RAW264.7 and EL4 cells were procured from the Bioresource Collection and Research Center (Hsinchu, Taiwan). RAW264.7 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA) and EL4 cells were cultured in DMEM containing 10% horse serum (HyClone, Logan, UT, USA) supplemented with 2 mM glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin. Unless otherwise specified, all chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA).

RAW264.7 cells were treated with varying concentrations of hydroxychloroquine (HCQ) or chloroquine (CQ) (1.25-40 μM; Cayman Chemical, Ann Arbor, Michigan) for 24 hours. Cell viability was determined via the MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; BioVision, Milpitas, CA, USA) following the manufacturer’s protocol.

Twenty-one 6-week-old female BALB/c mice were obtained from LASCO (Taipei, Taiwan) and allocated into three groups: the normal group (Group 1, n=7), the PIL group induced with pristane (Group 2, n=7), and the HCQ-treated PIL group (Group 3, n=7). The Institutional Animal Care and Use Committee of China Medical University Hospital, Taichung, Taiwan, approved the study protocol (CMUIACUC-2018-282-1, June 2018-June 2019), and the experiments adhered to ARRIVE guidelines. PIL was induced via a single intraperitoneal injection of pristane (0.5 mL; Sigma Chemical, St. Louis, MO, USA) on day 0. HCQ was administered via oral gavage (3 mg/kg/day; ACROS, Geel, Belgium), starting on day 1 and continued for 180 consecutive days. Peritoneal lavage and plasma were collected at 1, 3, and 6 months post-induction for the analysis of lupus-related markers, including autoantibody production, inflammatory cytokines, and proteinuria. Mice were sacrificed at 6 months to evaluate lupus nephritis through immune complex deposition in the kidneys. In the short-term treatment study, mice received HCQ (30 mg/kg/day) by oral gavage for seven consecutive days.

Paraffin-embedded kidney tissues from mice were sectioned into 5-µm serial slices, which were dried overnight at 60°C. The sections were deparaffinized in purified xylene (Leica) for two 10-minute immersions, followed by rehydration through a graded ethanol series (100%, 95%, 80%, and 70%) for 5 minutes each, with a final 5-minute wash in tap water. Antigen retrieval was achieved using a heat-induced epitope retrieval method with Citrate Antigen Retrieval Solution (Scytek) for 30 minutes. The slides were then washed in TBST (Scytek) for 5 minutes, blocked for endogenous peroxidase activity with Peroxide Block (ScyTek) for 10 minutes, and washed again twice with TBST. Non-specific binding was minimized by incubating slides with Protein Block (Leica Novolink) for 5 minutes, followed by two additional TBST washes. A goat anti-mouse IgG (H+L)-HRP antibody was diluted 1:200 in antibody dilution buffer (Ventana) and applied to the sections for 30 minutes, then washed twice with TBST. The sections were developed with DAB (Leica Novolink) for 5 minutes and rinsed with tap water. Counterstaining was performed with Hematoxylin Gill II (Leica) for 2 minutes, followed by another 5-minute wash in tap water. Finally, slides were dehydrated through two 5-minute immersions in 100% alcohol, cleared in xylene, and mounted with Micromount medium (Leica). The immunostained areas of IgG in kidney tissues were captured using an Axio Observer A1 light microscope (ZEISS, Germany) and subsequently analyzed with ImageJ software (NIH, Bethesda, MD, USA). Glomerular IgG staining was evaluated and graded on a scale from 0 (no staining) to 4 (strong, global mesangial staining) as previously described (34). For quantification, all IgG-positive cells within each glomerular tuft were counted and normalized to the total number of glomeruli present in the biopsy. A minimum of 25 glomeruli were assessed per sample to ensure robust quantification.

Prior to staining with cell surface markers, cells were stained using the Zombie Aqua fixable viability kit (BioLegend, San Diego, CA, USA) according to the manufacturer’s instructions. Surface expression of target proteins was evaluated by staining with fluorochrome-conjugated antibodies (BioLegend, San Diego, CA), including anti-F4/80-Alexa488 (clone BM8), anti-CD19-PE (clone 6D5), anti-MerTK-PE/Cy7 (clone 2B10C42), and anti-CD11b-BV421 (clone M1/70) for 20 minutes at 4°C. After washing, the cells were analyzed with a BD FACSCelesta flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) and data were processed using FlowJo software (Tree Star, Ashland, OR, USA). Apoptosis was performed by using ApoScreen Annexin V Apoptosis Kit (Southern Biotech, Birmingham, AL, USA) according to the manufacturer’s protocol.

RAW264.7 cells were stained with 5 μM carboxyfluorescein succinimidyl ester (CFSE; eBioscience, Santa Clara, CA, USA) for 20 minutes at 37°C and subsequently treated with HCQ for 24 hours. EL4 cells were labeled with 0.5 μM MitoTracker Deep Red FM (Thermo Fisher Scientific, Waltham, MA, USA) for 30 minutes at 37°C, followed by UVB exposure at 500 J/m² to induce apoptosis (34). One day post-exposure, the apoptotic EL4 cells were co-cultured with RAW264.7 cells at a 10:1 ratio for 30 minutes to 2 hours at 37°C. After washing with PBS, the cells were stained with Zombie Aqua live-dead dye (BioLegend, San Diego, CA, USA) and analyzed using the BD FACSCelesta flow cytometer. Efferocytosis was quantified as the percentage of macrophages engulfing apoptotic cells (Deep Red⁺/CSFE⁺) out of the total live macrophages (Zombie^-/CSFE⁺) (35). For pHrodo staining, apoptotic EL4 cells in serum-free RPMI 1640 medium were labeled by pHrodo red succinimidyl ester (Thermo Fisher Scientific, Waltham, MA, USA) at a final concentration of 1 μM and incubated with CSFE-stained RAW264.7 cells for 1 hour at 37°C in the dark. After labeling, cells were washed three times with PBS and resuspended in complete culture medium and analyzed using the BD FACSCelesta flow cytometer.

Thymocytes were harvested from BALB/c mouse thymi and induced to undergo apoptosis via serum starvation for 24 hours (36). Peritoneal macrophages were isolated from untreated, PIL, and PIL+HCQ mice on day 7 post-PIL. After a 4-hour adherence period, the macrophages were co-incubated with MitoTracker Deep Red-labeled apoptotic thymocytes at a 10:1 ratio for 30 minutes at 37°C. Efferocytosis was quantified as the percentage of macrophages engulfing apoptotic thymocytes (Deep Red⁺/F4/80⁺) relative to total live macrophages (Zombie^-/F4/80⁺) using the BD FACSCelesta flow cytometer.

After engulfing apoptotic EL4 cells (Deep Red⁺) for 30 min, RAW264.7 cells (CSFE⁺) were washed three times with PBS and analyzed using an Axio Observer A1 inverted fluorescence microscope (ZEISS, Germany). Effercytosis is defined by the counts of Deep Red⁺ cells within CSFE-positive region using ImageJ software (NIH, Bethesda, MD, USA). Acquired images were converted to 8-bit grayscale and processed using the “Threshold” function to segment positive signals from the background. A region of interest (ROI) was selected according to CFSE-green image using the selection tool and subsequently applied to the corresponding red-channel image to define the analysis area. The counts of positively Deep Red-stained cells within the defined ROI were automatically detected using the Analyze Particles function and then exported for statistical analysis. CFSE-green signals were collected using an Argon ion laser (488nm) and 525 bandpass filter, while Deep Red were detected with Cy5 filter at 40x magnification.

Levels of inflammatory cytokines IL-6, TNF-α, IFN-α, IFN-γ, IL-17A, and IL-10 were measured in peritoneal lavage and plasma samples using ELISA kits (BioLegend, San Diego, CA, USA) following the manufacturer’s instructions. Anti-nuclear antibody (ANA) and anti-double-stranded DNA (anti-dsDNA) antibodies were detected using kits from Alpha Diagnostic International (San Antonio, TX, USA) and Chondrex (Redmond, WA, USA), respectively.

Total RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA, USA) and reverse-transcribed into complementary DNA (cDNA) using iScript gDNA Clear cDNA Synthesis Kit with T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA). Q-PCR was conducted with IQ SYBR Green Supermix (Bio-Rad, Hercules, CA, USA) on the CFX Connect Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) using the following parameters: initial denaturation at 95°C for 2 minutes, followed by 40 cycles of 95°C for 5 seconds, 60°C for 30 seconds, 72°C for 20 seconds. Primer sequences are listed in Table 1 , with GAPDH used as the reference gene. Gene expression levels were quantified using the comparative ΔΔCt method, and RAW264.7 cells without EL4 cell co-incubation were designated as the calibrator group.

Frozen tissue samples (100 mg) were homogenized in grinding ball with two cycles of 60Hz of oscillation frequency for 30 sec using tissue grinders (High-throughput Tissue Grinders NANBEI, NB-48P). Protein lysates were lysed in RIPA buffer (Chemicon, Millipore, USA) and quantified by Assay Dye Reagent Concentrate (Bio-Rad #5000006). Proteins were separated by SDS-PAGE, transferred onto polyvinylidene difluoride (PVDF) membranes (Immobilon, Millipore, Eschborn, Germany) using Wet/Tank Blotting Systems (Bio-Rad, Hercules, CA, USA), and blocked with 3% BSA in TBS (pH 7.4) for 1 hour at room temperature. The membranes were incubated overnight at 4°C with primary antibodies against MerTK (AF591, R&D), p-MerTK (p186-749, PhosphoSolutions), Gas6 (PA5-72882, Thermo Fisher Scientific), p62 (A11250, ABclonal), LC3B (E-AB-70053, Elabscience), PPARγ (E-AB-60059, Elabscience), LXR (A04523-2, Boster), and actin (3700, Cell Signaling). After washing, membranes were probed with secondary antibodies (Elabscience), and detection was performed using ECL Plus reagent (PerkinElmer, MA, USA) on a BOX Chemi XRQ imaging system (Syngene, Cambridge, UK). Protein intensity was normalized to actin in each lane first and then calculated fold change relative to compared to the control condition.

Statistical analyses were conducted using GraphPad Prism software. Data are expressed as mean ± standard error of the mean (SEM) unless otherwise stated. The Shapiro-Wilk test was used to assess normality of data distribution. For comparisons between two independent groups, an unpaired Student’s t-test was performed for normally distributed data, whereas the Welch’s t-test was applied when homogeneity of variances was not met. For non-normally distributed data, the Mann-Whitney U test was utilized. For multiple-group comparisons, Welch’s ANOVA, followed by an appropriate post hoc test, was applied when variances were unequal. For non-normally distributed data, the Kruskal-Wallis test, followed by Dunn’s post hoc test, was used for multiple comparisons. Statistical significance was defined as p < 0.05.

To investigate the underlying mechanism by which HCQ may slow the progression of SLE, a mouse model was established using pristane to induce lupus-like symptoms. A total of 21 female mice were randomly assigned to one of three experimental groups. Group 1 served as the control, receiving sterile saline injections (normal group, n=7); Group 2 was administered pristane to induce a lupus phenotype (PIL group, n=7); and Group 3 was comprised of pristane-induced lupus (PIL) mice treated with HCQ (PIL+HCQ group, n=7). HCQ was given orally at a dose of 3 mg/kg per day, beginning on day 1 after PIL induction, with daily treatment continuing for a period of 180 days to assess long-term effects on disease progression. To evaluate the preventive effects of HCQ, plasma and urine samples were collected at 1, 3, and 6 months post-PIL induction and analyzed for inflammatory cytokine levels, anti-double-stranded DNA (anti-dsDNA) antibody titers, and proteinuria as indicators of disease activity and progression. After 6 months, all mice were sacrificed to permit in-depth examination of immune complex deposition and lupus nephritis through histopathological and immunological analysis ( Figure 1A ). Our results revealed that HCQ treatment had a significant suppressive effect on pro-inflammatory cytokine production, including interleukin-6 (IL-6) (p=0.006) and tumor necrosis factor-alpha (TNF-α) (p=0.0012) in peritoneal lavage of PIL mice at the 6-month time point ( Figure 1B ). Additionally, HCQ significantly reduced circulating anti-dsDNA antibody (p<0.0001), anti-nuclear antibody (ANA) (p<0.0041), and total IgG levels (p=0.0003), a hallmark of lupus, at 6 months post-PIL induction ( Figure 1C , Supplementary Figure S1A ). Histological examination showed that HCQ treatment led to a substantial reduction in immune complex deposition within the kidneys ( Figure 1D ) and complement C3 deposition ( Supplementary Figure S1B ), which correlated with an alleviation of proteinuria (p=0.0385) when compared to untreated PIL mice ( Figure 1E ). These findings indicate that HCQ effectively mitigates inflammatory signals, especially IL-6 and TNF-α, and protects against renal damage associated with lupus nephritis.

Given the observed reduction of pro-inflammatory cytokines IL-6 and TNF-α in PIL mice treated with HCQ, we hypothesized that HCQ enhances macrophage-mediated efferocytosis, thus promoting anti-inflammatory signaling pathways. To investigate this, we used murine macrophage-like RAW264.7 cells as an in vitro model. Cytotoxicity assays revealed that the half-maximal inhibitory concentration (IC50) of HCQ in RAW264.7 cells after 24 hours was 14 μM, while the IC50 of CQ was notably lower at 5 μM, indicating HCQ’s relatively milder cytotoxicity profile ( Figures 2A, B ). To evaluate the effect of HCQ on macrophage efferocytosis, apoptotic cells were generated by UV irradiation of EL4 cells and subsequently prepared for co-incubation with macrophages ( Supplementary Figure S2 ). RAW264.7 cells were labeled with CSFE and treated with sub-toxic concentrations of HCQ and CQ (2.5, 5, and 10 μM) for 24 hours, followed by co-incubation with apoptotic EL4 cells labeled with MitoTracker Deep Red to facilitate tracking. Efferocytosis was quantified by flow cytometry, with the efferocytic macrophages defined as the percentage of double-positive MitoTracker Deep Red and CFSE cells (Deep Red⁺/CFSE⁺) out of the live macrophages (Zombie^-/CFSE⁺). HCQ treatment significantly enhanced macrophage efferocytosis in a dose-dependent manner, with a pronounced increase observed at 5 μM (p=0.0014). Similarly, CQ treatment showed significant efferocytic enhancement at 5 μM (p<0.0001) ( Figures 2C, D , Supplementary Figure S3 ). We utilized pHrodo-Red, a pH-sensitive fluorescent dye that selectively fluoresces in acidified lysosomes, to label apoptotic cells and assess efferocytosis in macrophages. Following a 4-hour co-incubation period, flow cytometric analysis revealed a significant increase in the proportion of pHrodo-positive RAW264.7 cells (pHrodo⁺/CFSE⁺). Treatment with HCQ led to a marked elevation in pHrodo fluorescence (p=0.0072 at 5 μM), indicating HCQ facilitates the uptake and lysosomal processing of apoptotic cells by macrophages ( Figure 2E ). To visually confirm these observations, immunofluorescence analysis was performed, which demonstrated an increased phagocytosis of apoptotic cells to HCQ-treated macrophages compared to untreated controls (p=0.0206 at 5 μM), further supporting the efferocytic capacity induced by HCQ treatment ( Figure 2F ). These findings indicate that HCQ, within sub-toxic dose ranges, effectively enhances macrophage efferocytosis.

To further elucidate the mechanism by which HCQ enhances efferocytosis, we analyzed the gene expression profiles of apoptotic cell recognition receptors on macrophages, focusing on a variety of receptors essential for efferocytosis. These included Mertk, Axl, Tyro3, Tim4, Itgb3 (integrin β3), and Cd36, both in the absence and presence of apoptotic cells. Additionally, we assessed the gene expression of soluble phosphatidylserine (PtdSer)-binding proteins, namely Pros (protein S), Gas6, and Mfge8, which facilitate apoptotic cell recognition and clearance. Our findings reveal that HCQ treatment at a concentration of 10 μM significantly upregulates Mertk gene expression in RAW264 macrophages, under conditions both with (p=0.0179) and without (p=0.0055) apoptotic cell co-incubation, indicating a consistent HCQ effect on this pivotal efferocytosis receptor ( Figure 3A ). HCQ also markedly increased the expression of Gas6, a key ligand for TAM receptors, with significant upregulation observed at 10 μM both in the absence (p=0.002) and presence (p=0.0202) of apoptotic cells ( Figure 3C ). While increases in Axl and Pros expression were observed, these trends did not reach statistical significance following apoptotic cell co-incubation ( Figures 3B, D ). Other gene expressions of efferocytosis-related molecules (Tyro3, Tim4, Itgb3, Cd36, Mfge8) in macrophages were not changed after HCQ treatment. At the cell surface level, flow cytometric analysis revealed that HCQ also significantly increased the surface expression of MerTK on macrophages, observed at 5 μM(p=0.045), 7.5 μM (p=0.0346), and 10 μM (p=0.0144) in the absence of apoptotic cells ( Figure 3E ). After expose to apoptotic cells, HCQ-treated macrophages also exhibited increased surface MerTK expression. This upregulation is particularly notable, as it highlights HCQ’s ability to enhance MerTK surface availability, which is essential for efficient efferocytosis. The reduced MerTK levels in macrophages treated with 10 μM HCQ were attributed to the deterioration of death following co-incubation with apoptotic cells. Additionally, HCQ administration effectively inhibited autophagy, as evidenced by increased protein levels of p62 and LC3B, markers indicative of disrupted autophagic flux ( Figure 3F ). Notably, the levels of p-MerTK and Gas6 was elevated in a dose-dependent manner following HCQ treatment, suggesting that HCQ modulates MerTK and Gas6 proteins at both transcriptional and translational levels ( Figure 3F ). These results support the notion that HCQ upregulates MerTK and Gas6, highlighting HCQ’s role in promoting macrophage-mediated clearance of apoptotic cells through efferocytic pathways.

The impact of HCQ on enhancing efferocytosis and MerTK expression was further examined in a PIL mouse model. Peritoneal macrophages from four experimental groups: untreated control, HCQ-treated mice (30 mg/kg daily by oral gavage), PIL-induced mice, and PIL-induced mice treated with HCQ were isolated 7 days following PIL induction. After a 30-minute co-incubation with Mitotracker Deep Red-labeled apoptotic thymocytes, efferocytotic capacity was defined as the percentage of double-positive Deep Red and F4/80 cells (Deep Red⁺/F4/80⁺) out of the live macrophages (Zombie^-/F4/80⁺) and quantified using flow cytometry. The PIL model led to a marked reduction in macrophage efferocytosis, decreasing from a baseline of 57.5% ± 4.8% in the control group to 41.6% ± 3.6% (p=0.0002). Peritoneal macrophages isolated from HCQ-treated mice exhibited enhanced efferocytosis (68.9% ± 5.1%) and showed modest improvement in efferocytosis levels following PIL treatment, achieving 50.6% ± 4.0% (p=0.0259), suggesting a protective effect of HCQ against PIL-mediated impairment of efferocytosis ( Figure 4A ). Efferocytosis capacity of peritoneal macrophages were further confirmed by a 4-hour attachment in a culture dish and subsequently exposed to apoptotic thymocytes. Immunofluorescence staining demonstrated enhanced efferocytosis of peritoneal macrophages in the PIL+HCQ group compared to the PIL group alone (p<0.0001), indicating improved efferocytosis following HCQ treatment ( Figure 4B ). In vivo condition, pristane treatment resulted in a substantial reduction in the total number of peritoneal macrophages ( Supplementary Figure S4 ). HCQ administration significantly increased surface MerTK expression in the peritoneal macrophage population, regardless of pristane exposure ( Figure 4C ). In the spleen, HCQ administration enhanced MerTK (p=0.0174) and Gas6 protein expression (p=0.0034) following pristane treatment ( Figure 4D ). Further analysis of apoptosis levels revealed that HCQ treatment significantly reduced the proportion of apoptotic cells in both the peritoneal lavage and spleen, suggesting modulatory effect of HCQ in eliminating apoptotic cells within these immune compartments ( Supplementary Figure S5 ). Beyond cellular markers, HCQ treatment alleviated splenomegaly induced by pristane (p<0.0001), suggesting that HCQ mitigates inflammation linked to SLE pathology ( Figure 4E ). HCQ significantly reduced plasma IL-17A levels ( Figure 4F ) and Il17a gene expression in the peritoneal lavage following PIL treatment (p<0.0001) ( Supplementary Figure S6A ). In addition, HCQ markedly increased gene expression of anti-inflammatory cytokine Il10 in peritoneal macrophages post-PIL (p=0.0038) ( Supplementary Figure S6B ). By suppressing IL-17A and enhancing IL-10 production as early as one week after pristane exposure, HCQ appears to attenuate pro-inflammatory signals, potentially contributing to decreased autoantibody production. These findings support the anti-inflammatory effects of HCQ at early stage of actions in vivo.

To elucidate the role of MerTK in HCQ-enhanced efferocytosis, both RAW264.7 cells and peritoneal macrophages were treated with the specific MerTK inhibitor, UNC2025. Western blot analysis verified that UNC2025 effectively suppressed MerTK phosphorylation ( Figure 5A ), which corresponded with a significant reduction in HCQ-mediated increase in efferocytosis in RAW264.7 cells (p<0.0001) ( Figure 5B ). Similarly, peritoneal macrophages isolated from PIL mice exhibited impaired efferocytosis capacity, and the enhancement of efferocytosis by HCQ was attenuated upon MerTK inhibition ( Figure 5C ). After co-incubation with apoptotic cells, gene expressions of pro-inflammatory cytokines Ifnα and Il6 were upregulated in macrophages compared to untreated controls. HCQ treatment significantly reduced the expression of both Ifnα and Il6 (p<0.0001), an effect that was reversed after MerTK inhibition (p=0.0011 and p=0.001, respectively) ( Figures 5D, E ). In the PIL mice model, peritoneal macrophages exposed to apoptotic cells exhibited significantly increased Il6 expression, which was suppressed by HCQ treatment. This HCQ-mediated reduction in Il6 expression was reversed by MerTK inhibition with UNC2025 in both normal and PIL mice (p=0.0001 and p<0.0001, respectively) ( Figure 5G ). In contrast, exposure to apoptotic cells led to increased Il10 expression in macrophages, and this response was further enhanced by HCQ treatment ( Figure 5F , Supplementary Figure S6B ). Inhibition of MerTK signaling with UNC2025 significantly attenuated Il10 expression, and HCQ-induced Il10 expression was restored by UNC2025, indicating that Il10 induction is dependent on MerTK activity ( Figure 5F ). Additionally, HCQ treatment upregulated the protein expression of anti-inflammatory transcription factors, PPARγ and LXR, which are key regulators of macrophage-mediated immune modulation. Treatment of UNC2025 significantly suppressed HCQ-induced PPARγ and LXR following exposure to apoptotic cells (p<0.0001) ( Figure 5H ). These findings demonstrate that HCQ enhances efferocytosis and anti-inflammatory responses via MerTK activation, promoting IL-10 upregulation and IL-6 downregulation in macrophages.