Heat shock protein 70 surface-positive tumor exosomes stimulate migratory and cytolytic activity of natural killer cells.

Detergent-soluble membrane vesicles are actively released by human pancreas (Colo-/Colo+) and colon (CX-/CX+) carcinoma sublines, differing in their capacity to present heat shock protein 70 (Hsp70)/Bag-4 on their plasma membranes. Floating properties, acetylcholine esterase activity, and protein composition characterized them as exosomes. An enrichment of Rab-4 documented their intracellular transport route from early endosomes to the plasma membrane. After solubilization, comparable amounts of cytosolic proteins, including tubulin, Hsp70, Hsc70, and Bag-4, but not ER-residing Grp94 and calnexin, were detectable in tumor-derived exosomes. However, with respect to the exosomal surface, only Colo+/CX+ but not Colo-/CX- derived exosomes were Hsp70 membrane positive. Therefore, concomitant with an up-regulated cell surface density of activation markers, migration and Hsp70 reactivity of natural killer (NK) cells was stimulated selectively by Hsp70/Bag-4 surface-positive exosomes, but not by their negative counterparts and tumor cell lysates. Moreover, the exosome-mediated lytic activity of NK cells was blockable by Hsp70-specific antibody. As already shown for TKD stimulation, NK cells preincubated with Hsp70 surface-positive exosomes initiated apoptosis in tumors through granzyme B release. In summary, our data provide an explanation how Hsp70 reactivity in NK cells is induced by tumor-derived exosomes.

Treatment of colon and lung cancer patients with ex vivo heat shock protein 70-peptide-activated, autologous natural killer cells: a clinical phase i trial.

PURPOSE: The 14 amino acid sequence (aa(450-463)) TKDNNLLGRFELSG (TKD) of heat shock protein 70 (Hsp70) was identified as a tumor-selective recognition structure for natural killer (NK) cells. Incubation of peripheral blood lymphocyte cells with TKD plus low-dose interleukin 2 (IL-2) enhances the cytolytic activity of NK cells against Hsp70 membrane-positive tumors, in vitro and in vivo. These data encouraged us to test tolerability, feasibility, and safety of TKD-activated NK cells in a clinical Phase I trial. EXPERIMENTAL DESIGN: Patients with metastatic colorectal cancer (n = 11) and non-small cell lung cancer (n = 1) who had failed standard therapies were enrolled. After ex vivo stimulation of autologous peripheral blood lymphocytes with Hsp70-peptide TKD (2 microg/ml) plus low-dose IL-2 (100 units/ml), TKD was removed by extensive washing, and activated cells were reinfused i.v. The procedure was repeated for up to six cycles, applying a dose escalation schedule in 4 patients. RESULTS: The percentage of activated NK cells in the reinfused leukapheresis products ranged between 8 and 20% of total lymphocytes, corresponding to total NK cell counts of 0.1 up to 1.5 x 10(9). Apart from restless feeling in 1 patient and itching in 2 patients, no negative side effects were observed. Concomitant with an enhanced CD94 cell surface density, the cytolytic activity of NK cells against Hsp70 membrane-positive colon carcinoma cells was enhanced after TKD/IL-2 stimulation in 10 of 12 patients. Concerning tumor response, 1 patient was in stable disease during therapy by formal staging criteria and another patient showed stable disease in one metastases and progression in another. CONCLUSIONS: Reinfusion of Hsp70-activated autologous NK cells is safe. Immunological results warrant additional studies in patients with lower tumor burden.

Heat shock protein 70-reactivity is associated with increased cell surface density of CD94/CD56 on primary natural killer cells.

Previously we described an involvement of the C-type lectin receptor CD94 and the neuronal adhesion molecule CD56 in the interaction of natural killer (NK) cells with Hsp70-protein and Hsp70-peptide TKD. Therefore, differences in the cell surface density of these NK cell-specific markers were investigated comparatively in CD94-sorted, primary NK cells and in established NK cell lines NK-92, NKL, and YT after TKD stimulation. Initially, all NK cell types were positive for CD94; the CD56 expression varied. After stimulation with TKD, the mean fluorescence intensity (mfi) of CD94 and CD56 was upregulated selectively in primary NK cells but not in NK cell lines. Other cell surface markers including natural cytotoxicity receptors remained unaffected in all cell types. CD3-enriched T cells neither expressing CD94 nor CD56 served as a negative control. High receptor densities of CD94/CD56 were associated with an increased cytolytic response against Hsp70 membrane-positive tumor target cells. The major histocompatibility complex (MHC) class I-negative, Hsp70-positive target cell line K562 was efficiently lysed by primary NK cells and to a lower extent by NK lines NK-92 and NKL. YT and CD3-positive T cells were unable to kill K562 cells. MHC class-I and Hsp70-positive, Cx + tumor target cells were efficiently lysed only by CD94-sorted, TKD-stimulated NK cells with high CD94/CD56 mfi values. Hsp70-specificity was demonstrated by antibody blocking assays, comparative phenotyping of the tumor target cells, and by correlating the amount of membrane-bound Hsp70 with the sensitivity to lysis. Remarkably, a 14-mer peptide (LKD), exhibiting only 1 amino acid exchange at position 1 (T to L), neither stimulated Hsp70-reactivity nor resulted in an upregulated CD94 expression on primary NK cells. Taken together our findings indicate that an MHC class I-independent, Hsp70 reactivity could be associated with elevated cell surface densities of CD94 and CD56 after TKD stimulation.

[(2-Phenylindol-3-yl)methylene]propanedinitriles inhibit the growth of breast cancer cells by cell cycle arrest in G(2)/M phase and apoptosis.

Cell cycle arrest of malignant cells is an important option for cancer treatment. In this study, we modified the structure of antimitotic 2-phenylindole-3-carbaldehydes by condensation with malononitrile. The resulting methylene propanedinitriles inhibited the growth of MDA-MB 231 and MCF-7 breast cancer cells with IC(50) values below 100 nM. Though they exhibited similar structure-activity relationships as the aldehydes, they did not inhibit tubulin polymerization but were capable of blocking the cell cycle in G(2)/M phase. The cell cycle arrest was accompanied by apoptosis as demonstrated by the activation of caspases 3 and 9. Since the new 2-phenylindole derivatives also inhibited the growth of transplanted MXT mouse mammary tumors, they are interesting candidates for further development.

Antimitotic activities of 2-phenylindole-3-carbaldehydes in human breast cancer cells.

Small molecules such as indoles are attractive as inhibitors of tubulin polymerization. Thus a number of 2-phenylindole-3-carbaldehydes with lipophilic substituents in both aromatic rings was synthesized and evaluated for antitumor activity in MDA-MB 231 and MCF-7 breast cancer cells. Some 5-alkylindole derivatives with a 4-methoxy group in the 2-phenyl ring strongly inhibit the growth of breast cancer cells with IC(50) values of 5-20nM. Their action can be rationalized by the cell cycle arrest in G(2)/M phase due to the inhibition of tubulin polymerization.

Alternative mechanism by which IFN-gamma enhances tumor recognition: active release of heat shock protein 72.

IFN-gamma exhibits differential effects depending on the target and can induce cellular activation and enhance survival or mediate cell death via activation of apoptotic pathways. In this study, we demonstrate an alternative mechanism by which IFN-gamma enhances tumor recognition, mediated by the active release of Hsp72. We demonstrate that stimulation of 4T1 breast adenocarcinoma cells and K562 erythroleukemic cells with IFN-gamma triggers the cellular stress response, which results in the enhanced expression of total Hsp72 expression without a significant increase in cell death. Intracellular expression of Hsp72 was abrogated in cells stably transfected with a mutant hsf-1 gene. IFN-gamma-induced Hsp72 expression correlated with enhanced surface expression and consequent release of Hsp72 into the culture medium. Pretreatment of tumors with compounds known to the block the classical protein transport pathway, including monensin, brefeldin A, tunicamycin, and thapsigargin, did not significantly block Hsp72 release. However, pretreatment with intracellular calcium chelator BAPTA-AM or disruption of lipid rafts using methyl beta-cyclodextrin completely abrogated IFN-gamma-induced Hsp72 release. Biochemical characterization revealed that Hsp72 is released within exosomes and has the ability to up-regulate CD83 expression and stimulate IL-12 release by naive dendritic cells. Pretreatment with neutralizing mAb or depletion of Hsp72 completely abrogated its chaperokine function. Taken together, these findings are indicative of an additional previously unknown mechanism by which IFN-gamma promotes tumor surveillance and furthers our understanding of the central role of extracellular Hsp72 as an endogenous adjuvant and danger signal.

Interaction of heat shock protein 70 peptide with NK cells involves the NK receptor CD94.

Full-length Hsp70 protein (Hsp70) and the C-terminal domain of Hsp70 (Hsp70C) both stimulate the cytolytic activity of naive natural killer (NK) cells against Hsp70-positive tumor target cells. Here, we describe the characterization of Hsp70-NK cell interaction with binding studies using the human NK cell line YT. Binding of recombinant Hsp70 protein (Hsp70) and the C-terminal domain of Hsp70 (Hsp70C) to YT cells is demonstrated by immunofluorescence studies. A phenotypic characterization revealed that none of the recently described HSP-receptors (alpha2-macroglobulin receptor CD91, Toll-like receptors 2, 4, 9, CD14) are expressed on YT cells. Only the C-type lectin receptor CD94 is commonly expressed by YT cells and Hsp70 reactive NK cells. A correlation of the cell density-dependent, variable CD94 expression and the binding capacity of Hsp70 was detected. Furthermore, Hsp70 binding could be completely abrogated by preincubation of YT cells with a CD94-specific antibody. Competition assays using either unlabeled Hsp70 protein or an unrelated protein (GST) in 20-fold excess and binding studies with escalating doses of Hsp70 protein provide evidence for a specific and concentration-dependent interaction of Hsp70 with YT cells. In addition to Hsp70 and Hsp70C, a 14-mer Hsp70 peptide termed TKD is known to exhibit comparable stimulatory properties on NK cells. Similar to full-length Hsp70 protein (10 microg/ml-50 microg/ml), a specific binding of this peptide to YT cells was observed at 4 degrees C, at equivalent concentrations (2.0 microg/ml-8.0 microg/ml). Following a 30 min incubation period at 37 degrees C, membrane-bound Hsp70 protein and Hsp70 peptide TKD were completely taken up into the cytoplasm.

The cell surface-localized heat shock protein 70 epitope TKD induces migration and cytolytic activity selectively in human NK cells.

Profiling of surface-bound proteins uncovers a tumor-selective heat shock protein 70 (Hsp70) membrane expression that provides a target structure for human NK cells. Hsp70 peptide TKD (TKDNNLLGRFELSG; aa 450-463) was found to enhance the cytolytic activity of NK cells. In this study, we demonstrate that TKD-activated CD3-CD56+CD94+ NK cells are selectively attracted by Hsp70 membrane-positive tumor cells, and supernatants derived thereof. Hsp70 membrane-negative tumors failed to attract these NK cells. The capacity to migrate was associated with a substantial lytic activity against Hsp70-positive tumor cells. Because NK cell migration was independent of cell-to-cell contact, the involvement of a soluble factor was assumed. Interestingly, synthetic Hsp70 protein and Hsp70 peptide TKD, mimicking surface-bound Hsp70, initiates migration of NK cells in a concentration-dependent (1-5 microg/ml), highly selective, and chemokine-independent manner. In summary, our results indicate that Hsp70 peptide TKD not only stimulates cytolysis but also chemotaxis in CD3-CD56+CD94+ NK cells.

Effects of antineoplastic agents on cytoplasmic and membrane-bound heat shock protein 70 (Hsp70) levels.

Here we report on the study of the effects of different antineoplastic agents, including cytarabine, 4-hydroperoxyifosfamide, the activated form of ifosfamide, vincristine, and paclitaxel, with regard to their capacity to modulate the amount of cytoplasmic and membrane-bound heat shock protein 70 (Hsp70). Hsp70 levels were measured in the myelogenous leukemic cell line K562, in the human colon carcinoma cell line CX2, and in peripheral blood lymphocytes (PBL) under physiological conditions (37 degrees C), and following non-lethal heat shock at 41.8 degrees C. A concentration of 1 microM and an incubation period of 2 h were determined as non-lethal, since none of the different antineoplastic agents induced necrosis or apoptosis in untreated or heat-shocked cells under these conditions. Our results show that tubulin-interacting agents, including vincristine and paclitaxel, but not DNA-interacting agents, including cytarabine and ifosfamide, selectively increase the amount of cytoplasmic Hsp70 in tumor and normal cells, as measured by semi-quantitative Western blot analysis. Mechanistically, a vincristine- and paclitaxel-induced tubulin assembly, as demonstrated by immunofluorescence microscopy, might be responsible for the elevated cytoplasmic Hsp70 levels. Interestingly, an increased membrane expression of Hsp70 following treatment with vincristine or paclitaxel was selectively observed on tumor cells, but not on normal cells.

Tumor-derived heat shock protein 70 peptide complexes are cross-presented by human dendritic cells.

Our study demonstrates that tumor-derived heat shock protein (HSP)70 chaperones a tyrosinase peptide and mediates its transfer to human immature dendritic cells (DCs) by receptor-dependent uptake. Human tumor-derived HSP70 peptide complexes (HSP70-PC) thus have the immunogenic potential to instruct DCs to cross-present endogenously expressed, nonmutated, and tumor antigenic peptides that are shared among tumors of the melanocytic lineage for T cell recognition. T cell stimulation by HSP70-instructed DCs is dependent on the Ag bound to HSP70 in that only DCs incubated with HSP70-PC purified from tyrosinase-positive (HSP70-PC/tyr(+)) but not from tyrosinase-negative (HSP70-PC/tyr(-)) melanoma cells resulted in the specific activation of the HLA-A*0201-restricted tyrosinase peptide-specific cytotoxic T cell clone. HSP70-PC-mediated T cell stimulation is very efficient, delivering the tyrosinase peptide at concentrations as low as 30 ng/ml of HSP70-PC for T cell recognition. Receptor-dependent binding of HSP70-PC and active cell metabolism are prerequisites for MHC class I-restricted cross-presentation and T cell stimulation. T cell stimulation does not require external DC maturation signals (e.g., exogenously added TNF-alpha), suggesting that signaling DC maturation is an intrinsic property of the HSP70-PC itself and related to receptor-mediated binding. The cross-presentation of a shared human tumor Ag together with the exquisite efficacy are important new aspects for HSP70-based immunotherapy in clinical anti-cancer vaccination strategies, and suggest a potential extension of HSP70-based vaccination protocols from a patient-individual treatment modality to its use in an allogeneic setting.