Gsα Regulates Macrophage Foam Cell Formation During Atherosclerosis

Abstract

Background:

Many cardiovascular pathologies are induced by signaling through G-protein-coupled receptors via G protein stimulatory α subunit (Gsα) proteins. However, the specific cellular mechanisms that are driven by Gsα and contribute to the development of atherosclerosis remain unclear.

Methods:

High-throughput screening involving data from single-cell and bulk sequencing were used to explore the expression of Gsα in atherosclerosis. The differentially expression and activity of Gsα were analyzed by immunofluorescence and cAMP measurements. Macrophage-specific Gsα knockout (Mac-Gsα^KO) mice were generated to study the effect on atherosclerosis. The role of Gsα was determined by transplanting bone marrow, and performing assays for foam cell formation, Dil-ox-LDL uptake, chromatin immunoprecipitation and luciferase reporter assays.

Results:

ScRNA-seq showed elevated Gnas in atherosclerotic mouse aorta’s cholesterol metabolism macrophage cluster, while bulk sequencing confirmed increased GNAS expression in human plaque macrophage content. A significant upregulation of Gsα and active Gsα occurred in macrophages from human and mouse plaques. Ox-LDL could translocate Gsα from macrophage lipid rafts in short-term and promote Gnas transcription through ERK1/2 activation and C/EBPβ phosphorylation via oxidative stress in long-term. Atherosclerotic lesions from Mac-Gsα^KO mice displayed decreased lipid deposition compared with those from control mice. Additionally, Gsα deficiency alleviated lipid uptake and foam cell formation. Mechanistically, Gsα increased the levels of cAMP and transcriptional activity of the cyclic adenosine monophosphate response element binding protein, which resulted in increased expression of CD36 and SR-A1. In the translational experiments, inhibiting Gsα activation with suramin or cpGN13 reduced lipid uptake, foam cell formation, and the progression of atherosclerotic plaques in mice in vivo.

Conclusions:

Gsα activation is enhanced during atherosclerotic progression and increases lipid uptake and foam cell formation. The genetic or chemical inactivation of Gsα inhibit the development of atherosclerosis in mice, suggesting that drugs targeting Gsα may be useful in the treatment of atherosclerosis.

Graphical abstract

Introduction

Atherosclerosis is considered to be a chronic inflammatory disease that is characterized by the accumulation of lipid deposits (mainly cholesterol) in macrophages that are located in both large and medium-sized arteries.¹ It is the dominant cause of cardiovascular diseases, including myocardial infarction, heart failure, and stroke, which has overtaken communicable diseases to become the leading cause of death and disability worldwide.² During the development of atherosclerosis, circulating monocytes transmigrate into the subintima layers of the arteries and differentiate into macrophages that can take up large amounts of oxidized low-density lipoprotein (ox-LDL) via the scavenger receptor pathway, which results in their transformation into lipid-rich foam cells.³ As a hallmark of atherosclerosis, foam cells can develop into fat streaks in blood vessel walls; therefore, understanding the causal molecular mechanisms involved in the formation of foam cells is essential for developing novel strategies to better manage atherosclerosis.

G protein-coupled receptors (GPCRs) and the heterotrimeric G-protein α subunit (Gsα) are involved in a plethora of human diseases, physiological and pharmacological activities, making them one of the most extensively studied drug targets. Approximately 34% of FDA-approved drugs act on 108 unique GPCR targets.⁴ Gsα is encoded by the guanine nucleotide binding protein, α-stimulating (GNAS) (Gnas in mice).⁵ Signaling by G-protein-coupled receptors via Gsα proteins induces cyclic adenosine monophosphate (cAMP) production and the subsequent activation of protein kinase A, which phosphorylates various downstream targets, such as the cAMP response element binding protein (CREB).⁶ Mutations in GNAS have been found to be related to many human diseases, such as McCune-Albright syndrome.^7–9 We recently reported that Gsα regulates endothelial permeability¹⁰ and angiogenesis.¹¹ Moreover, smooth muscle Gsα modulates intestinal contractility¹² and inhibits angiotensin II-induced abdominal aortic aneurysm formation in mice in vivo.¹³ However, the role of macrophage Gsα in foam cell formation and atherosclerosis remains unknown.

To explore the expression and function of macrophage Gsα in the development of atherosclerosis, we employed high-throughput data to investigate the expression and cellular preferences of Gsα, and generated mice lacking Gsα in macrophages and defined the consequences of this intervention on macrophage function and the underlined mechanism in atherosclerosis.

Methods

Data Availability.

All sequencing data sets in this article are deposited in international public repository, Gene Expression Omnibus (GEO), under accession ID GSE252233 for bulk RNA sequencing from mouse macrophage. Please see the Major Resources Table in the Supplemental Materials.

Animals

The mice used in this study are on a C57BL/6 background. Gsα^flox/flox mice¹⁴ were kindly provided by Dr. Lee S. Weinstein (Metabolic Diseases Branch, National Institutes of Health). We generated macrophage-specific Gsα-knockout mice (Mac-Gsα^KO) by crossing Gsα^flox/flox mice with transgenic mice expressing Cre recombinase under the control of a Lyz2 promoter (Lyz2-Cre, Cat. NO. NM-KI-215037, purchased from Shanghai Model Organisms Center, Inc.) Male and female mice aged 8 weeks were used in this study. Each mouse was given a single tail-vein injection of recombinant adeno-associated virus of murine proprotein convertase subtilisin/kexin type 9 mutants (rAAV/D377Y-mPCSK9) at 1.5 × 10¹¹ pfu as previously described¹⁵ and fed the Paigen diet for 10 weeks.

The inclusion/exclusion criteria for enrollment and the end point (atherosclerotic lesions) were pre-defined. The inclusion criteria were all survival mice that received treatment and fed up to 8 weeks on the Paigen diet. The exclusion criteria were the mice that do not survive until the end of the study (up to 16 weeks on Paigen diet). Male Ldlr^−/− mice from Vital River (Beijing, China) aged 7 weeks were acclimatized for 1 week and fed the Paigen diet for 2 months. When they had reached 12 weeks of age, the mice were randomly divided into two groups and treated with sterile water (vehicle) or suramin via subcutaneously-implanted osmotic minipumps. The Gsα antagonist, suramin (50 mg/kg/day), was dissolved (200 mg/mL) in sterile water.

Male Ldlr^−/− mice from Vital River (Beijing, China) aged 7 weeks were acclimatized for 1 week and fed the Paigen diet for 2 months. When they had reached 12 weeks of age, the mice were randomly divided into two groups and treated with sterile water (vehicle) or cpGN13 via pump-driven continuous intravenous microinfusing. The Gsα antagonist, cpGN13 (3 mg/kg/day), was dissolved in sterile normal saline.

The mice were euthanized using profound anesthesia with pentobarbital (40 mg/kg, intraperitoneal injection) followed by exsanguination and tissue removal. All of the mice had access to food and water ad libitum and were maintained at 23 ºC with 60% relative humidity and a 12-h light/12-h dark cycle. All of the animal studies complied with the Management Rules of the Chinese Ministry of Health and were approved by the Ethical Committee.

Clinical study participants

Intermediate (type III) and advanced (type VI) atherosclerotic plaques were obtained from the same human carotid endarterectomies.¹⁶ All of the protocols were approved by the Ethical Committee of Shandong University Qilu Hospital. Blood coagulation was blocked using ethylene diamine tetraacetic acid, and peripheral blood mononuclear cells (PBMCs) were isolated as previously described.¹⁷ Briefly, PBMCs were isolated by centrifugation on a Ficoll gradient (Ficoll-Paque PLUS, GE Healthcare Life Sciences). Monocytes were selected using antibody-coated magnetic beads and magnetic columns according to the manufacturer’s instructions (CD14 MicroBeads, Miltenyi Biotec). This study was performed in accordance with the Declaration of Helsinki, and the participants provided informed consent.

Assay for activated Gsα (Gsα-GTP)

We performed Gsα-GTP assays following manufacturer’s guidelines and previous research.¹⁸ In brief, cell lysates were incubated with an antibody against active Gsα-GTP (NewEast Bioscience 26906), then pulled down with Protein A/G magnetic beads (Bimake, B23201). Activated Gsα and total Gsα proteins were identified via western blotting with a Gsα antibody. Additionally, plaque slices and cell slides underwent in situ immunofluorescence using an active Gsα-GTP-specific antibody as the primary antibody. We also utilized isoproterenol stimulation of BMDMs in immunofluorescence and western blotting experiments as a positive control of Gsα activation.

Statistical analysis

The sample size for the in vivo experiments was determined based on power calculations using lesion area in mice aortas. Using our preliminary experimental results, the difference in means (δ) of 19.0 and a standard deviation (σ) of 12.0, 1-β (power) was defined 80% and α was 0.05. Based on the PASS (version 15.0.5) sample calculation software, we used the Two-Sample T-Tests Assuming Equal Variance to estimate that the required sample size of at least 8.

For the statistical analyses, GraphPad Prism software (version 9.4) was used with default settings and two-sided P values. The Shapiro–Wilk test was used to check for normality firstly. For normally distributed data, observed results are expressed as mean ± standard error of the mean (SEM), while if the data was not normally distributed, it was described as median and interquartile range (IQR). The evaluation of statistical significance between groups were conducted as follows: (1) For two independent groups, demonstrating normal distribution with homogeneous variances, analysis was performed using the unpaired two-tailed Student’s t-test. In scenarios of heterogeneity in variances, the unpaired two-tailed Student’s t-test incorporating Welch’s correction was applied. For distributions deviating from normality (including n ≤ 4), the Mann Whitney test was applied. (2) For datasets involving more than two independent groups, with a singular factor under normal distribution, the analysis involved the use of one-way ANOVA for homogenous variances, complemented by Tukey’s multiple comparisons test for post-hoc insights. In cases of variance heterogeneity under normal distribution, the Brown-Forsythe ANOVA test was employed, followed by Dunnett’s T3 multiple comparisons test for further comparative analysis. For groups not conforming to normal distribution, the Kruskal-Wallis test was utilized, with subsequent post-hoc analysis via Dunn’s multiple comparisons test. And (3), in situations with more than two independent groups, characterized by two factors and normal distribution, the two-way ANOVA was the preferred statistical approach, followed by post-hoc examination using Tukey’s multiple comparisons test. P < 0.05 was considered to be statistically significant.

Results

Gsα activation is enhanced during human atherosclerotic progression

In order to understand role of Gsα in the human and rodent atherosclerosis, we firstly re-analyzed the single-cell sequencing data of aorta of Ldlr^−/− mice from GSE116240, which consisted of eleven CD45⁺ cluster leukocytes. All macrophage cells were divided into a total of eight subgroups: endocytosis macrophage (Cluster 0 and 3), inflammatory macrophage (Cluster 1 and 5), replicative macrophage (Cluster 2), cholesterol metabolism macrophage (Cluster 4), interferon-responsive macrophage (Cluster 6), and proliferating macrophage (Cluster 7) (Figure S1A). The Gene ontology (GO) analysis showed that cluster 4 macrophage was enriched in cholesterol metabolism and immune process (Figure S1B). Violin Plot and tSNE Plot indicated that Gnas was escalated in cholesterol metabolism macrophage, which expression of Gnas was 1.7 folds to other macrophage (Figure S1C and S1D).

To address expression of GNAS in human atherosclerosis, we analyzed the BiKE (Biobank of Karolinska Endarterectomy) dataset of human carotid endarterectomy samples (GSE21545). After characterizing macrophage with CIBERSORT from bulk sequencing data of BiKE, we identified the expression of GNAS in macrophages of human carotid endarterectomy samples, GNAS mRNA expression increased with plaque macrophage content (Figure S1E).

To further determine the expression of Gsα in the development of atherosclerosis, we measured the Gsα levels in human carotid atherosclerotic plaques. Colocalization studies indicated increased expression levels of Gsα in intimal macrophages from advanced atherosclerotic plaques (Figure 1A). We observed a higher number of intimal macrophages from advanced plaques than from intermediate plaques (Figure 1C). Within the intimal thickening of advanced atherosclerotic lesions, there was a 2.84-fold increase in the number of macrophages expressing Gsα compared to those in intermediate lesions (Figure 1D and 1E). Furthermore, we utilized antibody against active Gsα to detect the expression of activated guanosine triphosphate (GTP)-bound Gsα (active Gsα). The specificity of this antibody was demonstrated by the stimulation of BMDMs with isoproterenol in immunofluorescence and western blotting experiments (Figure S1F and S1G). The colocalization of active Gsα with intimal macrophages was highly expressed in the advanced atherosclerotic plaques (Figure 1B, 1F, and 1G), which suggested that Gsα activation is elevated in the development of atherosclerosis. In addition, CD14⁺ monocytes were isolated from peripheral blood mononuclear cells (PBMCs) using CD14 MicroBeads (> 90% purity, Figure S2) in coronary artery disease patients (Table S1). The Gensini score defines the severity of coronary artery disease and can be used to predict cardiovascular prognoses.¹⁹ Using a western blotting analysis, we found a progressive increase in the total and active Gsα expression levels in human CD14⁺ monocytes from patients with increasing plaque burden. Additionally, there was a gradual increase in the proportion of activated Gsα (Figure 1H, 1I). These results were further confirmed after measuring the cAMP content (Figure 1J). Gsα activation is therefore enhanced during human atherosclerotic progression. Furthermore, we measured the Giα levels in human carotid atherosclerotic plaques. In the intimal thickening of the advanced atherosclerotic lesions, the expression and activation of Giα was increased compared with that in the intermediate lesions (Figure S3A, S3B).

Figure 1. Gsα activation is enhanced during human atherosclerotic progression.

A–B, Representative immunofluorescence images for detecting Gsα, active Gsα and CD68 (specific for human monocytes and macrophages) in intermediate and advanced atherosclerotic plaques (n = 4 in each). Nuclear staining by DAPI. Scale bar = 50 μm. C–D, Quantification of CD68 (C) and macrophage Gsα (D) expression within the intimal thickening of intermediate and advanced atherosclerotic plaques. Statistical analyses were performed using the Mann Whitney test. Relative values are compared against those of the intermediate group. The unit is Manders overlap coefficients used for quantifying colocalization. E, Analysis of the colocalization of Gsα (red curves) and macrophages (green curves) in advanced lesions. F, Quantification of active Gsα expression by macrophages within the intimal thickening of intermediate and advanced atherosclerotic plaques. Statistical analyses were performed using the Mann Whitney test. Relative values are compared against those of the intermediate group. The unit is Manders overlap coefficients used for quantifying colocalization. G, Analysis of the colocalization of active Gsα (red curves) and macrophages (green curves) in advanced lesions. H, Representative western blotting analysis of active Gsα in human CD14+ monocytes (n = 8 in each). I, Quantification of the protein expression of active Gsα and total Gsα. Statistical analyses were conducted using the one-way ANOVA. Relative values are compared against those of the Gensini score 0–24 group. J, ELISA was used to detect the cAMP levels in the supernatants of human CD14⁺ monocytes (n = 6 in each). Statistical analyses were conducted using the one-way ANOVA. Relative values are compared against those of the Gensini score 0–24 group.

Taken together, we found that GNAS was highly expressed in CD14⁺ monocytes from PBMCs and plaques of atherosclerosis, and localized to cholesterol metabolism macrophage, which may be involved in the cholesterol metabolism and foam cell formation.

Active Gsα is upregulated by oxidized low-density lipoprotein in macrophages

We further detected the expression and activation of Gsα in mouse atherosclerotic plaques. A model of progressive atherosclerosis in Ldlr^−/− mice was established by administering the Paigen diet to the mice for different time durations, which resulted in progressive macrophage infiltration in the atherosclerotic lesions. The numbers of total and active Gsα-expressing macrophages gradually increased in the mice fed the Paigen diet for 12 weeks (Figure 2A and 2B), which was consistent with the results from the human samples. Furthermore, we observed a gradual increase in the number of total and active Giα-expressing macrophages in the mice fed the Paigen diet for up to 12 weeks (Figure S3C, S3D), which was consistent with the results from the human samples. To evaluate the effects of atherosclerotic stimuli on Gsα expression and activation in macrophages, primary bone marrow-derived macrophages (BMDMs) were incubated with oxidized low-density lipoprotein (ox-LDL). Cell viability and apoptosis were evaluated using the CCK-8 assay and the TUNEL assay, respectively. The results implied that cell viability remained relatively stable over the 48-hour duration, while a statistically significant increase in cell apoptosis was detected at the 48-hour time point (Figure S4A, S4B). As shown in Figure 2C–2D , the active Gsα protein level was upregulated after 6 h of the ox-LDL treatment in the BMDMs and was further increased after 12 h of the treatment, which was confirmed by the elevation in the intracellular cAMP levels following 6 h of ox-LDL treatment (Figure 2E). However, the total Gsα protein level was significantly increased only after 24 h of the ox-LDL treatment (Figure 2C and 2F). A reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis indicated that no statistically significant difference was observed in the mRNA levels of Gsα in response to ox-LDL in the BMDMs until 24 h had passed (Figure 2G). Thus, short-term stimulation within 12 h of the ox-LDL treatment led to Gsα activation without affecting the transcription of the Gnas gene. After 24 h, ox-LDL could upregulate the mRNA and protein levels of Gsα. We also observed that non-oxidized LDL (nLDL) could increase Gsα mRNA and protein levels, although its effect was notably weaker compared to that of ox-LDL (Figure S4C, S4D). Given that ox-LDL exerts a more pronounced impact on atherosclerosis,²⁰ we have placed greater emphasis on investigating the role of ox-LDL and Gsα in atherosclerosis. Taken together, these data suggest that ox-LDL activates Gsα in macrophages through different molecular mechanisms of action.

Figure 2. Active Gsα is upregulated by oxidized low-density lipoprotein (ox-LDL) in macrophages.

A–B, Representative immunofluorescence images for detecting Gsα (A), active Gsα (B) and MOMA-2 in atherosclerotic plaques from Ldlr^−/− mice (male) fed the Paigen diet for 0, 4, 8 and 12 weeks, respectively. Scale bar = 50 μm. C, Western blotting analysis of active and total Gsα protein level in BMDMs treated with 50 μg/ml Ox-LDL at different times (n = 6 in each). D, Quantification of protein expression of active Gsα. Statistical analyses were conducted using one-way ANOVA. E, ELISA was used to detect the cAMP level in BMDMs treated with 50 μg/ml Ox-LDL at different times. Statistical analyses were conducted using one-way ANOVA. F, Quantification of the protein expression of total Gsα. Statistical analyses were conducted using the one-way ANOVA. G, RT-qPCR analysis of Gsα mRNA level in BMDMs treated with 50 μg/ml Ox-LDL at different times (n = 6). Statistical analyses were conducted using one-way ANOVA.

Ox-LDL translocates Gsα from macrophage lipid rafts in the short term and promotes Gnas transcription in the long term

GPCR is often required for the Gsα activation. Subsequently, we assessed whether a GPCR was necessary for Gsα activation in our study. We adopted a systematic approach to identify potential GPCRs. Specifically, we intersected the highly expressed molecules from a pool of GPCR candidates, chosen based on their expression patterns in vertebrate animals according to the IUPHAR/BPS Guide to Pharmacology database, with single-cell sequencing databases of atherosclerotic macrophages in humans (GSE155512).²¹ The Venn diagram illustrates that we have identified 27 GPCRs through our screening process (Figure S5A). Subsequently, we curated a comprehensive library of GPCR inhibitors and administered them to macrophages before subjecting them to ox-LDL stimulation. Following a 12-hour treatment with ox-LDL, we employed high-content imaging technology for a quantitative assessment of the activation status of Gsα. However, no statistically significant difference was observed in the activation state of Gsα in response to treatment with any of the GPCR inhibitors (Figure S5B). As a result, at least in the short term, the activation of Gsα is independent of known GPCR signaling pathways studied. Lipid rafts are membrane microdomains that are enriched in specific lipids and proteins. Transient structural variations are associated with the activation or inhibition of a signal transduction cascade.^22,23 To explore the mechanisms underlying the ox-LDL-mediated activation of Gsα, BMDMs were incubated with ox-LDL for 12 h. The levels of activated Gsα increased by 5.7-fold and were accompanied by rearranged membrane structures labelled with flotillin, a lipid raft marker²⁴ (Figure 3A–3C). Colocalization experiments indicated decreased levels of Gsα in the lipid rafts after the ox-LDL treatment (Figure 3C and 3D). Furthermore, GM1, a ganglioside highly concentrated in membrane lipid rafts, was subjected to immunofluorescence analysis to examine its distribution and its relationship with active Gsα. GM1 was visualized in a diffuse distribution in resting macrophages and clustering in ox-LDL-stimulated macrophages (Figure S5C). Colocalization experiments revealed a significant reduction in Gsα levels within the lipid rafts following ox-LDL treatment, as corroborated by Figure S5D. Additionally, we endeavored to detect the interaction between GM1 or flotillin and Gsα through proximity ligation assay (PLA). Regrettably, PLA failed to yield positive signals, underscoring that the molecular proximity between GM1 or flotillin and Gsα on the cell surface is not less than 40nm (Figure S5E and S5F). To further confirm whether ox-LDL could translocate Gsα from the lipid rafts, Raw 264.7 cells were exposed to ox-LDL for 12 h, and lipid raft-associated proteins were isolated. Compared with the vehicle treatment, ox-LDL significantly reduced the level of Gsα in the lipid rafts (Figure 3E and 3F). No statistically significant difference was observed in the total Gsα level, which ruled out the possibility of decreased Gsα levels in the lipid rafts due to changes in the total level of Gsα.

Figure 3. Ox-LDL translocates Gsα from macrophage lipid rafts in the short term and promotes Gnas transcription in the long term.

A, Representative immunofluorescence analysis to detect active Gsα in BMDMs treated with 50 μg/ml ox-LDL for 12 hours (n = 6 in each). Nuclear staining by DAPI. Scale bar = 25 μm. B, Quantification the fluorescence intensity of active Gsα. Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group. C, Colocalization of active Gsα with flotillin (a lipid raft marker) in BMDMs treated with 50 μg/ml ox-LDL for 12 hours. The white line is the path of the colocalization analysis. Scale bar = 100 μm. D, Analysis of the colocalization of Gsα (red curves) and flotillin (green curves). The x-axis represented distance relative to the starting point of the colocalization analysis. E–F, Western blotting analysis (E) and quantification (F) of Gsα in lipid rafts (lipid raft; fractions 4–5), nonlipid rafts (non-rafts; fractions 9–12), and whole cellular Gsα protein (total) in Raw 264.7 cells treated with 50 μg/ml ox-LDL for 12 h (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group. G, Venn diagram showing the overlap of promoters in PROMO, Animal TFDB4, TRANSFAC and ox-LDL mOTM. H–J, Western blotting analysis and RT-qPCR of Gsα in BMDMs from WT mice transfected with CTR siRNA or C/EBPβ siRNA for 48 hours followed by treatment with ox-LDL for 24 hours (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the CTR siRNA + vehicle group. K, Predicted C/EBPβ site in the mouse Gnas promoter. L-M, Mouse BMDMs transfected with CTR siRNA or C/EBPβ siRNA for 48 hours followed by treatment with ox-LDL for 24 hours, chromatin immunoprecipitation was then performed to assay the C/EBPβ binding site in the Gnas promoter (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the CTR siRNA + vehicle group. N, HEK 293T cells were transfected with CTR siRNA or C/EBPβ siRNA, wild-type or mutant Gnas promoter luciferase constructs for 24 hours followed by treatment with ox-LDL for 24 hours, and then luciferase activity was measured (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the CTR siRNA group + vehicle + WT. O–Q, Western blotting and RT-qPCR analysis of Gsα in BMDMs treated with ox-LDL or NAC for 24 hours (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the vehicle group. ns indicates not significant.

CD36 is a scavenger receptor associated with atherosclerosis. In our in vitro study, it was shown that the suppression of CD36 using siRNA reversed the ox-LDL-induced increase in Gsα expression in the long term. However, the use of the nuclear hormone receptor PPARγ inhibitor, GW9662, did not affect the ox-LDL-induced Gsα expression (Figure S5G, S5H). Therefore, it can be concluded that the upregulation of Gsα expression in response to ox-LDL is contingent upon the presence of CD36, but independent of PPARγ. To understand the regulation of Gnas transcription by ox-LDL in the long term, potential transcriptional factors in the Gnas promoter region were predicted using bioinformatics tools, including PROMO (ALGGEN), Animal Transcription Factor DataBase (Animal TFDB4), TRANSFAC (gene regulation) and open chromatin mapping and enhancer and transcription factor mapping in primary human macrophages exposure to ox-LDL (ox-LDL mOTM).²⁵ We found the most promising transcription factor, C/EBPβ binding to the Gnas promoter in primary macrophages exposure to ox-LDL, based on intersection of multiple datasets (Figure 3G). A subsequent in vitro study demonstrated that the knockdown of C/EBPβ with siRNA significantly reversed ox-LDL-induced Gsα expression (Figure 3H–3J). A chromatin immunoprecipitation (ChIP) assay was then performed and revealed the direct binding of C/EBPβ to the Gnas promoter region (−1110 to −1123), which could be enhanced by treatment with ox-LDL (Figure 3K–3M). Using a dual luciferase reporter gene assay, we found that the Gnas luciferase reporter was activated after stimulation with ox-LDL. This effect was abolished in the C/EBPβ binding site mutation in the Gnas promoter (Figure 3N). Furthermore, C/EBPβ is a reactive oxygen species (ROS)-sensitive transcription factor that is triggered by the oxidative stress response,²⁶ and the antioxidant effects of N-acetylcysteine significantly prevented ox-LDL-induced Gsα upregulation (Figure 3O–3Q). To further confirm the role of ROS in ox-LDL-induced Gsα expression and related signaling pathway, we used a mitochondria-targeted antioxidant tempol, which could effectively prevent the ox-LDL-induced ERK1/2 activation, C/EBPβ phosphorylation and upregulation of Gsα (Figure S5I, S5J). Moreover, the inhibition of ERK1/2 catalytic activity using SCH772984 effectively countered the ox-LDL-induced ERK1/2 activation, C/EBPβ phosphorylation, and the upregulation of Gsα (Figure S5K, S5L). We confirmed that ox-LDL, which depend on mitochondrial ROS generation, induced ERK1/2 activation and subsequently C/EBPβ phosphorylation, promote Gsα expression in macrophages. Collectively, these results indicate that ox-LDL activates Gsα through different molecular mechanisms by which ox-LDL translocates Gsα from macrophage lipid rafts in the short term and promotes Gnas transcription through ERK1/2 activation and C/EBPβ phosphorylation via oxidative stress in the long term.

Macrophage Gsα deficiency ameliorates the development of atherosclerosis

To elucidate the biological significance of Gsα expression in macrophages in vivo, Gsα^flox/flox mice were crossbred with Lyz2-Cre mice to generate Gsα^flox/+/Lyz2-Cre mice, which were further intercrossed to obtain Gsα^flox/flox/Lyz2-Cre (Mac-Gsα^KO) mice (Figure 4A). The littermate Gsα^flox/flox/Cre^- mice were used as controls (CTR). Western blotting and RT-qPCR analyses confirmed significantly reduced Gsα expression in the macrophages of the Mac-Gsα^KO mice (Figure 4B–4D). However, RT-qPCR results indicated that the expression of Gsα associated transcripts, namely NESP55 and XLalpha(s), remains unaltered in Mac-Gsα^KO mice (Figure S6). The male CTR and Mac-Gsα^KO mice were further injected with rAAV/D377Y-mPCSK9 and then fed the Paigen diet for 10 weeks. As shown in Table S3, no differences in body weight, systolic and diastolic blood pressure, and the levels of serum triglycerides, serum cholesterol were observed in these two groups of mice, whereas the proportions of atherosclerotic surface lesions were significantly reduced in the Mac-Gsα^KO mice compared to those in the CTR mice (Figure 4E and 4F). Hematoxylin and eosin (HE) staining of the aortas and oil red O (ORO)-stained aortic roots displayed significantly decreased lesion areas in the Mac-Gsα^KO mice compared to those in the CTR mice (Figure 4G and 4H). In the aortic arches from the Mac-Gsα^KO mice, the number of MOMA-2-positive macrophages within the atherosclerotic plaques was decreased by 46% compared with that in the CTR mice (Figure 4G and 4H). To determine whether the observed phenotype was reproducible in the female mouse models, atherosclerosis was induced in the female CTR and Mac-Gsα^KO mice. The mean lesion areas in the aortas and aortic sinuses were significantly decreased in the Mac-Gsα^KO mice (Figure S7). Taken together, these data demonstrate that macrophage Gsα deficiency leads to ameliorated atherosclerotic development. Considering that the effects of macrophage-specific knockout of Gsα on atherosclerosis were found to be highly similar in both male and female mice, and in line with the “3Rs” guideline for animal welfare, subsequent animal experiments were conducted exclusively with male mice.

Figure 4. Macrophage Gsα deficiency ameliorates the development of atherosclerosis.

A, Schematic diagram of transgenic mice used to generate adult Mac-Gsα^KO mice. B–D, Western blotting and RT-qPCR analysis of Gsα in BMDMs from male CTR and Mac-Gsα^KO mice (n = 6 in each group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction or Mann–Whitney test, respectively. Relative values are compared against those of the CTR group. E–F, Representative ORO staining and en face analysis of atherosclerotic lesions in the whole aorta (n = 10 in each group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the CTR group. G-H, ORO staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). HE staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). Immunofluorescence analysis to detect MOMA-2 in frozen aortic root sections from CTR and Mac-Gsα^KO mice (male) (n = 10 per group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the CTR group. Scale bar = 100 μm. I–J, Representative ORO staining and en face analysis of atherosclerotic lesions in the whole aorta (n = 10 in each group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the CTR BM in Ldlr^−/− group. K–L, ORO staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). HE staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). Immunofluorescence analysis to detect MOMA-2 in frozen aortic root sections (n = 10 per group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test or Unpaired two-tailed Student’s t-test with Welch’s correction. Relative values are compared against those of the CTR BM in Ldlr^−/− group. Scale bar = 100 μm.

To further determine whether the observed phenotype was reproducible, lethally-irradiated 8-week-old male Ldlr^−/− mice were reconstituted with bone marrow (BM) from either the CTR or Mac-Gsα^KO mice. Flow cytometry confirmed successful BM chimeras after 4 weeks in peripheral myeloid cells (Figure S8). The mice were further fed the Paigen diet for 18 weeks. Body weight, systolic and diastolic blood pressure, and total cholesterol, triglyceride levels were comparable among the groups (Table S4). However, the extent of atherosclerotic lesion progression in the whole aortas and cross-sectional areas of the atherosclerotic lesions in the aortic sinuses was significantly decreased in the Mac-Gsα^KO BM→ Ldlr^−/− mice compared with that in the CTR BM → Ldlr^−/− mice (Figure 4I–4L). In the aortic arches of the Mac-Gsα^KO BM → Ldlr^−/− mice, the number of MOMA-2-positive macrophages within the atherosclerotic plaques was reduced by 57% compared to that in the CTR BM → Ldlr^−/− mice (Figure 4K–4L). Taken together, these findings demonstrate that macrophage Gsα contributes to atherosclerotic lesion progression.

Macrophage Gsα regulates foam cell formation

To explore the molecular mechanisms underlying the Gsα loss-mediated amelioration of atherosclerosis, transcriptome analysis was performed on macrophages derived from the CTR and Mac-Gsα^KO mice. There were 255 differentially-expressed genes (DEGs) in the Gsα-deficient macrophages compared with the CTR (Figure S9A). Among these DEGs, 53 genes were related to macrophage lipid metabolism by GSEA analysis (Figure S9B). We therefore speculate that Gsα deficiency may cause the amelioration of atherosclerotic development by regulating lipid accumulation and foam cell formation.

Macrophage-derived foam cell formation is a crucial step in the development of atherosclerosis. Tissue macrophages become foam cells when exposed to chemically-modified lipoproteins, such as ox-LDL. To determine whether macrophage Gsα affects foam cell formation, BMDMs isolated from the CTR and Mac-Gsα^KO mice were treated with ox-LDL or Dil-ox-LDL. As shown in Figure 5A and 5B, the BMDMs from the Mac-Gsα^KO mice displayed lower levels of ORO staining and Dil-ox-LDL uptake compared with those from the CTR. We also conducted intracellular measurements of total cholesterol (TC), free cholesterol (FC), and cholesteryl ester (ChE) to evaluate intracellular lipid accumulation. In comparison to the CTR group, Mac-Gsα^KO BMDMs exhibited a substantial reduction in intracellular TC content and FC content. Furthermore, the ChE/TC ratio was lower in Mac-Gsα^KO BMDMs (Figure S10A). Macrophage scavenger receptors, particularly CD36 and the class A1 scavenger receptor (SR-A1), are the principal receptors responsible for the uptake of modified lipoproteins and have been implicated as factors that contribute to early foam cell formation. The BMDMs from the Mac-Gsα^KO mice led to significantly-decreased protein and mRNA levels of both CD36 and SR-A1 (Figure 5C–5E). However, statistically-significant differences were not observed in the ABCA1, ABCG1, and SR-B1 protein and mRNA expression levels (Figure 5C–5E), which are involved in cholesterol efflux. Thus, Gsα deficiency regulates macrophage foam cell formation by decreasing lipid uptake and accumulation.

Figure 5. Macrophage Gsα regulates foam cell formation.

A–B, BMDM isolated from CTR and Mac-Gsα^KO mice were incubated with ox-LDL (50 μg/ml) for 24 hours followed by ORO staining, or treated with 10 μg/ml Dil-ox-LDL for 4 hours followed by immunofluorescence analysis (n = 6 in each). Statistical analyses for ORO staining were performed using the unpaired two-tailed Student’s t-test. Statistical analyses for Dil-ox-LDL were performed using the Mann–Whitney test. Relative values are compared against those of the CTR group. Scale bar = 50 μm. C–E, Western blotting (C-D) and RT-qPCR analysis (E) of 6 indicated proteins in BMDMs from CTR and Mac-Gsα^KO mice (n = 6 in each). Statistical analyses for western blotting of SR-A1, LOX-1, ABCG1, SR-B1 and RT-qPCR analysis of CD36, SR-A1, LOX-1, SR-B1 were performed using the unpaired two-tailed Student’s t-test. Statistical analyses for western blotting of CD36, ABCA1 and RT-qPCR analysis of ABCA1, ABCG1 were performed using the Mann–Whitney test. Relative values are compared against those of the CTR group. F–G, BMDM isolated from CTR mice infected with Ad-GFP or Ad-Gsα were incubated with ox-LDL (50 μg/ml) for 24 hours followed by ORO staining, or treated with 10 μg/ml Dil-ox-LDL for 4 hours followed by immunofluorescence analysis (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction or Unpaired two-tailed Student’s t-test. Relative values are compared against those of the Ad-GFP group. Scale bar = 50 μm. H–J, Western blotting (H-I) and RT-qPCR analysis (J) of 6 indicated proteins in BMDMs isolated from CTR mice infected with Ad-GFP or Ad-Gsα (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction or unpaired two-tailed Student’s t-test or Mann Whitney test. Relative values are compared against those of the Ad-GFP group. ns indicates not significant.

To further determine whether macrophage Gsα facilitates foam cell formation, BMDMs were infected with Ad-GFP or Ad-Gsα followed by treatment with ox-LDL or Dil-ox-LDL. The adenoviral overexpression of Gsα resulted in a significant increase in ORO and Dil-LDL staining, as well as in intracellular measurements of TC, FC, ChE and the ChE/TC ratio (Figure 5F and 5G, Figure S10B). Moreover, Gsα overexpression in the BMDMs led to the significantly-increased protein and mRNA levels of both SR-A1 and CD36, while no statistically significant difference was observed in the expression of Lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1), ABCA1, ABCG1, and SR-B1 (Figure 5H–5J). We conducted further assessments to determine whether the uptake could be restored by activating Gsα using cholera toxin (CT) after Gsα knockdown with siRNA. CT induces persistent activation of adenylate cyclase and subsequent accumulation of cAMP.²⁷ ORO staining and Dil-ox-LDL uptake experiments revealed that CT had the capacity to enhance lipid deposition and intracellular lipid uptake. Nevertheless, this effect was attenuated by the knockdown of Gsα with siRNA (Figure S10D). Macrophage Gsα therefore facilitates foam cell formation by promoting the uptake of modified lipoproteins.

Gsα modulates lipid accumulation through CREB

We next explored the mechanisms through which Gsα regulates CD36 and SR-A1 expression. Gsα is responsible for receptor-stimulated cAMP generation and subsequent CREB activation. Based on the Transcription Factor Database (http://jaspar.genereg.net), CRE sites were predicted as the CD36 and SR-A1 promoters (Figure 6A). To examine whether Gsα regulates the expression of CD36 and SR-A1 via CREB, a ChIP-PCR assay was performed, which revealed that the overexpression of Gsα resulted in significantly-increased CREB binding with the CD36 and SR-A1 promoters (Figure 6B and 6C). To further test whether CREB directly targets CD36 and SR-A1, a dual-luciferase reporter assay was applied and demonstrated that wild-type CD36 and SR-A1 luciferase reporters were activated by Gsα overexpression. Their activation was abolished in the CREB binding site mutant promoters (Figure 6D). These results suggest that Gsα regulates the expression of CD36 and SR-A1 through the direct binding of CREB to their promoters.

Figure 6. Gsα regulates lipid accumulation through CREB.

A, Predicted CRE site in the mouse CD36 and SR-A1 promoters. B–C, BMDMs were infected with Ad-GFP or Ad-Gsα for 48 hours, followed by chromatin immunoprecipitation assay to detect the CREB binding site in the CD36 and SR-A1 promoters (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction. Relative values are compared against those of the Ad-GFP group. D, HEK-293T cells were infected with Ad-GFP or Ad-Gsα and subsequent transfected with wild-type or mutant CD36 or SR-A1 promoter luciferase constructs for 24 hours followed by luciferase activity measurement (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the WT + Ad-GFP group. E–G, Western blotting analysis and RT-qPCR of CD36 and SR-A1 in BMDMs infected with Ad-GFP or Ad-Gsα treated with KG-501 (10 μM) for 24 hours (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the Ad-GFP + vehicle group. H–I, BMDMs were infected with Ad-GFP or Ad-Gsα, and treated with KG-501 and ox-LDL followed by ORO staining. BMDMs were infected with Ad-GFP or Ad-Gsα, and treated with KG-501 and Dil-ox-LDL followed by immunofluorescence analysis (n = 6 in each). Statistical analyses were performed using the two-way ANOVA. Relative values are compared against those of the Ad-GFP+ vehicle group. Scale bar = 20 μm. ns indicates not significant.

To determine whether CREB contributes to modified lipoprotein uptake and Gsα-mediated foam cell formation, the CREB inhibitor, KG-501, was used. The overexpression of Gsα in BMDMs resulted in significantly-increased CD36 and SR-A1 protein and mRNA expression, which was specifically abrogated following treatment with KG-501 (Figure 6E–6G). Moreover, Gsα overexpression led to significantly-enhanced foam cell formation, as indicated by the ORO staining, intracellular measurements of TC, FC, ChE and the ChE/TC ratio and Dil-ox-LDL uptake, which was attenuated in the presence of KG-501 (Figure 6H–6I, Figure S10C). Taken together, these data demonstrate that Gsα promotes modified lipoprotein uptake and foam cell formation through CREB-mediated CD36 and SR-A1 upregulation.

The Gsα inhibitor, suramin, reduces macrophage lipid uptake and atherosclerotic progression

Suramin acts as a Gsα inhibitor because it inhibits the rate-limiting step with the exchange of guanosine diphosphate for GTP in the activation of Gsα.²⁸ Here, we explored the effects of suramin on foam cell formation and atherosclerosis in vitro and in vivo. Suramin inhibited the activation of Gsα in a dose-dependent manner (Figure S11A and S11B). Treatment with suramin further resulted in suppressed modified lipoprotein uptake and macrophage foam cell formation in a dose-dependent manner (Figure S11C–S11E). The protein and mRNA levels of CD36 and SR-A1 were also reduced in BMDMs after treatment with suramin (Figure S11F and S11G). In vivo, the Ldlr^−/− mice were fed the Paigen diet for 8 weeks and then treated with suramin for the last 4 weeks of the experiment. Body weight, heart rate, systolic and diastolic blood pressure, and total cholesterol, triglyceride levels were comparable among the groups (Table S5). Immunofluorescence analyses in frozen aortic root sections confirmed significantly-reduced Gsα activation in the macrophages of Ldlr^−/− mice with suramin compared with that in the vehicle mice (Figure S11J and S11K). Treatment with suramin led to reduced aortic atherosclerotic plaque sizes (Figure S11H and S11I). The HE staining and ORO-stained aortic roots revealed that treatment with suramin resulted in reduced lesion areas (Figure S11J and S11K). The number of MOMA-2-positive macrophages within the atherosclerotic plaques was decreased in the aortic arches from the Ldlr^−/− mice following treatment with suramin (Figure S11J and S11K). Thus, suramin inhibits the activation of Gsα and attenuates the development of atherosclerosis.

The state-selective inhibitor of Gsα, cpGN13, reduces macrophage lipid uptake and atherosclerotic progression

We employed an another Gsα inhibitor to further confirm the role of Gsα in the development of atherosclerosis. Given that cell-permeable cyclic peptide GN13 (cpGN13) can recognize the active conformation of Gsα/GTP, hinder the binding of Gsα to adenylyl cyclase, and thereby inhibit Gsα-mediated cAMP synthesis, cpGN13 exhibits at least 40–100 fold conformational selectivity. Furthermore, they do not interact with other members of the Gα family such as Giα, Gα13 and Gαq.²⁹ We then proceeded to further investigate the effects of cpGN13 on foam cell formation and atherosclerosis in vitro and in vivo. Immunofluorescence showed cpGN13 inhibited the activation of Gsα (Figure 7A and 7B). Treatment with cpGN13 further resulted in suppressed modified lipoprotein uptake and macrophage foam cell formation (Figure 7C and 7D). The protein and mRNA levels of CD36 and SR-A1 were also reduced in BMDMs after treatment with cpGN13 (Figure 7E–7G). In vivo, based on the IC50 and intravenous bioavailability,³⁰ we selected an intravenous dosage of 3 mg/kg for administration. To evaluate the in vivo effectiveness of cpGN13 at different time points, we conducted angular vein blood sampling and isolated mononuclear cells. The results from immunofluorescence analysis demonstrated that intravenous administration effectively inhibited Gsα activation for up to 12 hours. However, beyond the 24-hour mark post-administration, Gsα activation was no longer suppressed (Figure 7H). Therefore, we adapted an established in vitro model of pump-driven continuous intravenous microinfusing at 3 mg/kg/d. The Ldlr^−/− mice were fed the Paigen diet for 8 weeks and then treated with cpGN13 for the last 4 weeks of the experiment. Body weight, heart rate, systolic and diastolic blood pressure, and total cholesterol, triglyceride levels were comparable among the groups (Table S6). Western blotting analyses confirmed significantly-reduced Gsα activation in the macrophages of Ldlr^−/− mice with cpGN13 compared with that in the vehicle mice (Figure 7I). Treatment with cpGN13 led to reduced aortic atherosclerotic plaque sizes (Figure 7J and 7K). The HE staining and ORO-stained aortic roots revealed that treatment with cpGN13 resulted in reduced lesion areas (Figure 7L and 7M). The number of MOMA-2-positive macrophages within the atherosclerotic plaques was decreased in the aortic arches from the Ldlr^−/− mice following treatment with cpGN13 (Figure 7L and 7M). Thus, cpGN13 inhibits the activation of Gsα and attenuates the development of atherosclerosis.

Figure 7. The state-selective inhibitor of Gsα, cpGN13, reduces macrophage lipid uptake and atherosclerotic progression.

A–B, Representative immunofluorescence analysis to detect active Gsα in BMDMs treated with 25 μM cpGN13 for 12 hours (n = 6 in each). Nuclear staining by DAPI. Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction. Relative values are compared against those of the vehicle group. Scale bar = 25 μm. C–D, ORO staining of BMDM isolated from WT mice were stimulated with cpGN13 (25 μM) and subsequent treated with ox-LDL (50 μg/ml) for 24 hours. BMDMs elicited from WT mice were stimulated with cpGN13 (25 μM) and subsequent treated with 10 μg/ml Dil-ox-LDL for 4 hours (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group. Scale bar = 50 μm. E–G, Western blotting analysis and RT-qPCR of CD36 and SR-A1 in BMDMs from WT mice stimulated with cpGN13 for 24 hours (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction. Relative values are compared against those of the vehicle group. H, Representative immunofluorescence analysis to detect active Gsα. Mice received a 3 mg/kg/day dosage of cpGN13 via tail vein injection. At various time intervals, eye globe blood samples were collected and mononuclear cells were isolated for analysis (n = 6 in each). Scale bar = 50 μm. I, Western blotting analysis of active Gsα in mouse monocytes (n = 6 in each). Statistical analyses were performed using the unpaired two-tailed Student’s t-test with Welch’s correction or unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group. J–K, Representative ORO staining and en face analysis of atherosclerotic lesions in the whole aorta (n = 10 in each group). Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group. L–M, HE staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). Scale bar = 50 μm. ORO staining and cross-sectional analysis of atherosclerotic lesions in the aortic root (n = 10 per group). Scale bar = 50 μm. Immunofluorescence analysis to detect active Gsα and MOMA-2 in frozen aortic root sections from Ldlr^−/− mice with or without cpGN13 (3 mg/kg/day). Scale bar = 100 μm. Statistical analyses were performed using the unpaired two-tailed Student’s t-test. Relative values are compared against those of the vehicle group.

Discussion

In this study, we discovered that Gsα activation was enhanced during atherosclerotic progression in both humans and mice, and ox-LDL could activate Gsα through different mechanisms of action by which ox-LDL translocated Gsα from macrophage lipid rafts in the short term and promoted Gnas transcription through ERK1/2 activation and C/EBPβ phosphorylation via oxidative stress via oxidative stress in the long term. Macrophage-specific Gsα deficiency inhibited the development of atherosclerosis in mice. Mechanistically, Gsα regulates the expression of CD36 and SR-A1 through cAMP/CREB-mediated transcription, which plays an important role in foam cell formation and atherosclerotic progression (Figure 8). Overall, our findings reveal the functional roles of macrophage Gsα in atherosclerosis, increase our understanding of the multiple functions of Gsα, and thereby highlight new potential targets and strategies that may be further explored for the treatment of atherosclerosis.

Figure 8. Proposed mechanisms underlying the Gsα-mediated regulation of macrophage foam cell formation during atherosclerosis.

In macrophages, ox-LDL can induce Gsα translocation from lipid rafts to promote Gsα activation in the short term and increase ERK1/2 activation and C/EBPβ phosphorylation via oxidative stress to enhance Gnas transcription in the long term. Active Gsα increases cAMP levels and the transcriptional activity of CREB and subsequently elevates the expression of CD36 and SR-A1, thereby promoting cholesterol influx and atherosclerotic progression. Inhibiting Gsα activation with cpGN13 in macrophages reduces lipid uptake, foam cell formation, and atherosclerotic progression. aGsα represents active Gsα. GPCR represents G protein-coupled receptors (created with BioRender.com).

Lipid rafts are specialized membrane microdomains that are defined by their cholesterol- and sphingomyelin-rich nature, enrichment in glycosylphosphatidylinositol-anchored proteins, cytoskeletal association, and their resistance to detergent extraction.²³ Lipid rafts harbor numerous signaling molecules, such as receptors, G proteins, and other membrane-associated proteins. They therefore act as platforms for signal transduction.³¹ It has been demonstrated that intact cytoskeletons suppress the Gsα-adenylate cyclase signaling pathway by interacting with components in lipid raft fractions.³² Ox-LDL induces the hydrolysis of sphingomyelin and internalization of cholesterol, which lead to rearrangements in the membrane structure and biochemical alterations in proteins.^33,34 For example, ox-LDL can promote the translocation of caveolin (a marker of rafts) from the plasma membrane to intracellular compartments and thereby displaces raft-associated signaling molecules from lipid rafts.³⁵ In the present study, we found that treatment with ox-LDL resulted in decreased amounts of Gsα in lipid rafts, which resulted in the activation of Gsα and increase in the amount of intracellular cAMP in the short term. Ox-LDL treatment may therefore facilitate the activation of adenylate cyclase by translocating Gsα from lipid rafts to non-lipid raft fractions. This process may involve a conformational change in the Gsα protein, the release of GDP, the attachment of GTP to Gsα, and the dissociation of Gsα from the Gβγ complex. The specific mechanism necessitates further empirical validation by experts in structural biology. However, ox-LDL induced Gnas transcription through C/EBPβ via oxidative stress in the long term. These findings highlight the mechanisms through which ox-LDL activates Gsα in macrophages. It is noteworthy that Giα is also activated during the process of atherosclerosis. However, the specific mechanisms of this activation and the roles in atherosclerosis remain unclear. Future research and exploration are required to elucidate these aspects.

The accumulation of lipids is essential for the differentiation of macrophages and formation of foam cells. Macrophages ingest extracellularly-modified LDL through phagocytosis and pinocytosis mediated by receptors such as CD36 and SR-A1³⁶, which are principal contributors to cholesterol uptake that comprises up to 90% of ox-LDL in macrophages.³⁷ CD36 is an 88-kDa heavily-glycosylated transmembrane protein that belongs to the scavenger receptor class B family.³⁸ CD36 bound to ox-LDL is endocytosed through a raft-mediated pathway that promotes the trapping of macrophages in the arterial intima.³⁹ Various studies, such as those utilizing functional blockage with antibodies, have shown that macrophages from individuals lacking CD36 decrease ox-LDL uptake by approximately 50%.⁴⁰ Some studies have reported that the transcription factors PPAR-γ and Nrf2 can regulate the expression of macrophage CD36.^41,42 SR-A1 is a 77-kDa trimeric transmembrane glycoprotein with six distinct domains⁴³ and binds to ox-LDL to form a ligand–receptor complex that plays a role in membrane transport, recycling, and internalization.⁴⁴ Mutations in the cytoplasmic domain of SR-A1 significantly reduce the SR-A1-mediated cellular internalization of modified LDL and cell adhesion.⁴⁵ It has previously been reported that mimicking the phosphorylation of human SRA at serine 48 reduces SR-A1 expression and that the upstream kinase LKB1 can directly phosphorylate SRA and promote its degradation.^46,47 In our study, we discovered that CREB directly bound to the promoters of CD36 and SR-A1 and induced transcriptional activation. In vitro, the CREB-specific inhibitor KG‑501 significantly reduced the CD36 and SR-A1 protein and mRNA levels, thereby inhibiting lipid uptake and accumulation. Our results implicate CREB as a new and important transcription factor that is involved in the induction of CD36 and SR-A1 and suggest a novel mechanism of action for foam cell regulation and atherosclerosis.

Gsα plays a crucial role in the pathophysiology of numerous diseases. Gsα dysfunction contributes to smooth muscle myopathy, tumoral development in the endocrine glands, and the fibrous dysplasia of bones.^9,12,48 Somatic Gsα-activating mutations can lead to the development of endocrine tumors⁴⁹, which indicates the value of Gsα as a potential therapeutic target for tumors. Gsα can mediate most parathyroid hormone actions in the renal proximal tubule, and alterations in this signaling pathway result in phenotypes that involve parathyroid hormone activity.⁵⁰ Gsα downregulation in juxtaglomerular cells results in kidney injury, which indicates that the Gsα–cAMP signaling pathway is involved in the protection of the renal microvascular endothelium.⁵¹ Gsα mutations result in constitutive elevations in cAMP levels, which can lead to alterations in the expression of several target genes, such as c-fos, c-jun, IL-6, and IL-11. These changes result in alterations in osteoblast recruitment and function in dysplastic bone lesions.⁵² Our previous study showed that Gsα deletions in intestinal smooth muscles result in decreased contractile responses through CREB-mediated Foxf1 expression.¹² Gsα deletions in endothelial cells resulted in impaired microvascular permeability and phenotypes of edema, anemia, hypoproteinemia, and hyperlipoproteinemia via cAMP/CREB-mediated PLVAP expression.¹⁰ Previous study have shown that endothelial Gsα deficiency results in increased endothelial inflammation and progression of atherosclerosis in areas of disturbed flow.⁵³ Besides, cAMP and CREB are a second messenger systems common to many pathways. Previous research has indicated that in obesity models, the paracrine hormone prostaglandin E2 (PGE2) enhances M2 polarization through the activation of the cAMP/CREB-mediated induction of Kruppel-like factor 4 (KLF4) pathways, thereby inhibiting insulin resistance in the context of a high-fat diet.⁵⁴ MaR1, which is an inducer of efferocytosis, is upregulated during coagulation and activate cAMP, ERK1/2 and CREB signaling pathways, promoting leukocyte antimicrobial responses.⁵⁵ A cAMP-CREB cascade induces M2 macrophage-specific gene expression and promotes muscle injury repair.⁵⁶ In the present study, macrophage-specific Gsα deficiency resulted in the inhibited development of atherosclerosis, and the Gsα inhibitor, suramin or cpGN13, reduced the sizes of atherosclerotic plaques in mice. Thus, treatments with Gsα inhibitors may comprise new therapeutic strategies to treat atherosclerosis. However, systemic administration is not the most optimized delivery strategy. In our future work, we will further investigate more optimized methods of drug delivery targeting macrophages, such as nanoparticles and microbubbles specifically designed for macrophages.

In summary, we demonstrated that Gsα deficiency plays a previously-unrecognized role in inhibiting the development of atherosclerosis. We further revealed a new mechanism through which Gsα regulates the expression of CD36 and SR-A1 through CREB-mediated transcription and leads to the promotion of modified lipoprotein uptake and foam cell formation. These findings reveal the regulatory role of macrophage Gsα in the development of atherosclerosis and highlight new strategies and targets for the treatment of atherosclerosis.

Novelty and Significance.

What is known?

• The G protein stimulatory α subunit (Gsα) is essential for hormone-stimulated cAMP production, plays an essential role in the regulation for many physiological processes.

• CD36 and SR-A1 contributes to foam cell formation that drives atherosclerotic progression.

What new information does this article contribute?

• Gsα is gradually activated by translocation from lipid rafts and transcription with ox-LDL treatment and its deficiency in macrophages ameliorates atherosclerosis.

• Gsα increases lipid accumulation by mediating expression of CD36 and SR-A1 through CREB in macrophages, promoting foam cell formation and atherosclerotic progression in mice.

• Inhibition of Gsα in macrophages could be a potential therapeutic strategy for atherosclerosis.

Summary.

Gsα activation is enhanced during atherosclerotic progression, which increases lipid uptake and foam cell formation. In this study, we found that Gsα is gradually activated by translocation from lipid rafts in the short-term and transcription in the long-term. Atherosclerotic progression and Gsα activation increases lipid accumulation by mediating expression of the CD36 and SR-A1 through the CREB, promoting foam cell formation and atherosclerotic progression, providing new insights into the role of lipid uptake in atherosclerosis. Inhibition of Gsα in macrophages could be a potential therapeutic strategy for atherosclerosis.

Non-standard Abbreviations and Acronyms:

BM

bone marrow

BMDM

bone marrow-derived macrophage

ChIP

chromatin immunoprecipitation

CREB

cAMP response element binding protein

CTR

control

DEG

differentially-expressed gene

GNAS

guanine nucleotide binding protein, α-stimulating

GTP

guanosine triphosphate

HE

hematoxylin and eosin

Ox-LDL

oxidized low-density lipoprotein

PBMC

peripheral blood mononuclear cell

RT-qPCR

reverse transcription-quantitative polymerase chain reaction

SR-A1

class A1 scavenger receptor

cAMP

cyclic adenosine monophosphate