PFKFB3-Driven Macrophage Glycolytic Metabolism Is a Crucial Component of Innate Antiviral Defense

This information is current as Hui Jiang, Hengfei Shi, Man Sun, Yafeng Wang, Qingzhou of October 2, 2021. Meng, Panpan Guo, Yanlan Cao, Jiong Chen, Xiang Gao, Erguang Li and Jianghuai Liu J Immunol 2016; 197:2880-2890; Prepublished online 26 August 2016; doi: 10.4049/jimmunol.1600474 http://www.jimmunol.org/content/197/7/2880 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2016/08/26/jimmunol.160047

Material 4.DCSupplemental http://www.jimmunol.org/ References This article cites 62 articles, 20 of which you can access for free at: http://www.jimmunol.org/content/197/7/2880.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 2, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

PFKFB3-Driven Macrophage Glycolytic Metabolism Is a Crucial Component of Innate Antiviral Defense

Hui Jiang,*,1 Hengfei Shi,†,1 Man Sun,* Yafeng Wang,* Qingzhou Meng,* Panpan Guo,* Yanlan Cao,* Jiong Chen,* Xiang Gao,* Erguang Li,*,† and Jianghuai Liu*

Signaling by viral nucleic acids and subsequently by type I IFN is central to antiviral innate immunity. These signaling events are also likely to engage metabolic changes in immune and nonimmune cells to support antiviral defense. In this study, we show that cytosolic viral recognition, by way of secondary IFN signaling, leads to upregulation of glycolysis preferentially in macrophages. This metabolic switch involves induction of glycolytic activator 6-phosphofructose-2-kinase and fructose-2,6-bisphosphatase (PFKFB3). Using a genetic inactivation approach together with pharmacological perturbations in mouse cells, we show that

PFKFB3-driven glycolysis selectively promotes the extrinsic antiviral capacity of macrophages, via metabolically supporting Downloaded from the engulfment and removal of virus-infected cells. Furthermore, the antiviral function of PFKFB3, as well as some contribution of its action from the hematopoietic compartment, was conﬁrmed in a mouse model of respiratory syncytial virus infection. There- fore, different from the long-standing perception of glycolysis as a proviral pathway, our ﬁndings establish an antiviral, immunometabolic aspect of glycolysis that may have therapeutic implications. The Journal of Immunology, 2016, 197: 2880–2890.

igher organisms have evolved a powerful innate immune program in a broad range of cell types to exert powerful antiviral http://www.jimmunol.org/ system to defend against viral infections. The activation effects (2, 3). H of innate antiviral immunity is based on recognition of It has been increasingly recognized that innate immune activa- virus-associated molecular patterns, including specific nucleic acid tion is associated with significant metabolic changes (4, 5). Such species, by cell/endosome surface TLRs in professional innate cross-talks may represent a strategic shift of cellular metabolism to immune cell types or by another group of cytosolic viral sensors support the immune responses. As far as glucose metabolism is expressed in virtually all cell types (1). Viral sensing by these concerned, it was found that in macrophages and dendritic cells, receptors engage different, yet overlapping signaling pathways, TLR activation results in significant increases in oxygen-independent which eventually converge on the production of the type I IFN as glycolysis and often decreases in the aerobic oxidative phosphory- well as a number of proinflammatory cytokines. Type I IFN, in lation (OXPHOS) metabolism (6–12). Such metabolic changes have by guest on October 2, 2021 particular, trigger a Stat1/2- and IRF9-dependent transcriptional been proven essential for some key immunoregulatory events downstream of TLR activation. However, few studies have investigated the features of cellular metabolic changes in the context of innate antiviral *State Key Laboratory of Pharmaceutical Biotechnology and Ministry of Education immune activation by live viruses and have dissected the contributing Key Laboratory of Model Animals for Disease Study, Model Animal Research Cen- signaling pathways. Moreover, whether immune cells and non- † ter of Nanjing University, Nanjing 210061, China; and School of Medicine, Nanjing immune cells undergo differential metabolic regulation under University, Nanjing 210093, China 1 viral challenges is not known. H.J. and H.S. contributed equally to this work. To fill such gaps in knowledge, we examined virus-elicited ORCIDs: 0000-0003-1170-0224 (H.J.); 0000-0001-5305-3112 (H.S.); 0000-0001- regulation of glycolytic metabolism in macrophages and mouse 9301-9736 (Y.W.); 0000-0002-3065-1336 (E.L.); 0000-0002-2382-9140 (J.L.). embryonic fibroblasts (MEFs), as representatives of innate immune Received for publication March 18, 2016. Accepted for publication August 1, 2016. cells types and nonimmune cell types, respectively. We established This work was supported by Chinese National Science and Technology Pillar Pro- gram Grant 2015BAI08B02 and National Science Foundation of China Grants that type I IFN triggered by viral infection acts to upregulate 31471313, 81371772, and 31271499. glycolytic activator 6-phosphofructose-2-kinase and fructose-2,6- Address correspondence and reprint requests to Dr. Jianghuai Liu or Dr. Erguang Li, bisphosphatase (PFKFB3) preferentially in macrophages. We State Key Laboratory of Pharmaceutical Biotechnology and Ministry of Education provided further evidence to demonstrate PFKFB3 as a novel, Key Laboratory of Model Animals for Disease Study, Model Animal Research Cen- ter of Nanjing University, 12 Xuefu Road, Nanjing 210061, China (J.L.) or School of metabolic effector of innate antiviral immune response. Medicine, Nanjing University, 22 Hankou Road, Nanjing 210093, China (E.L.). E-mail addresses: [email protected] (J.L.) or [email protected] (E.L.) Materials and Methods The online version of this article contains supplemental material. Ethics statement Abbreviations used in this article: BMDM, bone marrow–derived macrophage; ECAR, extracellular acidification rate; F2,6BP, fructose-2,6-bisphosphate; ISG, IFN-stimulated In this study, all experiments involving mice usage by the collaborating gene; MEF, mouse embryonic fibroblast; MOI, multiplicity of infection; 2-NBDG, laboratories were approved by the Institutional Animal Care and Use 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose; NJU, Nanjing Univer- Committee of the Model Animal Research Center at Nanjing University sity; OCR, oxygen consumption rate; OXPHOS, oxidative phosphorylation; PFK, (MARC-NJU project license LJH12) and by the Institutional Animal Care phosphofructose-1-kinase; PFKFB3, 6-phosphofructose-2-kinase and fructose-2,6- bisphosphatase; PM, peritoneal macrophage; poly(I:C), polyinosinic–polycytidylic and Use Committee of the School of Medicine at NJU (project license LiE- acid; qPCR, quantitative PCR; RSV, respiratory syncytial virus; VSV, vesicular sto- 02). The animal care and use protocols were in strict accordance with matitis virus; WT, wild-type. Regulation for Management of Laboratory Animals (1988) and Guidelines for Care and Use of Laboratory Animals (2006), both issued by the Ministry Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 of Science and Technology of PR China. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1600474 The Journal of Immunology 2881

Mice Extracellular lactate assay All mice were of C57/BL6 background. They were produced and main- In brief, the culture medium samples were deproteinated (equal volume of tained on ad libitum food and water by Nanjing Biomedical Research 0.5 M of metaphosphoric acid). The supernatant was then neutralized with Institute-NJU, an American Association for the Accreditation of Labora- 1/20 volume of 5 M of K2CO3. After centrifugation to remove the pre- tory Animal Care–accredited specific pathogen-free animal facility. For cipitated salts, the samples were analyzed using the lactate assay kit experiments involving viral infection, the mice were transferred to a BSL2- according to manufacturer’s instruction (Cayman). level animal facility at NJU School of Medicine. The Pfkfb3 heterozygous knockout mice were generated by clustered reg- 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose ularly interspaced short palindromic repeats/CRISPR-associated 9 technique uptake assay that caused a 73-bp indel in exon 2 of Pfkfb3. The mice were genotyped by PCR using two primers (forward: 59-GGTATAGGACTCACACCATTAAG-39; The cells were incubated for 30 min to 1 h in DMEM without serum and m reverse: 59-GACCTGGCTTACCTTTCGTTGGA-39) to amplify a 202-bp glucose. Cells were next incubated with 150 g/ml of a fluorescent glucose fragment for the wild type (WT) allele and 129-bp fragment for the deletion analog 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose mutant allele. In all experiments involving Pfkfb3+/2 mice, WT littermates of (2-NBDG; Invitrogen) for 15 min. After washing, cells were collected by the same sex were always used as controls. trypsin solution (MEF) or PBS containing 0.5% EDTA (macrophage). Cells were maintained on ice and promptly assayed by flow cytometry on a Reagents and Abs FACSCalibur platform (BD). Unless otherwise indicated, all chemicals were purchased from Sigma- Seahorse XF24-3 Metabolic flux analyses Aldrich. Polyinosinic–polycytidylic acid [poly(I:C)] was purchased from Invivogen. Mouse IFN-b (12405-1) and human IFN-a were from PBL and PMs or BMDMs were plated at 100,000 cells/well in a specialized 24- Sangon, respectively. Wortmannin (9951) was from Cell Signaling Tech- well plate. On the next day, untreated or IFN-treated cells were subjected nology. JAK Inhibitor I (420099) was from Merck Millipore. PFK15 was to analysis following instructions bundled with the instrument. In brief, purchased from Abcam. Lipofectamine 2000 (Invitrogen) and INTERFERin cells were washed with assay media and incubated for 1 h at 37˚C Downloaded from (Polyplus) were used to mediate poly(I:C) transfection of nonimmune cells without CO2. Glycolytic flux and cellular respiration were then quan- and macrophages, respectively. titated by recording extracellular acidification rate (ECAR; mpH/min) Primary Abs were purchased from Sangon (Stat1, AB55186), Proteintech and oxygen consumption rate (OCR; pmol/min), respectively. The in- (PFKFB3, 13763-A-AP), Santa Cruz (GAPDH, A1713), Genscript (VSV-G, dicated stress reagents were injected at preset time points. After the A00199) Sigma-Aldrich (Flag, F3165), Abcam (PFKFB3, AB181861), and assay, cell lysates were harvested. Results were normalized by protein Cell Signaling Technology (Abs for AKT pathway). Neutralizing Abs against concentrations. mouse IFNAR1 (127302), blocking Ab against mouse TIM4 (130004), the RNA extraction and quantitative real-time PCR http://www.jimmunol.org/ fluorophore-conjugated Abs against CD11b (101217), CD11c (117311), or F4/80 (123122), and the ELISA kit for mIFN-b (439407) were purchased Total RNA was extracted using the RNAiso Plus (Takara) according to the from Biolegend. manufacturer’s protocol. Reverse transcription and SYBR-based quantitative PCR (qPCR) analyses were performed as previously described (15). Primary cells and cell culture The 18S rRNA or GAPDH was used as internal controls for normalization. Peritoneal macrophages (PMs) and bone marrow–derived macrophages Primer sequences were generally derived from the publicized database of (BMDMs) were prepared from the mice (same sex, 8–12 wk old) PrimerBank. Sequences of more frequently used primers are listed in according to standard protocols (13). MEFs were obtained from 12.5-d-old Supplemental Fig. 1A. Others can be distributed upon request. embryos as described previously (14). The cell lines used in this study Cell lysis and Western analysis

(L929, Raw264.7, Hela, Vero, U937, 293T) were all previously obtained by guest on October 2, 2021 from American Type Culture Collection. The lentiviral tet-on shRNAmir When not indicated specifically, cells lysis was performed in buffer con- expression system pTRIPZ (Open Biosystems) was modified to drive in- taining 1% Triton-X 100, 0.5% Nonidet P-40, 150 mM of NaCl, 10 mM ducible expression of PFKFB3 as described previously (15). The packaged of Tris-HCl pH 7.5, 1 mM of EDTA, 1 mM of NaF, 1 mM of sodium recombinant lentivirus was used to transduce RAW264.7 cells. Stable orthovanadate, protease inhibitor mixture (1:500 dilution; Sigma) and transductants were selected with 1 mg/ml puromycin. For transient trans- freshly added 1 mM of PMSF. Western blots were performed as previously fection of PFKFB3 into MEFs, a Flag-tagged construct based on the described (17). pcDNA plasmid (Invitrogen) was used. DMEM (Life Technologies) supplemented with 10% FBS, 1% penicillin/ Flow cytometry analyses of cell-surface markers streptomycin were used to culture most of the cells used in the study. These include primary PMs and MEFs. For differentiation of BMDMs, mouse BMDMs and PMs were dissociated from the culture vessels using PBS containing 0.5% EDTA and were incubated with fluorophore- bone marrow aspirates were cultivated in the presence of 30% L929 su- conjugated Ab against F4/80, CD11b, or CD11c and the correspond- pernatant (containing endogenous M-CSF) for 7 d. U937 and L929 cells were maintained in RPMI 1640 (Invitrogen) supplemented with 10% FBS, ing isotype controls. The analyses were performed on a FACSCalibur 1% penicillin/streptomycin. All cultures were maintained using a 37˚C platform. humidified incubator supplied with 5% CO2. Lung macrophage depletion Vesicular stomatitis virus infection in vitro Ten-week-old mice were intranasally instilled with 200 mlofPBSor clodronate-containing liposomes (ClodronateLiposomes.com). Forty- Vesicular stomatitis virus (VSV; Indian strain) was propagated using Vero m cells. For in vitro infection, cells were inoculated (1 h) with the virus at eight hours later, cells were injected i.p. with 50 g of poly(I:C). indicated multiplicity of infections (MOIs). The culture supernatant and cell Twelve hours later, mice were sacrificed and lung tissues were har- samples were harvested at different time points afterward. For UV inac- vested for analyses. Histological processing of the tissue and immunofluorescence microscopy were performed as previously described tivation, the viral stock was exposed under the UV lamp in a tissue culture (17, 18). hood for 30 min. For coculture experiments, infected MEFs (MOI of 1 for 14 h to achieve significant cytopathic effect) were added at a ratio of 2:1 to Efferocytosis of apoptotic thymocytes BMDMs and incubated further. In some experiments, MEFs were added to an upper transwell apparatus (3-mm pore size) separated from the BMDMs Thymocytes were induced to undergo apoptosis (∼70%) using dexa- in the lower chambers. VSV titer was determined using plaque assay in methasone (6 h). After extensive washing, the apoptotic thymocytes were Hela cells (16). In experiments to visualize cell engulfment, MEFs were then added to PMs at a 20:1 ratio. Four hours later, cells were fixed and initially labeled with fluorescent probe CFSE (eBioscience) according to stained (H&E). The numbers of engulfed thymocytes/uptaking macro- manufacturer’s instructions. After incubating the infected MEFs with phage and the percentage of macrophages undergoing phagocytosis were macrophages, the cells were fixed, stained using F4/80 Ab (macrophage), determined in at least 10 random fields by light microscopy (.500 total and analyzed by immunofluorescence microscopy (17). In some experi- cells). As expected, there was often a positive correlation between the ments, engulfment of infected MEFs was assessed via flow cytometry. two measurements. Phagocytic indices were calculated as the product of After incubation with the infected MEFs, unfixed phagocytes were stained the latter two measurements. For measuring the extent of cargo adhesion, with anti-F4/80 and subsequently subjected to flow cytometry analyses. macrophages were incubated with a great excess (in a 1:100 ratio) of Median fluorescent intensity was used for quantitation. apoptotic thymocytes for 15 min (19). 2882 GLYCOLYSIS AND INNATE ANTIVIRAL DEFENSE

Fructose-2,6-bisphosphate measurements because of the insufficient amount of VSV-G to engage TLR4 The assay was adapted from an established protocol based on the use of under the experiment condition (MOI of 1). Likewise, UV- potato-derived phosphofructose-1-kinase (PFK) that is highly sensitive to the inactivated VSV failed to trigger glycolytic enhancement in PMs allosteric activation by fructose-2,6-bisphosphate (F2,6BP) (20). In brief, (Fig. 1C). These results suggested that the observed metabolic re- 50 mg of cell lysates (as a source of F2,6BP) was added to an assay buffer sponse in PMs by live VSV was likely to be dependent on cyto- containing the pyrophosphate-dependent, potato PFK (PPi-PFK; Sigma), solic viral RNA-sensing pathway (26). Consistently, transfection of aldolase, triosephosphate isomerase, GAPDH, as well as NAD+. The assay was initiated by adding a mixture of substrates including glucose-6- macrophages using poly(I:C), an analog of virus-associated dsRNA phosphate/fructose-6-phosphate and pyrophosphate. The conversion of (27), also resulted in increased glycolysis (Fig. 1D, Supplemental NAD+ to NADH was indicated by increases in OD at 340 nm. No activity Fig. 1D). was recorded when PPi-PFK was omitted from the reaction mix. Next, to examine the potential glycolysis-enhancement activities by viral RNA-induced secreted factors, we harvested conditioned Respiratory syncytial virus infection in vivo medium from poly(I:C)-transfected MEFs. An aliquot of condi- Seven-week-old mice were instilled intranasally with 5 3 106 respiratory tioned medium was then added, respectively, to PMs and MEFs, m syncytial virus (RSV; A2 subtype) in a volume of 100 l. Five days after and analyzed for its activity (8-h treatment) to regulate cellular gly- inoculation, mice were sacrificed and lung tissues were harvested for further analyses. To prepare bone marrow chimeras, we lethally irradiated (10 Gy) colysis (uptake of a glucose analog 2-NBDG and lactate produc- 6-wk-old C57/BL6 mice according to an established protocol (21). A total of tion; Supplemental Fig. 1E, 1F). Notably, mirroring the effects by 2 5 3 106 WT or Pfkfb3+/ donor bone marrow cells were injected to the re- VSV infection (Fig. 1A), the extracellular fraction from poly(I:C)- cipients via tail vein. After 6 wk of recovery, the mice were subjected to RSV transfected MEFs sufficed to enhance glycolysis in PMs, but not in challenges. RSV titers were determined by plaque assay using Hela cells (22). MEFs. Consequently, this led us to examine the involvement of type Downloaded from Statistical methods I IFN. All type I IFNs act via one receptor composed of subunits IFNAR1 and IFNAR2. Therefore, PMs were infected in the presence When not specifically indicated, data presented in this study are representatives of at least two independent experiments. Every sample for of a neutralizing Ab against type I IFN receptor (anti-IFNAR1). quantitative measurements was analyzed in replicates. For representative Importantly, VSV-induced macrophage glycolysis was abrogated results, mean values were presented (qPCR and metabolic flux: quadru- in IFNAR1-blocked cells (Fig. 1E, Supplemental Fig. 1G, 1H), plicates [6 SD]; lactate: triplicates [6 SD]; viral titer and F2,6BP: du- whereas such inhibition of IFN pathway expectedly caused greater 6 plicates [ range]; the extent of efferocytosis: 10 microscopic fields http://www.jimmunol.org/ [6 SD]). Some experiments were repeated more robustly to show average VSV-G protein accumulation. These results demonstrate a role of values (6 SEM) from independent experiments, and statistical analyses type I IFN, rather than cytosolic viral RNA signaling, directly re- (unpaired t tests) were performed. In some assays where the primary sponsible for VSV-induced glycolysis in macrophages. measurements tend to fluctuate while the trends of change were similar Consistently, rIFN-b treatment of PMs for 8 h increased glucose between independent experiments, paired t tests were performed. In such uptake (Fig. 1F), lactate production (Fig. 1G), as well as the ECAR cases, the graphs presented are still based on representative results. Fur- thermore, for some experiments, statistical analyses were performed to measured by Seahorse metabolic analyzer (Fig. 1H). Contrastingly, compare the values of fold changes. For such type of data presentation, the IFN failed to elicit a hyperglycolytic state in MEFs (Fig. 1F, 1G). To average fold changes from multiple repetitions (6 SEM) are marked on ensure that IFN-induced increase of glycolysis in PMs was not top of the graphs and p values are provided in the figure legends. In ex- caused by secondary effects arising from changes in OXPHOS, we periments involving RSV infection, data from each mouse were deter- by guest on October 2, 2021 mined individually (6 SEM) and Student t tests were performed. also assayed IFN-treated macrophages for the OCR indicative of mitochondria respiration. Despite apparent changes in glycolytic rates, the levels of OCR were comparable between treatment groups Results (Fig. 1I). Collectively, our results have demonstrated that innate VSV infection leads to macrophage-preferential, antiviral signaling engages a type I IFN–dependent, direct induction IFN-dependent activation of glycolysis of a hyperglycolytic state that occurs preferentially in macrophages. Because the oxygen-independent glycolytic pathway represents a This represents a previously unknown, intriguing phenotypic link major point of regulation in cellular glucose metabolism (23, 24), between antiviral immunity and cellular metabolism. we reasoned that antiviral immune signaling events may influence the rate of glycolysis. We tested this hypothesis in mouse Type I IFN induces PFKFB3 to enhance glycolysis in thioglycollate-elicited PMs and MEFs, representing an innate macrophages immune cell type and nonimmune cell type, respectively. PMs and Our analyses of macrophage metabolism were carried out at a time MEFs were infected with VSV (Fig. 1A, 1B), a negatively point when a transcriptional response to IFN is expected to be stranded RNA virus. Both cell types potently engaged innate im- prevalent (8 h). We therefore reasoned that the observed increase in mune signaling and the ensuing transcriptional response, as evi- macrophage glycolysis (Fig. 1F–H) is likely to be attributed to denced by the time-dependent induction of Ifnb1, and of Mx2,a changes in the expression of glycolytic enzymes. Indeed, when downstream target of type I IFN (Fig. 1B). Indeed, clear accu- the mRNA levels of a series of glycolysis-associated proteins/ mulation of VSV-G protein was observed in PMs and more readily enzymes were analyzed, those of PFKFB3, PFK1 (platelet iso- in MEFs, indicative of effective infection (Supplemental Fig. 1B). form), and lactate dehydrogenase A were found to be markedly Notably, at both 5 and 8 h postinfection, we observed time- upregulated in IFN-treated PMs, but not MEFs (Fig. 2A). The dependent increases of glycolysis in PMs, measured by the lev- correlation between gene expression analysis (Fig. 2A) and the els of glycolytic end-product lactate (Fig. 1A). In contrast, no such metabolic measurements of these two cell types (Fig. 1F, 1G) upregulation of glycolysis was observed in parallel using VSV- suggest increased levels of PFKFB3, PFK1, and/or LDH under- infected MEFs (Fig. 1A). It is plausible that the differential gly- lying IFN-dependent glycolytic activation in macrophages. colytic responses to VSV infection in PMs and MEFs may be The dual-activity, PFKFB family of enzymes catalyze the bi- attributed to the macrophage-restricted TLR4-TRAM pathway directional conversion between fructose-6-phosphate and F2,6BP. trigged by the VSV-G envelope protein (25). To test this possi- F2,6BP is an allosteric activator of PFK1, which in turn mediates a bility, we stimulated PMs using UV-inactivated VSV. In contrast key rate-limiting step of glycolysis (Fig. 2B). In the PFKFB family, with the live VSV, its inactivated counterpart was not capable of PFKFB3 exhibits a much higher kinase/phosphatase ratio than inducing the mRNAs of IFN (Supplemental Fig. 1C), possibly other family members (28), essentially mediating net synthesis of The Journal of Immunology 2883 Downloaded from http://www.jimmunol.org/

FIGURE 1. VSV infection leads to macrophage (Mf)-preferential, IFN-dependent enhancement of glycolysis. (A and B)Mfs (peritoneally elicited) and

MEFs were subjected to infection with VSV at MOI of 1. At indicated time points, the culture supernatant (A) and the cell layer (B) were harvested for by guest on October 2, 2021 lactate and mRNA analyses, respectively. The relative levels of mRNA induction (fold) were plotted in (B). (C) PMs were inoculated with live VSV or UV- treated VSV at an MOI of 1 and further incubated for 8 h. (D) PMs were mock-transfected or transfected with poly(I:C) (pI:C) for 4 h and then fed with fresh medium for an additional 6 h. (E) PMs were infected with VSV (or mock-infected) in the presence of control IgG or blocking Ab against IFNAR1 (10 mg/ml) for 8 h. (C–E) Graphs show the results from analyses of extracellular lactate levels. (F–I) PMs and/or MEFs were treated with mIFN-b at 1000 IU/ml for 8 h. (F) Cells were pulsed with 2-NBDG and subjected to flow cytometry. In contrast, the levels of extracellular lactate are plotted in (G). (H and I)IFN- treated PMs were assay for ECAR (H)andOCR(I). Various stressing compounds were added at indicated time points. Mean lactate measurements (6 SEM) were from three to four independent experiments. Other data presented in this figure are representative of at least two independent experiments. Representative quantitative results were graphed using the means of triplicate or quadruplicate measurements (6 SD). Student t tests were performed. Statistical information is marked on the graphs (*p , 0.05, **p , 0.01, ***p , 0.001). N.S., not significant.

F2,6BP to enhance glycolysis (Fig. 2B). Therefore, we placed our To verify that IFN induces PFKFB3 in a macrophage- attention on PFKFB3 because of its upstream regulatory status preferential manner in vivo (experiment scheme in Fig. 2E), we relative to PFK1 and LDH in glycolysis. When the levels of all focused on the lung, a tissue where the macrophages are present four members of the PFKFB family were examined in macro- in a significant portion, together with a diverse array of nonim- phages and MEFs at basal states, the relative expression of Pfkfb3 mune cell types. i.p. injection of poly(I:C) can elicit a robust, mRNA was found to be particularly high in PMs (Fig. 2C). It is systemic type I IFN response in mice (29). Such treatment led to interesting to note that in contrast with that of Pfkfb3, the levels a notable upregulation of PFKFB3 in the lung tissue, together of Pfkfb1, 2, 4 mRNAs were downregulated by IFN in PMs with Stat1 (Fig. 2F). To deplete lung macrophages, we adminis- (Supplemental Fig. 2A). Nevertheless, no further analyses re- tered clodronate liposomes intranasally to the mice. Immuno- garding the latter PFKFB members were carried out in this study. fluorescence staining of macrophage marker F4/80 and qPCR Consistent with the patterns of mRNA expression, PFKFB3 analysis of Cd68 confirmed the efficiency of macrophage depletion protein levels were also notably induced by IFN or VSV in PMs and in the lung tissue (Fig. 2G, Supplemental Fig. 2E). Importantly, BMDMs, but not in MEFs (Fig. 2D, Supplemental Fig. 2B), the liposome treatment markedly reduced i.p. poly(I:C)-induced whereas the induction of a canonical IFN-stimulated gene (ISG), upregulation of PFKFB3 (Fig. 2F). In comparison, upregulation that is, Stat1, was comparable in all cell types. A similar macrophage- of Stat1, a canonical ISG product, was not affected in the lung preferential induction pattern of PFKFB3 by IFN was observed using tissue. Similar contrasting patterns were observed between the human cell lines (Supplemental Fig. 2C). Mechanistically, Jak ac- levels of Pfkfb3 mRNA and those of another canonical ISG, that tivity was required for IFN-dependent induction of PFKFB3 in mac- is, Isg15 (Supplemental Fig. 2E). Collectively, the earlier results rophages (Supplemental Fig. 2D), suggesting Pfkfb3 as a bona fide verified macrophages as a major cell type that uses the IFN- macrophage-restricted ISG. PFKFB3 axis. 2884 GLYCOLYSIS AND INNATE ANTIVIRAL DEFENSE

FIGURE 2. Type I IFN induces glycolytic regulator PFKFB3 (“P.FB3”) in macrophages. (A) PMs and MEFs were treated with mIFN-b for 8 h. The changes of indicated mRNAs were presented. Red dotted lines are used to highlight the basal levels of each mRNA (normalized to “1”) in both cell types. (B) A schematic de- scription of glycolysis and associated metabolic pathways. The irreversible steps are highlighted using red boxes, and activation of PFK1 by PFKFB3 and its product F2,6BP is highlighted (green). (C) The mRNA levels of different isoforms of PFKFB were analyzed in untreated PMs and MEFs (relative to those of PFKFB1).

(D) PMs, BMDMs, and MEFs were treated with Downloaded from IFN for 8 h. Cell lysates were analyzed using Western blotting. Two major bands representing PFKFB3 (marked by a bracket) or nonspecific signal (N.S.) were indicated. (E–G) Mice were intranasally instilled with PBS or clodronate liposome (CLOD). Two days later, the mice were injected i.p. with 50 mg of poly(I:C) (E). http://www.jimmunol.org/ Twelve hours later, lungs were harvested from the mice and subjected to Western blot (F) and immunofluorescence (G). Data presented in this figure are representative of at least two independent experiments. The qPCR analyses were graphed using the means of quadruplicate measurements (6 SD). by guest on October 2, 2021

To functionally validate the role of PFKFB3 in macrophage bolic phenotypes correlate with the levels of PFKFB3 protein metabolism, we first used a stable macrophage cell line (Raw264.7) (Fig. 3E, right panel, Supplemental Fig. 3E). Despite such an ap- transduced with a tetracycline-inducible PFKFB3 expression parent change in glycolysis (∼25% in lactate levels), the Pfkfb3+/2 cassette (Fig. 3A, 3B). Doxycycline treatment induced PFKFB3 macrophages did not exhibit evident changes in cellular respira- expression to a similar extent achieved by IFN in primary mac- tion (Fig. 3G, Supplemental Fig. 3G). Therefore, our results confirm rophages (see Fig. 2D) and led to upregulation of glycolysis PFKFB3 as an important glycolytic regulator in macrophages. (Fig. 3A). To examine the metabolic function of endogenous Compared with the macrophages, Pfkfb3+/2 MEFs showed a more PFKFB3, we generated heterozygous Pfkfb3-knockout mice. modest (∼13%) decrease in overall glycolytic rates (Supplemental Cas9-mediated cleavage caused a 73-bp indel in exon 2 of Pfkfb3 Fig. 3H), likely attributed to a lack of specific enrichment of PFKFB3 that led to frameshift in the mutant allele (Supplemental Fig. 3A). (among all PFKFBs) (Fig. 2C). Conversely, forced expression of Accordingly, the levels of Pfkfb3 mRNA and its encoded protein PFKFB3 in MEFs increased the basal glycolytic rates, while the in multiple tissues from the heterozygous mutant mice were vis- cells were still unable to enhance glycolysis in response to IFN ibly reduced (Fig. 3C, 3D). Pfkfb3+/2 mice were morphologically treatment (Supplemental Fig. 3I). normal, consistent with a previous report (30). In addition, equivalent numbers of BMDMs or PMs could be derived from the PFKFB3-driven glycolysis promotes efferocytosis-dependent, WT and the Pfkfb3+/2 mice, and the cells showed similar patterns cell-extrinsic, antiviral activity in macrophages of macrophage-associated cell-surface markers (Supplemental The macrophage-preferential, IFN-dependent induction of PFKFB3 Fig. 3B, 3C). Furthermore, compared with the WT counterparts, hinted its role in regulating viral resistance of this particular cell Pfkfb3+/2 BMDMs engaged similar induction of Ifnb1 and two type. In contrast, PFKFB3-driven glycolysis may paradoxically ISGs (Mx2 and Isg15) in response to transfected dsRNA, showing play a proviral role via fueling biosynthesis and other critical steps an intact IFN system (Supplemental Fig. 3D). Nevertheless, of viral life cycles (31–35). Interestingly, Pfkfb3 heterozygosity Pfkfb3+/2 macrophages exhibited a visible reduction in glycolysis, in BMDMs or MEFs did not substantially affect their viral measured by lactate levels (Fig. 3E, left panel, Supplemental Fig. burdens (Fig. 4A, 4B, Supplemental Fig. 4A), indicating that a 3E) and by ECAR (Fig. 3F, Supplemental Fig. 3F). These meta- moderate reduction of glycolysis did not suffice to markedly The Journal of Immunology 2885

FIGURE 3. PFKFB3 enhances glycolysis in macrophages. (A and B) Raw264.7 macrophages (Mfs) containing an inducible cassette of PFKFB3 were treated with doxycycline (DOX; 2 mg/ml) for 16 h. Levels of extracellular lactate (A) and PFKFB3 protein (B) were analyzed. (C and D) Indicated tissue samples from WT and Pfkfb3+/2 mice were subjected to qPCR (C) and Western blotting (D). Tail DNA samples were genotyped (inset in C). (E–G) Indicated BMDMs were treated with IFN for 8 h. Levels of the extracellular lactate and indicated cellular proteins (E), ECAR (F), or OCR (G) were determined. Data presented are representative of at least two independent experi- Downloaded from ments. Quantitative data were graphed using the means of triplicate or quadruplicate measurements (6 SD). (A and E) Red dotted lines are used to highlight the control levels of lactate measurements within each experiment. http://www.jimmunol.org/

impact either VSV replication or the intrinsic antiviral activities VSV production in either cell type alone (Fig. 4G, 4H), similar in these two cell types. treatment of BMDM/infected MEFs coculture reproducibly resulted by guest on October 2, 2021 Compared with the nonimmune MEFs, macrophages may ad- in a higher overall viral burden (Fig. 4I). In contrast, no changes in ditionally exert cell-extrinsic antiviral function, where they can IFN pathway were apparent (Supplemental Fig. 4D). These results restrict viral load in a non–cell-autonomous manner (36, 37). We correlate well with the data from genetically modified BMDMs therefore considered an additional possibility that PFKFB3 may (Fig. 4A–D), and in conjunction present a model where PFKFB3- regulate the cell-extrinsic antiviral activities of macrophages. To driven glycolysis promotes macrophage-extrinsic antiviral activities this end, the WT or Pfkfb3+/2 BMDMs were incubated directly without affecting IFN pathway. with replicates of virus-infected C57/BL6 MEFs. After overnight Through their potent phagocytic activity, macrophages remove coculture, the supernatant was subsequently analyzed for viral the pathogen particles as well as host-derived apoptotic cells (40– titer. As a control, incubation of naive MEFs with virus-infected 42). The latter is referred to as efferocytosis. Efferocytosis of MEFs fueled secondary infections, causing further viral produc- virus-infected, apoptotic cells may serve as an important cell- tion (data not shown). In contrast, coculture of BMDMs with in- extrinsic antiviral mechanism to prevent viral spread (43–45). fected MEFs leads to a substantial decrease of released virus, Indeed, when the WT or Pfkfb3+/2 BMDMs were incubated with reflecting an extrinsic antiviral activity by the BMDMs (Fig. 4C). VSV-infected MEFs (labeled with CSFE) and subjected to im- Importantly, WT BMDMs exhibited a greater cell-extrinsic anti- munofluorescence microscopy, a higher content of MEF corpses viral activity compared with the Pfkfb3+/2 BMDMs (Fig. 4C). was found within the WT BMDMs (Fig. 5A). Consistent results Contrastingly, the induction of Ifnb1 and several ISGs was com- were obtained when engulfment of infected MEFs were quantified parable between the two coculture groups (Fig. 4D). When the via flow cytometry (Supplemental Fig. 4E). To prevent effer- IFN responses of VSV-infected BMDM monocultures were di- ocytosis of infected MEFs by BMDMs, we first physically seg- rectly measured, the WT and Pfkfb3+/2 BMDMs were found to regated the two cell types using permeable membranes (3-mm secrete comparable levels of IFN-b protein (Fig. 4E) and to en- pore-sized transwell culture system) that otherwise allowed for gage similar upregulation of ISG mRNAs (Fig. 4F). Furthermore, free movement of soluble factors and viruses. Consequently, the despite previous suggestions of NO as a potential effector medi- cell-extrinsic antiviral effects by BMDMs were markedly reduced ating macrophage cell-extrinsic antiviral activity (38), we did not and viral burden became comparable between the WT and mutant detect its enhanced production by BMDMs in the coculture system BMDM coculture groups (Fig. 5B). Similar results were also (data not shown). obtained comparing the direct or transwell-separated coculture of We next used a complementary pharmacological approach PMs with infected MEFs (Supplemental Fig. 4F). to target the catalytic activity of PFKFB3 in vitro. A recently We next sought for a more specific method of efferocytosis established PFKFB3 inhibitor, that is, PFK15 (39), visibly inhib- inhibition. Macrophages recognize apoptotic cells via multiple ited 2-NBDG uptake in BMDMs and MEFs (Supplemental Fig. 4B, redundant phosphatidylserine receptors including BAI1, TIM4, and 4C). Although an effective dose (10 mM) of PFK15 did not affect STABILIN-2 whose expression patterns vary in different types of 2886 GLYCOLYSIS AND INNATE ANTIVIRAL DEFENSE

FIGURE 4. PFKFB3 promotes the cell-extrinsic antiviral defense of macrophages. (A) WT and Pfkfb3+/2 BMDMs were infected with VSV at an MOI of 5 for 12 h. The viral titers in the culture supernatants were determined. Cell lysates were analyzed by Western blotting. (B) WT and Pfkfb3+/2 MEFs were infected with VSV at an MOI of 1 for 14 h and analyzed as in (A). (C and D) WT MEFs were infected as in (B). Unattached and loosely attached cells were collected and added at a 2:1 ratio to wells containing (or without) indicated BMDMs. In 12 h, the coculture supernatants and RNA samples were analyzed for viral titer (C) and by qPCR (D). (E and F) WT and Pfkfb3+/2 BMDM monocultures (three biological replicates) were infected as in (A). (E) Culture supernatants were then harvested for ELISA analyses of IFN-b (6 SD). (F) RNA samples from the infected cells were analyzed for G I markers of IFN pathway. ( – ) BMDMs or MEFs were Downloaded from treated, respectively, as in (A), (B), and (C), except that the cells were pretreated with or without PFK15 (10 mM). Data presented are representative of three independent experiments. Viral titers were graphed using the means of duplicate measurements (6 range). Paired Student t tests were performed on the viral titer quantitations and statistical information is marked on http://www.jimmunol.org/ the graphs. N.S., not signiﬁcant.

macrophages (41, 42). However, we took advantage of the fact driven glycolysis may be coupled to efferocytosis-associated sig- that TIM4 functions as the major phosphatidylserine receptor in naling to provide instant energy support (i.e., rapid generation of BMDMs (46, 47). Indeed, we verified that a blocking Ab against ATP) for cargo engulfment. The latter hypothesis was supported by guest on October 2, 2021 TIM4 notably reduced BMDM-mediated efferocytosis activity by the fact that efferocytosis is delicately regulated by phospho- against apoptotic thymocytes (Fig. 5C), as well as against VSV- lipid signaling activators including PI3K (40, 47), which is known infected MEFs (Supplemental Fig. 4G). Importantly, Ab-mediated to engage glycolysis (23). As expected, addition of apoptotic TIM4 blockage not only moderately inhibited the cell-extrinsic thymocytes into a PM culture led to a time-dependent phosphor- antiviral activity of BMDMs, but also reduced the difference in ylation of AKT and its target PRAS40, indicative of PI3K acti- viral burden between the WT and Pfkfb3+/2 coculture groups vation (Fig. 5G). Importantly, such treatment of PMs also resulted (Fig. 5D). Collectively, our data establish that PFKFB3 promotes in a PI3K-dependent increase in the levels of F2,6BP (Fig. 5H), the cell-extrinsic antiviral effector function of macrophages in an possibly resulting from AKT-mediated phosphorylation/activation efferocytosis-dependent manner. of PFKFB3 as suggested previously (51). Correlating with F2,6BP Next, to further generalize the contribution of PFKFB3 to measurements, the levels of 2-NBDG uptake were also upregu- efferocytosis, the WT and Pfkfb3+/2 macrophages were added lated after initiation of efferocytosis, in a PI3K-dependent manner with standard cargo of apoptotic thymocytes. Indeed, Pfkfb3+/2 (Fig. 5I). Furthermore, consistent with a role of IFN-induced BMDMs exhibited a .2-fold decrease in basal efferocytosis ac- PFKFB3 to promote such metabolic engagement, IFN pretreat- tivities compared with WT BMDMs, which was not attributed to ment could further increase apoptotic thymocyte-triggered 2-NBDG their differences in cargo adherence (Supplemental Fig. 4H). A uptake (Supplemental Fig. 4J). These results strongly support similar efferocytosis defect was observed in Pfkfb3+/2 PMs that PFKFB3-driven glycolysis provides a metabolic basis for (Supplemental Fig. 4I). Although IFN treatment of PMs markedly optimal efferocytosis and serves as a critical regulatory node. In increased their efferocytosis activity as shown previously (45, 48), contrast, despite a recent report suggesting a role of PFKFB3 as IFN-treated Pfkfb3+/2 PMs were still less potent in efferocytosis a feed-forward regulator of PI3K signaling (52), no significant than similarly treated WT PMs (Fig. 5E). Furthermore, the defects in the latter pathway were observed in Pfkfb3+/2 PMs efferocytosis activities of PMs were respectively enhanced and (Supplemental Fig. 4K). suppressed by established hyperglycolytic (high glucose culture medium) and hypoglycolytic (2-deoxyglucose treatment) condi- PFKFB3 exhibits antiviral function in vivo tions (Fig. 5F), clearly establishing a role of glycolytic meta- Our in vitro data thus far showed the antiviral, immunometabolic bolism in efferocytosis. aspects of PFKFB3-driven glycolysis. To establish the in vivo role Particle engulfment after phagocytic receptor activation involves of PFKFB3 under viral infection, we used a mouse model of RSV rapid extension of cellular processes to eventually enclose the infection. It was shown previously that lung macrophages con- cargos. Subcellularly, such a dynamic sequence involves active tributed to controlling the initial RSV load (53–55). Because RSV polymerization of actin filaments that is an energy-demanding mainly replicates in epithelial cells (55), this in vivo model was process (49, 50). Therefore, we hypothesized that PFKFB3- conducive for examining the cell-extrinsic antiviral function of The Journal of Immunology 2887 Downloaded from http://www.jimmunol.org/

FIGURE 5. PFKFB3-driven glycolysis metabolically supports macrophage-mediated efferocytosis of infected cells to limit viral propagation. (A) VSV- infected, CFSE-labeled MEFs (green) were added to BMDM cultures at a ratio of 5:1. In 4 h, the cells were fixed and subjected to immunofluorescence by guest on October 2, 2021 analysis (white: F4/80). Scale bars, 50 mm. (B) BMDMs were cocultured with infected MEFs as in Fig. 4C. Alternatively, the infected MEFs were segregated from the BMDMs by 3-mm pore-sized transwell membranes. The viral titers were determined. (C and D) BMDMs were preincubated with IgG or anti-TIM4 (100 mg/ml) and were then subjected to efferocytosis assay using apoptotic thymocytes (C), or extrinsic antiviral assay (D) as in Fig. 4C. (D) Mean fold changes (FC) from three independent experiments are marked (6 SEM). A p value ,0.0001 was achieved when comparing the viral production fold changes between groups. (E and F) WT or mutant PMs were pretreated with or without IFN as indicated (E). (F) PMs were cultured under indicated metabolic conditions for 2 h. PMs were then subjected to efferocytosis assay using apoptotic thymocytes. (G–I) Serum-starved PMs that had undergone pretreatment with or without wortmannin as indicated (200 nM) were added with apoptotic thymocytes (A.T., 20:1). (G) Cell lysates were harvested at various time points and analyzed by Western blotting. (H and I) Cells were harvested in 1 h and were either lysed for determination of F2,6BP levels (H)or were subjected to assay of 2-NBDG uptake (I). (H) Lysate from LPS-treated cells was used as positive control (“POS ctrl”). All graphs are plotted based on representative experiments. Viral titers and F2,6BP levels were graphed using the means of duplicate measurements (6 range). The levels of efferocytosis were graphed using the mean counts from 10 microscopic fields (6 SD). In particular, F2,6BP in cells treated with A.T. has been measured in three independent experiments (p , 0.05 by paired t test). 2-DG, 2-deoxyglucose. macrophages. WT and Pfkfb3+/2 mice (n = 5 for each group) were induction of Cd68 mRNA indicative of infiltrated macrophages challenged with RSV and the lung tissues were harvested 5 d later. (Fig. 6F). We notice that the differences in viral susceptibility 2 Importantly, the Pfkfb3+/ mice had significantly higher viral between the WT and Pfkfb3+/2 bone marrow–chimeric mice were burdens compared with the WT mice (Fig. 6A, 6B), clearly less significant than those from nontransplanted mice (compare demonstrating the antiviral function of PFKFB3 in vivo. In com- Fig. 6A and 6D). It is possible that the chosen recovery period parison, the markers indicative of macrophages were elevated to after bone marrow transplantation (6 wk) was not yet sufficient for similar extent in the WT and mutant lungs (Fig. 6C). Interestingly, complete reconstitution of resident alveolar macrophages (56), the mRNAs of several ISGs were moderately higher in the mutant where PFKFB3 may exert initial cell-extrinsic antiviral function. lungs (Fig. 6C), likely to be resulting from a persistent, greater Nevertheless, these in vivo results corroborate our in vitro data and viral load. supported a role of PFKFB3 in promoting cell-extrinsic antiviral Lastly, we sought to verify the contribution of hematopoietic effector function of innate immune cells. compartment-originated PFKFB3 to viral resistance by using WT or Pfkfb3+/2 bone marrow chimeras. The mice reconstituted with Discussion 2 either the WT or Pfkfb3+/ bone marrow were subsequently sub- Viruses hijack the host cells’ synthetic machineries to support jected to RSV challenge (Fig. 6D–F). Indeed, Pfkfb3 heterozy- various stages of their life cycles. Because pathways of energy gosity in the hematopoietic compartment caused a visible increase metabolism can regulate biosynthesis, as well as various steps in viral burden (Fig. 6D, 6E), without apparently affecting the associated with viral replication/assembly/propagation, they are 2888 GLYCOLYSIS AND INNATE ANTIVIRAL DEFENSE

FIGURE 6. PFKFB3 exhibits antiviral function in vivo. (A–C) Seven-week-old male mice were inoculated intranasally with 5 3 106 PFU of RSV- A2 (n = 5). Five days later, the lung RSV titer (A)or some viral/cellular mRNAs (B and C) in individual lung tissue was determined (6 SEM). (D–F)WTor Pfkfb3+/2 bone marrow cells were transferred into lethally irradiated WT mice. The mice were then subjected to RSV challenge (n = 8) as done earlier. The viral titer (D) and levels of RSV-N mRNA (E) and of indicated cellular mRNAs in lung or bone marrow tissues (F) were analyzed. RSV-N mRNA levels relative to those of 18S rRNA (3 a conversion factor of 106) were presented in arbitrary units (a.u.). **p , 0.005 (B), *p , 0.05 (E). Downloaded from

believed to significantly influence the infectivity of intruding efferocytosis process involves active actin filaments assembly http://www.jimmunol.org/ viruses (31–35). However, few studies have explored whether an (41, 42), where PFKFB3-driven glycolytic reaction may be in- optimal antiviral innate immune program also requires significant volved to supply the required ATP, similar to an earlier model for metabolic inputs. endothelial tip cell migration (50). Our data extend the latter In this study, we have provided evidence that cytoplasmic viral model to further suggest that PFKFB3 actively participates in recognition, via the action of type I IFN, induces a PFKFB3-driven efferocytosis via establishing the temporal and spatial associa- hyperglycolytic state in primary macrophages, but not in the tion between glycolytic metabolism and actin polymerization nonimmune MEFs (Figs. 1, 2). Such a macrophage-preferential (Fig. 5G–I). In addition to metabolically supporting effer- effect by IFN is consistent with a model of “immune-centric” ocytosis, PFKFB3 may possibly also affect other downstream by guest on October 2, 2021 metabolic reprogramming that serves to enhance the functions of events. Interestingly, efferocytosis was previously shown to in- innate immune cells. Importantly, despite an apparent effect on teract with the autophagic pathway (46). Because autophagy is glycolysis, the IFN-PFKFB3 axis did not affect cellular respira- intricately connected to the cells’ metabolic states (59) and it is tion. This is different from a recently described scenario in den- well-known to control viral resistance (60), PFKFB3-regulated dritic cells where a dominant inhibition of OXPHOS coupled with extrinsic antiviral activity may potentially involve changes in a corresponding increase of glycolysis upon in vivo poly(I:C) autophagic pathway. Such a possibility awaits future investiga- stimulation (11). Given the known functional diversity of IFN tion. Nevertheless, our study has provided strong evidence for a (57), the macrophage-preferential IFN-PFKFB3 axis therefore rather intuitive model linking PFKFB3-driven glycolysis to represents a unique system to study the delicate cross-talks be- macrophage phagocytic behavior. Such a new immunometabolic tween metabolism and immunity. role by PFKFB3 complements previous reports that a TLR4- Methodologically, we exploited a convenient Pfkfb3 hetero- PFKFB3 axis modulates the M1 macrophage gene expression zygotic knockout mouse model (Fig. 3), which was morphologi- circuitry (7, 61). It is also worth noting that because most TLRs cally normal, unlike the homozygous mutants (30). Despite a can trigger IFN production to varying degrees (1), the IFN- moderate reduction (∼25%) in glycolysis, single allele loss of PFKFB3 axis may be commonly engaged by TLRs. Our data Pfkfkb3 in macrophages resulted in little compensatory changes in based on a TLR3 activation model indeed support such a notion cellular respiration. An even lesser decrease in glycolysis was (Supplemental Fig. 2F). observed in Pfkfb3+/2 MEFs. These results are consistent with the In retrospect, it is interesting to note that the lipid metabolic role of PFKFB3 as a regulator, rather than a mediator, of gly- pathways were previously known as critical regulators of the colysis (50, 58) and in conjunction validate Pfkfb3+/2 mice/cells efferocytosis process (62). In contrast, glucose metabolism has as a straightforward partial loss-of-function model. been surprisingly underinvestigated in the field. Interestingly, Functionally, PFKFB3 selectively regulates efferocytosis- different from our observation, an earlier study using a phagocytic dependent, macrophage-specialized extrinsic antiviral activity Chinese hamster ovary cell line showed that high concentration of (Figs. 4, 5), which correlates well with its macrophage-preferential glucose had a negative effect on efferocytosis (63). We speculate induction by IFN. In addition to preventing viral propagation, that the cellular contexts, that is, macrophages versus other cell efferocytosis of infected or damaged cells is also likely to contribute types, primary versus transformed cells, or even differences in mac- to limiting virus-induced immunopathology via physical actions rophage activation states, may markedly influence the metabolism/ and secretion of anti-inflammatory cytokines (41, 42). Pfkfb3+/2 efferocytosis connections. Because different cell identities/fates are macrophages exhibited lower efferocytosis ability, suggesting that characterized by distinctive metabolic landscapes (24), the earlier the events leading to cargo engulfment are critically sensitive to discrepancies in others’ and our findings may suggest an intriguing changes in PFKFB3 levels and glycolysis. Mechanistically, the possibility that interactions among various metabolic pathways The Journal of Immunology 2889 coordinately control cells’ efferocytosis behaviors. The latter early glycolytic reprogramming via the kinases TBK1-IKKε supports the ana- bolic demands of dendritic cell activation. Nat. Immunol. 15: 323–332. appears as an exciting avenue for future research. 11. Pantel, A., A. Teixeira, E. Haddad, E. G. Wood, R. M. Steinman, and By using an RSV infection mouse model, we have confirmed the M. P. Longhi. 2014. Direct type I IFN but not MDA5/TLR3 activation of den- antiviral function of PFKFB3, and at least some contribution of its dritic cells is required for maturation and metabolic shift to glycolysis after poly IC stimulation. PLoS Biol. 12: e1001759. action from the hematopoietic compartment (Fig. 6), in agreement 12. Palsson-McDermott, E. M., A. M. Curtis, G. Goel, M. A. Lauterbach, with our in vitro results. Even though the RSV susceptibility phe- F. J. Sheedy, L. E. Gleeson, M. W. van den Bosch, S. R. Quinn, R. Domingo- notype associated with Pfkfb3 heterozygosity was moderate, it Fernandez, D. G. Johnston, et al. 2015. Pyruvate kinase M2 regulates Hif-1a activity and IL-1b induction and is a critical determinant of the warburg effect in demonstrates that tuning the levels/activities of PFKFB3 is suffi- LPS-activated macrophages. [Published erratum appears in 2015 Cell. Metab. cient to alter the virus–host balance. It is formally possible that 21: 347.] Cell Metab. 21: 65–80. PFKFB3 may exert other uncharacterized antiviral functions de- 13. Zhang, X., R. Goncalves, and D. M. Mosser. 2008. The isolation and charac- terization of murine macrophages. Curr. Protoc. Immunol. Chapter 14: Unit 14.1. pendent on the viral and cellular contexts. Further efforts to test the DOI: 10.1002/0471142735.im1401s83. latter notion in diverse viral infection models are highly warranted. 14. Xu, J. 2001. Preparation, Culture, and Immortalization of Mouse Embryonic Nevertheless, our data have provided genetic evidence for an anti- Fibroblasts. Curr. Protoc. Mol. Biol. Chapter 28: Unit 28.1. DOI: 10.1002/ 0471142727.mb2801s70. viral aspect of glycolysis at the organism level, a significant finding 15. Du, Y., Q. Meng, J. Zhang, M. Sun, B. Shen, H. Jiang, N. Kang, J. Gao, considering that many previous reports have suggested glycolysis as X. Huang, and J. Liu. 2015. Functional annotation of cis-regulatory elements in human cells by dCas9/sgRNA. Cell Res. 25: 877–880. a proviral pathway (34, 35). In this regard, it is worth investigating 16. Liu, J., W. C. HuangFu, K. G. Kumar, J. Qian, J. P. Casey, R. B. Hamanaka, whether the featured expression of PFKFB3 (compared with other C. Grigoriadou, R. Aldabe, J. A. Diehl, and S. Y. Fuchs. 2009. Virus-induced PFKFB isoforms; Fig. 2C) in macrophages, an innate immune cell unfolded protein response attenuates antiviral defenses via phosphorylation- dependent degradation of the type I interferon receptor. Cell Host Microbe 5: type, may contribute to its predominant role in immunometabolism, 72–83. Downloaded from but not virus-promoting metabolism. 17. Jiang, H., Y. Lu, L. Yuan, and J. Liu. 2011. Regulation of interleukin-10 receptor Overall, our work has established PFKFB3-driven glycolysis as a ubiquitination and stability by beta-TrCP-containing ubiquitin E3 ligase. PLoS One 6: e27464. critical metabolic effector mechanism of innate antiviral immunity. 18. Tong, Y., F. Li, Y. Lu, Y. Cao, J. Gao, and J. Liu. 2013. Rapamycin-sensitive Importantly, the in vivo data from this study challenge the sim- mTORC1 signaling is involved in physiological primordial follicle activation in plified, virus-central view of cellular metabolism and highlight the mouse ovary. Mol. Reprod. Dev. 80: 1018–1034. 19. Todt, J. C., B. Hu, and J. L. Curtis. 2004. The receptor tyrosine kinase MerTK significance of immunometabolism in innate antiviral defense. activates phospholipase C gamma2 during recognition of apoptotic thymocytes Interestingly, PFKFB3 was also suggested previously as a positive by murine macrophages. J. Leukoc. Biol. 75: 705–713. http://www.jimmunol.org/ 20. Van Schaftingen, E., B. Lederer, R. Bartrons, and H. G. Hers. 1982. A kinetic regulator of T cell functions (64). Therefore, we suggest that study of pyrophosphate: fructose-6-phosphate phosphotransferase from potato PFKFB3 may represent a promising drug target whose therapeutic tubers. Application to a microassay of fructose 2,6-bisphosphate. Eur. J. Bio- activation may enhance both the innate and the adaptive arms of chem. 129: 191–195. 21. Spangrude, G. J. 2008. Assessment of lymphocyte development in radiation antiviral response. bone marrow chimeras. Curr. Protoc. Immunol. Chapter 4: Unit 4.6. 22. Guo, X., T. Liu, H. Shi, J. Wang, P. Ji, H. Wang, Y. Hou, R. X. Tan, and E. Li. 2015. Respiratory syncytial virus infection upregulates NLRC5 and major his- Acknowledgments tocompatibility complex class I expression through RIG-I induction in airway We are grateful to the laboratories of G. Liu and Q. Zhang (Nanjing Univ.) epithelial cells. J. Virol. 89: 7636–7645. for technical help. We thank the Nanjing Biomedical Research Institute- 23. Vander Heiden, M. G., L. C. Cantley, and C. B. Thompson. 2009. Understanding by guest on October 2, 2021 NJU and the animal facility of NJU Medical School for excellent mouse the Warburg effect: the metabolic requirements of cell proliferation. Science 324: 1029–1033. services. We acknowledge the Collaborative Liver Disease Research Pro- 24. Ghesquie`re, B., B. W. Wong, A. Kuchnio, and P. Carmeliet. 2014. Metabolism of gram of NJU Medical School for instrumental support. stromal and immune cells in health and disease. Nature 511: 167–176. 25. Georgel, P., Z. Jiang, S. Kunz, E. Janssen, J. Mols, K. Hoebe, S. Bahram, M. B. Oldstone, and B. Beutler. 2007. Vesicular stomatitis virus glycoprotein G Disclosures activates a specific antiviral Toll-like receptor 4-dependent pathway. Virology The authors have no financial conflicts of interest. 362: 304–313. 26. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui, S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, et al. 2006. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441: 101– References 105. 1. Takeuchi, O., and S. Akira. 2010. Pattern recognition receptors and inflamma- 27. Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi, tion. Cell 140: 805–820. M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA helicase RIG-I 2. Platanias, L. C. 2005. Mechanisms of type-I- and type-II-interferon-mediated has an essential function in double-stranded RNA-induced innate antiviral re- signalling. Nat. Rev. Immunol. 5: 375–386. sponses. Nat. Immunol. 5: 730–737. 3. MacMicking, J. D. 2012. Interferon-inducible effector mechanisms in cell- 28. Mor, I., E. C. Cheung, and K. H. Vousden. 2011. Control of glycolysis through autonomous immunity. Nat. Rev. Immunol. 12: 367–382. regulation of PFK1: old friends and recent additions. Cold Spring Harb. Symp. 4. Pearce, E. J., and B. Everts. 2015. Dendritic cell metabolism. Nat. Rev. Immunol. Quant. Biol. 76: 211–216. 15: 18–29. 29. Rohlmann, A., M. Gotthardt, R. E. Hammer, and J. Herz. 1998. Inducible 5. Kelly, B., and L. A. O’Neill. 2015. Metabolic reprogramming in macrophages inactivation of hepatic LRP gene by cre-mediated recombination confirms and dendritic cells in innate immunity. Cell Res. 25: 771–784. role of LRP in clearance of chylomicron remnants. J. Clin. Invest. 101: 689– 6. Krawczyk, C. M., T. Holowka, J. Sun, J. Blagih, E. Amiel, R. J. DeBerardinis, 695. J. R. Cross, E. Jung, C. B. Thompson, R. G. Jones, and E. J. Pearce. 2010. Toll- 30. Chesney, J., S. Telang, A. Yalcin, A. Clem, N. Wallis, and R. Bucala. 2005. like receptor-induced changes in glycolytic metabolism regulate dendritic cell Targeted disruption of inducible 6-phosphofructo-2-kinase results in embryonic Biochem. Biophys. Res. Commun. activation. Blood 115: 4742–4749. lethality. 331: 139–146. 7. Rodrı´guez-Prados, J. C., P. G. Trave´s, J. Cuenca, D. Rico, J. Aragone´s, P. Martıń- 31. Munger, J., S. U. Bajad, H. A. Coller, T. Shenk, and J. D. Rabinowitz. 2006. Dynamics of the cellular metabolome during human cytomegalovirus infection. Sanz, M. Cascante, and L. Bosca´. 2010. Substrate fate in activated macrophages: PLoS Pathog. 2: e132. a comparison between innate, classic, and alternative activation. J. Immunol. 32. Munger, J., B. D. Bennett, A. Parikh, X.-J. Feng, J. McArdle, H. A. Rabitz, 185: 605–614. T. Shenk, and J. D. Rabinowitz. 2008. Systems-level metabolic flux profiling 8. Haschemi, A., P. Kosma, L. Gille, C. R. Evans, C. F. Burant, P. Starkl, B. Knapp, identifies fatty acid synthesis as a target for antiviral therapy. Nat. Biotechnol. R. Haas, J. A. Schmid, C. Jandl, et al. 2012. The sedoheptulose kinase CARKL 26: 1179–1186. directs macrophage polarization through control of glucose metabolism. Cell 33. Vastag, L., E. Koyuncu, S. L. Grady, T. E. Shenk, and J. D. Rabinowitz. 2011. Metab. 15: 813–826. Divergent effects of human cytomegalovirus and herpes simplex virus-1 on 9. Tannahill, G. M., A. M. Curtis, J. Adamik, E. M. Palsson-McDermott, cellular metabolism. PLoS Pathog. 7: e1002124. A. F. McGettrick, G. Goel, C. Frezza, N. J. Bernard, B. Kelly, N. H. Foley, et al. 34. Sanchez, E. L., and M. Lagunoff. 2015. Viral activation of cellular metabolism. 2013. Succinate is an inflammatory signal that induces IL-1b through HIF-1a. Virology 479–480: 609–618. Nature 496: 238–242. 35. Goodwin, C. M., S. Xu, and J. Munger. 2015. Stealing the keys to the kitchen: 10. Everts, B., E. Amiel, S. C. Huang, A. M. Smith, C. H. Chang, W. Y. Lam, viral manipulation of the host cell metabolic network. Trends Microbiol. 23: V. Redmann, T. C. Freitas, J. Blagih, G. J. van der Windt, et al. 2014. TLR-driven 789–798. 2890 GLYCOLYSIS AND INNATE ANTIVIRAL DEFENSE

36. Stohlman, S. A., J. G. Woodward, and J. A. Frelinger. 1982. Macrophage anti- 51. Humphrey, S. J., G. Yang, P. Yang, D. J. Fazakerley, J. Sto¨ckli, J. Y. Yang, and viral activity: extrinsic versus intrinsic activity. Infect. Immun. 36: 672–677. D. E. James. 2013. Dynamic adipocyte phosphoproteome reveals that Akt di- 37. Mogensen, S. C. 1985. Genetic aspects of macrophage involvement in natural rectly regulates mTORC2. Cell Metab. 17: 1009–1020. resistance to virus infections. Immunol. Lett. 11: 219–224. 52. Xu, Y., X. An, X. Guo, T. G. Habtetsion, Y. Wang, X. Xu, S. Kandala, Q. Li, 38. Benencia, F., and M. C. Courreges. 1999. Nitric oxide and macrophage antiviral H. Li, C. Zhang, et al. 2014. Endothelial PFKFB3 plays a critical role in an- extrinsic activity. Immunology 98: 363–370. giogenesis. Arterioscler. Thromb. Vasc. Biol. 34: 1231–1239. 39. Clem, B. F., J. O’Neal, G. Tapolsky, A. L. Clem, Y. Imbert-Fernandez, 53. Pribul, P. K., J. Harker, B. Wang, H. Wang, J. S. Tregoning, J. Schwarze, and D. A. Kerr, II, A. C. Klarer, R. Redman, D. M. Miller, J. O. Trent, et al. 2013. P. J. Openshaw. 2008. Alveolar macrophages are a major determinant of early Targeting 6-phosphofructo-2-kinase (PFKFB3) as a therapeutic strategy against responses to viral lung infection but do not influence subsequent disease de- cancer. Mol. Cancer Ther. 12: 1461–1470. velopment. J. Virol. 82: 4441–4448. 40. Flannagan, R. S., V. Jaumouille´, and S. Grinstein. 2012. The cell biology of 54. Kolli, D., M. R. Gupta, E. Sbrana, T. S. Velayutham, H. Chao, A. Casola, and phagocytosis. Annu. Rev. Pathol. 7: 61–98. R. P. Garofalo. 2014. Alveolar macrophages contribute to the pathogenesis of 41. Elliott, M. R., and K. S. Ravichandran. 2010. Clearance of apoptotic cells: human metapneumovirus infection while protecting against respiratory syncytial implications in health and disease. J. Cell Biol. 189: 1059–1070. virus infection. Am. J. Respir. Cell Mol. Biol. 51: 502–515. 42. Poon, I. K., C. D. Lucas, A. G. Rossi, and K. S. Ravichandran. 2014. Apoptotic 55. Goritzka, M., S. Makris, F. Kausar, L. R. Durant, C. Pereira, Y. Kumagai, cell clearance: basic biology and therapeutic potential. Nat. Rev. Immunol. 14: F. J. Culley, M. Mack, S. Akira, and C. Johansson. 2015. Alveolar macrophage- 166–180. derived type I interferons orchestrate innate immunity to RSV through recruit- 43. Fujimoto, I., J. Pan, T. Takizawa, and Y. Nakanishi. 2000. Virus clearance ment of antiviral monocytes. J. Exp. Med. 212: 699–714. through apoptosis-dependent phagocytosis of influenza A virus-infected cells by 56. Matute-Bello, G., J. S. Lee, C. W. Frevert, W. C. Liles, S. Sutlief, K. Ballman, macrophages. J. Virol. 74: 3399–3403. V. Wong, A. Selk, and T. R. Martin. 2004. Optimal timing to repopulation of 44. Hashimoto, Y., T. Moki, T. Takizawa, A. Shiratsuchi, and Y. Nakanishi. 2007. resident alveolar macrophages with donor cells following total body irradiation J. Immunol. Methods Evidence for phagocytosis of influenza virus-infected, apoptotic cells by neu- and bone marrow transplantation in mice. 292: 25–34. 57. Trinchieri, G. 2010. Type I interferon: friend or foe? J. Exp. Med. 207: 2053– trophils and macrophages in mice. J. Immunol. 178: 2448–2457. 2063. 45. Yańguez,€ E., A. Garcıá-Culebras, A. Frau, C. Llompart, K.-P. Knobeloch, 58. Xie, N., Z. Tan, S. Banerjee, H. Cui, J. Ge, R. M. Liu, K. Bernard, S. Gutierrez-Erlandsson, A. Garcıá-Sastre, M. Esteban, A. Nieto, and S. Guerra. V. J. Thannickal, and G. Liu. 2015. Glycolytic reprogramming in myofibroblast 2013. ISG15 regulates peritoneal macrophages functionality against viral in- Downloaded from differentiation and lung fibrosis. Am. J. Respir. Crit. Care Med. 192: 1462–1474. PLoS Pathog. PLoS Pathog. fection. [Published erratum appears in 2013 9.] 9: 59. Martinez, J., K. Verbist, R. Wang, and D. R. Green. 2013. The relationship e1003632. between metabolism and the autophagy machinery during the innate immune 46. Martinez, J., J. Almendinger, A. Oberst, R. Ness, C. P. Dillon, P. Fitzgerald, response. Cell Metab. 17: 895–900. M. O. Hengartner, and D. R. Green. 2011. Microtubule-associated protein 1 light 60. Levine, B., N. Mizushima, and H. W. Virgin. 2011. Autophagy in immunity and chain 3 alpha (LC3)-associated phagocytosis is required for the efficient clear- inflammation. Nature 469: 323–335. ance of dead cells. Proc. Natl. Acad. Sci. USA 108: 17396–17401. 61. Ruiz-Garcıá, A., E. Monsalve, L. Novellasdemunt, A. Navarro-Sabate´, 47. Flannagan, R. S., J. Canton, W. Furuya, M. Glogauer, and S. Grinstein. 2014. A. Manzano, S. Rivero, A. Castrillo, M. Casado, J. Laborda, R. Bartrons, and The phosphatidylserine receptor TIM4 utilizes integrins as coreceptors to effect M. J. Dıáz-Guerra. 2011. Cooperation of adenosine with macrophage Toll-4 http://www.jimmunol.org/ phagocytosis. Mol. Biol. Cell 25: 1511–1522. receptor agonists leads to increased glycolytic flux through the enhanced ex- 48. Wang, E., J. Michl, L. M. Pfeffer, S. C. Silverstein, and I. Tamm. 1984. Inter- pression of PFKFB3 gene. J. Biol. Chem. 286: 19247–19258. feron suppresses pinocytosis but stimulates phagocytosis in mouse peritoneal 62. Han, C. Z., and K. S. Ravichandran. 2011. Metabolic connections during apo- macrophages: related changes in cytoskeletal organization. J. Cell Biol. 98: ptotic cell engulfment. Cell 147: 1442–1445. 1328–1341. 63. Park, D., C. Z. Han, M. R. Elliott, J. M. Kinchen, P. C. Trampont, S. Das, 49. Korn, E. D., M. F. Carlier, and D. Pantaloni. 1987. Actin polymerization and S. Collins, J. J. Lysiak, K. L. Hoehn, and K. S. Ravichandran. 2011. Continued ATP hydrolysis. Science 238: 638–644. clearance of apoptotic cells critically depends on the phagocyte Ucp2 protein. 50. De Bock, K., M. Georgiadou, S. Schoors, A. Kuchnio, B. W. Wong, Nature 477: 220–224. A. R. Cantelmo, A. Quaegebeur, B. Ghesquie`re, S. Cauwenberghs, G. Eelen, 64. Telang, S., B. F. Clem, A. C. Klarer, A. L. Clem, J. O. Trent, R. Bucala, and et al. 2013. Role of PFKFB3-driven glycolysis in vessel sprouting. Cell 154: J. Chesney. 2012. Small molecule inhibition of 6-phosphofructo-2-kinase sup- 651–663. presses t cell activation. J. Transl. Med. 10: 95. by guest on October 2, 2021 1 Supplemental Figures and Legends

Supplemental Fig. 1

A. B. M MEF

VSV -+-+

70 VSV-G

Actin 40

E. pI:C-induced supernatant

M Mock

pI:C C. D. (-) pI:C Tfxn 1200 VSV 10000

Counts MEF Mock 600 UV-VSV 1000 pI:C 100 10 10 5 Relative mRNA levels 0 mRNA induction (fold) 1 Ifnb1 Ifna4 Cxcl11 Ifnb1 Mx2 FL1-H (2-NBDG)

F. G. H. IFNAR1 -+ pI:C-induced 60 supernatant (-) VSV ++ 50 1.5 MphM VSV 95 MEF 40 pStat1 VSV+R1 1 30 () 95 Stat1 0.5 20 Actin Relative mRNA levels 40 10

Ext lactate change (fold) 0 0 2 Mx2 Isg15 Il1b Ccl3

4 Supplemental Fig. 1: VSV infection leads to IFN-dependent enhancement of

5 glycolysis preferentially in macrophages. (A) Sequences (5’ to 3’) for the major

6 qPCR primers (in pairs) used in the present study. (B) PMs or MEFs were inoculated

7 with VSV at MOI of 1. Cell lysates were harvested 8 h later and analyzed on WB. (C)

8 PMs were inoculated with live VSV or UV-treated VSV at MOI of 1. Eight hours

1 1 following removing of the inoculum, total RNA was harvested for analysis of

2 indicated markers. The mRNA levels of untreated cells were used for normalization.

3 (D) PMs were mock-transfected or transfected with poly(I:C) (pI:C, 5 g/ml) for 4 h

4 and the cells were then fed with fresh medium. Six hours later, total RNA was

5 harvested for analysis. The levels of induction of indicated mRNAs were normalized

6 to the control levels and presented in log scale. (E and F) MEFs were transfected with

7 poly(I:C) (3 g/P100) for 4 h. The cells were fed with fresh medium and further

8 incubated for 6 h. The conditional medium was harvested. PMs and MEFs were

9 stimulated with the latter conditional medium for 8 h. In (E), 2-NBDG was then added

10 to the culture medium for 15 min. The cells were subjected to flow cytometry to

11 determine the level of 2-NBDG uptake. In (F), the relative levels of extracellular

12 lactate over those in untreated cells were plotted. (G) PMs were infected with VSV

13 (or mock-infected) in the presence of control IgG or a blocking antibody against

14 IFNAR1. RNA samples were harvested 8 h after infection and subjected to qPCR

15 analysis. (H) PMs were infected as in (G), but only for 4 h. The levels of pStat1 and

16 Stat1 in the cell lysates were analyzed by WB. At this earlier time point, pStat1 but

17 not Stat1 level was sensitive to receptor blockage. In this figure, data presented are

18 representative of at least two independent experiments. Quantitative data were

19 graphed using the means of triplicate or quadruplicate measurements ( STDEV).

2 Supplemental Fig. 2

A. B. BMDM PM MEFs C. U937 M 293T V V 1 I 6 2 F ( ( h ( ( N 1.2 - V - h - V - I ) ) ) ) F ( - N (-) 70 ) 1 P.FB3 70 IFN P.FB3 P.FB3 70 0.8 N.S.* 55 95 55 55 0.6 Stat1

0.4 Actin 40 95 95 0.2 70 Stat1 Stat1 VSV-G Relative mRNA levels 0

4 B1 KF KFB Actin 40 Actin 40 F PF PFKFB2P

D. E. F. IFNAR1 12 ADD -pI:C-pI:C JAK-I - - + + PBS IFN - + - + 8 70 pI:C P.FB3 70 55 P.FB3 4 CLOD+PBS 55 CLOD+pI:C 2 95 Stat1 95 Stat1 1 Cox2 70 Relative mRNA levels mRNA Relative Actin 40 0 Pfkfb3 Isg15 Cd68 Actin 40 1

3 Supplemental Fig. 2: Type I IFN induces glycolytic regulator PFKFB3 in

4 macrophages. (A) PMs were treated with 1000 IU/ml of mIFN for 8 h and RNA

5 samples were analyzed for the levels of other Pfkfb family members. (B) BMDMs

6 (MOI of 5), PMs (MOI of 1) and MEFs (MOI of 1) were treated with VSV (V) for 8 h,

7 if not specifically indicated. The cell lysates were subjected to Western analysis with

8 indicated antibodies. ‘P.FB3’ is the abbreviation for PFKFB3. Two major bands

9 representing PFKFB3 (marked by a bracket) and non-specific signal (N.S.) are

10 indicated. (C) U937 macrophages or the non-immune 293T cells were stimulated with

11 1000 IU/ml of hIFN for 8 h as indicated. Cell lysates were analyzed on Western. (D)

12 (D) PMs were pre-treated with Jak-inhibitor (5 M) for 1 h and were then stimulated

13 with IFN for 8 h. Cell lysates were analyzed on immunoblot. (E) Mice were

3 1 intranasally instilled with PBS or chlodronate liposome (CLOD). Two days later, the

2 mice were injected i.p. with 50 g poly(I:C). 12 h later, lungs were harvested from the

3 mice and the RNA samples were subjected to qPCR analyses. (F) PMs were

4 stimulated with poly(I:C) (50 g/ml) for 8 h in the presence in the presence of control

5 IgG or blocking antibody against IFNAR1. Cell lysates were subjected to Western

6 analysis. In this figure, data presented are representative of at least two independent

7 experiments. The qPCR analyses were graphed using the means of quadruplicate

8 measurements ( STDEV).

4 Supplemental Fig. 3

A. Mouse Pfkfb3 genome structure

3’-UTR

Exon 1 Exon 2 …… Exon 17

TCCCTTTCTTGATCTTTCCAGCCTGT‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐GACT‐‐‐‐‐‐GGAGCTGACTCGCTATCTCAACTGGATAGGT (Exon2‐73bp,A>G,C>T)

B.WT C. F4/80 CD11b CD11c D. +/- Pfkfb3 WT pI:C Tfxn P.fb3 +/- 100000 4 10 WT Pfkfb3+/- /well) 6 BMDM 10000

3 /mouse) 7.5 6

1000 2 5 FL-4H FL-1H FL-1H

1 2.5 100 mRNA induction (fold) induction mRNA P.Mph 10 # (x PM Ifnb1 Mx2 Isg15 BMDM cellBMDM # (x 10 0 0

FL-4H FL-1H FL-1H

E. F. G. 100 60 WT (-) 2 WT 50 He t 80 IFN He t 40 1.5 60 30 40 1 20 g h) protein/8

OCR (pmol/min) 20  ECAR (mpH/min) ECAR

Ext lactate 10 0.5 0 0 (nmol/ Bas al OLM FCCP 0 Bas al OLM WT Pfkfb3+/- IFN ---++ + + - H. MEF I. MEF 3 70 2.5 (-) (-) P.FB3 55 IFN 2 IFN 2

Actin 40 1.5 g h) protein/8 1  Ext lactate lactate Ext g h) protein/8

 1 Ext lactate 0 (nmol/ 0.5 pcDNA P.FB3-Flag (nmol/ IFN - + - + 0 WT Pfkfb3+/- 70 Flag IFN -+-+ 55 70 P.FB3 70 55 P.FB3 55 Stat1 95 95 Stat1

Actin Actin 40 1 40 2

3 Supplementary Fig. 3: PFKFB3 enhances glycolysis in macrophages. (A) Genome

4 structure of mouse Pfkfb3 gene is depicted. The exon harboring the indel mutation is

5 highlighted. (B) Left: Bone marrow cells were plated (triplicates) at a density of 2x

6 106 in a 60-mm dish in the presence of macrophage-differentiating L-medium. On day

5 1 7, the fully differentiated macrophages were counted (STDEV). Right: five days

2 after i.p. injection of thioglycollate, the peritoneal macrophages were harvested (n=2).

3 The numbers of PMs isolated from the WT and Pfkfb3+/- mice were counted ( range).

4 (C) The WT and Pfkfb3+/- BMDMs (upper) and PMs (lower) were subjected to flow

5 cytometry analysis using the indicated antibodies. (D) BMDMs were

6 mock-transfected or transfected with poly(I:C) (pI:C, 5 g/ml) for 4 h and the cells

7 were then fed with fresh medium. Eight hours later, total RNA was harvested for

8 analysis. The levels of induction of indicated mRNAs were normalized to the control

9 levels in the WT cells and presented in log scale. (E) Thioglycollate-elicited PMs

10 from the WT or Pfkfb3+/- mice were stimulated with IFN for 8 h. The culture

11 supernatant was analyzed for the levels of extracellular lactate and the cell lysates

12 were analyzed using indicated antibodies. (F and G) WT and Pfkfb3+/- PMs were

13 treated as in (E). Cells were then subjected to ECAR (F) and OCR (G) analysis using

14 Seahorse metabolic analyzer. Under given nutrient/stress conditions, the average

15 levels of the latter parameter at several time points were averaged. (H and I) WT or

16 Pfkfb3+/- primary MEFs (H) and MEFs transfected with empty plasmid or Flag-tagged

17 PFKFB3 (I) were stimulated and analyzed as in (E). In this figure, data presented are

18 representative of at least two independent experiments. Metabolic analyses were

19 graphed using the means of triplicate or quadruplicate measurements ( STDEV).

6 Supplemental Fig. 4

A. B.BMDM + PFK15 C. MEF + PFK15

MEF (-) (-) MOI 0.1 0.5 1 5 +/- +/- +/- WT +/- WT WT WT 70 1 M VSV-G 5 M 70 P.FB3 55 10 M 5 M Actin 40

100 101 102 103 104 100 101 102 103 104 FL1-H (2-NBDG) FL1-H (2-NBDG)

D. (-) co-culture E. F. + CFSE-labeled, infected MEFs

PFK15 co-culture 50 7 1x10 Transwell co-culture MEFs 20 WT Pfkfb3+/- 40 30 15 7.5x106

MFI 20 Co-culture 10 10 Counts 5x106 0 5 nduction (fold) nduction 5x10 5 VSV titer (pfu/ml) 5

mRNA i 2.5x10 0 Isg15 Mx2 FL1-H 0 PM P.fb3 +/+ +/- +/+ +/- N.A. genotype G. IgG H. I. WT WT TIM4 BMDM PM 40 Pfkfb3+/- Pfkfb3+/- 1 1 Adhesion index 1.5 30 1 0.5 0.5 20 0.5 infected MEFs) 10 Phagocytic index

MFI (Engulfmentof 0 0 Phagocytic index 0 Engulfed Adhered 0

J. K. WT Pfkfb3+/-

Ins (min) 0 15 30 0 15 30 (-) pAKT 70 A.T. AKT 70 IFN+A.T.

Counts pPRAS40 40

PRAS40 40 70 PFKFB3 FL1-H (2-NBDG) 55 40 1 GAPDH

3 Supplementary Fig. 4: PFKFB3 on macrophage extrinsic antiviral activity and

4 efferocytosis. (A) WT and Pfkfb3+/- primary MEFs were infected with VSV at

5 indicated MOIs. 14 h after infection, the lysates were harvested and analyzed using

6 the indicated antibodies. (B and C) BMDMs (B) or MEFs (C) were treated with

7 1 indicated concentration of PFK15 for 3 h. Cells were then incubated with glucose

2 analog, 2-NBDG, and subjected to flow cytometry analysis. A left-ward shift of the

3 peak indicates a decreased capacity to uptake glucose. (D) WT MEFs were first

4 infected with VSV at MOI of 1 for 14 h. Unattached and loosely attached cells were

5 collected. Infected MEFs were then added in a ratio of 2:1 into BMDMs that had been

6 pre-treated with DMSO or PFK15 for 3 h. RNA samples were harvested after 12 h

7 and later analyzed (means of quadruplicate measurements, STDEV). (E) MEFs were

8 labeled with CSFE and were then infected and collected as in (D). Infected MEFs

9 were then added in a ratio of 10:1 into BMDMs. Four h later, the cells were washed

10 extensively and stained with F4/80 antibody (FL4-H). To measure efferocytosis of

11 infected MEFs, the CSFE (FL1-H) fluorescence of F4/80-positive cells was

12 determined by flow cytometry. For quantitation, the MFI was determined and shown

13 in the inset. (F) The infected MEFs were prepared as in (D) and added at a ratio of 1:2

14 to wells containing (or w/o) WT or Pfkfb3+/- PMs. In some groups, infected MEFs

15 were segregated from the PMs by 3 m pore-sized transwell membranes. The culture

16 supernatants were harvested in 12 h for viral titer analysis (means of duplicate

17 measurements,  range). (G) BMDMs were pre-incubated with IgG or anti-TIM4

18 (100 g/ml) and were then subjected to the flow cytometry-based efferocytosis assay

19 as in (E). The MFI levels of CSFE fluorescence were shown. (H) The efferocytosis

20 activities of WT and Pfkfb3+/- BMDMs were measured using apoptotic thymocytes

21 (20:1). Four h following cargo addition, the average number of engulfed apoptotic

22 cells per total macrophage in 10 different microscopic fields were determined

8 1 (STDEV) and were shown on the left. To determine the capacity of BMDMs to

2 initially bind the cargos, the BMDMs were initially added with a greater excess (ratio

3 of 100:1) of apoptotic thymocyte. After a brief incubation of 15 min, the attached

4 apoptotic thymocytes were quantitated similarly as above and the results were shown

5 on the right. (I) WT and Pfkfb3+/- PMs were subjected to efferocytosis assay using

6 apoptotic thymocytes. (J) Serum-starved PMs that had undergone pre-treatment w/

7 or w/o IFN (8 h) were added with A.T. (20:1). Cells were harvested in 1 h for assay of

8 2-NBDG uptake. (K) WT and Pfkfb3+/- PMs were added with 100 nM of insulin for

9 indicated times. The cell lysates were harvested and analyzed in Western. Data

10 presented in this figure are representative of two independent experiments.