Article Images (0)
×![]()
featured_play_listSave article to folder
X
Create new folder
Full Text
PDF
Table of Contents
MLKL, The Protein That Me
Regulates Endosomal Trafficking And Extracellular
Vesicle Generation
Graphical Abstract
Highlights
Immunity
Article
MLKL, The Protein That Mediates Necroptosis,
Also Regulates Endosomal Trafficking
Seongmin Yoon,1 Andrew Kovalenko,1 Konstantin Bogdanov,1 And David Wallach1,2,*
SUMMARY
INTRODUCTION
RESULTS
MLKL Affects Endosomal Transport Independently Of RIPK3
MLKL Expression Is Required For Effective Generation Of Intraluminal And Extracellular Vesicles
Phosphorylation By RIPK3 Triggers Phosphorylated MLKL Extrusion From Cells In Extracellular Vesicles
Release Of Ph-MLKL In Extracellular Vesicles Seems To Withhold Necroptotic Cell Death
DISCUSSION
AUTHOR CONTRIBUTIONS
ACKNOWLEDGMENTS
SUPPORTING CITATIONS
KEY RESOURCES TABLE
Continued
Continued
Continued
CONTACT FOR REAGENT AND RESOURCE SHARING
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Mice
Cell Culture
METHOD DETAILS
Reagents
Reagents For RNA Interference
Antibodies
Collection And Quantification Of Extracellular Vesicles
Ligand And Receptor Uptake Assays, And Assessment Of Phosphorylation Of Signaling Proteins
Immunoprecipitation And Western Blot Analysis
Mass Spectrometry-Based Proteomics Analysis
Sample Preparation
Fluorescence Microscopy
Transmission Electron Microscopy
Immunogold Electron Microscopy
Quantification Of Cell Death
Real-Time PCR Analysis
Expression Analysis Of NanoString Inflammatory Panel RNA
QUANTIFICATION AND STATISTICAL ANALYSIS
Article
MLKL, the Protein that Me
diates Necroptosis, Also
Regulates Endosomal Trafficking and Extracellular
Vesicle Generation
Graphical Abstract
Highlights
d MLKL facilitates endosomal trafficking independently of induction of necroptosis d RIPK3 boosts release of phospho-MLKL-containing vesicles from cells d This release appears to withhold death and might contribute to cellular communication d The structural requirements for endosomal and necroptotic MLKL functions overlap Yoon et al., 2017, Immunity 47, 1–15 July 18, 2017 ª 2017 Elsevier Inc. http://dx.doi.org/10.1016/j.immuni.2017.06.001 Authors Seongmin Yoon, Andrew Kovalenko, Konstantin Bogdanov, David Wallach Correspondence d.wallach@weizmann.ac.il In Brief When phosphorylated by RIPK3, MLKL triggers necroptotic death. Yoon et al. show that MLKL also contributes to endosomal trafficking and generation of extracellular vesicles. This function is independent of RIPK3 but can be enhanced by it, yielding phospho-MLKL release within the vesicles, thereby apparently withholding death mediation by MLKL.
Immunity
Article
MLKL, the Protein that Mediates Necroptosis,
Also Regulates Endosomal Trafficking
and Extracellular Vesicle Generation
Seongmin Yoon,1 Andrew Kovalenko,1 Konstantin Bogdanov,1 and David Wallach1,2,*
1Department of Biomolecular Sciences, The Weizmann Institute of Science, 76100 Rehovot, Israel 2Lead Contact *Correspondence: d.wallach@weizmann.ac.il http://dx.doi.org/10.1016/j.immuni.2017.06.001
SUMMARY
Activation of the pseudokinase mixed lineage kinase domain-like (MLKL) upon its phosphorylation by the protein kinase RIPK3 triggers necroptosis, a form of programmed cell death in which rupture of cellular membranes yields release of intracellular components. We report that MLKL also associated with endosomes and controlled the transport of endocytosed proteins, thereby enhancing degradation of receptors and ligands, modulating their induced signaling and facilitating the generation of extracellular vesicles. This role was exerted on two quantitative grades: a constitutive one independent of RIPK3, and an enhanced one, triggered by RIPK3, where the association of MLKL with the endosomes was enhanced, and it was found to bind endosomal sorting complexes required for transport (ESCRT) proteins and the flotillins and to be excluded, together with them, from cells within vesicles. We suggest that release of phosphorylatedMLKLwithin extracellular vesicles serves as amechanism for self-restricting the necroptotic activity of this protein.
INTRODUCTION
Research on the mechanisms of cell-death induction by members of the tumor necrosis factor (TNF) cytokine family has revealed a signaling pathway that triggers programmed necrotic cell death (Han-Min and Vandenabeele, 2014). Unlike the extrinsic apoptotic cell-death pathway that is also activated by these cytokines, and where the initiation of signaling and the execution of death are mediated by members of the caspase family of cysteine proteases (Budihardjo et al., 1999), death via the ‘‘necroptotic’’ pathway is induced by activation of the protein kinaseRIPK3 (Cho et al., 2009; He et al., 2009; Zhang et al., 2009) and the pursuant RIPK3-mediated phosphorylation of the pseudokinase mixed lineage kinase domain-like (MLKL) (Sun et al., 2012; Zhao et al., 2012). Some necroptosis-inducing agents employ the protein kinase RIPK1 to activate RIPK3 (Degterev et al., 2008; Holler et al., 2000). Caspase-8, the proximal enzyme in the extrinsic apoptotic pathway, strongly inhibits the induction of necroptosis (Holler et al., 2000; Oberst et al., 2011; Vercammen et al., 1998). In vivo exploration of the functions of individual proteins participating in the necroptotic signaling pathway suggests that activation of this pathway triggers inflammation. Blocking of the kinase function of RIPK1, either by deletion of the kinase gene or pharmaceutically, is found to ameliorate a variety of inflammatory pathologies (Christofferson et al., 2014). Conversely, deficiency of caspase-8 or of FADD (the adaptor protein to which caspase-8 binds) is found to trigger inflammation (e.g., Bonnet et al., 2011; Wallach et al., 2014), which can be withheld by concomitant deletion of the Ripk1, Ripk3, orMlkl genes (Newton et al., 2016; Wallach et al., 2011). In the present study, we show that MLKL also serves to facilitate endosomal function and generation of extracellular vesicles (EVs) and that it does so independently of RIPK1 or RIPK3. Its phosphorylation by RIPK3 further enhances EV generation, independently of death induction, and prompts incorporation of the phosphorylated MLKL molecules into the EVs. We and others have previously shown that, in dendritic cells, triggering of MLKL phosphorylation by RIPK3 prompts assembly of the NLRP3 inflammasome and release of interleukin (IL)-1b and IL-18, without any sign of death (Conos et al., 2017; Kang et al., 2013). Here we present evidence suggesting that the release of RIPK3-phosporylated MLKL within EVs antagonizes the necroptotic function of this protein; it withholds death and, in the case of dendritic cells, contributes to its complete arrest.
RESULTS
MLKL Affects Endosomal Transport Independently of RIPK3
To gain further knowledge of the function of MLKL, we sought to analyze comprehensively the impact of its deficiency on the cellular response to TNF. Besides their resistance to the necroptotic effect of this cytokine, MLKL-depleted cells also showed a marked reduction in the rate of intracellular degradation of TNF after its binding to the TNF receptor. The slight decrease in TNF receptor 1 (TNFR1) observed in some cells after its binding of TNF (e.g., in HepG2 cells; see Figure 1A) was also withheld by MLKL depletion. This slowdown occurred not only in RIPK3-expressing cells, such as HT-29, but also in HeLa and HepG2 cells, which are deficient in RIPK3 (Figure S1), as well as in fibroblasts derived from Ripk3 / mice. This finding implied that the Immunity 47, 1–15, July 18, 2017 ª 2017 Elsevier Inc. 1 (legend on next page) 2 Immunity 47, 1–15, July 18, 2017 slowdown reflects an MLKL function that is independent of MLKL phosphorylation by RIPK3 and of necroptosis (Figures 1A and 1B). The slowdown of TNF degradation in MLKL-depleted cells was associated with pronounced alterations in TNF-induced signaling, as well as with increased basal concentrations of phospho-IkB (Figure 1B). NanoString and real-time PCR analyses of TNF-induced gene activation revealed that MLKL deficiency resulted in dramatically enhanced induction of some of these genes, but not of others. It also resulted in increased basal expression of some inflammatory genes (Figures 1C and 1D). To determine whether the mechanism accounting for the observed slowdown also affects other ligands, we examined the impact of MLKL deficiency on the cellular response to epidermal growth factor (EGF), a ligand of the tyrosine kinase receptor EGFR. Like TNF, EGF is taken up and degraded, along with its receptor, in lysosomes (Gur et al., 2006). As in the case of TNF, we found that RNA silencing of MLKL resulted in marked slowdown of the intracellular degradation of EGF and EGFR (Figure 1E). Likewise, the degradation of EGF and EGFR in vivo, observed in the livers of mice injected with EGF, was slower in Mlkl / mice than in the wild-type (Figure 1F). As with TNF, the slowed degradation of EGF and its receptor resulted in boosting of several EGF-induced signaling activities as well as enhanced gene induction by EGF. It also resulted in some increase in the basal intensity of these signaling activities and of cellular amounts of EGFR (Figures 1E–1G). The slowdown of intracellular degradation of TNF by MLKL depletion and the resulting modulation of gene induction was also detectable in cells cultured in the absence of EGF (Figures 1A and S2A). Conversely, slowdown of intracellular degradation of EGFR could also be observed in cells with deletion of Tnf (Figure S2B). It thus appeared that the effects of MLKL depletion on these two ligands and their receptors reflects a mechanism that exerts a general effect on receptor uptake and degradation. Immunocytochemical analysis of the intracellular fate of the EGFR in EGF-treated cells showed that it was taken up into the cells at about the same rate in both MLKL-expressing and MLKL-depleted cells. However, subsequent translocation of the receptor from the early to the late endosomes, and its ensuing degradation, were delayed (Figures 2A–2C). Treatment of the EGF-stimulated cells with lysosomal inhibitors arrested Figure 1. Degradation of Internalized Ligands and Receptors Is Delaye (A) Immunoblot analysis of the kinetics of intracellular degradation of biotinylated T on it. To exclude any impact of EGFR signaling on the observed MLKL-depletion (B) Immunoblot analysis of the impact of MLKL RNA silencing on the induction of TNF. NT, not treated. (C) Effect of MLKL RNA silencing on the TNF-induced expression of inflammatio Control or MLKL-siRNA-silenced HT29 cells were treated with TNF for the indic specified genes relative to control cells at 0 hr. Shown are the 20most strongly upr 6 hr siRNA-silenced sample. Darker fill color indicates higher expression. (D) Real-time PCR validation of the expression kinetics of genes analyzed in (C) (E) Immunoblot analysis of the kinetics of intracellular degradation of EGF and EG are the cellular amounts of phosphorylated EGFR and of three targets of EGFR s (F) Effects of MLKL deficiency on degradation of EGF and EGFR and on EGFR sign biotinylated EGF. (G) Real-time PCR analysis of the expression kinetics of several EGF-induced ge Values shown in (D) and (G) are averages of two independent experiments. The Please also see Figures S1 and S2. the EGF-induced decrease in EGFR to the same extent in MLKL-expressing andMLKL-depleted cells (Figures 2D and 2E). The observed effects of MLKL depletion on the rates of intracellular degradation of TNF, EGF, and their receptors pointed to the possibility of a constitutive role for MLKL in maintaining the endosomal trafficking of proteins. The enhancement of both basal and induced signaling by TNFR and EGFR seen in MLKL-depleted cells is consistent with prior reports attributing such enhancement to interference with endosomal trafficking (Babst et al., 2000; Mami nska et al., 2016; Vieira et al., 1996).
MLKL Expression Is Required for Effective Generation of Intraluminal and Extracellular Vesicles
A critical step in the endosomal transport of cell-surface receptors to the lysosome is the budding of membrane vesicles containing these receptors into the endosome lumen. These so-called intraluminal vesicles (ILVs) accumulate within multivesicular bodies (MVBs), which then fuse with or mature into lysosomes (Katzmann et al., 2002). Our electron microscopic analysis revealed a pronounced decrease in protein content and diminished count of ILVs in the MVBs of MLKL-depleted cells, along with an increase in diameter of the MVB-limiting membrane (Figures 3A–3C). Whereas in MLKL-expressing cells EGFR molecules could be discerned within the MVB cavity, in MLKL-depleted cells these molecules were largely restricted to the region of the MVB-limiting membrane (Figure S3A). MVBs fuse not only with lysosomes but also with the cell membrane. The ILVs released in this way correspond to the exosomes found in the extracellular milieu. Consistently with the observation that MLKL was required for effective generation of ILVs, we found that MLKL deficiency resulted in amarked reduction in the generation of EVs (Figures 3D–3H), whose shape, size, and sedimentation ratewere characteristic of exosomes (Figures S3B and S3C; Colombo et al., 2014) and which contained proteins known to occur in exosomes (Figure 3G). That this reduction was due to deficient generation of ILVs and not their deficient release was indicated by the fact that it could not be reversed by treatment with ionomycin (Figures 3D, 3F, and 3G), which is known to boost exosome release but not ILV generation (Colombo et al., 2014). Deficiency of RIPK1 or RIPK3 did not affect EV yield (Figure S4), nor was RIPK3 expression required for the observed d in MLKL-Depleted Cells, with Resulting Signaling Augmentation NF after its binding to the indicated cells and the impact of MLKL RNA silencing effect, TNF uptake was assessed in the absence of growth factors. phosphorylation of the ERK and p38 MAP kinases and of NF-kB activation by n-related genes, as assessed with the NanoString nCounter Analysis System. ated times. Numbers in the table record the fold increase in expression of the egulated genes (averages of duplicate tests), sorted in descending order for the by the NanoString system. FR in HepG2 cells, and the impact of MLKL RNA silencing on them. Also shown ignaling—AKT, STAT3, ERK—and their phosphorylated forms. aling in the livers of wild-type andMlkl / mice at various times after injection of nes in the livers of mice, 2 hr after their injection with EGF. bars show the range of the results. Immunity 47, 1–15, July 18, 2017 3 (legend on next page) 4 Immunity 47, 1–15, July 18, 2017 impact of MLKL RNA silencing on EV generation (Figure 3H). These findings further indicated that the role of MLKL in controlling endosomal trafficking is mediated independently of MLKL phosphorylation by RIPK3. The Intracellular Trafficking and Necroptotic Functions of MLKL Share Structural Requirements To explore the mechanistic interrelationships between the roles of MLKL in endosomal trafficking and in necroptosis, we compared the effects of MLKLmutations on endosomal function and on cell viability by inducibly expressing the various mutants in cells whose endogenous MLKL expression had been siRNA silenced. As mentioned above, we found that the contribution of MLKL to the cellular generation of EVs did not depend on its phosphorylation by RIPK3. Nevertheless, an MLKL mutant in which threonine (T357A) and serine (S358A), the two residues phosphorylated by RIPK3, were replaced by alanine failed to reconstitute EV generation in MLKL-depleted cells, suggesting that these two residues do in fact contribute to this constitutive function (Figures 4A and 4B). Moreover, expression of MLKL molecules harboring phosphomimetic mutations of threonine 357 and serine 358 (T357E/S358D), besides triggering death (Figure 4E; Sun et al., 2012), also enhanced EV generation (Figures 4C–4E). Even stronger enhancement was found when we usedMLKLmolecules harboring, within the ATP-binding pocket, mutations that impose the conformational change attained upon MLKL phosphorylation (K230M and Q356A) (Figures 4C–4E; Murphy et al., 2013). Expression of the latter mutant also enhanced intracellular degradation of EGF and EGFR (Figures 4F and 4G). The conformational change attained uponMLKL phosphorylation exposes an N-terminal region structured as a bundle of several a helices (referred to here as HB), allowing the region to bind negatively charged phospholipids in cellular membranes. Concomitant mutation of five basic residues at the HB (the ‘‘5A’’ mutation) that are believed to be required for that binding (Dondelinger et al., 2014; Hildebrand et al., 2014; Su et al., 2014; Wang et al., 2014) compromises death induction (Quarato et al., 2016). We found that it also compromised the ability of MLKL to support EV generation (Figures 4A and 4B). Fusion of N-terminal tags upstream of the HB also reportedly interferes with lipid binding and death induction by MLKL (Chen et al., Figure 2. Endosomal Trafficking Is Slowed in MLKL-Depleted Cells (A and B) Immunocytochemical analysis of EGFR uptake after application of EG localization of the receptor with EEA1 and Rab7-markers of early and late endos (A) Immunofluorescence images. Yellow, co-localization of EGFR (green) and EEA (B and C) Quantification of the data presented in (A). (B) Total amounts of EGFR in the cells and the amounts of EGFR associated with th amounts of EGFR in the cells. Values are averages of two independent experime (C) Amounts of EEA1 and Rab7 that colocalize with EGFR expressed respective fication of the data presented in (A). (D and E) Effects of arrest of lysosomal degradation by chloroquine (CQ) and ba (D) Immunoblot analysis, 2 hr after binding of biotinylated EGF to control and ML (E) Immunocytochemical analysis of a test performed as in (B), using LysoTracker ( marker Rab7. Shown are the results of immunostaining and their superposition on staining; white arrows indicate co-localization of EGFR, LysoTracker, and Rab7 sta almost fully degraded in the control cells, whereas in the MLKL-siRNA-silenced lysosomal degradation by CQ or Baf A1 resulted in equal EGFR accumulation in Scale bar, 10 mm. 2014; Hildebrand et al., 2014; Yoon et al., 2016). We found that, likewise, such fusion also compromised EV generation (data not shown). We did, however, observe one difference between the structural requirement for the induction of EVs and for necroptotic death by MLKL. An MLKL deletion mutant corresponding to the HB (‘‘1 180’’) is known to be capable of binding negatively charged phospholipids. However, whereas its mere overexpression triggers cell death (Chen et al., 2014; Dondelinger et al., 2014; Hildebrand et al., 2014), such overexpression did not yield enhanced generation of EVs. In fact, we found the 1 180mutant to be even less effective than the wild-type protein in facilitating EV generation (Figures 4C–4E). Altogether, these findings imply that some but not all of the structural requirements for the facilitation of endosomal function by MLKL are identical to those required for its necroptotic function and that the conformational change attained by MLKL upon induction of necroptosis further facilitates its endosomal function.
Phosphorylation by RIPK3 Triggers Phosphorylated MLKL Extrusion from Cells in Extracellular Vesicles
Having found that the conformational change attained by MLKL as a consequence of its phosphorylation by RIPK3 enhances its endosomal function, we next examined whether such enhancement is indeed yielded by activation of RIPK3. Such activation was achieved by treatment of RIPK3-expressing cells (HT-29 cells and mouse embryonic fibroblasts [MEFs]) with a combination of TNF, the SMAC mimetic agent BV6, and the caspase inhibitor z-VAD-fmk (TBZ) (Figures 5A–5E), or—when caspase8-deficient MEFs were used (Figure 5F)—by their treatment with only TNF plus BV6. Such treatments did not affect either the size distribution or the shape of the EVs, but they did result in an increase in their number (Figures 5B, S3B, and S3C). The increase occurred shortly after the stimulation was initiated, at which time only a few dead cells could be observed (Figures 5C and 5D). These results confirmed that, besides the constitutive contribution of MLKL to endosomal trafficking and EV generation, which does not depend on RIPK3, MLKL can also contribute to these functions at an enhanced grade that does depend on its phosphorylation by RIPK3. F to control and MLKL-siRNA-silenced HepG2 cells, and the kinetics of coomes, respectively. 1 (red); cyan, co-localization of EGFR (green) and Rab7 (blue). Scale bar, 10 mm. e cell membrane at different times, expressed as percentages of the initial total nts. The bars show the range of the results. ly as percentages of the total amounts of EEA1 and Rab7 in the cells. Quanti- filomycin A1 (Baf A1) on the cellular amounts of EGF and EGFR. KL-siRNA-silenced HepG2 cells, as in Figure 1B. a lysosome-staining reagent) and antibodies to EGFR and to the late endosome transmission pictures. Cyan arrows indicate co-localization of EGFR and Rab7 ining. At the time of the test (2 hr after ligand binding), the EGFRs taken upwere cells, some EGFRs remained in the late endosome compartment. Arrest of that compartment in control and MLKL-siRNA-silenced cells. Immunity 47, 1–15, July 18, 2017 5 (A C) Transmission electron microscopy (TEM) of the multivesicular bodies in wild-type and MLKL-siRNA-silenced HepG2 cells. (A) Representative pictures (scale bar, 100 nm). (B and C) Quantification of the sizes and numbers of arbitrarily chosen MVBs (132 MVBs in control cells and 115 in MLKL-siRNA-silenced cells) identified by the presence of BSA-gold in them (arrows) and their ILV content. The bars show the standard variations. (D H) Effects of MLKL RNA silencing (in HepG2 cells and in MEFs derived from Ripk3 / mice) and ofMlkl deficiency achieved by gene targeting (in BMDCs and in MEFs), and of ionomycin treatment (IONO; 1 mM for 12 hr) on EV generation, as assessed by (D F and H) nanoparticle tracking analysis (NTA), and (G) by immunoblotting. (D and E) Averages of quadruplicate assessments done in one of two qualitatively similar tests. The bars show the standard variations. (H) Averages of individual values in two independent tests. The bars show the range of the results. Please also see Figures S3 and S4. Activation of RIPK3 in vivo by injection of mice with TNF plus z-VAD-fmk triggered in wild-type mice, but not in Mlkl / mice, necroptotic death of cells, as reflected in increased serum concentrations of lactic dehydrogenase (LDH) (Figure 5G; Duprez 6 Immunity 47, 1–15, July 18, 2017 et al., 2011). This treatment also resulted in increased EV numbers in the serum (Figure 5H). Immunocytochemical staining of cells with antibodies toMLKL and to endosome markers revealed that MLKL associates (A E) Yield of EVs and extent of cell death in MLKL-siRNA-silenced HT-29 cells inducibly expressing MLKL and its indicated mutants. (A and C) Immunoblot analysis of the expressed proteins. (F and G) Effects of inducible expression of wild-type MLKL and the K230M/Q356A MLKL mutant on the amounts of EGF and EGFR in MLKL-siRNA-silenced HepG2 cells after binding of EGF to these cells. EGF was applied to the cells 7 hr after induction of the mutants was initiated. (F) Immunoblot analysis. (G) Densitometric quantification of the results. Expression of the K230M/Q356Amutant caused the death of only about 1%of the HepG2 cells within the time limit of the test. (B, D, and E) Averages of individual values in two independent tests. The bars show the range of the results. constitutively with both early and late endosomes and that such association was enhanced shortly after TBZ treatment (Figures 5J–5L). The enhancement occurred prior to translocation of MLKL to the cellular membrane and before cell death (Figures 5J–5L and S5A). Immunoelectron microscopic analysis with antibodies to MLKL revealed that the protein was taken up into the MVBs (Figure S5B), from which the exosomes are released. Consistently, both in cell culture and in vivo, we found that preparations of EVs generated after stimulation of the necroptotic pathway Immunity 47, 1–15, July 18, 2017 7 (legend on next page) 8 Immunity 47, 1–15, July 18, 2017 contained phosphorylated MLKL (ph-MLKL) (Figures 5A, 5C, and 5I). Sucrose gradient fractionation confirmed that this phMLKL was indeed associated with EVs and not with some protein aggregates that chanced to co-sediment with them (Figure S6A). Incubation of EVs with trypsin did not lead to degradation of their associated ph-MLKL, suggesting that the ph-MLKL occurs within the vesicles (Figure S6B). RIPK3 Activation Triggers MLKL Association with ESCRT Proteins and with Flotillins and then Their Release from the Cell To investigate themechanism(s) underlying the enhanced generation of EVs and their release of ph-MLKL after RIPK3 activation, we first used mass spectrometry to identify the MLKL-associated proteins within EVs released from TBZ-treated HT-29 cells. This analysis, which was further confirmed by immunoblotting, revealed that the MLKL found within the EVs of TBZ-treated cells was associated with some (but not with other) components of the endosomal sorting complexes required for transport (ESCRT) system, a series of protein complexes that serve sequentially to dictate transport of proteins in the endosomes and their incorporation into ILVs (Williams and Urbé, 2007). Two other proteins found to associate with MLKL in the vesicles were flotillin 1 and 2 (Figure 5M and Table S1), membrane-associated proteins suggested to contribute to targeting of specific cargo proteins to exosomes (Meister and Tikkanen, 2014). The total amounts of some of the MLKL-associated ESCRT proteins and, even more so, of the flotillins, in the EVs generated by TBZ-treated cells were higher than in EVs of untreated cells (Figure 5M). Using cells expressing MLKL fused to an affinity-purification tag, we then examined the identity of the proteins that associate with MLKL within cells. Both the ESCRT proteins shown here to associate with MLKL within EVs and the flotillins were found to already associate with MLKL within the cells soon after TBZ application (Figure 5N). These findings suggest that the enhanced endosomal function of MLKL after its phosphorylation by RIPK3, as well as the induced extrusion of ph-MLKLwithin the EVs, occurs as a result of an induced association of MLKL with the ESCRT proteins. Figure 5. RIPK3 Activation Enhances Release of EVs Containing both (A F) Activation of RIPK3 in HT-29 cells and in MEFs whose Casp8 gene was ind The proteins Alix, Hrs, TSG101, CD9, HSP70, flotillin 1(FLOT-1), and flotillin 2 (FLO expression is restricted to the mitochondria, and LDH is a cytoplasmic protein. ( (G I) Effects of triggering RIPK3 activation in vivo, achieved by injecting wild-typ and z-VAD-fmk (250 mg/mouse, 3 hr prior to plasma collection) on (G) the plasma c (I) on the presence of ph-MLKL in the EVs. (J and K) Immunocytochemical staining using rhodamine-tagged phalloidin, whi (green), and antibodies to endosomal markers (magenta). (J) Staining with an ant CD63 (a marker of late endosomes and of MVBs). White, co-localization of MLK Scale bar, 10 mm. (L) Quantification of the data presented in (J) and (K) (averages of the values in 5 (M andN) RIPK3-activation-induced association ofMLKLwith ESCRT proteins and RIPA buffer extracts of EVs of untreated cells and of cells treated for 3 hr with TBZ vesicles using anti-MLKL antibody. (N) Immunoblot analysis of the proteins that as TBZ, as determined by assessment of their co-precipitation with MLKL from HT (D, G, and H) Averages of individual values in two independent tests. The bars s Please also see Figures S5 and S6 and Table S1.
Release of ph-MLKL in Extracellular Vesicles Seems to Withhold Necroptotic Cell Death
We have previously shown that in mouse bone marrow-derived dendritic cells (BMDCs) that are sensitized by caspase-8 deficiency to activation of RIPK3 by lipopolysaccharide (LPS) or TNF, such activation does not result in necroptotic death but rather in activation of the NLRP3 inflammasome, with consequent release of IL-1b and IL-18 (Kang et al., 2013). Figures 6A and 6B show that BMDCs, like the other cells examined in this study, responded to RIPK3 activation by enhanced generation of EVs and the release of ph-MLKL within them. Consistently, LPS injection of mice in which Casp8 was deleted only in the BMDCs resulted both in increased plasma concentrations of EVs and in the presence of ph-MLKL in those circulating EVs, but without any increase in serum LDH concentrations (Figures 6C–6E). The relative amounts of ph-MLKL in EVs released from the LPS-treated BMDCs were higher than in EVs released by HT-29 cells, which do die upon RIPK3 activation. In HT-29 cells, ph-MLKL could be discerned both within the cells and in the EVs that they generated (Figures 5A, 5C, and 6B), whereas in the case of BMDCs, ph-MLKL was barely detectable in the BMDCs themselves but occurred profusely in the EVs (Figure 6B). BMDCs also exhibited only slow and limited death when treated with LPS in the presence of the pan-caspase inhibitor z-VAD-fmk. In contrast, mouse BMDMs are effectively killed by such treatment (Figure 6F; He et al., 2011; Kang et al., 2013). Whereas in BMDMs, combined treatment with LPS and z-VAD-fmk resulted in accumulation of phosphorylated MLKL both within the cells and in their EVs, in BMDCs that were exposed to this treatment, ph-MLKL could hardly be discerned in the cellular lysate, despite reaching high concentrations in the EVs (Figure 6G). These findings indicated that EV extrusion of ph-MLKL in BMDCs is more effective than in cells sensitive to the necroptotic effect of MLKL. To examine the possibility of causal relationship between the effectiveness of ph-MLKL release in EVs and the resistance of cells to necroptotic death, we attempted to assess the impact of RNA silencing of Rab27a and Rab27b—small G proteins needed for exosomal release (Ostrowski et al., 2010)—on the cellular response to RIPK3 activation. Unfortunately, our attempts at RNA silencing resulted in only partial decrease of MLKL and ESCRT Proteins that Bind to it ucibly deleted facilitates generation of EVs. (B) NTA size distribution patterns. T-2) are frequently found in exosomes. TOM40 and VDAC are proteins whose D) Kinetics of death induction. e andMlkl / mice (5 mice for each experimental point) with TNF (9 mg/mouse) oncentrations of LDH (as ameasure of the extent of cell lysis) and of (H) EVs and ch stains the plasma membrane-associated F-actin (red), antibodies to MLKL ibody to EEA1 (a marker of early endosomes). (K) Staining with an antibody to L with the endosomal marker; yellow, co-localization of MLKL with phalloidin. 0 cells). The bars show the standard variations. the flotillins. (M) Immunoblot analysis of the content of the indicated proteins in , and of the proteins co-immunoprecipitated with MLKL from the extracts of the sociate withMLKLwithin cells exposed for the indicated times to treatmentwith 29-cell lysates expressing MLKL fused C-terminally to a Strep-tag. how the range of the results. Immunity 47, 1–15, July 18, 2017 9 (A) Quantification of EVs released by wild-type, caspase-8-deficient, and caspase-8- and MLKL-deficient BMDCs, and its alteration by stimulation with LPS for 3 hr. Shown are averages of individual values in two independent tests. The bars show the range of the results. (B) Immunoblots depicting cellular and EV amounts of ph-MLKL in HT-29 cells and BMDCs, and their modulation by treatment with TBZ or LPS and by combined RNA silencing of Rab27a and Rab27b. (Cont, control). (C E) Effects of triggering of RIPK3 activation in dendritic cells in vivo (achieved by injecting Casp8 fl/-:Itgax-Cremice (5 mice for each experimental point) with LPS (5 mg/kg, 3 hr prior to serum collection) on (C) the serum concentrations of LDH (as ameasure of the extent of cell lysis), and (D) of EVs, and (E) on the presence of ph-MLKL in them (immunoblot). The bars in (C) and (D) show the standard variations. (F) Viability of BMDCs and BMDMs at different times after LPS +z-VAD-fmk application. (G) Cellular and EV amounts of ph-MLKL in the BMDCs and BMDMs. (legend continued on next page) 10 Immunity 47, 1–15, July 18, 2017 the cellular amounts of Rab27a and Rab27b (Figure 6B). However, even that relatively small decrease sufficed to yield a clear decrease in EV generation by both BMDCs and HT-29 cells (Figure S7). It also resulted in increased intracellular amounts of phMLKL in both of these cell types (Figure 6B). Moreover, Rab27a andRab27b RNA silencing rendered the BMDCs somewhat sensitive to cell death induction by LPS (Figure 6H) and enhanced the death of HT29 cells in response to TBZ (Figure 6I). These findings are consistent with the notion that the release of ph-MLKL in EVs serves to downregulate the cellular content of ph-MLKL, thereby withholding initiation of necrotic cell death. Extracellular Vesicles Released by BMDCs upon Activation of the Necroptotic Pathway Contain IL-1b Activation of the NLRP3 inflammasome in BMDCs by triggering RIPK3-induced phosphorylation of MLKL yields MLKL-dependent proteolytic processing of the cytoplasmic precursors of IL-1b and IL-18, followed by release of the processed proteins into the extracellular milieu (Figure 7A; Conos et al., 2017; Kang et al., 2013). In other cells, this release occurs partly by rupture of the cell membrane in the course of necrotic death (Eder, 2009; Lopez-Castejon and Brough, 2011). Since RIPK3 activation in the BMDCs does not lead to death, the release of IL-1b and IL-18 from these cells must occur in some other way. Generation of EVs reportedly constitutes one of the ways by which such release can occur without compromising cell viability (Eder, 2009; Lopez-Castejon and Brough, 2011). Our immunoblot analysis of the vesicles released by BMDCs after RIPK3 activation showed that they indeed contained both the (H) Enhancement of induced necroptosis in BMDCs by combined RNA silencing (I) Enhancement of induced necroptosis in HT-29 cells by combined RNA silenci (F, H, and I) Averages of individual values in two independent tests. The bars sho Please also see Figure S7. precursor and processed forms of IL-1b. They also contained both precursor and processed forms of caspase-1, the protease that activates IL-1b by cleaving it (Figure 7B). As with ph-MLKL, the processed form of caspase-1 was barely detectable in the BMDCs themselves despite their abundant presence in the EVs (Figure 7B). As expected from the fact that in this experimental setup the activation of the NLRP3 inflammasome depends on RIPK3-mediated phosphorylation of MLKL, EVs that were generated by non-stimulated BMDCs, or by LPS-stimulated BMDCs deficient in both caspase-8 and MLKL, did not contain the processed forms of either IL-1b or caspase-1. Those vesicles were also found not to contain the precursor form of IL-1b, implying that not only the proteolytic processing of this cytokine but also its incorporation into EVs depends on RIPK3 activation. In contrast, release of the caspase-1 precursor within the EVs was found to occur independently of BMDC stimulation (Figure 7B).
DISCUSSION
Studies of MLKL have until now been largely restricted to the role of this protein in necroptotic death. The present study shows that MLKL also serves a non-deadly function: it associates with the endosomes and provides assistance for endosomal transport and for the generation of ILVs and EVs. This function is manifested at two quantitative grades: a basal, RIPK3-independent contribution, which was revealed when we assessed the consequences of arrest of MLKL expression, and an enhanced grade, observed subsequently to phosphorylation of MLKL by activated of Rab27a and Rab27b. ng of Rab27a and Rab27b. w the range of the results. Immunity 47, 1–15, July 18, 2017 11 RIPK3. Such phosphorylation or, alternatively, expression of MLKLmutants that attain the conformation that RIPK3-mediated phosphorylation imposes on MLKL, was found to enhance MLKL association with the endosomes and to enhance endosomal trafficking. It also resulted in enhanced release of EVs and in extrusion of cellular ph-MLKL in them. This enhancement of endosomal trafficking and of EV formation following RIPK3mediated phosphorylation of MLKL was not a consequence of necroptotic death. In cells sensitive to the necroptotic function of MLKL, we found that the enhancement occurred before MLKL associated with the cell membrane and before cell death. We also found such enhancement in BMDCs, in which necroptotic death was not induced at all. In fact, comparison of the consequences of RIPK3 activation in cells that do die necroptotically to those in BMDCs suggested that the endosomal effect of ph-MLKL serves rather to antagonize its induction of death. Differing vulnerabilities of cells to the necroptotic effect of phMLKL seemed to relate inversely to the effectiveness with which ph-MLKL triggers in them its own extrusion from the cells within EVs. In BMDCs, but not in BMDMs or HT29 cells that are vulnerable to necroptosis, this extrusion was found to be effective enough to result in practical depletion of intracellular ph-MLKL. This might well account for the survival of BMDCs after RIPK3 activation in them. Restriction of signaling by exosomal release of the proteins that mediate it has also been described for some other signaling proteins (Chairoungdua et al., 2010; Verweij et al., 2011). Apart from containing ph-MLKL, EVs released from BMDCs in which RIPK3 was activated were also found to contain processed IL-1b. It thus seems possible that the effect of phMLKL on endosomal trafficking can serve to convert the signaling for necroptosis, a form of death believed to initiate inflammation by release of damage-associated molecular patterns through the ruptured membranes of the dying cells (Pasparakis and Vandenabeele, 2015), to a non-deadly form of intercellular delivery of inflammatory mediators. Mechanisms that self-restrict death induction and instead divert the signaling to non-deadly consequences are known to affect, in addition, the function of some other death-inducing proteins. The death domain of the TNF receptor, for example, can initiate signaling both for death and for mechanisms of resistance to death (Wallach et al., 2008). Further evidence that the endosomal function of MLKL is independent of its necroptotic function emerges from our finding that, besides its mediation of enhanced endosomal transport and vesicle release following its phosphorylation by RIPK3, MLKL also serves a ‘‘housekeeping’’ role by maintaining a basal grade of these activities. This function, unlike its necroptotic function, is independent of RIPK3 activation, or indeed even of mere expression of RIPK3 in the cell. Its housekeeping role is manifested in decreased rates of endosomal transport and of lysosomal degradation of TNF, EGF, and EGFR in MLKLdepleted cells. It is also manifested in a dramatic reduction in the contents harbored by MVBs in these cells and reduced release of vesicles from them. Our assessment of the impacts of MLKL mutations on its role in endosomal trafficking suggested that the structural requirements for this non-deadly function are in part identical to those for death induction and in part distinct from them. Although the 12 Immunity 47, 1–15, July 18, 2017 contribution of MLKL to constitutive EV generation was found to be independent of RIPK3, we found that this contribution was compromised when we mutated the two RIPK3 target residues (T537 and S358) in MLKL. This finding suggested that mediation of this constitutive function does depend on these residues and perhaps also requests their modification. Such modification might be attributable to the function of a protein kinase distinct from RIPK3 (acting too transiently, or at concentrations too low, to be discerned by our use of anti-ph-MLKL antibodies for immunoblot analysis) or to another kind of enzyme that imposes another modification of these residues. Further evidence for the similarity of the structural requirements for MLKL-mediated cell death and for MLKL-facilitated endosomal trafficking was provided by the observed enhancement of EV generation in cells that express full-length MLKLmolecules with mutations that yield the protein’s conformation attained upon its phosphorylation. It is also evidenced by the fact that both functions are obliterated by the ‘‘5A’’ mutation, which interferes with binding of the N terminus of MLKL to negatively charged lipids. These findings suggested that, similarly to the necroptotic function of MLKL, this protein’s effect on endosomal function depends on its binding to certain lipids. Several lipid-binding domains, such as the FYVE and the PX (phox) domains, are known to contribute to the functions of proteins that participate in the regulation of endosomal dynamics (Bissig and Gruenberg, 2013). It might well be that, likewise, the mere association of MLKL with certain lipids contributes to its effect on endosomal trafficking and on ILV generation. However, we also noticed one marked difference between the structural requirements for the necroptotic function of MLKL and its function in maintaining EV generation. The necrotic activity of MLKL is fully mediated by its N-terminal 4HB domain, as shown by the fact that the MLKL deletion mutant ‘‘1 180,’’ which corresponds to this domain, is capable of triggering cell death in the absence of any stimulus (Chen et al., 2014; Dondelinger et al., 2014; Hildebrand et al., 2014). If the mechanism for the facilitation of EV generation byMLKL were identical to the mechanism of its necroptotic function, expression of that deletion mutant would be expected to dictate generation of EVs and to do so more effectively than the wild-type MLKL. We repeatedly found, however, that cells expressing the 1 180 mutant generate much lower amounts of EVs than the amounts generated by cells expressing thewild-type protein. This finding raised the possibility that the endosomal function of MLKL, in addition to its dependence on the interaction of its N terminus with lipids, is also affected by some interaction(s) of its C-terminal region. These latter interactions perhaps account for the observed association of MLKL with ESCRT proteins and the flotillins. The relationship of the EVs whose generation was found here to be affected by MLKL to EVs described in numerous other studies remains to be further clarified. Currently, it is customary to distinguish between two classes of such vesicles: (1) those derived from the ILV and termed exosomes and (2) shed plasma-membrane vesicles, dubbed microvesicles or ectosomes (Colombo et al., 2014). The latter are larger than exosomes but have similar composition. The impact of MLKL deficiency on endosomal function and on ILV generation suggests that it affects the generation of exosomes. The association of MLKL with the endosomes and its occurrence inside the MVBs suggest that it is also extruded from cells within exosomes. The size distribution of the EVs generated by the cells used in this study—both those generated by nonstimulated cells and those generated by cells in which RIPK3 phosphorylation was enhanced—is indeed characteristic of exosomes. The impact of Rab27a and Rab27b RNA silencing on EV generation and on the effectiveness of necroptosis is also indicative of the identity of the EVs with exosomes. However, a recent study by Gong et al. (2017), which was published while our paper was under evaluation, showed that MLKL activation also prompts shedding of vesicles derived from damaged plasma membranes. It was suggested that the shedding serves as a means of repair of the cellular damage caused by MLKL. Those shed membranes were shown to be ‘‘broken,’’ implying that they are permeable to macromolecules. In that respect, they seem to be distinct from the EVs that we found to harbor ph-MLKL, in which neither MLKL nor TSG101 was accessible to external trypsin. Taken together, our findings and those of Gong et al. (2017) suggest that the subcellular sites of EV generation upon activation of RIPK3 might be heterogeneous. They also suggest, however, that the mechanisms controlling the generation of these various vesicles are similar. The ESCRT proteins, which we found to associate withMLKL and are known to control endosomal trafficking and contribute to the translocation of cargo proteins into exosomes (Williams and Urbé, 2007), are also required for the shedding of damaged plasma-membrane patches (Gong et al., 2017; Jimenez et al., 2014). Prompted by the increasing evidence that necroptotic death contributes to the pathology of a number of severe diseases (Pasparakis and Vandenabeele, 2015; Zhao et al., 2015), attempts are now underway to develop drugs that block MLKL function. Our finding that MLKL, besides mediating necroptotic death, also contributes to endosomal trafficking raises the need to further delineate the differences between the structural requirements for mediation of these different MLKL functions. Based on such a distinction, it might be possible to develop drugs that will arrest the necroptotic effect of MLKL without compromising its contribution to endosomal function. STAR+METHODS Detailed methods are provided in the online version of this paper and include the following: d KEY RESOURCES TABLE d CONTACT FOR REAGENT AND RESOURCE SHARING d EXPERIMENTAL MODEL AND SUBJECT DETAILS B Mice B Cell Culture d METHOD DETAILS B Reagents B Reagents for RNA Interference B Antibodies B Inducible Expression of MLKL Mutants and Inducible Deletion of the Casp8 Gene B Collection and Quantification of Extracellular Vesicles B Ligand and Receptor Uptake Assays, and Assessment of Phosphorylation of Signaling Proteins B Immunoprecipitation and Western Blot Analysis B Mass Spectrometry-Based Proteomics Analysis B Fluorescence Microscopy B Transmission Electron Microscopy B Immunogold Electron Microscopy B Quantification of Cell Death B Real-Time PCR Analysis B Expression Analysis of NanoString Inflammatory Panel RNA d QUANTIFICATION AND STATISTICAL ANALYSIS SUPPLEMENTAL INFORMATION Supplemental Information includes seven figures and one table and can be found with this article online at http://dx.doi.org/10.1016/j.immuni.2017. 06.001.
AUTHOR CONTRIBUTIONS
S.Y., A.K., and D.W. were responsible for project conception, data analysis, andmanuscript editing. K.B. generated DNA constructs. All other experiments were performed by S.Y. The manuscript was written by D.W., who also supervised the study.
ACKNOWLEDGMENTS
We thank Dr. Tae-Bong Kang for useful advice, Helena Sabany for advice and help with the electron microscopic analysis, Vladimir Kiss and Dr. Reinat Nevo for advice on use of the immunocytochemical analysis, Tatiana Shalevich for maintaining cultured cells, and both Inna Kolesnik and Shoshana Grossfeld for genotyping the mice. The mass spectrometry-based proteomics analysis was conducted by Dr. Alon Savidor at the De Botton Institute for Protein Profiling, The Nancy & Stephen Grand Israel National Center for Personalized Medicine, The Weizmann Institute of Science. Received: October 28, 2016 Revised: April 14, 2017 Accepted: May 31, 2017 Published: June 27, 2017
SUPPORTING CITATIONS
The following reference appears in the Supplemental Information: Chiou and Ansel, 2016.
KEY RESOURCES TABLE
REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies Mouse anti-Caspase-1 (p20) mAb AdipoGen Cat# AG-20B-0042 RRID:AB_2490248 Rabbit anti-Human MLKL Polyclonal Antibody GeneTex Cat# GTX107538, RRID:AB_2037439 Rabbit anti-mouse MLKL (C-TERM) antibody Sigma-Aldrich Cat# SAB1302339 Mouse anti-beta-Actin Monoclonal Antibody, Clone AC-15 Sigma-Aldrich Cat# A5441, RRID:AB_476744 Rabbit anti-MAP Kinase (ERK-1, ERK-2) antibody Sigma-Aldrich Cat# M5670, RRID:AB_477216 Mouse anti-ERK-1 / ERK-2, diphospho Monoclonal Antibody, Clone MAPK-YT Sigma-Aldrich Cat# M8159, RRID:AB_477245 Rabbit anti-MVB12B Sigma-Aldrich Cat# HPA043683 Rabbit anti-Human Annexin II Abcam Cat# ab41803, RRID:AB_940267 Rabbit anti-human MLKL (phospho S358), Clone EPR9514 Abcam Cat# ab187091, RRID:AB_2619685 Rabbit anti-mouse MLKL (phospho S345), Clone EPR9515(2) Abcam Cat# ab196436 Rabbit anti GFP antibody Abcam Cat# ab6556, RRID:AB_305564 Rabbit anti-VDAC1 / Porin antibody Clone EPR10852(B) Abcam Cat# ab154856 Mouse anti-Flotillin-1, Clone 18 BD Biosciences Cat# 610820, RRID:AB_398139 Mouse anti-Flotillin-2, Clone 29 BD Biosciences Cat# 610383, RRID:AB_397766 Mouse anti-TSG101 Clone 51 BD Biosciences Cat# 612696, RRID:AB_399936 Mouse anti-Phosphotyrosine Clone PY20 BD Biosciences Cat# 61000 Mouse anti-human Rab27 Clone 20/RAB27A BD Biosciences Cat#558532 Rabbit anti-RIP3 (E1Z1D) mAb Cell Signaling Technology Cat# 13526 Rabbit anti-Akt Antibody Cell Signaling Technology Cat# 9272, RRID:AB_329827 Mouse anti-phospho-Akt (Ser473) (587F11) mAb antibody Cell Signaling Technology Cat# 4051, RRID:AB_331158 Mouse anti-phospho-Ikappa B alpha (Ser32/36) (5A5) mAb antibody Cell Signaling Technology Cat# 9246L, RRID:AB_2267145 Rabbit anti-p38 MAPK, phospho (Thr180 / Tyr182) Monoclonal Antibody, Clone 12F8 Cell Signaling Technology Cat# 4631L, RRID:AB_331766) Rabbit anti-phospho-NF-kappaB p65 (Ser536) Antibody Cell Signaling Technology Cat# 3031, RRID:AB_330559 Rabbit anti-phospho-Stat-3 (tyr705) antibody Cell Signaling Technology Cat# 9131, RRID:AB_331586 Rabbit anti-HSP70 antibody System Biosciences Cat# EXOAB-Hsp70A-1 Rabbit anti-CD9 antibody System Biosciences Cat# EXOAB-CD9A-1 Mouse anti-Alix Purified Monoclonal Antibody, Clone 3A9 BioLegend Cat# 634502, RRID:AB_2162471 Hrs and Hrs-2, mAb (A-5) antibody Enzo Life Sciences Cat# ALX-804-382-C050, RRID:AB_2051626 Mouse anti-EGFR Monoclonal antibody, Clone 6f1 Enzo Life Sciences Cat# CSA-330E, RRID:AB_1083161 Rabbit anti-caspase-1 p10 (M-20) antibody Santa Cruz Biotechnology Cat# sc-514, RRID:AB_2068895 Goat anti-I kappa B alpha Polyclonal antibody Santa Cruz Biotechnology Cat# sc-371, RRID:AB_2235952 Mouse anti-NFkappaB p65 (F-6) antibody Santa Cruz Biotechnology Cat# sc-8008, RRID:AB_628017 Mouse anti-Stat3 (F-2) antibody Santa Cruz Biotechnology Cat# sc-8019, RRID:AB_628293 Rabbit anti-mouse Rab 27A antibody Synaptic Systems Cat# 168 013, RRID:AB_887766 (Continued on next page) e1 Immunity 47, 1–15.e1–e7, July 18, 2017
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER Rabbit anti-mouse Rab 27B antibody Synaptic Systems Cat# 168 103, RRID:AB_887767 Mouse IL-1 beta/IL-1F2 Affinity Purified Polyclonal Ab antibody R and D Systems Cat# AF-401-NA, RRID:AB_416684 Rabbit anti-human RAB27B antibody Proteintech Group Cat# 13412-1-AP, RRID:AB_2176732 Sheep anti-CHMP3 (N-terminal) polyclonal antibody Creative Diagnostics Cat# DPABH-18107, RRID:AB_2514553 Rabbit anti-VPS25 (aa 116-144) polyclonal antibody Creative Diagnostics Cat# DPABH-05576, RRID:AB_2521296 Rabbit anti-VPS37A Antibody LSBio Cat# LS-C409401 Rabbit anti-VPS28 Antibody Novus Cat# NBP1-85976, RRID:AB 11004858 Peroxidase-IgG Fraction Monoclonal Mouse Anti-Sheep IgG, Light Chain Specific antibody Jackson ImmunoResearch Labs Cat# 213-032-177, RRID:AB_2339251 Peroxidase-AffiniPure Donkey Anti-Goat IgG (H+L) antibody Jackson ImmunoResearch Labs Cat# 705-035-147, RRID:AB_2313587 Peroxidase-AffiniPure Goat Anti-Mouse IgG, F(ab’)2 Fragment Specific Jackson ImmunoResearch Labs Cat# 115-035-072, RRID:AB_2338507 Peroxidase-AffiniPure Goat Anti-Mouse IgG, Light Chain Specific Jackson ImmunoResearch Labs Cat# 115-035-174, RRID:AB_2338512 Peroxidase-AffiniPure Goat Anti-Rabbit IgG (H+L) antibody Jackson ImmunoResearch Labs Cat# 111-035-003, RRID:AB_2313567 Rabbit anti-EGF Receptor (D38B1) XP mAb (Alexa Fluor 647 Conjugate) antibody Cell Signaling Technology Cat# 5588S, RRID:AB_10694773 Rabbit anti-RAB7 antibody [EPR7589] Abcam Cat# ab137029, RRID:AB_2629474 Mouse Anti-EEA1 Monoclonal Antibody, Clone 14 BD Biosciences Cat# 610456, RRID:AB_397829 Cy2-AffiniPure Goat Anti-Rabbit IgG (H+L) (min X Hu,Ms,Rat Sr Prot) antibody Jackson ImmunoResearch Labs Cat# 111-225-144, RRID:AB_2338021 Goat anti-Mouse IgG, Cy3 conjugate antibody Millipore Cat# AP124C, RRID:AB_92459 H5C6 (CD63) antibody, deposited by August, J.T. / Hildreth, J.E.K. DSHB Cat# h5c6, RRID:AB_528158 Sheep Anti-Epidermal Growth Factor (EGF) Receptor Polyclonal Antibody Fitzgerald Industries International Cat# 20-ES04, RRID:AB_231428 12 nm Colloidal Gold-AffiniPure Donkey Anti-Sheep IgG (H+L) antibody Jackson ImmunoResearch Labs Cat# 713-205-147, RRID:AB_2340733 18nm Colloidal Gold-AffiniPure Goat Anti-Mouse IgG (H+L) antibody Jackson ImmunoResearch Labs Cat# 115-215-166, RRID:AB_2338739 Chemicals, Peptides, and Recombinant Proteins Bafilomycin A1 Santa Cruz Biotechnology Cat# SC-201550A Chloroquine Sigma-Aldrich Cat# C6628 4-hydroxytamoxifen Sigma-Aldrich Cat# H7904 Ionomycin Sigma-Aldrich Cat# I9657 Brij O10 (POLYOXYETHYLENE 10 OLEOYL ETHER) Sigma-Aldrich Cat# P6136 Human EGF Prospec Cat# CYT-217 Mouse biotin-EGF Prospec Cat# CYT-841 human TNF YBDY Biotech Cat# REC110 EZ-Link Sulfo-NHS-biotin Thermo Fisher Scientific Cat# 21326 BSA-gold CMC Utrecht Cat# BSA 5nm The bivalent IAP antagonist BV6 WuXi AppTec N/A z-VAD-fmk WuXi AppTec N/A Strep-Tactin Sepharose 50% suspension IBA Cat# 2-1201-010 Mouse Dendritic Cell Nucleofector Kit Lonza Cat# VVPA-1011 Rhodamine Phalloidin Thermo Fisher Scientific Cat# R415 Lipofectamin 3000 Transfection reagent Thermo Fisher Scientific Cat# L3000015 ( on next page) Immunity 47, 1–15.e1–e7, July 18, 2017 e2
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER Critical Commercial Assays Mouse IL-1beta Platinum ELISA 2 3 96 tests Kit antibody Thermo Fisher Scientific Cat# BMS6002TWO, RRID:AB_2575605 Cytotoxicity Detection Kit (LDH) Sigma-Aldrich Cat# 11644793001 PerfeCta FastMix II Quanta Biosciences Cat# 95119-012 SuperSignal West Pico Chemiluminescent Substrate Thermo Fisher Scientific Cat# 34077 Streptavidin Protein, HRP Thermo Fisher Scientific Cat# 21124 BCA protein assay kit Thermo Fisher Scientific Cat# 23225 Experimental Models: Cell Lines Human: HT-29 ATCC Cat# HTB-38, RRID:CVCL_0320 Human: Hep-G2 ATCC Cat# HB-8065, RRID:CVCL_0027 Human: HeLa ATCC Cat# CRL-7923, RRID:CVCL_0030 Murine Embryonic fibroblasts (MEFs) Laboratory of D. Wallach (Krelin et al., 2008) Mouse bone marrow-derived macrophages (BMDMs) Laboratory of D. Wallach (Kang et al., 2004) Mouse bone marrow-derived dendritic cells (BMDCs) Laboratory of D. Wallach (Kang et al., 2013) Experimental Models: Organisms/Strains Mouse: C57BL/6J The Jackson Laboratory RRID:IMSR_JAX:000664 Mouse: 129/SvEv-C57BL/6 (Mlkl / ) Taconic Cat# TF2780 Mouse: Ripk3 / Laboratory of V. Dixit (Newton et al., 2004) Mouse: Ripk1 / Laboratory of M. Kelliher (Kelliher et al., 1998) Mouse: Casp8 / Laboratory of D. Wallach (Varfolomeev et al., 1998) Mouse: Casp8 fl/fl Laboratory of D. Wallach (Kang et al., 2004) Mouse: Expressing Cre under control of the integrin alpha X gene (Itgax or Cd11c) promoter region Laboratory of B. Reizis (Caton et al., 2007) Mouse: B6.129S-Tnftm1Gkl/J (Tnf / ) The Jackson Laboratory RRID:IMSR_JAX:005540 Oligonucleotides Human Rab27a RNAi duplex (50-UAUGUUUGUCCCA UUGGCAGCTT-30) Integrated DNA Technologies N/A Human Rab27b RNAi duplex(50-UACUGUAGUGAUG AAUUUGGGTT-30) Integrated DNA Technologies N/A Mouse Rab27a RNAi duplex (50-CCAGUUUAAGAGAA GUGUU-30) Integrated DNA Technologies N/A Mouse Rab27b RNAi duplex (50- CUGAGACAAUG UCAAACCA-30) Integrated DNA Technologies N/A 30-UTR-targeting lentiviral human MLKL shRNA Sigma-Aldrich Cat# TRCN0000003227 Lentiviral mouse MLKL shRNA Sigma-Aldrich Cat# TRCN0000022599 Recombinant DNA GEV16/pF5x UAS system (Dunning et al., 2007) N/A pBABE-puro (Morgenstern and Land, 1990) Addgene plasmid #1764 pBABE-puro-STR2-3xFlag Laboratory of D. Wallach; described in (Gloeckner et al., 2007) N/A Software and Algorithms NanoSight NTA 3.1 Malvern https://www.malvern.com/en/support/ events-and-training/webinars/ W150326NanoSightSoftwareRelease.html Expressionist software version 10.5 Genedata https://www.genedata.com/products/ expressionist/ PeptideProphet algorithm (Keller et al., 2002) Imaris 8.3 Bitplane http://www.bitplane.com/imaris8 ImageJ NIH https://imagej.nih.gov/ij/download.html ( on next page) e3 Immunity 47, 1–15.e1–e7, July 18, 2017
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER Other MiniCollect TUBEs Greiner Bio-One Cat# 450475 CO2-independent medium Thermo Fisher Scientific Cat# 18045-054 Medium kit with serum and without growth factor Cell systems Cat# 4Z0-500-S
CONTACT FOR REAGENT AND RESOURCE SHARING
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, David Wallach (d.wallach@weizmann.ac.il).
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Mice
on aC57BL/6 backgroundwere used in this study.Mlkl / micewere obtained from Taconic.Ripk3 / micewere obtained from Dr. Vishva Dixit (Newton et al., 2004) and Ripk1 / mice were from Dr. Michelle Kelliher (Kelliher et al., 1998). Casp8 / and Caspfl/fl were established in our laboratory (Kang et al., 2004; Varfolomeev et al., 1998). Mice expressing Cre under control of the integrin alpha X gene (Itgax or Cd11c) promoter region were from Dr. Boris Reizis (Caton et al., 2007). Tnf / mice (strain B6; 129S6Tnftm1Gkl/J) were obtained from The Jackson Laboratory. All animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of The Weizmann Institute of Science.
Cell Culture
Cells of the human HT-29 colorectal adenocarcinoma line were grown in McCoy’s 5A medium. Cells of the HeLa cervical adenocarcinoma line and of the HepG2 hepatocellular carcinoma line, as well as mouse embryonic fibroblasts (MEFs) immortalized by expression of the SV40 large T antigen, were cultured in Dulbecco’s modified Eagle’s medium. Both media were supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 mg/ml streptomycin. Mouse bone marrow-derived macrophages (BMDMs), as well as wild-type and caspase-8-deficient mouse bone marrow-derived dendritic cells (BMDCs), were produced as described previously (Kang et al., 2004; Kang et al., 2013).
METHOD DETAILS
Reagents
Bafilomycin A1 from Santa Cruz Biotechnology was applied to the cells at 100 nM. Chloroquine applied at 25 mM, 4-hydroxytamoxifen applied at 1 mM, ionomycin applied at 1 mMand Brij were from Sigma. Human EGF from Prospec, mouse EGF from ProSpec and human TNF from YBDY Biotech were biotinylated using an EZ-Link biotinylation kit (Thermo Fisher Scientific) and then applied to cells at 5 mg/ml or injected intravenously into mice (10 mg/mouse). Bovine serum albumin (BSA) tagged with gold particles (BSA gold, CMC Utrecht) was applied to cells at an optical density of 2. To trigger necroptosis, TNF (1000 units/ml) was applied together with the bivalent IAP (inhibitor of apoptosis protein) antagonist BV6 (Varfolomeev et al., 2007) and the caspase inhibitor z-VAD-fmk, both from WuXi App Tec, at concentrations of 1 mM and 20 mM, respectively. IL-1b ELISA kit and Streptavidin-HRP (21124) were from Thermo Fisher Scientific. Strep-Tactin beads were from IBA Life Sciences and Rhodamine-Phalloidin (R415) from Thermo Scientific.
Reagents for RNA Interference
To siRNA silence the expression of Rab27a and Rab27b in human HT-29 cells, we transfected these cells with RNAi duplex (Rab27a, 50-UAUGUUUGUCCCAUUGGCAGCTT-30; Rab27b, 50-UACUGUAGUGAUGAAUUUGGGTT-30) (Integrated DNA Technologies), using Lipofectamine 3000 Reagent (Thermo Fisher Scientific). To siRNA silence the expression of Rab27a and Rab27b in BMDCs, we transfected these cells with siRNA duplex (Rab27a, 50-CCAGUUUAAGAGAAGUGUU; mouse Rab27b, 50-CUGAGACAAUGU CAAACCA-30, Integrated DNA Technologies) using Mouse Dendritic Cell Nucleofector Kit (Lonza). To siRNA silence the expression of human MLKL we used 30-UTR-targeting lentiviral shRNA (Sigma). Use of this shRNA, which binds to the 30 UTR of MLKL, allowed us to re-express the cDNAs of various structural mutants of MLKL in siRNA silenced cells by reconstituting the cells with cDNAs corresponding only to the coding region of the MLKL mutants. To siRNA silence mouse MLKL expression we used lentiviral shRNA (Sigma).
Antibodies
The following antibodies were applied for western blotting analysis: anti-caspase-1 (AG-20B-0042) from AdipoGen; anti-human MLKL (GTX107538) from GeneTex; anti-mouse MLKL (Sab1302339), anti b-actin (A5441), anti-ERK (M5670), anti-phospho-ERK Immunity 47, 1–15.e1–e7, July 18, 2017 e4 (M8159) and anti-MVB12B (HPA043683) from Sigma; anti-Annexin II (ab41803), anti-human phospho MLKL (ab187091), anti-mouse phospho-MLKL (ab196436), anti-GFP (ab6556) and anti-VDAC (ab154856) from Abcam; anti-flotillin-1 (610820), anti-flotillin-2 (610383), anti-TSG101 (612696), anti-phosphotyrosine (61000) and anti-human Rab 27a (558532) from BD Biosciences; anti-human RIPK3 (13526), anti-AKT (9272), anti-phospho AKT (4051), anti-phospho IkBa (9246), anti-phospho p38 kinase (4631), anti-phospho p65 (3031) and anti-phospho-STAT3 (9131) fromCell Signaling; anti-HSP70 (EXOAB-hsp70A-1) and anti-CD9 (EXOAB-CD9A-1) from System Biosciences; anti-Alix (3A9) from BioLegend; anti-Hrs (A-5) and anti-EGFR (6F1) from Enzo Life Sciences; anti-caspase-1 (SC-514), anti IkBa (SC-371), anti-p65 (SC-8008) and anti-STAT3 (SC-8019) from Santa Cruz; anti-mouse Rab 27a (168013) and anti-mouse Rab 27b (168103) from Synaptic Systems, anti-IL-1b from R&D Systems (AF-401-NA) and anti-human Rab 27b (13412-1AP) from Proteintech, anti-VPS25 (DPABH-18107) and anti-CHMP3 (DPABH-18107) from Creative diagnostics, antiVPS37A (LS-C409401) from LSbio, and anti-VPS28 (NBP1-85976) from Novus biologicals. HRP-conjugated antibodies were from Jackson ImmunoResearch. For immunofluorescenceanalysisweusedEGFRconjugated toAlexa647 (5588,CellSignaling); antibodiesagainstRab7 (ab137029, Abcam) and EEA1 (610456, BD Biosciences); Cy2-conjugated goat anti-rabbit IgG (111-225-144, Jackson ImmunoResearch), and Cy3-conjugated anti-mouse IgG (AP124C, EMDMillipore). For immune electron microscopy we used anti-CD63 (MEM-259, DSHB) and anti-EGFR (20-ES04, Fitzgerald). Secondary antibodies were 12 nm colloidal gold-conjugated donkey anti-sheep IgG (713-205-147), and 18 nm colloidal gold-conjugated goat anti-mouse IgG (115-215-166, Jackson ImmunoResearch). Inducible Expression of MLKL Mutants and Inducible Deletion of the Casp8 Gene The various mutants of MLKL were expressed inducibly in cells in which the endogenous MLKL mRNA was constitutively RNA silenced. The cells were infected with lentiviral vectors encoding MLKL in the GEV16/pF5x UAS system (Dunning et al., 2007) and induced to express MLKL by their treatment with 4 hydroxytamoxifen (1 mM) for the indicated durations. Inducible deletion of ‘floxed’ caspase-8 gene inMEFs was achieved by expressing the enzyme Cre in these cells for 48 hr using the GEV16/pF5x UAS system as above.
Collection and Quantification of Extracellular Vesicles
In all experiments aimed at assessing the generation of EVs by cultured cells, the FBS supplementing the growth medium was predepleted of bovine exosomes by centrifugation at 100,000 3 G for 18 hr. The EVs generated by unstimulated cells were collected after incubation for 12 hr. In cells stimulated with TBZ or LPS the cell-growth media were replaced with fresh media immediately before the stimulants were applied and were collected at the end of the stimulation period. Unless otherwise indicated, TBZ was applied for 4 hr and LPS for 3 hr. For recovery of EVs from the mice, mouse blood samples were collected in MiniCollect TUBEs (Greiner Bio-One) and centrifuged at 3000 rpm for 15min. The EVswere isolated from these samples of mouse plasma,mouse serum and cell-growth media, as described before (Ostrowski et al., 2010; Thery et al., 2006), by passage through 0.2 mm filters followed by centrifugation at 100,0003G for 90min, and the pellet was then suspended in phosphate-buffered saline (PBS) and resedimented by centrifugation as above. Size spectra and amounts of particles in the EV preparations were determined by Nanoparticle Tracking Analysis (NTA) using the NanoSight NS300 device (Malvern Instruments) according to the manufacturer’s instructions. Samples of particles derived from 2.5 3 107 cultured cells or from 0.8 mL of plasma were applied for particle quantification by NTA. Samples derived from 2.5 3 107 HT-29 cells or from 5 3 107 BMDCs were applied for particle size spectra analysis by NTA. For western blot analysis, samples of particles derived from 2.5 3 107 cells or from 3.2 mL of plasma were applied to each lane.
Ligand and Receptor Uptake Assays, and Assessment of Phosphorylation of Signaling Proteins
Prior to ligand treatment the cells were incubated for 12 hr in serum-free medium. Biotinylated EGF and TNF were applied to the cells for 30 min on ice in CO2-independent medium (Thermo Fisher Scientific), and this was followed by rinsing and further incubation without those ligands at 37 C for the indicated times, either in normal growth medium plus FCS or, where indicated, with Medium Kit with Serum and Without Growth Factors (Cell System). For immunofluorescence microscopy, cells were fixed in 4% paraformaldehyde (PFA) in PBS and then stained with antibodies conjugated to a fluorescent dye. For western blot analysis, cell and liver samples were extracted at the indicated times in RIPA buffer (20mMTris-HCl pH 7.5, 150mMNaCl, 1mMEDTA, 1mMEGTA, 1%NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), protease and phosphatase inhibitors) and in 1%SDS, respectively. Protein concentrations in the samples were determined using the BCA protein assay kit (Thermo Fisher Scientific).
Immunoprecipitation and Western Blot Analysis
Interaction of MLKL with the proteins found in the extracellular vesicles (EVs) was assessed by, extraction of EVs for 10 min at 4 C in RIPA buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS), followed by centrifugation for 10min at 180003G and immunoprecipitation of MLKL along with its bound proteins. Samples (8 mg) of proteins from EV extracts and of proteins immunoprecipitated from 150 mg aliquots of the EV extracts were analyzed by SDS polyacrylamide gel electrophoresis (PAGE). To assess the interaction of these proteins with MLKL within the cell we used HT-29 cells constitutively expressing MLKL that was fused C-terminally to the STR2-3xFlag tag (Gloeckner et al., 2007). The cells were extracted for 1 hr on ice with a buffer containing 1% Brij, 10 mM Tris pH 7.4, 150 mM NaCl, 1 mM MgCl2, 1 mM EDTA, and protease and phosphatase inhibitors (1 mM PMSF, 40 mM beta-glycerophosphate, 50 mM NaF, and 1 mM sodium vanadate). e5 Immunity 47, 1–15.e1–e7, July 18, 2017 Following centrifugation at 16,0003 G for 10 min, the supernatant was cleared for 6 hr by the use of protein-G beads and was then further incubated with Strep-Tactin beads for 16 hr. The beads were rinsed 5 times in a washing buffer containing 0.5% NP-40, 20 mM Tris-HCl pH 7.4, 150 mMNaCl and 1 mM EDTA, and the bound proteins were eluted by incubation in washing buffer to which 1 mM of biotin was added. Samples (10 mg) of proteins from the cell extracts and of proteins immunoprecipitated from 6 mg aliquots of the extracts were analyzed by PAGE. For immunoblotting, protein samples were loaded on SDS PAGE gels, electrophoresed, and transferred onto nitrocellulosemembranes (Bio-Rad). The membranes were treated for 1 hr at room temperature with PBS containing 5% skimmed milk and 0.05% Tween-20, or, in the case of immunoblotting with anti-mouse ph-MLKL antibody, with PBS containing 1% BSA and 0.05% Tween-20, and further incubated for 16 hr at 4 C either with the primary antibody in PBS containing 5% BSA or (for detection of biotinylated TNF or EGF) with horseradish peroxidase (HRP)-linked streptavidin (Strep-HRP) in PBS containing 0.025% Tween-20 (PBS Tween). After threewashes in PBS Tween themembraneswere incubated for 1 hr at room temperature with HRP-conjugated secondary antibody in PBS Tween. After threemorewashes in PBS Tween, blots were developed using the SuperSignalWest Pico Chemiluminescent Substrate (Thermo Fisher Scientific). To quantify band intensity we used ImageJ software.
Mass Spectrometry-Based Proteomics Analysis
Sample preparation
Samples of the proteins immunoprecipitated with anti-MLKL antibody from RIPA extracts of EVs as above were subjected to insolution, on-bead, tryptic digestion. 8M urea in 0.1M Tris, pH 7.9 was added onto PBS washed beads and incubated for 15min in room temperature. Proteins were reduced by incubation with dithiothreitol (5 mM; Sigma) for 60 min at room temperature, and alkylated with 10 mM iodoacetamide (Sigma) in the dark for 30 min at room temperature. Urea was diluted to 2M with 50mM ammonium bicarbonate. 250ng trypsin (Promega; Madison, WI, USA) was added and incubated overnight at 37 C followed by addition of 100ng trypsin for 4hr at 37 C. Digestions were stopped by addition of trifluroacetic acid (1% final concentration). Following digestion, peptides were desalted using Oasis HLB mElution format (Waters, Milford, MA, USA), vacuum dried and stored in 80 C until further analysis. Liquid chromatography ULC/MS grade solvents were used for all chromatographic steps. Each sample was loaded using split-less nano-Ultra Performance Liquid Chromatography (10 kpsi nanoAcquity; Waters, Milford, MA, USA). The mobile phase was: A) H2O + 0.1% formic acid and B) acetonitrile + 0.1% formic acid. Desalting of the samples was performed online using a reversed-phase Symmetry C18 trapping column (180 mm internal diameter, 20mm length, 5 mmparticle size;Waters). The peptides were then separated using a T3HSS nanocolumn (75 mm internal diameter, 250 mm length, 1.8 mm particle size; Waters) at 0.35 mL/min. Peptides were eluted from the column into the mass spectrometer using the following gradient: 4% to 35%B in 65 in, 35% to 90%B in 5 min, maintained at 90% for 5 min and then back to initial conditions. Mass Spectrometry The nanoUPLCwas coupled online through a nanoESI emitter (10 mm tip; NewObjective; Woburn, MA, USA) to a quadrupole orbitrap mass spectrometer (Q Exactive HF, Thermo Scientific) using a FlexIon nanospray apparatus (Proxeon). Data was acquired in data dependent acquisition (DDA) mode, using a Top20 method. MS1 resolution was set to 120,000 (at 200 m/z), mass range of 300-1650 m/z, AGC of 3e6 and maximum injection time was set to 20msec. MS2 resolution was set to 30,000, quadrupole isolation 1.7 m/z, AGC of 1e6, dynamic exclusion of 60sec and maximum injection time of 60msec. Data processing and analysis Raw data was imported into the Expressionist software version 10.5 (Genedata) and processed as described (Shalit et al., 2015). The software was used for retention time alignment and peak detection of precursor peptides. Amaster peak list was generated from all MS/MS events and sent for database searching using Mascot v2.5.1 (Matrix Sciences). Data was searched against the human sequences from UniprotKB (http://www.uniprot.org/) version 2016_02 appended with common laboratory contaminant proteins. Fixed modification was set to carbamidomethylation of cysteines and variable modifications were set to oxidation of methionines and deamidation of N or Q. Search results were then filtered using the PeptideProphet algorithm to achievemaximum false discovery rate of 1% at the protein level. Peptide identifications were imported back to Expressionist to annotate identified peaks. Quantification of proteins from the peptide data was performed using an in-house script. Data was normalized base on the total ion current. Protein abundance was obtained by summing the three most intense, unique peptides per protein. A Student’s t-Test, after logarithmic transformation, was used to identify significant differences across the biological replica. Fold changes were calculated based on the ratio of arithmetic means of the case versus control samples.
Fluorescence Microscopy
Immunostaining and its microscopic analysis were performed as described previously (Yoon et al., 2016).
Transmission Electron Microscopy
HepG2 cells were incubated for 30 min at 37 C in cell-growth medium containing bovine serum albumin (BSA) tagged with gold (BSA gold) and 100 ng/ml EGF, and then fixed for 2 hr with Karnovsky’s fixative (4% PFA, 2% glutaraldehyde, 5 mM CaCl2 in 0.1 M cacodylate buffer, pH 7.4). The fixed cells were scraped, pelleted by centrifugation at 400 3 G, embedded in 1.7% Nobel agar, and postfixed with 1% osmium tetroxide, 0.5% potassium dichromate and 0.5% potassium hexacyanoferrate in 0.1 M Immunity 47, 1–15.e1–e7, July 18, 2017 e6 cacodylate buffer. The pellet was stained and blockedwith 2%aqueous uranyl acetate, then dehydratedwith ethanol and embedded in graded Epon 812. Ultrathin sections (70 100 nm) were cut with Leica Ultracut UCT ultramicrotome and analyzed using an FEI T12 Spirit electron microscope. Images were obtained with an Eagle CCD camera and processed by ImageJ software.
Immunogold Electron Microscopy
Cells were fixed for 2 hr at room temperature in freshly prepared 3%PFA, 0.1% glutaraldehyde in 0.1 M cacodylate buffer containing 5 mM CaCl2. Pelleted fixed cells were infiltrated for 30 min in 10% gelatin at 37 C. Excess gelatin was removed by centrifugation at 400 3 G at 37 C, and this was followed by incubation at 4 C for 24 hr. The fixed cell pellets were cryoprotected by overnight infiltration with 2.3 M sucrose in cacodylate buffer, then frozen by injection into liquid nitrogen. Ultrathin (75 nm) frozen sections were sliced at 110 C on a Leica EM FC6 ultramicrotome and transferred to Formvar-coated 200-mesh nickel grids. Sections were treated for 5minwith 3%NGS, 0.5%BSA, 0.1%glycine and 1%Tween 20 in PBS to block nonspecific binding, and this was followed by incubation for 2 hr with the primary antibodies. After extensive rinsing with 0.1% glycine in PBS (PBS glycine) the cells were further incubated for 30 min in colloidal gold-conjugated rabbit anti-mouse antibody. The grids were then washed in PBS glycine and stained with 2% methyl cellulose and 2% uranyl acetate.
Quantification of Cell Death
Cell death was quantified based on the concentration of lactic dehydrogenase (LDH) in the mouse cell media, plasma or serum, determined by use of the Cytotoxicity Detection Kit (Sigma-Aldrich).
Real-Time PCR Analysis
mRNAs were extracted using the RNeasy kit (QIAGEN), with quantified with PerfeCta FastMix II (Quanta Biosciences) on a StepOnePlusTM real-time PCR system. The results are presented as described (Kovalenko et al., 2009).
Expression Analysis of NanoString Inflammatory Panel RNA
Samples of RNA isolated from HT29 cells were hybridized to the NanoString nCounter Human Inflammation v2 code set. This code set contains probes against a panel of 249 genes encoding key inflammatory mediators. Scanning and data collection onto a digital analyzer were followed by data normalization against positive and negative control oligonucleotides and six housekeeping genes. Normalized results are represented as the relative mRNA amounts.
QUANTIFICATION AND STATISTICAL ANALYSIS
Except where otherwise indicated, all the presented data are representative results of at least two independent experiments. In all diagrams with error bars the values correspond to mean values and the bars show either the range of the results (in the case of duplicate samples) or standard variations (in the case of larger numbers of samples). e7 Immunity 47, 1–15.e1–e7, July 18, 2017
Article Images (0)
×![]()
Bioz is the world’s most comprehensive AI search engine for scientific experimentation. The Bioz search engine offers researchers billions of data-driven product, technique, and protocol recommendations. Bioz taps into the latest advances in Artificial Intelligence (AI) – including Natural Language Processing (NLP), Machine Learning (ML) and Generative AI to mine and structure hundreds of millions of pages of complex and unstructured scientific papers. By harnessing the power of AI, Bioz provides researchers with an unprecedented amount of summarized scientific experimentation knowledge right at their fingertips, ultimately helping to speed up drug discovery and the rate of success in finding cures for diseases. Bioz is used by over 15 Million researchers from over 17,000 different universities and companies in 196 countries.
