The Proteasome Inhibitor Bortezomib Acts Differently in Combination With

Home , Bortezomib

ORIGINAL ARTICLE The proteasome inhibitor bortezomib acts differently in combination with p53 gene transfer or cytotoxic chemotherapy on NSCLC cells J Neukirchen1, A Meier1, A Rohrbeck1, G Garcia-Pardillos1, U Steidl2, R Fenk1, R Haas1, R Kronenwett1 and U-P Rohr1 1Klinik fu¨r Haematologie, Onkologie und klinische Immunologie, Heinrich-Heine-Universitaet Duesseldorf, Duesseldorf, Germany and 2Beth Israel Deaconess Medical Center, Harvard Institutes of Medicine, Boston, MA, USA

In this report, the effects of a combined treatment with the proteasome inhibitor bortezomib and either a recombinant adeno- associated virus type 2 (rAAV-2)-mediated p53 gene transfer or chemotherapeutic agents, docetaxel and pemetrexed, were tested on p53 positive and p53negative non-small cell lung cancer (NSCLC) cell lines. The combination of bortezomib and rAAV-p53 led to a significant synergistic inhibition of cell growth between 62–82% depending on the p53 status of the cell line and drug concentration. Surviving cells of the combined treatment showed a significant reduced ability to form colonies. Enhanced cell toxicity was associated with a 5.3–14.4-fold increase of the apoptotic rate and intracellular p53 level up to 50.4% following vector- mediated p53 restoration and bortezomib treatment. In contrast, an antagonistic effect on tumor cell growth and colony formation was observed for the combination of bortezomib and docetaxel or pemetrexed as a reduction of cell growth between 31 and 48% was found in comparison to 50% using the single agents. Lower cytotoxic effects were associated with significantly reduced apoptosis and an increase of clonogenic growth. The observed antagonistic effects between bortezomib and docetaxel or pemetrexed might influence clinical trials using these compounds. Conversely, p53 restoration and bortezomib treatment led to enhanced, synergistic tumor cell toxicity. Cancer Gene Therapy (2007) 14, 431–439. doi:10.1038/sj.cgt.7701029; published online 19 January 2007 Keywords: bortezomib; rAAV-p53; docetaxel; pemetrexed

6,7 Introduction restore wild-type p53. Unfortunately, gene therapy strategies restoring p53 alone or in combination with Non-small cell lung cancer (NSCLC) represents the chemotherapy failed to improve the outcome of the 8 majority of lung cancers and is incurable in advanced patients. stages. This is reflected by a 5-year survival rate of less A new class of biologically active agents in lung cancers than 3% and a median survival of 8–11 months after a are proteasome inhibitors. Bortezomib is a dipeptide standard chemotherapy.1,2 Considering the poor prog- boronic acid and a potent, highly selective and reversible 9 nosis, second-line therapies for patients with progressive inhibitor of the 26S proteasome. Next to lung cancer, disease have been approved, such as docetaxel and in vitro and in vivo studies have shown antitumor activity as pemetrexed.3,4 In a phase III trial, both administered a single agent in a wide range of solid and hematological as single agents showed comparable median survival of malignancies, including prostate, breast, colon, pancreatic 10,11 7.9–8.3 months and a 1-year survival of 30% for cancers and myeloma. Consequently, it was approved docetaxel and pemetrexed, respectively.5 in 2003 for the treatment of advanced multiple myeloma. Gene therapy is a possible modality for the treatment The 26S proteasome plays a critical role in the degrada- of patients with lung cancer. As p53 is lost or mutated in tion of proteins involved in multiple cellular processes, numerous cancers, a number of trials used viral vectors to including the cell cycle, transcription-factor activation, apoptosis and intracellular signal-transduction traffic, which are relevant for tumor progression and drug Correspondence: Dr U-P Rohr, Ha¨ matologie, Onkologie und 12 Klinische Immunologie, University of Du¨ sseldorf, Moorenstrae5, resistance. The proteasome degrades the tumor sup- Du¨ sseldorf 40225, Germany. pressor p53, which is required for the transcription of a E-mail: [email protected] number of genes involved in cell-cycle control and DNA 13 Received 31 July 2006; revised 16 October 2006; accepted 18 synthesis. Wild-type p53 is stabilized by proteasome November 2006; published online 19 January 2007 inhibition in cancer cell lines,14,15 suggesting a mechanism Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 432 of bortezomib-induced apoptosis that may be relevant in resuspended with RPMI medium in a final volume of suppressing cancer progression.16,17 1000 ml. Viable cells were assessed as stated below. We focussed on bortezomib as a promising agent to To determine the concentration required for 50% investigate if its combination with p53 gene transfer might inhibition of growth (inhibitory concentration 50%, enhance cytotoxicity in NSCLC cell lines. Further, the IC50) for rAAV-p53, cells were incubated with rAAV- two chemotherapeutic drugs, pemetrexed (Alimta) and p53 in different multiplicities of infections (MOIs) docetaxel (Taxotere), were used in combination with between 10 and 1000. For bortezomib, pemetrexed and bortezomib to analyze the effects on NSCLC cell lines. docetaxel experiments, dose–response curves for each Our preclinical findings describe drug interactions drug were determined by incubating the cells with between bortezomib, p53 and cytotoxic agents, which dilutions between 0.0001 and 100 mM of the drug. For might influence further clinical studies that are currently further experiments, the IC50 of rAAV-p53, pemetrexed planned for patients with recurrent NSCLC. and docetaxel was chosen for combination with the IC50 and IC25 of bortezomib. These concentrations were considered as suitable to analyze growth inhibitory effects on the lung cancer cell lines. Material and methods In an attempt to evaluate subadditive, additive or Cell culture overadditive growth inhibitory effects of bortezomib, The NSCLC cell lines H460 showing wild-type p53 rAAV-p53, pemetrexed and docetaxel on NSCLC cell expression18 and the p53-negative cell line H1299, in lines, we calculated hypothetical combined survival rates which both copies of the p53 gene are deleted,19 as well as by multiplying the effects of each drug given as a single T47D cells (ductal carcinoma of the breast) with mutant agent. The calculated survival rates were compared with p53 were cultured in Rosewell Park Memorial Institute the measured survival rates of the combined treatment. (RPMI)-1640 medium (Sigma, Deissenhofen, Germany) supplemented with 10% heat-inactivated fetal calf serum Assessment of viable cells (PAA Laboratories, Linz, Austria), 2 mML-glutamine, The viability of cells was assessed by Trypan blue 100 U/ml penicillin and 0.1 mg/ml streptomycin (all exclusion and by counting using a Neubauer counting Sigma). All cell lines were cultured in a humidified chamber. The survival rate in percent was calculated by atmosphere with 5% CO2 at 371C. The H1299 and the dividing the number of viable cells by the number of H460 cells were highly susceptible for recombinant adeno- viable untreated control cells multiplied by 100. Inhibition associated virus type 2 (rAAV-2)-mediated gene transfer in percent was calculated by the following formula: with a mean transduction rate of 82%.20 The human inhibition in % ¼ 100%Àsurvival rate in %. embryonic kidney cells 293Twere cultured under the same conditions using Dulbecco’s modified Eagle’s Clonogenic assay medium (Sigma) instead of RPMI-1640. Clonogenic assays were performed with surviving cells following treatment with bortezomib, rAAV-p53, docetaxel or pemetrexed as described above. After incubation Production of recombinant AAV-2 vector stocks of the cells with different drugs in different concentrations Recombinant AAV-2 vector stocks containing the p53 2 (IC25 and IC50) for 3 days, 5 Â 10 viable cells were gene (AAV-2-p53) and the green fluorescent protein assessed by Trypan blue exclusion and placed in 25 cm2 (GFP)-gene (AAV-2-GFP) were produced as described 21 flasks with fresh medium. After 7 days, colonies were previously. Briefly, 293Tcells were cotransfected with counted using a light microscope. Only colonies contain- 20 mg vector plasmid (pAAV-p53-AR6,20 respectively, 22 23 ing more than 50 cells were counted. The percentage of pAAV-GFP-AR6 ) and 20 mg helper plasmid (pDG ) colonies was calculated by dividing the number of using calcium phosphate precipitation. Three days later, colonies derived from treated cells by those from 293Tcells were harvested and lysed by three freeze–thaw untreated cells and multiplying by 100. cycles following sonication. The lysate was centrifuged and the supernatant was subjected to an iodixanol 24 Apoptosis assay gradient to purify viral particles. After centrifugation, To determine and quantify the induction of apoptosis, fractions of 500 ml were collected and dialysed against DNA fragmentation was measured using a cell death 1 RPMI-1640 for 24 h at 4 C. AAV-2-p53 and AAV-2-GFP detection enzyme-linked immunosorbent assay (ELISA) fractions were quantified using a quantitative real-time kit (Roche, Molecular Biochemicals, Mannheim, PCR for titration of infectious particles as described 25 Germany). The enrichment of mono- and oligonucleo- previously. somes in the cytoplasm of apoptotic cells is attributable to the fact that DNA degradation occurs several Cell culture and treatment with drugs hours before plasma membrane breakdown. After treat- 5 Â 104 lung cancer cells were cultured in 24-well plates ment of the cells as described above, 1 Â 104 cells were and incubated with different concentrations of bortezo- used for the ELISA procedure, according to the mib, rAAV-p53, docetaxel and pemetrexed, alone and manufacturer’s protocol. DNA fragmentation was in combination. After 3 days, cells were detached quantified at 405 nm in a 1420 Multilabel Counter with trypsin/ethylenediaminetetraacetic acid (Sigma) and (EG&G Wallac, Turku, Finland). The enrichment of

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 433 mono- and oligonucleosomes released into the cytoplasm a H1299 was calculated as absorbance of sample cells/absorbance no M O I MOI MOI MOI MOI MOI MOI of untreated control, whereas the measured apoptosis rate virus 10 50 100 150 500 700 1000 from the untreated cells was defined as 1. 120% bortezomib docetaxel Western blot analysis 100% pemetrexed Western blot was performed as already described.20 rAAV-p53 Briefly, 1 Â 105 cells were plated in 24-well plates and 80% incubated with an IC50 of rAAV-p53 and bortezomib 60% alone and in combination with bortezomib for 24 h. As primary antibodies, a monoclonal mouse antihuman 40% p53 antibody (Clone DO-7, Dako Cytomation, Hamburg, Germany; dilution 1:500) and a monoclonal mouse 20% antirabbit glyceraldehyde-3-phosphate dehydrogenase 0% (GAPDH) antibody (Clone 6C5, Acris Antibodies, Hiddenhausen, Germany; dilution 1:25000) were used. 0 µM 1 µM 10 µM 0.1 µM 0.5 µM

The primary antibodies were detected with a peroxidase- 100 µM 0.01 µM 0.03 µM 0.05 µM 0.06 µM 0.001 µM

coupled rabbit antimouse secondary antibody (P0161, 0.0001 µM Dako Cytomation, Hamburg, Germany; dilution 1:1000). Immuncomplexes were visualized with an enhanced b H460 chemiluminescence kit (Amersham, Buckinghamshire, no M O I MOI MOI MOI MOI MOI MOI MOI MOI England) according to the manfacturer’s instructions. virus 10 50 100 150 200 400 500 700 1000 120% As a positive control, T47D cells, which express p53 at bortezomib docetaxel high levels, were used. T47D p53 has a point mutation in 100% codon 194 that leads to stabilization and accumulation pemetrexed 26 rAAV-p53 of nonfunctional p53 protein. To compare the different 80% bands, an AlphaImager Imaging System (Biozym Scientific GmbH, Oldendorf, Germany) and AlphaEase 60% FC software (Biozym Scientific GmbH) were used. To 40% determine relative changes in the p53 expression, the value of the p53 band was divided by the value of the 20% corresponding GAPDH control band and multiplied with 100. 0% 1 µM Statistical analysis 0 µM 0.1 µM 0.5 µM 0.01 µM 0.03 µM 0.05 µM 0.06 µM 0.005 µM All assays were performed in triplicate and repeated at 0.001 µM least for three times. The mean results of the nine 0.0001 µM independent experiments and the s.d. are presented in Figure 1 Cytotoxic effects of each administered drug on the the text. For statistical analysis, the two-sided Student’s NSCLC cell lines H1299 (a) and H460 (b) after an incubation period t-test was used. Statistical significance was defined as of 3 days. The concentrations of the rAAV-p53 vector are given as MOI on top. Po0.05. the H460 cells. The rAAV-p53 vector led to a 50% growth inhibition of the H1299 and H460 cells at a MOI of 150 Results and 400, respectively. The IC50 of each drug and the IC25 Effects of bortezomib, rAAV-p53, docetaxel and of bortezomib were used for the following experiments. pemetrexed as single agents on NSCLC cell lines First, we investigated the cytotoxic effects of bortezomib, Effects of bortezomib combined with rAAV-2-mediated rAAV-p53, docetaxel and pemetrexed on the NSCLC cell p53 gene transfer lines H1299 and H460 after a 3-day incubation period of Having evaluated the effects of the single agents on each single agent using concentrations ranging between NSCLC cells, we asked next whether a coincubation of 0.0001 and 100 mM for the chemotherapeutics and between bortezomib and rAAV-p53 might lead to a pronounced a MOI of 10 and 1000 for rAAV-p53. The three drugs led growth inhibition. Therefore, cell lines were incubated to a dose–dependent growth inhibition in both cell lines with an IC25 and IC50 of bortezomib and an IC50 of (Figure 1). According to these data, we evaluated the IC50 rAAV-p53. After a 3-days incubation period, an for each agent and the IC25 for bortezomib. A concentra- additive effect of p53 and bortezomib was observed in a tion of 0.03 mM bortezomib led to a 25% and 0.05 mM to a dose–dependent manner (Figure 2a and b). When cells 50% growth inhibition in both cell lines. The IC50 for were treated with bortezomib at an IC25 and rAAV-p53, pemetrexed was 0.1 mM for both cell lines. For docetaxel, the growth reduction was 62% (s.d.73.8%) and the IC50 was 0.01 mM for the H1299 cells and 0.005 mM for 69% (s.d.73.4%) for the H1299 and the H460 cells,

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 434 a b 100% *p<0.0001

80%

60%

% survival 40% * * * H1299 control H460 control 20% *

0% control inhibition inhibition calculated rAAV-GFP calculated + rAAV-p53 + rAAV-GFP bortezomib + rAAV-GFP bortezomib IC25 IC25+ rAAV-p53 bortezomib IC50 bortezomib IC50 bortezomib IC50 + bortezomib IC50 rAAV control + rAAV control c *p<0.0001 100%

80%

60% ** **

% colonies 40% bortezomib IC50 + bortezomib IC50 + 20% rAAV-p53 rAAV-p53

0% control rAAV-p53 rAAV-GFP rAAV-p53 rAAV-p53 rAAV-GFP rAAV-GFP bortezomib IC50 + bortezomib IC25 + bortezomib IC25 + bortezomib IC50 + Figure 2 Effects of bortezomib and rAAV-2-mediated p53 gene transfer on NSCLC cell lines performing viability assay and clonogenic assay. (a) Viability assays of NSCLC cell lines H1299 (black bars) and H460 (gray bars) after treatment with bortezomib and rAAV-p53. The combined treatment led to a significant additive growth inhibition of both cell lines in comparison to the cells treated with bortezomib and the control vector rAAV-GFP (Po0.0001). (b) Phase-contrast microscopy of H1299 and H460 cells from a representative experiment following incubation with bortezomib and rAAV-p53 at an IC50.(c) Effects of bortezomib and rAAV-p53 on clonogenic growth of NSCLC cell lines H1299 (black bars) and H460 (white bars). Following combined treatment, surviving cells were inhibited in their ability to form colonies. Consequently, cells pretreated with bortezomib and rAAV-p53 had a significantly lower ability to form colonies than cells pretreated with bortezomib and rAAV-GFP (Po0.0001).

respectively, in comparison with untreated cells. Using an Following combined bortezomib and rAAV-p53 treat- IC50 for bortezomib, the growth reduction was 74% ment, a fraction of the tumor cells survived. Therefore, we (s.d.72.4%) in the p53-negative cell line H1299. Compar- were interested in the ability of those surviving tumor cells ing the reduction of viable cells with the calculated to form colonies. We found that bortezomib and rAAV- survival rates of the combined treatment, an additive p53 treated cells were significantly inhibited in their effect was found for H1299 cells with a growth reduction ability to form colonies in comparison to cells that were of 74%, which was almost identical to the calculated treated with bortezomib-and the control vector rAAV- value of 75% using the IC50. In the p53 positive cell line GFP (Figure 2c). In particular, the combined treatment H460, the additive growth inhibitory effect of the with bortezomib and rAAV-p53-led to a significant combined treatment was more evident. This is reflected colony reduction of 55% (s.d.74.0%) and 58% by a growth inhibition of 84% (s.d.74.0%) using an IC50 (s.d.73.4%) when IC25 and IC50 were used in H1299 for bortezomib, significantly exceeding the calculated cells (Po0.001). The inhibitory effect on clonogenic growth inhibition of 75% (Po0.002). In contrast to colony growth for the p53-positive cell line H460 was rAAV-p53, no growth inhibition was observed for the more distinct. The combined treatment with IC25 and control vector rAAV-GFP. IC50 led to a significant reduction of the colony growth

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 435 of 66% (s.d.76.0%) and 81% (s.d.75.2%), respectively observed, respectively, indicating successful inhibition of (Po0.0001). p53 degradation and restoration. The combined treatment of bortezomib and rAAV-p53 led to the strongest intracellular increase of p53 of 50.4% in comparison with p53 expression after rAAV-2-mediated p53 gene transfer untreated cells. and bortezomib treatment The expression of p53 in H1299 (p53-null) and H460 Apoptotic activity following p53 and bortezomib (WTp53) cells after treatment with bortezomib and treatment in NSCLC cell lines rAAV-p53 alone and in combination was examined by In an attempt to ascertain if the synergistic effects of immunoblotting (Figure 3). Owing to the homozygous bortezomib and rAAV-2-mediated p53 gene transfer are deletion of the p53 gene, the p53-null H1299 cells did associated with an increased apoptosis rate in the NSCLC not express p53 after treatment with bortezomib. After cell lines, apoptotic activity following various treatments infection of the cells with rAAV-p53 using an MOI of were determined by relative changes of DNA fragmenta- 150 infectious particles per cell, restoration of p53 tion of treated cells in comparison to untreated cells expression was detected. The combined treatment with (Figure 4). bortezomib and rAAV-p53 led to an increase of p53 A significant dose–dependent increase of apoptotic expression of 29.3% in comparison to the single treatment activity was seen in both cell lines using a combination of with rAAV-p53. rAAV-p53 and bortezomib in comparison to untreated The WT p53 positive H460 cells exhibit a low p53 cells. In particular, for H1299 cells a significant increase expression. After treatment with either bortezomib or of apoptosis ranging from a factor of 3.9 (s.d.70.9) to 5.3 rAAV-p53, an increase of 7.4 and 32.7% of p53 level was (s.d.70.6) was seen when cells were treated with bortezomib with an IC25 and IC50 in combination with rAAV-p53, respectively (Po0.001). In the H460 cells, a a significant dose–dependent increase of the apoptotic activity of the factor 9.2 (s.d.71.9) and 14.5 (s.d.72.4) was observed when using rAAV-p53 and bortezomib at an IC and IC , respectively (P 0.001). In contrast, p53 25 50 o this effect was not seen when cells were treated with the control vector rAAV-GFP. GAPDH Effects of the combined treatment of bortezomib and the cytotoxic chemotherapeutic agents docetaxel and pemetrexed Control H1299 H1299 + H1299 + H1299 +rAAV- p53+ p53- bortezomib rAAV-p53 p53 + Next, the effects of a combination of the cytotoxic agents bortezomib docetaxel or pemetrexed with bortezomib on NSCLC cell b p53 18 *p<0.001 * *

* * * * GAPDH 10

H460 H460 + H460 + H460 + rAAV-

p53+ bortezomib rAAV-p53 p53 + relative apoptotic activity bortezomib 00 Figure 3 Stabilization of p53 after incubation with rAAV-p53 and bortezomib as shown by Western blot analysis. (a) The p53-null H1299 cells show a strong p53 expression after restoration of the control

p53-gene with rAAV-p53. The combined treatment with bortezomib rAAV-GFP and rAAV-p53 led to an inhibition of p53 degradation and an rAAV-p53 IC50 IC50 + rAAV-53

increased p53 concentration. As a p53-positive control, p53 bortezomib IC25 bortezomib IC50 IC25 + rAAV-p53 expressing T47D cells were used. Of note, the point-mutated p35 IC25 + rAAV-GFP IC50 + rAAV-GFP protein of control T47D cells had a marginal divergence in the Figure 4 Induction of apoptosis following treatment of NSCLC cell electrophoretical band in comparison to the WT-transfected p53 in lines H1299 (black bars) and H460 cells (white bars) with bortezomib H1299 cells. (b) The untreated H460 cells (WT-p53) exhibit a low and rAAV-p53. Data are shown as relative increase of nucleosomal concentration of p53 that increased following transfection with rAAV- DNA fragmentation in comparison with the untreated control cells. p53. The combined treatment with rAAV-p53 and bortezomib led to Cells treated with bortezomib and rAAV-p53 showed a significant highest intracellular p53 concentrations. increase of apoptosis in contrast to the untreated control (Po0.001).

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 436 lines H1299 and H460 were tested. For H1299 cells, a treatment with bortezomib at an IC25 and IC50 in *p<0.05 docetaxel pemetrexed 100% combination with docetaxel at an IC50 led to a growth inhibition of 47% (s.d.73.6%) and 45.7% (s.d.72.3%), 80% * respectively. Consequently, the addition of bortezomib to * ** docetaxel led to an inferior growth reduction compared 60% with those of the single agents. Similar to this observation, the same effect was found in H460 cells, showing a % survival 40% growth inhibition of 48.2% (s.d.71.9%) and 41.5% 20% (s.d.77.0%) using an IC25 and IC50 of bortezomib, respectively. 0% These experiments were repeated for pemetrexed in IC50 IC50

combination with bortezomib. For H1299, a growth control inhibition of 44% (s.d.71.7%) and 35% (s.d.72.4%) was observed. Similar to these results, in H460 cells treated with bortezomib at an IC25 and an IC50 in

combination with pemetrexed, a growth inhibition of + bortezomib IC25 + bortezomib IC50 + bortezomib IC25 + bortezomib IC50 31% (s.d.74.3%) and 41% (s.d.74.3%) was observed. As a result, the addition of either docetaxel or pemetrexed b *p<0.05 *p<0.001 to bortezomib led to a significantly antagonistic effect 100% * bortezomib docetaxel pemetrexed with regard to growth inhibition in both cell lines * * * (Figure 5a). 80% * * * * Next, we asked if the surviving cells of the combined 60% treatment were able to form more colonies than the cells

treated with one drug alone in a clonogenic assay. As a % colonies 40% result, the cells of the combined treatment had the same or higher ability for clonogenic growth in comparison to 20% cells just treated with one agent (Figure 5b). This is 0% exemplified at its best when looking at pemetrexed as a

single agent. Colonies were reduced by 44 (s.d.79.8%) IC25 IC50 IC50 IC50 and 47% (s.d.76.2%) for H1299 and H460, compared to control untreated cells. In contrast, the combination of pemetrexed and bortezomib led to an increase of clonogenic growth. The reduction of colonies compared to untreated + bortezomib IC25 + bortezomib IC50 + bortezomib IC25 + bortezomib IC50 cells was between 30% (v8.4%) and 42% (s.d.75.0%) depending on the cell type and IC of bortezomib. The Figure 5 Effects of bortezomib and docetaxel or pemetrexed on NSCLC cell lines H1299 and H460 in a viability-assay and a reduction of the antiproliferative efficacy of the combined clonogenic assay. (a) Antagonistic effects of bortezomib and the treatment in comparison with the single agent pemetrexed chemotherapeutic agents docetaxel or pemetrexed on cell growth of was statistically significant for both concentrations and NSCLC cell lines H1299 (black bars) and H460 (white bars). Cells

both cell lines (Po0.05). treated with IC50 bortezomib and docetaxel or pemetrexed showed a significant decreased growth inhibition compared to docetaxel or Apoptotic activity of bortezomib in combination with pemetrexed treatment alone at an IC50 (Po0.05). (b) Effects of the chemotherapeutic agents bortezomib and docetaxel or pemetrexed on clonogenic growth of To find out if the antagonistic effects of bortezomib and NSCLC cell lines H1299 (black bars) and H460 (white bars). The the cytotoxic agents were associated with decreased combined treatment of the cells led to a significantly increased apoptosis, the apoptotic activities for the different colony-forming ability compared to the single treatment regimen (P 0.05). treatments were determined. Docetaxel as a single agent o increased the apoptotic activity in H1299 and H460 cells by the factor 3.8 (s.d.70.8) and 4.8 (s.d.70.8) in comparison to untreated cells. significantly decreased apoptotic activity between a factor The apoptotic activity was significantly lower when the of 1.4 and 2.7 for the different cell types and concentra- combination of docetaxel and bortezomib was used tions (Figure 6; Po0.05). (Po0.05). Depending on the cell line and concentration of bortezomib, the apoptotic activity was increased between a factor of 1.9 and 4.2. Using pemetrexed as a single agent, the apoptotic Discussion activity was 2.5 (s.d.71.0) and 2.9 (s.d.70.4) fold increased for H1299 and H460 compared to untreated In this study, we were interested in the growth inhibitory control, respectively. Similar to previous results, the and proapoptotic effects of the proteasome inhibitor combination of pemetrexed and bortezomib led to a bortezomib on NSCLC cell lines in the presence or

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 437 6 combination with chemotherapy were disappointing. docetaxel *p<0.05 bortezomib pemetrexed Although the approach was safe and restoration of p53 5 *p<0.001 was successful, the clinical benefit for the patient was * 7,33–35 4 marginal. Here we were able to demonstrate a synergism between p53 restoration and a new class of * * 3 * * cancer drugs, the proteasome inhibitors. The overadditive * 2 effects of bortezomib and p53 gene transfer might be relevant for clinical trials based on tumor suppressor gene 1 replacement. The combination with bortezomib seems to relative apoptotic activity be beneficial with regard to the inhibition of p53 0 degradation in p53-positive and p53-negative NSCLCs.

IC25 IC50 IC50 IC50 We next looked at the biological effects of the

control proteasome inhibitor bortezomib and the cytotoxic agents pemetrexed and docetaxel. Preliminary experimental data have demonstrated that bortezomib is able to sensitize cancer cell lines to chemotherapeutics such as doxorubicin + bortezomib IC50 + bortezomib IC25 + bortezomib IC50 + bortezomib IC25 or melphalan as seen for patients with multiple myelo- Figure 6 Apoptotic activity in NSCLC cell lines H1299 (black bars) ma.12,27 Others found sequential effects with regard to the and H460 (white bars) measured by DNA fragmentation after time of administration using gemcitabine and carboplatin combined treatment with bortezomib and cytotoxic agents docetaxel in combination with bortezomib.14 We chose docetaxel or pemetrexed. A significantly decreased apoptotic activity was seen and bortezomib as these two agents have shown to for the combined treatment in comparison to the single agent (Po0.05). improve survival and life quality of patients with NSCLC as single agents in the second-line therapy. In contrast to p53 gene transfer, no synergism was found for the combination between bortezomib and the cytotoxic absence of p53. We found that bortezomib inhibits tumor drugs. The simultaneous treatment of NSCLC cells with cell growth in p53-negative and p53-positive NSCLC cell bortezomib and docetaxel as well as with pemetrexed did lines, suggesting p53-independent effects. After rAAV-2- not lead to an enhanced inhibition of cell growth. mediated p53 restoration in p53-negative cells, significant Moreover, we found a significant reduction of growth additive effects on growth inhibition in different assays inhibition and reduced apoptosis for the combination of and an increased induction of apoptosis were observed. bortezomib and chemotherapy, suggesting an antagonism The inhibition of p53 degradation by the proteasome between the drugs. Our results are consistent with the inhibition was reflected by an increased intracellular p53 observations by others, demonstrating that bortezomib concentration as assessed in Western blot analysis. failed to enhance docetaxel-induced apoptosis in human Interestingly, looking at the p53-positive NSCLC cell pancreatic tumor xenografts.36 Although the antitumoral line, the synergistic effects between p53 gene transfer and effects of pemetrexed are well studied in a variety of bortezomib were even more distinct. Enhancement of p53 clinical investigations,37 there are no data about the concentrations in p53 positive or negative tumor cells and effects of pemetrexed in combination with proteasome the treatment with bortezomib led to significant over- inhibition in NSCLCs. Pemetrexed, as a multitargeted additive effects with regard to inhibition of tumor cell antifolate, causes an arrest in the S phase of the cell growth and induction of apoptosis. This effect possibly cycle.38 Docetaxel is lethal towards cells in S phase by results from the intracellular increase of p53 and a lack of affecting centrosome organization39 whereas bortezomib 40 p53 degradation caused by the treatment with bortezo- causes a G2/M-phase arrest. Consequently, when mib. These results are in line with the finding that p53 bortezomib and a chemotherapeutic drug are adminis- rapidly accumulates in cells treated with proteasome tered simultaneously, the bortezomib-mediated G2/M inhibitors9,27 and underline the observation of p53- arrest prevents cells from entering the part of the cell mediated sensitization for proteasome inhibitor-induced cycle in which the chemotherapeutical agents develop apoptosis.16 p53 gene transfer was chosen for combina- their definitive lethal activity as observed by Nawrocki tion with bortezomib for several reasons. First, p53 plays et al.36 an essential intracellular role with regard to cell cycle Based on the data of our in vitro studies, the design of regulation, DNA repair and apoptosis.28 Second, muta- clinical trials using bortezomib and cytotoxic drugs in tions in the p53 gene are frequently observed in up to combination might be without improvement for the 60% of all NSCLC,29 playing a critical role in tumor patients. Although the antagonistic effects of bortezomib formation, progression and chemoresistance, thus leading in combination with chemotherapeutical agents have not to an inferior prognosis for the patient.30,31 Third, p53 been studied in xenograft models, there is evidence that associated apoptosis is regulated by proteasome degrada- our results are of relevance in the clinical setting. This is tion.17 For p53 gene transfer, an rAAV-2 vector was used exemplified by a clinical phase II study for patients with as we previously demonstrated that NSCLC cells are advanced NSCLC published at the ASCO in 2005. highly susceptible for rAAV-2.20,32 The results of previous Fanucchi et al.41 treated 155 patients with advanced clinical trials using p53 gene replacement therapy in NSCLC with bortezomib alone and in combination with

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 438 docetaxel. Bortezomib plus docetaxel was generally well lung cancer cell line. Cancer Chemother Pharmacol 2004; 54: tolerated with manageable toxicities but no advantage 343–353. for the patients’ outcome in terms of overall–response 15 Yang Y, Ikezoe T, Saito T, Kobayashi M, Koeffler HP, rate and survival was observed compared to bortezomib Taguchi H. Proteasome inhibitor PS-341 induces growth monotherapy. Consequently, a combination of docetaxel arrest and apoptosis of non-small cell lung cancer cells via or pemetrexed with bortezomib should be reconsidered in the JNK/c-Jun/AP-1 signaling. Cancer Sci 2004; 95: 176–180. 16 Chen F, Chang D, Goh M, Klibanov SA, Ljungman M. further clinical trials. Role of p53 in cell cycle regulation and apoptosis following exposure to proteasome inhibitors. Cell Growth Differ 2000; Acknowledgements 11: 239–246. 17 Lopes UG, Erhardt P, Yao R, Cooper GM. P53-dependent induction of apoptosis by proteasome inhibitors. J Biol This work was supported by the Research commission of Chem 1997; 272: 12893–12896. the Medical Faculty, University Duesseldorf and the 18 Ren J, Shi M, Liu R, Yang Q-H, Johnson T, Skarnes WC Leukaemie Liga eV. Duesseldorf. et al. The Birc6 (Bruce) gene regulates p53 and the mitochondrial pathway of apoptosis and is essential for mouse embryonic development. Proc Natl Acad Sci 2005; 102: 565–570. References 19 Mitsudomi T, Steinberg SM, Nau MM, Carbone D, D’Amico D, Bodner S et al. P53 gene mutations in non- 1 Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, small cell lung cancer cell lines and their correlation with the Ward E et al. Cancer statistics, 2004. CA Cancer J Clin 2004; presence of ras mutations and clinical features. Oncogene 45: 8–29. 1992; 7: 171–180. 2 Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, 20 Rohr UP, Wulf MA, Stahn S, Heyd F, Steidl U, Fenk R Krook J et al. Comparison of four chemotherapy regimens et al. Non-small lung cancer cells are prime targets for p53 for advanced non-small cell lung cancer. N Engl J Med 2002; gene transfer mediated by a recombinant adeno-associated 346: 92–98. virus type-2 vector. Cancer Gene Ther 2003; 10: 898–906. 3 Engels FK, Sparreboom A, Mathot RAA, Verweij J. 21 Rohr UP, Wulf MA, Stahn S, Steidl U, Haas R, Kronenwett Potential for improvement of docetaxel-based chemo- R. Fast and reliable titration of recombinant adeno- therapy: a pharmacological review. Br J Cancer 2005; 93: associated virus type-2 using quantitative real-time PCR. 173–177. J Virol Methods 2002; 106: 81–89. 4 Adjei AA. Pemetrexed (Alimta), a novel multitargeted 22 Ho¨ rster A, Teichmann B, Hormes R, Grimm D, Kleinsch- antineoplastic agent. Clin Cancer Res 2004; 10: 4276–4280. midt J, Sczakiel G. Recombinant AAV-2 harboring gfp- 5 Hanna N, Sheperd F, Rosell R, Pereira JR, De Marinis F, Fossella F et al. Randomized phase III trial of pemetrexed antisense/ribozyme fusion sequences monitor transduction, versus docetaxel in patients with locally advanced or gene expression and show anti-HIV-1 efficacy. Gene Therapy metastatic non-small cell lung cancer previously treated with 1999; 6: 1231–1238. chemotherapy. Am Soc Clin Oncol 2003; 22: 622. 23 Grimm D, Kern A, Rittner K, Kleinschmidt JA. Novel tools 6 Roth JA, Grammer SF. Gene replacement therapy for non- for production and purification of recombinant adeno- small cell lung cancer: a review. Hematol Oncol Clin N Am associated virus vectors. Hum Gene Therapy 1998; 9: 2004; 18: 215–229. 2745–2760. 7 Swisher SG, Roth JA, Nemunaitis J, Lawrence DD, Kemp 24 Zolotukhin S, Byrne BJ, Mason E, Zolotukhin I, Potter M, BL, Carrasco CH et al. Adenovirus-mediated p53 gene Chesnut K et al. Recombinant adeno-associated virus transfer in advanced non-small-cell lung cancer. J Nat purification using novel methods improves infectious titer Cancer Inst 1999; 91: 763–771. and yield. Gene Therapy 1999; 6: 973–985. 8 Maione P, Rossi A, Airoma G, Ferrara C, Castaldo V, 25 Rohr UP, Heyd F, Neukirchen J, Wulf MA, Queitsch I, Gridelli C. The role of targeted therapy in non-small cell lung Kroener-Lux G et al. Quantitative real-time PCR for cancer. Crit Rev Oncol Hematol 2004; 51: 29–44. titration of infectious recombinant AAV-2 particles. J Virol 9 Adams J. Development of the proteasome inhibitor PS-341. Methods 2005; 127: 40–45. Oncologist 2002; 7: 9–16. 26 Vayssade M, Haddada H, Faridoni-Laurens L, Tourpin S, 10 Ludwig H, Khayat D, Giaccone G, Facon T. Proteasome Valent A, Benard J et al. p73 functionally replaces p53 in inhibition and its clinical prospects in the treatment of adriamycin-treated, p53-deficient breast cancer cells. Int J hematologic and solid malignancies. Cancer 2005; 104: Cancer 2005; 116: 860–869. 1794–1807. 27 Richardson PG, Hideshima T, Anderson KC. Bortezomib: a 11 Scagliotti G. Proteasome inhibition in lung cancer. Crit Rev novel, first-in-class proteasome inhibitor for the treatment of Oncol Hematol 2006; 58: 177–189. multiple meyloma and other cancers. Cancer Control 2005; 12 Adams J. The proteasome: structure, function and role in the 10: 361–369. cell. Cancer Treat Rew 2003; 29: 3–9. 28 Campling BG, El-Deiry WS. Clinical implication of p53 13 Ling Y-H, Liebes L, Jiang J-D, Holland JF, Elliott PJ, mutation in lung cancer. Mol Biotech 2003; 24: 141–156. Adams J et al. Mechanisms of proteasome inhibitor PS-341- 29 Hainaut P, Hernandez T, Robinson A, Rodriguez-Tome P, induced G2–M-phase arrest and apoptosis in human non- Flores T, Hollstein M et al. IARC database of p53 gene small cell lung cancer cell lines. Clin Cancer Res 2003; 9: mutations in human tumors and cell lines: updated compila- 1145–1154. tion, revised formats and novel visualisation tools. Nucleic 14 Mortenson MM, Schlieman MG, Virudachalam S, Bold RJ. Acids Res 1998; 26: 205–213. Effects of the proteasome inhibitor bortezomib alone and in 30 Levine AJ. P53, the cellular gatekeeper for growth and combination with chemotherapy in the A549 non-small cell division. Cell 1997; 88: 323–331.

Cancer Gene Therapy Effects of bortezomib, p53 and chemotherapy on NSCLC cells J Neukirchen et al 439 31 O’Connor PM, Jackman J, Bae I, Myers TG, Fan S, Mutoh of a multicenter phase II study. J Clin Oncol 2001; 19: M et al. Characterization of the p53 tumor suppressor 1750–1758. pathway in cell lines of the national cancer institute 36 Nawrocki ST, Sweeny-Gotsch B, Takamori R, McConkey anticancer drug screen and correlation with the growth DJ. The proteasome inhibitor bortezomib enhances the inhibitory potency of 123 anticancer agents. Cancer Res activity of docetaxel in orthotopic human pancreatic tumor 1997; 57: 4285–4300. xenografts. Mol Cancer Ther 2004; 3: 59–70. 32 Rohr UP, Kronenwett R, Grimm D, Kleinschmidt J, Haas 37 Kut V, Patel JD, Argiris A. Pemtrexed: a novel antifolate R. Primary human cell differ in their susceptibility to rAAV- agent enters clinical practice. Expert Rev Anticancer Ther 2-mediated gene transfer and duration of reporter gene 2004; 4: 511–522. expression. J Virol Methods 2002; 105: 265–275. 38 Tonkinson JL, Marder P, Andis SL, Schultz RM, Gossett 33 Swisher SG, Roth JA, Komaki R, Hicks M, Ro J, Dreiling L LS, Shih C et al. Cell cycle effects of antifolate antimetabo- et al. A phase II trial of adenoviral mediated p53 gene lites: implications for cytotoxity and cytostasis. Cancer transfer (RPR/INGN 201) in conjunction with radiation Chemother Pharmacol 1997; 39: 521–531. therapy in patients with localized non-small cell lung cancer 39 Hennequin C, Giocanti N, Favaudon V. S-phase specificity (NSCLC). Am Soc Clin Oncol 2000; 19: 461a. of cell killing by docetaxel (Taxotere) in synchronized HeLa 34 Nemunaitis J, Swisher SG, Timmons T, Connors D, cells. Br J Cancer 1995; 71: 1194–1198. Mack M, Doerksen L et al. Adenovirus-mediated p53 40 Ling Y-H, Liebes L, Ng B, Buckley M, Elliott PJ, Adams J gene transfer in sequence with cisplatin to tumors of patients et al. PS-341, a novel proteasome inhibitor, induces Bcl-2 with non-small cell lung cancer. J Clin Oncol 2000; 18: phosphorylation and cleavage in association with G2 Mphase 609–622. arrest and apoptosis. Mol Cancer Ther 2002; 1: 841–849. 35 Schuler M, Herrmann R, De Greve JL, Stewart AK, 41 Fanucchi MP, Fossella F, Fidias P. Bortezomib7docetaxel Gatzemeier U, Stewart DJ et al. Adenovirus-mediated in previously treated patients with advanced non-small cell wild-type p53 gene transfer in patients receiving chemo- lung cancer (NSCLC): a phase 2 study. Am Soc Clin Oncol therapy for advanced non-small cell lung cancer: results 2005; Abstract No: 7024.

Cancer Gene Therapy