The prevalence of obesity is increasing, and the coexistence of obesity and systemic lupus erythematosus (lupus) is possible. A high-fat diet (HFD) was orally administered for 6 months in female 8-week-old Fc gamma receptor IIb deficient (FcgRIIb−/−) lupus or age and gender-matched wild-type (WT) mice. Lupus nephritis (anti-dsDNA, proteinuria, and increased creatinine), gut barrier defect (fluorescein isothiocyanate dextran), serum lipopolysaccharide (LPS), serum interleukin (IL)-6, liver injury (alanine transaminase), organ fibrosis (liver and kidney pathology), spleen apoptosis (activated caspase 3), and aorta thickness (but not weight gain and lipid profiles) were more prominent in HFD-administered FcgRIIb−/− mice than the obese WT, without injury in regular diet-administered mice (both FcgRIIb−/− and WT). In parallel, combined palmitic acid (PA; a saturated fatty acid) with LPS (PA + LPS) induced higher tumor necrotic factor-α, IL-6, and IL-10 in the supernatant, inflammatory genes (inducible nitric oxide synthase and IL-1β), reactive oxygen species (dihydroethidium), and glycolysis with reduced mitochondrial activity (extracellular flux analysis) when compared with the activation by each molecule alone in both FcgRIIb−/− and WT macrophages. However, the alterations of these parameters were more prominent in PA + LPS-administered FcgRIIb−/− than in the WT cells. In conclusion, obesity accelerated inflammation in FcgRIIb−/− mice, partly due to the more potent responses from the loss of inhibitory FcgRIIb against PA + LPS with obesity-induced gut barrier defect.

© 2022 The Author(s). Published by S. Karger AG, Basel

Introduction

The incidence of obesity-related deaths has considerably increased during the past decades [1], as it is a risk factor of diabetes, hypertension, hyperlipidemia, and adverse cardiovascular outcomes, including myocardial infarction and heart failure [2]. Obesity can also cause chronic activation of inflammatory processes, leading to atherosclerosis and other major vascular complications [3] through multiple mechanisms, including hypoxia and apoptosis of adipocytes from cell hypertrophy [4-6], reduced adiponectin with leptin elevation [7], saturated fatty acid-induced mitochondrial dysfunctions [8], and metabolic endotoxemia from gut barrier defects [9]. Among these, immune responses against endotoxins may lead to the most potent response because immune activation by the organism’s pathogen-associated molecular patterns is usually more robust than the responses toward the host cell’s damage-associated molecular patterns [10]. The activation of toll-like receptor 4 (TLR-4) by the lipid component, especially saturated fatty acid [11], might induce intestinal inflammation that leads to gut barrier defect and metabolic endotoxemia, increasing serum lipopolysaccharide (LPS) with systemic inflammation [12]. Indeed, LPS is a major cell wall component of Gram-negative bacteria (the most abundant gut organisms) with molecular weight (MW) between 10 and 100 kDa. LPS is usually too large to pass through the healthy gut barrier which allows the passive transport of molecules with MW less than 0.6 kDa [13]. Obesity and a high-fat diet (HFD) can induce gut dysbiosis [14], an alteration in intestinal organisms [15] that causes gut-mucosal barrier defect (leaky gut), leading to translocation of high-MW molecules, such as LPS, from the gut into the systemic circulation, so-called gut translocation [16, 17]. Such LPS translocation is also involved in the interplay between a high-fat diet and gut dysbiosis in the pathogenic mechanisms of nonalcoholic fatty liver disease or steatohepatitis, a serious consequence of obesity [18].

Systemic lupus erythematosus (SLE) is an autoimmune disease influenced by environmental factors and genetic defects, including the dysfunction polymorphism of the Fc gamma receptor IIb (FcgRIIb) gene (the only inhibitory receptor in the FcgR family), especially in the Asian populations [19-21]. Despite the non-prominent gastrointestinal symptoms in patients with lupus, peritonitis is found in 60–70% of postmortem autopsy in patients with active lupus [22], partly due to intestinal immune complex deposition, resulting in gut permeability defect and endotoxemia in active lupus [23]. Because of the specificity of anti-dsDNA autoantibody in lupus, the formation of an immune complex between anti-dsDNA and nucleic acid might be the most common component in lupus pathogenesis [24].

As gut translocation of endotoxin induces systemic inflammation in both obesity [14] and lupus [23], both diseases might have some connection. As such, there was a significant increase in the risk of SLE development among women with obesity compared to those with average body weight [25]. Moreover, SLE patients with obesity are at higher risk of developing disability [26] and insulin resistance [27-29]. Because obesity causes systemic inflammation (partly from endotoxemia), an important factor for lupus exacerbation, inflammation might be a key factor in enhancing disease severity among patients with combined lupus and obesity. Because both lupus and obesity can induce gut dysbiosis, causing leaky gut, endotoxin translocation [12, 30, 31], and worsening systemic inflammation, the combination of both conditions might additively enhance systemic inflammation. Although there are several observational data on the additive adverse effects of both conditions [25, 26], understanding of the direct correlation between obesity and lupus is still lacking. Therefore, systemic inflammation was tested in a FcgRIIb deficient (FcgRIIb−/−) lupus mouse model with and without HFD-induced obesity. We also conducted the in vitro macrophage experiments to test the impact of endotoxin and saturated fatty acid on lupus condition.

Materials and MethodsAnimals and Animal Model

Animal care and use protocol was approved by the Institutional Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (SST 025/2563) in compliance with the US National Institutes of Health standards. As such, FcgRIIb−/− mice on a C57BL/6 background were provided by Dr. Silvia Bolland (NIAID, NIH, Maryland, USA). The wild-type (WT) mice were purchased from the Nomura Siam International (Pathumwan, Bangkok, Thailand). Due to age-dependent lupus characteristics [20, 21, 32, 33], 8-week-old FcgRIIb−/− female mice represented asymptomatic lupus-prone mouse model, and age-matched female WT mice were used in all experiments. The HFD containing fat, mostly from lard (60% w/w), with energy content calculated at 8.64 kcal/g [34] were used. The regular mouse diet was a standard laboratory chow containing fat (4.5% w/w) with energy content calculated at 3.04 kcal/g (Mouse Feed Food No. 082, CP Company, Bangkok, Thailand). Blood was collected through tail vein nicking (baseline and 3 months post-HFD) and via cardiac puncture under isoflurane anesthesia at sacrifice (6 months post-HFD). Organ tissues were collected after sacrifice and processed in 10% neutral buffered formalin fixation for histology evaluation.

Analysis of Mouse Blood Samples and Organs

After fasting for 12 h, lipid profiles were measured by the quantification assay for triglyceride (TG), total cholesterol (Sigma-Aldrich, St. Louis, MO, USA), low-, and high-density lipoprotein cholesterol (LDL and HDL) (Crystal Chem Inc., Downners grove, IL, USA). Serum creatinine (QuantiChrom Creatinine-Assay, DICT-500; BioAssay, Hayward, CA, USA), proteinuria, and serum anti-dsDNA were measured for lupus characteristics. Urine protein creatinine index (UPCI) was used to represent the severity of proteinuria according to the following equation: UPCI = urine protein (mg)/urine creatinine (mg/dL). Urine protein and creatinine were measured by Bradford Bio-Rad Protein Assay (Bio-Rad; Hercules, CA, USA) and QuantiChrom Creatinine-Assay (DICT-500) (BioAssay), respectively. The anti-dsDNA was evaluated with a protocol using coated-calf thymus DNA (Invitrogen, Carlsbad, CA, USA) [21]. Liver damage and serum cytokines were determined by EnzyChrom Alanine Transaminase assay (EALT-100; BioAssay) and enzyme-linked immunosorbent assays (ELISA) (Invitrogen), respectively. In the detection of cytokines from organ tissue, the samples were weighed, cut, thoroughly sonicated (High Intensity Ultrasonic Processor, Newtown, CT, USA) in 500 mL of ice-cold phosphate buffer solution (PBS)-containing protease inhibitor Cocktail (I3786; Sigma-Aldrich), and the cytokines were measured from the supernatant by ELISA (Invitrogen).

For histology, paraffin-embedded sections (4 µm thick) stained by different colors from 10% formalin-fixed samples were evaluated. There were several scoring systems in different organs with Hematoxylin and Eosin (H&E) color as follows; (i) livers (obesity-induced liver damage) [35]; steatosis (0–3), lobular inflammation (0–3), and hepatocellular ballooning degeneration (0–2), (ii) subcutaneous fat thickness [36], (iii) kidney (renal injury score defined from the area per high power fields with pathological characteristics; tubular epithelial swelling, loss of brush border, vacuolar degeneration, necrotic tubules, cast formation, and desquamation) [21]: 0, area of damage <5%; 1, area of damage 5–10%; 2, area of damage 10–25%; 3, area of damage 25–50%; and 4, area of damage >50%, and (iv) aorta thickness was evaluated in the cross-sectional and longitudinal sections at aortic arch and 0.5 cm above renal arterial branching. Fibrosis in organs (livers and kidneys) and aortic atherosclerosis were determined by Masson’s trichrome staining with the fibrosis area measurement using computerized image analysis software (ImageJ©software, Bethesda, MD, USA) with a ×200 magnification field, 10 fields per sample [37].

Gut Leakage Measurement

Gut permeability was determined by fluorescein isothiocyanate dextran (FITC-dextran) assay and endotoxemia [38-40]. FITC-dextran, a nonabsorbable molecule with 4.4 kDa molecular weight (Sigma-Aldrich) at 12.5 mg per mouse, was orally administered at 3 h before detecting FITC-dextran in serum by Fluorospectrometer (NanoDrop 3300; ThermoFisher Scientific, Wilmington, DE, USA). Serum endotoxin (LPS) was measured by HEK-Blue LPS Detection (InvivoGen, San Diego, CA, USA). The data were recorded as 0 when LPS values were less than 0.01 EU/mL because of the limited lower range of the standard curve.

RNA Sequencing Analysis from LPS-Activated Macrophages

Bone marrow-derived macrophages were prepared from mice as previously described [41]. Briefly, the bone marrow from the long bone of mice (tibia and fibula) was collected by centrifugation at 6,000 rpm for 4°C and incubated for 7 days with modified Dulbecco’s Modified Eagle’s Media (DMEM) in a humidified 5% CO2 incubator at 37°C. Conditioned media of the L929 cell line, containing macrophage-colony stimulating factor, at 20% weight by volume (w/v), was used to induce macrophages from the pluripotent stem cells. Then, macrophages at 1 × 105 cells/well were incubated with LPS (Sigma-Aldrich) (100 ng/mL) for 6 h before processing with the RNA sequencing of BGISEQ-50 platform by the BGI Company [42]. The sequencing quality was determined using Fast-QC. The abundance transcripts were quantified by Kallisto, and the transcripts were converted to genes by the Tx-import r package. The read count data were determined for differentially expressed genes (DEGs) using edge-R. The up- and down-regulated genes were analyzed in a biological process (GO Term) using Enrich-r. The gene lists associated with glycolysis, mitochondrial oxidative phosphorylation, and ATP synthesis were selected based on the GO Term of DEGs, and the heatmaps were generated by the p-heatmap r package.

Lipid Burdens, Cytokines, and Gene Expression in Macrophages with LPS and/or Saturated Fatty Acid

Macrophages at 1 × 105 cells/well were incubated with DMEM control media or palmitic acid (PA; a representative saturated fatty acid) or oleic acid (a representative monounsaturated fatty acid) (Sigma-Aldrich) at 0.5 mM/well or LPS from E. coli O26:B6 (Sigma-Aldrich) at 100 ng/mL or PA with LPS (PA + LPS) for 18 h. Then, lipid abundance, supernatant cytokines, gene expressions, and cell energy were evaluated. Accumulation of lipid droplets (Oil-Red-O staining) [43] was determined to test lipid metabolism in the activated macrophages [8]. Briefly, activated cells were washed twice with PBS before staining with 0.3% Oil-Red-O solution (Sigma-Aldrich) for 10 min, fixed with 15 min 4% paraformaldehyde, and evaluated under a microscope with 10 random fields from each well using ImageJ (NIH) for the color intensity evaluation.

Gene expression was identified by real-time polymerase chain reaction (PCR) using the total RNA (RNA-easy mini kit; Qiagen, Hilden, Germany) and reverse transcription assay (Applied Biosystems, Warrington, UK) on Applied Biosystems QuantStudio 6 Flex Real-Time PCR System with SYBR® Green PCR Master Mix (Applied Biosystems). The comparative threshold (delta-delta Ct) method (2−∆∆Ct) normalized by β-actin (an endogenous housekeeping gene) was performed. The primers of target genes are listed in Table 1.

Table 1.

List of primers used in study

Reactive Oxygen Species, Total ATP, and Mitochondrial Staining in Macrophages

Because reactive oxygen species (ROS) are byproducts of mitochondrial activity and ATP production [44], several parameters are measured in 18 h-activated macrophages. Total ROS production was determined by dihydroethidium (DHE) fluorescent dye [8], using 20 min of DHE (20 µM) (Sigma-Aldrich) incubation at 37°C before DHE measurement at 520 nm by a microplate reader (ThermoFisher Scientific). In parallel, ATP content [8] was identified by Luminescent ATP Detection Assay (Abcam, San Francisco, CA, USA) with a microplate reader. Mitotracker Red CMxROS or Mitotracker green (Molecular probe) is used for fluorescent staining for mitochondrial mass and mitochondrial function (membrane potential), respectively [8]. As such, each color was added to each well at 37°C for 15 min, fixed with cold methanol at −20°C for 15 min, washed twice with PBS, and photographed by an inverted microscope (Olympus, Tokyo, Japan). Additionally, mitochondrial ROS was evaluated by mitochondrial ROS detection assay (Cayman chemical, Ann Arbor, MI, USA) according to the manufacturer’s protocol.

Extracellular Flux Analysis (without and with the Manipulations of Cell Energy)

To explore the continuous cell-energy alteration after LPS activation in control and PA-activated macrophages, the cell-energy measurement of 18 h-preconditioning macrophages (by PA or DMEM control) was started immediately after LPS administration in the Seahorse XFp Analyzers (Agilent, Santa Clara, CA, USA). The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) represent mitochondrial function (respiration) and glycolysis activity, respectively, in Seahorse XFp Analyzers (Agilent). The preconditioning macrophages (18 h of PA or DMEM control) at 1 × 105 cells/well were incubated with LPS 100 ng/mL in the Seahorse system. The real-time OCR and ECAR were measured and analyzed by Seahorse Wave 2.6 software without and with manipulating cell energy. For cell-energy intervention, several agents, including glucose, oligomycin, 2-Deoxy-D-glucose (2-DG), and rotenone/antimycin A, were sequentially injected according to the manufacturer’s protocol [45-47]. Data from Seahorse Wave 2.6 software based on the following equations: maximal respiration = OCR between FCCP and rotenone/antimycin A–OCR after rotenone/antimycin A; respiratory reserve = OCR between FCCP and rotenone/antimycin A–OCR before oligomycin; glycolysis = ECAR between glucose and oligomycin; maximal glycolysis (glycolysis capacity) = ECAR between oligomycin and 2-DG–ECAR after 2-DG; glycolysis reserve = ECAR between oligomycin and 2-DG–ECAR between glucose and oligomycin.

Separation of Blood Mononuclear Cells and the Activation by LPS and/or Saturated Fatty Acid

Due to the possibility of the interference of culture processes of bone marrow-derived macrophages and the immune activity in lupus mice on the cell activities, the isolation of peripheral blood mononuclear cells (PBMC) from 3 months old mice by Ficoll-Paque density gradient centrifugation was performed following a previous publication [48]. Briefly, the heparinized mouse blood was diluted with an equal volume of PBS, pH 7.4 before mixing with the same volume of Ficoll-Paque (Robbins Scientific Corporation, Sunnyvale, CA, USA) and centrifuged at 600 g for 30 min at room temperature in a swinging bucket rotor (Backman Coulter, USA) without the brake applied. Then, the PBMC interface was carefully removed by pipetting and washed with PBS by centrifugation at 600 g for 5 min before suspending in lysing buffer for 5 min at room temperature with gentle mixing to lyse contaminating red blood cells. Subsequently, the PBMC at 1 × 105 cells/well was used for the activation in the same procedures for bone marrow-derived macrophages as previously mentioned.

Statistical Analysis

Mean ± standard error (SE) was used for data presentation. The differences between groups were examined for statistical significance by one-way analysis of variance (ANOVA) followed by Tukey’s analysis or Student’s t test for comparisons of multiple groups or 2 groups, respectively. All statistical analyses were performed with SPSS 11.5 software (SPSS, IL, USA) and Graph Pad Prism version 7.0 software (La Jolla, CA, USA). A p value of <0.05 was considered statistically significant.

ResultsObesity Facilitated More Prominent Systemic Inflammation and Lupus Characteristics in FcgRIIb−/− Mice

Although HFD similarly induced obesity in WT and FcgRIIb−/− mice as early as 4 months of the experiments, lupus characteristics (anti-dsDNA, serum creatinine, and proteinuria), permeability defect (FITC-dextran assay and endotoxemia), and systemic inflammation (serum interleukin (IL)-6) were more prominent in obese FcgRIIb−/− mice than regular diet (RD) FcgRIIb−/− mice (Fig. 1a–g). In FcgRIIb−/− mice, anti-dsDNA, endotoxemia, and serum IL-6 increased as early as 3 months after HFD administration. The renal injury (serum creatinine and proteinuria) increased later at 6 months post-HFD (Fig. 1b–g), supporting the possibility that renal anti-dsDNA deposition and systemic inflammation as the causes of lupus nephritis [21]. Serum anti-dsDNA in RD FcgRIIb−/− mice at 3 and 6 months of the experiment were higher than age-matched WT control mice but lower than HFD FcgRIIb−/− mice (Fig. 1b–d), implying an age-dependent spontaneous anti-dsDNA elevation in FcgRIIb−/− mice [23, 39, 49, 50]. Despite non-lupus characteristics in WT mice, HFD WT mice demonstrated gut permeability defect (FITC-dextran and endotoxemia) and higher serum IL-6 at 3 and 6 months post-HFD (Fig. 1b–g).

Fig. 1.

Characteristics of WT or FcgRIIb−/− mice with RD or HFD as determined by the time-course of body weight, lupus activity (serum anti-dsDNA, serum creatinine, and urine protein creatinine index), gut barrier defect (FITC-dextran assay and endotoxemia), and serum IL-6 (a–g) with the parameters at 6 months of the experiments as indicated by fasting blood lipid profile (total cholesterol, low-density lipoprotein, and high-density lipoprotein) (h–j), adipose tissue depots in several sites (mesenteric fat, perigonadal fat, perirenal fat, and subcutaneous) (k–n), and liver injury (serum alanine transaminase) (o) are demonstrated (n = 6–8/time-point or group).

The parameters of obesity, including blood lipid profile (total cholesterol, low- and high-density lipoprotein) and fat deposit in several sites (mesentery, perigonad, perirenal, and subcutaneous fat), were not different between FcgRIIb−/− and WT mice with HFD, while liver enzyme (alanine transaminase) was more prominent in FcgRIIb−/− mice (Fig. 1h–o). Additionally, obesity-induced liver injury in HFD FcgRIIb−/− mice was more severe than obese WT mice as indicated by liver weight, liver fibrosis area, hepatocyte apoptosis, and liver tumor necrotic factor-α (TNF-α) but not liver injury score (H&E staining) and other liver cytokines (IL-6 and IL-10) (Fig. 2a–g, 3). Similarly, organ damage (kidney, spleen, and aorta) in obese FcgRIIb−/− mice was more prominent than obese WT mice as indicated by renal injury score (H&E staining), fibrosis area, renal cytokines (TNF-α, IL-6, and IL-10), spleen apoptosis, and abdominal aorta thickness (but not at the aortic arch) (Fig. 2h–o, 4, 5, 6). In the RD groups, FcgRIIb−/− mice demonstrated higher renal injury score, renal TNF-α, and spleen apoptosis than WT mice (Fig. 2h, j, m) possibly be associated with age-related lupus characteristics of FcgRIIb−/− mice [49, 51]. Notably, atherosclerosis lesions were demonstrated only in obese FcgRIIb−/− mice but not WT mice (Fig. 6).

Fig. 2.

Characteristics of WT or FcgRIIb−/− mice with RD or HFD as determined by the parameters at 6 months of the experiments of liver injury, including liver weight, H&E stained liver injury score, Masson’s Trichrome stained liver fibrosis area, activated caspase 3 stained hepatocyte apoptosis, and liver cytokines (a–g), kidney injury (H&E stained injury score, Masson’s Trichrome fibrosis area, and kidney cytokines) (h–l), spleen apoptosis (m), and aorta thickness at the abdominal aorta and aortic arch (n, o) are demonstrated (n = 6–8/time-point or group).

Fig. 3.

Representative pictures of livers with different color staining, including H&E, Masson’s Trichrome (Masson), activated caspase 3 (Caspase 3) from WT or FcgRIIb−/− mice with RD or HFD (original magnification, ×20).

Fig. 4.

Representative pictures of kidneys with different color staining, including H&E and Masson’s Trichrome (Masson) from WT or FcgRIIb−/− mice with RD or HFD (original magnification, ×20 and ×40).

Fig. 5.

Representative pictures of spleen apoptosis with activated caspase 3 staining from WT or FcgRIIb−/− mice with RD or HFD (original magnification, ×20).

Fig. 6.

Representative pictures of aortas with different color staining, including H&E, Masson’s Trichrome (Masson), activated caspase 3 (Caspase 3) from WT or FcgRIIb−/− mice with RD or HFD (original magnification, ×4 and ×10).

Prominent Additive Pro-Inflammatory Effect of Saturated Fatty Acid with LPS in FcgRIIb−/− Macrophages

Because cell energy is important for cytokine production and fat metabolism is an important cellular energy source [52], LPS activation might be associated with lipid metabolisms. From RNA sequencing analysis, LPS activation in FcgRIIb−/− macrophages alter the expression of several genes when compared with control FcgRIIb−/− macrophages as indicated by heatmap and a volcano plot with the list of genes in several categories (Fig. 7a–e). With the comparison between LPS-activated versus control FcGRIIb−/− macrophages, there were 1,647 up-regulated and 2,064 down-regulated genes, respectively, which could be divided into several clusters using unsupervised hierarchical clustering (Euclidean distance; Ward.D2 method) as evaluated from the DEGs (Fig. 7a, b). The clustering on DEGs demonstrated some differences between LPS activation versus control FcgRIIb−/− macrophages (Fig. 7a). The categorization of DEGs demonstrated the highest difference between up- versus down-regulated DEGs numbers in the metabolic pathways after LPS stimulation with the down- and up-regulated genes in approximately 280 and 140 DEGs numbers (Fig. 7c). However, the up-regulated DEGs in other pathways, such as AMPK, TNF, FoxO, and HIF-1, were more prominent than the down-regulated genes after LPS stimulation (Fig. 7c). Then, the genes that might be associated with metabolic pathways were categorized into several groups with the color code in Figure 7d. Among these groups, there were 1,390 metabolism-associated genes and 147 genes were correlated with the lipid metabolism after LPS activation (Fig. 7d), at least in part, supporting LPS-associated lipid metabolisms [53]. Moreover, LPS interfered with the lipid in the categories of overall lipid metabolism and fatty acid processes (biosynthesis, elongation, and degradation) (Fig. 7e).

Fig. 7.

The RNA sequencing analysis on macrophages from Fc gamma receptor IIb deficient (FcgRIIb−/−) and WT mice indicates an impact of LPS of FcgRIIb−/− macrophages by heatmap of gene expression, Volcano plot, enrichment pathways, and the number of genes expression as categorized by functions (a–d) are demonstrated. e The lists of lipid metabolism-associated genes are expanded compared to the RNA sequencing analysis of WT macrophages. The color scale-bars are demonstrated by log2 fold change. IBD, inflammatory bowel disease.

Because LPS alone could induce cellular lipid alteration, the administration of free fatty acid with LPS might induce the synergistic effects. Indeed, lipid accumulation in macrophages (both WT and FcgRIIb−/− cells) was similar between the incubation by LPS and PA (a representative of saturated fatty acid); however, the lipid contents after combined PA with LPS (PA + LPS) was more profound than the stimulation by LPS or PA alone (both WT and FcgRIIb−/− cells) (Fig. 8a, b). The similar lipid accumulation after PA incubation in WT and FcgRIIb−/− macrophages (Fig. 8a, b) implied the less importance of FcgRIIb on lipid uptake of macrophages [49]. There was no macrophage activity with PA activation alone, while LPS alone induced supernatant cytokines (TNF-α, IL-6, and IL-10), inflammatory genes (inducible nitric oxide synthase (iNOS) and IL-1β), and reactive oxygen species (DHE) (more prominent in FcgRIIb−/− macrophages) (Fig. 8c–j). Only IL-1β and Arginase-1 in PA + LPS WT macrophages were higher than in non-PA-LPS WT cells, while more parameters (supernatant TNF-α, iNOS, IL-1β, and Arginase-1) in PA + LPS FcgRIIb−/− macrophages were higher than non-PA-LPS FcgRIIb−/− cells (Fig. 8c–j). Notably, there was no additive proinflammatory effect of oleic acid (OA, a representative monounsaturated fatty acid) in both WT and FcgRIIb−/− macrophages as indicated by similar levels of supernatant cytokines (TNF-α, IL-6, and IL-10) between LPS versus OA + LPS (Fig. 8k–m). Because the cell differentiation during the bone marrow-derived procedures and the micro-environment in mice at 3 months of the experiments (anti-dsDNA and endotoxemia) (Fig. 1b, f) might interfere with the cell activities, the mononuclear cells from peripheral blood were tested. As such, there was no activation of isolated cells (WT and FcgRIIb−/−) without activations and PA + LPS-stimulated FcgRIIb−/− cells showed the highest level of supernatant cytokines (TNF-α and IL-6 but not IL-10) with the highest expression of proinflammatory genes (iNOS and IL-1β but not Arginase-1 and transforming growth factor β (TGF-β)) compared with PA + LPS-activated WT or LPS-induced FcgRIIb−/− cells (online suppl. Fig. 1; for all online suppl. material, see www.karger.com/doi/10.1159/000526206). The additive effect on PA + LPS when compared with LPS stimulation alone was demonstrated in both WT and FcgRIIb−/− cells and the alterations in mice at 3 months of experiments did not enough to induce proinflammatory activation (online suppl. Fig. 1). These data indicated an additive stimulation of saturated fatty acid on LPS macrophage responses and the more potent inflammatory responses in FcgRIIb−/− macrophages over WT cells (perhaps due to the loss of inhibitory FcgRIIb receptor) [54].

Fig. 8.

Characteristics of WT or Fc gamma receptor IIb deficient (FcgRIIb−/−) macrophages after activation by control media (DMEM), PA saturated fatty acid, LPS, and PA with LPS (PA + LPS) as indicated by Oil-red O-stained lipid-containing cells (representative pictures and abundance) (a, b), supernatant cytokines (TNF-α, IL-6, and IL-10) (c–e), proinflammatory genes (iNOS and IL-1β) (f, g), anti-inflammatory genes (Arginase-1 and TGF-β) (h, i), and DHE oxidative stress (j) are demonstrated. Likewise, the characteristics of macrophages after activation by OA monounsaturated fatty acid, PA, LPS, OA + LPS, and PA + LPS (k–m) are also shown (independent triplicate experiments were performed for all experiments).

Enhanced Cell-Energy Status by Saturated Fatty Acid and Alteration of Cell-Energy Reserve with LPS Stimulation in Macrophages

Continuous cell-energy measurement (extracellular flux analysis) in 18 h pre-conditioning macrophages with PA (PA-primed) or media control before and after LPS stimulation was conducted because of a possible connection between cell-energy status and cell activities [8]. As such, PA-primed macrophages increased mitochondrial activity (OCR) in both WT and FcgRIIb−/− cells, while increased glycolysis (ECAR) only in PA-LPS WT cells (Fig. 9a, b). Meanwhile, LPS similarly increased glycolysis function (ECAR) but not mitochondrial activity (OCR) in both WT and FcgRIIb−/− macrophages (Fig. 9a, b). Energy phenotype profile before- versus after-LPS demonstrated the movement of cell energy from a quiescent state to aerobic state with saturated fatty acid (PA-primed) and toward glycolysis by LPS (both PA-primed and control) without differences between FcgRIIb−/− and WT; indicated by an area under the curve (Fig. 9c, d). There were higher total ATP in LPS-activated macrophages than control cells (both FcgRIIb−/− and WT), higher mitochondrial ROS (only in PA primed + LPS in FcgRIIb−/− cells) without the nonsignificant changes in mitochondrial alteration, abundance, and functions (membrane potential), with (Fig. 9e–g). Total ATP increased by LPS stimulation, while PA priming enhanced cellular ATP only in LPS-stimulated WT cells (Fig. 9e), partially supported PA-facilitated mitochondria [31] and glycolysis-dependent LPS macrophage responses [55, 56]. Interestingly, mitochondrial ROS increased only in PA primed with LPS in FcgRIIb−/− macrophages but not in other groups indicating an additional impact of PA-LPS synergy and the loss of inhibitory signals on mitochondrial injury (Fig. 9f).

Fig. 9.

Characteristics of cell-energy alteration in extracellular flux analysis (without reagents for energy manipulation) in WT or FcgRIIb−/− macrophages preconditioned by PA saturated fatty acid or control media (DMEM) followed by LPS stimulation as indicated by indicators for mitochondrial function, using OCR, and glycolysis, using ECAR, in continuous recording (x-y graph) and at 0.25 h before and after activation (bar graph) (a, b) with energy phenotype profile (c), and AUC of OCR and ECAR (after LPS in relative to before LPS) (d) are demonstrated. In parallel, cell energy as determined by total ATP, mitochondrial ROS, and mitochondrial mass (mitotracker CMxROS) (e–g) is also demonstrated (independent triplicate experiments were performed for all experiments). MFI, mean fluorescent intensity. AUC, area under the curve.

With energy manipulation (extracellular flux reagents), PA priming in WT and FcgRIIb−/− macrophages (without LPS) were similarly higher than control cells (Fig. 10a–c). In FcgRIIb−/− macrophages, LPS and PA priming demonstrated the highest maximal glycolysis capacity (an ability to induce energy from glycolysis pathway after terminating mitochondrial function) and glycolysis reserve with the lowest maximal respiration (an ability to induce energy from mitochondria after stopping glycolysis activity) and respiratory reserve (Fig. 10a–c). The loss of inhibitory FcgRIIb in macrophages might enhance the utilization of cell-energy reserve on glycolysis while minimizing mitochondrial activity (Fig. 10a–c). The higher inflammatory responses of FcgRIIb−/− macrophages over WT cells after PA + LPS stimulation (Fig. 8) might result from the higher maximal glycolysis (Fig. 10a–c) partially supporting glycolysis-dependent cytokine production [55].

Fig. 10.

Characteristics of cell-energy alteration in extracellular flux analysis (with reagents for energy manipulation) in WT or FcgRIIb−/− macrophages preconditioned by PA saturated fatty acid or control media (DMEM) followed by LPS stimulation as indicated by indicators for mitochondrial function, using OCR, and glycolysis, using ECAR (a, b) with the graph presentation of several parameters (c) are demonstrated (independent triplicate experiments were performed for all experiments).

Discussion

An impact of systemic inflammation from obesity-induced endotoxemia was more prominent in FcgRIIb−/− mice than WT, which induced an earlier onset of lupus characteristics in FcgRIIb−/− mice.

High-Fat Diet and Obesity-Induced Inflammation, Neglected Lupus Exacerbating Factor

While the 6 months HFD administration in mice might be a good representative model for the chronic exposure to saturated fatty acids and a clear model of obesity, the influence of nonspecific or non-HFD-associated factors might interfere with the model due to the long duration of the model. For example, the stress from the procedures and the increased age during the experiments might interfere with the results; however, the matched gender and age with good control of the experiments might reduce this general limitation of a model with a long duration experiment. Although body weight of both FcgRIIb−/− and WT mice were significantly higher than the control mice after 4 months post-HFD, serum IL-6, anti-dsDNA (a specific autoantibody for lupus), and endotoxemia in HFD FcgRIIb−/− mice was detectable as early as 3 months post-HFD (before obesity). At 6-month post-HFD in FcgRIIb−/− lupus mice, some characteristics of the mice were different from the control RD-administered FcgRIIb−/− mice, including serum anti-dsDNA, gut barrier defect (FITC-dextran assay and endotoxemia), serum IL-6, liver damage (ALT, apoptosis, fibrosis, and tissue TNF-α), renal injury (serum creatinine, pathological score, fibrosis, and tissue cytokines), spleen apoptosis, and thickness of abdominal aorta. Hence, some lupus characteristics (serum anti-dsDNA and serum creatinine but not proteinuria), not all lupus features, were enhanced by HFD suggesting a possible impact of obesity-induced inflammation in some certain lupus manifestations.

Indeed, saturated fatty acid-induced systemic inflammation (increased c-reactive protein and serum IL-6) both in humans and mice [57-59], through the direct stimulation on TLR-4 [60, 61], amplifying macrophage LPS responses [62], hypertrophic adipocytes [63], and enterocyte injury (gut permeability defect [31] and gut dysbiosis [64]). Perhaps, the similarity between the lipid portion of LPS and saturated fatty acid and dimerization of TLRs are responsible for saturated fatty acid-activated TLR [59]. In parallel, obesity also increases serum LPS, a potent TLR-4 activator, from gut translocation through obesity-induced gut barrier defect [65]. Then, the synergistic TLR-4 activation by LPS and saturated fatty acid might induce more prominent inflammation, especially at 6 months post-HFD. Although obesity-induced inflammation between WT and FcgRIIb−/− mice was similar, the inflammatory responses of FcgRIIb−/− mice were more prominent than WT, perhaps, due to the loss of inhibitory FcgRIIb [33, 39], as indicated by serum IL-6, organ injuries (liver, kidney, and spleen), and aorta thickness (with atherosclerosis lesions). At 6 months of the experiment, FcgRIIb−/− mice with a RD (32-week-old) spontaneously developed anti-dsDNA, mild proteinuria with renal fibrosis (but normal serum creatinine), and spleen apoptosis without other injuries (gut permeability, liver, spleen, and aorta) supported the possible initial phase of lupus nephritis [54]. However, the inflammation in RD FcgRIIb−/− mice was not severe enough to increase serum IL-6, different from HFD mice (FcgRIIb−/− and WT) at 6 months post-HFD. Despite an impact of FcgRIIb on lipid uptake of endothelial cells and hepatocytes [66, 67], fatty liver and atherosclerosis in obese FcgRIIb−/− mice were more severe than HFD WT mice, possibly through other lipid-entry receptors (such as scavenger receptors, hyaluronan, and mannose receptor) [68-71] with inflammation [72].

Saturated Fatty Acid Enhanced the Crosstalk between TLR-4 and Activating FcgRs, Possible Pathogenesis of Enhanced Inflammation in Lupus with Obesity

The main pathogenesis of age-related lupus characteristics in FcgRIIb−/− mice is the loss of inhibitory regulator on plasma cells that causes antibody hyper-production (autoantibodies) generating immune complex deposition in several organs [73]. On the other hand, the FcgRIIb defect in macrophages induces hyper-functioning cells with prominent responses to the stimuli possibly through several pathways. For example, trained immunity, the increased responses to the stimuli due to the pre-conditioning by some molecules, such as beta-glucan (from fungi), Bacille Calmette-Guerin (BCG, a vaccine for tuberculosis), oxidized low-density lipoprotein, and some saturated fatty acids [74-76] and the synergistic responses against several microbial molecules. Here, FcgRIIb−/− macrophages were also hyper-responsive against LPS, partly through the cross-link between activating FcgRs and TLR-4 [77-80], as demonstrated in several models [23]. Due to a variety of ligands of TLR-4, both endogenous (host molecules) and exogenous (pathogen molecules) (such as heat-shock proteins, nucleic acids, saturated fatty acids, and LPS) [81, 82] with the crosstalk between TLR-4 and FcgRs, the molecules from both infection and non-infection might induce TLR-4-associated inflammation.

LPS is associated with lipid metabolism, as demonstrated by altering several genes from RNA sequencing analysis and Oil-red-O staining, similarly in WT and FcgRIIb−/− macrophages. Because saturated fatty acids facilitate TLR-4 dimerization on macrophages [59], which enhances TLR-4 responses by recruiting essential adapter proteins [83], the dimerized TLR-4 by saturated fatty acid might lead to sub-sequentially higher responses (Fig. 11). With PA as a representative saturated fatty acid, combined PA + LPS induced more prominent inflammation than LPS activation alone, and PA-primed macrophages elevated macrophage cell-energy status (Fig. 9c). The PA + LPS altered FcgRIIb−/− macrophages into more proinflammatory M1 polarization (TNF-α, iNOS, and IL-1β) with enhanced cell injury (oxidative stress) compared to WT cells. Notably, FcgRIIb was less important in macrophage lipid-endocytosis (similar Oil-red-O staining between WT and FcgRIIb−/− cells), different from the endothelium and hepatocytes [66, 67]. Furthermore, saturated fatty acid upwardly shifted the energy phenotype profile (Fig. 9c) similarly in both WT and FcgRIIb−/− macrophages, indicating an augmented mitochondrial activity (OCR), possibly because of the energy produced by lipid metabolism [84]. Also, LPS rapidly enhanced glycolysis activity (ECAR) in both WT and FcgRIIb−/− macrophages as the graph (Fig. 9c) shifted to the right (increased glycolysis activity), supporting a previous report on glycolysis as a primary source of macrophage energy after LPS stimulation [85]. With the manipulation on cell energy (extracellular flux reagents), PA with LPS, but not PA alone, induced higher maximal glycolysis capacity and enhanced glycolysis reserve in FcgRIIb−/− macrophages, but not in WT cells, implied an impact of the loss of inhibitory FcgRIIb on cell-energy status. In parallel, PA with LPS in FcgRIIb−/− macrophages profoundly reduced maximal mitochondrial activity (maximal respiration) and respiratory reserve suggesting limited mitochondrial functions. However, mitochondrial injury could not be detected by mitochondrial fluorescent testes (Fig. 9f, g). Perhaps, the loss of inhibitory signaling in FcgRIIb−/− macrophages enhance the positive signals from the cross-link of TLR-4 with activating FcgRs that further facilitate more profound inflammatory responses in FcgRIIb−/− cells compared with WT (Fig. 11). Hence, the cell-energy status might directly control cell activities, and the manipulations of cell energy possibly harness macrophage activities and immune responses [86].

Fig. 11.

Illustration of the working hypothesis demonstrates a synergy of SF and LPS on TLR-4 and FcgRs of macrophages as follows; (i) SF facilitates TLR-4 dimerization that recruits the essential adapter proteins for the further activation, while directly endocytosed into the cells and activates cytosolic receptors [59] (ii) the cross-link between TLR-4 with activating FcgRs enhanced the responses, while inhibitory FcgRIIb reduced the activities [32, 94], and (iii) TLR-4 is activated by LPS and SF [59]. With the loss of inhibitory FcgRIIb in FcgRIIb−/− cells, the responses toward the combined saturated fatty acid and LPS are augmented when compared with WT. Figure created by Biorender.com. SF, saturated fatty acid; NLRs, nucleotide-binding and oligomerization domain (NOD)-like receptors; DAG, diacylglycerol; PKC, protein kinase C; NADPH, nicotinamide adenine dinucleotide phosphate.

On the other hand, the anti-inflammatory property of monounsaturated fatty acid [87] is possibly due to the structure that easier bind to peroxisome proliferator-activated receptors than TLR-4 when compared with the saturated fatty acid [88, 89] is previously mentioned. However, OA did not induce an anti-inflammatory effect in WT and FcgRIIb−/− macrophages here. While type I interferon, mainly produced by plasmacytoid dendritic cells (the innate immune cells effectively facilitate the adaptive immunity), is an important immune adjuvant that possibly induces autoantibody production from self-reactive plasma cells [90], the influence of other innate immune responses in lupus is currently mentioned [91]. As such, chronic or acute inflammation is one of the lupus exacerbation factors [92] and the imbalance of pro- versus anti-inflammatory macrophages (M1/M2 macrophages) have been suggested as a part of lupus pathogenesis [93]. Here, we hypothesize that saturated fatty acids might be one of the factors causing macrophage imbalance that exacerbate lupus activities possibly through TLR-4 and FcgRs (Fig. 11). However, the in vivo micro-environment is generally more complex than the in vitro experiments and the simple explanation from the in vitro experiments can explain only some parts of the pathogenesis. Hence, the co-activation of saturated fatty acids and LPS might induce other signals and cell homeostasis pathways. More studies on this topic are interesting.

In conclusion, HFD and obesity possibly accelerate lupus activities in FcgRIIb−/− mice in several pathways, but at least in part, through the additive stimulation by saturated fatty acid plus endotoxemia that enhanced macrophage responses. Thus, obesity in lupus should be more concerned in the clinical context.

Acknowledgments

Asada Leelahavanichkul is under Center of Excellence on Translational Research in inflammation and Immunology Research Unit (CETRII), Department of Microbiology, Chulalongkorn University, Bangkok, Thailand.

Statement of Ethics

The animal procedure, in accordance with the protocol of the National Institutes of Health (NIH; USA), was approved by the Institutional Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand, approval number (SST 025/2563).

Conflict of Interest Statement

The authors have no conflicts of interest to disclose.

Funding Sources

This research was supported by Chulalongkorn University through Fundamental Fund 66, the National Research Council of Thailand (Grant No. NRCT-N41A640076) and (811/2563) with NSRF via the Program Management Unit for Human Resources & Institutional Development, Research, and Innovation (B16F640175 and B05F640144), and Rachadapisak Sompoch Endowment Fund (CU-GR_60_13_23_02).

Author Contributions

The followings are the authors’ contribution: conceptualization: Kanyarat Udompornpitak and Asada Leelahavanichkul; methodology: Kanyarat Udompornpitak, Awirut Charoensappakit, Kritsanawan Sae-Khow, Thansita Bhunyakarnjanarat, Cong Phi Dang, Wilasinee Saisorn, and Peerapat Visitchanakun; validation: Kanyarat Udompornpitak, Awirut Charoensappakit, and Asada Leelahavanichkul; formal analysis: Kanyarat Udompornpitak, Awirut Charoensappakit, and Asada Leelahavanichkul; investigation: Kanyarat Udompornpitak, Pornpimol Phuengmaung, and Somkanya Tungsanga; resources: Tanapat Palaga, Patcharee Ritprajak, and Asada Leelahavanichkul; data curation: Kanyarat Udompornpitak, and Asada Leelahavanichkul; writing-original draft preparation: Asada Leelahavanichkul; writing-review and editing: Kanyarat Udompornpitak, and Asada Leelahavanichkul; supervision: Asada Leelahavanichkul; and funding acquisition: Tanapat Palaga, Patcharee Ritprajak, Somkanya Tungsanga, and Asada Leelahavanichkul. All the authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.