GX15-070

A small molecule pan-Bcl-2 family inhibitor, GX15-070, induces apoptosis and enhances cisplatin-induced apoptosis in non-small cell lung cancer cells

Jiannong Li · Jean Viallet · Eric B. Haura

Received: 22 November 2006 / Accepted: 8 April 2007 / Published online: 16 May 2007 © Springer-Verlag 2007

Abstract
Purpose Overexpression of Bcl-2 family members as well as deregulated apoptosis pathways are known hall- marks of lung cancer. Non-small cell lung cancer (NSCLC) cells are typically resistant to cytotoxic chemotherapy and approaches that alter the balance between pro-survival and pro-death Bcl-2 family members have shown promise in preclinical models of NSCLC.
Methods Here we evaluated the eVects of a novel pan- Bcl-2 inhibitor GX15-070 on NSCLC survival and when combined with epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors as well as traditional cytotoxic agents. GX15-070 is a small molecule agent that binds anti- apoptotic Bcl-2 proteins and interferes with their ability to interact with pro-apoptotic proteins. We evaluated the eVect of GX15-070 and correlated the eVect on EGFR status as well as Bcl-2 family protein expression.
Results We show that GX15-070 can disrupt Mcl-1:Bak interactions in lung cancer cells. We identiWed diVerential sensitivity of a panel of lung cancer cells to GX15-070 and no clear relationship existed between EGFR status or Bcl-2 family protein expression and sensitivity to GX15-070.
GX15-070 could induce apoptosis in a subset of lung can- cer cell lines and this correlated with the eVects on cell via- bility. GX15-070 combined with geWtinib was synergistic in a cell line dependent on EGFR for survival but GX15- 070 could not reverse resistance to geWtinib in cell lines not dependent on EGFR for survival. Finally, we observed syn- ergy between GX15-070 and cisplatin in NSCLC cells. Conclusions Based on these results, GX15-070 can trig- ger apoptosis in NSCLC cells and can enhance chemother- apy-induced death. These data suggest that clinical trials with GX15-070 in combination with cytotoxic chemother- apy are indicated.

Keywords Mcl-1 · Bcl-2 · GX15-070 · Lung cancer ·
Apoptosis · Chemotherapy

Introduction

Overexpression of anti-apoptotic Bcl-2 family members and deregulation of pathways that regulate pro-apoptotic family members have been observed in lung cancers [1]. It is now the general consensus that the relative ratio of pro-apoptotic to anti-apoptotic family members ultimately dictate the

J. Li · E. B. Haura (&)
Thoracic Oncology and Experimental Therapeutics Program,
H. Lee MoYtt Cancer Center and Research Institute, MRC3 East, Room 3056, 12902 Magnolia Drive, Tampa, FL 33612-9497, USA e-mail: [email protected]

J. Li · E. B. Haura
Department of Interdisciplinary Oncology, University of South Florida College of Medicine, Tampa, FL 33612, USA

J. Viallet
GeminX, Inc, Malvern, PA 19355, USA
cell’s fate toward survival or death [2, 3]. The Bcl-2 family consists of three main subfamilies including Bcl-2, Bcl-xL, and Mcl-1 which inhibit apoptosis, the Bax subfamily con- sisting of Bax, Bak, and Bok that promote apoptosis, and the BH3-only subfamily consisting of Bid, Bim, Bad, and others that also promote apoptosis [4, 5]. Elevated levels of Bcl-xL and to some extent Bcl-2 have been observed in non-small cell lung cancer (NSCLC) [6–18]. Cellular models have demonstrated a role of Bcl-2 proteins in maintaining lung cancer survival and control of apoptosis in response to cyto- toxic chemotherapy [19–24]. Inhibition of Bcl-2 or Bcl-xL

with antisense oligonucleotides can trigger apoptosis in lung

cancer cells [19–24]. Similarly, enforced Bcl-2 overexpres- sion can prevent apoptosis resulting from inhibition of the epidermal growth factor receptor (EGFR) [25]. In addition to Bcl-2 and Bcl-xL, Mcl-1 is overexpressed in NSCLC and knockdown of Mcl-1 levels result in apoptosis of lung can- cer cells and sensitizes lung cancer cells to apoptosis induced by cytotoxic agents [8, 26].
A

B
GX (uM)

0

1

5 15

Because apoptosis is deregulated in lung cancer as well as in many cancers, apoptosis targets such as Bcl-2 proteins are being explored for identiWcation of new cancer drugs that can be used in clinic [27]. For example, development of antisense oligonucleotide therapy directed against Bcl-2 has been evaluated as therapy for patients with advanced lung
IP: Mcl-1 IB: Mcl-1

IP: Mcl-1 IB: Bak
Heavy chain Mcl-1

Bak
Light chain

cancer, as well as other tumor types [28–31]. Clinical stud- ies using antisense oligonucleotides directed against Bcl-2 have been reported in small cell lung cancer and a current trial in NSCLC is comparing single agent docetaxel to docetaxel plus antisense Bcl-2 [32]. An alternative approach to antisense strategies is to develop small molecule inhibitors of pro-survival Bcl-2 family inhibitors [33]. One promising early compound is GX15-070 (2-[5[(3,5-Dimethyl-1H-pyr- rol-2-ylmethylene)-4-methoxy-5H-pyrrol-2-yl]-1H-indole), an indole-derivative novel small molecule inhibitor of Bcl-2 proteins (Fig. 1a). GX15-070 binds anti-apoptotic Bcl-2 proteins and interferes with their ability to interact with and negatively regulate pro-apoptotic proteins through the BH3 domain. This disruption of binding allows the pro-apoptotic Bcl-2 proteins induce caspase-dependent apoptosis in cancer cells. GX15-070 can inhibit the interaction between recom- binant Bcl-2 and Bax as well as the interaction between Mcl-1 and Bak both in vitro and in vivo. Finally, GX15-070 can lead to phosphorylation of the pro-apoptotic BH3-only protein Bim and can induce Bim translocation to the mito- chrondria.
Based on the importance of Bcl-2 proteins in lung cancer, we evaluated the eVect of GX15-070 in human NSCLC cells. We used lung cancer cell lines with known EGFR mutational status and sensitivity to EGFR tyrosine kinase inhibitors (TKI). The level of pro-survival and pro-death Bcl-2 proteins were examined in these cell lines and the eVect of GX15-070 was evaluated in regard to Bcl-2 family member expression level and ability of GX15-070 to induce apoptosis. GX15- 070 in combination with either EGFR TKI or cytotoxic che- motherapy was examined for its eVect on NSCLC cells.

Materials and methods Cell lines and reagents
Human lung cancer cell lines were purchased from ATCC and maintained in RPMI-1640 medium supplemented with
Fig. 1 EVect of GX15-070 on Mcl-1:Bak protein–protein interac- tions. a Chemical structure of GX15-070 b PC9 cells were exposed to indicated concentrations of GX15-070 for 6 h before being harvested for total proteins. Protein lysates were immunoprecipitated with anti- Mcl-1 antibody as described and immunoprecipitate was immunoblot- ted with antibodies that recognize Mcl-1 and Bak. Heavy chain and light chains indicate contaminating IgG from anti-Mcl-1 antibody used for immunoprecipitation. GX GX15-070

10% bovine calf serum (BCS, Gibco). H3255 cells were provided by Dr Pasi Janne (Dana Farber, Boston, MA, USA) and grown in ACL-4 media [34]. HCC827, HCC2279, and H4006 cells were a gift of Dr Jon Kurie (MD Anderson, Houston, TX, USA) [35]. GeWtinib-resistant H1650 cells were provided by Dr JeVrey Settleman (Har- vard, Boston, MA, USA) [36]. Stock solutions of geWtinib, paclitaxel, cisplatin, and GX15-070 in 100% DMSO and gemcitabine in sterile PBS were diluted directly into the media to indicated concentrations. GeWtinib was provided by Astra-Zeneca (Wilmington, DE, USA) and GX15-070 by GeminX, Inc. (Malvern, PA, USA). Paclitaxel and cis- platin were purchased from Sigma (St Louis, MO, USA) and gemcitabine was provided by Dr Doug Cress (MoYtt Cancer Center, Tampa, FL, USA).

Immunoprecipitation assays

Whole cell lysates were prepared using an ice-cold IP buVer (10 mM Tris, pH 8.0, 60 mM KCl, 1 mM EDTA, 1 mM DTT, 0.5% NP-40, 10 mM Na3VO4, 50 mM NaF, 1 mM PMSF, 1 ti g/ml aprotinin, 1 ti g/ml leupeptin, 1 ti g/ml pepstatin. 300 ti g of total protein was incubated with 1.5 ti g of rabbit Mcl-1 antibody (Santa Cruz, CA, USA) overnight at 4°C with rotating, and then incubated with Protein A-agarose beads (Roche, Indianapolis, IN, USA) for 4 h at 4°C. Beads were washed three times with IP buVer contain- ing protein inhibitors and immunoprecipitated proteins
were eluted from beads with 30 ti l of 2£ SDS-polyacryl- amide gel electrophoresis (PAGE) sample buVer from BIO- RAD (Hercules, CA, USA) at 90°C for 3 min. Proteins

were separated on 12% SDS-PAGE gels and immunoblots were performed on membranes incubated overnight with primary antibody mouse anti-Mcl-1 (Pharmingen, San Diego, CA, USA) and rabbit anti-Bak (Upstate, Lake Placid, NY, USA). Detection of signals was performed using the corresponding secondary HRP-conjugated anti- body and enhanced chemiluminescence (ECL) purchased through Amersham (Piscataway, NJ, USA).

Cytotoxicity and apoptosis assays

Cytotoxicity assays (MTT) were performed according the manufacturers recommendations of Cell Proliferation Kit
from Roche (Indianapolis, IN, USA). 5 £ 103 cells in 5% BCS complete media were placed into single wells in a 96- well plate from FALCON (Franklin Lakes, NJ, USA), exposed to indicate agents, and viability assessed following 72 h. The IC50 was deWned as the drug concentration that induced a 50% reduction in cell viability in comparison with DMSO controls and was calculated by non-linear regression analyses. Data presented represents three sepa- rate experiments with eight data points per condition. Apoptosis was assayed using cleaved PARP antibody From Cell Signaling (Beverly, MA, USA).
For combination cell viability assays, we Wrst used MTT to identify relevant and eVective concentrations of inhibi- tors and cytotoxic agents. For Wnal data analysis and report- ing, viable cells were identiWed and counted using trypan
blue exclusion. 5 £ 104 cells in 5% BCS complete media were placed into single wells in a 24-well plate, exposed to indicated agents, and viable and non-viable cells were counted using trypan blue solution (Sigma, St Louis, MO, USA) following 72 h. Analysis of synergism between diVerent agents in inducing apoptosis of cells was per- formed by median dose–eVect analysis and calculation of combination indices (CI) using commercially available software (Calcusyn; Biosoft, Ferguson, MO, USA) [38]. According the recommendations of this methodology, the value of CI represents as following: 0.9–1.10 is mildly additive; 0.85–0.9 is slight synergism; 0.7–0.85 is moderate synergism; 0.3–0.7 is synergism; 0.1–0.3 is strong syner- gism; less than 0.1 is very strong synergism.

Protein expression analysis

Cell lysates were prepared using RIPA buVer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM NaF, 2 mM Na3VO4, 0.25% sodium deoxycholate, 1% NP-40, 1 mM PMSF, 60 ti g/ml aprotinin, 10 ti g/ml leupeptin, 1 ti g/ml pepstatin), were normalized for total protein content (50 ti g) and subjected to SDS-PAGE.
Primary antibodies used in these studies consisted of Bcl-2, Bcl-xL, Bim, and Mcl-1 from Santa Cruz (Santa

Cruz, CA, USA), cleaved PARP, Bax and Bak from Cell Signaling (Beverly, MA, USA) and ti -actin from Sigma (St Louis, MO, USA). The cleaved PARP (Asp214) antibody detects endogenous levels of the large fragment (89 kDa) of human PARP1 produced by caspase cleavage. The anti- body does not recognize full length PARP1 or other PARP isoforms. Detection of proteins was accomplished using horseradish-peroxidase conjugated secondary antibodies and ECL purchased through Amersham (Piscataway, NJ, USA) [37].

Results

GX15-070 disrupts Mcl-1:Bak complexes in vivo

To assess the eVect of GX15-070 on pro-survival and pro- death Bcl-2 family protein complexes, we evaluated the eVect on Mcl-1:Bak complexes using immunoprecipitation assays. We chose to focus on Mcl-1:Bak complexes since Bcl-2 is rarely expressed in NSCLC cells and our previous results as well as results of others suggest the importance of Mcl-1 in the survival of epithelial tumors [26]. PC9 cells were exposed to increasing concentrations of GX15-070 for a total of 6 h and then total proteins harvested for immuno- precipitation. Lysates were immunoprecipitated with a Mcl-1 antibody and pulldowns evaluated for total Mcl-1 as well as co-immunoprecipitated Bak. These results are shown in Fig. 1b and demonstrate the ability of GX15-070 to disrupt Mcl-1:Bak complexes in cells in a concentration- dependent manner. Because re-association of Mcl-1 and Bak proteins into complexes may result secondary to dilu- tion of GX15-070 during the process of creating protein lystates and immunoprecipitation, our results may underes- timate the concentrations required to disrupt the complex in whole cells.

EVect of GX15-070 on cell proliferation and apoptosis in NSCLC cell lines

To assess the eVects of GX15-070 on NSCLC cell lines, cell lines with distinct EGFR status and sensitivity to EGFR TKI were exposed to increasing concentrations of GX15- 070 and cell viability was assayed. We chose to stratify cell lines based on the EGFR status (wildtype vs. mutant) as well as their sensitivity to EGFR tyrosine kinase inhibitors [39–41]. These data are shown in Fig. 2 and the IC50 sum- marized in Table 1. These lung cancer cell lines demon- strated diVerential sensitivity to GX15-070 and there was no clear relationship between EGFR status and sensitivity to GX15-070. In cells with wildtype EGFR, A549 cells were the most sensitive to GX15-070 with an approximate IC50 of 1.3 ti M (Fig. 2a). Of the lung cancer cells lines that

A100 90

Table 1 Activity of GX15-070 in human lung cancer cell lines

80
70
60
50
40
30
20
10
0

A549
H460
H1299
H358

0 0.039 0.078 0.156 0.312 0.625 1.25 2.5 5 10 20
GX (µM)
Cell lines

A549
H460
H1299
H358
PC9 HCC827 H4006 H1650
EGFR status

WT
WT
WT
WT
Mutant, geWtinib-sensitive Mutant, geWtinib-sensitive Mutant, geWtinib-sensitive Mutant, geWtinib-sensitive
IC50 (ti M)

1.33 § 0.47 3.85 § 1.60 7.62 § 2.80 15.4 § 2.09 0.26 § 0.12 1.83 § 0.39 2.89 § 0.75 5.99 § 1.55

B
100
90
80
70
60
50
40
30
20
10
0
H3255 Mutant, geWtinib-sensitive 4.77 § 1.52
H1975 Mutant, geWtinib-resistant 3.04 § 1.27
H1650-R Mutant, geWtinib-resistant 7.78 § 2.10
HCC2279 Mutant, geWtinib-resistant 7.32 § 1.90 Mean § SD (N = 3) IC50 values for GX15-070 in the evaluated cell
lines along with EGFR status are listed

to EGFR TKI. This includes the H1975 cell with a L858R + T790M mutation, H1650 cells that were engi- neered to be resistant to EGFR TKI, and HCC2279 cells

0 0.039 0.078 0.156 0.312 0.625 1.25 2.5 5 10 20
GX (µM)
that have an activating EGFR mutation but have reduced ErbB-3 and HER ligand expression [35, 36, 42]. As shown

C 100 90 80 70 60 50 40 30 20 10 0

H1975 H1650-R HCC2279

0 0.039 0.078 0.156 0.312 0.625 1.25 2.5 5 10 20
GX (µM)
in Fig. 2c, these cells have IC50 of greater than 1 ti M. Because GX15-070 prevents Bcl-2:Bax and Mcl-1:Bak
interactions, we next examined the expression of both pro-survival and pro-death Bcl-2 proteins using western analysis to determine if a relationship exists between target protein expression and GX15-0970 sensitivity (Fig. 3). We included Bcl-2, Mcl-1, Bcl-xL, Bim, Bax, and Bak in our analysis. Consistent with prior reports, Bcl-2 expression is relatively rare in NSCLC cells with high levels of expres- sion only found in H460 and H1299 cells with no obvious expression in other cells including the cells with activating EGFR mutations. Consistent with our previous studies, the

Fig. 2 EVect of GX15-070 on cell viability in lung cancer cell lines. Indicated lung cancer cells were exposed to increasing concentrations of GX15-070 and cell viability assessed after 72 h. Each bar represents the standard deviation of three separate experiments. Cell survival was normalized to DMSO-treated control cells. a Wildtype EGFR lung cancer cells (A549, H460, H1299, H358), b mutant EGFR and geWti- nib-sensitive lung cancer cells (PC9, HCC827, H4006, H1650, and H3255), and c mutant EGFR and geWtinib-resistant lung cancer cells (H1975, H1650-R, and HCC2279)

show strong dependence on EGFR for survival, the PC9 cell with the deletion EGFR mutation was very sensitive to GX15-070 with an approximate IC50 of 0.25 ti M (Fig. 2b). All the other cells that have been reported either by us or others to be dependent on EGFR for survival demonstrated IC50 > 1 ti M for GX15-070. Finally, we also explored cells that have mutations in EGFR but are nonetheless insensitive
majority of the lung cancer cell lines expressed Mcl-1 pro- tein levels albeit to diVerent levels and this did not correlate with EGFR mutation status [26]. Similarly, Bcl-xL was expressed in the majority of these NSCLC cell lines and again no obvious relationship existed between EGFR muta- tion status and protein expression. Expression of Bax was identiWed to diVerent extents in all the cell lines evaluated and Bak expression was seen in most cells but not all. When the expression level of these pro-survival and pro- death Bcl-2 proteins was examined in relation to the sensi- tivity of the cells to GX15-070, we could not identify an obvious relationship between protein expression and drug sensitivity.
We next examined the eVect of GX15-070 on lung cancer cell apoptosis. These data are shown in Fig. 4. In order to assess whether GX15-070 can induce apoptosis directly in these lung cancer cells, we evaluated cleavage of PARP as a

EGFR-WT
Cell lines A549 H460 H1299 H358 Bcl-2

Mcl-1

Bcl-xL

Bim

Bax

Bak

Actin

EGFR-Mutant
PC9 HCC827 H4006 H1975 H1650

PARP cleavage starting at 48 h and increasing at 72 h. Simi- lar eVects were seen in H460 cells where little PARP cleav- age was identiWed after 24 h (data not shown) but was clearly evident by 48 h (Fig. 4b). In contrast, H1299 and H358 cells demonstrated little to no PARP cleavage above baseline in untreated cells and this observation corresponds to the higher IC50 identiWed in the viability assays in Fig. 2 and Table 1. We next examined apoptosis using PARP cleavage in the cells with EGFR mutation. As shown in Fig. 4c, PC9 cells demonstrate PARP cleavage as early as 24 h and at a concentration of 0.5 tiM. Of the other cells that have demonstrated dependence on EGFR for survival, we only identiWed PARP cleavage at a 1 ti M concentration of GX15-070 in the HCC827 cells but not H4006, H3255, or H1650 (Fig. 4d). We again noted the correlation with PARP cleavage with the IC50 since PC9 and HCC827 cells have lower IC50 that corresponds to the ability of GX15-070 to induce PARP cleavage. Finally, we also evaluated the apop- totic eVect of GX15-070 on H1975 cells. Here we observed

Fig. 3 Bcl-2 family protein expression in lung cancer cell lines. Total protein was collected from lung cancer cells growing in serum and Bcl- 2, Mcl-1, Bcl-xL, Bim, Bax, and Bak protein levels evaluated using western analysis. Equal protein loading was conWrmed by evaluation of actin
PARP cleavage at a concentration of 1 ti M. From these stud- ies, we conclude that GX15-070 can induce apoptosis in a subset of lung cancer cell lines and the induction of apopto- sis based on PARP cleavage correlates with the IC50 identi- Wed in cell viability assays.

biochemical marker of apoptosis and evaluated the eVect of

GX15-070 exposure time on PARP cleavage. As shown in Fig. 4a, a 1 tiM concentration of GX15-070 induces a time- dependent eVect on PARP cleavage in A549 cells. While no PARP cleavage is evident at 24 h, we identiWed the onset of
EVect of EGFR TKI and GX15-070 on NSCLC survival We next examined the eVect of combined EGFR tyrosine
kinase inhibition and GX15-070 on lung cancer cell survival.

Fig. 4 EVect of GX15-070 on biochemical apoptosis. Lung

A

Cell line Time (hrs)

24

A549
48

72

D

cancer cells were exposed to GX (µM) 0 1 0 1 0 1
Cell line

GX15-070 at diVerent concen- trations and for indicated times. The cells were collected and to- tal protein harvested. PARP cleavage was evaluated using western analysis and equal pro- tein loading conWrmed by evalu-

Cleaved PARP

Actin

BCell line

GX (µM) 0 1 2 5

HCC827
GX (µM) 0 1 2 5
Cleaved PARP

Actin

Cleaved PARP

ation of actin. a A549 cells were exposed to either control solvent or GX15-070 for indicated times and cleaved PARP assayed.
b Wildtype EGFR lung cancer cells were assayed for cleaved PARP after 48 h. c PC9 cells were exposed to indicated con- centrations of GX15-070 for
24 h and cleaved PARP assayed. ZD ZD1839 (geWtinib) was used as a positive control for apopto- sis in these mutant EGFR cells. d PARP cleavage following 48 h exposure to GX15-070 in mutant EGFR lung cancer cells

C
Cleaved PARP
H460
Actin

Cleaved PARP
H1299
Actin

Cleaved PARP
H358
Actin

Cell line PC9
GX (µM) 0 0.1 0.5 1 2 5 ZD 1µM
Cleaved PARP

Actin
H4006

H3255

H1650

H1975

Actin

Cleaved PARP

Actin

Cleaved PARP

Actin

Cleaved PARP

Actin

This was based in part on studies that demonstrated that Bcl-2 or Mcl-1 could inhibit death induced by geWtinib [25, 26]. These data are shown in Fig. 5. First, we examined if combination of geWtinib, an EGFR TKI, with GX15-070 could produce enhanced apoptosis in lung cancer cells with activating EGFR mutation and sensitivity to EGFR TKI. For these experiments, PC9 cells were exposed to GX15- 070, geWtinib, or the combination and the eVect on cell via- bility quantiWed. As shown in Fig. 5a, both GX15-070 and geWtinib result in a concentration-dependent increase in cell

death and the combined eVect of both agents results in enhanced cell death. Analysis of synergism between geWti- nib and GX15-070 in inducing cell death was performed by median dose–eVect analysis. Importantly, exposure to the combination of geWtinib and GX15-070 exerted synergistic apoptotic eVect in PC9 cells, as determined by the median dose–eVect analysis, which revealed combination index
values of <1.0 (0.77 § 0.12).
We next determined if GX15-070 could reverse resis- tance of lung cancer cells to geWtinib. These data are shown

A

70

60

50

40

30

20

10

0

0.25-2.5 0.5-5.0 1.0-10 1.5-15
Dose of GX (µM) and Gefitinib (nM)

B
70

60

50

40

30

20

10

0

70

60

50

40

30

20

10

0

1.25-0.25 2.5-0.5 5.0-1.0 10-2.0 1.25-0.25 2.5-0.5 5.0-1.0 10-2.0
Dose of GX and Gefitinib (µM) Dose of GX and Gefitinib (µM)

70
H460-GX+Gefitinib
70

GX
H358-GX+Gefitinib

60 Gefitinib 60

GX+Gefitinib
GX

50 50 Gefitinib
GX+Gefitinib

40

30

20

10

0
40

30

20

10

0

1.25-0.25 2.5-0.5 5.0-1.0 10-2.0 1.25-0.25 2.5-0.5 5.0-1.0 10-2.0
Dose of GX and Gefitinib (µM) Dose of GX and Gefitinib (µM)

Fig. 5 Combination eVect of GX15-070 and geWtinib in lung cancer cells. (a PC9 cells were exposed to increasing concentrations of GX15- 070, geWtinib, or the combination of both agents in a Wxed ratio. Viable cells were assayed by trypan blue exclusion and combination indices were calculated using Calcusyn software using percent of viable cells. GX GX15-070, GEF geWtinib. b A549, H460, H1299, and H358 cells

were exposed to increasing concentrations of GX15-070, geWtinib, or the combination of both agents in a Wxed ratio. Viable cells were as- sayed and combination indices (CI) were calculated using Calcusyn software as described. The values above each bar represent CI for that dose combination

in Fig. 5b and in Table 2. Lung cancer cell lines that do not have activating EGFR mutations were exposed to increasing concentrations of GX15-070, geWtinib, or the combination of both agents and the eVect on cell viability was assessed. Analysis of synergism between geWtinib and GX15-070 in inhibiting cell viability was performed by median dose– eVect analysis. In none of the cell studied did we observe the ability of GX15-070 to enhance the anti-proliferative eVect of geWtinib. From these studies we conclude that GX15-070 can synergistically enhance apoptosis in lung cancer cells dependent on EGFR for survival but cannot reverse geWti- nib-resistance in lung cancer cells with wildtype EGFR.

EVect of cytotoxic chemotherapy and GX15-070 on NSCLC survival

We next examined the eVect of GX15-070 on cytotoxic chemotherapy-induced cell death. Lung cancer cell lines that do not have activating EGFR mutations were exposed to increasing concentrations of GX15-070, cytotoxic che- motherapy agents commonly used in clinic for the treat- ment of advanced lung cancer (cisplatin, gemcitabine, paclitaxel), or the combination of both agents. MTT was used initially to identify dose ranges for further study. The eVect on cellular viability was examined by counting viable cells identiWed with trypan blue staining and analysis of synergism between cytotoxics and GX15-070 was per- formed by median dose–eVect analysis. These data are shown in Fig. 6 and are summarized in Table 3. For A549 cells, combination of cisplatin and GX15-070 resulted in evidence of synergism since all CI were below 1.0
(0.677 § 0.186) while combination of GX15-070 and either gemcitabine or paclitaxel had CI that averaged around 1.0. Similarly, in H460 cells, cisplatin and GX15- 070 led to synergistic cell kill with all CI below 1.0
(0.657 § 0.139). H1299 cells however demonstrated little eVect of combined GX15-070 and cytotoxics and no evi- dence of synergy on cell viability was identiWed.

Discussion

One approach to improving the outcome of advanced NSCLC is to alter the balance of pro-apoptotic to anti-apoptotic

mechanisms and therefore sensitize cancer cells to apop- tosis triggered either endogenously through oncogenic signals that trigger apoptosis or through exogenous stim- uli such as exposure to cytotoxic chemotherapy. A previ- ous study demonstrated that another small molecule Bcl-2 inhibitor (ABT-737) could induce regressions in solid tumors [43]. Here we evaluated the eVect of a novel pan- Bcl-2 protein inhibitor on lung cancer cell lines with deW- ned EGFR status and evaluated the combination eVect with either EGFR TKI or cytotoxic chemotherapy. The major Wndings include demonstrating that GX15-070 is cytotoxic in some NSCLC cells and the ability to induce apoptosis, based on PARP cleavage, correlates with a lower IC50 in cell viability assays. Our results also suggest that prolonged exposure time may result in increased cell death. We identiWed no obvious relationship between the studied Bcl-2 proteins and the sensitivity of the cells to GX15-070. GX15-070 in combination with geWtinib was synergistic in a cell line with mutant EGFR sensitive to geWtinib. However, in cells with wildtype EGFR that were not dependent on EGFR for survival, GX15-070 cannot reverse geWtinib resistance. Finally, GX15-070 exhibited synergistic eVects with cisplatin in two of the three cell lines studied.
To our knowledge this is the Wrst examination of Bcl-2 family members in cells with activating EGFR mutations. Consistent with prior reports suggesting that Bcl-2 protein expression is rare in NSCLC, we found no expression of Bcl-2 protein in cells with activating EGFR mutations while Bcl-xL and Mcl-1 expression were more evident. This suggest that agents targeting Bcl-2 are unlikely to demonstrate eYcacy in mutant EGFR cell lines, generally of adenocarcinoma origin, since Bcl-2 protein is not expressed. These current students reiterate our previous work suggesting that Mcl-1 is a viable anti-apoptotic target in lung cancer cells [26].
We were unable to Wnd a clear relationship between Bcl-2 family protein expression and sensitivity to GX15- 070. However, the more interesting Wnding was that cells undergoing apoptosis as evidenced by cleavage of PARP was correlated with lower IC50 in cellular viability assays. This suggests one possible biomarker of GX15-070 activ- ity may be early evidence of apoptosis in tumor cells. For example, induction of apoptosis was monitored using

Table 2 Combination eVect of GX15-070 and EGFR TKI on cell proliferation

Cell line

PC9

Combination drug

GX and geWtinib

Dose range (ti M)

0.25–1.5 and 0.0025–0.015

CI

0.765 § 0.123

Listed are the cell line studied, dose range of GX15-070 and geWtinib, respectively, and the combination indices
(mean § SD)
A549
H460
H1299
H358
GX and geWtinib GX and geWtinib GX and geWtinib GX and geWtinib
1.25–10 and 0.25–2 1.25–10 and 0.25–2 1.25–10 and 0.25–2 1.25–10 and 0.25–2
0.957 § 0.203 3.441 § 1.247
0.984 § 0.105 1.121 § 0.211

100
90
80
70
60
50
40
30
20
10
0

A549-GX+Cisplatin

100
90
80
70
60
50
40
30
20
10
0

100
90
80
70
60
50
40
30
20
10
0

H1299-GX+Cisplatin

0.06-1.25 0.12-2.5 0.25-5.0 0.5-10
Dose of GX and Cisplatin (µM)
0.25-0.31 0.5-0.62 1-1.25 2-2.5
Dose of GX and Cisplatin (µM)
0.25-1.25 0.50-2.5 1.0-5.0 2.0-10
Dose of GX and Cisplatin (µM)

100
90
80
70
60
50
40
30
20
10
0

A549-GX+Gemcitabine

100
90
80
70
60
50
40
30
20
10
0

H460-GX+Gemcitabine
100
90
80
70
60
50
40
30
20
10
0

0.12-0.04 0.25-0.08 0.5-0.16 1-0.33
Dose of GX and Gemcitabine (µM)
0.25-0.04 0.50-0.08 1.0-0.16 2.0-0.33
Dose of GX and Gemcitabine (µM)
0.25-0.04 0.50-0.08 1.0-0.16 2.0-0.33
Dose of GX and Gemcitabine (µM)

100
90
80
70
60
50
40
30
20
10
0

A549-GX+Paclitaxel

100
90
80
70
60
50
40
30
20
10
0

H460-GX+Paclitaxel

100
90
80
70
60
50
40
30
20
10
0

0.12-1.2 0.25-2.5 0.5-5.0 1-10.0 0.25-0.62 0.50-1.25 1.0-2.5 2.0-5.0 0.25-0.012 0.5-0.025 1.0-0.05 2.0-0.1
Dose of GX and Paclitaxel (µM) Dose of GX and Paclitaxel (µM) Dose of GX and Paclitaxel (µM)

Fig. 6 Combination eVect of GX15-070 and chemotherapy in lung cancer cells. A549, H460, and H1299 cells were exposed to increasing concentrations of GX15-070, cytotoxic agents (cisplatin, gemcitabine, paclitaxel), or the combination of both agents in a Wxed ratio. Viable

cells were counted using trypan blue staining and combination indices (CI) were calculated using Calcusyn software as described. The values above each bar represent CI for that dose combination. GX GX15-070, Cis cisplatin, Gem gemcitabine, Tax paclitaxel

Table 3 Combination eVect of GX15-070 and chemotherapy on cell proliferation

Cell line

A549

Combination drug

GX and cisplatin

Dose range (ti M)

0.06–0.5 and 1.25–10

CI

0.677 § 0.186

GX and gemcitabine 0.12–1.0 and 0.04–0.33 1.197 § 0.400
GX and paclitaxel 0.12–1.0 and 1.25–10 1.174 § 0.448
H460 GX and cisplatin 0.25–2.0 and 0.31–2.5 0.657 § 0.139
GX and gemcitabine 0.25–2.0 and 0.04–0.33 1.612 § 1.322

Listed are the cell line studied, compounds, dose range of GX15-070 and cytotoxic, respectively, and the combina- tion indices (mean § SD)

H1299
GX and paclitaxel GX and cisplatin
GX and gemcitabine GX and paclitaxel
0.12–1.0 and 0.62–5.0 0.25–2.0 and 1.25–10 0.25–2.0 and 0.04–0.33 0.12–1.0 and 0.012–0.1
1.061 § 0.379 1.572 § 0.555 1.194 § 0.403 1.524 § 0.452

plasma oligonucleosomal DNA complexes in patients with chronic lymphocytic leukemia treated with GX15- 070 [44]. While diYcult to perform, another approach could be rapid serial biopsies in patients with accessible solid tumors to assess apoptosis. Alternatively, non-inva- sive imaging strategies that can evaluate in vivo tumor apoptosis could be an early marker for GX15-070 activity in patients. Further studies are indicated to evaluate this hypothesis.

Based on our studies, the combination of cisplatin and GX15-070 results in some degree of synergistic tumor cell kill while combinations of other cytotoxics does not demonstrate such eVect. These studies are obviously lim- ited since the data are generated using cell culture models instead of in vivo models. In addition, we found that GX15- 070 cannot reverse geWtinib resistance in lung cancer cells not dependent on EGFR for survival. While previous stud- ies demonstrated that exogenous Bcl-2 or Mcl-1 expression

can prevent geWtinib-induced apoptosis, our results suggest that simply downregulating the activity of these proteins may be insuYcient for reversing EGFR TKI sensitivity [25, 26]. This may not be surprising given the numerous parallel pathways that exist in lung cancer cells that can aVect apoptosis [1].
Early studies of GX15-070 in humans with advanced cancer are underway and the agent appears to be safe and well tolerated. Based on our results as well as the literature, studies of GX15-070 in combination with cytotoxic chemo- therapy are planned for patients with advanced NSCLC.

Acknowledgments We thank Dr Mark Watson (GeminX Biotech- nologies Inc., Montreal, QC, Canada) for help with coimmunoprecipi- tation assays and Drs Pasi Janne (Dana Farber, Boston, MA, USA), Jon Kurie (MD Anderson, Houston, TX, USA), and JeVrey Settleman (Massachusetts General Hospital Cancer Center, Boston, MA, USA) for providing cell lines. We thank Drs Warren Fiskus and Kapil Bhalla (Medical College of Georgia Cancer Center, Augusta, GA, USA) for help with synergy analysis, and TiVany Dyn for administrative assis- tance. This work has been supported in part by the Molecular Imaging Core, Molecular Biology and Sequencing Core, and the Flow Cytom- etry Core at the H. Lee MoYtt Cancer Center & Research Institute. This work was partially funded by the H. Lee MoYtt Cancer Center &
Research Institute.

References

1.Haura EB, Cress WD, Chellappan S, Zheng Z, Bepler G (2004) Antiapoptotic signaling pathways in non-small-cell lung can- cer: biology and therapeutic strategies. Clin Lung Cancer 6:113–122
2.Reed JC (1996) Mechanisms of Bcl-2 family protein function and dysfunction in health and disease. Behring Inst Mitt 97:72–100
3.Sedlak TW, Oltvai ZN, Yang E, Wang K, Boise LH, Thompson CB, Korsmeyer SJ (1995) Multiple Bcl-2 family members demon- strate selective dimerizations with Bax. Proc Natl Acad Sci U S A 92:7834–7838
4.Adams JM, Cory S (1998) The Bcl-2 protein family: arbiters of cell survival. Science 281:1322–1326
5.Cory S, Huang DC, Adams JM (2003) The Bcl-2 family: roles in cell survival and oncogenesis. Oncogene 22:8590–8607
6.Anton RC, Brown RW, Younes M, Gondo MM, Stephenson MA, Cagle PT (1997) Absence of prognostic signiWcance of bcl-2 im- munopositivity in non-small cell lung cancer: analysis of 427 cas- es. Hum Pathol 28:1079–1082
7.Apolinario RM, van der Valk P, de Jong JS, Deville W, van Ark- Otte J, Dingemans AM, van Mourik JC, Postmus PE, Pinedo HM, Giaccone G (1997) Prognostic value of the expression of p53, bcl- 2, and bax oncoproteins, and neovascularization in patients with radically resected non-small-cell lung cancer. J Clin Oncol 15:2456–2466
8.Borner MM, Brousset P, Pfanner-Meyer B, Bacchi M, Vonlanthen S, Hotz MA, Altermatt HJ, Schlaifer D, Reed JC, Betticher DC (1999) Expression of apoptosis regulatory proteins of the Bcl-2 family and p53 in primary resected non-small-cell lung cancer. Br J Cancer 79:952–958
9.Fontanini G, Vignati S, Bigini D, Mussi A, Lucchi M, Angeletti CA, Basolo F, Bevilacqua G (1995) Bcl-2 protein: a prognostic factor inversely correlated to p53 in non-small-cell lung cancer. Br J Cancer 71:1003–1007

10.GaVney EF, O’Neil AJ, Staunton MJ (1994) Bcl-2 and prognosis in non-small-cell lung carcinoma. N Engl J Med 330:1757–1758
11.Higashiyama M, Doi O, Kodama K, Yokouchi H, Tateishi R (1995) High prevalence of bcl-2 oncoprotein expression in small cell lung cancer. Anticancer Res 15:503–505
12.Higashiyama M, Doi O, Kodama K, Yokouchi H, Tateishi R (1996) Bcl-2 oncoprotein expression is increased especially in the portion of small cell carcinoma within the combined type of small cell lung cancer. Tumour Biol 17:341–344
13.Higashiyama M, Doi O, Kodama K, Yokouchi H, Nakamori S, Tateishi R (1997) Bcl-2 oncoprotein in surgically resected non- small cell lung cancer: possibly favorable prognostic factor in association with low incidence of distant metastasis. J Surg Oncol 64:48–54
14.Ikegaki N, Katsumata M, Minna J, Tsujimoto Y (1994) Expression of bcl-2 in small cell lung carcinoma cells. Cancer Res 54:6–8
15.Katz HR (1994) Bcl-2 protein in non-small-cell lung carcinoma. N Engl J Med 330:221
16.Pezzella F, Turley H, Kuzu I, Tungekar MF, Dunnill MS, Pierce CB, Harris A, Gatter KC, Mason DY (1993) Bcl-2 protein in non- small-cell lung carcinoma. N Engl J Med 329:690–694
17.Ritter JH, Dresler CM, Wick MR (1995) Expression of bcl-2 pro- tein in stage T1N0M0 non-small cell lung carcinoma. Hum Pathol 26:1227–1232
18.Sartorius UA, Krammer PH (2002) Upregulation of Bcl-2 is in- volved in the mediation of chemotherapy resistance in human small cell lung cancer cell lines. Int J Cancer 97:584–592
19.Gautschi O, Tschopp S, Olie RA, Leech SH, Simoes-Wust AP, Ziegler A, Baumann B, Odermatt B, Hall J, Stahel RA, Zangemei- ster-Wittke U (2001) Activity of a novel bcl-2/bcl-xL-bispeciWc antisense oligonucleotide against tumors of diverse histologic ori- gins. J Natl Cancer Inst 93:463–471
20.Kataoka M, Wiehle S, Spitz F, Schumacher G, Roth JA, Cristiano RJ (2000) Down-regulation of bcl-2 is associated with p16INK4- mediated apoptosis in non-small cell lung cancer cells. Oncogene 19:1589–1595
21.Koty PP, Zhang H, Levitt ML (1999) Antisense bcl-2 treatment in- creases programmed cell death in non-small cell lung cancer cell lines. Lung Cancer 23:115–127
22.Leech SH, Olie RA, Gautschi O, Simoes-Wust AP, Tschopp S, Haner R, Hall J, Stahel RA, Zangemeister-Wittke U (2000) Induc- tion of apoptosis in lung-cancer cells following bcl-xL anti-sense treatment. Int J Cancer 86:570–576
23.Ouyang N, Ran P, Qiu Z (2000) [Bcl-2 antisense oligodeoxyribo- nucleotide increases apoptosis of lung carcinoma cells induced by cisplatin]. Zhonghua Jie He He Hu Xi Za Zhi 23:722–724
24.Ziegler A, Luedke GH, Fabbro D, Altmann KH, Stahel RA, Zangemeister-Wittke U (1997) Induction of apoptosis in small- cell lung cancer cells by an antisense oligodeoxynucleotide target- ing the Bcl-2 coding sequence. J Natl Cancer Inst 89:1027–1036
25.Janmaat ML, Kruyt FA, Rodriguez JA, Giaccone G (2003) Re- sponse to epidermal growth factor receptor inhibitors in non-small cell lung cancer cells: limited antiproliferative eVects and absence of apoptosis associated with persistent activity of extracellular sig- nal-regulated kinase or Akt kinase pathways. Clin Cancer Res 9:2316–2326
26.Song L, Coppola D, Livingston S, Cress D, Haura EB (2005) Mcl- 1 regulates survival and sensitivity to diverse apoptotic stimuli in human non-small cell lung cancer cells. Cancer Biol Ther, 4:267– 276
27.Fesik SW (2005) Promoting apoptosis as a strategy for cancer drug discovery. Nat Rev Cancer 5:876–885
28.Banerjee D (1999) Technology evaluation: G-3139. Curr Opin Mol Ther 1:404–408
29.Banerjee D (2001) Genasense (Genta Inc). Curr Opin Investig Drugs 2:574–580

30.Morris MJ, Tong WP, Cordon-Cardo C, Drobnjak M, Kelly WK, Slovin SF, Terry KL, Siedlecki K, Swanson P, RaW M, DiPaola RS, Rosen N, Scher HI (2002) Phase I trial of BCL-2 antisense oligo- nucleotide (G3139) administered by continuous intravenous infu- sion in patients with advanced cancer. Clin Cancer Res 8:679–683
31.Rudin CM, Otterson GA, Mauer AM, Villalona-Calero MA, Tomek R, Prange B, George CM, Szeto L, Vokes EE (2002) A pilot trial of G3139, a bcl-2 antisense oligonucleotide, and paclitaxel in patients with chemorefractory small-cell lung cancer. Ann Oncol 13:539–545
32.Rudin CM, KozloV M, HoVman PC, Edelman MJ, Karnauskas R, Tomek R, Szeto L, Vokes EE (2004) Phase I study of G3139, a bcl-2 antisense oligonucleotide, combined with carboplatin and etoposide in patients with small-cell lung cancer. J Clin Oncol 22:1110–1117
33.Reed JC, Pellecchia M (2005) Apoptosis-based therapies for he- matologic malignancies. Blood 106:408–418
34.Tracy S, Mukohara T, Hansen M, Meyerson M, Johnson BE, Janne PA (2004) GeWtinib induces apoptosis in the EGFRL858R non-small-cell lung cancer cell line H3255. Cancer Res 64:7241– 7244
35.Fujimoto N, Wislez M, Zhang J, Iwanaga K, Dackor J, Hanna AE, Kalyankrishna S, Cody DD, Price RE, Sato M, Shay JW, Minna JD, Peyton M, Tang X, Massarelli E, Herbst R, Threadgill DW, Wistuba II, Kurie JM (2005) High expression of ErbB family members and their ligands in lung adenocarcinomas that are sensi- tive to inhibition of epidermal growth factor receptor. Cancer Res 65:11478–11485
36.Kwak EL, Sordella R, Bell DW, Godin-Heymann N, Okimoto RA, Brannigan BW, Harris PL, Driscoll DR, Fidias P, Lynch TJ, Ra- bindran SK, McGinnis JP, Wissner A, Sharma SV, Isselbacher KJ, Settleman J, Haber DA (2005) Irreversible inhibitors of the EGF receptor may circumvent acquired resistance to geWtinib. Proc Natl Acad Sci USA 102(21):7665–7670
37.Song L, Turkson J, Karras JG, Jove R, Haura EB (2003) Activa- tion of Stat3 by receptor tyrosine kinases and cytokines regulates

survival in human non-small cell carcinoma cells. Oncogene 22:4150–4165
38.Chou TC, Talalay P (1984) Quantitative analysis of dose-eVect relationships: the combined eVects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 22:27–55
39.Haura EB, Zheng Z, Song L, Cantor A, Bepler G (2005) Activated epidermal growth factor receptor-stat-3 signaling promotes tumor survival in vivo in non-small cell lung cancer. Clin Cancer Res 11:8288–8294
40.Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, Herman P, Kaye FJ, Lindeman N, Boggon TJ, Naoki K, Sasaki H, Fujii Y, Eck MJ, Sellers WR, Johnson BE, Meyerson M (2004) EGFR mutations in lung cancer: correlation with clinical response to geWtinib therapy. Science 304:1497–1500
41.Sordella R, Bell DW, Haber DA, Settleman J (2004) GeWtinib-sen- sitizing EGFR mutations in lung cancer activate anti-apoptotic pathways. Science 305:1163–1167
42.Pao W, Miller VA, Politi KA, Riely GJ, Somwar R, Zakowski MF, Kris MG, Varmus H (2005) Acquired resistance of lung adenocar- cinomas to geWtinib or erlotinib is associated with a second muta- tion in the EGFR kinase domain. PLoS Med 2:1–11
43.Oltersdorf T, Elmore SW, Shoemaker AR, Armstrong RC, Augeri DJ, Belli BA, Bruncko M, Deckwerth TL, Dinges J, Hajduk PJ, Jo- seph MK, Kitada S, Korsmeyer SJ, Kunzer AR, Letai A, Li C, Mit- ten MJ, Nettesheim DG, Ng S, Nimmer PM, O’Connor JM, Oleksijew A, Petros AM, Reed JC, Shen W, Tahir SK, Thompson CB, Tomaselli KJ, Wang B, Wendt MD, Zhang H, Fesik SW, Rosenberg SH (2005) An inhibitor of Bcl-2 family proteins induc- es regression of solid tumours. Nature 435:677–681
44.O’Brien S, Kipps TJ, Fader S, Crump M, Keating M, Anderson B, Soho C, Bole J, Turner R, Viallet J, Cheson B (2004) A phase I tri- al of the small molecule pan-bcl-2 family inhibitor GX15-070 administered intravenously (IV) every 3 weeks to patients with previously treated chronic lymphocytic leukemia (CLL). Ameri- can Society of Hematology (ASH), Washington