| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1,2
* Childhood Cancer Research Unit, Department of Woman and Child Health,
Department of Medical Biochemistry and Biophysics, and
Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden;
Department of Cell Biology and Histology, Faculty of Medicine, University of Tromsö, Tromsö, Norway; and
|| Northern Institute for Cancer Research, Newcastle University, Newcastle upon Tyne, UK
2Correspondence: Childhood Cancer Research Unit, Astrid Lindgren Children’s Hospital, Q6:05, Karolinska Hospital, S-171 76 Stockholm, Sweden, E-mail: baldur.sveinbjornsson{at}ki.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: eicosanoids • 5-lipoxygenase • apoptosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
Neuroblastoma, the most common extracranial solid tumor in children, is characterized by a heterogeneous clinical behavior. Some tumors may undergo spontaneous regression or differentiation, whereas the majority of metastatic neuroblastomas have poor prognosis despite intensive therapy (2)
.
In comparison with nonmalignant nervous tissue, neuroblastoma cells contain increased levels of arachidonic acid, the main substrate for eicosanoid biosynthesis catalyzed by cyclooxygenases (COXs) and lipoxygenases (LOs; refs.3
4
5
). The COX pathway generates prostaglandins and thromboxane, whereas the LO pathway generates leukotrienes, lipoxins, and hydroxyecosatetraenoic acid (Fig. 1
). The role of COX-2 and prostaglandins in cancer has been widely studied and implicated in resistance to apoptosis as well as induction of metastasis and angiogenesis (3)
.
|
In contrast to prostaglandins, leukotrienes are mostly produced by inflammatory cells, even though leukotriene production in cells of nonhematopoetic origin also has been reported (6
7
8)
. Leukotrienes are produced in a sequence of events that include phospholipase A2-mediated release of arachidonic acid and translocation of 5-LO to the nuclear envelope, where it associates with 5-LO activating protein (FLAP). Subsequently, 5-LO oxidizes arachidonic acid to 5-HPETE, which may be reduced to 5-hydroxyeicosatetranoic acid (5-HETE). Alternatively, 5-HPETE is dehydrated into the unstable epoxide leukotriene (LT) A4. This intermediate can then be converted into LTB4 by LTA4 hydrolase (LTA4H) or conjugated with glutathione to form LTC4 in a reaction catalyzed by a specific LTC4 synthase (LTC4S). After extracellular transfer by multidrug-resistance-associated protein-1 (MRP1), LTC4 may be converted to LTD4 and further into LTE4 (5
, 6)
. LTC4, LTD4, and LTE4 are collectively known as cysteinyl leukotrienes (CysLT; ref. 6
; Fig. 1
).
LTB4 is a potent chemoattractant that exerts its effect through two G-protein- coupled receptors, designated BLT1 and BLT2. BLT1 is a high affinity receptor for LTB4 and is expressed mainly on leukocytes, whereas BLT2 is a low affinity receptor with high expression on leukocytes, liver, and spleen and weak expression in most other tissues (9)
. Likewise, cysteinyl leukotrienes signal through two G-protein-coupled receptors, CysLT1 and CysLT2, which differ in ligand specificities and distribution (10)
. Cysteinyl leukotrienes have been shown to have a prominent role in inflammatory diseases including asthma and inflammatory bowel diseases (6)
.
Aberrant expression of 5-LO has been detected in various adult cancers (11
12
13
14
15
16)
. Moreover, the chemotherapeutic and anticancer effects of 5-LO enzyme pathway inhibitors have been demonstrated in experimental animal models (14
, 17
18
19)
. Yet, little is known about whether the 5-LO enzyme pathway and leukotrienes play a role in progression of tumors of neural origin.
The aim of this study was to assess the expression of the leukotriene enzyme pathway in childhood neuroblastoma and in particular the effect of leukotrienes on neuroblastoma cell growth and survival.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) were analyzed. Three childhood ganglioneuromas and 3 samples of nonmalignant adrenals from children aged 12–25 months were also included. Ethical approval was obtained by the Karolinska University Hospital Research Ethics Committee (approval 03-308).
|
Immunohistochemistry
Formalin-fixed and paraffin-embedded tissue sections were deparaffinized in xylene and graded alcohols, hydrated, and washed in PBS. After antigen retrieval in sodium citrate buffer (pH 6) in a microwave oven, the endogenous peroxidase was blocked by 0.3% H2O2 for 15 min. Sections were incubated overnight at 4°C with a monoclonal mouse anti-5-LO antibody (Research Diagnostics, Concord, MA, USA). As a secondary antibody, the anti-mouse-horseradish peroxidase (HRP) SuperPicTure Polymer detection kit was used (Zymed Laboratories, San Francisco, CA, USA). For identification of LTA4H and leukotriene receptors, sections were incubated overnight at 4°C with rabbit polyclonal antibody against LTA4H, CysLT1, CysLT2, and BLT1, respectively (Cayman Chemicals, Ann Arbor, MI, USA). Immunodetection of LTC4 synthase (LTC4S) and FLAP was performed by using rabbit polyclonal antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Sections were subsequently washed and processed with anti-rabbit-HRP SuperPicTure Polymer detection kit (Zymed). A matched isotype control was used as a control for nonspecific background staining.
Chemicals
AA-861, SC 22716, baicalein, caffeic acid, esculetin, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoleum bromide (MTT) were purchased from Sigma-Aldrich (Solna, Sweden). MK-886, Rev-5901, NDGA, and LY171883 were purchased from BioMol (Plymouth Meeting, PA, USA). MK-571, LY255283, U-75302, and 5(S)-HETE were purchased from Cayman Chemicals. Montelukast sodium was a generous gift from Merck MSD (Rahway, NJ, USA). All reagents were dissolved in DMSO and further diluted in culture medium (final DMSO concentration always <0.5%). Arachidonic acid was from Nu-Check Prep (Elysian, MN, USA). The pan-caspase inhibitor carbobenzoyloxy-Val-Ala-Asp (Ome) fluoromethyl ketone (zVADfmk; R&D Systems, Abingdon, UK) was used at a 10 µM final concentration in culture medium.
Cell lines
The human neuroblastoma cell lines SH-SY5Y, SK-N-BE(2), SK-N-SH, SK-N-AS, SK-N-FI, SK-N-DZ, and IMR-32 were cultured as described previously (20)
. The human myelocytic cell line U937 was grown in RPMI 1640 using the same supplement as above.
Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis
RT-PCR analysis was performed as described previously (21)
. Modifications of PCR running conditions and nucleotide sequences of PCR primers used are shown in Table 2
. The human myelocytic cell line U937 was used as a positive control.
|
Western blot analysis
Isolation of proteins was performed as described previously (20)
. Equal quantities were separated by SDS-PAGE, transferred to nylon membranes (Millipore, Sundbyberg, Sweden), and probed with antibodies against 5-LO (Research Diagnostics), FLAP (FL-161), and LTC4S (H-100) (Santa Cruz Biotechnology), LTA4H, CysLT1, CysLT2 (Cayman Chemicals), cleaved caspase-3, cleaved caspase-9 (Cell Signaling, Beverly, MA, USA), and β-actin (Sigma-Aldrich). Anti-mouse IgG or anti-rabbit IgG (Pharmacia Biosciences, Uppsala, Sweden) conjugated with HRP served as secondary antibodies. Pierce Super Signal (Pierce, Rockford, IL, USA) was used for detection.
Analysis of BLT1 by flow cytometry
Quantitation of the BLT1 receptor on neuroblastoma cell lines was performed with a phycoerythrin (PE) -conjugated monoclonal anti-BLT1 antibody according to the manufacturer’s instructions (R&D Systems). PE-conjugated mouse IgG1 isotype control (R&D Systems) was used as a control for nonspecific background staining. After being labeled, cells were washed twice in PBS supplemented with 0.5% bovine serum albumin and analyzed on the FL2 channel on a FACSCalibur flow cytometer, using Cell Quest Software (Becton-Dickinson, San Jose, CA, USA).
Analysis of leukotriene biosynthesis in primary neuroblastomas and cell cultures by enzyme immunoassay
Five fresh primary tumor samples (Table 3
) were suspended in Ca2+- and Mg2+-free PBS together with a cocktail of protease inhibitors (Roche Diagnostics GmbH, Mannheim, Germany), homogenized, and sonicated 3 times for 10 s on ice. Homogenates were incubated with 2 mM ATP, 2 mM Ca2+, and 80 µM arachidonic acid for 10 min at 37°C and subsequently quenched with an equal volume of methanol. The same procedure was used for analysis of leukotriene production in neuroblastoma cell cultures. After acidification to pH 3–4 (for LTB4 assay) or pH 5–6 (for cysteinyl leukotriene assay), the samples were purified by solid-phase extraction (Supelclean LC-18, Supelco, Sigma-Aldrich) and eluted in methanol. Samples were dried under nitrogen and resuspended in enzyme immunoassay buffer. The concentration of LTB4 and cysteinyl leukotrienes was determined with enzyme immunoassay according to manufacturer's protocol (Cayman Chemicals).
|
Cell viability of neuroblastoma cells on leukotriene stimulation in vitro
The neuroblastoma cell lines SK-N-AS (1x104 cells/well) and SK-N-BE(2) (3x103cells/well) were seeded in 96-well culture plates. Cells were cultivated for 24 h and then changed to serum-free RPMI for 24 h followed by replacement with the same medium containing different concentrations of either LTB4 or LTD4. Cells were incubated further for 48, 72, or 96 h with medium changes every second day. Cell viability was assessed by the MTT assay.
Cytotoxicity assay and fluorescence-activated cell-sorting analysis
Cells were incubated in 96-well culture plates with the indicated concentrations of drugs dissolved in OptiMEM (Gibco-BRL, Sundbyberg, Sweden) supplemented with 2 mM L-glutamine, 100 IU/ml penicillin G, and 100 µg/ml streptomycin (Life Technologies, Carlsbad, CA, USA) for 48 h. Cell viability was measured using the MTT assay. Mean values of optical density measurements were calculated from six separate wells. To determine the effect of exogenous addition of leukotrienes and 5(S)-HETE on cytotoxicity mediated by 5-LO inhibition, SK-N-BE(2) cells (1x104 cells/well) were cultivated with the 5-LO inhibitor AA-861 (6 µM) in the absence or presence of 125 nM LTB4, 125 nM LTD4, or 250 nM 5(S)-HETE respectively, for 48 h.
The mitochondrial transmembrane potential was assessed in SH-SY5Y and SK-N-BE(2) after 24 h of incubation with AA-861 (10 µM), MK-886 (7 µM), or montelukast (10 µM), respectively, using tetramethylrhodamine ethyl ester (TMRE; Molecular Probes, OR, USA), as described previously (20)
. DNA content was assessed by flow cytometry essentially as described previously (22)
. Briefly, SH-SY5Y and SK-N-BE(2) cells were treated with AA-861 (10 µM), MK-886 (7 µM), or montelukast (10 µM) for 48 h, harvested, and stained with DAPI and subjected to cell cycle analysis by using single-parameter DNA flow cytometry. Assessment of early apoptosis in NB cells treated as described above was done using annexin V-fluorescein isothiocyanate and monitored by flow cytometry as recommended by the manufacturers (Sigma-Aldrich).
5-LO and CysLT1 siRNA
SK-N-BE(2) cells were seeded in 6-well culture plates in RPMI medium at a 30–50% confluence. Cells were transfected with target-specific 5-LO (sc-29596), CysLT1 (sc-43712), control (Fluorescein Conjugate)-A (sc-36869), or scrambled control (sc-37007) siRNA (Santa Cruz Biotechnology), respectively, at a concentration of 33 nM using Lipofectamin 2000 (Invitrogen, Carlsbad, CA, USA) in OptiMEM. To evaluate cell viability and siRNA efficiency, Western blot analysis of protein extracts and trypan blue exclusion assay were performed 72 h after the initial transfection. Transfection efficiency was assessed in SK-N-BE(2) cells transfected with control (Fluorescein Conjugate)-A using flow cytometry.
For measurement of leukotriene production in transfected cells, 107 SK-N-BE(2) cells were seeded in 15 cm Petri dishes and the cells were grown to 50% confluence before transfection with 5-LO using Lipofectamine 2000 as described above. Measurement of leukotriene biosynthesis was performed 72 h after the initial transfection, as described above.
Statistical analysis
Two-sided unpaired t-tests were performed to evaluate leukotriene synthesis, proliferation, and cell viability, the data was log transformed when indicated.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Twenty-six of 27 neuroblastoma samples showed specific expression of 5-LO protein in the perinuclear area and cytoplasm of the tumor cells (Fig. 2a
). Nonmalignant adrenals from children showed significant staining for 5-LO in the cortex, whereas the medulla, where neuroblastomas typically arise, showed no staining (Fig. 2a
). Three ganglioneuromas were investigated and showed 5-LO immunopositivity in the tumor-derived ganglion cells but not in the surrounding stroma (Fig. 2a
). Leukotriene synthesis is dependent not only on 5-LO but also on FLAP, which acts to present arachidonic acid released from the cellular membranes to 5-LO, facilitating leukotriene production (23)
. All human neuroblastoma cell lines investigated showed both 5-LO and FLAP mRNA and protein expression as detected by RT-PCR and Western blot (Fig. 2i, j
). Furthermore, all neuroblastoma tumors investigated were shown to express FLAP in the perinuclear area of tumor cells (Fig. 2b
). Expression of FLAP in the ganglioneuroma and normal adrenal was consistent with the expression of 5-LO (Fig. 2a, b
).
|
LTA4H is widely distributed in most tissues and cells, and clearly it is not restricted to cells expressing 5-LO (24
, 25)
. Specific immunostaining for LTA4H was detected in the cytoplasm and nucleus of tumor cells and adjacent stromal cells. Ganglioneuroma cells, as well as cortex and medullary regions of the adrenal, were also immunopositive (Fig. 2c
). LTC4S conjugates LTA4 with glutathione to form LTC4, the parent compound of CysLTs (26)
. Strong nuclear staining for LTC4S was observed in a majority of cells within the tumor samples investigated (Fig. 2d
). Ganglioneuroma cells were strongly immunopositive, whereas the cortical and medullary regions of the nonmalignant adrenals were weakly positive (Fig. 2d
). We further confirmed the expression of LTA4H and LTC4S in neuroblastoma cell lines by RT-PCR and Western blotting (Fig. 2i, j
).
Leukotrienes exert their effect through G-protein coupled receptors. Hence, we analyzed the expression of the leukotriene receptors BLT1, CysLT1, and CysLT2 in neuroblastoma primary tumors and cell lines. CysLT1 receptor staining was evident in all clinical tumor specimens and in addition was detected in the vasculature of the adjacent stroma (Fig. 2e
). Similarly, the immunopositivity for BLT1 was abundant in the cell membrane of the tumor cells and in the tumor vasculature (Fig. 2g
). Adrenal gland showed expression of CysLT1, CysLT2, and BLT1 in both the cortical and medullary compartments (Fig. 2f
). RT-PCR revealed transcripts for BLT1, CysLT1, and CysLT2 in all neuroblastoma cell lines investigated (Fig. 2e-g
). At the protein level, the leukotriene receptors CysLT1, CysLT2, and BLT1 were expressed in all neuroblastoma cell lines, as shown by Western blot and flow cytometry, respectively (Fig. 2j, k
).
Endogenous production of leukotrienes by neuroblastoma cells increases cell viability
Having established the expression of leukotriene pathway enzymes and receptors in neuroblastomas, we investigated endogenous production of leukotrienes in neuroblastoma cells. The leukotriene content in primary tumors and neuroblastoma cell lysates was measured by enzyme immunoassay. Incubation of cell homogenates with 80 µM arachidonic acid resulted in a significant leukotriene production in both primary tumors and cells (Table 3
; Fig. 3a
).
|
Moreover, given the high expression of leukotriene receptors, we analyzed the survival rate of neuroblastoma cells in response to exogenous leukotrienes. Both LTB4 and LTD4 significantly induced a concentration-dependent increase in cell viability of SK-N-AS cells (P<0.001; Fig. 3b
). Interestingly, both leukotrienes induced cell proliferation of SK-N-BE(2) cells (P<0.001; Fig. 3b
). Taken together, these results indicate the presence of a leukotriene-driven autocrine survival loop in neuroblastoma cells.
The rationale for using SK-N-BE(2) cells in the majority of the experiments in our study was based on the fact that this cell line is a multiresistant, MYCN-amplified, p53-mutated neuroblastoma cell line that differs from the majority of other neuroblastoma cell lines in that it is considerably more resistant to most conventional cancer therapeutic drugs. The other two cell lines SH-SY5Y and SK-N-AS do not exhibit any MYCN-amplification or p53 mutations. On the other hand, the SK-N-AS cell line is multiresistant whereas SH-SY5Y is not.
Leukotriene enzyme pathway inhibitors have profound effect on neuroblastoma cell growth
We proceeded to investigate whether inhibition of the leukotriene pathway could affect neuroblastoma cell growth and survival. Hence, we treated six neuroblastoma cell lines with increasing concentrations of different LO inhibitors, a FLAP inhibitor, a LTA4H inhibitor, or leukotriene receptor antagonists. Irrespective of the drug, the treatment resulted in a dose-dependent inhibition of neuroblastoma cell growth. The drug concentration needed to inhibit 50% of cell viability, EC50, is shown for the different drugs and cell lines in Table 4
. The most effective drugs were MK-886, an inhibitor of FLAP; montelukast, a CysLT1 receptor antagonist; and AA-861, an inhibitor of 5-LO. Inhibition of leukotriene function with montelukast was as effective in reducing neuroblastoma cell growth as either inhibition of 5-LO or FLAP (Table 4)
. Furthermore, exogenous addition of LTD4 and LTB4 or 5(S)-HETE to SK-N-BE(2) cells treated with AA-861 (6 µM) was shown to significantly inhibit the AA-861-mediated cytotoxicity (P<0.001; Fig. 3c
). Neither MYCN-amplification nor a drug-resistant phenotype had any effect on the sensitivity to LO pathway inhibitors (Table 4)
.
|
Leukotriene pathway inhibitors induce cell cycle arrest and apoptosis of neuroblastoma cells
Depolarization of the mitochondrial trans-membrane potential and the subsequent release of proapoptotic factors are required for the induction of the intrinsic apoptotic pathway that is affected in aggressive neuroblastomas (27)
. Treatment of SH-SY5Y and SK-N-BE(2) cells with AA-861, MK-886, or montelukast induced depolarization of the mitochondrial membrane potential (Fig. 4a
) activation of caspase-9 and caspase-3 [(Fig. 4b
; similar data for SK-N-BE(2) not shown] followed by an accumulation of neuroblastoma cells in the sub-G1 phase of the cell cycle (Fig. 4c
). Moreover, the broad caspase inhibitor zVAD-fmk almost completely blocked the cytotoxic effects of MK886, AA861, and montelukast, suggesting that all three drugs induced caspase-dependent apoptosis of SH-SY-5Y and SK-N-BE(2) cells (data not shown). Treatment of SH-SY5Y cells with MK-886 resulted in a 97% accumulation of cells in sub-G1, whereas treatment with AA-861 or montelukast resulted in a 68 and 43% accumulation of cells with sub-G1 DNA content, respectively. For the p53-mutated, MYCN-amplified, multidrug-resistant cell line SK-N-BE(2), >80% of the cells showed a sub-G1 content when treated with montelukast, whereas MK-886 and AA-861 induced prominent arrest of cells in G1 (Fig. 4c
). However, no accumulation of cells with a sub-G1 DNA content was observed with these drugs (Fig. 4c
). We therefore performed annexin V-staining to monitor early apoptotic stages. SK-N-BE(2) cells treated with MK-886 or AA-861 demonstrated positive annexin V-staining that corresponded to the activation of caspases (Fig. 4b
and data not shown). These results demonstrate that inhibitors of the 5-LO enzyme pathway and the CysLT1 receptor antagonist induce a combination of cell cycle arrest and apoptosis of neuroblastoma cells, acting via the intrinsic apoptotic pathway. To further confirm the role of 5-LO and leukotrienes in neuroblastoma growth, SK-N-BE(2) cells were transfected with 5-LO, CysLT1 or scrambled siRNA. As shown in Fig. 4d
, transfection with either 5-LO or CysLT1 siRNA resulted in a significant decrease in cell growth compared with cells transfected with scrambled siRNA (P<0.05). Specific suppression of 5-LO and CysLT1 protein by siRNA transfection was verified by Western blot (Fig. 4d
) together with a significant decrease in leukotriene production (P<0.05; Fig 4e
). Transfection efficiency was shown to be 
30% as evaluated using flourescein-labeled scrambled siRNA and flow cytometry analysis (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Eicosanoid-producing enzymes are over-expressed in adult epithelial cancers and result in production of a spectrum of mediators that may be involved in resistance to apoptosis, induction of metastasis, and angiogenesis (28
, 29)
.
We have previously reported that COX-2 is expressed in human neuroblastoma and that inhibitors of COX have significant effects in the treatment of neuroblastoma in vivo (20
, 30
, 31)
. Furthermore, simultaneous inhibition of COX and LO by diclofenac and a pan-LO inhibitor, NDGA, in the presence of arachidonic acid inhibited cell viability more potently than inhibiting either pathway alone (20)
. The current study was designed to investigate the role of the 5-LO enzyme pathway in neuroblastoma growth and survival. We analyzed neuroblastoma primary tumors from different biological subsets and all clinical stages and detected high 5-LO expression in 26 out of 27 samples. 5-LO expression was also detected in tumor cells of the more differentiated ganglioneuromas (Fig. 2a
) but not in the nonmalignant adrenal medullas, where neuroblastomas typically arise (Fig. 2a
).
The significance of the expression of 5-LO in cells of neuronal origin is not well understood. It has been shown that the expression of 5-LO mRNA and protein was higher in proliferating immature cortical neurons than in mature cultures (32)
. Furthermore, the presence of 5-LO seems to be essential for maintaining normal proliferation of immature neuroblasts, as treatment with 5-LO antisense or 5-LO inhibitors reduced cell number and proliferation (32
, 33)
. LO activity has also been reported to be involved in cell cycle progression of neuroblastoma cells (34)
. Together with our results these data indicate that 5-LO is particularly expressed in proliferating immature neuronal cells.
Aberrant expression of 5-LO has been detected in various adult cancers (11
12
13
14
15
16)
. Furthermore, polymorphisms in the promoter region of the 5-LO gene are associated with a lower risk of colon cancer (35)
. A recent study (36)
suggests that 5-LO may be a marker for early pancreatic intraepithelial neoplastic lesions. In our study, we could not detect any differences in 5-LO expression between different stages and biological subsets of neuroblastomas. Also, results from analysis of leukotriene production in primary tumor samples could not be related to different biological stages of tumors (Table 3)
.
Expression of FLAP was consistent with the expression of 5-LO in neuroblastoma samples, ganglioneuroma, and adrenal cortex (Fig. 2b
), and both 5-LO and FLAP mRNA and protein were detected in all neuroblastoma cell lines investigated (Fig. 2i, j
). The data on FLAP expression in tumor cells are limited. FLAP expression has been shown to be up-regulated in esophageal squamous cell carcinoma, and a recent study demonstrated an association between FLAP expression and poor prognosis in breast cancer (37
, 38)
. Although the mechanism of action for FLAP has not been fully elucidated, it has been suggested that its role is to present arachidonic acid to 5-LO. FLAP was shown to bind arachidonic acid, and this binding could be competed by compounds like MK-886 (39)
. FLAP also stimulates the utilization of arachidonic acid by 5-LO and increases the efficiency by which 5-LO converts 5-HPETE into LTA4 (40)
. The requirement of FLAP for 5-LO function was demonstrated when human osteosarcoma cells transfected with 5-LO required a concomitant transfection with FLAP for ionophore-induced leukotriene production (41)
.
LTA4 is an unstable epoxide intermediate from which all other leukotrienes are synthesized along two major metabolic pathways, the first one involving enzymatic hydrolysis by LTA4H, which leads to the production of LTB4. The second metabolic pathway involves conjugation of LTA4 with gluthathione by LTC4S to form LTC4, the parent compound of cysteinyl leukotrienes (Fig. 1
; ref. 26
). As shown by immunohistochemistry, strong expression of both LTA4H and LTC4S was observed in neuroblastoma cells, in ganglioneuroma, and in the adrenal gland (Fig. 2c, d
). Furthermore, both enzymes were expressed in all neuroblastoma cell lines investigated, as shown by RT-PCR and Western blot (Fig. 2i, j
). The significance of 5-LO enzyme pathway expression in the adrenal gland is unclear, although it has been previously reported and suggested that 5-LO metabolites may play a role in hormone production and secretion in adrenal cortical cells (42
43
44
45)
.
After having shown that 5-LO pathway enzymes were highly expressed in neuroblastomas, we wanted to examine whether the cells were able to produce leukotrienes. Primary neuroblastomas were shown to produce leukotrienes (Table 3)
. When neuroblastoma cultures were incubated in the presence of arachidonic acid, significant levels of leukotrienes were produced in all three cell lines investigated, indicating that the leukotriene synthesis is a general feature of neuroblastoma cells (Fig. 3a
). Biosynthesis of leukotrienes may also involve transport of LTA4 from the cytoplasm of the cell into another cell containing LTA4H and LTC4S by a process called transcellular biosynthesis (46)
. In the context of the tumor environment, LTA4 could theoretically be provided by infiltrating inflammatory cells and further converted into leukotrienes by neuroblastoma cells. However, the fact that the leukotriene pathway enzymes are present in neuroblastoma cells and leukotriene production can be detected in both primary tumors and cells in vitro clearly demonstrates endogenous leukotriene production.
Leukotrienes possess a wide variety of effects attributed to tumor growth such as increased vascular permeability, chemotaxis, and cell adhesion, as well as cell survival and proliferation (47
48
49)
. In addition, a role for leukotrienes in neoangiogenesis has been suggested, because they have been shown to increase migration and viability of endothelial cells (6
, 50)
. The effects of cysteinyl leukotrienes are mediated through CysLT1 and CysLT2, whereas BLT1 and BLT2 are activated by LTB4 (9
, 10)
. A third receptor for LTB4 is the nuclear peroxisome proliferator-activated receptor 
(PPAR-), involved in termination of inflammatory responses (51)
. Increased expression of leukotriene receptors in human tumors has previously been reported and was shown to negatively correlate with patient survival (11
, 16
, 52)
. In this study, we demonstrate the presence of CysLT1, CysLT2, and BLT1 in neuroblastoma cell lines and tumors (Fig. 2e-g
). Both LTB4 and LTD4 induced a survival and concentration-dependent proliferative response in neuroblastoma cells (Fig. 3b
). Our results demonstrate that these lipid mediators can influence the growth and survival of neuroblastoma cells in an autocrine and/or paracrine manner. Other studies (18
, 53
, 54)
suggest that the mechanisms underlying increased proliferation and cell survival response induced by leukotrienes involve activation of extracellular related kinase 1/2 and phosphoinositide 3-kinase and increased expression of antiapoptotic proteins such as BCL-2.
In vitro inhibition of the 5-LO enzyme pathway in a variety of cancer cells results in growth inhibition and apoptosis (13
, 17
, 18
, 55)
. To investigate whether the leukotriene signaling pathway represents a possible therapeutic target in neuroblastomas, we assessed the effect of different inhibitors of the 5-LO enzyme pathway and leukotriene receptor antagonists on a panel of neuroblastoma cells in vitro. Out of 13 different drugs investigated, the 5-LO inhibitor AA-861, the FLAP inhibitor MK-886, and the CysLT1 receptor antagonist montelukast were the most effective in reducing neuroblastoma cell growth. The drug concentration that was needed to inhibit 50% of cell viability (EC50) ranged from 2–14 µM for these drugs (Table 4)
. The cytotoxic effect of AA-861 on neuroblastoma cells was inhibited by addition of leukotrienes or 5(S)-HETE to the cultures (Fig. 3c
), indicating the importance of 5-LO metabolites in neuroblastoma growth and survival.
Cell death by apoptosis after inhibition of the 5-LO enzyme pathway in adult epithelial cancer cells has been reported, and specific 5-LO enzyme pathway inhibitors have been shown to be effective in experimental tumor models in vivo (14
, 17
, 18
, 55)
. We observed that treatment of neuroblastoma cells with AA-861, MK-886, or montelukast resulted in a depolarization of the mitochondrial transmembrane potential and subsequent induction of apoptosis (Fig. 4a, b
). The three different drugs have different primary targets in the leukotriene pathway, and their cytotoxic effect over time is different. Therefore, a distinct caspase cleavage pattern was not unexpected. Hence, we monitored induction of early apoptosis by annexin V staining. Treatment of NB cells with MK-886 or AA-861 resulted in positive annexin V staining. These data together with the measurements of the mitochondrial transmembrane potential, activation of caspases, and cell cycle analysis demonstrate that inhibitors of the 5-LO pathway induce apoptosis of NB cells (Fig. 4a-c
and data not shown). The fact that a transient siRNA transfection was less potent in neuroblastoma cell growth inhibition compared with chemical inhibitors or leukotriene receptor antagonist may be due to the low transfection efficiency of the siRNA constructs used in our study or the existence of a 5-LO enzyme pathway-independent mechanism of the drugs (Fig. 4d, e
).
Montelukast had a significant cytotoxic effect on neuroblastoma cells in vitro (Table 4
; Fig. 4
). In addition to its function as a CysLT1 receptor antagonist, montelukast may also have a direct inhibitory effect on 5-LO at therapeutically relevant concentrations (56)
. Moreover, inhibitory effects of montelukast on nuclear factor-kβ activity (NF-
B) and VEGF expression has been reported (57
, 58)
. Therefore, it cannot be excluded that targets of montelukast other than CysLT1 may be involved in the tumor cell growth inhibition. Of particular interest for a potential future application of montelukast for neuroblastoma therapy is the fact that the drug is available for oral clinical treatment of pediatric asthma and the extraordinary tolerability shown in clinical use (59
, 60)
.
In comparison with nonmalignant nervous tissue, neuroblastomas contain increased levels of arachidonic acid, the main substrate for eicosanoid biosynthesis catalyzed by COX-2 and 5-LO (4)
. Interestingly, cysteinyl leukotrienes and prostaglandins are transported out of producing cells by MRP1 and MRP4, respectively. In addition to inhibiting prostaglandin synthesis, some nonsteroidal anti-inflammatory drugs may inhibit MRP4 directly (61)
. Similarly, MRP1 can be inhibited by leukotriene receptor antagonists (62)
. Expression of both MRP1 and MRP4 is correlated to MYCN expression and malignancy in neuroblastoma and confers resistance to chemotherapy (63
, 64)
. Hence, the leukotriene pathway inhibitors may potentiate the effect of chemotherapeutic agents.
Our results demonstrate for the first time that a fully active and functional leukotriene synthesis pathway is present in childhood neuroblastoma and that neuroblastoma cells produce leukotrienes that can promote survival in an autocrine manner. Inhibition of the leukotriene pathway also induced apoptosis of neuroblastoma cells in vitro. Our findings provide new insight into the pathobiology of neuroblastoma. Pharmacological interventions that target the leukotriene signaling pathway may be an important adjuvant therapy for children with neuroblastoma, but further preclinical in vivo studies are warranted.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication December 4, 2007. Accepted for publication May 15, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-glutamyl transpeptidase, DP: dipeptidase).
