Divergence of Acetate Uptake in Proinflammatory and Inflammation-Resolving Macrophages: Implications for Imaging Atherosclerosis

Abstract

Background:

Metabolic divergence of macrophages polarized into different phenotypes represents a mechanistically relevant target for non-invasive characterization of atherosclerotic plaques using positron emission tomography (PET). Carbon-11 (¹¹C)-labeled acetate is a clinically available tracer which accumulates in atherosclerotic plaques, but its biological and clinical correlates in atherosclerosis are undefined.

Methods and Results:

Histological correlates of ¹⁴C-acetate uptake were determined in brachiocephalic arteries of western diet fed apoE^−/− mice. The effect of polarizing stimuli on ¹⁴C-acetate uptake was determined by pro-inflammatory (interferon-γ+lipopolysaccharide) vs. inflammation-resolving (interleukin-4) stimulation of murine macrophages and human carotid endarterectomy specimens over two days. ¹⁴C-acetate accumulated in atherosclerotic regions of arteries. CD68-positive monocytes/macrophages vs. smooth muscle actin-positive smooth muscle cells were the dominant cells in regions with high vs. low ¹⁴C-acetate uptake. ¹⁴C-acetate uptake progressively decreased in proinflammatory macrophages to 25.9±4.5% of baseline (P<0.001). A delayed increase in ¹⁴C-acetate uptake was induced in inflammation-resolving macrophages, reaching to 164.1±21.4% (P<0.01) of baseline. Consistently, stimulation of endarterectomy specimens with interferon-γ+lipopolysaccharide decreased ¹⁴C-acetate uptake to 66.5±14.5%, while interleukin-4 increased ¹⁴C-acetate uptake to 151.5±25.8% compared to non-stimulated plaques (P<0.05).

Conclusions:

Acetate uptake by macrophages diverges upon pro-inflammatory and inflammation-resolving stimulations, which may be exploited for immuno-metabolic characterization of atherosclerosis.

Introduction

Immuno-metabolism, i.e., the cross-talk of intra-cellular metabolic pathways and immune cell function, plays a pivotal role in the pathogenesis of inflammatory diseases, including atherosclerosis^1–3. For example, hypoxia, a major trigger of enhanced glycolysis in atherosclerotic plaques, promotes the production of pro-inflammatory cytokines, e.g., interleukin-1β, by macrophages^4–6, and expansion of the necrotic core⁷. Moreover, pro-inflammatory stimulation of macrophages by lipopolysaccharide (LPS) enhances glucose uptake and glycolysis^8–10, which in turn accentuates the production of interleukin-1β through succinate-mediated stabilization of hypoxia-inducible factor-1α and development of a pseudo-hypoxic state⁸. The unprecedented recognition of the vessel wall immuno-metabolic heterogeneity and its mechanistic contribution to atherogenesis provides the opportunity to explore the potential applications of metabolic imaging in cardiovascular diseases, analogous to similar efforts in oncology^{11, 12}.

Imaging of glucose utilization by ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET) has been instrumental in elucidating the in vivo metabolic heterogeneity of the vessel wall, supplementing the information obtained by ex vivo metabolomics techniques¹. However, the non-specificity of ¹⁸F-FDG uptake, which targets a nearly ubiquitous metabolic process upregulated in both pro-inflammatory and inflammation-resolving (reparative) states, makes it suboptimal for characterization of the inflammatory response as a stand-alone imaging marker^{9, 10, 13–15}. Additionally, myocardial uptake of ¹⁸F-FDG has been a major challenge for its utilization in coronary artery disease^{15, 16}. These limitations have prompted investigators to exploit the utility of metabolic substrates other than glucose, e.g., glutamine⁹ and acetate^{17, 18}, in non-invasive characterization of plaque immuno-metabolism. Notably, the safety of several metabolic substrates has already been established in oncological or cardiac imaging studies^{17, 19}, which facilitates their applications in imaging atherosclerosis.

Carbon-11 (¹¹C)-labeled acetate (¹¹C-acetate) has been extensively studied to detect cancer-associated lipogenesis and myocardial blood flow and oxidative metabolism^{20, 21}. More recently, pre-clinical¹⁸ and pilot clinical¹⁷ studies have demonstrated the feasibility of imaging ¹¹C-acetate uptake in atherosclerosis. However, the biological correlates of acetate uptake in plaques and its clinical implications have not been addressed. The purpose of this study was to determine the histological correlates of acetate uptake in atherosclerotic plaques and its link to inflammatory states of macrophages, using ¹⁴C-acetate where the long half-lived β-emitting C-14 allows ex vivo imaging via autoradiography. We hypothesized that acetate uptake in macrophage-rich plaques differentiates their pro-inflammatory vs. inflammation-resolving polarization states. Histological correlates of acetate uptake were determined by combined immuno-histology and high-resolution ¹⁴C-acetate autoradiography in murine atherosclerotic brachiocephalic arteries. The effect of pro-inflammatory vs. inflammation-resolving stimuli on acetate uptake was assessed in murine macrophage and human carotid endarterectomy specimens.

Materials and Methods

Animals

Wild-type (N=12) and apoE^−/− (N=3) mice on a C57BL/6J background were purchased from Jackson Laboratories. Atherosclerosis was induced by feeding with a western diet (21% anhydrous fat, TD88137, Envigo, 33 weeks) in apoE^−/− mice. Experiments were conducted in accordance with a protocol approved by Institutional Animal Care and Use Committee.

Cell Culture

Commercial experimental reagents are listed in Supplemental Table 1. Elicited peritoneal cells were harvested from wild-type mice through lavage, three days after intra-peritoneal injection of 1.5 mL of 3% thioglycolate²². Adherent macrophages were cultured for two days in RPMI-1640 supplemented with 10% fetal bovine serum, glucose (5.5 mM), sodium pyruvate (1 mM), glutamine (2 mM), non-essential amino acids (1X), HEPES buffer (20 mM), penicillin (50 U/mL) and streptomycin (50 μg/mL), as described^{9, 10, 13}. Macrophages which were not activated by the addition of polarizing stimuli were considered as non-activated (Mφ₀). Polarization was induced by addition of the conventional pro-inflammatory (M1) or inflammation-resolving (M2) stimuli: i) Mφ_IFNγ+LPS: recombinant murine interferon-γ (50 ng/mL) plus LPS (10 ng/mL); and ii) Mφ_IL4: recombinant murine IL-4 (10 ng/mL).

Primary aortic vascular smooth muscle cells (VSMCs) obtained from C57BL/6J mice²³ were cultured for two days in DMEM/F12 supplemented with 10% fetal bovine serum, L-glutamine (2 mM), HEPES buffer (10 mM), penicillin (100 U/mL) and streptomycin (100 μg/mL). Polarization was induced by addition of the conventional pro-inflammatory or inflammation-resolving stimuli: i) VSMC_IFNγ+LPS: recombinant murine interferon-γ (50 ng/mL) plus LPS (10 ng/mL); and ii) VSMC_IL4: recombinant murine IL-4 (10 ng/mL).

Culture of Human Endarterectomy Specimens

Anonymized atherosclerotic specimens were collected from three patients who underwent carotid endarterectomy as the standard-of-care, under a “No Human Subject involvement” classification by Institutional Review Board. Freshly collected specimens were cut into ~3-mm thick slices and cultured ex vivo in complete RPMI-1640 medium^{24, 25}. Plaques were incubated for two days in the absence (control) or presence of i) recombinant human interferon-γ (50 ng/mL) plus LPS (10 ng/mL) vs. ii) recombinant human IL-4 (10 ng/mL).

Uptake of ¹⁴C-Acetate by Macrophages and VSMCs

¹⁴C-acetate uptake was measured in macrophages and VSMCs after 6, 24 or 48 hours of polarization. Immediately prior to the assay, cells were washed with phosphate-buffered saline and were incubated with an uptake buffer (NaCl: 140 mM, KCl: 5.4 mM, CaCl₂: 1.8 mM, MgSO₄: 0.8 mM, D-glucose: 5.5 mM, HEPES: 25 mM) supplemented with 2 μCi/mL (74 kBq/mL) ¹⁴C-acetate at 37°C for 2 hours. Subsequently, cells were washed and lysed with 0.1 mM NaOH. ¹⁴C-acetate uptake was quantified by a liquid scintillation counter (Beckman Coulter LS6500), normalized to the DNA content of the cell lysates (PicoGreen, Invitrogen), and expressed relative to the uptake values of unstimulated MФ₀ or VSMCs^{9, 10, 13}.

Micro-Autoradiography of ¹⁴C-Acetate Uptake

Excised murine brachiocephalic arteries or human endarterectomy specimens were incubated with the uptake buffer (described above) supplemented with 2 μCi/mL (74 kBq/mL) ¹⁴C-acetate at 37°C for 2 hours. Specimens were then washed with a cold buffer and embedded in optimal cutting temperature (OCT) compound for cryosectioning (Leica CM1860) at 10-μm-thickness. Tissues, along with known amounts of ¹⁴C-acetate blotted on papers as standards, were exposed to high-resolution phosphor screens (BAS-IP SR2025 Super Resolution, GE Healthcare). Phosphor screens were scanned by a Sapphire Biomolecular Imager (Azure Biosystems) at 10-μm resolution. Autoradiography images were adjusted using the Fire Look-Up Table and a smoothing algorithm (Image J). The average uptake of ¹⁴C-acetate by endarterectomy specimens stimulated with pro-inflammatory and inflammation-resolving stimuli was quantified (ImageJ) by drawing regions of interests around the tissues (2–3 sections per specimen at 250 μm intervals). Uptake values are expressed relative to control (non-stimulated) plaques.

Histology

Histological analysis of the brachiocephalic arteries was performed after completion of the autoradiography by Movat’s pentachrome staining per standard protocols. Smooth muscle cells and macrophages were identified by immunostaining with anti-mouse α-smooth muscle actin (αSMA) and anti-mouse CD68, respectively (Supplemental Table 1). Immunostaining for inducible nitric oxide synthase (iNOS) and CD206 was performed as conventional markers of M1 and M2 polarization, respectively. Slides were mounted using ProLong^™ Gold Antifade with 4′,6-diamidino-2-phenylindole (DAPI) for nuclear counter-staining. Tissues were photographed using an AxioVert Microscope (Zeiss) equipped with a digital camera, controlled by ZEN software (Zeiss).

High-Resolution Quantification of ¹⁴C-Acetate Uptake

AzureSpot software was used to place a 5×5 μm grid on top of overlayed autoradiography and immunostaining (both CD68 and αSMA) (Supplemental Figure 1). ¹⁴C-aceate intensity data were transferred to FlowJo 10.5.3 software. Each 5×5 μm box was ranked from the lowest to the highest based on its ¹⁴C-acetate uptake and assigned a quartile (Q1 = lowest, Q4 = highest). CD68⁺ and αSMA⁺ percent areas were determined in each quartile of ¹⁴C-acetate uptake (averaged from 2–3 tissues sections per mouse).

Gene Expression Analysis

Messenger RNA (mRNA) extraction and reverse transcription were performed on cultured mouse macrophages and endarterectomy specimens two days after stimulation using TRIzol® reagent and QuantiTect® Reverse Transcription Kit, per standard protocols^{9, 10, 13}. Quantitative PCR (qPCR) was performed using a QuantStudio-3 Real-Time PCR System (ThermoFisher) and TaqMan® gene expression assays (Supplemental Table 2). Transcript amplification data were analyzed using QuantStudio v1.4.3 software (Applied Biosystems) and were normalized to the expression level of 18S ribosomal RNA (Rn18s), as the housekeeping gene. mRNA expression levels are expressed as relative to the level of expression in MФ₀ macrophages and summarized as a heat map. Transcript levels of macrophage polarization markers in endarterectomies are expressed relative to the expression of CD68 to account for variations in the macrophage contents of different specimens.

Statistical Analysis

Data were expressed as mean ± standard error of the mean. Statistical analyses were performed with Prism-8 software (GraphPad). Student t test or analysis of variance (followed by Fisher’s Exact post hoc test) were used to test the differences between two or multiple groups, respectively. P < 0.05 was considered statistically significant.

Results

Uptake of ¹⁴C-Acetate by Macrophage-Rich Murine Brachiocephalic Artery Plaques

Micro-autoradiography of murine brachiocephalic arteries from western diet fed apoE^−/− mice demonstrated focal areas of increased ¹⁴C-acetate uptake in the vessel wall, corresponding to atherosclerotic plaques, as confirmed by overlaid histological (Movat’s pentachrome staining) and autoradiography images (Figure 1A). Consistent with the visual assessment, quantitative assessment of the brachiocephalic arteries trended towards higher ¹⁴C-acetate uptake in plaques compared to plaque-free regions (Figure 1B). To determine the cellular composition of plaque regions with focal ¹⁴C-acetate uptake, we performed immunofluorescent staining of brachiocephalic arteries using CD68 and αSMA antibodies, delineating macrophages and smooth muscle cells, as the two most abundant cells within the plaques. As demonstrated in representative images of combined autoradiography and immunostaining (Figure 2A–D), CD68⁺ macrophages were highly abundant in plaque regions with high ¹⁴C-acetate uptake. Quantification of CD68⁺ and αSMA⁺ areas (Figure 2E) confirmed that macrophages were the predominant cells in regions with the highest quartile (Q4) of ¹⁴C-acetate uptake (CD68⁺ macrophages: 41.2 ± 0.8% vs. αSMA⁺ smooth muscle cells: 18.1 ± 3.5%, P < 0.01). On the other hand, vessel wall regions with the lowest quartile (Q1) of ¹⁴C-acetate uptake were predominantly comprised of αSMA⁺ smooth muscle cells (43.3 ± 1.3%) compared with CD68⁺ macrophages (30.5 ± 2.1%) (Figure 2E, P < 0.01). Of note, ¹⁴C-acetate uptake within the macrophage-rich regions of plaques was heterogeneous, and CD68⁺ regions consisted of areas with high (marked by the red oval shape in Figure 2A and 2D) and low (marked by the red box in Figure 2A and 2D) ¹⁴C-acetate uptake. On the other hand, smooth muscle cells both within the plaques (blue arrows in Figure 2B and 2D) and normal regions of arteries (blue arrowheads in Figure 2B and 2D) had relatively low ¹⁴C-acetate uptake.

Figure 1: Focal uptake of ¹⁴C-acetate by murine atherosclerotic plaques.

Micro-autoradiography (A, left), Movat’s pentachrome (A, middle) and their superimposed (A, right) images demonstrate that vessel wall ¹⁴C-acetate uptake colocalizes with brachiocephalic artery atherosclerotic plaques of apoE^−/− mice fed with a western diet. Plaque uptake of ¹⁴C-acetate (B) trended higher in arterial regions with plaques rather than plaque-free regions.

Figure 2: Colocalization of ¹⁴C-acetate uptake and macrophage-rich regions of atherosclerotic plaques.

Combine micro-autoradiography and immunofluorescent staining for CD68 (macrophage marker) and αSMA (smooth muscle cell marker) (A–D) demonstrates that areas of focal uptake of ¹⁴C-acetate primarily correspond to macrophage-rich regions of murine brachiocephalic artery atherosclerotic plaques. Of note, ¹⁴C-acetate uptake within the macrophage-rich regions of plaques was heterogeneous. Some macrophage-rich regions demonstrated high ¹⁴C-acetate uptake (marked by the red oval shape in A and D), while other regions had low uptake (marked by the red box in A and D). Smooth muscle cells both within the plaque (blue arrows in B and D) and normal regions of artery (blue arrowheads in B and D) had relatively low ¹⁴C-acetate uptake. Quantification of CD68⁺ and αSMA⁺ areas in vessel wall regions with increasing quartiles of ¹⁴C-acetate uptake (E) confirms that macrophages are the most abundant cells in regions with higher ¹⁴C-acetate uptake, while smooth muscle cells comprise the most abundant cells in regions with lower ¹⁴C-acetate uptake (N = 3 mice, 2–3 tissue sections per each mouse). ** indicates P < 0.01 for within quartile comparison of abundance of macrophages and VSMCs; P values above the brackets indicates between quartiles comparisons of the abundance of macrophages and VSMCs.

Divergence of ¹⁴C-Acetate Uptake upon M1 and M2 Polarization of Macrophages

The striking heterogeneity of ¹⁴C-acetate uptake in CD68⁺ regions of plaques led us to examine the association between acetate uptake and different polarization states of macrophages. We determined through immunofluorescence staining that both M1-like (CD68⁺/iNOS⁺) and M2-like (CD68⁺/CD206⁺) macrophages were present in murine brachiocephalic plaque (Supplemental Figure 2).

We utilized a conventional approach to induce pro-inflammatory (M1) and inflammation-resolving (M2) polarization states in murine macrophages using IFN-γ plus LPS (Mφ_IFNγ+LPS) and IL-4 (Mφ_IL4), respectively, over a 2-day period. Successful conversion of macrophages into different polarization states was confirmed by markedly different transcript levels of a panel of conventional M1 (Arg2, Cxcl9, Cxcl10, Il1b, Nos2 and Tnf) and M2 (Arg1, Cd36, Chil3, Mrc1, Tfrc, Tgfb) markers (Figure 3).

Figure 3: Distinct inflammatory profiles of Mφ₀, Mφ_IFNγ+LPS and Mφ_IL4.

Heat map illustration of macrophage polarization transcript levels (relative to MΦ₀) confirms the development of distinct polarization phenotypes upon 2-day culture of murine macrophages with by IFN-γ + LPS vs. IL-4. (N = 5)

The uptake of ¹⁴C-acetate by macrophages were determined at 6, 24, and 48 hours after the induction of polarization. As demonstrated in Figure 4A, there was a steady decline in ¹⁴C-acetate uptake in Mφ_IFNγ+LPS, which was noticeable as early as 6 hours after stimulation (59.5 ± 3.7% of the baseline level, P < 0.001) and reached to its minimum at 48 hours (25.9 ± 4.5% of the baseline level, P < 0.001). On the other hand, Mφ_IL4 demonstrated a significant, but delayed, increase in ¹⁴C-acetate uptake, reaching to 164.1 ± 21.4% of the baseline level at 48 hours after stimulation (P < 0.01). The ¹⁴C-acetate uptake by VSMCs, however, was essentially unchanged upon stimulation by either IFNγ + LPS or IL-4 (Supplemental Figure 3).

Figure 4: Divergence of ¹⁴C-acetate uptake in Mφ_IFNγ+LPS and Mφ_IL4.

Stimulation of murine macrophages with IFN-γ + LPS vs. IL-4 resulted in the divergence of ¹⁴C-acetate (A), characterized by a progressive decline by ~4 folds in Mφ_IFNγ+LPS and a delayed ~64% increase in Mφ_IL4. mRNA level of Slc16a1 (B) is increased in both Mφ_IFNγ+LPS (3.9-fold) and Mφ_IL4 (6.3-fold). Expression of Slc16a3 transcript (C) was also increased in Mφ_IFNγ+LPS (6.1-fold); but the apparent overexpression of Slc16a3 in Mφ_IL4 (3.3-fold) did not reach statistical significance (P = 0.14). The expression of Slc16a7 (D) was not significantly changed by the polarization state of macrophages (N = 7 for uptake assays and = 5 for gene expression assays).

The transport of acetate through cell membrane occurs through members of the monocarboxylate transporter family²⁶. Therefore, we determined the mRNA expression of Slc16a1, Slc6a3, and Slc16a7, encoding monocarboxylic acid transporter −1, −3, and −2, respectively (Figure 4B–D). The expression of Slc16a1, representing the most abundant of the three transporters^{26, 27}, was 3.9 ± 0.3 (P = 0.05) and 6.3 ± 1.6 (P < 0.01) fold higher in Mφ_IFNγ+LPS and Mφ_IL4 compared to Mφ₀, respectively (P < 0.001). The expression of Slc16a3 was also 6.1 ± 1.3 (P < 0.01) and 3.3 ± 1.2 (P = 0.14) fold higher in Mφ_IFNγ+LPS and Mφ_IL4 compared to Mφ₀, respectively. The expression of Slc16a7, the least abundant transporter, was comparable among the different polarization states of macrophages.

¹⁴C-Acetate Uptake by Human Carotid Plaques Stimulated by IFN-γ + LPS vs. IL-4 Mirrors the Changes in Polarized Macrophages

In order to determine the effect of different polarizing agents on acetate uptake in the vessel wall micro-environment, we utilized a previously validated technique of ex vivo culture of human carotid endarterectomy specimens^{24, 25}. Induction of different polarization states in endarterectomy specimens were confirmed by the increased expression of M1 polarization markers, interleukin-1β (IL-1β) and nitric oxide synthase-2 (NOS2), by IFN-γ + LPS (Figure 5A) vs. M2 polarization markers, CD163 and mannose receptor C-type 1 (MRC1), by IL-4 (Figure 5B). Consistent with our observation in polarized murine macrophages, stimulation of human endarterectomy specimens with IFN-γ + LPS led to reduction of ¹⁴C-acetate uptake to 66.5 ±14.5% of non-stimulated plaques over a 2-day period (Figure 5C–D). On the other hand, stimulation by IL-4 increased ¹⁴C-acetate uptake to 151.5 ± 25.8% (Figure 5C–D) (P<0.05).

Figure 5: Divergence of ¹⁴C-acetate uptake by human atherosclerotic plaques upon pro-inflammatory vs. inflammation-resolving stimulation.

Induction of distinct polarization states in human carotid endarterectomy specimens is confirmed by quantification of M1 polarization markers (interleukin-1β (IL-1β) and nitric oxide synthase-2 (NOS2), panel A) vs. M2 polarization markers (CD163 and mannose receptor C-type 1 (MRC1), panel B). Quantitative autoradiography demonstrates that stimulation of plaques with IFN-γ + LPS leads to a significant decline in ¹⁴C-acetate uptake, while IL-4 stimulation increases ¹⁴C-acetate uptake (C–D). N = 3 endarterectomy specimens (2–3 tissue sections per each specimen quantified).

Discussion

As summarized in schematic Figure 6, our study demonstrated that uptake of acetate is primarily localized to macrophage-rich regions of atherosclerotic plaques. Additionally, we showed a divergence in acetate uptake by macrophages and endarterectomy specimens stimulated with pro-inflammatory (IFNγ + LPS) vs. inflammation-resolving (IL-4) stimuli, suggesting the potential of ¹¹C-acetate PET in characterizing the inflammatory state of the vessel wall.

Figure 6: Schematic summary of the study.

¹⁴C-acetate uptake is primarily localized within the macrophage-rich regions of atherosclerotic plaques in brachiocephalic arteries of apoE^−/− mice. Ex vivo polarization of murine peritoneal macrophages and human carotid endarterectomy specimens into pro-inflammatory and inflammation-resolving states, by IFN-γ + LPS vs. IL-4, respectively, leads to a marked divergence in their uptake of ¹⁴C-acetate, suggesting a potential venue for immuno-metabolic characterization of atherosclerotic plaques by PET.

The growing appreciation of the contribution of plaques’ immuno-metabolic heterogeneity to the vessel wall biology¹ underscores the potential role of metabolic imaging in non-invasive characterization of atherosclerosis, which may improve the risk stratification of patients and monitoring the response to preventive or therapeutic interventions^{15, 16}. To further enhance the translational potential of this study, we used murine brachiocephalic plaques, rather than aortic plaques, as the former more accurately models the complexities of human plaque pathophysiology^28–30.

Despite initial promising results indicating an association between ¹⁸F-FDG uptake and the pro-inflammatory state of vessel wall macrophages^{15, 16, 31}, emerging evidence supports that glucose uptake, as a stand-alone marker, does not reveal sufficient immuno-metabolic information to allow discrimination of plaques macrophages from smooth muscle cells^{14, 32} or pro-inflammatory from inflammation-resolving macrophages^{5, 9, 10, 13}.

A promising approach to overcome the non-specificity of ¹⁸F-FDG in elucidating vessel wall immuno-metabolism is a combined imaging approach to target other key metabolic pathways, which may aid in differentiation of different immune cell phenotypes. For example, combined imaging of glutamine and glucose uptake improves the identification of different polarization states of macrophages⁹. Considering the central roles of acetate in various immuno-metabolic pathways, including tricarboxylic acid cycle, lipogenesis and protein acetylation³³, ¹¹C-acetate PET represents a promising, but largely unexplored, approach in inflammatory diseases. Notably, ¹¹C-acetate uptake has been reported in ~30% of atherosclerotic plaques¹⁷. However, biological/immuno-metabolic correlates and clinical implications of acetate uptake in atherosclerosis remained to be determined. In this study we showed that acetate uptake is primarily localized to macrophage-rich regions of atherosclerosis.

In order to identify the immuno-metabolic implications of acetate uptake, we determined the time-course of changes in cultured macrophages over a 2-day period of ex vivo polarization. We identified a progressive decline in acetate uptake by pro-inflammatory macrophages, but a delayed increase in acetate uptake in inflammation-resolving macrophages. The distinct temporal pattern of acetate uptake in polarized macrophages may reflect differences in IFN-γ + LPS vs. IL-4 induced cellular metabolic reprogramming, particularly in M2-like macrophages which rely on mitochondrial biogenesis for activation and subsequent upregulation of specific metabolic pathways (e.g., tricarboxylic acid cycle, β-oxidation, oxidative phosphorylation)^{10, 34, 35}. By contrast, IFN-γ + LPS stimulation affects macrophage metabolism by more rapid regulation of flux through certain pathways, including enhanced glycolysis and suppression of the tricarboxylic acid cycle and oxidative phosphorylation cycle⁸. This divergence of acetate uptake in differentially polarized macrophages highlights the potential role of ¹¹C-acetate PET in non-invasive discrimination of inflammatory states of atherosclerotic plaques. As a proof-of-principle ex vivo experiment, we confirmed that IFN-γ + LPS reduces, while IL-4 increases, the acetate uptake in human endarterectomy specimens.

Cellular transport of acetate is primarily through monocarboxylate transporters²⁶. Consistent with prior reports, Slc16a1 and Slc16a3 were the most abundant transporters at the transcriptional level^{26, 27}. Interestingly, Slc16a1 was ~6-fold over-expressed in Mφ_IL4, consistent with their higher acetate uptake, compared to Mφ₀. There was also ~4- and ~6-fold increase in the expression of Slc16a1 and Slc16a3 in Mφ_IFNγ+LPS, despite their markedly reduced acetate uptake, which likely reflects their role in transport of other substrates, particularly lactate, which is required for glycolytic reprogramming of macrophages in response to LPS³⁶.

Intra-cellular metabolic fate of various substrates, e.g., glucose and glutamine, plays crucial role in phenotypic alterations of macrophages in response to different stimuli^{8, 37–39}. Through acetyl-CoA synthetase mediated conversion into acetyl-CoA and subsequent condensation with oxaloacetate to generate citrate, intra-cellular acetate is at the cross-road of metabolic pathways and may contributes to lipogenesis and tricarboxylic acid (TCA) cycle¹⁸. Analysis of rabbit atherosclerotic arteries administered with ¹⁴C-acetate has revealed that acetate contributes to both aqueous (e.g., glutamate) and fat-soluble (e.g., triglycerides and cholesterol esters) metabolites^{18, 40, 41}. Our study demonstrated profound divergence of overall uptake of acetate by polarized murine macrophages and human atherosclerotic plaques. However, potential alterations in intra-cellular utilization of acetate by its differential fluxes through different pathways and the potential impact of these metabolic changes on macrophage biology and atherogenesis remained to be determined. Importantly, the dichotomy of IFNγ + LPS (i.e. M1-like) and IL-4 (i.e. M2-like) stimulated macrophages fails to capture the profound and nuanced complexity of macrophage phenotypes, and their unique metabolic and functional profiles⁴².

In this study, we took advantage of the higher spatial resolution of lower energy β⁻ decay through ¹⁴C-acetate autoradiography, in contrast to high energy low-resolution β⁺ by ¹¹C-acetate, to determine the histological distribution of acetate uptake within the vessel wall. The long half-life of ¹⁴C-acetate was also compatible with the lengthy tissue processing steps required for micro-autoradiography, which would have been challenging for the short half-life of ¹¹C (~20 min). Although, the use of ¹⁴C-acetate eliminated the possibility of PET, the feasibility of in vivo detection of ¹¹C-acetate uptake in atherosclerosis has been previously demonstrated¹⁷; hence, our primary purpose was to determine the histological correlates of acetate uptake to unravel its association with plaque biology and eventually its clinical implications. Future clinical studies are needed to correlate in vivo ¹¹C-acetate uptake with molecular and histological markers of plaque vulnerability, like the macrophage burden, the abundance of different polarization states of macrophages, and pro/anti-inflammatory cytokine production, and disease outcome.

While acetate is a promising molecular imaging probe for atherosclerotic plaques¹⁷, certain limitations common to other immune-metabolic tracers, including low spatial resolution with respect to plaque size and low signal-to-background, remain roadblocks for translational application. Additionally, high ¹¹C-acetate uptake by metabolically active organs, particularly heart and liver, may limit its usefulness in imaging coronary artery disease^{17, 43–45}. ¹¹C-acetate PET may therefore be more suited for assessment of atherosclerosis in larger arteries, particularly carotid arteries, to predict and monitor response to conventional, e.g., statin, or emerging immunomodulatory therapies.

In summary, our data demonstrate that acetate uptake within the vessel wall primarily correlates with macrophage-rich atherosclerotic plaques. Moreover, classical pro-inflammatory polarization of macrophages and atherosclerotic plaques significantly reduces their uptake of acetate, while alternative activation upregulates the uptake. Together, these data suggest a potential link between the enhanced uptake of acetate and the induction of an inflammation-resolving state in atherosclerotic plaques. Further in vivo imaging studies are required to determine if ¹¹C-acetate PET, alone or in combination with other metabolic substrates, e.g., ¹⁸F-FDG and ¹⁸F-fluoroglutamine, may play a role in identifying vulnerable plaques and in monitoring the response to anti-inflammatory interventions, such as interleukin-1β blocking antibody, Canakinumab⁴⁶.

New Knowledge Gained:

Our results demonstrate that arterial uptake of acetate is mostly localized to macrophage-rich regions of atherosclerotic plaques compared to regions enriched in smooth muscle cells. Additionally, we showed that a profound divergence in acetate uptake distinguishes the classical (pro-inflammatory) and alternative (inflammation-resolving) polarization states of macrophages, induced by interferon-γ + lipopolysaccharide vs. interleukin-4, respectively, in two different ex vivo models, i.e., cultured murine macrophages and human endarterectomy specimens. These data provide a new potential venue for in vivo immuno-metabolic characterization of plaques by non-invasive imaging.

Abbreviations:

¹⁸F-FDG

¹⁸F-fluorodeoxyglucose

PET

Positron emission tomography

IL-1β

Interleukin-1β

IL-4

Interleukin-4

IFN-γ

Interferon-γ

LPS

Lipopolysaccharide

NOS2

Nitric oxide synthase-2