I-kappa-kinase-2 (IKK-2) inhibition potentiates vincristine cytotoxicity in non-Hodgkin's lymphoma
© Al-Katib et al. 2010
Received: 4 January 2010
Accepted: 1 September 2010
Published: 1 September 2010
IKK-2 is an important regulator of the nuclear factor-κB (NF-κB) which has been implicated in survival, proliferation and apoptosis resistance of lymphoma cells. In this study, we investigated whether inhibition of IKK-2 impacts cell growth or cytotoxicity of selected conventional chemotherapeutic agents in non-Hodgkin's lymphoma.
Two established model systems were used; Follicular (WSU-FSCCL) and Diffuse Large Cell (WSU-DLCL2) Lymphoma, both of which constitutively express p-IκB. A novel, selective small molecule inhibitor of IKK-2, ML120B (N-[6-chloro-7-methoxy-9H-β-carbolin-8-yl]-2-methylnicotinamide) was used to perturb NF-κB in lymphoma cells. The growth inhibitory effect of ML120B (M) alone and in combination with cyclophosphamide monohydrate (C), doxorubicin (H) or vincristine (V) was evaluated in vitro using short-term culture assay. We also determined efficacy of the combination in vivo using the SCID mouse xenografts.
ML120B down-regulated p-IκBα protein expression in a concentration dependent manner, caused growth inhibition, increased G0/G1 cells, but did not induce apoptosis. There was no significant enhancement of cell kill in the M/C or M/H combination. However, there was strong synergy in the M/V combination where the vincristine concentration can be lowered by a hundred fold in the combination for comparable G2/M arrest and apoptosis. ML120B prevented vincristine-induced nuclear translocation of p65 subunit of NF-κB. In vivo, ML120B was effective by itself and enhanced CHOP anti-tumor activity significantly (P = 0.001) in the WSU-DLCL2-SCID model but did not prevent CNS lymphoma in the WSU-FSCCL-SCID model.
For the first time, this study demonstrates that perturbation of IKK-2 by ML120B leads to synergistic enhancement of vincristine cytotoxicity in lymphoma. These results suggest that disruption of the NF-κB pathway is a useful adjunct to cytotoxic chemotherapy in lymphoma.
NHL is the fifth most common type of cancer in the US representing 4.5% of cancer cases. Since the early 1970's the incidence of NHL has doubled . It is a group of heterogeneous diseases resulting from malignant transformation of lymphocytes. Eighty-five percent of NHLs are B-cells that can be broadly classified as aggressive (50%) and indolent (40%). Diffuse Large B-cell NHL (DLBCL) is the most common subtype (30%) of all lymphomas and is the prototype of aggressive but curable NHL. Follicular lymphoma (FL) is the second most common subtype, representing 22% and is the most common indolent NHL [2, 3].
To date, there is no consensus concerning the best treatment algorithm, but combination chemotherapy has been the mainstay for treatment of NHL. Incorporation of the anti-CD20 monoclonal antibody, Rituximab, has led to improvements in overall survival [4, 5]. More than half of patients with DLBCL can be cured with combination of Rituximab (R) and cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP). Incorporating Rituximab into conventional chemotherapy for follicular lymphoma has lead to higher response rates and longer durations of remission compared with chemotherapy alone . The success of Rituximab suggests that additional targeted therapeutics might improve the efficacy of cytotoxic regimens.
Constitutively active NF-κB in lymphoma is known to induce resistance to intrinsic and extrinsic apoptosis pathways . NF-κB is a transcription factor comprised of homo- and heterodimers, p50/p105 (NF-κB1), p52/p100 (NF-κB2), c-Rel, RelB, and p65 (RelA) . Inhibitors of kappa B (IκBα, IκBβ and IκBε) contain ankyrin-like repeats that mediate sequestration of NF-κB in the cytosol . The interaction between IκBα and NF-κB is regulated by IκB kinase (IKK-1 and IKK-2). Phosphorylation of IκBα leads to its degradation and release of NF-κB. NF-κB is then able to translocate to the nucleus where it controls a number of molecules involved in vital cellular functions, such as proliferation, apoptosis, and resistance to chemotherapy [10–16].
Clinically, aberrant NF-κB activation has been linked to poor outcome in lymphomas [17, 18]. Therefore, these and other studies prompted us to investigate potential therapeutic effects of inhibiting components of the NF-κB activation pathway in our lymphoma models.
Small molecule inhibitors (SMI) are used to selectively target molecules involved in survival pathways. ML120B (N-[6-chloro-7-methoxy-9H-β-carbolin-8-yl]-2-methylnicotinamide) is a potent and selective inhibitor of IKK-2, acting through blockade of the ATP-binding site in the kinase. ML120B has been shown to inhibit tumor necrosis factor-α (TNF-α)-induced nuclear translocation of p65 subunit of NF-κB and block TNF-α-stimulated cytokine production in human fibroblast-like synovial cell cultures isolated from patients with rheumatoid arthritis . ML120B inhibits both baseline and TNF-α-induced NF-κB activation in multiple myeloma cells. It was also shown to inhibit the growth of multiple myeloma cells in vitro and in vivo SCID mouse models .
In this report, we show that ML120B inhibits the phosphorylation of IκBα, hinders the growth of lymphoma cell lines in a concentration- and time-dependent manner and reduces progression out of G0/G1 phase of the cell cycle. More importantly, ML120B has a synergistic interaction with vincristine, a common cytotoxic agent used in the treatment of hematological malignancies. Our studies suggest that IKK-2 inhibition has a therapeutic role in lymphoma when used alone or in combination with cytotoxic agents.
Inhibition of IKK-2 Leads to Growth Inhibition of Lymphoma Cell Lines
The WSU-FSCCL and WSU-DLCL2 cells exhibit constitutive NF-κB activation as shown by the baseline expression of the phosphorylated form of IκBα (p-IκBα). ML120B inhibited phosphorylated IκBα in a concentration-dependent manner within one hour incubation with these cells. At 20 μM, ML120B inhibited p-IκBα by 100% in WSU-FSCCL and 40% in WSU-DLCL2 compared to control (Figure 1C).
IKK-2 Inhibition in Combination with Conventional Chemotherapeutic Agents
ML120B: Vincristine Combination Induces G2/M Cell Cycle Arrest and Apoptosis
Mechanism of Interaction Between IKK-2 Inhibition and Vincristine
Antitumor Activity of ML120B in Lymphoma-bearing SCID Mice
In this study we show that inhibition of IKK-2 by a small molecule inhibitor, ML120B, enhances the cytotoxic effect of the microtubule inhibitor, vincristine in lymphoma cells. IKK-2 inhibition leads to sequestration of p65 in the cytosol and prevention of vincristine-induced nuclear translocation. It was previously shown that NF-κB activation is involved in vincristine resistance . This is believed to be due, at least in part, to the positive effect of NF-κB on cell cycle progression .
There are multiple approaches to target IKK-2/NF-κB pathway. Several specific IKK-2 inhibitors are under development (reviewed by Karin et. al ). These inhibitors have a wide range of IC50 in relationship to inhibiting IKK-2. For example, the IC50 of SPC-839, PS-1145, and BMS-345541 are 62 nM, 0.15 μM and 0.3 μM, respectively. ML120B inhibits IKK-2 at an IC50 of 62 nM. ML120B, in the nM range is highly specific to IKK-2, but is capable of inhibiting IKKε and other enzymes at an IC50 greater that 100 μM . Other novel SMIs, such as GS143 suppress IκBα ubiquitination, but not IκBα phosphorylation. Thus, inhibition of NF-κB activation is as complex as the activation pathway itself with multiple sites as targets for inhibition .
The present study makes several key observations regarding IKK-2 as a potential therapeutic target in lymphoma. First, we demonstrated that inhibition of IKK-2 by ML120B can cause growth inhibition in a concentration- and time-dependent manner. The cause of the growth inhibition was due to the increase of cells in G0/G1 phase of the cell cycle. Our results suggest that ML120B alone acts by blocking cell growth and not via apoptosis. Second, we demonstrated that ML120B can inhibit constitutive activation of NF-κB in indolent and aggressive lymphoma cell lines in a concentration dependent manner similar to what was observed in myeloma cells . These observations suggest a broad application of IKK-2 inhibition in lymphoid tumors.
Interestingly, our data shows that IKK-2 inhibition synergizes the cytotoxic effects of microtubule inhibitor, vincristine. This synergy was found at 1/100 the dose of vincristine alone required to induce comparable G2/M arrest and apoptosis (520 pM in combination with ML120B (40 μM) vs 50 nM when used alone, Figure 3A-C). Furthermore, our results suggest that the ML120B: vincristine combination induces cell cycle arrest followed by apoptosis out of G2/M. Vincristine is a microtubule depolymerizing agent. It was shown that depolymerization of microtubules activates NF-kB and induces NF-kB-dependent gene expression . Our data indicate that prevention of vincristine-induced nuclear translocation of p65 and activation of NF-κB is a major mechanism of synergy between IKK-2 inhibition and vincristine. This synergy is selective since we did not observe similar interaction between IKK-2 inhibition and cyclophosphamide or doxorubicin. Cell death induced by the ML120B: vincristine combination is through the apoptosis pathway since there was evidence for caspase 3 and PARP cleavage (Figure 6). Constitutive activation of NF-κB in lymphoma and consequent activation of downstream molecules like cIAP2 , p21 , and Bcl-2  increases the threshold for apoptosis. This cell survival mechanism is accentuated by exposure of cells to vincristine [24, 33]. IKK-2 inhibition, by sequestering NF-κB in the cytosol and consequent down regulation of pro-survival molecules, lowers the threshold of apoptosis in response to cytotoxic agents like vincristine (Figures 3, 4, 5).
In vivo, we showed that ML120B: CHOP combination was well tolerated by the animals and induced higher anti-tumor efficacy compared with each agent alone in our WSU-DLCL2-SCID model (Figure 8). We have previously shown that genistein (30 μM) sensitizes DLCL2 cells to CHOP . Bharti et al., have shown that curcumin, a natural inhibitor of NF-κB, may sensitize the cytotoxic effects of vincristine (50 μM) . Sanda et al., showed that IKK inhibition by ACHP (10 μM) led to growth inhibition of MM cells and potentiation of vincristine cytotoxicity .
In summary, our study shows the feasibility of inhibiting a constitutively active NF-κB pathway in lymphoma cells. Such inhibition is associated with therapeutically beneficial biological effects in vitro and in vivo. When used alone, ML120B elicited modest therapeutic gains. However, there was significant synergy with the microtubule inhibitor, vincristine. Our data indicate that approaches to NF-κB pathway inhibition are best used in combination with cytotoxic chemotherapy rather than single agents. The major future challenge is to develop a more effective IKK-2 inhibitor with lower cellular IC50 in order to make them more attractive clinically.
Materials and methods
Cell Culture and Reagents
The cell lines used in the study have been previously described; Follicular Lymphoma (WSU-FSCCL)  and Diffuse Large Cell Lymphoma (WSU-DLCL2) ; The WSU-FSCCL cell line has been karyotyped at least 4 times since our initial publication in 1993. The recent analysis in September of 2009 revealed the same chromosomal abnormalities as previously reported; 47,XY,+der(1)i(1)(q10)del(1)(q32),t(1;13)(p31;q12),del(6)(q21q27),t(8;11)(q24;q22),t(14;18)(q32;q21). The WSU-DLCL2 has been similarly karyotyped several times since its establishment in 1990. The cell line acquired an additional abnormality, add(8q24), that was detected for the first time in 1997. Since then the karyotype profile has remained stable with no further changes. The most recent karyotype in September of 2009 revealed: 48,XY,t(1;2)(p36;q37),der(3)t(3;7)(q13;p15),t(4;14)(q27;q32),+7,i(7)(p10),der(7)t(3;7)(q21;p11.2),+8,add(8)(q24),t(14;18)(q32;q21),del(15)(q26.1),del(16)(q22). Furthermore, fluorescent in situ hybridization (FISH) using LSI MYC dual color break-apart DNA probe (Vysis Inc.) revealed a deletion of the telomeric 3' region of CMYC gene most likely due to unbalanced translocation affecting the CMYC gene region. Cells were maintained in RPMI-1640 medium containing 10% heat-inactivated fetal bovine serum (FBS), 1% L-glutamine, 100 U/ml penicillin G and 100 μg/ml streptomycin and incubated at 37°C in a humidified incubator with 95%/5% CO2. Primary antibody specific for Actin was obtained from Santa Cruz Biotechnology, (Santa Cruz, CA). Primary antibodies specific for Caspase-3, Caspase-9, PARP, p-IκBα and IκBα were obtained from Cell Signaling, (Danvers, MA). G3PDH was obtained from Trevigen, Inc (Gaithersburg, MD). Protein concentrations were determined using the Micro BCA protein assay (Pierce Chemical Company, Rockford, IL). Cyclophosphamide monohydrate was obtained from Mead Johnson (Evansville, IN). Doxorubicin hydrochloride was obtained from Bedford Inc (SA, Australia). Vincristine was obtained from Pharma Inc. (Bloomington, MN). ML120B was synthesized by Millennium Pharmaceuticals, Inc (Cambridge, MA) and dissolved in DMSO. Concentration of DMSO in the final culture was 0.44%.
Western Blot Analysis
Proteins obtained from cell extracts were collected 24, 48, or 72 h after single or combination treatment with the IKK-2 inhibitor (ML120B) and vincristine in lysis buffer containing protease inhibitors. Cytosolic protein extracts were prepared from control and treated cells using Nuclear/Cytosolic Fractionation Kit according to manufacturer's protocol (BioVision, Mountain View, CA). All proteins were resolved using 12% SDS-PAGE and transferred to Hybond C-extra membranes (Amersham Life Science, Arlington Heights, IL). Membranes were blocked with 5% milk in Tris buffer saline containing 0.05% Tween 20 (TBST) for 1 h at 25°C and incubated overnight at 4°C with rabbit anti-caspase 9 (H-170, Santa Cruz), rabbt anti-caspase 8 (H-134, Santa Cruz), rabbit anti-PARP (#9542, Cell Signaling), mouse anti-caspase 3 (#9668, Cell Signaling) or rabbit anti-NF-κB (H-286, Santa Cruz) in 2% Bovine serum albumin in TBST (1:1000 dilutions in BSA-TBST). Following incubation, membranes were washed with TBST and incubated with corresponding horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA; 1:5000 dilution in 5% milk-TBST) for 1 h at 25°C and then washed before proteins were visualized using picoglow HRP substrate (Michigan Diagnostics, LLC, Royal Oak, MI).
Flow Cytometric Analysis of Cell Cycle and Apoptosis
Cell cycle analysis and sub G0/G1 DNA content were determined by flow cytometry using propidium iodide (PI) staining. Cells were grown in the presence or absence of ML120B or vincristine then centrifuged and washed. The cells were then fixed with 75% ice-cold ethanol overnight and stained with 50 μg of PI and analyzed. To determine DNA fragmentation (as indication of apoptosis) induced by treatment agents, we utilized standard terminal deoxynucleotidyl transferase of dUTP nick end labeling (TUNEL) assay and propidium iodide (PI) staining. The kit used in this method (ApoDirect In Situ DNA Fragmentation Assay Kit, BioVision, Mountain View, CA; Catalog #K402-50) utilizes terminal deoxynucleotidyl transferase (TdT) to catalyze incorporation of DUTP at the 3'-hydroxyl ends of the fragmented DNA. The fluorescein-labeled DNA was detected by flow cytometry (horizontal axis in Figure 4). PI staining was simultaneously used to separate cells into G0/G1, S, G2 M and sub-G0 compartments based on DNA content (vertical axis, Figure 4). The dual staining (dUTP and PI) allowed us to assign dUTP-positive cells to a cell cycle phase. In this method, it is accepted that dUTP-positive cells are considered apoptotic . To confirm induction of apoptosis, we stained WSU-FSCCL cells with 7-AAD as previously published from our laboratory . All flow cytometry analysis of cells was done on FACScan (Becton-Dickinson, San Jose, CA).
WSU-FSCCL cells, treated and untreated, were harvested, washed once with PBS and fixed for 10 min with 3.7% formaldehyde in PBS. All procedures were carried out at room temperature. Following fixation, cells were washed 3 times with PBS, blocked for 45 min with 0.5% BSA in PBS and then incubated for 3 hr in 200 μl PBS containing 0.1% saponin (PBS-S), 1 μg/ml each of two primary antibodies, mouse anti-human NF-κBp65 and rabbit anti-tubulin. After incubation with primary antibodies, cells were carefully washed 3 times with PBS-S and then resuspended in PBS-S containing 5% goat sera and 10 μg/ml each of two fluorescently-labeled secondary antibodies and DAPI (10 μg/ml) for nuclear staining, if used. Cells were incubated for 1 hour, washed X3 with PBS-S and then fixed for 1 min with 3.7% formaldehyde. Following the final fixation, cells were washed 3 times with PBS containing no saponin. Cell suspensions were mounted on 1% gelatin-coated slide, dried, sealed with coverslips and visualized using an Olympus BX40 microscope equipped with laser light and fluorescence filter cubes for UV, green and red fluorescence. Visual recordings were captured separately using an RT-Spot Color Camera (Diagnostic Instruments, Inc, Sterling Heights, MI) and merged using Super Spot software (Diagnostic Instruments) to complete the overlay and final pictures.
All primary antibodies were purchased from Cell Signaling Technologies (Danvers, MA). Slow Fade Light, DAPI and Alexa Fluor 488 (green) and Alexa Fluor 568 (red) fluorescently labeled secondary antibodies were purchased from Molecular Probes (a division of InVitrogen, Carlsbad, CA).
Establishment and Propagation of Xenografts
3-4 week old female ICR mice with severe combined immune deficiency (SCID) were purchased from Taconic Farms (Germantown, NY). Animals were housed in special protective environment and left to adapt for few days before beginning the experiments. To initiate the WSU-DLCL2-SCID xenografts, (5-10) × 106 WSU-DLCL2 cells in serum-free RPMI 1640 medium were injected subcutaneously (SC) in the flank areas of each animal. Palpable tumors were detected by clinical examination in about 2 weeks. When tumor weight reached 1000-1500 mg, animals were euthanized; tumors dissected out, placed in RPMI 1640 medium in sterile environment and minced into small fragments (20-30 mg each). To propagate the xenografts, tumor fragments were implanted SC, using a trocar, into flanks of 3-4 week old female ICR-SCID mice. Forty animals were implanted with WSU-DLCL2 tumors for the single agent (ML120B alone) experiment and forty for the combination study (CHOP plus ML120B). The WSU-FSCCL-SCID is a systemic model which is established by injecting 107 WSU-FSCCL cells in serum-free medium intravenously via tail vein of ICR-SCID mice. The growth pattern and assessment of response of this model to ML120B were the same as previously published from our laboratory .
Efficacy Trial Design
WSU-DLCL2 tumor-bearing animals were randomly assigned to control or one of 3 treatment dose/schedules of ML120B; 10 animals in each group. Therapy was started one week after tumor implantation. Group 1 received one dose of ML120B at 120 mg/kg. Group 2 received 60 mg/kg twice (every 12 hours). Group 3 received 60 mg/kg twice a day for 28 days. All treatments were given through oral gavage. ML120B compound was dissolved in 5% (hydroxypropyl) methylcellulose. Control group animals received vehicle alone. CHOP MTD in SCID mice was previously determined in our laboratory  for one injection (i.e.40 mg/kg, i.v. cyclophosphamide; 3.3 mg/kg,i.v. doxorubicin; 0.5 mg/kg,i.v. vincristine; and 0.2 mg/kg orally prednisone every day for 5 days). Animals were monitored 3 times per week for signs of toxicity, weight changes and tumor measurements. They were euthanized to avoid discomfort if the tumor burden reached ~2000 mg (approximately 10% of the body weight). All animal experiments were done according to protocols approved by the Animal Investigation Committee (AIC) of Wayne State University.
Statistical significance of drug-treated versus control measurements was determined by the student t-test. The interaction between ML120B and vincristine was analyzed using Calcusyn V2 software program to determine if the combinations were synergistic. Calcusyn is based on the Chou-Talalay method , which calculates a combinational index (CI) to indicate synergistic effects where CI < 0.9, is considered synergistic. Survival functions were estimated using the Kaplan-Meier method and compared by the log-rank test. P-values <0.05 were considered statistically significant.
All statistical analyses were evaluated using GraphPad Prism 4 (San Diego, CA).
Flow cytometry core: Eric Van Buren and Claire Lee. We thank Dr Ben Chen, Kristina Cole, Anton Scott Goustin, and Anshu Giri for their valuable input. Grant support: Leukemia and Lymphoma Society grant 6028-8 (R.M. Mohammad); National Institute of Health grant R01 CA109389. NIH/National Institute of General Medical Sciences (NIGMS) grant GM058905B (J. Dunbar - Initiative for Maximizing Student Diversity (IMSD) - program director). This work was funded in part by Millennium Pharmaceuticals, The Takeda Oncology Company.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.