Background
Tumor cells depend on the suppression of apoptosis to survive the multitude of stressors they encounter on a daily basis, such as genetic mutations, hypoxia, chemotherapy and radiation. Chemotherapy induces the intrinsic pathway of apoptosis that is tightly regulated by mitochondrial resident Bcl-2 family proteins whose structure is comprised of one to four highly conserved Bcl-2 homology (BH) domains [
1]. Within this family Bcl-2 proteins with multiple BH domains can be either anti-apoptotic (Bcl-2, Bcl-X
L, Bcl-w, Mcl-1, Bfl-1/A1) or pro-apoptotic (Bax, Bak) and a subset containing only the BH3 domain are all pro-apoptotic (Bim, Bid, Puma, Bad, Bik, etc.) [
2]. The BH3 domain forms an amphipathic alpha helix that fits snuggly into the hydrophobic groove of multidomain members, leading to the specific binding of certain BH3 proteins to select multidomain members. Tumor cells often increase the expression and/or activity of anti-apoptotic Bcl-2 members to bind to and sequester stress-activated BH3 proteins to prevent their binding to Bax and or Bak to inhibit apoptosis. For example, a cancer dependent on Bcl-2 for survival, such as chronic lymphocytic leukemia (CLL), has an abundance of Bcl-2 that actively sequesters the BH3 protein, Bim, making the cancer cell “primed to die” in the event that other cellular stressors activate additional BH3 proteins to overcome the Bcl-2 binding capacity within the cell [
3].
Knowledge of the unique primed state of cancer cells led to the development of novel classes of pro-apoptotic small molecules that disrupt critical Bcl-2 family protein:protein interactions to reinstate apoptosis (e.g., HA14-1 [
4], AT-101 [
5], ABT-737 [
6], ABT-263 [
7]). The most potent candidates designed to antagonize Bcl-2, Bcl-X
L and Bcl-w includes ABT-737 and its clinical homologue, ABT-263 (Navitoclax, Abbvie) [
6,
8]. ABT-737 mimics the BH3 protein Bad and targets the hydrophobic pocket of Bcl-2, Bcl-X
L and Bcl-w with low nanomolar affinity.
Navitoclax has been tested in clinical trials for adult cancers like CLL, AML and Small Cell Lung Carcinoma [
9,
10]. Navitoclax treatment in these trials caused an immediate drop in the patient’s peripheral platelet count, potentially putting them at increased risk for bleeding. This dose limiting toxicity of Navitoclax is caused by its Bcl-X
L -targeting properties as mature circulating platelets depend on Bcl-X
L for survival [
11,
12]. Recently, a more selective Bcl-2 antagonist also from Abbvie, ABT-199 (Venetoclax), became clinically available [
13]. When tested in a phase 1 trial for adult CLL, ABT-199 showed remarkable tumor killing effects, causing overwhelming tumor lysis due to its potent on target disruption of Bcl-2 and its binding partners [
13]. ABT-199 selectively targets the hydrophobic pocket of Bcl-2 and not Bcl-X
L, safeguarding against unacceptable platelet effects [
14]. Given these Bcl-2 select antagonists are well tolerated and performing well in adult cancer trials, they have high potential to translate to the pediatric setting to treat Bcl-2 dependent tumors. Here we provide evidence that this holds true in a highly lethal pediatric solid tumor known to have functional dependence on Bcl-2, neuroblastoma.
Patients with high-risk neuroblastoma (HR NB) initially respond to treatment yet > 50 % still die of disease due to the emergence of chemotherapy resistance [
15]. We and others have shown that chemotherapy induced apoptosis is inhibited in NB due to alterations in Bcl-2 family protein interactions at the mitochondria [
16]. Bcl-2 dependence in NB, like other cancers, is not due to an overabundance of Bcl-2 in the cell compared to other anti-apoptotic members, but due to the functional dependence on Bcl-2 to bind to and sequester activated Bim [
17]. Differences in post-translational modifications of Bim seem to play a part in determining its selectivity to Bcl-2 over Mcl-1 in HR NB [
18]. We have shown that ABT-737 in combination with cyclophosphamide, induces a complete durable tumor response in mice bearing large xenografts of even the most therapy resistant NB cell lines, including those with MYCN amplification and ALK mutations [
19]. Functional characterization of Bcl-2 dependence patterns by co-immunoprecipitation (co-IP) confirmed that ABT-737 sensitive NB cell lines depend on Bcl-2 and not on Bcl-X
L to sequester Bim for survival, making a large subset of NBs prime candidates for select Bcl-2 antagonism [
19].
Thus our work and others supports a role for the functional identification and targeting of Bcl-2 mediated resistance mechanisms in HR NB. We have tested relapsed NB cell lines as well as primary human high-risk NB tumors and validated that similar Bcl-2 dependence patterns exist in relapsed as well as primary tumors [
19]. In order to facilitate a rapid translation of ABT-199 into clinical trials for patients with high-risk NB, we demonstrate with these studies that ABT-199 is just as potent pre-clinically as the Bcl-2/ Bcl-X
L/Bcl-w antagonists that preceded it.
Methods
Cell lines
Neuroblastoma cell lines with MYCN amplification
(IMR-5, SMS-SAN and NB-1643) were used (courtesy of COG Cell Line Repository and Michael D. Hogarty, MD, Laboratory, CHOP). Neural cells were grown in RPMI-1640 (Life Technologies) supplemented with 10 % fetal bovine serum, 2 mM L-Glutamine, 100 U/mL of penicillin. Tissue culture was at 37 °C in a humidified atmosphere of 5 % CO
2. All cell lines were validated as unique using STR based genotyping and confirmed using the COG cell line STR genotype database (
www.cog.org).
Cell viability assays
2 × 104 NB cells/well were plated in triplicate in 96-well plates in phenol red-free media and allowed to adhere for 24 h. Cells were then treated with ABT-199, doxorubicin, melphalan or vehicle controls. After 48 h, WST-1 reagent (Roche) was added and absorbance at 590 nM was recorded and normalized to the background. Error bars represent the average of the three technical triplicates and are representative of results seen in two separate biologic experiments.
PARP cleavage
Cells were treated with the indicated ABT-199 doses for 24 h. Cells were lysed with CHAPS buffer (10 mmol/L HEPES, 150 mmol/L NaCl, 2 % CHAPS [Sigma Aldrich, MO]) as previously described [
19]. Protein from cell lysate (25 μg per well) was loaded and an anti-PARP antibody (Cell Signaling, MA; #9542) was used to detect PARP cleavage.
Caspase 3/7 assay
2 × 104 cells/well were plated in a 96-well plate and treated with the appropriate dose of ABT-199 or Etoposide (Sigma Aldrich, MO) for 24 h. Caspase-Glo 3/7 Assay kit (Promega, WI; G8090) was used to quantify caspase activity according to the manufacturer’s specifications. Error bars represent the average of three separate biologic experiments.
Annexin/PI apoptosis assay
NB cells were treated with 5 nM ABT-199 or 10 μM etoposide for 24 h. Following the incubation, cells were harvested, washed twice with PBS and re-suspended in Binding Buffer (8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO 4•7H20, 0.24 g KH2PO4) at a concentration of 1 × 106 cells/mL. Cells were incubated with Annexin V-PE and 7-AAD (BD Biosciences, CA) for 15 min and then analyzed immediately by flow cytometry using the FACSCanto II flow cytometer (BD Biosciences, CA). Histograms represent percentage of cells found to be Annexin V positive. Error bars represent the average of two separate experiments.
Co-immunoprecipitation
Cells were lysed with CHAPS buffer with 300 μg of protein lysate added to antibody-matrix complex as previously described [
18]. Xenograft tumor sample were harvested and immediately dissociated though a 0.45 μm sterile filter, lysed, and immunoprecipitated as above. Antibodies used as previously described [
19].
Murine xenograft studies
Xenografts (IMR5 and NB-1643) were established subcutaneously in the flank of 4–6 week old nu/nu athymic mice (Jackson Laboratories, ME) as described [
19]. When tumor volume reached 150 to 200 mm
3, mice were treated either (i) by intraperitoneal (IP) injection of cyclophosphamide (CPM: 75 mg/kg) or saline solution IP twice weekly for 2 weeks, (ii) by oral gavage of ABT-199 (100 mg/kg/day for 2 weeks) or vehicle solution, (iii) by a combination of CPM with ABT-199 at monotherapy dosing/duration. ABT-199 was formulated in a mixture of 60 % Phosal 50 PG, 30 % PEG 400, and 10 % Ethanol. One animal from the combination treatment was retreated with the same therapy schedule when the recurrent tumor reached 200 mm
3. Tumor volumes were evaluated using calipers and calculations using the ellipsoid formula (Length × width
2 × 0.52). Animals were sacrificed when tumor volumes exceeded 2000 mm
3, according to the Emory Institutional Animal Care and Use Committee Guidelines. One mouse with recurrent tumor from the combination treatment was sacrificed for tumor co-immunoprecipitation studies. Mice were housed in a temperature- and light-controlled pathogen-free facility and given access to sterilized food and water ad libitum. All animal work abided by and was carried out under a protocol approved by the Emory Institutional Animal Care and Use Committee.
Statistical evaluation
Survival analysis were conducted according to the method of Kaplan-Meier and analyzed by a Mantel-Cox test. GraphPad Prism software was used to determine statistical significance between treatments and the IC50 in the cell survival analyses. P-values < 0.05 were considered statistically significant.
Discussion
We have previously shown that high-risk NB cell lines derived from human tumors with the poorest prognosis can be functionally dependent on Bcl-2 for survival through Bcl-2’s tonic sequestration of Bim [
19]. While the majority of NB cell lines and primary tumors express Bcl-X
L protein, Bcl-X
L fails to sequester Bim or other activator BH3 proteins to functionally repress programmed cell death. Given the absence of Bcl-X
L dependence, we hypothesized that select Bcl-2 antagonism would be equally as potent as ABT-737 in killing NB tumors and eventually result in less hematologic side effects when translated in the clinic. These studies confirmed the potency of select Bcl-2 antagonism in HR NB models.
Given ABT-199 disrupts Bim binding to Bcl-2 at low nanomolar concentrations in vitro, the superiority of the combination regimen over ABT-199 alone in vivo does not indicate that ABT-199 is less effective at inhibiting its target in vivo. Interestingly, the same in vivo experiments we performed with ABT-737 against NB xenografts also showed that ABT-737 was more potent in combination with CPM compared to ABT-737 when given alone. The second remission noted in the recurrent xenograft retreated with ABT-199/CPM suggests that a longer treatment duration (>14 days) with ABT-199 either alone or in combination may reveal the true potency of this drug for the clinic. Further mechanistic studies are warranted to understand the contribution a cytotoxic like cyclophosphamide has in augmenting the impressive tumor regression seen in the in situ setting.
We cannot discount the potent effects of cytotoxic chemotherapy in curing the largest portion of patients with HR NB and other solid tumors over the years, frankly at a far higher rate than any targeted therapy has thus far. Thus the results of these combination studies makes an argument that novel pro-apoptotic agents like ABT-199 should not be used as a substitute for standard cytotoxics but as an addition to chemotherapy, to permit the lowering of cytotoxic dosing to decrease normal cell toxicity while enhancing cytotoxic induced tumor death by reinstating functional apoptotic machinery to the tumor. Given the platelet effects of Navitoclax are not seen in patients clinically treated with ABT-199 supports that this agent has a higher potential to be used in combination with cytotoxic drugs in patients with multiply recurrent NB, who have heavily pretreated marrows with mild thrombocytopenia at baseline.
Like many tumors, Bcl-2 dependence in neuroblastoma cannot be determined by expression-based methods alone but by a functional determination of Bim:anti-apoptotic Bcl-2 protein binding patterns. Functional assays such as mitochondrial BH3 profiling, co-IP and more recently dynamic BH3 profiling can be performed on very small amounts of solid tumor tissue to successfully characterize Bcl-2 dependence patterns leading to apoptosis resistance in chemotherapy refractory solid tumors like NB and others [
17,
19,
22]. Further validation of such assays and honing them to be performed in the clinical setting using limited numbers of cells from tumor samples will improve our ability to match patients to the most effective pro-apoptotic therapies like ABT-199 and will further pave the way for using small molecule Bcl-2 antagonists in the clinic for those pediatric NB patients most likely to respond.
Competing interests
The authors report no financial or non-financial conflicts of interest related to this work.
Authors’ contributions
RT carried out the xenograft studies, confirmed co-immunoprecipitations (CoIP) results with technical replicates, performed apoptosis assays as well as statistical analysis, and drafted the manuscript. DK participated by running the in vitro drug combination assays. SN participated in the co-IPs as well as monotherapy ABT-199 and ABT-263 experiments. KG conceived the study, participated in the design, coordination, and data analysis as well as drafted and finalized the manuscript. All authors read, edited and approved the final manuscript prior to submission.