Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Sep 20;190(6):1053-65.
doi: 10.1083/jcb.201001028. Epub 2010 Sep 13.

Phosphoinositide 3-kinase δ Regulates Membrane Fission of Golgi Carriers for Selective Cytokine Secretion

Affiliations
Free PMC article

Phosphoinositide 3-kinase δ Regulates Membrane Fission of Golgi Carriers for Selective Cytokine Secretion

Pei Ching Low et al. J Cell Biol. .
Free PMC article

Abstract

Phosphoinositide 3-kinase (PI3K) p110 isoforms are membrane lipid kinases classically involved in signal transduction. Lipopolysaccharide (LPS)-activated macrophages constitutively and abundantly secrete proinflammatory cytokines including tumor necrosis factor-α (TNF). Loss of function of the p110δ isoform of PI3K using inhibitors, RNA-mediated knockdown, or genetic inactivation in mice abolishes TNF trafficking and secretion, trapping TNF in tubular carriers at the trans-Golgi network (TGN). Kinase-active p110δ localizes to the Golgi complex in LPS-activated macrophages, and TNF is loaded into p230-labeled tubules, which cannot undergo fission when p110δ is inactivated. Similar blocks in fission of these tubules and in TNF secretion result from inhibition of the guanosine triphosphatase dynamin 2. These findings demonstrate a new function for p110δ as part of the membrane fission machinery required at the TGN for the selective trafficking and secretion of cytokines in macrophages.

Figures

Figure 1.
Figure 1.
PI3K inhibition and TNF secretion. (A) RAW264.7 macrophages were stimulated with LPS in the presence of vehicle DMSO or either 15 nM wortmannin or 50 µM LY294002, and supernatants were sampled at the times indicated to measure TNF secretion by ELISA. (B) TNF secretion 2 h after LPS activation and treatment with p110-selective inhibitors (YM024 [p110α], TGX221 [p110β], IC87114 [p110δ], or AS252524 [p110γ]) at the concentrations indicated. The following indicate statistical significance at the lowest effective concentrations: ** (IC87114), P = 0.0025; ** (TGX221), P = 0.0073; * (YM024), P = 0.116. (C) Epifluorescence of RAW264.7 cells stimulated with LPS for 1 h in the presence of a TACE inhibitor, TAPI, to retain surface TNF. Cells treated with DMSO alone or with 50 µM LY294002. Immunostaining of surface TNF (red) on unpermeabilized (unperm) cells was followed by staining of intracellular TNF (green) after permeabilization (perm). DAPI (blue) labeling depicts nuclei. Bars, 10 µm. (D) Flow cytometric analysis of intracellular TNF (Alexa Fluor 488) and surface TNF (PE) staining in activated cells treated with DMSO, 15 nM wortmannin, 25 µM LY294002, or 5 µM IC87114 for 2 h. Mean fluorescence intensity of TNF staining in Alexa Fluor 488 or PE channel expressed as mean ratio ± SEM relative to DMSO. Corresponding representative overlay histograms are displayed. (A–D) All results are representative of three independent experiments. Error bars indicate mean ± SD.
Figure 2.
Figure 2.
siRNA knockdown of p110δ impairs TNF secretion. (A) Levels of p110δ protein (∼115 kD; loading control actin, ∼40 kD) in cell lysates measured by Western blotting 48 h after transfection with specific siRNAs or controls. (B) Immunostaining of p110δ (red) and TNF (green) in cells after LPS activation in the presence of TAPI. siRNA3-transfected cells show reduced staining for both proteins. (C) Supernatants from mock-transfected (control) cells and cells transfected with p110δ-targeting siRNAs were used to measure TNF secretion by ELISA. Normalized to total protein, data are expressed as mean ± SEM from triplicate transfections. *, P = 0.0218. (D) Rescue of p110δ siRNA knockdown. Expression of siRNA-resistant WT p110δ (immunostained in red) in p110δ-si–transfected cells. Rescued cells show surface staining for TNF (green). LPS-activated cells were treated with TAPI, fixed, and immunostained. Phalloidin (white) was used to outline cells. Bars, 10 µm.
Figure 3.
Figure 3.
Impaired trafficking and secretion of TNF in macrophages from genetically inactivated p110δ mice. (A and B) Supernatants from LPS-stimulated WT and p110δD910A/D910A BMM cultures were collected every 2 h over a 10-h time course for ELISA measurements of secreted TNF (A) and IL-6 (B). Results indicate mean ± SEM from three littermates of each genotype. ND, not detected. (C) Flow cytometric analysis of intracellular and surface TNF staining in WT and p110δD910A/D910A BMMs treated with LPS and TAPI for 2 h. Mean fluorescence intensity in each TNF channel expressed as mean ratio ± SEM relative to LPS-stimulated WT from three independent experiments. ***, P < 0.001. Representative overlay histograms are shown for LPS-stimulated WT and p110δD910A/D910A BMMs. (D and E) Confocal analysis of TNF staining in activated WT and p110δD910A/D910A BMMs in the presence (D) or absence (E) of TAPI for 2 h. (D) Surface (red) and intracellular (green) TNF staining. Magnified images in D show BMMs costained with phalloidin (blue) to depict surface actin. (E) Intracellular TNF (green) in BMMs costained with the Golgi marker GM130 (red). Differential interference contrast images are included to define cells. Bars, 10 µm.
Figure 4.
Figure 4.
Localization of p110δ on Golgi membranes. (A) Confocal imaging reveals the diffuse cytoplasmic and Golgi immunostaining of p110δ (red) colocalized with GM130 (green; magnified) in RAW264.7 macrophages activated by LPS for 2 h. p110γ (red) gave mostly cytoplasmic immunostaining. (B) Stacked Golgi membranes prepared from activated RAW264.7 macrophages were incubated in vitro with cytosol and GTP-γS to generate budded carriers. The remnant Golgi membranes, budded carriers, and cytosol were collected by ultracentrifugations. Samples of these fractions were run on gels (25% of cytosol supernatants; whole membrane pellets) and immunoblotted for p110δ and γ-adaptin as a peripheral protein marker of TGN-derived budded carriers and TNF as cargo. (C) Coverslip-adherent RAW264.7 macrophages were stimulated with LPS over an acute time course. Immunostaining for p110δ (red) and GM130 (green) was performed at times 0, 30 min, and 2 h on ripped-open cells that exposed intact Golgi membranes. At least 20 cells were analyzed per three independent experiments. Arrows highlight colocalization of p110δ with GM130. DAPI (blue) labeling depicts nuclei. Bars, 10 µm.
Figure 5.
Figure 5.
Site of action for PI3K in TNF secretory pathway. (A–C) LPS-activated RAW264.7 macrophages were incubated with DMSO or 50 µM LY294002 for 2 h, fixed, and immunostained for TNF (red) and, in green, either TfnR (A) or GM130 (B and C). Arrows in A point to multiple TNF-labeled tubules extending from the Golgi complex, and the arrow in B indicates one of these tubules labeled for TNF but not GM130. (C) 3D tomographic reconstructions of confocal z sections (0.48 µm) show multiple discrete TNF-positive tubules (arrows) after treatment with LY294002. DAPI (blue) labeling depicts nuclei. (D) TNF tubules were counted in cells from five fields (40–50 cells/field). Error bars indicate mean ± SEM. **, P = 0.0077. (E) RAW264.7 macrophages were transiently transfected with YFP-p230GRIP (green) and TNF-mCherry (red) for 2 h and treated with 50 µM LY294002 for 1 h before fixation for confocal microscopy. Arrows indicate two tubules (∼3 and 13 µm in length). (A–C and E) Regions of colocalization or interest are indicated by boxes and are magnified in insets. Bars: (A, B, and E) 20 µm; (C) 10 µm.
Figure 6.
Figure 6.
p110δ regulates fission of TGN-derived, TNF-containing tubular carriers. (A–C) RAW264.7 macrophages coexpressing YFP-p230GRIP (green) and TNF-mCherry (red) were treated with LPS only (A; control), LPS and 15 nM wortmannin (B), or LPS and 5 µM IC87114 (C) for at least 30 min before live cell imaging. (A) Selected stills from Video 2 depict tubular carriers labeled for YFP-p230GRIP emerging from the TGN and rapidly undergoing fission (white and green arrows) to release boluses of TNF-mCherry as cargo (dotted circles in grayscale insets). (B and C) Stills from Videos 3 and 4, respectively, also show YFP-p230GRIP tubules loaded with TNF-mCherry that were formed but cannot undergo fission. (D) Counting of fission events in TNF-mCherry–loaded YFP-p230GRIP tubules in live cells treated with LPS only (control), LPS and wortmannin (wortm), or LPS and IC87114 (IC87114). 30–40 cells were analyzed over eight transfections for each treatment. Error bars indicate mean ± SEM. **, P = 0.0022; ***, P < 0.0001. Regions of colocalization or interest are indicated by boxes and are magnified in insets. Bars, 20 µm.
Figure 7.
Figure 7.
Kinase activity of p110δ is required for TGN recruitment of Dyn2. (A and B) Unstimulated RAW264.7 macrophages (unstim) and macrophages activated by LPS for 2 h in the absence or presence of 5 µM IC87114 were fixed for immunostaining for PtdIns(3,4,5)P3 (red) and TNF (green). DAPI (blue) labeling depicts nuclei. Appearance of PtdIns(3,4,5)P3 upon activation at perinuclear Golgi membranes colocalizing with TNF. This PtdIns(3,4,5)P3 detection was selectively lost in the presence of IC87114. (C) RAW264.7 cells transiently transfected with YFP-p230GRIP (green) and untagged WT Dyn2 before activation by LPS for 2 h in the presence of DMSO (control) or 5 µM IC87114 were fixed for Dyn2 immunostaining (red). (B and C) Arrows indicate examples of colocalization of Dyn2 with p230 on TGN membranes that disappeared upon IC87114 treatment. (D) Activated cells coexpressing YFP-p230GRIP and Dyn2 treated with DMSO (control) or IC87114 as described in C were analyzed by image analysis to quantify relative distribution intensities of Dyn2 (TGN vs. cytoplasm). Data are represented as mean ratio ± SEM from 30 cells over three transfections. ***, P < 0.0001. Insets show magnified views of boxed regions. Bars, 10 µm.

Similar articles

See all similar articles

Cited by 22 articles

See all "Cited by" articles

References

    1. Ali K., Bilancio A., Thomas M., Pearce W., Gilfillan A.M., Tkaczyk C., Kuehn N., Gray A., Giddings J., Peskett E., et al. 2004. Essential role for the p110delta phosphoinositide 3-kinase in the allergic response. Nature. 431:1007–1011 10.1038/nature02991 - DOI - PubMed
    1. Bard F., Malhotra V. 2006. The formation of TGN-to-plasma-membrane transport carriers. Annu. Rev. Cell Dev. Biol. 22:439–455 10.1146/annurev.cellbio.21.012704.133126 - DOI - PubMed
    1. Beutler B.A. 1999. The role of tumor necrosis factor in health and disease. J. Rheumatol. Suppl. 57:16–21 - PubMed
    1. Braun V., Deschamps C., Raposo G., Benaroch P., Benmerah A., Chavrier P., Niedergang F. 2007. AP-1 and ARF1 control endosomal dynamics at sites of FcR mediated phagocytosis. Mol. Biol. Cell. 18:4921–4931 10.1091/mbc.E07-04-0392 - DOI - PMC - PubMed
    1. Cao H., Thompson H.M., Krueger E.W., McNiven M.A. 2000. Disruption of Golgi structure and function in mammalian cells expressing a mutant dynamin. J. Cell Sci. 113:1993–2002 - PubMed

Publication types

MeSH terms

Feedback