The transcription factors TFEB and TFE3 link the FLCN-AMPK signaling axis to innate immune response and pathogen resistance

TFEB and TFE3 are transcriptional regulators of the innate immune response, but the mechanisms regulating their activation upon pathogen infection are poorly elucidated. Using C. elegans and mammalian models, we report that the master metabolic modulator 5’-AMP-activated protein kinase (AMPK) and its negative regulator Folliculin (FLCN) act upstream of TFEB/TFE3 in the innate immune response, independently of the mTORC1 signaling pathway. In nematodes, loss of FLCN or overexpression of AMPK conferred pathogen resistance via activation of TFEB/TFE3-dependent antimicrobial genes, while ablation of total AMPK activity abolished this phenotype. Similarly, in mammalian cells, loss of FLCN or pharmacological activation of AMPK induced TFEB/TFE3-dependent pro-inflammatory cytokine expression. Importantly, a rapid reduction in cellular ATP levels in murine macrophages was observed upon lipopolysaccharide (LPS) treatment accompanied by an acute AMPK activation and TFEB nuclear localization. These results uncover an ancient, highly conserved and pharmacologically actionable mechanism coupling energy status with innate immunity.


Introduction
Innate immune responses constitute the first line of defense against pathogenic infections in simple metazoans, invertebrates, and mammals [1][2][3][4]. While much effort has been put into elucidating the functions of downstream mediators of immune response including antimicrobial peptides, C-type lectins, cytokines and chemokines, less is known regarding how host cells recognize foreign infections and trigger the activation of transcription factors that coordinate the anti-microbial response. Among the few well-characterized transcription factors, NF-κB, was shown to be an important factor in controlling host defense gene expression, mediated through toll like receptor (TLR) and nucleotide-binding leucine-rich repeat containing (NLR) ligand pathways [5]. However, another under-appreciated host-defense transcription factor was recently identified in Caenorhabditis elegans (C. elegans), which lacks the NF-κB pathway [6]. Using this model, HLH-30, the C. elegans ortholog of TFEB and TFE3, was identified as an important evolutionarily conserved transcriptional regulator of the host response to infection [6][7][8]. TFEB and TFE3 are basic helix-loop-helix leucine zipper transcription factors that multi-task in regulating a similar set of genes involved in lipid metabolism, autophagy, lysosomal biogenesis and stress response genes [9][10][11][12][13]. Several studies have reported a similar mechanisms underlying TFEB/TFE3 activation in response to nutrient deprivation and metabolic stress. In nutrient-rich environments, the kinases ERK2 and mTORC1 phosphorylate TFEB/TFE3 on specific serine residues and retain them in the cytoplasm in an inactive state [13][14][15][16][17]. The mTORC1-dependent phosphorylation of TFEB (S211) and TFE3 (S321) promotes binding to 14-3-3. It has been suggested that this interaction masks the Nuclear Localization Signal (NLS), thus inhibiting TFEB and TFE3 nuclear translocation [14,16]. Conversely, under starvation, this repressive phosphorylation is lifted, resulting in their translocation to the nucleus and activation of their downstream transcriptional targets that encode components of the lysosomal biogenesis and autophagy pathways [13,14,16,17]. Despite these remarkable similarities between TFEB and TFE3, it is still unclear whether these transcription factors have cooperative, complementary, or partially redundant roles under different environmental conditions. Importantly, in murine macrophages, both TFEB and TFE3 were shown to be activated and translocated to the nucleus upon pathogen infection or stimulation with TLR ligands, where they collaborate in mediating the transcriptional upregulation of several cytokines and chemokines involved in antimicrobial immune response [6,18,19]. This functional conservation of the TFEB/TFE3 pathway is further supported by a recent study showing that bacterial membrane pore-forming toxin induces cellular autophagy in an HLH-30-dependent manner in C. elegans [20]. However, the mechanisms by which nematode and mammalian TFEB/TFE3 are activated during infection are still poorly understood. Recently, TFEB activation was found to involve phospholipase C and protein kinase D pathways both in C. elegans and mammals upon pathogen infection [21]. Subsequent studies showed that lipopolysaccharide (LPS)-stimulated TFEB/TFE3 activation in murine macrophages induced cytokine production and secretion independent of mTORC1, but the specific pathway by which their activation was mediated was not elucidated [18]. Folliculin (FLCN) is a binding partner and negative regulator of 5'-AMP-activated protein kinase (AMPK) [22,23], which was identified as a tumor suppressor protein responsible for the Birt-Hogg-Dubé (BHD) neoplastic syndrome in humans [24]. Importantly, the interaction of FLCN with AMPK is mediated by two homologous FLCN-binding proteins FNIP1 and 2 [22,23]. Pathogenic mutations from BHD patients lead to a loss of FNIP/AMPK binding pointing to the functional significance of this interaction in tumor suppression [22]. AMPK is a heterotrimeric enzyme, which monitors the energy status and maintains energy homeostasis under metabolic stress by activating catabolic processes and inhibiting anabolic pathways [25][26][27]. We have previously shown that loss of FLCN or expression of a FLCN mutant unable to bind FNIP/AMPK led to chronic AMPK activation, resulting in increased ATP levels through an elevated glycolytic flux, oxidative phosphorylation and autophagy [28][29][30]. Importantly, we have shown that loss of FLCN mediates resistance to oxidative stress, heat, anoxia, obesity, and hyperosmotic stresses via AMPK activation in C. elegans and mammalian models [28][29][30][31].
While a role for FLCN in regulating immune responses has not been reported, the functional role for AMPK in innate immunity seems to be context and cell-type dependent [32,33]. In this study, we demonstrate a novel evolutionarily conserved pathogen resistance mechanism mediated by FLCN and AMPK via TFEB/TFE3. Specifically, we show that loss of flcn-1 in C. elegans, which leads to chronic AMPK activation, enhances the HLH-30 nuclear translocation and induces the expression of hlh-30-dependent antimicrobial genes upon infection, mediating resistance to bacterial pathogens. Using RNA-seq, we show that many hlh-30dependent antimicrobial genes are regulated by AMPK upon S. aureus infection. AMPK loss reduces HLH-30 nuclear translocation and abrogates the increased resistance of flcn-1(ok975) mutant animals to pathogens. Furthermore, we show that constitutive activation of AMPK C. elegans nematodes leads to an HLH-30-dependent increase in pathogen resistance, similar to what we observe upon loss of flcn-1. Importantly, we show that this pathway of regulation is evolutionarily conserved and that FLCN and AMPK regulate TFEB/TFE3-driven cytokine and inflammatory genes in mouse embryonic fibroblasts and macrophages. Overall our data suggest an essential role of the FLCN/AMPK axis in the regulation of host-defense response via TFEB/TFE3, highlighting a possible mechanism likely to contribute to tumor formation in BHD patients. Our findings also shed new light on the potential use of AMPK activators in the stimulation of the innate immune response and defense against pathogens.

Loss of flcn-1 in C. elegans increases the expression of anti-microbial genes and confers resistance to bacterial pathogens
To understand the physiological role of FLCN-1, we compared gene expression profiles of wild-type and flcn-1(ok975) mutant animals. Among differentially expressed genes, 243 transcripts were up-regulated in flcn-1(ok975) mutant animals compared to wild-type animals at basal level (Table S1) and were classified based on their biological functions (Table 1 and Tables   S2 and S3). Genes associated with stress response, innate immune response, defense mechanisms and response to stimulus processes, including heat shock proteins, C-type lectins, lysozymes and cytochrome P450 genes, were induced in flcn-1(ok975) unstressed mutant animals compared to wild-type animals (Table 1, S1-3, Figure 1A). Selected genes were validated using RT-qPCR ( Figure 1B and Table S3). On the other hand, 704 genes were shown to be downregulated in flcn-1(ok975) mutant animals (Table S4) and are involved in various processes that control proliferation and growth (Table S5). These results indicate that a differential gene expression might be providing advantage to the flcn-1 mutant worms prior to stress or pathogen attacks. This is in accordance with our previously reported results where loss of flcn-1(ok975) conferred resistance to oxidative stress, heat stress, anoxia and hyperosmotic stress in C. elegans [28][29][30][31]34]. Since it was demonstrated that the osmo-sensitive gene expression mimics the transcriptional profiles of pathogen infection [35], we compared the overlap between genes upregulated in flcn-1(ok975) mutant animals and genes induced by infection of C. elegans nematodes with pathogens [36,37]. Indeed, we found a significant overlap of the transcriptome especially upon Staphylococcus aureus (S. aureus) ( Figure S1A and Table S6) and Pseudomonas aeruginosa (P. aeruginosa) infection ( Figure S1B and Table S7).
Next, we asked whether flcn-1(ok975) mutant animals display enhanced resistance to pathogens. Strikingly, we found that the flcn-1(ok975) mutant animals are more resistant than wild-type animals to S. aureus and P. aeruginosa infection ( Figure 1C

Loss of flcn-1 increases pathogen resistance via HLH-30 activation
HLH-30, the worm ortholog of TFEB/TFE3, has been reported to modulate longevity and pathogen resistance in C. elegans through activation of autophagy and expression of antimicrobial genes [7,12]. Importantly, we found a significant overlap between genes that were upregulated in flcn-1(ok975) mutant animals and downregulated in hlh-30(tm1978) mutant animals (Table S10) [6]. Thus, we asked whether HLH-30 is induced in flcn-1 mutants using an hlh-30::GFP transgenic reporter strain [6,7]. Upon infection with S. aureus, as shown in this study and others [6 ], HLH-30 translocated to the nucleus (Figure 2A). In particular, close to 40% of the wild-type animals displayed an HLH-30 nuclear localization after 20 min of infection with S. aureus. Importantly, we observed that the percentage of animals displaying a constitutive nuclear HLH-30 translocation in uninfected worms was significantly higher upon loss of flcn-1 ( Figure 2B, time 0). Strikingly, we show that upon S. aureus infection, the percentage of animals with HLH-30 nuclear translocation increased further in flcn-1 mutant animals. Specifically, we found that after 20 min of infection with S. aureus, more than 80% of the flcn-1 mutant animals displayed an HLH-30 nuclear localization in comparison to less than 40% for wild-type animals.
Overall, this highlights an important role for HLH-30 in the increased pathogen resistance conferred by loss of flcn-1 ( Figure 2B).
Importantly, loss of hlh-30 significantly impaired the survival advantage upon both S. aureus ( Figure 2C) and P. aeruginosa infections ( Figure 2D) that was conferred by loss of flcn-1, demonstrating its involvement in pathogen resistance (Tables S8A-B). Accordingly, loss of hlh-30 also suppressed the increased resistance of flcn-1 to hyperosmotic stress [28] supporting that the adaptation to the two stresses requires a similar transcriptional profile dictated by HLH-30 ( Figure S2).
To further assess the involvement of HLH-30 in the transcriptional response downstream of FLCN-1, we measured the gene expression of known HLH-30 target genes [6]. Using RT-qPCR, we found a significant upregulation in many hlh-30-dependent antimicrobial and infection-associated genes in uninfected flcn-1 mutant worms ( Figures 2E-K). Furthermore, after 4 h of infection with S. aureus, we show that loss of hlh-30 strongly reduced the expression of antimicrobial peptide genes and infection-related genes in both wild-type and flcn-1 mutant animals ( Figures 2E-K), supporting a role for HLH-30 in the pathogen transcriptional signature downstream of flcn-1. Collectively, we found that loss of flcn-1 activates the transcription of HLH-30 antimicrobial peptide genes at basal level, which is further induced upon S. aureus infection.

The regulation of TFEB/TFE3 by FLCN is evolutionarily conserved through an mTOR independent mechanism
Because the role of HLH-30 in host defense is evolutionarily conserved [6], we tested whether the FLCN-HLH-30 axis that we uncovered in C. elegans is conserved from worms to mammals. Indeed, we observed that Flcn deletion in mouse embryonic fibroblasts (MEFs) promoted TFEB and TFE3 nuclear localization at basal levels compared to wild-type MEFs as detected by subcellular fractionation and immunofluorescence assays ( Figure 3A-C). The difference in the cytosolic TFEB molecular weight can be attributed to the phosphorylation forms of TFEB [17]. Consequently, known TFEB and TFE3 targets were upregulated at the mRNA level upon Flcn deletion ( Figure 3B), including genes involved in innate host response, such as IL-6. Addition of Torin1, a specific inhibitor of mTORC1 and mTORC2, induced TFEB nuclear localization in Flcn knockout (KO) MEFs to a higher extent than wild-type MEFs ( Figure 3C), evoking an mTOR independent pathway. Moreover, loss of Flcn did not affect mTOR signaling, as measured by immunoblotting for the phosphorylated form of the S6 ribosomal protein (S6), a well-described mTORC1 downstream target ( Figure 3D). In line with this, in C. elegans, inhibition of let-363, the C. elegans TOR homolog, increased the HLH-30 nuclear translocation at basal level similar to what has been previously reported [7] ( Figure 3E). Importantly, loss of flcn-1 further increased the HLH-30 nuclear translocation upon inhibition of let-363 at basal level supporting a TOR-independent pathway governing HLH-30 regulation ( Figure 3E). Moreover, infection with S. aureus increased to a similar extent the HLH-30 nuclear localization both in wild-type and flcn-1 (ok975) animals fed with let-363 RNAi, presumably because the infection happens rapidly masking the effects of let-363 RNAi on HLH-30 1 0 nuclear localization through a mechanism distinct from the canonical mTOR pathway both in nematodes and mammalian cells.
To further assess whether the transcriptional up-regulation of cytokines and chemokines upon loss of Flcn was mediated by TFEB and TFE3, we knocked down their endogenous expression simultaneously using shTFEB and shTFE3 in wild-type and Flcn KO MEFs and determined the expression of IL-6 following TNFα stimulation ( Figure 3F-G). Notably, we found that the significant induction of IL-6 mRNA levels upon TNF-α stimulation in both wildtype and Flcn KO MEFs was abrogated to levels observed in unstimulated cells upon knockdown of TFEB/TFE3 ( Figure 3F). To confirm the observed effects in a relevant cellular system for innate immune response, we used RAW264.7 murine macrophages and reduced the endogenous expression of Flcn using shRNA-mediated knockdown approaches. Importantly, we show a significant increase in IL-6 production, at both mRNA ( Figure 3H) and protein levels ( Figure 3I), in FLCN KD macrophages compared to empty vector (EV) in response to LPS stimulation. To further assess the role of FLCN in inflammation and innate immune response, we determined the cytokine and chemokine secretion profiles in wild-type and shFLCN macrophages after 3h and 24h of LPS stimulation using mouse protein cytokine arrays. Notably, we show a significant and prominent increase in many cytokines in FLCN KD macrophages as compared to EV upon LPS stimulation ( Figures 3J-K). These cytokines encompass key mediators of the inflammatory response.

FLCN depletion in macrophages enhances their energy metabolism and phagocytic potential
Next, we investigated the metabolic consequences of FLCN depletion in RAW264.7 macrophages. We found that glucose consumption and lactate production were increased in FLCN KD macrophages compared to control macrophages ( Figure 4A-B), and this was accompanied by an augmented extracellular acidification rate (ECAR) ( Figure 4C-D) and oxygen consumption rate (OCR) (Figure 4E-F) at basal level and upon the sequential addition of oligomycin (an ATP synthase inhibitor), Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) (for maximum respiratory capacity), followed by rotenone/antimycin A (to block mitochondrial electron transport). In line with these results, we also report an increase in ATP production in FLCN KD macrophages compared to controls ( Figure 4G). Next, we investigated whether FLCN depletion enhances the phagocytic potential in macrophages ( Figure 4I). Using pHrodo Red S. aureus Bioparticles, we report a 30% increase in phagocytic capacity of FLCN KD macrophages compared to control cells, as shown by the fold change in the mean florescence intensity ( Figure 4I). To test whether this increase in phagocytic activity in FLCN KD macrophages is dependent on TFEB/TFE3 activation, we knocked down FLCN in TFEB/TFE3 DKO RAW macrophages ( Figure 4H), and showed that the phagocytic activity of these cells decreased by almost 50% compared to FLCN depleted macrophages, upon stimulation with pHrodo Red S. aureus Bioparticles ( Figure 4I). Taken together, we show that depletion of FLCN in macrophages prompts a metabolic transformation toward increased cellular bioenergetics, accompanied by an augmented TFEB/TFE3-dependent phagocytic capacity, which might further enhance the innate immune response.

AMPK regulates HLH-30 activation and antimicrobial response upon infection with bacterial pathogens.
Given that we have previously reported that loss of flcn-1 leads to chronic AMPK activation, which increases resistance to energy [29] and hyperosmotic stresses [34] in nematodes, we tested whether flcn-1 mutant animals confer pathogen resistance via AMPKmediated regulation of HLH-30. Importantly, simultaneous loss of aak-1 and aak-2 (C. elegans orthologs of AMPK α 1/α2) completely abolished the increased survival to both S. aureus and P.
aeruginosa in wild-type and flcn-1 mutant animals, demonstrating that this phenotype requires AMPK ( Figures 5A-B and Tables S8A-B). Furthermore, transgenic overexpression of a constitutively active catalytic subunit of AMPK (aak-2 oe) in nematodes confers pathogen resistance similar to flcn-1(ok975) mutants, which is mostly dependent on HLH-30 ( Figure 5C and Table S8 A). Moreover, loss of both AMPK catalytic subunits significantly reduced the nuclear translocation of HLH-30 upon S. aureus infection ( Figure 5D). Additionally, we found that loss of aak-2(ok524) alone was insufficient to reduce the nuclear translocation of HLH-30 upon S. aureus infection, suggesting that complete abrogation of both AMPK catalytic activities is required for this phenotype ( Figure S3B). To further elaborate the role of AMPK in pathogen response and specifically in the transcription of antimicrobial and stress response genes upon infection, we used RNA-seq technology to measure differential gene expression in wild-type and aak-1(tm1944); aak-2(ok524) mutant animals at basal level and after 4 h infection with S. aureus ( Figure S3A, Tables S11-S15). We identified more than 800 genes induced upon S. aureus infection that are dependent on AMPK ( Figure 5E and S3A, Tables S16-S17). Furthermore, we found a significant overlap of 112 genes down-regulated in aak-1(tm1944); aak-2(ok524) mutant animals and genes regulated by hlh-30 upon S. aureus infection [6] ( Figure 5F, Table S18). Gene ontology classification highlights important pathways regulated by AMPK during S. aureus infection, including defense response and stress response pathways ( Figure 5G and Tables S16-S19  Additionally, we show that IL-6, a TFEB/TFE3 target, was transcriptionally upregulated when treated with GSK-621 and its expression was abrogated upon down-regulation of TFEB/TFE3 using shTFEB/TFE3 ( Figure 6D) implying that AMPK impinges on TFEB/TFE3-mediated transcription in mammalian cells similarly to what we have observed in C. elegans. Moreover, treatment of RAW264.7 macrophages with GSK-621 activated AMPK without affecting mTOR signaling ( Figure 6E), promoted the nuclear translocation of TFEB ( Figure 6F), and led to a strong increase in production and secretion of various cytokines and chemokines even in the absence of LPS treatment or pathogen infection ( Figure 6G). To substantiate our findings in a more physiological context, we tested whether acute LPS treatment of macrophages could affect cellular bioenergetics, which could be sensed by AMPK. Indeed, we observed an acute reduction in cellular ATP levels ( Figure 6H), accompanied by AMPK activation (Figure 6I), and a significant increase in TFEB nuclear localization ( Figure 6J) as early as 30 minutes after addition of LPS in RAW macrophages.
Collectively, both the mammalian and worm results demonstrate an important role for AMPK in the regulation of the innate host immune response through TFEB/TFE3 activation.

Discussion
We have previously shown that loss of FLCN activates AMPK, increasing the resistance to oxidative stress, heat, anoxia, hyperosmotic stresses, and obesity in C. elegans and mammalian models [28][29][30][31]40]. Here, we report for the first time evidence supporting a novel and evolutionary conserved role of FLCN in innate host defense mediated through AMPK and TFEB/TFE3 activation.
Given that the gene profile upon osmotic stress mimics that of pathogen infection [36], we found a significant overlap in the transcriptional profile in flcn-1 mutant animals when compared to wild-type animals infected with pathogens. We report that flcn-1 mutant animals confer resistance to pathogen infection through nuclear localization and activation of HLH-30.
Increased nuclear localization and activation of TFE3 were previously reported in renal tumors from Birt Hogg-Dubé syndrome patients, a syndrome associated with a germline mutation of the Bacterivorous nematodes, such as C. elegans induce the expression of transcriptional host-defense responses including the HLH-30/TFEB pathway that promote organismal survival [6,37,[51][52][53]. However, these invertebrates appear to lack the NLR and TLR pathogen sensing pathways as well as NF-κB and other transcription factor pathways that regulate innate immunity in higher organisms [3,54]. Our findings shed light on an ancient, highly conserved pathogen sensing and signal transduction mechanism, which involves AMPK and the transcription factor TFEB/TFE3. LPS, which is part of the outer membrane of Gram-negative bacteria, was shown to inhibit respiration and energy production in cells and isolated mitochondria [55][56][57][58]. We show here, that LPS treatment of macrophages leads to an acute reduction of cellular energy levels resulting in AMPK activation, induction of TFEB/TFE3 and inflammatory cytokines (Figure 6H-K; Figure 3H, I). Therefore, this ancient pathogen-sensing pathway may have evolved by the fact that pathogen infection leads to an energy shortage, which is sensed by cellular AMPK.
Activated AMPK will in turn promote TFEB/TFE3 nuclear translocation and induction of an innate host defense. Furthermore, in this study we place AMPK at the center of FLCN-TFEB/TFE3 axis.
Several direct AMPK activators are being developed for treatment of type-2 diabetes, obesity, and metabolic syndrome [64]. We propose that some of these specific AMPK activating compounds could be repurposed to enhance host defense against pathogens or treat other immunodeficiency syndromes through AMPK-mediated activation of TFEB/TFE3, providing new druggable strategies in innate immune modulation and therapy of bacterial infections. To this end, pharmacological inhibition of mTOR is currently being investigated in human clinical trials to treat age-associated immune dysfunction, also dubbed "immune senescence" (resTORbio, Inc). 0 MOP-142259. J.A.M. and R.P. were supported by the Intramural Research Program of the National Institutes of Health, National Heart, Lung, and Blood Institute (NHLBI).

C. elegans strains, maintenance, and RNAi treatments
All strains used in this study are described in supplementary table S20 (Table S20).
Nematodes were maintained and synchronized using standard culture methods [65]. The RNAi feeding experiments were performed as described in [66], and bacteria transformed with empty vector were used as control. For all RNAi experiments, phenotypes were scored with the F1 generation.

Pathogen resistance assay
To measure pathogen stress resistance, synchronized L4 worms were transferred to Tryptic Soya Agar (TSA) plates with 8 µg/ml Nalidixic acid that were seeded with 1:50 S.

RNA extraction and real-time PCR in C. elegans
Synchronized young adult nematodes were exposed to pathogenic S. aureus bacteria or

RNA seq method
Synchronized wild-type and aak-1(tm1944); aak-2(ok524) animals were harvested at the late L4 stage, washed with M9, and flash frozen in liquid nitrogen. RNA was extracted using Trizol and purified using Qiagen RNeasy columns. RNA samples were processed for RNA-seq analysis at Novogene Inc.

Mouse Protein Cytokine Array
RAW 264.7 cells were seeded in triplicates in 6-well plates at 1 x 10 6 cells per well in DMEM medium supplemented with 10% FBS. After incubation for 24 h at 37 °C, 5% CO 2 , cells were treated with LPS or GSK-621 or vehicle for 3, or 24 h, and the conditioned medium was harvested and centrifuged at 1,500 × g to remove cell debris. 32 cytokine/chemokine/growth factor biomarkers were simultaneously quantified by using a Discovery

Protein extraction and immunoblotting
For AMPK immunoblotting, cells were washed twice with cold PBS, lysed in AMPK lysis buffer (10 mM Tris-HCl (pH 8.0), 0.5 mM CHAPS, 1.5 mM MgCl 2 1 mM EGTA, 10% glycerol, 5 mM NaF, 0.1 mM Na 3 VO 4 , 1 mM benzamidine, 5 mM NaPPi), supplemented with complete protease inhibitor (Roche) and DTT (1 mM), and cell lysates were cleared by centrifugation at 13000 x g. For all other immunoblotting, cells were washed twice with cold PBS and lysed directly in Laemmli buffer (62.5 mM Tris-HCl (pH 6.8), 2% (w/v) sodium bromophenol blue). Proteins were separated on SDS-PAGE gels and revealed by western blot as we previously described [29] using the antibodies listed above.

ATP quantification
Cells were plated in triplicates in 96-well plates. After 24 h, cells were lysed and mixed for 10 min according to manufacturer's instructions (CellTiter-Glo luminescent cell viability assay, Promaga). Luminescence was measured using Fluostar Omage (BMG Labtech) directly in plates.

Phagocytosis Assay
Phagocytosis in EV or shFLCN RAW264.7 cells was assessed using Red pHrodo S.aureus BioParticles conjugate assay (Thermofisher) according to the manufacturer's protocol.
In brief, EV or shFLCN RAW264.7 were plated in triplicates in 96-well plates at 80,000 cells/well in growth medium for 2 h before treatment. After 2 h, cells were treated with the BioParticles (after homogenization in serum-free DMEM) at a final concentration of 1 mg/ml and incubated at 37 o C for 3 h. Subsequently, cells were collected and analyzed using FACSDiva analyzer (Becton Dickison).

Statistical Analyses
Data are expressed as mean ± SEM. Statistical analyses for all data were performed using student's t-test for comparisons between 2 groups, one-way ANOVA for comparisons between 3 or more groups, and Log-rank Mantel Cox test for survival plots, using GraphPad Prism 7 software. Statistical significance is indicated in figures (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) or included in the supplemental tables.  Significance was determined using one-way ANOVA with the application of Bonferroni correction (*p<0.05, **p<0.01, ***p<0.001). replicate. Fold increase was normalized against EV and color-coded (dark red indicates 2 or 1 more-fold increase, dark blue indicates no change). (L) Fold increase in cytokine and chemokine secretion levels as described in (J). Data represent the average of three independent experiments, each done in triplicates ± SEM. Significance was determined using student's t-test in comparison to the EV stimulated with LPS for 3h and 24, respectively (*p<0.05, **p<0.01, ***p<0.001). Significance was determined using one-way ANOVA with the application of Bonferroni correction (**p<0.01, ****p<0.0001).  Significance was determined using student's t-test (*p<0.05, **p<0.01, ***p<0.001).  Table S9 for details on number of animals utilized and number of repeats Statistics obtained by Mantel-Cox analysis on the pooled curve.

Fig S3: RNA Seq heat map and gene ontology analysis in wild-type and aak-1(tm1944); aak-2(ok524) at basal level and upon S. aureus infection.
(A) Heat map showing differential gene expression in wild-type and aak-1(tm1944); aak-2(ok524) mutant animals grown on OP50 or exposed to S. aureus for 4 h. Red color indicates genes that are differentially upregulated while blue color indicates gene sets that are downregulated in comparison to untreated wild-type animals. RNAseq data were analysed by Novogene Inc. using DEGseq 1.12.0 (B) Nuclear translocation of HLH-30 in aak-2(ok524); hlh-30::GFP at basal level and upon S. aureus infection. Data represent the mean ± SEM with 3 independent repeats, n ≥ 30 animals/condition for every repeat. Significance was determined using student's t-test.                 -1(tm1944); aak-2(ok524) animals in comparison to wild-type animals at basal level.

wild-type animals upon S. aureus infection and downregulated in infected aak-1(tm1944);
aak-2(ok524) mutant animals.    l  i  c  u  l  i  n  e  n  c  o  d  e  d  b  y  t  h  e  B  H  D  g  e  n  e  i  n  t  e  r  a  c  t  s  w  i  t  h  a  b  i  n  d  i  n  g  p  r  o  t  e  i  n  ,  F  N  I  P  1  ,  a  n  d  A  M  P  K  ,  a  n  d  i  s  i  n  v  o  l  v  e  d  i  n  A  M  P  K  a  n  d  m  T  O  R  s  i  g  n  a  l  i  n  g  .  P  r  o  c  N  a  t  l  A  c  a  d  S  c  i  U  S  A  ,  2  0  0  6  .  1  0  3  (  4  2  )  :  p  .  1  5  5  5  2  -7  .  2  3  .  T  a  k  a  g  i  ,  Y  .  ,  e  t  a  l  .  ,  I  n  t  e  r  a  c  t  i  o  n  o  f  f  o  l  l  i  c  u  l  i  n  (  B  i  r  t  -H  o  g  g  -D  u  b  e  g  e  n  e  p  r  o  d  u  c  t  )  w  i  t  h  a  n  o  v  e  l  F  n  i  p  1  -l  i  k H  o  o  g  e  w  i  j  s  ,  D  .  ,  e  t  a  l  .  ,  S  e  l  e  c  t  i  o  n  a  n  d  v  a  l  i  d  a  t  i  o  n  o  f  a  s  e  t  o  f  r  e  l  i  a  b  l  e  r  e  f  e  r  e  n  c  e  g  e  n  e  s  f  o  r  q  u  a  n  t  i  t  a  t  i  v  e  s  o  d  g  e  n  e  e  x  p  r  e  s  s  i  o  n  a  n  a  l  y  s  i  s  i  n  C  .  e  l  e  g  a  n  s  .  B  M  C  M  o  l  B  i  o  l  ,  2  0  0  8  .  9  :  p  .  9  .  6  9  .  V  i  n  c  e  n  t  ,  E  .  E  .  ,  e  t  a  l  .  ,  D  i  f  f  e  r  e  n  t  i  a  l  e  f  f  e  c  t  s  o  f  A  M  P  K  a  g  o  n  i  s  t  s  o  n  c  e  l  l  g  r  o  w  t  h  a  n  d  m  e  t  a  b  o  l  i  s  m  .  O  n  c  o  g  e  n  e  ,  2  0  1  5  .  3  4  (  2  8  )