Myristoylated, Alanine-rich C-kinase Substrate (MARCKS) regulates Toll-like receptor 4 signaling in macrophages

MARCKS (Myristoylated Alanine-rich C-kinase Substrate) is a membrane protein expressed in many cell types, including macrophages. MARCKS is functionally implicated in cell adhesion, phagocytosis, and inflammation. LPS (lipopolysaccharide) triggers inflammation via TLR4 (Toll-like receptor 4). The presence of MARCKS and the formation of phospho-MARCKS in macrophages have been described, but the role(s) of MARCKS in regulating macrophage functions remain unclear. To investigate the role of MARCKS during inflammation, we activated macrophages using LPS with or without the addition of a PKC inhibitor. We found that PKC inhibition substantially decreased macrophage IL6 and TNF cytokine production. In addition, confocal microscopy revealed that MARCKS and phospho-MARCKS increased localization to endosomes and the Golgi apparatus upon LPS stimulation. CRISPR-CAS9 mediated knockout of MARCKS in macrophages downregulated TNF and IL6 production, suggesting a role for MARCKS in inflammatory responses. Our comprehensive proteomics analysis together with real-time metabolic assays comparing LPS-stimulation of WT and MARCKS knock-out macrophages provided insights into the involvement of MARCKS in specific biological processes and signaling pathways, uncovering specific proteins involved in regulating MARCKS activity upon LPS stimulation. MARCKS appears to be a key regulator of inflammation whose inhibition might be beneficial for therapeutic intervention in inflammatory related diseases.


Introduction
The myristoylated alanine-rich C kinase substrate (MARCKS) is a ubiquitous and highly conserved membrane protein which is present in many cell types including macrophages 1 .MARCKS is attached to the cell membrane via a myristoylated site at its N-terminus and a cationic AA rich effector domain (ED) 2,3 .The lysine rich ED contains four serine residues which are potential phosphorylation sites.
MARCKS was rst identi ed as a protein kinase C (PKC) substrate 4 .PKC phosphorylation or calmodulinbinding on the ED leads to the migration of MARCKS from the cell membrane to the cytosol followed by activation of several signal transduction pathways 5 .MARCKS returns to the cell membrane after dephosphorylation 3,6 .MARCKS is involved in a wide variety of functions such as brain development 7 , phagocytosis 8 , cell migration 9 , cell adhesion and in ammation 10,11 .
Macrophages are innate immune cells that play key roles in in ammation.Pathogen associated molecular patterns (PAMPs) are recognized by pattern recognition receptors (PRRs), which participate in the initiation of speci c immune responses 12 .One of the well-known PAMPs is lipopolysaccharide (LPS), a major gram-negative bacterial cell wall component, which is a potent activator of Toll-like receptor 4 (TLR4) of macrophages leading to in ammatory responses.The transcription level of MARCKS is increased in response to LPS stimulation 13 .LPS also induces MARCKS phosphorylation 14 .Moreover, LPS triggers secretion of proin ammatory cytokines such as TNF 15 and IL6 16 which contribute to the pathogenicity of many diseases including asthma 17 , rheumatoid arthritis and sepsis 18 .Elevation of TNF and IL6 in the serum are commonly found in those patients 18,19 .The functions of MARCKS in response to the LPS are still controversial.MARCKS was reported to block the effect of LPS stimulation in a study that used MARCKS de cient embryonic broblast cells 20 .In contrast, another group reported that MARCKS promotes proin ammatory cytokine expression in macrophages.Inhibition of MARCKS using the myristoylated N-terminal sequence (MANS) peptide suppresses pro-in ammatory cytokines and attenuates sepsis in a mouse model 21 .The discovery of the mechanism by which MARCKS contributes to the in ammatory response may provide new strategies to manipulate in ammation-related diseases.
Our laboratory has shown that MARCKS phosphorylation was upregulated during LPS stimulation using phosphoproteome pro ling of immortalized mouse macrophages (IMMs) 22 .In this work, we investigated the effects of MARCKS in the context of macrophage functions in response to LPS stimulation.We have used CRISPR CAS9 technique to create a MARCKS knockout in IMMs and characterized the knockout in comparison to wild type IMMs.We used mass-spectrometry-based proteomics to compare proteome pro les of wildtype and MARCKS de cient macrophages after LPS stimulation to provide a global view of MARCKS-dependent changes in the macrophage proteome.We found that while MARCKS promotes IL6 and TNF production, MARCKS de cient macrophages suppress the OXPHOS (oxidative phosphorylation) pathway to reduce the cytokine production and inhibit the pro-in ammatory functions.

MARCKS is upregulated during LPS stimulation
The functions of MARCKS protein in the context of LPS signaling in macrophages are not fully understood.To examine the relationship between MARCKS protein expression during LPS stimulation, we exposed wild type immortalized mouse macrophages (WT IMMs) to LPS for 6 hours.Then, MARCKS mRNA and protein levels were measured using real-time PCR and western blots.We found that the mRNA of MARCKS increased 5-fold after LPS stimulation (Fig. 1A).The protein expression level of MARCKS also increased after LPS exposure (Fig. 1B).
Mass spectrometry was used to quantitatively compare the proteomes of unstimulated and LPS stimulated WT IMMs.Comparison between LPS-treated and control revealed 270 proteins differentially expressed after LPS treatment (101 down-regulated and 169 up-regulated proteins).As expected, MARCKS was signi cantly (nearly 3-fold) increased in macrophages exposed to LPS compared to unstimulated control (Fig. 1C).Next, we performed gene ontology enrichment analysis (GO terms) for the up-and down-regulated proteins.The top enriched biological processes highlight GO terms that re ect macrophage functions such as innate immune response, in ammatory response, and cytokine production (Fig. 1D).We further analyzed cellular components, pathways and molecular functions (Fig. 1E, F, G).KEGG pathway analysis indicated enrichment in pathways associated with Toll-like receptor signaling (Fig. 1H).Our data suggest that MARCKS might play a regulatory role in the LPS signaling in macrophages since we found both its mRNA and protein levels change following LPS stimulation.

MARCKS increases its colocalization with the endosome post LPS stimulation
Intracellular translocation of TLR4 prevents the constitutive response of cells that are exposed to bacteria and their products.Since both MARCKS and the TLR4 regulator TRAM, that associates with TLR4 and is essential for TLR 4 signal transduction 23 contain a myristoyl tail, we wanted to investigate whether MARCKS localizes to the same cellular compartments as TLR4 post LPS stimulation in macrophages.We exposed WT IMMs to LPS for 0 15, 30 and 60 minutes.Endogenous localization of MARCKS and EEA1, a marker for early endosomes, were determined by immuno uorescence microscopy.With no stimulation, MARCKS is understood to be tethered at the membrane by its myristoyl tail, and its effector domain is supposed to interact with the membrane.MARCKS was located at the cell membrane prior to LPS stimulation.After LPS stimulation for 30 minutes, MARCKS colocalized with the endosome (Fig. 2A-F) which is in agreement with previous reports 20 .This colocalization increased post LPS stimulation, and returned to basal level 60 min after stimulation (Fig. 2G).

MARCKS colocalizes with TLR4 at early time points post LPS stimulation
Since MARCKS colocalizes with the endosome post LPS stimulation and it is known that TLR4 is translocating from the membrane to the endosome post LPS stimulation, we wanted to investigate whether MARCKS and TLR4 colocalized and for what amount of time.To answer this question, WT IMMs were stimulated with LPS for 15, 30, or 60 minutes or left unstimulated, and colocalization of endogenous TLR4 and MARCKS was determined with confocal imaging.Based on the Pearson's coe cient, TLR4 and MARCKS colocalized in the absence of LPS but was disrupted upon stimulation (Supplementary Fig. 1A).However, the amount of MARCKS that colocalized with TLR4 didn't change signi cantly during our experimental time, whereas the amount of TLR4 that colocalized with MARCKS decreased during the 60 min interval of the experiment (Supplementary Fig. 1B).Since MARCKS colocalized with TLR4 post LPS stimulation, our results suggest that MARCKS might be physiologically relevant during LPS-activation of TLR4 signaling pathway.

Phospho-MARCKS localizes to the Golgi post LPS treatment
Numerous studies have shown that upon LPS stimulation, the effector domain of MARCKS is phosphorylated 24,25 , and it is thought that phosphorylation elicits a switch from MARCKS interacting with the plasma membrane to interacting with actin in the cytoplasm.MARCKS phosphorylation on sites Ser152, 156 and 163, or any combinations of these phosphorylated residues 22 could be contributing to this activity of MARCKS.To investigate the localization of endogenous phospho-MARCKS Ser163, we labeled xed IMMs with antibodies against phospho-MARCKS Ser163 conjugated to uorescent secondary antibodies and performed immuno uorescence microscopy.The intensity of phospho-MARCKS in confocal images was very low without LPS treatment but it increased dramatically after LPS treatment.Furthermore, after LPS treatment, the colocalization of phospho-MARCKS with Golgin-97, a trans-Golgi network marker, increased dramatically (Fig. 3).

MARCKS expression is TLR4-dependent
We inquired whether the upregulation of MARCKS post LPS stimulation is dependent on TLR4 signaling.
To address this question, we used TLR4 knockout (ΔTLR4) IMMs and WT IMMs and stimulated them with LPS for 0, 1, 6, and 24 hours.ΔTLR4 IMMs failed to increase the TNF mRNA level whereas the WT IMMs showed nearly 20-fold increase in TNF mRNA level after LPS stimulation (Fig. 4A).The level of MARCKS mRNA in ΔTLR4 IMMs was slightly increased but not statistically signi cant whereas it was signi cantly increased (~ 6 fold) in WT IMMs treated with LPS (Fig. 4B).Of note, MARCKS protein level did not change after LPS stimulation in TLR4 knockout IMMs (Fig. 4C, D).This is in contrast to the LPSinduced increase in MARCKS protein observed in WT IMMs (Fig. 1B), suggesting that MARCKS mRNA expression and protein abundance are TLR4 dependent.

MARCKS promotes proin ammatory cytokine responses
Macrophages play a pivotal role in the in ammatory response, but the link between MARCKS and macrophage-mediated in ammation remains to be elucidated.During LPS treatment, MARCKS is thought to be phosphorylated by PKC.To examine whether MARCKS has an effect on the in ammatory responses, we pre-incubated IMMs with rottlerin, a PKC inhibitor, for one hour before stimulation with LPS and compared them to cells that did not receive rottlerin.We found that PKC inhibition attenuated the proin ammatory cytokine secretion of both TNF and IL6 (Fig. 5A, B).Next, we used CRISPR CAS9 mediated gene knockout to generate MARCKS knockout in IMMs (ΔMARCKS IMMs).The schematic diagram for generating the knockout cells is shown in Fig. 5C.The level of MARCKS protein expression was con rmed using western blot, which showed nearly no MARCKS protein in the ΔMARCKS IMMs when compared to WT IMMs (Fig. 5D, E).To further con rm the knockout was successful, we used mass spectrometry to compare the levels of MARCKS protein between WT IMMs and ΔMARCKS IMMs.The results showed that there is no protein detected in ΔMARCKS IMMs while WT IMMs showed 3.6X10 7 protein abundance (Fig. 5F).Of note, both WT and ΔMARCKS IMMs had comparable levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a housekeeping protein we used as a control (Fig. 5G).Moreover, looking at the tryptic peptides identi ed by mass spectrometry, we detected no MARCKS peptides in ΔMARCKS IMMs whereas in WT IMMs we identi ed 6 peptides from MARCKS.
(Supplementary table 1).These data con rmed that we successfully created MARCKS de cient macrophages.
To con rm the role of MARCKS in the pro-in ammatory cytokine responses, we exposed WT and ΔMARCKS IMMs to LPS, then and collected the conditioned media for measurements of cytokine production.We found that ΔMARCKS IMMs downregulated both TNF and IL6 levels.Our data suggest that MARCKS knockout suppressed LPS-induced IL6 and TNF expression (Fig. 5H, I).To further con rm the role of MARCKS in the cytokine production, we cloned MARCKS sequence to pLenti-C-mGFP-P2A-Puro, a lentiviral vector with C-terminal mGFP tag and P2A-Puromycin and then we transduced it into ΔMARCKS IMMs to generate stable cells (MARCKS knock-in IMMs).In the stable cell line, we activated MARCKS knock-in IMMs with LPS and compared them to WT and ΔMARCKS IMMs.MARCKS knock-in IMMs rescued the levels of TNF cytokine production (Fig. 5J).These data are in agreement with the PKC inhibitor and LPS stimulated MARCKS knockout macrophage data.The results further suggest that MARCKS promotes TNF and IL6 production in macrophages post LPS stimulation.

MARCKS de cient macrophages show distinct proteomics pro le during LPS stimulation
Because MARCKS is also associated with a wide variety of functions in cells such as signal transduction, cell migration, cell proliferation, cell differentiation and cytokine production, we hypothesized that phenotypic changes in ΔMARCKS macrophages may be associated with other proteins that regulate macrophage functions.To test this hypothesis, we stimulated WT and ΔMARCKS IMMs with the LPS for 6 hours and conducted a proteomic analysis using label free quanti cation.We found 94 up-regulated proteins and 60 down-regulated proteins in ΔMARCKS after LPS treatment compared to WT IMMs after LPS treatment (Fig. 6A).The identi ed proteins were primarily associated with biological processes involved in the macrophage functions such as "Positive regulation of response to cytokine stimulus", "Regulation of response to the cytokine stimulus" and "Innate immune response".The enriched cellular components represented "Polysomal ribosome" and "Mitochondrion".The molecular functions included "ATP gate channel activity", "Electron transfer activity" and "Cytochrome C oxidase activity".Taken together, these data suggested that mitochondria might be involved in the phenotypic changes in IMMs upon LPS stimulation in the absence of MARCKS.Since we found several GO terms of the top 15 cellular components and molecular functions were associated with mitochondrial proteins (Fig. 6B), we performed KEGG pathway enrichment analysis of the signi cant proteins in ΔMARCKS IMMs with LPS treatment compared to WT IMMs.Signi cantly enriched pathways included "Oxidative phosphorylation" and other metabolic pathways including "Metabolism of xenobiotics by cytochrome P450" and "Cytokinecytokine receptor interaction" (Fig. 6C).We created a heatmap of the protein abundance in individual samples based on ontology enrichment analysis of the signi cant proteins focusing on innate defense response-related proteins and mitochondria related proteins.We found that the protein expression patterns in the individual samples of WT and ΔMARCKS IMMs after LPS treatment were clearly distinct (Fig. 6D), with many proteins showing an increase in KO cells.Our proteomics data suggest that the classic mitochondria function such as oxidative phosphorylation (OXPHOS) might be altered during LPS signaling in ΔMARCKS IMMs.

MARCKS promotes oxidative phosphorylation
Mitochondrial OXPHOS is the electron transfer process through the main 5 protein complexes in the inner mitochondrial membrane resulting in the production of adenosine triphosphate (ATP), the main source of energy in eukaryotic cells 26 .The schematic diagram for the OXPHOS and the oxygen consumption rate over time is shown in Fig. 7A, B. To directly test if MARCKS in uences OXPHOS, we used extracellular ux analysis to measure OXPHOS during LPS treatment of WT and ΔMARCKS IMMs compared to unstimulated control.In the untreated control group, we found that ΔMARCKS IMMs had lower OXPHOS when compared to the WT IMMs.After 6 h LPS treatment, the mitochondrial respiration was downregulated in both WT and ΔMARCKS IMMs.(Fig. 7C).We further analyzed the other parameters obtained from extracellular ux analysis.We did not observe any signi cant difference in basal respiration between unstimulated WT and ΔMARCKS IMMs (Fig. 7D), but when the cells were stimulated with LPS, basal respiration statistically decreased in ΔMARCKS cells compared to the WT IMMs.Maximal respiration and spare respiration were signi cantly decreased in ΔMARCKS IMMs when compared to WT IMMs.In the LPS treatment group, LPS reduced the maximal respiration in both cell types but that of ΔMARCKS IMMs was still lower compared to WT IMMs (Fig. 7E, F) Mitochondria play a crucial role in ATP production, and mitochondrial dysfunction may result in insu cient energy supply for the cells.Moreover, ATP is required during the immune response 27,28 .We aimed to determine whether ATP production is defective in ΔMARCKS IMMs, resulting in lower proin ammatory cytokine production.To address this, we examined the ATP production in OXPHOS during mitochondrial respiration.We found that ATP production in ΔMARCKS IMMs was reduced when compared to WT IMMs.ATP production was signi cantly lower after LPS treatment (Fig. 7G).In addition, LPS-treated-ΔMARCKS IMMs showed statistically signi cant lower ATP production than the WT IMMs.Our data suggest that MARCKS is involved in mitochondrial respiration, resulting in su cient energy production for macrophage in ammatory functions such as cytokine production.

Discussion
The macrophage response to the LPS resulting in the cytokine production involves many elements of regulation such as downstream signaling transduction, transcriptional regulation, regulation by adaptor proteins 29 , and subcellular localization.Our results showed that both MARCKS mRNA and protein levels (as detected by both western blots and proteomics) were upregulated after LPS stimulation.In addition, LPS induced changes of the macrophage proteome, and MARCKS was among the upregulated proteins.
A previous report showed that MARCKS transcription level is highly elevated after LPS stimulation 13 , but the mechanisms of action remained unclear.Our ndings provide evidence that MARCKS plays a key role during LPS stimulation in macrophages and signi cantly contributes to the in ammatory response.
TLR4 is translocated to the endosome after LPS stimulation 30 and many pieces of indirect evidence suggest that MARCKS is physically connected to the TLR4 signaling pathway.Considering the wellcharacterized function of the TIR-domain containing adaptor protein TRAM, which is a transmembrane protein containing the myristoylated site and PKC phosphorylation site 31 , which are both important for innate immune signaling, and the colocalization of TRAM with TLR4 in the plasma membrane and Golgi apparatus where the TLR4 signal transduction takes place 23,32 , we sought to nd out whether translocation of MARCKS protein from the cell membrane to cytoplasm may have a similar effect.We observed that MARCKS co-localized with TLR4 and endosome at the early time points following LPS stimulation.In addition, phosphoS163-MARCKS also co-localized with the Golgi apparatus which is the same cellular compartment where TLR4 translocates eventually after LPS stimulation.Colocalization of MARCKS and endosome agrees with the previous report by Mancek-Keber et al. 20 However, in that study, the cytokine production after LPS stimulation was attenuated by MARCKS and MARCKS effector peptide in cultured cell lines and mouse embryonic broblasts 20 .This result was surprising and is in contrast to two more recent studies which showed that MARCKS enhanced proin ammatory cytokine expression by regulation of p38/JNK and NF-kappaB, an effect which was further inhibited by MARCKS inhibitory peptide (MANS) in neutrophils 33 and macrophages and a mouse model 10 .Additionally, inhibition of MARCKS using MANS peptide resulted in a reduction of proin ammatory cytokines such as CXCL1, IL-1β, IL6, MCP-1 and TNF in bronchoalveolar lavage uid in a neutrophil elastase-induced murine bronchitis model 34 We sought to address this controversy, and our results using IMMs demonstrated that the increased expression of MARCKS protein is functionally associated with elevated pro-in ammatory cytokine production and that MARCKS de cient macrophages had signi cantly reduced levels of secreted IL6 and TNF.Furthermore, inhibiting MARCKS activation with a PKC inhibitor reduced cytokine production.Therefore, our results strongly favor the hypothesis that MARCKS is indeed a positive regulator of in ammatory function in macrophages.
MARCKS knockout mice are not well-characterized because MARCKS de ciency in mice resulted in abnormal brain development and the pups die within several hours after birth 7 .Therefore, we used instead the CRISPR CAS9 technique 35 for the rst time to generate MARCKS knockout macrophages and con rm that cytokine production is altered during LPS stimulation.These results further con rmed that MARCKS promotes IL6 and TNF expression during LPS stimulation in macrophages.Because in ammatory cytokine secretion by macrophages is a major cause of pathogenesis and progression of the in ammatory related diseases, a better understanding of the regulatory mechanisms of cytokine production would be bene cial for the identi cation of therapeutic targets.Because MARCKS is expressed in many innate immune cells including macrophages 13 , neutrophils 36,37 and monocytes 38 , which are the key player cells for the in ammatory responses, inhibition of MARCKS should lead to a reduction of the proin ammatory cytokines produced by many innate immune cell types.
In addition, we report for the rst time the global changes in proteome pro les of MARCKS de cient macrophages compared to WT IMMs.Interferon-induced protein with tetratrico-peptide repeats-1(IFIT1) and interferon-induced protein with tetratrico-peptide repeats-3 (IFIT3), both of which play an important role during viral infection, are upregulated in ΔMARCKS IMMs.IFITs play an important role during viral infection 39 .In contrast, IFIT1 and IFIT3 have been identi ed as negative regulators of LPS-induced TNF in human macrophages 40 .Upregulation of IFIT1 and IFIT3 in our proteomic data may be the cause of the lower cytokine production phenotype in ΔMARCKS IMMs.Gene ontology analysis and KEGG pathway enrichment analyses of the up-and down-regulated proteins pointed to the oxidative phosphorylation.Oxidative phosphorylation occurs mainly in the mitochondrial inner membrane.Glycolysis and mitochondria are the main sources of energy production during in ammatory responses 41,42 .Cellular energy production changes, and ATP dependent cytokine production involved in in ammation have been described 43,44 .Our data indicate that ΔMARCKS IMMs have decreased mitochondrial respiration compared to the WT IMMs.A previous study showed that more mitochondria or an increase in glycolytic respiration resulted in higher ATP production which supported macrophage cytokine production 45 .
Moreover, mitochondrial respiration is lower in the LPS-tolerant macrophages 46 .Our results point to the correlation between MARCKS, mitochondrial respiration and cytokine production.
In conclusion, MARCKS promotes proin ammatory cytokine production during LPS stimulation in part by increasing the cell's ATP production.The hypothesis we propose is shown in Fig. 8.The direct interactions between MARCKS and other in ammatory pathway proteins remain unknown, and the precise mechanisms of action need further investigation.MARCKS appears to be an emerging therapeutic target in in ammatory diseases.

Cell line
Immortalized mouse macrophages (IMM) (a generous gift from Dr. Eicke Latz 47,48 ) were cultured in complete DMEM (Gibco, Grand Island, NY, USA) complemented with 10% FBS (Gibco), in a 5% CO 2 incubator at 37 o C, maintained at low densities and passaged until reaching the con uent state, usually every 3-4 days on sterile tissue culture plates.

Real-time PCR
The RNeasy kit (QIAGEN, Hilden, Germany) was used for isolation of total RNA from cell culture.Total RNA (1ug) was converted to cDNA using MultiScribe reverse transcriptase (Thermo Scienti c, Rockford, IL, USA), and real-time PCR was performed using Power SYBR Green PCR Master Mix with speci c primer to determine gene expression level.The primers are listed in supplementary Table 2.The ΔΔCt method was used for determining relative gene expression and beta-actin was used as a housekeeping gene.

Stable expression by lentiviral transduction
HEK293FT cells (Thermo Scienti c) were seeded on 6 wells cell culture plates (250,000 cells/well) overnight prior to transfection.pLenti-C-mGFP with MARCKS sequence were transfected to HEK293FT cells via TranIT-TKO (MirusBio, Madison, WI, USA).Conditioned media containing lentiviral particles were harvested 24h and 48h post-transfection and ltered using 0.45 µM lters (Millipore, Carrigtwohill, Ireland) before being used to infect IMMs cells.Cells were monitored under uorescence microscopy.Stable cell transfectants were selected using puromycin (5 µg/mL) (Gibco) containing media.Single cells were isolated and cultured for 10-14 days to obtain homogenous clonal populations.MARCKS expression levels were con rmed using western blots.

Western blot analysis
IMMs were lysed in Pierce RIPA buffer (Thermo Fisher Scienti c) supplemented with Halt protease inhibitor (Thermo Fisher Scienti c) and phosphatase inhibitor cocktails (Thermo Fisher Scienti c).Total protein concentration was determined by BCA protein assay (Thermo Fisher Scienti c).Samples (20 µg total protein) were loaded into NuPAGE 4 to 12% Bis-Tris polyacrylamide gels (Invitrogen, Carlsbad, CA, USA) and run at 200 V for 1 h.The proteins were transferred to the PVDF membrane (Thermo Fisher Scienti c).The membranes were blocked in 5% BSA overnight and then incubated with speci c primary antibodies for one hour followed by incubation with secondary antibodies for one hour.The antibodies are listed in supplementary Table 3.The blots were developed using ECL substrate (Thermo Fisher Scienti c) The results were visualized using a ChemiDoc imaging system (Bio-Rad, Hercules, CA, USA).
The densitometric analysis of western blot results was performed using Image J. ELISA TNF and IL6 were quanti ed using ELISA kits (Invitrogen, Vienna, Austria) kit according to the manufacturer's instructions.Brie y, IMMs were treated with 100 ng/mL LPS (LPS from Salmonella minnesota R595, Enzo Life Sciences) for 6 hours in quadruplicate and the supernatants were collected at the indicated time points.Capture antibody was applied to 96 well plates overnight followed by blocking with 1x diluent buffer for one hour at room temperature, then supernatants with appropriate dilution factor were put in each well for two hours at room temperature.Detection antibody and streptavidin-HRP were sequentially added to the assay plate and incubated one hour and 30 minutes respectively.ELISA washing steps were performed 4 times with 0.05% tween 20 (Thermo Fisher Scienti c) in PBS (PBS-T).ELISAs were developed with TMB substrate for 10 mins followed by a stop solution.Absorbances were determined using a microplate reader.The results were interpreted in comparison to the standard curve.

Seahorse Assay
IMMs in different experimental groups (control vs LPS 6 hours) were dispersed into monolayers for the measurement.Mitochondrial stress tests were performed at 37°C using the Seahorse XFe96 bioanalyzer (Seahorse Bioscience).IMMs were seeded at 2× 10 5 cells per well on the Seahorse analysis plates.Oxygen consumption rates (OCR) and extracellular acidi cation rates (ECAR) for the mitochondria were measured in XF media (containing 25 mM glucose, 2 mM L-glutamine, and 1 mM sodium pyruvate) under basal conditions and in response to 1.5 µM oligomycin, 1 µM uoro-carbonyl cyanide phenylhydrazone (FCCP), and 0.5 µM rotenone and antimycin A. All of the values were normalized with total protein using Wave software (Agilent).

Mass spectrometry
Sample preparation.WT and MARCKS knockout IMMs were seeded on a 10 cm dish and cultured overnight, the cells were unstimulated or stimulated with 100 ng/mL LPS for 6 hours.The indicated samples were lysed using RIPA buffer (Thermo Fisher Scienti c).For each sample, 500 ug of total protein mass was run on Bis-Tris NuPAGE gel (Invitrogen) as described above.The gels were xed using xing solution (47.5% methanol and 5% glacial acetic acid) for 30 minutes at room temperature.The xed gels were stained using PageBlue® protein staining solution (Thermo Fisher Scienti c) for one hour at room temperature, and then destained overnight with ddH 2 O at 4°C.Each lane was cut into 1 mm 3 pieces using razor blades and the gel pieces were put in 1.5 mL Eppendorf tubes and processed using in-gel digestion according to the published protocol and summarized below 49 .
In-gel protein digestion: Brie y, 500 µL of acetonitrile (ACN) were added to the gel pieces and incubated for 10 minutes at room temperature before removing all the supernatant from the tube.For reduction of disul de bond, 50 µL of DTT in 100 mM ammonium bicarbonate (ABC) was added to the tube and incubated at 56 ° C for 30 minutes.Then, 50 µL of 55 mM of 2-chloroacetamide (CA) in 100 mM ABC solution were added and incubated for 20 minutes at room temperature in the dark.Then, trypsin solution (Promega, Madison, WI, USA) (1:25 w:w ratio) was added and the samples were incubated at 37°C for 17 hours.Following incubation, peptides were cleaned and desalted using C18 ZipTip tips (Millipore) according to the manufacturer's instruction.Mass Spectrometry.An Orbitrap Fusion Eclipse with an EASY-Spray ion source (Thermo Fisher Scienti c, San Jose, USA) coupled to a Thermo UltiMate 3000 (Thermo Fisher Scienti c) was used for LC-MS/MS experiments. 1 ug of total peptides were injected for LC-MS/MS analysis.Peptides were trapped on an Acclaim C18 PepMap 100 trap column (5 µm particles, 100 Å pores, 300 µm i.d.x 5 mm, Thermo Fisher Scienti c) and separated on a PepMap RSLC C18 column (2 µm particles, 100 Å pores, 75 µm i.d.x 50 cm, Thermo Fisher Scienti c) at 40 C. The LC steps were: 98% mobile phase A (0.1% v/v formic acid in H 2 O) and 2% mobile phase B (0.1% v/v formic acid in ACN) from 0 to 5 min, 2-35% linear gradient of mobile phase B from 5 to 155 min, 35-85% linear gradient of mobile phase B from 155 to 157 min, 85% mobile phase B from 157 to 170 min, 85-2% linear gradient of mobile phase B from 170 to at 172 min, 2% of mobile phase B from 172 to 190 min.Eluted peptides were ionized in positive ion polarity at a 2.1 kV of spraying voltage.MS 1 full scans were recorded in the range of m/z 375 to 1,500 with a resolution of 120,000 at 200 m/z using the Orbitrap mass analyzer.Automatic gain control and maximum injection time were set to standard and auto, respectively.Top 3sec data dependent acquisition mode was used to maximize the number of MS 2 spectra from each duty cycle.Higher-energy collision-induced dissociation (HCD) was used to fragment selected precursor ions with normalized collision energy of 27.MS 2 scans were recorded using an automatic scan range with a resolution of 15,000 at 200 m/z using the Orbitrap mass analyzer.Data analysis, label free quanti cation and statistical analysis were performed using Proteome Discoverer 2.5 (Thermo Fisher Scienti c).Brie y, the raw les were searched against the mouse uniport database with the list of common protein contaminants.The groups and conditions were speci ed to obtain the quanti cation ratios and adjusted p-values were calculated using Benjamini-Hochberg method.The data visualization was performed using R packages.The differentially expressed proteins were identi ed using log2 fold change less than − 0.5 (0.7-fold change) for downregulated proteins and log2 fold change more than 0.5 (1.4-fold change) for upregulated proteins combined with the adjusted p-value less than 0.05.Volcano plots were generated by ggplot2.All the signi cant proteins were used as input to identify the signi cant enriched Go and pathway analysis.KEGG pathway analyses were search against mouse KEGG genome database and the bubble chart was generated by path ndR.
The adjusted p-value were calculated using Bonferroni method.The heatmap was generated by pheatmap.Go enrichment analysis was performed using shiny v. 0.77 by searching against the mouse string database with the FDR cutoff 0.05.The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium 50 via the PRIDE 51 partner repository with the dataset identi er PXD042097 (temporary reviewer account details: Username: reviewer_pxd042097@ebi.ac.ukPassword: cyYoee0O ).

Confocal microscopy
IMMs were seeded in 24 well dishes with an inserted glass coverslip at 2x10 4 cells per well and incubated overnight to recover.The cells were then treated with 1 mg/mL LPS for 5, 10, 15, 20, 30, 45, 60 minutes or left untreated for the phospho-MARCKS colocalization study, 15, 30, 60 minutes and untreated for the endosome colocalization study and 1, 5 minutes and untreated for Golgi colocalization study.After LPS treatment, the cells were washed with ice cold PBS three times and xed with 2% paraformaldehyde for 20 minutes at room temperature.The coverslips were washed three times with PBS-T, the primary antibody was added in PBS-T with 0.1% BSA and incubated overnight at 4 o C.After extensive washing with PBS-T, the secondary antibody was added and incubated in the dark on the coverslips for 1 hour at room temperature.The coverslips were washed once with PBST, Hoechst dye was added (1:15000) and incubated in the dark for 15min after which the coverslips were again extensively washed with PBST.The coverslips were mounted on slides using ProLong Gold (Life Technologies, Grand Island, NY) and kept in the dark at 4 o C until visualization.The visualization of the image was performed using confocal Imaging was performed using a Leica SP8 confocal microscope and data quanti cation including colocalization analysis was performed using Imaris software.B) IMMs were stimulated with 100 ng/mL of LPS for 0, 1, 6 24 h.MARCKS protein expression was detected by western blot, and beta-actin was used as a loading control.The full-length blot image is included in the Supplementary Information le, Supplementary Figure 3.

Figures
Figures

C
) Densitometric analysis of western blot results.MARCKS levels were normalized to beta-actin.Data are representative of 2 independent experiments, shown as mean ± SEM. * p< 0.05 (one-way ANOVA).D) Volcano plot of the differential protein expression within WT IMMs after 6 h LPS stimulation vs unstimulated WT IMMs.Green dots represent up-regulated proteins, red dots represent down-regulated proteins and black dot represents MARCKS.E-G) Go enrichment analysis performed using shinyGO v. 0.77 to characterize the biological functions: D) Biological Processes E) Cellular Components F) Molecular functions of these differentially expressed proteins.H) KEGG pathway analysis of LPS-treated IMMs vs unstimulated IMMs.The x axis represents the fold enrichment value while y axis represents the enriched pathways.The bubble size shows the number of differentially expressed proteins in the indicated pathway.The color intensity indicates the -log10(lowestp) value, the darker the red, the more signi cant pathway enrichment.