Local anesthetic bupivacaine induced ovarian and prostate cancer apoptotic cell death and underlying mechanisms in vitro

Retrospective studies indicate that the use of regional anesthesia can reduce cancer recurrence after surgery which could be due to ranging from immune function preservation to direct molecular mechanisms. This study was to investigate the effects of bupivacaine on ovarian and prostate cancer cell biology and the underlying molecular mechanisms. Cell viability, proliferation and migration of ovarian carcinoma (SKOV-3) and prostate carcinoma (PC-3) were examined following treatment with bupivacaine. Cleaved caspase 3, 8 and 9, and GSK-3β, pGSK-3βtyr216 and pGSK-3βser9 expression were assessed by immunofluorescence. FAS ligand neutralization, caspase and GSK-3 inhibitors and GSK-3β siRNA were applied to further explore underlying mechanisms. Clinically relevant concentrations of bupivacaine reduced cell viability and inhibited cellular proliferation and migration in both cell lines. Caspase 8 and 9 inhibition generated partial cell death reversal in SKOV-3, whilst only caspase 9 was effective in PC-3. Bupivacaine increased the phosphorylation of GSK-3βTyr216 in SKOV-3 but without measurable effect in PC3. GSK-3β inhibition and siRNA gene knockdown decreased bupivacaine induced cell death in SKOV-3 but not in PC3. Our data suggests that bupivacaine has direct ‘anti-cancer’ properties through the activation of intrinsic and extrinsic apoptotic pathways in ovarian cancer but only the intrinsic pathway in prostate cancer.

Bupivacaine on cancer cell apoptosis. Caspase 3,8 and 9 were activated in SKOV-3 following 1 mM bupivacaine treatment at 24 hours ( Fig. 2A-C), with caspases 3 and 9 being cleaved in PC-3 ( Fig. 2D-F). Cleaved caspase 3 expression through western blot were both elevated in two cancer cell lines after the treatment of 1mM bupivacaine (Fig. 2G). Caspase 9 inhibitor partially reversed bupivacaine induced SKOV-3 and PC-3 cell death (Fig. 3B), whilst Caspase 8 inhibition was effective in SKOV-3 only (Fig. 3A). Cytotoxicity was independent from FAS receptor activity as FAS ligand neutralization antibody treatment did not yield significant results as detected by flow cytometry (Fig. 3C,D).
Bupivacaine on cancer cell migration. Wound healing assay was utilized to investigate the effects of bupivacaine on the migration potential of both cell lines. Following 24 hour bupivacaine treatment, fewer cells migrated towards the scratch midline in 1 mM bupivacaine treated group when compared to the control groups (Fig. 5A,B). Bupivacaine reduced migration potential of PC-3 cells by up to 60% (Fig. 5B). Lower concentrations of bupivacaine did not impact on cancer cell migration (data not shown). Bupivacaine on GSK-3β expression. Varying levels of increased expression in total GSK-3β , pGSK-3β tyr216 and pGSK-3β ser9 were demonstrated using immunofluorescence. In SKOV-3 cells, baseline levels of total GSK-3β and pGSK-3β tyr216 were similar and almost doubled following bupivacaine treatment when compared with control (Fig. 6A,B). The baseline level and elevation of pGSK-3β ser9 was relatively lower (Fig. 6C). With reference to the PC-3 cell line following bupivacaine treatment, similar basal expression levels of GSK-3β , pGSK-3β tyr216 and pGSK-3β ser9 were observed and these were not statistically significant ( Fig. 6D-F).

Discussion
This study indicates that bupivacaine possesses cytotoxic, anti-proliferative and anti-metastatic properties in both ovarian and prostate cancer cell lines. Following 24 hours treatment, 1mM bupivacaine induced similar levels of cytotoxicity in both cell lines. Cell death was more pronounced after treatment for 72 hours. This is contrasted by the lack of statistical significant cell death exhibited by HK-2 cells when treated with 1mM bupivacaine. This may indicate different cell lines (cancerous vs non-cancerous) have varying levels of sensitivity to bupivacaine. Furthermore, an synergistic effect was observed when bupivacaine was combined with the chemotherapy agent taxol. Data is limited on potential mechanisms, but as shown in (Fig. 9) the activation of the intrinsic and extrinsic apoptotic pathways in conjunction with the active form of GSK-3β are likely to be involved (Fig. 9). Data indicates there is an association between changes in cellular metabolism and the rate of cellular proliferation in cancer cells 16 . Beitner et al. reported that LA reduced melanoma cells glycolysis and ATP levels by downregulating two allosteric stimulatory signal molecules 17 . Lucchinetti et al. suggested that LA inhibited mesenchymal stem cells (MSC) proliferation 18 . Ki-67, a key marker of proliferation 19 , was used in our study to examine the effects of LA on cellular proliferation. Our results show decreased Ki67 expression in both SKOV-3 and PC-3 cell lines treated with 1 mM bupivacaine.
Lucchinetti et al. 18 reported that 100 μ M ropivacaine significantly inhibited mesenchymal stem cell migration as measured by wound healing assay. Our study demonstrates that 1 mM bupivacaine inhibits the migration potential of SKOV-3 and PC-3. Greater inhibition was observed in PC-3. Data indicates that prostate cancer exhibits higher levels of metastatic potential 20 . This may suggest that the anti-migration properties of LA may have a greater impact on more invasive tumors.
Reactive oxygen species (ROS) activity was also determined in the current study. It has been reported that increased ROS levels suppresses breast cancer cell proliferation 21 , whilst opposite finding also exists is that antioxidants can inhibit liver cancer cell proliferation 22 . Our study demonstrates increased ROS levels in SKOV-3, whilst decreased ROS levels in PC-3. These results are in keeping with the contradictory role ROS activity appears to have in cancer cell growth/proliferation 21,22 and requires further investigation.
Previous studies indicate that LA induced cancer cell death is caspase dependent 23 . Werdehausen et al. 24 used gene modulation techniques and demonstrated that low concentrations of lidocaine induced Jurkat cell death via the intrinsic apoptotic pathway. We found that caspase 3, 8 and 9 were activated in SKOV-3 following 1 mM bupivacaine treatment while caspase 3 and 9 were cleaved in PC-3. Furthermore, partial cell death inhibition was observed with caspase 9 inhibition in both cell lines. Caspase 8 inhibition was only effective in SKOV-3. It has been noticed that mutations in tumor suppressor genes; such as BRCA1 in ovarian cancer, have a significant function in DNA damaging related apoptosis in cancer chemotherapy 25 . Local anesthetics have also demonstrated to be potent DNA damaging agents. Kim et al. demonstrated that dibucaine induced DNA fragmentation and chromatin condensation in neuroblastoma cells 26 . It is to be determined if genomic defects increase the cytotoxic profile of bupivacaine in ovarian and prostate cancer.
The caspase cascade involves numerous cellular processes; of which FAS ligand activity is implicated. This study indicates FAS ligand receptor activity is not involved in bupivacaine interactions with SKOV-3 or PC-3. It has been reported that caspase 8 activation not only occurs via FAS/death receptor ligand 27 , but also via FAS  ligand independent caspase 8 induced cell death 28 . This suggests different cancer cells exhibit varying biological profiles which influence molecular signal transduction processes involved in growth, development and death.
Glycogen synthase kinase-3β (GSK-3β ) is a serine/threonine kinase which is implicated in numerous cell functions including cell differentiation, proliferation and apoptosis. GSK-3β leads cell apoptosis via the interaction with proapoptotic transcription factor p53 as its regulatory protein, and it also cause apoptosis by inducing mitochondrial injury and the caspase cascade 29,30 . Phosphorylation at its tyrosine residue (tyr-216) constitutes its active form whereas serine residue (ser-9) phosphorylation is its inactive form 31 . The role of GSK3β in tumor development is controversial. Previous studies have shown that GSK3β impaired tumor growth in several cancer cell lines 32,33 . However, Cao et al. 34 reported that the suppression of kinase inactive form GSK3β ser9 promoted ovarian cancer development, which indicated GSK3β is also necessary for tumor survival. Furthermore, another study showed that the suppression of Src-GSK3β axis could be a new target to treat prostate cancer 35 . GSK3β interactions with chemotherapy agents are complicated. There is increasing evidence which indicates that GSK3β activity modulates the effectiveness of chemotherapy on cancer cells. Downregulation of GSK3β expression level conferred resistance of ovarian cancer cells from cisplatin treatment 36 . In hepatoblastoma cell lines, GSK3β inhibition by pharmacological or gene knockdown/mutant techniques limited anti-cancer drug induced apoptosis 37 . Consistent with these findings, our study demonstrates GSK3β expression is essential for bupivacaine induced cell death. Total GSK3β , pGSK-3β ser9 and pGSK-3β tyr216 were all elevated in SKOV-3 cells following 24 hours of treatment with 1 mM bupivacaine. Greater levels of expression were observed in pGSK-3β tyr216 , the active form of GSK3β , when compared with the inactive form pGSK-3β ser9 . Previous reports have primarily focused on GSK-3β ser9 or GSK-3β tyr216 expression in isolation. Our findings indicate that whilst both were activated, an overall increase in the expression of GSK-3β is observed. Statistically significant changes in expression of GSK3β, pGSK-3β ser9 and pGSK-3β tyr216 were not observed in the PC-3 cell line. This indicates that bupivacaine induced prostate cancer cell death is unlikely to involve GSK3β activity.
To further examine GSK-3β activity and the relative expression of its residues, GSK3β inhibition demonstrated the partial suppression of cell death in bupivacaine treated SKOV-3 cells. This suggests GSK3β is pro-apoptotic in bupivacaine induced cell death. To investigate potential interactions between GSK3β and caspase activity in apoptosis, we demonstrated that caspase 3, 8 and 9 were down regulated following GSK3β siRNA treatment, which reaffirmed the hypothesis that GSK3β deactivation in the SKOV-3 confers resistance to bupivacaine induced cell death.
In summary, our findings suggest that bupivacaine has direct 'anti-cancer' properties in vitro. However, our work is not without limitations. Firstly, these in vitro experiments do not fully replicate an in vivo or clinical environment and thus warrant further study. Secondly, the concentrations of bupivacaine (up to 1 mM) tested in this study may not be applicable in particular clinical contexts. LA concentrations vary according to their mode of delivery. The concentration of LA on direct infiltration has been reported to reach 500 μ M. It can be even higher when administered topically 38 . An extension of this even though this is beyond the scope of this study is to acknowledge that the cancer microenvironment is complex and this in itself is an important determinant in tumor growth and metastasis. It has been shown that in the presence of TNF-alpha, a low dose of LA suppressed TNF-alpha induced ICAM-1 phosphorylation which is associated with LA anti-migration properties 14 . Finally, regional anesthesia combined general anesthesia with inhalational agents are often used in cancer patients 39,40 . The potential interaction between local anesthetics and inhalational anesthetics on cancer cell biology has not been investigated in this study. Interestingly, the survival rate from cancer recurrence is much higher with regional anesthesia combined with general anesthesia than with general anesthesia alone 9,41 . These observations are discussed in recently published literature 42,43 . They indicate that inhalational anesthetics (e.g. isoflurane) may promote cancer cell malignancy in vitro. This would require further investigation to establish if they hold true in vivo and ultimately in clinical practice. Nevertheless, the data reported here clearly demonstrates that the local anesthetic bupivacaine can directly "kill" cancer cells through mechanisms not explored before in this context, which is likely to stimulate further research and increase the call for more clinical trials to be conducted.

Materials and Methods
Cell Culture. Human ovarian carcinoma (SKOV-3), prostate carcinoma (PC-3) and human proximal tubular cell (HK-2) lines (all purchased from European cell culture collection, Salisbury, UK) were used for this study. SKOV-3 were cultured in McCoy's 5A medium (Sigma Aldrich, St. Louis, USA) and PC-3 and HK-2 were cultured in RPMI-1640 medium (Sigma Aldrich, St. Louis, USA). Culture mediums were supplemented with 10% newborn calf serum (HyClone, Auckland, New Zealand) with 1% L-glutamine and 1% penicillin-streptomycin (Sigma Aldrich, St. Louis, USA). Cells were maintained at 37 °C under a humidified atmosphere of 5% CO 2 and 95% air in an air jacket incubator (Triple Red, Buckinghamshire, UK). All cells were cultured in 24 wells plate and used for experiment when reach 70-80% confluence.
Cells were cultured with bupivacaine at concentrations ranging from 1 μ M to 1 mM for 24 or 72 hours. Other cohort cultures were treated with 100 μ M/1 mM bupivacaine or together with 100 nM chemotherapy drug taxol    Cell viability and death measurement. Cell viability was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) (EMD Chemicals, San Diego, CA) assay reported elsewhere 45 . MTT was dissolved in Opti-MEM (Gibco, Paisley, UK) to form 0.5 mg/ml working solution. Culture medium was removed after the incubation with or without indicated concentrations of LA and taxol, then 500 μ L MTT working solution was added into each well and incubated for 4 hours. Following this, supernatants were aspirated and 500 μL DMSO (Fisher Scientific, Leicestershire, UK) was added to dissolve the formazan crystals. In 96 well plates, absorbance was measured at 595 nm using micro plate reader analysis (Dynex technologies, Chantilly, VA, USA). Cell viability relative to the control was calculated and expressed as relative to control.
For further clarification, Propidium iodide (PI) (Sigma Aldrich, St. Louis, USA) staining was used to examine cell death as described previously 46 . Cells were harvested in a FACS tube and washed twice before re-suspension in FACS buffer. PI was added to make the final concentration to 1 μ g/ml and incubated in dark for 5 min. Cells were not washed then PI fluorescence was detected using flow cytometry (FL-2 channel). Single color positive cells were defined as dead cells.
Wound healing assay. Capability of migration was evaluated by wound healing assay reported previously 47 .
Briefly, cells were cultured in a 60 mm petri dish to form a confluent monolayer. Then, a wound was scratched on this monolayer by 1 ml pipet tip and washed twice with culture media. Before taking each image, a mark was made at the bottom of the dish to make sure that all the images were taken at the same site. After another 24 hours incubation with or without various concentrations of bupivacaine, the second image was taken at the same site. The wound healing status was compared by Image-Pro Plus software (Media Cybernetics, USA) based on these images. The results were presented as (initial wound area − remaining wound area)/(initial wound area).
Western blot. Cell samples were homogenized in lysis buffer. Cell lysates were centrifuged with the supernatant being collected. Then protein concentration was quantified in the supernatant by Bradford protein assay (Bio-Rad, United Kingdom). Protein extracts (40 μ g per sample) were heated, denatured, and loaded on a NuPAGE 4 to 12% Bis-Tris gel (Invitrogen, USA) for electrophoresis and transferred to a polyvinylidene difluoride membrane. The membrane was blocked with 5% non-fat milk for 1 h at room temperature and then probed with rabbit anti-cleaved caspase-3 and mouse anti-GSK-3β (1:1000; Santa Cruz, USA) primary antibody in tris-buffered saline and Tween-20 overnight at 4 °C, followed by the application of goat anti-rabbit or goat anti-mouse horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. The loading control was protein GAPDH (1:10,000; Millipore, USA). The blots were detected with enhanced chemiluminescence system (Santa Cruz, USA) and analyzed with GeneSnap (Syngene, United Kingdom). Protein band intensity was normalized with GAPDH and expressed as a ratio of control. Statistical analysis. All data were expressed as Mean ± SD. Data were analyzed by one-way analysis of variance, followed by the post hoc Student-Newman-Keuls test (GraphPad Prism 5.0 software, San Diego, CA) for comparison with Bonferroni corrections. A p value < 0.05 was considered to be statistically significant.