminated at a wavelength of 490 nm.Fluorescence images of the fluorescent compound 2,7-dichlorofluorescein oxidized by ROS were recorded using the ORCA-Flash2.8digital camera (Hamamatsu Photonics).Data collection and analyses were performed using the HCImage system (Hamamatsu Photonics).After the treatment with trypsin solution, isolated M 2 -MACs were fixed and permeabilized using a CytoFix/Perm Kit.The antibodies shown in Table S2 were labeled with an Alexa Fluor 488-conjugated secondary antibody.After seeding onto a glass bottom dish, fluorescence images were visualized using a confocal laser scanning microscopy system (Nikon A1R, Tokyo, Japan) [14].Image data were quantitatively analyzed using ImageJ software.To assess the P-Nrf2-positive cells, the images from more than 30 cells for each case (n = 1) were obtained.

Statistical Analysis
Statistical analyses were performed with the statistical software XLSTAT (version 2013.1).To assess the significance of differences between two groups and among multiple groups, the unpaired/paired Student's t-test with Welch's correction or a one-way ANOVA with Tukey's test was used.Results with a p value < 0.05 were considered to be significant.Data are shown as means ± standard error.

Conclusions
[K + ] e elevation by necrotic cancer and cancer-infiltrating cells in the TME increased the expression and production of IL-10 and IL-8 in M 2 -MACs.In addition to our previous findings showing the effectiveness of the activation of K Ca 3.1 [14], the present study demonstrated the effectiveness of the inhibition of LRRC8A for TME-mediated TAM dysregulation, which induced the accumulation of pro-tumorigenic and pro-angiogenic cytokines, such as IL-8 and IL-10.However, the clinical and therapeutic importance of LRRC8A inhibitors remains unclear.The present study provides novel insights that support the potential of potent and selective LRRC8A inhibitors as therapeutic applications in TAM-targeting cancer immunotherapy.Recent studies indicated that TAM-like M 2 macrophages enhanced resistance to anti-cancer drugs and the recruitment of immunosuppressive cells, such as T reg s and myeloid-derived suppressor cells (MDSCs), to the TME, and also that these effects were associated with LRRC8A [67,68].Therefore, the inhibition of LRRC8A may overcome resistance to anti-cancer drugs by downregulating drug-metabolizing enzymes and efflux transporters and may also prevent the accumulation of immunosuppressive cells (TAMs, T reg s, and MDSCs) by suppressing the production of their recruiting chemokines.Further studies under conditions that mimic the TME are needed to examine the clinical applications and limitations of LRRC8A inhibitors in cancer immunotherapy.

2. 2 .
Decreased Expression and Secretion of IL-8 and IL-10 by the Small Interfering RNA (siRNA)-Mediated and Pharmacological Inhibition of LRRC8A in M 2 -MACs

Figure 3 .
Figure 3. Effects of the siRNA-mediated and pharmacological inhibition of LRRC8A on IL-8 and IL-10 expression and secretion in M2-MACs.(A,C) Real-time PCR examination of IL-8 (A) and IL-10 (C) expression in M2-MACs 48 h after the transfection of LRRC8A siRNA (siLRRC8A).The mRNA expression level in the control siRNA (siCont)-transfected group is expressed as 1.0 (n = 4 for each).(B,D) Quantitative detection of IL-8 (B) and IL-10 (D) secretion by an ELISA assay in the siCont and siLRRC8A groups.The cytokine secretion level in the siCont group is expressed as 1.0 (n = 4 for each).(E,G) Real-time PCR examination of IL-8 (E) and IL-10 (G) expression in vehicle-and endovion (EDV: 1, 3, and 10 μM)-treated M2-MACs for 12 hr.The mRNA expression level in the vehicle control is expressed as 1.0 (n = 4 for each).(F,H) Quantitative detection of IL-8 (F) and IL-10 (H) secretion by an ELISA assay in the vehicle-and EDV (10 μM)-treated groups.The cytokine secretion level in the vehicle control is expressed as 1.0 (n = 4 for each).*, **: p < 0.05, 0.01 vs. siCont and the vehicle control.

Figure 3 .
Figure 3. Effects of the siRNA-mediated and pharmacological inhibition of LRRC8A on IL-8 and IL-10 expression and secretion in M 2 -MACs.(A,C) Real-time PCR examination of IL-8 (A) and IL-10 (C) expression in M 2 -MACs 48 h after the transfection of LRRC8A siRNA (siLRRC8A).The mRNA expression level in the control siRNA (siCont)-transfected group is expressed as 1.0 (n = 4 for each).(B,D) Quantitative detection of IL-8 (B) and IL-10 (D) secretion by an ELISA assay in the siCont and siLRRC8A groups.The cytokine secretion level in the siCont group is expressed as 1.0 (n = 4 for each).(E,G) Real-time PCR examination of IL-8 (E) and IL-10 (G) expression in vehicle-and endovion (EDV: 1, 3, and 10 µM)-treated M 2 -MACs for 12 h.The mRNA expression level in the vehicle control is expressed as 1.0 (n = 4 for each).(F,H) Quantitative detection of IL-8 (F) and IL-10 (H) secretion by an ELISA assay in the vehicle-and EDV (10 µM)-treated groups.The cytokine secretion level in the vehicle control is expressed as 1.0 (n = 4 for each).*, **: p < 0.05, 0.01 vs. siCont and the vehicle control.

Figure 6 .
Figure 6.Identification of CEBP isoforms as downstream transcriptional factors of Nrf2 i MACs.(A) Real-time PCR examination of CEBP isoform expression in M2-MACs.Expression are shown as the ratio to the ACTB (n = 4 for each).(B,C) Real-time PCR examination of IL-8 (B IL-10 (C) expression in M2-MACs 48 h after the transfection of CEBPB siRNA (siCEBPB).The m expression level in the control siRNA (siCont)-transfected group is expressed as 1.0 (n = 4 for (D,E) Real-time PCR examination of CEBPB expression in vehicle-, endovion (EDV; 10 μM and ML385 (5 μM)-treated (E) M2-MACs for 12 h.The mRNA expression level in the vehicle c is expressed as 1.0 (n = 4 for each).**: p < 0.01 vs. siCont and the vehicle control.
-AMPK and AMPK were bot tected at similar levels in the cytosolic regions of vehicle-(Figure S7A(a),B,C(a),D EDV-(Figure S7A(b),B,C(b),D) treated M2-MACs.In M2-MACs stained with the A Fluor 488-conjugated anti-mouse IgG antibody alone, very weak fluorescent signals observed under the same acquisition settings (Figure S7E).

Figure 6 .
Figure 6.Identification of CEBP isoforms as downstream transcriptional factors of Nrf2 in M 2 -MACs.(A) Real-time PCR examination of CEBP isoform expression in M 2 -MACs.Expression levels are shown as the ratio to the ACTB (n = 4 for each).(B,C) Real-time PCR examination of IL-8 (B) and IL-10 (C) expression in M 2 -MACs 48 h after the transfection of CEBPB siRNA (siCEBPB).The mRNA expression level in the control siRNA (siCont)-transfected group is expressed as 1.0 (n = 4 for each).(D,E) Real-time PCR examination of CEBPB expression in vehicle-, endovion (EDV; 10 µM)-(D), and ML385 (5 µM)-treated (E) M 2 -MACs for 12 h.The mRNA expression level in the vehicle control is expressed as 1.0 (n = 4 for each).**: p < 0.01 vs. siCont and the vehicle control.2.5.Involvement of the NOX2-ROS Axis in the LRRC8A Inhibition-Induced Transcriptional Repression of IL-8 and IL-10 in M 2 -MACs -AMPK and AMPK were both detected at similar levels in the cytosolic regions of vehicle-(Figure S7A(a),B,C(a),D) and EDV-(Figure S7A(b),B,C(b),D) treated M 2 -MACs.In M 2 -MACs stained with the Alexa Fluor 488-conjugated anti-mouse IgG antibody alone, very weak fluorescent signals were observed under the same acquisition settings (Figure S7E).
-AMPK and AMPK were both detected at similar levels in the cytosolic regions of vehicle-(Figure S7A(a),B,C(a),D) and EDV-(Figure S7A(b),B,C(b),D) treated M2-MACs.In M2-MACs stained with the Alexa Fluor 488-conjugated anti-mouse IgG antibody alone, very weak fluorescent signals were observed under the same acquisition settings (Figure S7E).

Figure 7 .
Figure 7.No involvement of AKT or AMPK in the LRRC8A inhibition-induced downregulati IL-8 and IL-10 in M2-MACs.(A) Protein expression of P-Nrf2, Nrf2, P-AKT, and AKT in pr lysates of M2-MACs.Blots were probed with anti-P-Nrf2/Nrf2 (approx.100 kDa), anti-P-AKT/ (approx.60 kDa), and anti-ACTB (approx.45 kDa) antibodies.(B,C) Real-time PCR examinati IL-8 (B) and IL-10 (C) expression in vehicle-and AKT inhibitor AZD5363 (2 μM)-treated M2-M for 12 hr.The mRNA level in the vehicle control is expressed as 1.0 (n = 4 for each).(D) Pr expression of P-AMPK and AMPK in protein lysates of vehicle-and endovion (EDV; 10 μM)-tr M2-MACs for 2 hr.Blots were probed with anti-P-AMPK/AMPK (approx.65 kDa) and anti-A (approx.45 kDa) antibodies.(E) Summarized results of the relative expression of P-AMPK/A were obtained from the optical density of P-AMPK, AMPK, and ACTB band signals (n = 4 for e Apart from AKT and AMPK, ROS also activates the Nrf2 downstream signaling p way in macrophages [38].LRRC8A has been shown to activate NOX-dependent ROS duction and physically binds to NOX1, 2, and 4 [39,40].As shown in Figure 8A, am the seven NOX isoforms, NOX2 transcripts were predominantly expressed in M2-M NOX2 protein expression was observed in M2-MACs (Figure 8B).Correspondingly fluorescence signals of the Alexa 488-labeled anti-NOX2 antibody were observed a the plasma membrane of M2-MACs (Figure S8A(b)).With the same acquisition sett very weak signals were observed in native THP-1 cells stained with the Alexa 488-lab anti-NOX2 antibody and in native THP-1 cells and M2-MACs stained with the Alexa labeled anti-NOX4 antibody (Figure S8A(a),B).GLX351322 (GLX) is a NOX4 inhi with IC50 = 5 μM and also inhibits NOX2 with IC50 = 40 μM [40].On the other h GSK2796039 (GSK) is a selective NOX2 inhibitor with IC50 = 1 μM [41].No significan crease in the expression of IL-8 and IL-10 transcripts was found following the inhib of NOX4 in M2-MACs with 10 μM GLX for 12 h (Figure 8C,D); however, it was sig cantly decreased by approximately 50% following the inhibition of NOX2 with 100 GLX and 10 μM GSK (Figure 8C,D,H,I) for 12 hr.Correspondingly, IL-8 and IL-10 s tion levels were reduced following the inhibition of NOX2 (Figure 8F,G,K,L).The exp sion levels of CEBPB transcripts were also significantly decreased by the inhibitio NOX2 (Figure 8E,J).

Figure 8 . 8 .
Figure 8. Involvement of NOX2 in the LRRC8A inhibition-induced downregulation of IL-8 an 10 in M2-MACs.(A) Identification of NOX isoforms mainly expressed in M2-MACs by a real Figure 8. Involvement of NOX2 in the LRRC8A inhibition-induced downregulation of IL-8 and IL-10 in M 2 -MACs.(A) Identification of NOX isoforms mainly expressed in M 2 -MACs by a real-time PCR examination.Expression levels are shown as the ratio to the ACTB.The mRNA expression level in Int. J. Mol.Sci.2024, 25, x FOR PEER REVIEW 1 PCR examination.Expression levels are shown as the ratio to the ACTB.The mRNA expr level in the vehicle control is expressed as 1.0 (n = 4 for each).(B) Protein expression of NOX2 MAC protein lysates.Blots were probed with anti-NOX2 (approx.60 kDa) and anti-ACTB (a 45 kDa) antibodies.(C-E) Real-time PCR examination of IL-8 (C), IL-10 (D), and CEBPB (E) i cle-and NOX4/NOX2 inhibitor GLX351322 (GLX) (10 and 100 μM)-treated M2-MACs for 12 h mRNA expression level in the vehicle control is expressed as 1.0 (n = 4 for each).(F, G) Quan detection of IL-8 (F) and IL-10 (G) secretion by an ELISA assay in vehicle-and GLX (100 μM)-M2-MACs for 24 hr.The cytokine secretion level in the vehicle control is expressed as 1.0 (n each).(H-J) Real-time PCR examination of IL-8 (H), IL-10 (I), and CEBPB (J) in vehicle-and se NOX2 inhibitor GSK2796039 (GSK) (10 μM)-treated M2-MACs for 12 hr.The mRNA expressio in the vehicle control is expressed as 1.0 (n = 4 for each).(K,L) Quantitative detection of IL-8 ( IL-10 (L) secretion by an ELISA assay in vehicle-and GSK-treated M2-MACs for 24 hr.The cy secretion level in the vehicle control is expressed as 1.0 (n = 4 for each).**: p < 0.01 vs. the control.

2. 6 .
Decreased ROS Production by LRRC8A Inhibition in M 2 -MACs We examined the effects of the inhibition of LRRC8A with EDV on intracellular ROS levels in M 2 -MACs using an ROS assay kit.Fluorescence images of the fluorescent compound 2,7-dichlorofluorescein oxidized by ROS were assessed.As shown in Figure 10A,B, fluorescence intensity was significantly decreased by the treatments with 10 µM EDV and 10 µM GSK for 30 min (n = 4 for each, p < 0.01).Therefore, the inhibition of LRRC8A may downregulate the expression of IL-8 and IL-10 in M 2 -MACs through the NOX2-Nrf2-CEBPB transcriptional axis.Int.J. Mol.Sci.2024, 25, x FOR PEER REVIEW 11 of 21

Figure 10 .
Figure 10.Inhibition of ROS activity by treatments with EDV and GSK in M2-MACs.(A) Typical pseudo-colored images of the fluorescent compound 2,7-dichlorofluorescein oxidized by ROS 30 min after the treatments with vehicle (a), endovion (EDV, 1 μM) (b), and GSK2795039 (GSK, 10 μM) (c) in M2-MACs.The color bar is scaled in the same range, and the warmer colors represent higher ROS levels.(B) Summarized results of relative fluorescence intensity (1.0 at time 0 min) were obtained from the optical density of 2,7-dichlorofluorescein. Numbers used for experiments are shown in parentheses.The scale bar in 'A(a)' (same in A(b,c)) shows 20 μm.**: p < 0.01 vs. the vehicle control.

Figure 10 .
Figure 10.Inhibition of ROS activity by treatments with EDV and GSK in M 2 -MACs.(A) Typical pseudo-colored images of the fluorescent compound 2,7-dichlorofluorescein oxidized by ROS 30 min after the treatments with vehicle (a), endovion (EDV, 1 µM) (b), and GSK2795039 (GSK, 10 µM) (c) in M 2 -MACs.The color bar is scaled in the same range, and the warmer colors represent higher ROS levels.(B) Summarized results of relative fluorescence intensity (1.0 at time 0 min) were obtained from the optical density of 2,7-dichlorofluorescein. Numbers used for experiments are shown in parentheses.The scale bar in 'A(a)' (same in A(b,c)) shows 20 µm.**: p < 0.01 vs. the vehicle control.

4. 7 .
Confocal Imaging of the Nuclear and Cytosolic Distribution of Nrf2 and AMPK and the Plasma Membrane Localization of LRRC8A and NOX2