Sorting soluble tumor necrosis factor (TNF ... - Wiley Online Library

Sorting soluble tumor necrosis factor (TNF) receptor for storage and regulated secretion in hematopoietic cells Ying Gao,* Markus Hansson,* Jero Calafat,† Hans Tapper,‡ and Inge Olsson*,1 Departments of *Hematology, C14, and ‡Cell and Molecular Biology, B14, BMC, Lund, Sweden; and †The Netherlands Cancer Institute, Amsterdam

Abstract: Hematopoietic cells contain secretory lysosomes that degranulate at sites of inflammation. We envisage that secretory granules can act as vehicles for targeting inflammatory sites, including malignancies, and thereafter, locally release therapeutically active agents to these sites. Exogenous proteins, such as the soluble tumor necrosis factor receptor 1 (sTNFR1), have been shown previously to be targeted to secretory lysosomes [1]. In this work, we asked whether exogenous, secretory lysosome-targeted proteins were subject to regulated secretion. sTNFR1–transmembrane (tm)– cytosol-sorting signal (Y) and sTNFR1–tm–Y– enhanced green fluorescent protein (egfp) were expressed in rat basophilic leukemia cell clones having different secretory capacities. sTNFR1–tm–Y was targeted directly from the Golgi to secretory lysosomes, followed by generation of membranefree sTNFR1, whose secretion could be triggered by a Ca2ⴙ ionophore or immunoglobulin E receptor activation. In contrast, sTNFR1–tm–Y– egfp was targeted to the plasma membrane and then subjected to endocytosis and presumably, secretory lysosome targeting, as judged by results from antibody ligation and cell-surface biotinylation. Activation of protein kinase C with phorbol ester promoted ectodomain shedding at the cell surface, resulting in sTNFR1 release from sTNFR1–tm–Y– egfp. These results support a concept for using the storage organelles of hematopoietic cells as vehicles for targeting sites of inflammation with therapeutically active agents. J. Leukoc. Biol. 76: 876 – 885; 2004. Key Words: inflammation 䡠 secretory lysosome 䡠 endosome pathway

INTRODUCTION Hematopoietic cells, such as neutrophils, basophils, natural killer (NK) cells, and cytotoxic T lymphocytes (CTLs), migrate to an inflammatory focus. There, they play a critical role in host defense by releasing an array of bioactive granule agents that have been manufactured and stored during cell maturation [2, 3]. However, the granule-targeting mechanisms are not unique for endogenous granule proteins, as targeting and storage can 876

Journal of Leukocyte Biology Volume 76, October 2004

also be achieved for exogenous proteins [4, 5]. We therefore envisage that secretory storage granules of hematopoietic cells can act as vehicles for targeting sites of inflammation, including malignancies, with therapeutically active agents that would function after local release. In essence, the idea was to deliver a heterogeneous cargo at the inflamed site by using a ubiquitous transport pathway in hematopoietic cells. Hence, we have expressed a soluble tumor necrosis factor receptor 1 form (sTNFR1) in hematopoietic cells with the aim of targeting secretory lysosomes for storage. This was accomplished by using a transmembrane (tm) form of the protein to facilitate endoplasmic reticulum (ER) export and by incorporating a cytosol-sorting signal with a tyrosine motif (Y) for secretory lysosomes to overcome constitutive secretion and achieve targeting of sTNFR1–tm–Y [1, 6]. In the present work, we have gone beyond granule targeting and studied the regulated secretion of targeted proteins expressed in the rat basophilic leukemia (RBL)/mast cell line [7]. Cytolytic proteins are packed together with hydrolases in secretory lysosomes (lytic granules) [8]. These organelles contain dense cores of tightly packed proteins and internal membrane-bound vesicles and are able to combine storage, regulated secretion, and lysosome function. tm protein targeting to the secretory lysosome requires a sorting signal. Granule matrix protein-targeting mechanisms are largely unknown [3] with the exception of targeting lysosome hydrolases and granzyme by the mannose-6-phosphate receptor (MPR) system [9], which navigates cargo from the trans-Golgi network (TGN) along the endosome pathway to secretory lysosomes, after which, empty MPR is recycled. In contrast, a nonendosome pathway is a major route for regulated secretory granule formation in endocrine, neuroendocrine, and exocrine cells. In the latter cells, proteins are packed in dense secretory vesicles within the TGN, and mis-sorted proteins are removed before granule maturation [10 –12]. It is not known whether hematopoietic cells have a direct, nonendosome, secretory lysosome-targeting pathway in addition to an endosome pathway. Separate pathways could target separate secretory lysosome subcompartments. Hematopoietic cells must strictly regulate secretory lysosome delivery of soluble and membrane proteins at sites of

1

Correspondence: Department of Hematology, C14, BMC, SE-221 84 Lund, Sweden. E-mail: [email protected] Received November 26, 2003; revised May 19, 2004; accepted June 1, 2004; doi: 10.1189/jlb.1103593.

http://www.jleukbio.org

inflammation [13]. Cell-surface receptor activation induces 2⫹ Ca mobilization within the cell and triggers degranulation. During degranulation, fusion of the limiting membrane with the plasma membrane releases the matrix content of the secretory lysosomes. Furthermore, internal vesicles of the secretory lysosomes, designated exosomes, can be released intact [14]. Specific protein components such as synaptotagmin are required for the regulation of the exocytosis pathway of hematopoietic cells. These components may act as Ca2⫹ sensors for secretory lysosomes [15]. The membrane fusion processes are mediated by vesicle and target soluble N-ethylmaleimide-sensitive factor attachment protein receptors [16, 17] as well as by small isoforms of GTPases of the Rab protein family [18]. A common, critical mechanism for regulated secretion from secretory lysosomes is suggested by observations in the ChediakHigashi syndrome. In this syndrome, the immune function of cells with secretory lysosomes is impaired, and the nonsecretory function of conventional lysosomes is normal [8, 13]. We have now evaluated further a concept of using hematopoietic granules as delivery vehicles of therapeutic agents. We posed the question as to whether exogenous, secretory lysosome-targeted proteins could be secreted. To address this, RBL cells stably transfected with sTNFR1–tm–Y or sTNFR1–tm– Y– enhanced green fluorescent protein (egfp) were investigated. Our results showed direct targeting of secretory lysosomes with sTNFR1–tm–Y but suggested indirect targeting of sTNFR1–tm–Y– egfp after plasma membrane translocation and endocytosis. Regulated secretion of secretory lysosome-generated sTNFR1 could be triggered by a Ca2⫹ signal.

MATERIALS AND METHODS

(SDS; v/v), 5% Triton X-100 (v/v), 5% sodium deoxycholate (w/v)], and lysis buffer for immunoblotting [electrophoresis sample buffer; 86 mM Tris, pH 6.8, 11% (v/v) glycerol, 2.3% (w/v) SDS, 1.2% (v/v) ␤-ME, 0.005% (v/v) bromophenol blue] were used. Protease inhibitors were added to all buffers before use.

Construction of expression vectors The cDNA fusion constructs sTNFR1–tm–Y and sTNFR1–tm–Y– egfp were used. tm corresponds to the tm domain of human TNFR1 with flanking sequences (VKGTEDSGTTVLLPLVIFFGLCLLSLLFIGLMYRYORWKSKLYSIV), and Y corresponds to the cytosol tyrosine-sorting signal (SIRSGYEVM) for secretory lysosomes of CD63 [8]. All polymerase chain reactions (PCRs) were performed by 20-cycle reactions in a Perkin-Elmer 480 thermal cycler using Pfu polymerase (Stratagene, La Jolla, CA), per the manufacturer’s instructions. In the design of the primers, the Kozak consensus leader sequence for maximal translational efficiency [20] was introduced 5⬘ to the ATG initiation codon. The constructs were cloned into the pcDNA3 vector. The generation of sTNFR1–tm–Y cDNA was described previously [1]. To generate sTNFR1–tm–Y– egfp cDNA, two extension PCRs were performed. The first PCR used the sTNFR1–tm–Y cDNA as a template with the following primers: upstream primer, 5⬘-TTCGGAGGATCCGCCACCATGGGCCTCTCCACCGTG (A1); downstream primer, 3⬘-TTCGGAGAATTCTCAATCGATCATCACCTCGTAGCCACTTCT (A6 short), leading to the creation of sTNFR1– tm–Y (A6 short) with BamHI, ClaI, and EcoRI restriction-enzyme sites. The second PCR used the pEGFP-N1 as a template with the following primers: upstream primer, 5⬘-TTCGGAATCGATGCCACCGTGAGCAAGGGCGAGGAGCTG (A14); downstream primer, 3⬘- TTCGGAGAATTCTCACTTGTACAGCTCGTCCATGCC (A15). The product was cut by ClaI and EcoRI and ligated into the sTNFR1–tm–Y (A6 short) cDNA, leading to the creation of sTNFR1–tm–Y– egfp with BamHI and EcoRI restriction enzyme sites.

Transfections RBL-1 cells [7] were stably transfected by using the Bio-Rad Gene Pulser™ (Bio-Rad, Hercules, CA) with electrical settings of 960 ␮F and 260 –300 V as described previously [21, 22]. RBL-2H3 cells were transfected by using SuperFect威. Recombinant clones were selected with 1 mg/ml geneticin and were screened for the expression of the transfected protein by biosynthetic radiolabeling and immunoprecipitation.

Biosynthetic radiolabeling

Materials Constructs were created for eukaryotic expression using the vector pcDNA3 (Invitrogen, Groningen, The Netherlands) for transfection of RBL cells. SuperFect威 transfection reagent was from Qiagen (West Sussex, UK). [35S]-Methionine/[35S]-cysteine (cell-radiolabeling grade) was from ICN Pharmaceuticals (Costa Mesa, CA). Heat-inactivated fetal bovine serum (FBS) was from BioWhittaker (Verviers, Belgium). L-glutamine was from Gibco-BRL (Life Technologies, Grand Island, NY). Percoll and protein A-Sepharose CL-4B were from Pharmacia (Uppsala, Sweden). 2-Mercaptoethanol (2-ME) was from Sigma-Aldrich Co. (St. Louis, MO). Geneticin and Complete™ (proteaseinhibitor cocktail tablets) were from Boehringer Mannheim (Mannheim, Germany). Sulfo N-hydroxysuccinimide bound via a disulfide bond to biotin (NHS-SS-biotin) and streptavidin-agarose beads were from Pierce Chemical Co. (Rockford, IL). Novex Precast gels (10 –20% Tris-glycine gel) were from Novex (San Diego, CA). Polyclonal antiserum against sTNFR1 was obtained by immunization of rabbits and was used for immunoprecipitation [19]. A monoclonal antibody (mAb), anti-sTNFR1 (SC8436; Santa Cruz Biotechnology, Santa Cruz, CA) was used in detection of sTNFR1 by Western blotting. Goat serum was from Sigma-Aldrich Co. Rabbit anti-lysosome glycoprotein 120 (lgp120) antiserum was from Dr. Ira Mellman (Yale University School of Medicine, New Haven, CT). Calcium-magnesium ionophore was from Calbiochem (CN Biosciences, Darmstadt, Germany). Monoclonal antibody to rat mast cell protease-II (RMCP-II; MS-RM4) was from Moredun Scientific (Midlothian, Scotland).

Biosynthetic radiolabeling of newly synthesized protein was carried out using [35S]methionine/[35S]cysteine, as described previously [23].

Subcellular fractionation A cell homogenate was centrifuged in a continuous Percoll gradient, and nine fractions were collected as described previously [21]. All the cytosol was contained in fraction 9. The distribution of secretory lysosomes and Golgi elements was determined by assaying for ␤-hexosaminidase and galactosyl transferase, respectively [24, 25]. The peak activities of ␤-hexosaminidase and galactosyl transferase in Percoll fractions of RBL cells were localized in fractions 2 and 6, respectively [21]. Furthermore, the distribution of the secretory lysosome marker RMCP-II was determined by immunoprecipitation of radiolabeled cells.

Immunoprecipitation Cells (2⫻106/ml) were disrupted in lysis buffer for cells, frozen, and thawed. Percoll-containing subcellular fractions were diluted with 1 vol H2O and one-half of a volume of fivefold-concentrated RIPA. Samples were cleared by centrifugation at 32,000 g for 60 min. Antiserum, protein A-Sepharose, and protein G-Agarose were added, the mixture was rotated overnight, and the pellet was analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) followed by gel exposure to X-ray film at – 80°C. Densitometry was performed in a Molecular Imager威 FX (Bio-Rad).

Buffer systems Lysis buffer for cells [1 M NaCl, 50 mM Tris-HCl, pH 8.0, 0.5% Triton X-100 (v/v)], radioimmune precipitation buffer [radio immunoprecipitation assay (RIPA); 0.75 M NaCl, 0.15 M HEPES, pH 7.3, 0.5% sodium dodecyl sulfate

Western blotting Cell extracts were frozen and sonicated in lysis buffer for immunoblotting, boiled for 5 min, and cleared by centrifugation. A lysate aliquot was separated

Gao et al. Sorting for regulated secretion in hematopoietic cells

877

by SDS-PAGE and electroblotted onto Hybond-P nitrocellulose membranes and exposed to Hyperfilm™ ECL™ for 10 –30 s.

Immunoelectron microscopy

the cells in Eagle’s minimal essential medium with 20% FBS at 37°C and was terminated by placing the cultures on ice. Remaining biotinylated surface proteins were debiotinylated by incubation with 50 mM 2-ME sulfonic acid in 50 mM Tris, pH 8.7, 100 mM NaCl, 2.5 mM CaCl2 for 2 ⫻ 20 min at 4°C. After additional washing twice with 10 mM Tris, pH 7.0, 150 mM NaCl, the cells were lysed in 50 mM Tris, pH 7.0, 1% Triton X-100 with protease inhibitors. The internalized, biotinylated proteins were precipitated by 50 ␮l streptavidinagarose beads overnight, separated by SDS-PAGE, and analyzed by Western blotting.

RBL cells were fixed for 24 h in 4% paraformaldehyde (PFA) in 0.1 M PHEM buffer (pH 6.9) containing 240 mM piperazine diethane sulfonic acid (PIPES), 100 mM HEPES, 8 mM MgCl2, and 40 mM ethylene glycol tetra-acetic acid (EGTA) and then processed for ultrathin cryosectioning as described before [26]. Cryosections (45 nm) were cut at –125°C using diamond knives (Drukker Cuijk, The Netherlands) in an ultracryomicrotome (Leica Aktiengesellschaft, Vienna, Austria) and transferred with a mixture of sucrose and methylcellulose onto formvar-coated copper grids. The grids were placed on 35-mm petri dishes containing 2% gelatin. Double-immunolabeling was performed with the procedure described by Slot et al. [27] with 10- and 15-nm protein A-conjugated colloidal gold probes (EM Lab, Utrecht University, The Netherlands). After immunolabeling, the cryosections were embedded in a mixture of methylcellulose and uranyl acetate and examined with a Philips CM 10 electron microscope (Eindhoven, The Netherlands). For the controls, the primary antibody was replaced by a nonrelevant rabbit or mouse antiserum.

RESULTS

Immunofluorescence microscopy

Processing and targeting sTNFR1–tm–Y

For colocalization experiments, fixed RBL-2H3 cells (cultured overnight on glass coverslips) were permeabilized in 1 ml cytoskeletal buffer containing 100 mM KOH, 2 mM MgCl2, 5 mM EGTA, 0.05% (v/v) Triton X-100, and 100 mM piperazine-bis 2-ethane sulfonic acid (pH 6.8) for 10 min on ice and thereafter, were incubated in blocking solution [phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin (BSA; w/v), 0.2% Tween 20 (v/v), and 5% (v/v) goat serum] for 30 min at room temperature. Next, cells were incubated overnight at 8°C with the primary antibody (dilution 1:800 –1:1000) in blocking solution. Following washing, cells were incubated with secondary antibody (goat anti-rabbit Alexa FluorR 488 F(ab)2 fragment, dilution 1:500; or goat anti-mouse Alexa FluorR 594 F(ab)2 fragment, dilution 1:500; Molecular Probes, Eugene, OR) for 1 h in blocking solution. After washing twice, the samples were then overlaid with ProLong Antifade reagent (Molecular Probes) and were mounted. Images were recorded on a Nicon Eclipse TE300 inverted fluorescence microscope equipped with a Hamamatsu C4742-95 cooled charged-coupled device camera using a Plan Apochromat 100⫻ objective and a high numerical aperture oil-condenser. Plasma membrane-localized sTNFR1–tm–Y– egfp was labeled with red fluorescence by incubating transfected cells with a mouse mAb against the sTNFR1 (1:1000, 5 min, room temperature) and subsequently, with an Alexa FluorR 594 F(ab)2 fragment-labeled anti-mouse secondary antibody (1:200, 5 min). After several rinses, experiments were performed. Thereafter, the cells were immediately observed in the microscope or fixed using 1 ml 2% (v/v) PFA solution (Becton Dickinson, Franklin Lakes, NJ) for 15 min on ice and 45 min at room temperature. Fixed cells in suspension (RBL-1) were adhered to poly-L-lysine-coated coverslips. The samples were then mounted, and images were recorded as described above.

Results from Western blotting of the adherent RBL-2H3 cells transfected with sTNFR1–tm–Y showed full-length 37 kDa protein and a 29-kDa processed form consisting of sTNFR1 (data not shown). Results from immunofluorescence microscopy and immunoelectron microscopy demonstrated targeting of sTNFR1–tm–Y to secretory lysosomes (data not shown) similar to what has been shown before for RBL-1 cells [1]. Thus, the sTNFR1 signal colocalized with the rat lgp120 and RMCP-II, a major constituent of RBL cell granules. In conclusion, the data demonstrate targeting of sTNFR1–tm–Y to the secretory lysosomes of the adherent RBL-2H3 cells.

Secretion RBL-1 cells (20⫻106) or RBL-2H3 cells (10⫻106) were biosynthetically radiolabeled for 1 h followed by radiolabel chase for 2 h without or with 1 ␮M Ca2⫹ ionophore or 10 ng/mL phorbol myristate acetate (PMA), added 30 and 60 min before end of chase, respectively. Secretion was also induced by cross-linking the immunoglobulin (Ig)E receptors. For this, cells were biosynthetically radiolabeled for 45 min followed by a 90-min radiolabel chase in the presence of mouse monoclonal anti-dinitrophenyl (DNP) IgE (1 ␮g/106 cells/ ml). After washing, the cells were incubated 30 min with DNP-albumin conjugate (10 ␮g/106 cells/ml) at 37°C to achieve IgE receptor cross-linking. Cell lysates and supernatants were immunoprecipitated and analyzed by SDS-PAGE.

Biotinylation After washing with cold PBS, surface proteins of 3 ⫻ 106 RBL-2H3 cells expressing sTNFR1–tm–Y– egfp were labeled with 0.5 mg/ml NHS-SS-biotin in PBS with Ca2⫹ and Mg2⫹ for 60 min at 4°C and were then washed once with ice-cold 10 mM Tris, pH 7.0, 150 mM NaCl, and 50 mM glycin and twice with the same buffer without glycin. Internalization was initiated by incubation of

878


To show the regulated secretion of proteins targeted to secretory lysosomes, we used transfectants of RBL-1 cells, which could grow in suspension, and RBL-2H3 cells, which could grow in adherent layers. These sublines differed in their secretory capacity.

Secretion of sTNFR1 from sTNFR1–tm–Yexpressing cells Physiological Ca2⫹ stimulation can promote regulated secretion from secretory lysosomes in RBL cells [28, 29]. We tried to induce secretion with Ca2⫹ ionophore, IgE receptor activation, or PMA in RBL-1 (Fig. 1A) and RBL-2H3 cells (Fig. 1B) expressing sTNFR1–tm–Y. Using the endogenous secretory lysosome constituent RMCP-II as an indicator of secretory lysosome release, RBL-1 cells seemed to lack secretory capacity (Fig. 1A), although constitutive RMCP-II secretion was observed. This was consistent with the observed lack of sTNFR1 secretion in these cells upon incubation with Ca2⫹ ionophore (Fig. 1A) or IgE receptor cross-linking (data not shown). Furthermore, the endogenous RMCP-II as well as sTNFR1, which accumulated in the densest fractions, corresponding to the secretory lysosomes of sTNFR1–tm–Y-transfected cells, were unaffected by PMA (Fig. 1A). PMA is unable to stimulate regulated secretion from secretory lysosomes unless Ca2⫹ influx is stimulated with a Ca2⫹ ionophore [30]. In contrast, RBL-2H3 cells have a secretory capability. Thus, upon incubation with a Ca2⫹ ionophore, the sTNFR1 that accumulated in the densest fractions and the endogenous RMCP-II were secreted (Fig. 1B). In addition, IgE receptor activation, which is a more physiological stimulus for mast cells, induced regulated secretion, although to a lower degree than Ca2⫹ ionophore (Fig. 1B). Our results therefore show that secretory lysosome-targeted sTNFR1 was subject to regulated secretion in RBL-2H3-cells but not in RBL-1 cells. http://www.jleukbio.org

Fig. 1. Ca2⫹-induced secretion of sTNFR1 in RBL-2H3 but not RBL-1 cells expressing sTNFR1–tm–Y. RBL-1 cells (A) and RBL-2H3 cells (B) transfected with sTNFR1–tm–Y were biosynthetically radiolabeled for 1 h followed by radiolabel chase for 2 h without or with 1 ␮M Ca2⫹ ionophore or 10 ng/mL PMA added 30 or 60 min before the end of chase, respectively. For induction of secretion by IgE receptor activation, cells were biosynthetically radiolabeled for 45 min followed by 90 min radiolabel chase in the presence of mouse monoclonal anti-DNP IgE. After washing, the cells were incubated 30 min with DNP-albumin conjugate at 37°C for IgE receptor cross-linking. Cell lysates and medium were immunoprecipitated for detection of sTNFR1 and RMCP-II as described in Materials and Methods. The position of sTNFR1–tm–Y, sTNFR1, and RMCP-II is indicated. The flurogram of the IgE receptor activation experiment is overexposed, as secretion was lower as compared with the Ca2⫹-ionophore experiment.

Processing and targeting sTNFR1–tm–Y– egfp Results from pulse-chase radiolabeling of RBL-1 cells transfected with sTNFR1–tm–Y– egfp showed expression of the full-length, 61-kDa protein as well as the processed forms of the protein corresponding to the 37-kDa sTNFR1–tm–Y and the 29-kDa sTNFR1 (data not shown). Immunoprecipitation with anti-GFP detected only the full-length sTNFR1–tm–Y– egfp. This suggested that the egfp was eliminated by proteolysis during processing so that the processed forms lacked egfp and were therefore undetectable with anti-egfp (data not shown). Secretory lysosome targeting was investigated by pulsechase radiolabeling combined with subcellular fractionation. After radiolabeling, sTNFR1–tm–Y– egfp and sTNFR1–tm–Y were visible in the light-density fractions corresponding to the ER, Golgi, and plasma membrane elements (Fig. 2). During radiolabel chase, most of the sTNFR1–tm–Y– egfp disappeared from the light-density fractions, and processed sTNFR1–tm–Y remained in these fractions (Fig. 2). Concomitantly, proteolytically generated sTNFR1 was detected in the densest fractions, suggesting that secretory lysosome targeting occurred prior to release of sTNFR1. The density gradient distribution of secretory lysosomes was confirmed by pulse-chase radiolabeling of RMCP-II. After pulse, RMCP-II was present in the light fractions, and after subsequent radiolabel chase, a proportion of RMCP-II shifted to the densest fractions, indicating the location of the secretory lysosomes. The generated sTNFR1 showed a similar distribution to RMCP-II consistent with secretory lysosome targeting (Fig. 2). Cells expressing sTNFR1–tm–Y– egfp showed anti-sTNFR1 labeling of the Golgi and TGN and in granules, as judged by immunoelectron microscopy (Fig. 3, A, C, and D). A distinct labeling of the plasma membrane was also observed (Fig. 3B) in contrast to cells expressing sTNFR1–tm–Y, which lacked

Fig. 2. Subcellular distribution of radiolabeled sTNFR1–tm–Y– egfp and processed forms as well as endogenous RMCP-II in RBL-1 cells. Cells expressing sTNFR1–tm–Y– egfp were biosynthetically radiolabeled for 1 h (Pulse), followed by chase of the label for 2 h. At the pulse and chase intervals, 100 ⫻ 106 cells were removed and homogenized, and the postnuclear supernatant was subcellularly fractionated by centrifugation in Percoll. The fractions were immunoprecipitated and analyzed as described in Materials and Methods. The position of sTNFR1–tm–Y– egfp and the processed sTNFR1–tm–Y and sTNFR1 forms are indicated with arrows. The position of endogenous RMCP-II is also indicated. Molecular weight markers are shown to the left in kDa.


879

min at 37°C prior to incubation with anti-sTNFR1 led to loss of antibody reactivity with the cell surface (Fig. 5E, inset), indicating PMA-induced shedding consistent with the lack of cell-surface egfp signal (Fig. 5D). However, prelabeling cell surface-bound sTNFR1–tm–Y– egfp with anti-sTNFR1 (Fig. 5B) led to rapid internalization of the signal (Fig. 5E), not allowing enough time for shedding of prelabeled sTNFR1 to take place. Our conclusions were further supported by results from immunoelectron microscopy, which showed complete loss of the plasma membrane sTNFR1 signal after PMA incubation (data not shown).

Secretion of sTNFR1 from sTNFR1–tm–Y– egfpexpressing cells

Fig. 3. Colocalization of sTNFR1 with a matrix protein (RMCP) and a membrane-associated protein (lgp120) in secretory lysosomes of RBL-1 cells expressing sTNFR1–tm–Y– egfp. Ultrathin cryosections were labeled with rabbit anti-TNFR1 and protein A gold (10 nm; A, B), double-labeled with anti-TNFR1 (10 nm) and mouse anti-RMCP followed by protein A gold (15 nm; C), or double-labeled with anti-TNFR1 (10 nm) and rabbit anti-lgp120 (15 nm; D). (A) Area showing that the Golgi (G) and the TGN are labeled for sTNFR1. (B) Area showing that the plasma membrane from several cells is also labeled. (C) Area near the Golgi (G) with two granules (g) double-labeled for sTNFR1 and RMCP, and (D) area with two granules (g) double-labeled for lgp120 and sTNFR1. n, Nucleus. Original bar ⫽ 200 nm.

plasma-membrane labeling. Moreover, the results demonstrated colocalization of the sTNFR1 signal with RMCP-II and lgp120 in the granules/multivesicular bodies (Fig. 3, C and D), arguing for targeting of sTNFR1–tm–Y– egfp and/or its processed form(s) to secretory lysosomes. Results from immunofluorescence microscopy also showed egfp labeling at the cell surface (Fig. 4A). Furthermore, the egfp staining appeared mostly in structures likely to represent the ER, Golgi, and endosomal apparatus. The staining pattern for egfp (Fig. 4A) and the secretory lysosome marker lgp120 (Fig. 4B) showed only partial colocalization (Fig. 4D). This is consistent with egfp being eliminated by proteolysis only after appearing at the cell surface, during subsequent traffick to granules. What was detectable in granules by immunoelectron microscopy (Fig. 3) is therefore likely to be mainly processed in sTNFR1–tm–Y and/or sTNFR1 forms. As observed by electron microscopy, the presence of a sTNFR1 signal at the surface of fixed cells was also verified by immunofluorescent labeling (not shown). Immunofluorescence microscopy also showed cell-surface localization of sTNFR1–tm–Y– egfp on viable cells (Fig. 5A). Furthermore, PMA incubation resulted in loss of sTNFR1–tm– Y– egfp from the cell surface, judging by loss of egfp signal (Fig. 5D), consistent with protein kinase C (PKC)-induced sTNFR1 shedding. PMA incubation also resulted in some loss of intracellular egfp signal (Fig. 5D). PMA incubation for 15 880


Secretion of sTNFR1 was triggered by Ca2⫹ ionophore in RBL-1 transfected with sTNFR1–tm–Y– egfp (Fig. 6), although no secretion was triggered by IgE receptor activation (data not shown). In contrast, RBL-1 cells expressing sTNFR1– tm–Y lacked Ca2⫹ ionophore-induced secretion of sTNFR1 (Fig. 1A). In addition, PMA induced a prominent release of sTNFR1 from sTNFR1–tm–Y– egfp transfectants, presumably by ectodomain shedding at the cell surface (Fig. 6). The observed plasma membrane localization of sTNFR1– tm–Y– egfp in cells expressing this protein (Fig. 5) and the sTNFR1 release from these cells after PMA incubation (Fig. 5) are compatible with PKC-induced ectodomain shedding at the cell surface [19]. These results were investigated further by

Fig. 4. Partial colocalization of egfp with a membrane-associated protein (lgp120) of secretory lysosomes in RBL-1 cells expressing sTNFR1–tm–Y– egfp. Direct localization of egfp and indirect immunolocalization of lgp120 are shown. The cells were fixed, permeabilized, and stained, as described in Materials and Methods. Briefly, the cells were incubated with a rabbit antilgp120 antibody. Thereafter, the cells were stained with Alexa Fluor™-labeled anti-rabbit secondary antibody. The localization of egfp and lgp120 is shown in green (A) and red (B), respectively, whereas an overlay of the green and red staining pattern is seen (D), demonstrating partial colocalization (yellow). The corresponding Nomarski image is shown (C). Original size bar ⫽ 10 ␮m.


Fig. 5. PMA-induced shedding of sTNFR1 in RBL-1 cells expressing sTNFR1–tm–Y– egfp. Cells transfected with sTNFR1–tm–Y– egfp were incubated 5 min at room temperature with a mouse mAb against the sTNFR1 (1: 1000). Thereafter, the cells were stained with an Alexa Fluor™ 594-labeled anti-mouse secondary antibody (1: 200, 5 min). Following several rinses to remove unbound antibody, cells were incubated with PMA at 37°C and then fixed using 2% PFA and attached to poly-L-lysinecoated coverslips. The cells shown (A–C) were incubated with 50 nM PMA for less than 1 min, whereas the cells shown (D–F) were incubated 30 min with PMA. The insets show cells incubated with PMA for 15 min at 37°C prior to analysis. The egfp-derived green fluorescence of sTNFR1–tm–Y– egfp is shown (A and D); the red staining patterns of bound sTNFR1 antibody are shown (B and E). The corresponding Nomarski images are shown (C and F). PMA induced a loss of egfp signal from the cell surface and cell interior (D). Prelabeling of cell surface with anti-sTNFR1 (B) was combined with a rapid internalization of the signal (E). PMA incubation prior to antibody incubation resulted in loss of antibody reactivity with the cell surface (E, inset). Original size bar ⫽ 10 ␮m.

determining the effects on secretion at the subcellular level. RBL-1 cells expressing sTNFR1–tm–Y– egfp were radiolabeled for 1 h, and the radiolabel was chased for 2 h. PMA or Ca2⫹ ionophore was added during the second hour of chase, and the cells were fractionated subcellularly on Percoll gradients. The SDS-PAGE protein bands were analyzed by densitometry (Fig. 7). PMA promoted a decrease in sTNFR1–tm– Y– egfp, particularly in the light-density fractions (Fig. 7A). These results are consistent with PKC-induced ectodomain shedding at the cell surface with resulting sTNFR1 secretion and are consistent with the presence of sTNFR1–tm–Y– egfp at the cell surface (Figs. 3 and 5). Furthermore, PMA also promoted a decrease in sTNFR1–tm–Y in the light-density fractions (Fig. 7B), supporting PKC stimulation of plasma membrane targeting of this protein and ectodomain shedding. It is important that PMA had little effect on the sTNFR1, which accumulated in the densest fractions corresponding to secre-

tory lysosomes. This is compatible with PMA being unable to promote granule release unless Ca2⫹ influx is stimulated with a Ca2⫹ ionophore [30]. An ionophore-triggered Ca2⫹ signal had only a slight effect on the level of sTNFR1–tm–Y– egfp in the various subcellular fractions (Fig. 7C). In contrast, the Ca2⫹ signal triggered extensive elimination of the processed sTNFR1–tm–Y and sTNFR1 forms (Fig. 7D). The sTNFR1 loss from the densest fractions, which contain the secretory lysosomes, is consistent with a regulated secretion from the cell. Conversely, the disappearance of sTNFR1–tm–Y from all fractions may result from its plasma membrane targeting and ectodomain shedding. In support of this, only sTNFR1 was secreted under these conditions (Fig. 6). In conclusion, our results from sTNFR1–tm–Y– egfp-transfected cells suggest that a Ca2⫹-induced regulation of sTNFR1 secretion is generated in secretory lysosomes. Furthermore, the results suggest that PKC activation induces sTNFR1 release at the cell surface through ectodomain shedding.

Endosome reuptake of sTNFR1–tm–Y– egfp

Fig. 6. Ca2⫹-induced secretion of sTNFR1 in RBL-1 expressing sTNFR1– tm–Y– egfp. RBL-1 cells transfected with sTNFR1–tm–Y– egfp were radiolabeled biosynthetically for 1 h followed by radiolabel chase for 2 h without or with 1 ␮M Ca2⫹ ionophore or 10 ng/mL PMA added 30 or 60 min before end of chase, respectively. Lysates of cells and medium were immunoprecipitated for detection of sTNFR1 as described in Materials and Methods. The position of sTNFR1–tm–Y– egfp and the processed sTNFR1–tm–Y and sTNFR1 forms are indicated with arrows.

Results from fluorescence microscopy using egfp fluorescence and immunofluorescence with anti-sTNFR1 demonstrated the presence of sTNFR1–tm–Y– egfp at the cell surface in RBL-1 cells (Fig. 8, A and B). In contrast, cells transfected with sTNFR1–tm–Y did not show any sTNFR1 signal at this location (data not shown). Furthermore, antibody labeling of cellsurface sTNFR1–tm–Y– egfp of living cells at room temperature (Fig. 8B) was followed by rapid internalization of the antibody during subsequent chase at 37°C (Fig. 8D). These results implied that internalization of sTNFR1–tm–Y– egfp took place. Rapid internalization was confirmed by a timecourse experiment (Fig. 9). We also used biotinylated cell-surface sTNFR1–tm–Y– egfp to show constitutive internalization. First, all proteins on the cell surface were labeled with NHS-SS-biotin at 4°C. Then the cells were switched to 37°C to allow for internalization, and remaining biotinylated surface proteins were debiotinylated by


881

Fig. 7. Subcellular distribution of radiolabeled sTNFR1–tm–Y– egfp and processed forms after induced secretion. RBL-1 cells transfected with sTNFR1–tm– Y– egfp were radiolabeled biosynthetically for 1 h, followed by chase of the label for 2 h. During chase, the cells were incubated with or without PMA (A and B) or Ca2⫹ ionophore (C and D). For each condition, 100 ⫻ 106 cells were removed and homogenized, and the postnuclear supernatant was fractionated subcellularly by centrifugation in Percoll. The fractions were immunoprecipitated and analyzed as described in Materials and Methods. The SDS-PAGE bands corresponding to sTNFR1–tm–Y– egfp (I; A and C) and the processed forms sTNFR1–tm–Y (II) and sTNFR1 (III; B and D) are shown. The results from densitometry (relative density) of these bands are shown. The sTNFR1–tm–Y and sTNFR1 bands were combined for densitometry. PMA promoted a decrease in sTNFR1–tm–Y– egfp (A) and sTNFR1–tm–Y (B), in particular, of the light-density fractions. Ca2⫹ ionophore gave a slight decrease of sTNFR1–tm–Y– egfp (C) but an extensive elimination of sTNFR1–tm–Y and sTNFR1 (D).

cleavage of the NHS-SS-biotin disulfide bond. Internalized biotinylated proteins were precipitated by streptavidin, and sTNFR1–tm–Y– egfp was detected by Western blotting (Fig. 10). The total surface-biotinylated sTNFR1–tm–Y– egfp was determined in cells held on ice not to allow internalization. Using this method, we found almost quantitative internalization of the sTNFR1–tm–Y– egfp content present on the cell surface (Fig. 10). Taken together, our results indicated that sTNFR1–tm–Y– egfp might be targeted to secretory lysosomes after cell-surface translocation and endosome reuptake.

DISCUSSION Our results support a general concept that hematopoietic granules (the normal storage site of antimicrobial and other agents with a role in innate immunity) might be used as vehicles for 882


the deposition of therapeutically active agents (such as sTNFR1) at sites of inflammation and malignancy during degranulation. We have shown that a tm sTNFR construct, targeted to the secretory lysosome of hematopoietic cells, was subject to proteolytic cleavage and regulated secretion of the released soluble receptor. Hiding the carboxy-terminal sorting signal by incorporation of egfp promoted plasma membrane targeting and subsequent endosome uptake of the construct and finally, secretory lysosome targeting.

Targeting signals When the tyrosine-based, cytosol-sorting signal from CD63 (a normal tm secretory lysosome constituent) was placed at the C-terminus, secretory lysosome targeting of exogenous proteins expressed in RBL cells was successfully achieved. This signal conforms to the tm protein-sorting sequence YXXØ necessary for coat recruitment, membrane invagination, and transport vesicle formation at the TGN and plasma membrane [31]. Y http://www.jleukbio.org

Fig. 10. Internalization of sTNFR1– tm–Y– egfp determined by surface biotinylation and Western blotting. Surface proteins of 3 ⫻ 106 RBL-2H3 cells expressing sTNFR1–tm–Y– egfp were labeled at 4°C by sulfo-NHS-SS-biotin before initiation of constitutive sTNFR1–tm–Y– egfp internalization at 37°C. Total cell-surface, biotinylated sTNFR1–tm–Y– egfp was determined in cells held at 4°C to prevent internalization (A). Background internalization (at 4°C) of sTNFR1–tm–Y– egfp was measured in cells debiotinylated without prior initiation of internalization (B). After allowing internalization for 10 (C) and 30 min (D) at 37°C, debiotinylation was carried out before Western blotting analysis. Molecular weight markers are shown to the left in kDa.

Fig. 8. Internalization of plasma membrane-located sTNFR1–tm–Y– egfp in RBL-1 cells. Unfixed cells transfected with sTNFR1–tm–Y– egfp were incubated for 5 min at room temperature with a mouse mAb against the sTNFR1 (1:1000). Thereafter, the cells were stained with an Alexa Fluor™ 594 -labeled anti-mouse secondary antibody (1:200, 5 min). Following several rinses to remove unbound antibody, images were recorded immediately (A and B) or after a 45-min chase period at 37°C (C and D). The green fluorescence of sTNFR1–tm–Y– egfp is shown (A and C), whereas the images to the right show the red-staining patterns resulting from bound sTNFR1 antibody. If the primary antibody was omitted, no staining was observed. The red fluorescence is initially observed at the cell surface (B) but is transferred intracellularly after chase (D), indicating endosome reuptake. Original size bar ⫽ 10 ␮m.

corresponds to tyrosine and X to any amino acid, and the Ø position can accommodate residues with bulky hydrophobic side-chains. The distance from the tm domain of the sorting motif is of importance in lysosome-targeting and endocytosis [32]. First, a carboxy-terminal YXXØ sequence close to the tm domain serves as a lysosome-targeting signal at the TGN and/or

endosome. Second, a hidden YXXØ signal that was not placed carboxy-terminally might still serve as a signal for endocytosis. In the latter case, the hidden YXXØ motif prevents interactions with the sorting machinery at the TGN, thus leading to plasma membrane targeting but permitting endocytosis. Our results are consistent with this general model for targeting. The sTNFR1–tm–Y was targeted directly to secretory lysosomes by a carboxy-terminal YXXØ sequence placed close to the tm domain. The sTNFR1–tm–Y– egfp, conversely, was targeted through an indirect pathway to the secretory lysosome, presumably because the YXXØ motif was not carboxy-terminal but hidden amino-terminal to egfp. Internalization of sTNFR1–tm–Y– egfp via the plasma membrane agreed with an indirect pathway for secretory lysosome targeting. Endocytosis might be followed by the generation of sTNFR1–tm–Y after proteolytic release of egfp from the endosomes and/or secretory lysosomes. As a result, after egfp removal, Ø of YXXØ would be offered at a carboxy-terminal position and have the ability to signal for secretory lysosome targeting at an endosome-sorting station. Thus, newly synthesized sTNFR1–tm–Y– egfp is, at least in part, rapidly moved to the cell surface after exiting the TGN. sTNFR1–tm–Y– egfp then enters the endocytic system, presumably via activated protein (AP)-2-dependent sorting at the plasma membrane [31], with the subsequent generation of sTNFR1–tm–Y. We assume that a second sorting event in endosomes, possibly

Fig. 9. Time-course of the endosomal reuptake of sTNFR1–tm–Y– egfp in RBL-1 cells. Cells transfected with sTNFR1–tm–Y– egfp were incubated for 5 min at room temperature with a mouse mAb against the sTNFR1 (1:1000). Thereafter, the cells were stained with an Alexa Fluor™ 594 -labeled anti-mouse secondary antibody (1:200, 5 min). Following several rinses to remove unbound antibody, cells were chased for various times at 37°C and then fixed using 2% PFA and attached to poly-L-lysine-coated coverslips. The staining pattern of the bound sTNFR1 antibody after a 1-min (D), 3-min (E), and 10-min (F) chase is shown, together with the corresponding Nomarski images. Original size bar ⫽ 10 ␮m.


883

through the AP-3 complex, could be involved in secretory lysosome targeting. AP-3 has been localized to an endocytic compartment [33]. The reason that sTNFR1–tm–Y– egfp might not interact with the AP-3 complex at the TGN in the first place was thought to be insufficient exposure of the sorting sequence. In conclusion, our results can be explained on the basis of direct targeting of sTNFR1–tm–Y to secretory lysosomes and indirect targeting of sTNFR1–tm–Y– egfp to secretory lysosomes after plasma membrane targeting and endocytosis.

Generation of sTNFR1 for secretion The sTNFR1–tm–Y was processed into a lower molecular weight protein. As the processed protein was located to the densest subcellular fractions, it was likely to consist of sTNFR1 released into the secretory lysosomes as a result of the limited proteolysis of sTNFR1–tm–Y. Furthermore, immunoelectron microscopy verified colocalization of sTNFR1–tm–Y and/or its processed form with endogenous secretory lysosome constituents such as RMCP-II and lysosome-associated membrane protein-1 (LAMP-1). The sTNFR1 signal was, however, located mostly in the interior of the secretory lysosome in contrast to LAMP-1, which was on the limiting membrane. It can therefore be concluded that sTNFR1–tm–Y was first targeted to the limiting membrane followed by the transfer to internal vesicles in a similar manner to CD63 [34]. Transfer to internal vesicles has been reported to require a mono-ubiquitin tag on the cytosol tail and binding by an escorting complex, endosomal sorting complex required for transport 1 [35]. However, CD63 transfer to internal vesicles does not require ubiquitination of the cytosol-sorting sequence [36]. The cytosolsorting motif (Y) at the C terminus of sTNFR1–tm–Y was derived from CD63. Consequently, the internal vesicle transfer of sTNFR1–tm–Y was likely similar to the transfer of CD63 and independent of ubiquitination. As a result of internal vesicle transfer, the major N-terminal region of the sTNFR1– tm–Y would be on the outside of the internal vesicle. At this position, proteolysis would generate sTNFR1 for release into the lumen of the secretory lysosome. As a consequence, the sTNFR1 formed would be deposited in the secretory lysosome matrix and/or dense cores.

sTNFR1–tm–Y-transfected RBL-1 cells. This lack of Ca2⫹triggered degranulation indicated a secretion deficiency in RBL-1 cells. In spite of this, Ca2⫹ ionophore promoted release and secretion of sTNFR1 in RBL-1 cells transfected with sTNFR1–tm–Y– egfp. This may correspond to the regulated secretion from secretory lysosomes or ectodomain shedding at the cell surface. The discrepancy in Ca2⫹-dependent secretion between sTNFR1–tm–Y- and sTNFR1–tm–Y– egfp-transfected RBL-1 cells may reflect a difference in compartmentalization of the targeted products corresponding to the differences in intracellular transport discussed above. PMA activation of PKC is a strong signal for shedding of sTNFR at the cell surface [19]. In support of this, PMA promoted sTNFR1 release from sTNFR1–tm–Y– egfp-transfected RBL-1 cells. However, PMA did not promote secretory lysosome secretion (Fig. 1). This is consonant with an inability of PMA to stimulate granule release unless Ca2⫹ influx is generated with a Ca2⫹ ionophore [30].

CONCLUSIONS Taken together, our results show that an exogenous protein such as sTNFR1–tm–Y is not only targeted to secretory lysosomes in hematopoietic cells but is also subject to regulated secretion once stored. Taking advantage of a sorting signal should be generally applicable for tm protein targeting to secretory lysosomes of hematopoietic precursors during granule biogenesis. Moreover, targeting and regulated secretion might be more efficient in normal hematopoietic cells than in the transformed cell lines used here. A local deposition of cytokines and soluble cytokine receptors would strengthen specific effects and diminish systemic effects. It should now be possible to express sTNFR1–tm–Y, for instance, in normal NK cells and CTLs during differentiation of bone marrow cells for investigation of the intracellular localization, storage in secretory granules, and regulated secretion. Furthermore, it would be interesting to see whether sTNFR1 can be delivered by NK cells and CTLs in vivo. Azurophil granules of neutrophils, conversely, would be vehicles for delivering proteins into phagosomes to promote an antimicrobial defense.

Regulated secretion We found a discrepancy in the secretory capacity between RBL-2H3 cells, which grow in adherent layers, and RBL-1 cells, which grow in suspension. RBL-2H3 was able to secrete endogenous secretory lysosome constituents, but RBL-1 cells lacked such a capacity. Results from experiments with RBL2H3 cells clearly showed support for our concept in that regulated secretion of secretory lysosome-targeted sTNFR1 is induced by a Ca2⫹ ionophore or more physiologically, by IgE receptor activation. Furthermore, sTNFR1–tm–Y seemed to be rather stable in the secretory lysosome environment, with the exception of when the tm–Y fragment was cleaved by proteolysis, resulting in generation of membrane-free sTNFR1. The Ca2⫹-triggered secretion induced the release of the sTNFR1, consistent with discharge from the granule matrix and dense cores of the secretory lysosome. In contrast, neither RMCP-II nor sTNFR1 was, however, secreted upon Ca2⫹ influx in 884


ACKNOWLEDGMENTS This work was supported by the Swedish Cancer Foundation, the Swedish Childhood Cancer Foundation, the Swedish Research Council (Grants 1329 and 12613), the Crafoord Foundation, the Alfred O¨ sterlund Foundation, and the Greta and Johan Kock Foundation. We thank Hans Janssen, Nico Ong, and Ann-Maj Persson for their expert technical assistance.

REFERENCES 1. Gao, Y., Rosen, H., Johnsson, E., Calafat, J., Tapper, H., Olsson, I. (2003) Sorting of soluble TNF-receptor for granule storage in hematopoietic cells as a principle for targeting of selected proteins to inflamed sites. Blood 102, 682– 688.


2. Borregaard, N., Cowland, J. B. (1997) Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89, 3503–3521. 3. Gullberg, U., Bengtsson, N., Bulow, E., Garwicz, D., Lindmark, A., Olsson, I. (1999) Processing and targeting of granule proteins in human neutrophils. J. Immunol. Methods 232, 201–210. 4. Bulow, E., Gullberg, U., Olsson, I. (2000) Structural requirements for intracellular processing and sorting of bactericidal/permeability-increasing protein (BPI): comparison with lipopolysaccharide-binding protein. J. Leukoc. Biol. 68, 669 – 678. 5. Bulow, E., Nauseef, W. M., Goedken, M., McCormick, S., Calafat, J., Gullberg, U., Olsson, I. (2002) Sorting for storage in myeloid cells of nonmyeloid proteins and chimeras with the propeptide of myeloperoxidase precursor. J. Leukoc. Biol. 71, 279 –288. 6. Hansson, M., Jonsson, S., Persson, A. M., Calafat, J., Tapper, H., Olsson, I. (2003) Targeting proteins to secretory lysosomes of natural killer cells as a principle for immunoregulation. Mol. Immunol. 40, 363–372. 7. Seldin, D. C., Adelman, S., Austen, K. F., Stevens, R. L., Hein, A., Caulfield, J. P., Woodbury, R. G. (1985) Homology of the rat basophilic leukemia cell and the rat mucosal mast cell. Proc. Natl. Acad. Sci. USA 82, 3871–3875. 8. Dell’Angelica, E. C., Mullins, C., Caplan, S., Bonifacino, J. S. (2000) Lysosome-related organelles. FASEB J. 14, 1265–1278. 9. Griffiths, G. M., Isaaz, S. (1993) Granzymes A and B are targeted to the lytic granules of lymphocytes by the mannose-6-phosphate receptor. J. Cell Biol. 120, 885– 896. 10. Arvan, P., Castle, D. (1998) Sorting and storage during secretory granule biogenesis: looking backward and looking forward. Biochem. J. 332, 593– 610. 11. Gerdes, H. H., Rosa, P., Phillips, E., Baeuerle, P. A., Frank, R., Argos, P., Huttner, W. B. (1989) The primary structure of human secretogranin II, a widespread tyrosine-sulfated secretory granule protein that exhibits low pH- and calcium-induced aggregation. J. Biol. Chem. 264, 12009 – 12015. 12. Tooze, S. A., Martens, G. J., Huttner, W. B. (2001) Secretory granule biogenesis: rafting to the SNARE. Trends Cell Biol. 11, 116 –122. 13. Blott, E. J., Griffiths, G. M. (2002) Secretory lysosomes. Nat. Rev. Mol. Cell Biol. 3, 122–131. 14. Raposo, G., Nijman, H. W., Stoorvogel, W., Liejendekker, R., Harding, C. V., Melief, C. J., Geuze, H. J. (1996) B lymphocytes secrete antigenpresenting vesicles. J. Exp. Med. 183, 1161–1172. 15. Baram, D., Adachi, R., Medalia, O., Tuvim, M., Dickey, B. F., Mekori, Y. A., Sagi-Eisenberg, R. (1999) Synaptotagmin II negatively regulates Ca2⫹-triggered exocytosis of lysosomes in mast cells. J. Exp. Med. 189, 1649 –1658. 16. Brumell, J. H., Volchuk, A., Sengelov, H., Borregaard, N., Cieutat, A. M., Bainton, D. F., Grinstein, S., Klip, A. (1995) Subcellular distribution of docking/fusion proteins in neutrophils, secretory cells with multiple exocytic compartments. J. Immunol. 155, 5750 –5759. 17. Guo, Z., Turner, C., Castle, D. (1998) Relocation of the t-SNARE SNAP-23 from lamellipodia-like cell-surface projections regulates compound exocytosis in mast cells. Cell 94, 537–548. 18. Tardieux, I., Webster, P., Ravesloot, J., Boron, W., Lunn, J. A., Heuser, J. E., Andrews, N. W. (1992) Lysosome recruitment and fusion are early events required for trypanosome invasion of mammalian cells. Cell 71, 1117–1130.

19. Lantz, M., Gullberg, U., Nilsson, E., Olsson, I. (1990) Characterization in vitro of a human tumor necrosis factor-binding protein. A soluble form of a tumor necrosis factor receptor. J. Clin. Invest. 86, 1396 –1342. 20. Kozak, M. (1987) An analysis of 5⬘-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15, 8125– 8148. 21. Gullberg, U., Lindmark, A., Nilsson, E., Persson, A. M., Olsson, I. (1994) Processing of human cathepsin G after transfection to the rat basophilic/ mast cell tumor line RBL. J. Biol. Chem. 269, 25219 –25225. 22. Garwicz, D., Lindmark, A., Gullberg, U. (1995) Human cathepsin G lacking functional glycosylation site is proteolytically processed and targeted for storage in granules after transfection to the rat basophilic/mast cell line RBL or the murine myeloid cell line 32D. J. Biol. Chem. 270, 28413–28418. 23. Andersson, E., Hellman, L., Gullberg, U., Olsson, I. (1998) The role of the propeptide for processing and sorting of human myeloperoxidase. J. Biol. Chem. 273, 4747– 4753. 24. Hultberg, B., Lindsten, J., Sjoblad, S. (1976) Molecular forms and activities of glycosidases in cultures of amniotic-fluid cells. Biochem. J. 155, 599 – 605. 25. Bretz, R., Staubli, W. (1977) Detergent influence on rat-liver galactosyltransferase activities towards different acceptors. Eur. J. Biochem. 77, 181–192. 26. Calafat, J., Janssen, H., Stahle-Backdahl, M., Zuurbier, A. E., Knol, E. F., Egesten, A. (1997) Human monocytes and neutrophils store transforming growth factor-␣ in a subpopulation of cytoplasmic granules. Blood 90, 1255–1266. 27. Slot, J. W., Geuze, H. J., Gigengack, S., Lienhard, G. E., James, D. E. (1991) Immuno-localization of the insulin regulatable glucose transporter in brown adipose tissue of the rat. J. Cell Biol. 113, 123–135. 28. Baldassarre, M., Dragonetti, A., Marra, P., Luini, A., Isidoro, C., Buccione, R. (2000) Regulation of protein sorting at the TGN by plasma membrane receptor activation. J. Cell Sci. 113, 741–748. 29. Beaven, M. A., Guthrie, D. F., Moore, J. P., Smith, G. A., Hesketh, T. R., Metcalfe, J. C. (1987) Synergistic signals in the mechanism of antigeninduced exocytosis in 2H3 cells: evidence for an unidentified signal required for histamine release. J. Cell Biol. 105, 1129 –1136. 30. De Matteis, M. A., Di Tullio, G., Buccione, R., Luini, A. (1991) Characterization of calcium-triggered secretion in permeabilized rat basophilic leukemia cells. Possible role of vectorially acting G proteins. J. Biol. Chem. 266, 10452–10460. 31. Bonifacino, J. S., Traub, L. M. (2003) Signals for sorting of transmembrane proteins to endosomes and lysosomes. Annu. Rev. Biochem. 72, 395– 447. 32. Rohrer, J., Schweizer, A., Russell, D., Kornfeld, S. (1996) The targeting of Lamp1 to lysosomes is dependent on the spacing of its cytoplasmic tail tyrosine sorting motif relative to the membrane. J. Cell Biol. 132, 565– 576. 33. Simpson, F., Peden, A. A., Christopoulou, L., Robinson, M. S. (1997) Characterization of the adaptor-related protein complex, AP-3. J. Cell Biol. 137, 835– 845. 34. Jiang, L., Erickson, A., Rogers, J. (2002) Multivesicular bodies: a mechanism to package lytic and storage functions in one organelle? Trends Cell Biol. 12, 362–367. 35. Katzmann, D. J., Babst, M., Emr, S. D. (2001) Ubiquitin-dependent sorting into the multivesicular body pathway requires the function of a conserved endosomal protein sorting complex, ESCRT-I. Cell 106, 145–155. 36. Stoorvogel, W., Kleijmeer, M. J., Geuze, H. J., Raposo, G. (2002) The biogenesis and functions of exosomes. Traffic 3, 321–330.


885