Neutrophil elastase and proteinase 3 trafficking routes in myelomonocytic cells.

Neutrophil elastase (NE) and proteinase 3 (PR3) differ in intracellular localization, which may reflect different trafficking mechanisms of the precursor forms when synthesized at immature stages of neutrophils. To shed further light on these mechanisms, we compared the trafficking of precursor NE (proNE) and precursor PR3 (proPR3). Like proNE [1], proPR3 interacted with CD63 upon heterologous co-expression in COS cells but endogenous interaction was not detected although cell surface proNE/proPR3/CD63 were co-endocytosed in myelomonocytic cells. Cell surface proNE/proPR3 turned over more rapidly than cell surface CD63 consistent with processing/degradation of the pro-proteases but recycling of CD63. Colocalization of proNE/proPR3/CD63 with clathrin and Rab 7 suggested trafficking through coated vesicles and late endosomes. Partial caveolar trafficking of proNE/CD63 but not proPR3 was suggested by colocalization with caveolin-1. Blocking the C-terminus of proNE/proPR3 by creating a fusion with FK506 binding protein inhibited endosomal re-uptake of proNE but not proPR3 indicating "pro(C)"-peptide-dependent structural/conformational requirements for proNE but not for proPR3 endocytosis. The NE aminoacid residue Y199 of a proposed NE sorting motif that interacts with AP-3 [2] was not required for proNE processing, sorting or endocytosis in rat basophilic leukemia (RBL) cells expressing heterologous Y199-deleted proNE; this suggests operation of another AP-3-link for proNE targeting. Our results show intracellular multi-step trafficking to be different between proNE and proPR3 consistent with their differential subcellular NE/PR3 localization in neutrophils.

Secretory lysosome targeting and induced secretion of human soluble TNF-alpha receptor in murine hematopoietic cells in vivo as a principle for immunoregulation in inflammation and malignancy.

OBJECTIVE: Systemic administration of immunotherapeutics often gives rise to severe side effects. A local deposition, using secretory lysosomes of hematopoietic cells as vehicles for delivery, can overcome this problem. In the present study, the validity of this concept was investigated using retroviral transduction of the human soluble tumor necrosis factor-alpha receptor 1 (hsTNFR1) into murine bone marrow cells, followed by transfer of the genetically modified cells into irradiated mice. MATERIALS AND METHODS: Bone marrow cells from donor mice were transduced with retroviral vector containing cDNA for hsTNFR1, together with a transmembrane domain and a tyrosine-sorting signal in order to facilitate the endoplasmic reticulum export and to achieve secretory lysosome loading. Expression of hsTNFR1 in recipient mice was investigated using flow cytometry and Western blot. Enzyme-linked immunosorbent assay was used to measure levels of tumor necrosis factor-alpha, hsTNFR1, and murine TNFR1. RESULTS: Stable long-term expression of hsTNFR1 was achieved in transplanted mice. Hematopoietic cells, such as natural killer, T and B cells, and neutrophils contained hsTNFR1. Exposure of lipopolysaccaride (in vivo) or phorbole-myristrate esterase (in vitro) induced significant secretion of hsTNFR1. Release of endogeneous murine sTNFR1 did not differ between cells transduced with hsTNFR1 or an "empty" vector. CONCLUSION: Long-term expression in vivo and inducible secretion of hsTNFR1 in murine hematopoietic cells support the potential use of storage organelles in hematopoietic cells as vehicles for targeting inflamed/malignant sites with therapeutically active agents.

The tetraspanin CD63 is involved in granule targeting of neutrophil elastase.

Targeting mechanisms of neutrophil elastase (NE) and other luminal proteins stored in myeloperoxidase (MPO)-positive secretory lysosomes/primary granules of neutrophils are unknown. These granules contain an integral membrane protein, CD63, with an adaptor protein-3-dependent granule delivery system. Therefore, we hypothesized that CD63 cooperates in granule delivery of the precursor of NE (proNE). Supporting this hypothesis, an association was demonstrated between CD63 and proNE upon coexpression in COS cells. This also involved augmented cellular retention of proNE requiring intact large extracellular loop of CD63. Furthermore, depletion of CD63 in promyelocytic HL-60 cells with RNA interference or a CD63 mutant caused reduction of cellular NE. However, the proNE steady-state level was similar to wild type in CD63-depleted clones, making it feasible to examine possible effects of CD63 on NE trafficking. Thus, depletion of CD63 led to reduced processing of proNE into mature NE and reduced constitutive secretion. Furthermore, CD63-depleted cells showed a lack of morphologically normal granules, but contained MPO-positive cytoplasmic vacuoles with a lack of proNE and NE. Collectively, our data suggest that granule proteins may cooperate in targeting; CD63 can be involved in ER or Golgi export, cellular retention, and granule targeting of proNE before storage as mature NE.

Neutrophil elastase sorting involves plasma membrane trafficking requiring the C-terminal propeptide.

The primary granules/secretory lysosomes of neutrophils store mature neutrophil elastase (NE) as a luminal protein after proteolytic removal of N-terminal and C-terminal pro-peptides from a proform of NE. The N-terminal pro-peptide prevents premature activation that might be toxic to the cell, but the C-terminal pro-peptide has no defined function. In this study, we investigated the role of the C-terminal pro-peptide in trafficking of NE by expressing, in rat basophilic leukemia (RBL) cells, both wild-type NE and the mutant NE/Delta248-267, which lacks the C-terminal pro-peptide. Both transfected proteins were found to be targeted to secretory lysosomes. In addition, results from antibody ligation and cell-surface biotinylation indicated that proform of NE was targeted to the plasma membrane, and then subjected to endocytosis. The results were supported by the detection of targeting of the proform to the plasma membrane followed by internalization both in RBL cells and normal granulopoietic precursor cells. Targeting of NE to the plasma membrane required the C-terminal pro-peptide as NE/Delta248-267 expressed in RBL cells bypassed plasma membrane trafficking. Our results indicate targeting of a population of NE to the plasma membrane and internalization dependent on the C-terminal NE pro-peptide.

The Leukemia-associated ETO homologues are differently expressed during hematopoietic differentiation.

The Eight twenty-one (ETO) homologues are nuclear repressor proteins including ETO, myeloid-transforming gene-related protein 1 (MTGR1), and myeloid-transforming gene chromosome 16 (MTG16). ETO and MTG16 are both part of fusion proteins resulting from chromosomal translocations associated with acute myeloid leukemia. Expression of these chimeras results in a differentiation block that contributes to the onset of leukemia. In order to elucidate the relation between the ETO homologues and hematopoietic differentiation, we determined the expression of the homologues during differentiation of leukemic and normal hematopoietic cells. Our results showed MTGR1 and MTG16 to be ubiquitously expressed in leukemic cell lines, whereas expression of ETO was observed only in an erythroleukemic cell line. The MTGR1 and MTG16 proteins decreased during all trans-retinoic acid-, but not vitamin D(3)-induced differentiation of leukemic cells. The reduction seemed to reflect a decrease in transcript levels as well as in protein stability. MTGR1 transcripts were ubiquitously expressed in human bone marrow cells. The MTG16 transcripts of CD34(+) progenitor cells were rapidly downregulated by cytokine-induced differentiation into myeloid or erythroid lineages. ETO transcripts, present at very low abundance in CD34(+) progenitor cells, were transiently upregulated during erythroid differentiation. In conclusion, the differential expression of the ETO homologues suggests that they may have a potential role in hematopoietic differentiation.

Targeting proteins to secretory lysosomes of natural killer cells as a principle for immunoregulation.

Secretory lysosomes of natural killer (NK) cells combine storage, regulated secretion and lysosomal activity. We asked whether one could target exogenous proteins to the secretory lysosomes of NK-cells for final delivery into a tumor site upon degranulation. cDNAs for both soluble and transmembrane (tm) proteins were expressed in the human YT-Indy NK-cell line. Targeting of a soluble TNF receptor (sTNFR1) was achieved by expressing a cDNA construct with a transmembrane sequence to facilitate ER-export and by incorporating a cytosolic sorting signal (Y) from CD63 to overcome constitutive secretion. The resulting sTNFR1-tm-Y was targeted to secretory lysosomes as confirmed by results from biosynthetic radiolabeling in combination with subcellular fractionation, immunoelectron microscopy, and immunofluorescence microscopy. A soluble sTNFR1 form was generated in the secretory lysosome by endogenous proteolytic activity. Expression of exogenous normally secretory non-membrane proteins, such as alpha1-microglobulin (alpha1-m) and alpha1-antitrypsin (alpha1-at) resulted mostly in constitutive secretion although a small amount of alpha1-microglobulin was targeted to secretory lysosomes. Our results suggest a potential for delivery of pharmacologically active agents into tumor sites by use of the NK-cell secretory lysosome as a carrier.

Interactions between the leukaemia-associated ETO homologues of nuclear repressor proteins.

The eight-twenty-one (ETO) homologues, represented by ETO, myeloid transforming gene-related protein 1 (MTGR1) and myeloid transforming gene chromosome 16 (MTG16), are nuclear repressor proteins. ETO is part of the fusion protein acute myeloid leukaemia (AML)1-ETO, resulting from the translocation (8;21). Similarly, MTG16 is disrupted to become part of AML1/MTG16 in t(16;21). The aberrant expression of these chimeras could affect interplay between ETO homologues and contribute to the leukaemogenic process. We investigated possible interactions between the ETO homologues. Ectopic co-expression in COS-cells resulted in heterodimerisation of the various ETO homologues suggesting that they may co-operate. Similarly, the chimeric oncoprotein AML1-ETO interacted with both MTGR1 and MTG16. However, results from cell lines endogenously expressing more than one ETO homologue did not demonstrate co-precipitation. Results from IP-Western and size determination by gel filtration of deletion mutants expressed in COS-cells, indicated an important role of the HHR domain for oligomerisation. A role was also suggested for the Nervy domain in the homologue interactions. Our results suggest that ETO homologues can interact with each other as well as with AML1-ETO, although it is unclear as to what extent these interactions occur in vivo.