Immunometabolic modulation of retinal inflammation by CD36 ligand

In subretinal inflammation, activated mononuclear phagocytes (MP) play a key role in the progression of retinopathies. Little is known about the mechanism involved in the loss of photoreceptors leading to vision impairment. Studying retinal damage induced by photo-oxidative stress, we observed that cluster of differentiation 36 (CD36)-deficient mice featured less subretinal MP accumulation and attenuated photoreceptor degeneration. Moreover, treatment with a CD36-selective azapeptide ligand (MPE-001) reduced subretinal activated MP accumulation in wild type mice and preserved photoreceptor layers and function as assessed by electroretinography in a CD36-dependent manner. The azapeptide modulated the transcriptome of subretinal activated MP by reducing pro-inflammatory markers. In isolated MP, MPE-001 induced dissociation of the CD36-Toll-like receptor 2 (TLR2) oligomeric complex, decreasing nuclear factor-kappa B (NF-κB) and NLR family pyrin domain containing 3 (NLRP3) inflammasome activation. In addition, MPE-001 caused an aerobic metabolic shift in activated MP, involving peroxisome proliferator-activated receptor-γ (PPAR-γ) activation, which in turn mitigated inflammation. Accordingly, PPAR-γ inhibition blocked the cytoprotective effect of MPE-001 on photoreceptor apoptosis elicited by activated MP. By altering activated MP metabolism, MPE-001 decreased immune responses to alleviate subsequent inflammation-dependent neuronal injury characteristic of various vision-threatening retinal disorders.

The relationship between subretinal inflammation and photoreceptor degeneration is well established [18][19][20][21] . The inhibitory effect of MPE-001 on MP infiltration into the subretinal space motivated further study of the capacity of the azapeptide CD36 ligand MPE-001 to preserve photoreceptor integrity in blue light-challenged mice. Glial fibrillary acidic protein (GFAP) 22,23 is upregulated in response to MP activation, and can be used to index retinal degeneration. GFAP was mainly expressed in inner retinal layers in both CD36 +/+ and CD36 −/− mice (Fig. 1D,E). Blue light exposure of WT mice caused photoreceptor degeneration evidenced by a thinner outer nuclear layer (ONL), a corresponding decrease in their function substantiated by lower a-wave and dependent b-wave electroretinographic amplitudes, as well as an increase in GFAP expression throughout the retina (Fig. 1D-M). MPE-001 administered to blue light-challenged WT mice preserved photoreceptor ONL thickness, restored aand b-wave amplitudes, and reduced GFAP expression to values comparable to those seen in non-illuminated CD36 +/+ mice. Hence, blue light-induced photoreceptor injury involved CD36.

MPE-001 downregulates inflammatory markers of activated MPs and reduces photoreceptor degeneration in vivo.
Laser capture microdissection was used to determine mRNA levels of inflammatory markers in the area between the ONL and the RPE ( Fig. 2A, area delineated by green line) in mice treated with MPE-001 compared with vehicle-treated control mice (n = 3-4 mice/group). MPE-001-treated mice exhibited significant reductions in the expression of iNOS, IL-12 and IBA-1 mRNA respectively by 65, 56 and 47%; whereas anti-inflammatory IL-10 mRNA was augmented by MPE-001 ( Fig. 2A). Correspondingly, RPE flat mounts from animals treated with MPE-001 showed respectively 62% and 45% decreased numbers of iNOS + /F4/80 + -and IL-12 + /IBA-1 + -stained cells compared to those from control mice (Fig. 2B,C,E). In addition, MPE-001 elicited increased expression of CD206, a surface marker of anti-inflammatory (M2-like) MPs (Fig. 2D,E). Consistent with a decrease in pro-inflammatory MPs and ensued preservation of the ONL structure ( Fig. 1F) 26,27 , MPE-001 prevented against the degeneration of cone photoreceptor segments and the mislocalization of the S-opsin (Fig. 2F,G); it was also reducing apoptosis in the photoreceptors in WT mice exposed to blue light-illumination (Fig. 2H,I).
MPE-001 diminishes TLR2-mediated proinflammatory cytokine and chemokine release in vitro and protects against photoreceptor degeneration. Toll-like receptors (TLR) in association with cofactor proteins play crucial roles in innate immunity that trigger inflammatory responses 28 . The CD36, as co-receptor of TLR2/6 heterodimer, enhanced the TLR2-signaling pathway in the presence of its agonists, such as the diacylated lipoproteins LTA and R-FSL1 [29][30][31] . Upon stimulation by specific ligands, the TLR2/6-CD36 complex triggers the activation of NFκB and MAPKs (P38 and JNK) which elicit an inflammatory response in MPs 13,29 . On the other hand, TLR2/1 heterodimer can be activated independently of the co-receptor CD36 29 . The role of CD36 in the mitigating effects of MPE-001 on TLR2-mediated inflammation was investigated in purified systemic MPs (peritoneal) from CD36 +/+ and CD36 −/− mice, which were stimulated with IFNγ to induce a proinflammatory phenotype. The selectivity of MPE-001 to the CD36-TLR2 signaling pathway was demonstrated using a set of selective TLR agonists 29-31 : R-FSL1 and LTA for TLR2/6 32,33 , pgLPS for TLR2/4 34 , PAM3CSK4 for TLR2/1 35

MPE-001 decreases NLRP3 inflammasome. The intracellular nucleation of CD36 ligands has
been reported to trigger inflammasome activation 40 . We investigated next the ability of MPE-001 to mitigate CD36-dependent TLR2/6 stimulation of the inflammasome pathway. In IFNγ-primed peritoneal macrophages, induced IL-1β secretion on stimulation of TLR2/6 with R-FSL1 (n = 3-4/group) was associated with increased expression of NLRP3, pro-caspase1 and caspase1, all of which were attenuated by MPE-001 ( Fig. 5A-F, Table S2), underscoring the critical photoreceptor cytotoxic effects of IL-1β as reported 41 . The role of inflammasome-generated IL-1β in inducing photoreceptor cytotoxicity was again studied on retinal explants incubated with conditioned media of R-FSL1-stimulated and unstimulated BMDM (n = 3/group) (Fig. 5G). www.nature.com/scientificreports www.nature.com/scientificreports/ Conditioned media from these R-FSL1-stimulated MPs induced an increase in apoptotic cell number in the photoreceptor layer, which was attenuated by anti-IL-1β antibody as well as by MPE-001 (Fig. 5G,H).
Subretinal MPs in inflammatory retinal disease of elderly subjects display similar profiles of CD36 and TLR2 expression to those observed in light-induced inflammation in mice. Human retina from healthy elderly donors and from patients presenting age-related retinal inflammation were stained with IBA-1. Their subretinal myeloid cells were examined in RPE flat mounts (counter staining with rhodamine phalloidin, Fig. S6B,C) and compared to retina from blue light-illuminated mice (Fig. S6A). The immune cell accumulation that was found in retinas of elderly human donors with inflammatory eye disease was comparable to that observed in mice exposed to blue light irradiation (Fig. S6A,C). RPE flat mounts showed expression of CD36 (Fig. S6D) and TLR2 (Fig. S6E) in all subretinal IBA-1 + MPs. Atrophy of the RPE correlated with the accumulation of IBA-1 + MPs (Fig. S6C magnification) and contrasted with the regular shape of RPE cells from an elderly donor without subretinal inflammation (Fig. S6B magnification).

Discussion
Inflammatory processes play critical roles in the pathogenesis of various retinal diseases 48 . Although regulation of the complement system has garnered significant contemporary attention for the treatment of degenerative retinal diseases, such as age-related macular degeneration 49 , alternative strategies to curb chronic inflammation driven by MPs merit further study to address underlying causes of tissue damage. Neuronal cell death, such as that seen for photoreceptors and retinal ganglion cells, is often the consequence of MP activation through TLR pattern recognition and intracellular signaling to stimulate genes encoding pro-inflammatory cytokines 9 . TLR2 pro-inflammatory function has been postulated to contribute to RPE apoptosis due to inflammation and oxidant stress 8 . Considering their important roles in degenerative retinal disease pathology, TLR signaling pathways have emerged as promising targets for mitigating MP-driven inflammation; however, direct TLR inhibition and antagonism have to date had limited success in the clinical setting 50,51 . Alternatively, cofactor proteins which associate with TLRs and modulate their activity, represent an unexplored means for disrupting their signaling. Studying modulators of CD36, we have identified the azapeptide ligand MPE-001 that binds this cofactor protein and consequently interferes with TLR2 signaling. Herein, we provide mechanistic insights into the mode of action of MPE-001. After binding to the hydrophobic region of CD36 52,53 , MPE-001 disrupts the interaction between CD36 and the TLR2/6 heterodimer at the MP cell membrane, and subsequently perturbs downstream signaling by attenuating the relevant photo-oxidative stress-triggered pro-inflammatory cascade. Concomitantly, MPE-001 enhances the metabolic rate of MPs through PPAR-γ induction which in turn contributes to suppression of inflammation via a common NLRP3 link with the TLR2 pathway; abrogation of inflammation preserves photoreceptor integrity (see schematic diagram in Fig. 7).
The class B scavenger receptor CD36 was first identified as a fatty acid transporter involved in energy metabolism 54 , and later implicated in TLR-dependent inflammatory response and sterile inflammation featuring inflammasome activation in MPs 40 . The selective azapeptide CD36 ligand MPE-001 has now been shown to modulate activation of the TLR2/6 heterodimer and downstream signaling in MPs, resulting in decreased pro-inflammatory cytokine and chemokine release, and mitigation of the influx, activation and accumulation of MPs into the subretinal space, which is normally devoid of immune cells 24 . The consequences of MPE-001 treatment include attenuation of inflammation which causes RPE and photoreceptor layer degeneration 55 , as demonstrated in blue light-exposed mice in vivo. The efficacy of MPE-001 was further substantiated in the CX3CR1-deficient murine model in which the absence of CX3CR1 accelerates tissue damage and retinal degeneration due to increased presence of mononuclear phagocytes in the retina upon exposure to photo-oxidative injury 20 . By reducing proinflammatory cytokine levels and mononuclear phagocyte recruitment, MPE-001 exhibited cytoprotective effect, prevented photoreceptor loss, and preserved significantly retinal function after exposure to conditions of photo-oxidative stress that mimic chronic inflammation 56 . Moreover, MPE-001 modulated the assembly of cytoplasmic components of the inflammasome and decreased IL-1β release in subretinal MPs. Although CD36 has been shown to act as non-opsonic phagocytic receptor 57 , and to cooperate with TLR4 in bacterial endocytosis and phagocytosis by MPs 58 , MPE-001 interfered selectively with TLR2 stimulation without altering the MP phagocytic function. The manner by which MPE-001 preserved the innate immune response is suggestive of selective www.nature.com/scientificreports www.nature.com/scientificreports/ biased allosteric modulation of the CD36 interaction with TLR2; these combined observations were consistent in MPs from distinct sources, including in human monocytes.
The role of the CD36-TLR2 interaction in mediating inflammation and ensuing neurotoxicity offers a novel target for therapeutic intervention. The prototype MPE-001 disrupted association between CD36 and TLR2 proteins labeled with fluorescent probes as demonstrated by an observed reduction of the energy transfer caused on binding to the TLR2 agonist R-FSL-1. Consequently, the normal TLR2-signaling pathway was interrupted by MPE-001 as demonstrated by diminished phosphorylation of various downstream signals of TLR2 59,60 .
Cell-specific responses may be mediated by CD36, which interacts with multiple ligands and binding partners: e.g., TLR heterodimers, β1 and β2 integrins 61 , and tetraspanins 62 to activate NF-κB, NLRP3, Src/Lyn/Fyn, MAPKs and TGFβ signalling pathways. Among multiple CD36 lipid-related ligands, oxidized phospholipids were shown to promote the activation of TLR4/6-dependent innate immune response 63,64 . Binding of oxidized LDL is associated with the upregulation of inflammatory cytokine expression and inflammasome stimulation to trigger pro IL-1β and NLRP3 activation 40 . On binding to CD36, MPE-001 decreased IL-1β release through a modulatory effect on the activation of the TLR2/6-CD36 complex by ligands such as the diacylpeptide agonist R-FSL1. The modulatory roles of MPE-001 on TLR2-dependent inflammatory processes and sterile inflammation, are both mediated through NLRP3 activation. As a co-receptor of TLR2, CD36 activates AP-1 and triggers gene transcription of proinflammatory cytokines, primarily through activation of c-Jun N-terminal kinase (JNK) and P38 58 .
Our results indicate that azapeptide MPE-001 decreased AP-1 activation by reducing phosphorylation of JNK and P38 in activated MPs.
MPs in the subretinal space and many other tissues are characterized as resident and invading pathology-triggered inflammatory-types. Analogous to the adaptive immune system in which Th1 and Th2 cells have been characterized, MPs have also been subdivided based on their cytokine production 42 . MPs activated by DAMPs are pro-inflammatory, anti-angiogenic and potentially neurotoxic. Those stimulated by anti-inflammatory cytokines (e.g. IL-4) display pro-angiogenic properties, promote phagocytosis and are anti-inflammatory. However, the spectrum of MP phenotypes is broader than traditionally specifically-labelled M1 and M2 subtypes 65,66 . The inflammatory profile of MPs is also affected by metabolic rate, such that inhibition of glycolysis or oxidative phosphorylation alters respectively M1 or M2 activation 67,68 . CD36 affects metabolic pathways. CD36 modulators enhance influx of oxidized LDL and separately, efflux of cholesterol via ABCA1/G1 transporters through a PPAR-γ-dependent process 69 . PPAR-γ repression (antagonism) enhances the glycolytic metabolic pathway 70,71 , stimulates oxidative phosphorylation 72 , and augments concomitantly anti-inflammatory cytokines to curtail the pro-inflammatory actions of excessive CD36 induction 73 . Our findings in vitro and in vivo concur with these concepts, which reveal that CD36 azapeptide ligands exert anti-inflammatory properties by affecting the TLR2-inflammasome pathway and by shifting metabolic rate to increase oxygen consumption by influencing the PPAR-γ pathway. Accordingly, MPE-001 acts by inhibiting the signaling of CD36 to certain www.nature.com/scientificreports www.nature.com/scientificreports/ pathways (specifically NF-κB-inflammasome) and activating others (notably PPAR-γ-PGC1α), consistent with the biased signaling actions that we have reported for other CD36 ligands 16,69 .
The effects of MPE-001 in decreasing inflammation and IL-1β release may go beyond MPs 74 , and extend to other cells expressing CD36 and inflammasome components, such as the RPE and choroidal endothelium 75 . Notably, in the late stages of human RPE degeneration, NLRP3-inflammasome activation and increased levels of IL-1β correlate with oxidative stress that leads to lipid peroxidation end products such as 4-hydroxynonenal and carboxyethylpyrrole 75 . Considering IL-1β release elicits subretinal accumulation of MPs responsible for cone segment degeneration with loss of high visual acuity 21 , the potential for MPE-001 to reduce IL-1β release may be exploited to prevent cone cell loss in geographic atrophy 21 , consistent with observations made herein. In addition, downregulation of expression of pro-inflammatory inducible nitric oxide synthase and IL-12 in subretinal MPs on treatment with MPE-001 in vivo was also accompanied by increase in expression of anti-inflammatory IL-10 which may further dampen NLRP3 expression, inflammasome assembly and caspase-8 activation 76 .
In summary, we have shown for the first time that MPE-001, a selective azapeptide ligand of CD36, can specifically modulate the CD36-TLR2 interaction and the induction of PPAR-γ/PGC-1α. Consequently, MPE-001 mitigated inflammation and ensuing neurotoxicity, consequences that are regularly observed in degenerative outer-and sub-retinal disorders. Considering subretinal inflammation with accumulation of activated MPs is prevalent in retina from elderly human patients and mirrored in mice subjected to photo-oxidative stress, modulators such as MPE-001 offer promise as a novel prototype for therapeutic targeting of the CD36 receptor to mitigate chronic MP-driven inflammation in vision-threatening maladies, such as retinitis pigmentosa, diabetic retinopathy and age-related macular degeneration. www.nature.com/scientificreports www.nature.com/scientificreports/ ophthalmic atropine solution 1% (Alcon) was applied to both eyes daily. MPE-001 (289 nmol/kg) was administered s.c. at 24 h following blue light exposure for 7 consecutive days. At the end of the blue light exposure period, the mice were maintained on a 12:12 h light: dark cycle for 3 days before being sacrificed.

Isolation and culture of mouse peritoneal macrophages. Unstimulated peritoneal macrophages
were harvested by washing the peritoneal cavity of 12-week-old C57BL/6J and CD36 −/− male mice using 10 mL Dulbecco's Modified Eagle's Medium (DMEM) cell-culture medium. Peritoneal macrophages were purified by depletion of non-target cells using the Monocyte Isolation Kit (Miltenyi, 130-100-629) according to the manufacturer's instructions. Flow cytometry analysis indicated that the cell population contained above 98% F4/80 + CD80 + cells. Purified peritoneal macrophages were plated in DMEM containing 10% Fetal Bovine Serum (FBS) and 20 ng/mL interferon γ (IFNγ) at 37 °C in a 5% CO 2 -enriched atmosphere. After 48 h, cells were washed twice with PBS to remove IFNγ and FBS. Peritoneal macrophages were then weaned off FBS by incubation for 2 h with DMEM containing 0.2% Bovine Serum Albumin (BSA) prior to stimulation.
Colocalization of CD36-TLR2 in lipid rafts. Peritoneal macrophages (10 7 ) plated in petri dishes (10 mm 2 ) were stimulated for 5 min with the TLR2 agonist (R-FSL1, 300 ng/ml) or vehicle. Cells were lysed in 1 mL of RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, 1% Triton X-100, 50 mM NaF, 2 mM EDTA, pH 7.4) containing protease and phosphatase inhibitors (Pierce Biotechnology) for 1 h on ice. Cell lysates were mixed with an equal volume of 5% sucrose in RIPA and placed at the bottom of a centrifuge tube then subjected to sucrose density gradient. Samples were overlaid with 1 ml of 40%, 1 mL of 30%, 2 mL of 25%, 2 mL of 20%, 2 mL of 15% and 2 mL of 10% sucrose in RIPA, then centrifuged at 100,000 g for 16 h at 4 °C. Fractions (1 mL) were gently removed from the top of the gradient and subjected to centrifugal filtration Microcon (Millipore, MRCPRT010) with a membrane Nominal Molecular Weight Limit (NMWL) of 10 kDa for protein concentration and desalting. N-octylglucoside (25 mM) was added to each fraction to solubilize lipid rafts and equivalent portions of each fraction were analyzed by SDS-PAGE and immunoblotted using primary antibodies: anti-CD36 (NB400-145), anti-TLR2, anti-MyD88, anti-Flottilin1 or anti-TLR6, all used at 1:500 vol:vol. After the washing steps, blots were incubated for 1 h at room IL-1β and NLRP3 protein assays. Peritoneal macrophages (1.5 × 10 6 ) were seeded in 12-well plates in DMEM/10% FBS overnight. Cells were weaned off FBS for 2 h with DMEM containing 0.2% BSA, then were treated with MPE-001 (10 −7 M) or vehicle and stimulated with 300 ng/mL R-FSL1 for 4 h. Thirty minutes before the end of stimulation, 10 µM of ATP was added to the cells to stimulate IL-1β secretion. The supernatants were recovered, and the amount of IL-1β was measured by ELISA (eBiosciences, 88-7013). NLRP3, pro-Caspase-1, Caspase-1 and pro-IL-1β protein levels were assessed on cell lysates by western blot analysis.
Retinal tissue preparation and immunofluorescence staining. Human and mouse eyes were fixed in 4% PFA and cryoprotected using 30% sucrose. They were embedded in optimal cutting temperature (OCT) compound (Leica, Wetzlar, Germany), frozen in liquid nitrogen, and stored at −80 °C. Frozen sections (10 μm thick) were cut in a cryostat (Leica CM 3050 S) and mounted on gelatin-coated slides for immunofluorescence analysis. For flat mounts, human and mouse eyes were fixed in 4% PFA for 15 min at room temperature and sectioned at the limbus; the anterior segments were discarded. The posterior eye cups consisting of neuroretina/RPE/ choroid/sclera complex were collected and the neuroretina was carefully detached from RPE/choroid/sclera to be prepared separately for experiments.
MP and retinal explant incubation. BMDM were respectively left untreated or co-stimulated with 300 ng/mL R-FSL1 and 20 ng/mL IFNγ for 24 h with or without MPE-001 (10 −7 M). To study the role of PPAR-γ in the cytoprotective effect of MPE-001 in apoptosis of photoreceptors induced by activated MPs, PPAR-γ inhibitor (10 −6 M) (GW9662) was added to BMDM treated with MPE-001 for 24 h. Culture media was removed and the BMDM were cultured with fresh DMEM/10% FBS for 24 h, and the supernatant was used as the conditioned media (CM). Neuroretina explants were generated from 12-16-week-old C57BL/6J mice eyes. The neuroretina was carefully detached from RPE/choroid/sclera to be incubated for 18 h with the CM or on DMEM with BMDM prepared as mentioned in the previous section. To reveal the role of IL-1β released in CM-induced photoreceptor apoptosis, the CM was incubated or not with 150 ng/mL of anti-IL-1β neutralizing antibody (Abcam, 9722) for 15 min at room temperature. After 18 h of stimulation, the explants were carefully removed, and the detection of apoptotic cells was performed using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. www.nature.com/scientificreports www.nature.com/scientificreports/ system Observer Z1 (Zeiss, Oberkochen, Germany). The retinal section analysis was performed exclusively in the central retina and peripheral retina was excluded. Isolated retinal mRNA was transcribed into cDNA for quantitative real-time PCR analysis. TUNEL assay. Neuro-retinal flat mounts were fixed in 4% PFA for 30 min, washed with PBS, post-fixed in frozen acetic acid for 30 min, and washed with PBS. Neuro-retinal flat mounts and retinal cryosections were permeabilized with 0.1% sodium citrate and 0.1% triton for 2 min on ice, and washed with PBS. Samples were incubated with the terminal deoxynucleotidyl transferase (TdT) and fluorescein-dUTP using the In-Situ Cell Death Detection Kit (Roche) for 60 min at 37 °C in a humid chamber in the dark. The reaction was stopped by washing the slides three times with PBS. Nuclei were counterstained with DAPI. Samples were analyzed using confocal microscopy (Olympus FluoView 1000, Richmond Hill, ON, Canada).

Quantification of activated
Electroretinography. Electroretinographs (ERGs) were recorded from WT and age-matched CD36 −/− mice on an Espion ERG Diagnosys apparatus equipped with a ColorDome Ganzfeld stimulator (Diagnosys LLC, Lowell, MA). Mice were dark adapted overnight and anesthetized intraperitoneally with an aqueous solution containing a mixture of ketamine (100 mg/kg) and xylazine (20 m/kg). Pupils were dilated using atropine and phenylephrine. A drop of methylcellulose was placed on the corneal surface to prevent corneal dehydration. Mouse body temperature was maintained at 37 °C using a heated water pad.
Flash scotopic ERGs were measured using corneal DTL Plus electrodes (Diagnosys LLC). The electrodes were placed on the surface of the cornea. A needle electrode on the forehead served as the reference electrode. Another needle grounding electrode was inserted into the tail skin. Scotopic responses were simultaneously stimulated from both eyes of the dark-adapted animals at the following increasing light intensities: 0.5, 1.0, 3.0 and 10.0 candela*second/meter² (cd*s/m²). Ten waveform responses were averaged with an inter-stimulus interval (ISI) of 5 seconds (for 0.5 cd*s/m²) or 20 seconds (for 1, 3 and 10 cd*s/m²). All procedures were performed in a dark room under dim red-light illumination. The amplitude and latency of the major ERG components were measured with the Espion software (Diagnosys LLC). ERG results were recorded at the optimal light intensity of 3.0 cd*s/ m². The ERG a-wave amplitudes were measured from the baseline to the primary negative peak and the b-wave amplitudes were measured from the trough of the a-wave to the maximum of the fourth positive peak.

Bioenergetics.
Polarized M0 or M1-BMDM treated or not with MPE-001, or peritoneal macrophages isolated from MPE-001-or NaCl-treated mice were seeded at 2.5 × 10 5 in Seahorse plates. MPE-001 was injected directly to the Seahorse plate on M1-BMDM. Real-time analysis of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were analyzed with an XF-24 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, Billerica, MA, USA). Uncoupled respiration (proton leak) represents fraction of total respiration that is insensitive to oligomycin treatment (1μM). Coupled respiration is calculated by substracting proton leak reaction from total respiration and represents reaction coupled to ATP synthesis. The contribution of non-mitochondrial respiration is also substracted in these experiments.
Metabolite quantification by GC-MS. Following treatments, MPs plated in 6-well plates were rinsed 3 times with 9 g/L NaCl solution (4 °C), quenched with 1.2 mL dry ice-cooled 80% MeOH, and stored at −80 °C. Samples were treated by sonication using the bioruptor (Diagenode, Denville, NJ, USA) for 10 min at the highest setting, with pulses and rests of 30 sec. Samples were cleared by centrifugation, and 800 ng D 27 -myristic acid in pyridine was added as an internal standard. Supernatants were dried up overnight in a vacuum pump concentrator (Labconco, Kansas City, MO, USA) set at −4 °C. Pellets were resuspended in 10 mg/mL methoxyamine hydrochloride/pyridine and subjected to methoximation for 30 min at 70 °C and silylation with N-tert-Butyldimethylsilyl-N-methyltrifluoroacetamide (MTBSTFA) + 1% t-BDMCS for 1 h at 70 °C. Internal standards and derivatization reagents were from Millipore Sigma (Oakville, ON, CA). Samples (1 µL) were injected in splitless mode in a 5975 C GC-MS configured with a DB-5MS + DG (30 m × 259 µm × 0.25 µm) capillary column (Agilent, Santa Clara, CA, USA). Inlet temperature was set to 280 °C and the carrier gas was helium. The flow rate was set to lock the internal standard elution at 17.94 min. The quadrupole was set at 150 °C and the GC-MS interface at 285 °C. The oven program started at 60 °C for 1 min, then temperature was raised by 10 °C/ min until 320 °C. Bake-out was at 320 °C (10 min). Data was acquired in scan mode and showed no saturation. All metabolites measured were validated using authentic standards (Sigma Millipore). Data analysis was done using the ChemStation software (Agilent). Relative metabolite concentrations were obtained by correcting the peak areas of quantifying ions with those of D 27 -myristic acid, and by dividing this ratio with the average protein content associated with paired BMDM plates, thus providing data with arbitrary units. Principal component analyses (PCA) were performed with MetaboAnalyst 4.0 79,80 using data obtained from 2 independent experiments, each conducted with 2-3 independent BMDM cultures. Data was uploaded as.csv files in MetaboAnalyst prior to the generation of PCA plots (containing 2 principal components).
Quantitative real-time PCR. RNA was run on a 2100 Bioanalyzer using a Nano RNA chip to verify its integrity. Total RNA was treated with DNase and reverse transcribed using the Maxima First Strand cDNA synthesis kit with ds DNase (Thermo Scientific). Gene expression was determined using assays designed with the Universal Probe Library from Roche (www.universalprobelibrary.com), and when no probe was available, a SYBR Green assay was designed. For all qPCR assay, a standard curve was performed to ensure that the efficacy of the assay is between 90% and 110%. For UPL assays, qPCR reactions were performed using Taqman Advanced master mix (Life Technologies), 2 µM of each primer and 1 µM of the corresponding UPL probe. For SYBR green assays, a melt curve was performed to ensure only a single product was amplified, and qPCR reactions were performed using Fast SYBR Green Master Mix (Wisent) and 2 µM of each primer. The Viia7 qPCR instrument