Human NK cell lines migrate differentially in vitro related to matrix interaction and MMP expression.

Matrix metalloproteinases (MMPs) are thought to be of importance for the migratory ability of natural killer (NK) cells. Their expression and production may influence the amount of tumour-infiltrating NK cells and thereby any therapeutic capability. In this study, we sought to investigate the importance of MMPs for human NK cells' ability to degrade and migrate through the extracellular matrix (ECM). The two human NK cell lines, NK-92 and YT, migratory ability, MMP expression and production as well as their morphological appearance when cultured in the ECM equivalent Matrigel were analysed and compared. The quantitatively more migratory NK-92 cells were found to express invadopodia/podosomes at a significantly higher degree when cultured in Matrigel and gave rise to a general disintegration of the Matrigel. The NK-92 cells had a higher mRNA expression of MMP-2, -9, -13, MT1-, MT3- and MT6-MMP and a significantly higher production of MMP-9 compared to YT cells. These differences could explain the substantial functional difference observed between the two cell lines with respect to migratory capacity. In addition, the number of Matrigel invading NK-92 cells decreased significantly in the presence of the MMP inhibitor GM6001, demonstrating that MMPs have a critical function in their migration.

In Vitro Assessment of Human Natural Killer Cell Migration and Invasion.

Cell invasion assays are important to obtain valuable functional data relating to tissue migration and invasion of effector lymphocytes. Boyden chamber invasion assays represent a reductionist system that allows for easy manipulation using various extracellular matrix (ECM) components addressing migratory functions or invasion into/through a three-dimensional matrix where migration and invasion inhibitors as well as stimulators can be added. Presented here is a description using the Transwell(®) system where invasion and migration can be studied. It constitutes an inner cell culture well with a (PET) polycarbonate filter with 3-8 µm pores that allow for cells to transverse to the bottom chamber where they can be recorded by various methods (Fig. 1). The polycarbonate filters may be coated with ECM proteins for migration and adhesive studies or covered with a thick layer that occludes the pores to address matrix degradation, i.e., invasion as described in this chapter.

Different mechanisms of soy isoflavones in cell cycle regulation and inhibition of invasion.

BACKGROUND: Soy isoflavones, genistein, daidzein and glycitein, are thought to have beneficial effects on cancer prevention. MATERIALS AND METHODS: We used cell cycle analysis, invasion assay and immunoblotting to determine the effects of genistein and glycitein on Jurkat T cells. RESULTS: Glycitein inhibited Jurkat cell invasion at a level comparable to the inhibition by genistein. Both genistein and glycitein down-regulated MMP-13 proteolytic activity by 60-70% and MMP-8 expression. Caffeine could block G2/M arrest by genistein, but was unable to block the inhibition of invasion by genistein and glycitein. We also demonstrated that glycitein inhibited proteintyrosine phosphorylation in Jurkat cells. CONCLUSION: We determined, for the first time, that glycitein inhibited Jurkat cell invasion, in part through the down-regulation of MMP-13 activity and MMP-8 expression. Our findings also suggest that soy isoflavones may utilize different mechanisms to exert their effects on cell cycle progression and invasiveness of Jurkat cells.

2B4(CD244)-mediated activation of NK cells reduces metastases of B16F10 melanoma in mice.

BACKGROUND: Natural killer (NK) cells are a third population of lymphocytes that can kill certain tumor cells. This killing is regulated by signals received through activating and inhibitory receptors. 2B4 (CD244), a member of the CD2 subset of Immunoglobulin superfamily, was identified as an activating receptor on NK cells. Interaction of 2B4 with its ligand CD48 or anti-2B4 mAb stimulates NK cell cytolytic function as well as production of INF-gamma. MATERIALS AND METHODS: A murine tumor model was used to study the in vivo role of 2B4. 2B4 and CD48 were activated in vivo by injecting anti-2B4 and anti-CD48 monoclonal antibodies. RESULTS: Activation of 2B4 or CD48 resulted in a five-fold reduction in tumor metastasis. IFN-gamma knockout mice had a two-fold increase in metastasis as compared to wild-type after 2B4 activation. CONCLUSION: Activation of 2B4 and CD48 reduces metastasis of B16F10 melanoma cells and this anti-tumor effect involves both cytolytic function and cytokine production.

Bortezomib (millennium pharmaceuticals).

Bortezomib is a ubiquitin proteasome inhibitor under development by Millennium Pharmaceuticals (formerly LeukoSite Inc) for the potential treatment of various solid tumors [312219], [392555]. In the first quarter of 2001, Millennium initiated two phase II trials evaluating bortezomib for multiple myeloma (MM). A phase II trial in patients with chronic lymphocytic leukemia (CLL) was initiated in June 2001 [400636], [412848]. By November 2001, the agent was in a number of phase I trials and combination studies for various solid tumors, including prostate, pancreatic and colorectal carcinoma [412700], [429923], [435062], [452675]. In June 2002, bortezomib was awarded Fast Track status by the FDA [453557], and in the same month pivotal phase III trials evaluating bortezomib in MM were initiated in the US, Canada and Europe [454446].

Modulation of human NK cell lines by vascular endothelial growth factor and receptor VEGFR-1 (FLT-1).

Our laboratory has previously reported that natural killer (NK) cells bind to angiogenic microvessels in established cancer metastases. Vascular endothelial growth factor (VEGF) plays an important role in solid tumor angiogenesis by enhancing new blood vessel formation to transport nutrients and oxygen into tumors. Here we report that the human natural killer cell lines, NK-92 and YT, express the mRNA message and protein product for VEGF-B and its receptor, VEGFR-1/Flt-1. While stimulation of these cells by the potent angiogenic factor VEGF-A165, which also binds to VEGFR-1, does not alter the proliferation of the cells, it does increase adhesion to a model basement membrane-like extracellular matrix, Matrigel. VEGF-A165 also induces NK cell binding to human microvascular endothelial cells in newly forming but not established microvessels in vitro. These results suggest that human NK cells produce an angiogenic factor which may be involved in autocrine and paracrine regulations of angiogenesis. VEGF-A165 appears to stimulate NK cell adhesion to the microvasculature within established cancer metastases.

Combined chemo/anti-angiogenic cancer therapy against Lewis lung carcinoma (3LL) pulmonary metastases.

A6 is an eight amino acid peptide derived from the non-receptor binding region of urokinase plasminogen activator (uPA), which interferes with the uPA/uPA receptor system. A6 has been synthesized as a potential anti-angiogenic, anti-cancer agent. The current study has investigated the potential therapeutic activity of A6 in the Lewis lung carcinoma (3LL) model of pulmonary metastasis. A6 was found to have direct anti-tumor activity against established 3LL pulmonary metastases at a low tumor burden (10-20 colonies per lung) and was therapeutic in combination with cyclophosphamide at high tumor burdens (> 100 colonies per lung). Mechanistic studies have revealed that A6 directly inhibits the invasion of 3LL cells through a Matrigel model basement membrane by 40-45%. Moreover, treatment with either A6 or doxorubicin resulted in thicker tubes in endothelial tube formation studies. Our results suggest that A6, by virtue of its anti-invasive and anti-angiogenic properties, might work additively or synergistically with chemotherapeutic agents and thereby contribute to enhanced therapy of established 3LL cancer metastases.

Matrix metalloproteinases in cytotoxic lymphocytes impact on tumour infiltration and immunomodulation.

To efficiently combat solid tumours, endogenously or adoptively transferred cytotoxic T cells and natural killer (NK) cells, need to leave the vasculature, traverse the interstitium and ultimately infiltrate the tumour mass. During this locomotion and migration in the three dimensional environment many obstacles need to be overcome, one of which is the possible impediment of the extracellular matrix. The first and obvious one is the sub-endothelial basement membrane but the infiltrating cells will also meet other, both loose and tight, matrix structures that need to be overridden. Matrix metalloproteinases (MMPs) are believed to be one of the most important endoprotease families, with more than 25 members, which together have function on all known matrix components. This review summarizes what is known on synthesis, expression patterns and regulation of MMPs in cytotoxic lymphocytes and their possible role in the process of tumour infiltration. We also discuss different functions of MMPs as well as the possible use of other lymphocyte proteases for matrix degradation.

A KDR-binding peptide (ST100,059) can block angiogenesis, melanoma tumor growth and metastasis in vitro and in vivo.

A major goal of treatment strategies for cancer is the development of agents which can block primary tumor growth and development as well as the progression of tumor metastasis without any treatment associated side effects. Using mini peptide display (MPD) technology, we generated peptides that can bind to the human vascular endothelial growth factor (VEGF) receptor KDR. These peptides were evaluated for their ability to block angiogenesis, tumor growth and metastasis in vitro and in vivo. A D-amino acid peptide with high serum stability (ST100,059) was found to have the most potent activity in vitro as indicated by inhibition of VEGF stimulation of endothelial cells. It was also found to be the most active of the series in blocking VEGF-mediated activity in vivo, as measured in Matrigel-filled angioreactors implanted in mice. ST100,059 blocks VEGF-induced MAPK phosphorylation, as well as inhibits VEGF-induced changes in gene expression in HUVEC cells. In in vivo studies, treatment of female C57BL/6 mice inoculated with B16 mouse melanoma cells with ST100,059 resulted in a dose-dependent decrease in tumor volume and lung metastasis as compared to control groups of animals receiving vehicle alone. These studies demonstrate that by using MPD, peptides can be identified with enhanced affinity relative to those discovered using phage display. Based on these studies we have identified one such peptide ST100,059 which can effectively block tumor growth and metastasis due to its anti-angiogenic effects and ability to block intracellular signaling pathways involved in tumor progression.

Effects of IL-2 on MMP expression in freshly isolated human NK cells and the IL-2-independent NK cell line YT.

Interleukin-2 is an important activation factor for natural killer (NK) cells but its effect on NK cell matrix metalloproteinases (MMP) production and matrix degradation is less well investigated. We have used freshly isolated human NK cells and the IL-2-independent NK cell line, YT, to investigate the effects of IL-2 stimulation on NK cell invasion of Matrigel and on MMP expression and production. In YT cells, we found opposing early and late effects of IL-2 stimulation with an early (2 h) increase in MMP-9 protein level and enhanced migration in the Matrigel invasion assay and by 30 hours a decreased mRNA expression of MMP-2, MMP-9, MMP-13, MT3-MMP, and MT6-MMP. We also found a preculture period of 48 hours with IL-2 to negatively affect YT cell migration. We furthermore found that freshly isolated human NK cells Matrigel invasion was MMP-dependent and it increased in response to IL-2. Importantly, in freshly isolated human NK cells we did not see a downregulation of MMPs after 24 hours IL-2 stimulation, but instead a significant upregulation of MT6-MMP mRNA. Because of the cellular localisation of MT6-MMP, which ensures a focalized proteolytic activity, and its high expression compared with the other MMPs in freshly isolated human NK cells makes it of interest to study further.

Differential effects of proteasome inhibitors on cell cycle and apoptotic pathways in human YT and Jurkat cells.

Herein, we report differential effects of various proteasome inhibitors including clasto-lactacystin-beta-lactone, (-)-epigallocatechin gallate (EGCG) and N-Acetyl-Leu-Leu-Norleu-al (LLnL) on proteasomal activities of YT and Jurkat cells, human natural killer (NK) and T cell lines, respectively. The inhibitory rates of these inhibitors on the purified 20S proteasomal and 26S proteasomal chymotrypsin-like activity in whole cell extracts and intact cells did not show significant differences between the two cell lines. The viability of both cell lines was reduced in the presence of LLnL. Subsequent studies revealed a reduction of the mitochondrial membrane potential and caspase-3 activation in these two cell lines upon treatment with proteasome inhibitors; however, caspase-3 activation occurred much earlier in Jurkat cells. Cell cycle analysis indicated a sub-G(1) apoptotic cell population in Jurkat cells and G(2)/M arrest in YT cells after they were treated by proteasome inhibitors. Moreover, pretreatment of YT cells by a caspase inhibitor followed by a proteasome inhibitor did not increase the percentage of G(2)/M phase cells. In addition, accumulation of p27 and IkappaB-alpha was detected only in Jurkat cells, but not YT cells. In summary, proteasome inhibitors may act differentially in cell cycle arrest and apoptosis of tumors of NK and T cell origin, and may have similar effects on normal NK and T cells.

Physical association of uPAR with the alphaV integrin on the surface of human NK cells.

The urokinase-type plasminogen activator receptor (uPAR) serves as a receptor for urokinase plasminogen activator (uPA) and plays a role in invasion and migration of certain immune cells, including NK cells. Although uPAR is anchored to the plasma membrane via a glycosylphosphatidylinositol lipid moiety, we have previously shown that uPAR crosslinking results in MAP kinase signaling and increased integrin expression on the surface of the human NK cell line, YT. We report, herein, that the binding of uPA to uPAR also activates the MAP kinase signaling cascade. Furthermore, we show the physical association between uPAR and integrins on YT cells using cocapping and fluorescence microscopy. These results suggest that signaling initiated by either uPAR binding to uPA or by uPAR clustering may depend on the physical association of uPAR with integrins, a process that may be a prerequisite for NK cell accumulation within established tumor metastases during adoptive therapy.

Urokinase-type plasminogen activator receptor crosslinking in an NK cell line increases integrin surface expression by the MAP kinase/ERK 1/2 signaling pathway.

Urokinase-type plasminogen activator receptor (uPAR) is attached to cell membranes by a glycosylphosphatidylinositol (GPI) anchor, and as such is devoid of an intracellular domain, but is nevertheless able to initiate signal transduction. Herein, we report a relationship between integrins and uPAR on the surface of the human NK cell line, YT. Our data reveals that crosslinking uPAR, which mimics uPAR clustering at focal adhesion sites, causes increases in expression of the alpha(M), alpha(V), and beta(2) integrins on the surface of YT cells. Activation of the MEK/ERK signaling cascade occurs following uPAR crosslinking, as phosphorylation of both MEK 1/2 and ERK 1/2 results from receptor clustering. The MEK-specific inhibitors PD98059 and U0126 blocked MAP kinase phosphorylation; furthermore, PD98059 inhibited the increase in integrin expression induced by uPAR clustering. This study suggests that uPAR is a signaling receptor and regulator of integrins in NK cells and may impact NK cell function, including the potential for their accumulation within tumor metastases following adoptive transfer.

Evidence of inter- and intra-molecular crosslinking of tyrosine residues of calmodulin induced by photo-activation of ruthenium(II).

Tris(2,2'-bipyridyl)ruthenium(n) upon illumination with light at a wavelength of 450 nm in the presence of an electron acceptor induces dityrosine crosslinking in proteins.

Activation of multiple caspases and modification of cell surface fas (CD95) in proteasome inhibitor-induced apoptosis of rat natural killer cells.

The proteasome is a multi-subunit protease complex that is involved in intracellular protein degradation in eukaryotes. Previously, we have reported that selective, synthetic chymotryptic proteasome inhibitors inhibit A-NK cell-mediated cytotoxicity by approximately 50%; however, the exact role of the proteasome in NK cell-mediated cytotoxicity remains unknown. Herein, we report that proteasome inhibitors, MG115 and MG132, decreased the proteasome chymotrypsin-like activity in the rat natural killer cell line RNK16 by 85% at a concentration of 5 microM. The viability of RNK16 cells was also reduced in the presence of these inhibitors. Both inhibitors induced the apoptosis of RNK16 cells, as shown by DNA fragmentation, caspase-3 activation and the appearance of sub-G-cell populations. An increase in the fraction of apoptotic cells was observed in a dose- and time-dependent manner in our studies. In addition, the activity of caspase-1, -2, -6, -7, -8, and -9, was increased following the treatment of RNK16 cells with these inhibitors. Further investigation revealed that the expression of Fas (CD95) protein on the RNK16 cell surface was increased after the treatment by MG115 or MG132, indicating that apoptosis induced by proteasome inhibitors in RNK16 cells might be mediated through the Fas (CD95)-mediated death pathway as well. Our studies indicate, for the first time, that proteasomal chymotryptic inhibitors can reduce natural killer cell viability and therefore indirectly inhibit cell-mediated cytotoxicity via the apoptosis-inducing properties of these agents.

Conformational and enzymatic changes of 20S proteasome of rat natural killer cells induced by mono- and divalent cations.

We have been investigated the relation between activation of "neutral" and "acidic" chymotrypsin-like (ChT-L) activity and conformational changes in the 20S proteasome complex from the rat natural killer (NK) cells induced by SDS, mono- and divalent cations. The conformational changes were monitored by tryptophan fluorescence and light scattering. It was revealed that the changes in the maximum position and contribution of the short-wavelength spectral component correlated with the alteration of ChT-L activity of the proteasome. Statistical analysis was applied to assign the fluorescence components with tryptophan residues based on the classification of calculated structural parameters of the environment of tryptophan fluorophores in protein. It was proposed that the emission of W13 from alpha6-subunit located near the cluster of highly conserved proteasome residues is mostly sensitive to the activation of the enzyme. We concluded that the expression of maximal ChT-L activity of 20S proteasome is associated with the conformational changes occurs in this cluster that lead to the proteasome open conformation, allowing substrate access into the proteolytic chamber.

IL-2-mediated upregulation of uPA and uPAR in natural killer cells.

Urokinase plasminogen activator (uPA) and its receptor uPAR play a major role in immune cell-mediated, including natural killer (NK) cell-mediated, degradation of extracellular matrices. Herein, we investigate the effects of IL-2 on NK cell uPA and uPAR. RNA and protein analyses showed upregulation of uPA and uPAR following IL-2 stimulation. Gel-shift assays and Western blots detected uPA and uPAR mRNA binding proteins (mRNABPs), previously shown to destabilize uPA and uPAR mRNA. Following IL-2 stimulation, a downregulation of uPAR mRNABP and a reciprocal induction of uPAR mRNA were noted. The increase in uPA following IL-2 stimulation appeared to be more transcriptionally regulated. These data suggest that IL-2 upregulates both uPA and uPAR in NK cells through posttranscriptional as well as transcriptional mechanisms, partially explaining increases in NK cell invasiveness following IL-2 stimulation.

Tumor-localization by adoptively transferred, interleukin-2-activated NK cells leads to destruction of well-established lung metastases.

We have shown previously that i.v. injection of interleukin-2-(IL-2) activated natural killer (A-NK) cells together with IL-2 leads to a substantial localization of the A-NK cells into most, but not all, well-established B16 lung metastases in C57BL/6 mice within 12-24 hr. We demonstrate that the morphology of the lung metastases, (loose or more compact in appearance), and their location in the lungs (on the surface or deep in the lung parenchyma) are closely tied to the infiltration-permissiveness of the metastases as well as their sensitivity to treatment with A-NK cells. Although more than 1100 A-NK cells/mm(2) were found in deep metastases with a "loose" morphology (D-L), only 534, 90 and 89 cells/mm(2) were found in surface-loose (S-L), surface-compact (S-C) and deep-compact (D-C) metastases, respectively. The best infiltrated metastases responded best to the A-NK cell therapy. Thus, metastases of the D-L phenotype became reduced by 65-90% after treatment with 2 x 10(6) A-NK cells and IL-2 (120000 IU Peg-IL-2 every 12 hr for 3 days) compared to similar lesions in animals treated with PEG-IL-2 alone. In contrast, poorly infiltrated metastases, that is lesions of the compact phenotype (D-C and S-C) as well as loose metastases on the lung surface (S-L), did not react significantly to this treatment. We conclude that adoptively transferred A-NK cells are able to eliminate even well-established metastases. The existence of metastases that are resistant to infiltration by the transferred effector cells at time of treatment might reduce the efficacy of cell-based immuno-therapeutic ventures.

NK cells and the tumour microenvironment: implications for NK-cell function and anti-tumour activity.

Although it is clear that natural killer (NK) cells have the ability to recognize and kill tumour cells in vitro, their potential as a highly effective treatment for tumours has not yet been realized in the clinical setting. Following activation, endogenous and adoptively transferred NK cells can be found in tumours. However, not all tumours are equally well-infiltrated, and many of the infiltrating cells do not make target-cell contact but rather reside in the tumour stroma. New insights into the migration of NK cells, their activation status and production of matrix-degrading proteases might help to overcome this localization defect, with implications for the treatment of human cancer.