Bone marrow-derived macrophages (BMDMs) were isolated from the bone marrow of C57B/6 mice aged 10–12 weeks old. All procedures were approved by the Institutional Animal Care and Use Committee of the National Defense Medical Center, Taiwan and followed the National Institutes of Health guidelines for animal care. After euthanasia, the legs of the mice were sprayed with 70% ethanol, and the femurs and tibias were dissected, separated, and transferred to a sterile flow hood. After removing the skin and muscle tissue, the bone was sprayed with 70% ethanol and the ends of the femurs and tibias were cut with sterile scissors. The bone marrow was flushed with ice-cold PBS and filtered through a 70 μm cell strainer (BD Biosciences, CA, United States) to remove solid debris. The recovered cells were cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM; Gibco, ThermoFisher Scientific, Waltham, MA, United States) supplemented with 10% heat-inactivated fetal bovine serum (FBS; HyClone, Cytiva, Logan, UT, United States), 1% penicillin–streptomycin (Biological Industries, CT, United States), and 20 ng/ml recombinant monocytic colony-stimulating factor (M-CSF; Peprotech, Rocky Hill, NJ, United States) for 7 days to fully differentiate into macrophages.

BMDMs were seeded into 6-well culture plates at a density of 1 × 10⁶ cells/well and subjected to the respective experiments. BMDMs were treated with 1 μg/ml LPS (Escherichia coli serotype 026:B6, Sigma-Aldrich, St Louis, MO, United States) or 20 μg/ml IL-4 (Peprotech) to induce macrophage polarization. To modulate cell size, BMDMs were pretreated with 20 mM NaCl (final osmolarity of 400 mOsmol) for 30 min before LPS or IL-4 stimulation. In some experiments, the cells were pretreated with 1 μM HgCl₂ (Sigma-Aldrich). HgCl₂ was dissolved in PBS buffer. M1/M2 polarization, expression of AQPs, cell size, production of pro-inflammatory mediators, signaling pathways, cell autophagy, and mitochondrial stress were analyzed at 8 and 24 h.

Cells were scraped and stained with the following antibodies: APC-eFluor 780 anti-mouse F4/80 (clone BM8, eBioscience, San Diego, CA, United States), Alexa Fluor 488 anti-mouse CD11b (clone M1/70, BD Biosciences), PE anti-mouse iNOS (clone CXNFT, eBioscience), and eFluor450 anti-mouse CD38 (clone 90, eBioscience). The isotype controls were rat IgG2bκ (eBR2a) and rat IgG2bκ (DA/HA) from BD Biosciences (San Jose, CA, United States). The cells were washed and stained with Flow Cytometry Staining Buffer (eBioscience) for 30 min at 4°C. For intracellular staining, cells were fixed in IC Fixation Buffer (eBioscience) for 30 min at 4°C, permeabilized, and stained in permeabilization buffer (eBioscience). Results were obtained using the FACSVerse (BD Biosciences) and analyzed using FlowJo software (Tree Star, Ashland, OR, United States). The F4/80⁺ and CD11b⁺ cells were gated as fully differentiated macrophages.

Total RNA was isolated from cells using the commercial kit RNA Miniprep Plus (Zymo Research Corp., Irvine, CA, United States), and cDNA was prepared using 20–100 ng/μL total RNA with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, United States) following the manufacturer’s instructions. Real-time quantitative PCR (qPCR) was performed on the cDNA using TaqMan probes for nos2, tnf, ccl17 (Applied Biosystems). qPCR was performed on a QuantStudio™ five System (Applied Biosystems) using the TaqMan^® universal Master Mix II (Applied Biosystems). Fold changes in expression were calculated by the 2^−ΔΔCt method using mouse gapdh (glyceraldehyde-3-phosphate dehydrogenase) as an endogenous control for mRNA expression.

Cells were cultivated in a 3.5 cm glass bottom confocal dish (SPL Life Sciences, Pocheon, Korea) at a density of 1.2 × 10⁶ cells per dish. After incubating for 24 h, the cells were fixed with 4% paraformaldehyde for 20 min and blocked with 10% bovine serum albumin (BSA) for 20 min. The cells were incubated with conjugated antibodies eFluor 660 anti-mouse F4/80 (, clone BM8, eBioscience) (1:100), PE anti-mouse iNOS (clone CXNFT, eBioscience) (1:300), and DAPI (4′,6-diamidino-2-phenylindole, Invitrogen, Carlsbad, CA, United States) for 1 h. Images were captured using a laser scanning confocal microscope (LSM 880, Carl Zeiss, United States) with a ×40 objective (original magnification).

Cells in 6-well culture plates were lysed in a radioimmunoprecipitation assay (RIPA) cell lysis buffer containing protease and phosphatase inhibitor cocktail (Halt, ThermoFisher Scientific). To collect the nuclear fraction, cells were seeded and cultivated in 10 cm culture dishes and isolated with Nuclear and Cytoplasmic Extraction Reagent kits (NE-PER, ThermoFisher Scientific) following the manufacturer’s instructions. Protein concentrations were measured using a bicinchoninic acid protein assay kit (Pierce, ThermoFisher Scientific). The lysate (30 μg/lane) was resolved on a 10% SDS-polyacrylamide gel using a Hoefer electrophoresis blotting system (Hoefer, United States) and transferred to a 0.45 μm polyvinylidene fluoride membrane (Merck Millipore, United States). The membrane was probed with antibodies against JNK, p-JNK (Thr183/Tyr185), p38, p-p38 (Thr180/Tyr182), NF-κB (p65), p-NF-κB (p65) (Ser536), cleaved Caspase-3, LC3A/B, P62 (SQSTM1) purchased from Cell Signaling Technology (Danvers, MA, United States), PINK1 from Abcam, AQP1, and AQP9 from Biorbyt (Berkeley, CA, United States) (1:1000 dilution). β-actin (Sigma-Aldrich) and PCNA (OriGene Technologies, Rockville, MD, United States) were incubated at 1:10000 dilutions and used as internal controls. Densitometry was analyzed semi-quantitatively using ImageJ software (National Institutes of Health, Bethesda, MD, United States).

The culture medium was collected from the 6-well culture plates. The concentration of IL-1β was quantified using ELISA kits (R&D Systems, Minneapolis, MN, United States) following the manufacturer’s instructions.

LA-4 murine lung epithelium (CCL-196) purchased from American Type Culture Collection was seeded into 24-well culture plates at 80–90% confluency. Epithelial cells were then treated with 1 μg/ml LPS for 24 h. LPS- or HgCl₂-treated BMDMs labeled with green, fluorescent dye (Invitrogen) were seeded on top of the activated LA-4 monolayer and incubated for 1 h. Non-adherent cells were washed off with warm PBS. The remaining adhered BMDMs were counted from images captured at ×50 magnification using a fluorescence microscope (Leica DM 2500, Wetzlar, Germany). The experiment was repeated at least three times.

Mitochondrial stress was assessed using a Seahorse XFp Extracellular Flux Analyzer (Agilent Technologies, Santa Clara, CA, United States) following the manufacturer’s instructions. BMDMs were seeded in XFp cell culture miniplates (Agilent Technologies) at a density of 3.2 × 10⁵ cells/well. After 8 h of incubation, the medium was replaced with the XF medium. Oligomycin (1 µM), carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP, 4 µM), and antimycin A (0.5 µM) were added to the wells through a Seahorse cartridge. Real-time readings of the oxygen consumption rate (OCR) were calculated and recorded using Seahorse XFp software.

Mitochondrial reactive oxygen species (ROS) levels were measured using MitoSox-Red (Molecular Probes, Invitrogen). BMDMs were seeded in 12-well culture plates at a density of 4 × 10⁵ cells/well. The cells were double-stained with 5 μM MitoSox-Red and 50 nM MitoTracker Green FM (Molecular Probes) in HBSS for 30 min. Images were captured at ×200 and ×400 magnifications using a fluorescence microscope (Leica). The percentage of MitoSOX-positive cells was calculated at magnification of ×200.

All results are expressed as mean ± standard error of the mean (SEM). One-way analysis of covariance (ANOVA) and Student’s t-test were used to compare differences between the study groups. Tukey’s correction was used for post-hoc comparisons. Statistical significance was set at p < 0.05.

LPS is commonly used to polarize macrophages into the M1 phenotype. LPS (1 μg/ml) significantly promoted M1 polarization in BMDM, as characterized by the increase of iNOS⁺/CD38⁺ population (Figures 1C–E). Although hypertonicity (+20 mM NaCl) restricted cell swelling (Figures 1A,B), it further enhanced LPS-induced M1 polarization (Figures 1C–E). In contrast, hypertonicity did not affect M2 polarization in BMDM (Figures 1F–H).

Inhibition of AQPs by HgCl₂ (1 μM) restricted the increase in BMDM cell size caused by LPS and also abolished cell size changes caused by hypertonicity (Figures 2A,B). On the effects to polarization, HgCl₂ significantly inhibited LPS-induced M1 polarization (Figures 2E–I). Immunofluorescence staining showed that HgCl₂ significantly inhibited iNOS and CD38 expression in BMDM (Figures 2C,D), the percentage of macrophage presenting high amount of iNOS (iNOS^hi) and CD38 (CD38^hi) signals were reduced almost by half with HgCl₂ (Figures 2E,F). The effect of HgCl₂ on M1 polarization was also assessed by qPCR, which showed that the expression of M1-related genes nos2, tnf, and ccl17 was reduced by 76%, 21%, and 72%, respectively (Figures 2G–I).

Activation of nuclear factor kappa B (NF-κB) pathway and increased adhesion ability are two notably features for M1 macrophage. LPS increased the cytoplasmic abundance of phosphorylated p38 (Figures 3A–C) and the nuclear translocation of p-p65 NF-κB (Figures 3D–G) in BMDM, and produced large amounts of IL-1β (Figure 3H). LPS also significantly increased the adhesion ability of BMDM (Figures 3I–K). HgCl₂ effectively suppressed the activation of the NF-κB and p38 MAPK pathways and inhibited the production of IL-1β (Figures 3A–H). The presence of HgCl₂ also significantly reduced the adhesion ability (Figures 3I–K).

Mitochondrial ROS production plays an important role in macrophage immunity; however, over-activated mitochondria are associated with cellular stress and dysfunction. Mitochondrial ROS production was estimated using MitoSOX staining. LPS induced substantial mitochondrial ROS production in BMDM (Figure 4A) and significantly increased the percentage of MitoSOX-positive cells (Figure 4B). The LPS-induced mitochondrial ROS production was significantly suppressed by HgCl₂ treatment (Figure 4).

The Seahorse XF Cell Mito Stress Test was used to calculate the fundamental parameters of mitochondrial function (Figure 5A). LPS significantly suppressed oxygen consumption in BMDM under basal conditions (Figure 5C) and during maximal respiration (Figure 5D). LPS also reduced the ATP-linked OCR and spare respiratory capacity (Figures 5E,F). HgCl₂ significantly restored mitochondrial function (Figures 5C–F).

Autophagy is a recycling mechanism of cellular homeostasis responsible for maintaining healthy mitochondria. LPS increased light chain three II/light chain three I level in the presence of bafilomycin A1 (Figures 6A,B) and caused substantial p62/SQSTM1 abundance in BMDM (Figures 6A,C). HgCl₂ further augmented the LPS-induced increase in autophagic flux (Figures 6A–C). LPS also stimulated the expression of mitophagy marker PTEN-induced kinase 1 (PINK1), indicating mitochondrial stress. Pre-treatment of BMDM with HgCl₂ reduced LPS-induced PINK1 abundance, suggesting that the present of AQP inhibitors alleviated mitochondrial stress (Figure 6D).