Bacillus Calmette Guerin Treatment Of Bladder Cancer Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The majority of Bladder Cancer cases are Transitional Cell Carcinomas (TCC). Treatment with Bacillus Calmette-Guérin (BCG) bladder instillations is an established treatment modality for superficial urinary bladder cancer and carcinoma in situ (CIS) but only 70% of patients respond to treatment. The exact mechanism of how BCG induce tumour remission is still unknown. Earlier reports put forward the idea that BCG induces an inflammatory response which activates macrophages resulting in stimulation of cytotoxic factors. Nitric Oxide (NO) is believed to be responsible for the BCG-mediated cytotoxic effect. Recent evidences suggest that a mutual cross-talk exists between NOS2 and COX-2 in cancer. The present thesis was aimed to investigate NOS2 up-regulation in BCG-treated macrophages by inhibiting COX-2 using Celecoxib. As analyzed by NO measures and western blot, COX-2 inhibition had a dual impact on NO released in BCG-treated Raw cells. Difference in Raw cell viability after BCG-treatment was seen between the XTT and Trypan blue assays. Since XTT test relies on mitochondrial activity of cells, Immunofluorescence staining was performed to investigate the problem. Staining results confirm increased mitochondrial activity in BCG-treated macrophages. This could explain why increased cell viability was seen with the XTT assay and not with Trypan blue inclusion. Thus, the molecular pathway between NOS2 and COX-2 should be explored further to investigate the role of NO in BCG immunotherapy.


Celecoxib, macrophages, CIS, BCG, NO, NOS2, COX2,

1. Introduction

1.1 Urinary Bladder Cancer

Urinary Bladder Cancer is the ninth most common cancer worldwide and sixth most common in Sweden, accounting for a higher incidence in males when compared to females in a ratio of 3:1 (Parkin, 2008). Next to prostate cancer, carcinoma of the urinary bladder is the predominant malignancy of the urinary tract. The incidence rate increases with age and the average age of onset is 70 years. The Pathophysiology of Bladder Cancer suggests that the most prominent histological type is the Transitional cell carcinomas (TCC) accounting for more than 90%, proceeded by 5% of squamous cell carcinomas (SCC) and less than 2% of adenocarcinomas (Parkin, 2008). Bladder cancer seems to be more prevalent in Western European countries with an elevated frequency of TCC related to cigarette smoking (Parkin, 2008). High prominence of squamous cell carcinoma is recorded in countries such as Middle East and North Africa where urinary schistosomiasis is endemic (Jemal et al., 2011). Schistosomiasis is caused by Schistosoma Haematobium, which moves and infects the bladder. The other common risk factors that contribute to the development of the disease include chemical and environmental exposures, diet and genetic polymorphism (Preslan and Joshi, 2011). TCC also known as Urothelial cell carcinoma (UCC) rises from the epithelial tissue lining the inner surface of the organs in the urinary system. The standardized treatment for these carcinomas depends on the type of cancer. The tumours in the urothelium have been graded into four sub types, by World Health Organization (WHO) and International Society of Urological Pathology based on the degree of nuclear anaplasia, namely papilloma, papillary urothelial neoplasm of low malignant potential, low grade and high grade carcinoma (Epstein et al., 1998).

1.2 Staging of Bladder tumours

In order to study the extent of tumour spread in the bladder wall, TNM (Tumour Node Metastasis) classification is used for staging along with World Health Organization (WHO) histological grading systems. Approximately, 75-80% of the bladder cancer cases are superficial tumours or early bladder cancer which includes Non-muscle invasive bladder cancer (NMIBC) confined to mucosa (stage Ta), submucosa (stage T1) (fig-1) (Knowles, 2006). Deep or invasive bladder cancer is of high risk where the cancer cells would have spread to muscle and fat layer (T2 and T3 respectively) and in metastatic bladder cancer the tumour spreads outside the bladder and invades other parts of the body (T4). CIS is a flat, high grade surface spreading lession confined to the innermost lining of bladder. It is the primary diagnosis in 1-4% of all bladder cancers and is associated with papillary tumours in 13-20% of bladder cancer patients (Babjuk et al., 2011).

Figure 1: T-classification for Bladder Cancer. Staging of bladder tumours by TNM system. Adopted from (Knowles, 2006).

1.3 Various treatments for Bladder Cancer

Although patients with invasive bladder cancer are treated by surgery, chemotherapy, radiation therapy or through a combined modality of the above three therapies, around 70% of patients die during the first five year follow-up period (Barocas et al., 2012). The choice of therapy depends on the staging and grading of cancer, patient's age and physical condition, etc. CIS, commonly treated by either radical cystectomy or BCG, in severe cases has a high recurrence rate and notably 50% of untreated CIS patients develop invasive cancer within 5 years (Lamm et al., 1980). Immunotherapy with Bacillus Calmette-Guérin (BCG) instillations is a standard treatment for high risk superficial tumours and is directly administered to the bladder via the urethra. BCG is especially recommended for patients with primary CIS and tumours with intermediate- to high-risk stage progression (Gontero et al., 2010).

1.4 History of BCG

BCG vaccine, an attenuated live substrain of Mycobacteria Bovis was identified by Albert Calmette and Camille Guerin in 1921, when they were trying to develop a vaccine against Tuberculosis(TB) (Calmette et al., 1927). Studies suggest that BCG-treated tuberculosis patients had less cancer incidence which led to an interesting notion that BCG has anti-tumour response (Pearl, 1929). This was followed by experimental research on BCG to demonstrate its effects on the immune response of the patient and remission of tumours in rats and mice (Vitale and Allegretti, 1963). Meanwhile, the introduction of chemotherapy and radiation therapy reduced the interest of using BCG as an anti-cancerous agent. But further successful results about the strong delayed hypersensitivity to BCG immunization in Guinea pig bladder, again aroused attention towards efficacy of BCG (Coe and Feldman, 1966). In 1976, Morales et al., performed the intravesical administration of BCG clinically and showed tremendous decrease in recurrence rate of superficial bladder tumours. The above said observations were again confirmed (Lamm et al., 1980) and ever since, a standardized treatment regimen of BCG (as described by Morales et al.,) is followed in the treatment of superficial transitional cell urinary bladder cancer. Although BCG is regarded the most successful immunotherapy for bladder cancer to date, its usage is limited due to its high side effects and the exact mechanisms behind its ability to induce tumour remission still remains largely unknown(Martin and Kamat, 2009). Hence to control the adverse side effects, major efforts have been put by researchers to illustrate the mechanism of BCG through which it eradicates tumour to identify non-responders before treatment and develop successful therapies without side effects(Martin and Kamat, 2009).

1.5 BCG treatment for Bladder Cancer

Prior studies suggest that macrophages actively mediate part of the cancer removal induced by BCG in tumours (Klostergaard et al., 1991). Brandau et al ., has detailed about the sequential events that occurs inside the bladder after instillation of BCG (Fig-2) (Brandau and Suttmann, 2007). Initially, the attenuated mycobacteria (BCG) are internalized in the urothelial cells after adhering to fibronectin attachment protein, present in the walls lining bladder. As an effect, there is a major outflow of proinflammatory cytokines from the urothelial cells such as IL-6, IL-8, tumour necrosis factor (TNF-α) which subsequently activates the innate immune cells including neutrophils and macrophages leading to release of more cytokines and chemokines. This outcome produces a strong non-specific inflammatory reaction with Th-1 response and activation of cytotoxic T cells and natural killer cells (NK) along with macrophages finally destroying bladder tumour cells (Luo and Knudson, 2010). It has also been demonstrated that a strong establishment and expression of Th-1 response plays a major role in enhancing the anti-tumour response of BCG and the elimination of tumour cells is contributed by lymphocyte subsets CD4+, CD8+ T cells and NK cells (Riemensberger et al., 2002). It is evident that a complex cross-talk interaction between innate and adaptive immunity is involved in the mechanism of BCG, right after its intravesical administration into bladder lumen. Thus the immunocompetency of the host plays a major role for an effective BCG immunotherapy. Other significant factors include less number of cancer cells and close proximity between host (tumour) and antigen (BCG) in order to develop an immune response (Morales et al., 1976).

Figure 2: Effect of BCG on urothelial cells. Induction of BCG into the urothelial cells produces proinflammatory cytokines which in turn attracts neutrophiles and macrophages resulting in activation of cytotoxic T-cells and NK-cells finally eradicating the tumour. Adopted from (Brandau and Suttmann, 2007).

The optimum choice of therapy used in patients with intermediate risk of NMIBC is an immediate induction course of BCG with weekly instillation for six weeks after chemotherapy followed by a minimum 1 year follow up (Babjuk et al., 2011). Although there is a tremendous success in 70% of BCG-treated NMIBC patients, recent data mentions 30-35% patients either fail to respond to treatment or the disease recurs within five-years of clinical follow up (Barocas et al., 2012). As abovesaid, this may be due to the complexity in the immune response of BCG associated with release of pro-inflammatory cytokines (Ratliff et al., 1987). It is proven that several of the cytokines found in urine from BCG treated patients are able to stimulate Nitric Oxide (NO) synthesis and elevated NO concentrations have also been measured in human urinary bladder after treatment with BCG (Jansson et al., 1998). These results suggest that NO may play an important role in the BCG mediated anti-tumour effect since NO is considered to be one of the main factors responsible for the cytotoxic activity that macrophages exert after BCG instillation (John et al., 1988).

1.6 Nitric oxide synthases

NO is produced by three different isoforms of nitric oxide synthases (NOS) derived from three separate genes out of which two are constitutively expressed (cNOS) in most cells known as neuronal NOS (nNOS/NOS1) and endothelial NOS(eNOS/NOS3), named after cells where they were first found. Since their activation is calcium and calmodulin dependent which depends upon various physiological stimuli, they produce low levels of NO. The third isoform, inducible NOS (iNOS/NOS2) is tightly bound to calmodulin which makes it calcium independent resulting in the production of large amounts of NO when compared to cNOS (Knowles, 2006). NOS catalyses the conversion of L-arginine to L-citrulline in the presence of molecular oxygen and NADPH, which produces NO, which in turn reacts with guanylate cyclase (sGC) to form cyclic guanosine-3', 5'-monophosphate(cGMP) (Huwiler and Pfeilschifter, 2003). The NO-cGMP signalling promotes the physiological activities mediated by NO e.g. smooth muscle relaxation, neurotransmission, development of nervous system, angiogenesis, inhibition of platelet aggregation and adhesion (Knowles, 2006). Bladder cancers consist of tumour cells which express NOS2 and NO (Koskela et al., 2012), at low concentrations promotes cell growth whereas it exhibit cytostatic and cytotoxic effects in high concentrations (Morcos et al., 1999). Hence the effect of NO in tumour biology depends on tumour stage, activity and levels of NO.

1.7 Role of NO in the BCG treatment of Bladder Cancer

Some bladder tumours show increased nitric oxide (NO) activity per se and patients with superficial, high-risk bladder tumours have highly elevated intra-bladder NO levels after BCG treatment (Koskela et al., 2012). These results go in line with the increased NOS2 gene expression at transcriptional and protein level in patient tumours observed after BCG-treatment (Hosseini et al., 2006). Interestingly, NOS2 protein expression was not detected in bladder cancer cells after BCG-treatment in vitro, however when the same cells were treated with supernatant from BCG treated macrophages they were able to upregulate NOS2 protein. These data also correspond to previous results which states that NOS2 was detected in activated macrophages as a response to inflammatory cytokines after BCG treatment (Bohle et al., 1990). Taken together the above results suggest that NO might play a crucial role in the cytotoxic activity that macrophages exert on tumour cells after BCG-treatment (John et al., 1988).

In order to identify the additional factors needed for NOS2 upregulation after BCG treatment of urinary bladder cancer cells, an array experiment was performed in macrophages after BCG treatment. The expression of 96 "immune genes" was examined and 18 genes showed the same expression pattern as NOS2.These genes were considered as a starting point for further examination and one of these genes was COX-2 (PTGS2).

1.8 Role of Cyclooxygenases

Cyclooxygenases (COX) is an enzyme responsible for formation of prostanoids, containing prostaglandins, prostacyclins and thromboxanes which plays an important role in inflammatory response (Williams et al., 1999). COX exists in two isoforms, COX-1 and COX-2. Both isoforms possess the ability to transform arachidonic acid (polyunsaturated fatty acids) into cyclic endoperoxide prostaglandin H2 (PGH2). This occurs by peroxidation and a reductase reaction finally generating biological mediators such as prostaglandin E2 (PGE2) resulting in pain and inflammation (Minghetti, 2004). Prostaglandins (fatty-acid derivatives), beside their well known effects in inflammation and the immune response, also involved in various processes such as ovulation, blood clotting, renal function, differentiation of immune cells, nerve growth and bone metabolism (Fortier et al., 2008). Nonsteroidal anti-inflammatory drugs (NSAIDS) are used to treat inflammation and pain by directly inhibiting COX activity which in turn prevents prostaglandin production. COX-1 is constitutively expressed in most tissues and mediates many physiological functions e.g. maintaining gastric mucosa layer, regulating platelet function and renal blood flow by producing prostaglandins (Perrone et al., 2010). On the contrary, COX-2 is a highly regulated isoform and is readily inducible during inflammation by effect of proinflammatory cytokines, lipopolysaccharides and growth factors (Xuan et al., 2003). High PGE2 is produced only by COX-2 in inflammatory sites when compared to COX-1 which controls the formation of the prostaglandins involved in the normal function of many of our body's organs. Early NSAIDS were non-selective inhibitors which blocked both COX-1 and COX-2 identically. Since blockage of COX-1 results in severe side effects such as ulcers and kidney problem, a selective inhibitor of COX-2 was in need to prevent NSAID toxicity (Perrone et al., 2010).

1.9 Cross-talk between COX-2 and NOS-2

Several studies have revealed that the role of COX-2 in inflammatory response is dependent on nitric oxide pathways (Bansal et al., 2009). Earlier studies report that both the COX-2 and NOS2 support tumour growth by promotion of metastasis, invasion and angiogenesis (Cianchi et al., 2005). As both COX-2 and NOS2 are inducible isoforms, they both are known to be co-regulated in many ways when they are stimulated by proinflammatory cytokines (Rahman et al., 2001). In inflammation, NOS2 produces NO that enhances the activation and production of COX-2 which in turn produces PGE2 (Cianchi et al., 2005). It has also been suggested that PGE2 and NO produce cyclic nucleotide effectors, cyclic AMP(cAMP) and cyclic GMP(cGMP) that trigger the inflammatory system (Cianchi et al., 2005). On the contrary, another study suggest the inhibitory effects of NO (at high levels) on COX-2 (Swierkosz et al., 1994). Thus, both the positive and negative effects of NO on COX-2 suggests that the interactions depend on NO levels and cell type.

Figure 3: Proposed model of NOS2 mediated COX-2 signaling pathway. The inflammatory agents stimulate NOS2, which in turn induces both COX-2 and Notch-1 expression by p38 MAPK, JNK 1/2 pathway and JNK1/2 pathway respectively. Adopted from (Ishimura et al., 2005).

The above figure depicts that iNOS may not only mediate COX-2 through the p38 MAPK and JNK1/2 pathways to promote tumorigenecity in tumour cells, but that also induces Notch-1 expression (Ishimura et al., 2005). But our interest is to focus on the COX-2 and NO interaction in macrophages. Others have shown stimulation of COX-2 expression using NO donors in Raw macrophages (Bansal et al., 2009). Another interesting fact is that it has also been proved that a positive feedback loop exists between COX-2 and NO, with reports revealing that COX-2 products may also modulate the expression of NO and vice-versa (Marotta et al., 1992). Hence, it is very important to unravel the underlying mechanisms between COX and NO synthase pathway since their co-ordinated interaction is a key factor in pathophysiology of inflammation.

1.10 Mode of action of celecoxib

Inhibition of COX-2 is an effective cancer therapy since COX-2 is overexpressed in several human cancer types (Wu, 2006). Recent studies illustrate the anti tumour effects of a selective COX-2 inhibitor-celecoxib, a non-steroidal anti-inflammatory drug (Dhawan et al., 2008). Celecoxib reduces the synthesis of PGE2 by inhibiting COX-2 activity resulting in suppression of events involved in carcinogenesis. Previous evidences suggest that celecoxib blocks activation of P38 MAP kinase, AP-1 transcription factor (Chun et al., 2004) and also inhibits IKK, AKT activation thereby inhibiting activation of NF-kβ transcription factor (Shishodia et al., 2004) finally resulting in downregulation of expression of COX-2 gene.

1.11 Aims

The aims of the current study were to evaluate:

The stability of the drug celecoxib, previously not used in the current laboratory.

How toxic celecoxib is to macrophages, since celecoxib has different effects on different cell lines.

If celcoxib inhibits COX-2 protein expression in Raw 264.7 macrophages.

The effects of co-treatment with BCG and celecoxib.

The effect of COX-2 inhibition (by celecoxib) on NO release and NOS2 up-regulation after BCG-treatment of macrophages.

2. Results

2.1 Estimation of optimum Raw cell density

Raw cells were used in the study, since it has given promising results with its co-treatment with BCG in earlier studies. Initially, to determine the optimum cell density of Raw cells, a wide range of different cell concentrations were tried in the XTT viability assay. In fig-4, absorbance increases with increasing cell number. Cells appeared to be in a linear phase from 0.5-2*106 cells/ml proceeded by a constant stationary phase from 2*106 cells/ml and finally heading to death phase at 5*106 cells/ml. Thus 1*106 cells/ml was taken as appropriate cell density of Raw cells for the study (fig-4) so that we can study both increased and decreased cell viability.

Figure 4: Cell density test. Experiment was performed to determine the optimum cell concentration for further experiments. Cell densities were seeded from 0.5-5x106 cells/ml and viability was measured using the XTT test. The experiment was set up in duplicates and repeated three times. Error bars show standard error of mean (SEM).

2.2 Effect of celecoxib (µM) on cell viability of mouse macrophages (Raw cells)

The optimal cell density (1*106 cells/ml) was used to determine the effect of celecoxib (µM) on Raw cells using the XTT viability test. Increased celecoxib concentrations (1-100µM) were used to investigate the toxicity at different concentrations. As shown in fig-5, the drug doesn't have a notable toxic effect at 1-20µM but reduced viability was seen at concentrations ≥20µM and celecoxib concentrations ≥50µM killed almost all cells. Increased concentrations of the COX-2 inhibitor result in significant decrease of cell viability. Hence, 20-30µM was considered as an optimal toxic dose of the inhibitor since it is the highest dose that has marginal effect on the viability on the cell line in the above experiment.

Figure 5: Viability measurements of celecoxib. Experiment was performed to determine the optimum dose of celecoxib for further experiments. Cells were pre-treated with (1-100 µM) celecoxib concentration and viability was measured using the XTT test. There was a significant reduction in absorbance values at high celecoxib doses. The experiment was set up in pentalicates and was repeated three times. Error bars show standard error of mean (SEM).

2.3 Effect of DMSO solvent in Raw cells

Celecoxib was prepared in DMSO and to ensure that DMSO in itself does not have any effect on Raw cells, they were treated with 0.04% DMSO, corresponding to the amount of DMSO in celecoxib, and viability was measured by the XTT assay. Fig-6(a), shows no change in cell viability measured by XTT assay, after addition of 0.04% DMSO to Raw cells when compared with control (untreated cells). Effect of DMSO on NO-levels was also investigated using a chemiluminiscence reader. Fig 6(b) shows DMSO has no effect in NO-levels compared to untreated control. It is also seen NO is not produced by both control and DMSO. Hence, it is confirmed that DMSO has no impact on Raw cells at concentrations used in this study.

Figure 6: DMSO (0.04%) has no effect in Raw cells. (a) shows the effect of 0.04% DMSO on cell viability, in Raw cells analysed by the XTT assay. (b) shows the NO measurements of raw cells with/without treatment of 0.04% DMSO analysed by chemiluminiscence reader. Experiment was repeated three times in pentalicates and is presented with error bars showing SEM.

2.4 BCG-treatment of mouse macrophages

Raw cells were treated with different BCG doses (0.1-10*106 CFU/ml) and were analyzed by following methods.

2.4.1 Effect of BCG on cell viability of Raw cells by XTT assay

BCG dose titration curve was performed using XTT assay to study effects of BCG-treatment on cell viability. Increasing BCG doses (0.1-10*106 CFU/ml) were considered and as seen in fig-7(a), low BCG doses (0.1-1*106 CFU/ml) did not show high toxicity. A slight reduction in raw cell viability was seen at highest BCG dose-10*106 CFU/ml when compared with untreated control [fig-7(a)].

2.4.2 Effect of BCG on cell viability of Raw cells as measured by Trypan blue exclusion test

Number of viable cells were counted microscopically from the same XTT plates [presented in fig-7(a)] by staining them with trypan blue. From fig-7 (b), we can see proliferative effect of BCG until 1*106 CFU/ml whereas BCG at 5*106 CFU/ml dose reduces proliferation with approximately half compared to untreated control and cells are dead at 10*106 CFU/ml. Trypan blue assay results do not correlate with XTT viability assay results [fig-7(a-b)]. However, BCG at 5*106 CFU/ml dose seems to reduce viability in trypan blue assay and hence was used as an optimum dose for further experiments since it is the same dose given to patients.

2.4.3 Effect of NO release on BCG-treated Raw cells

The ability of BCG treated macrophages to produce NO was measured at different doses of BCG with a NO analyzer. Raw cells were treated with BCG for 24 hours before analysis and untreated cells were set as control. Air from the examination room was collected and its NO level was found to be the same as in the control flask. Fig-7(c), illustrates that BCG-treated Raw cells produce NO in a dose-dependent manner.

2.4.4 NOS2 and COX2 protein expression of BCG-treated Raw cells

After NO measures, expression of the proteins extracted from the same BCG treated cells were estimated by western blot. As seen above in fig-7(d), BCG induces COX-2 from 0.5-10*106 CFU/ml whereas NOS2 was expressed only at 1 and 10*106 CFU/ml BCG when it was reprobed on the same membrane. Actin, the house-keeping gene was loaded as the control and it confirms the equal loading of protein.


Figure 7: BCG Dose titration. (a) and (b) show the effect of different doses of BCG on Raw 264.7 mouse macrophages by XTT and Trypan blue assay respectively. Reduction in cell absorbance (at high doses) was seen in both tests. XTT test was set up in pentalicates and was repeated three times. Error bars show standard error of mean (SEM). (c) shows the NO measurements of Raw cells with different BCG doses analysed by chemiluminiscence reader. The increase in NO levels of BCG-treated cells is dose-dependent. Experiment was repeated four times and is presented with standard error bars. (d) Protein expression of NOS2 and COX2 in Raw cells treated with different doses of BCG was analyzed by immunoblotting, β-actin was used as the loading control.

2.4.5 Immunofluorescence staining of mouse macrophages mitochondria after BCG-treatment

Since the theory suggests that XTT viability assay uses the mitochondrial activity to measure viability and contradictory results were found between the XTT and the Trypan Blue assay results, mitochondrial staining was performed on raw cells for further investigation. In short, the XTT assay measures the reduction of tetrazolium salts by active mitochondria to formazan salts by metabolically viable cells. To address the reason behind the proliferative effect of BCG in the XTT assay, it is necessary to observe what happens in mitochondria after induction of BCG. Immunofluorescent mitochondrial staining was performed on the Raw cells which underwent 24 hour BCG treatment at different doses and untreated cells were set as control.

Figure 8: Effect of BCG on mitochondrial staining of live Raw cells. Immunofluorescent images of Raw cells after addition of different doses of BCG (0.1-10*106 CFU/ml). Red fluorescence shows the mitochondrial staining using mitotracker red CMXRos probe. Blue fluorescence shows the nuclei staining using DAPI mounting medium.

In the above figure-8, no significant difference was found in mitochondria of raw cells at low dose of BCG (0.1-0.5*106 CFU/ml). At higher doses (1-10*106 CFU/ml), mitochondria seems to be brighter and larger which indicates active hypopolarized mitochondria. Furthermore, only few cells were seen at the highest dose of BCG (10*106 CFU/ml) which correlates with Trypan Blue analysis [Fig-7(b)] indicating cell death.

2.5 Combined effect of BCG and celecoxib on Raw cells

Based on the above experiments, celecoxib (10-40µM) and BCG (5*106 CFU/ml) were analyzed for combined treatment on raw cells by the following methods.

2.5.1 Effect of BCG and celecoxib on cell viability of Raw cells by XTT assay

Raw cells were pre-treated with different celecoxib doses 1 hour prior to addition of BCG. Untreated cells were set as celecoxib control [point 0- straight line in fig-9(a)] and only BCG (5*106 CFU/ml) treated cells were set as BCG control [point 0-dashed line in fig-9(a)]. As shown in fig-9(a), higher doses of celecoxib when combined with BCG, decreases the cell viability analyzed by the XTT assay. Proliferative effect is evident after addition of BCG when compared to cells that are treated only with celecoxib (µM) or untreated control and this result corresponds to fig-7(a).

2.5.2 Effect of BCG and celecoxib on cell viability of Raw cells as measured by Trypan Blue exclusion test

Viable cell count, as assessed by trypan blue exclusion test [as shown in fig-9(b)], reveals that BCG only (5*106 CFU/ml) at point 0, doesn't exhibit a proliferative effect on Raw cells which correlates with the same observation already seen in Fig-7(b) for the dose 5*106CFU/ml. However, same effect of celecoxib was found in Raw cells when they were analysed by both XTT viability and trypan blue assay. Hence the proliferative effect of BCG in XTT assay should be investigated further.

2.5.3 Effect of NO-release on Raw cells co-treated with BCG and celecoxib

To assess the inhibitory role of celecoxib on NO, macrophages were treated with the combination of BCG and celecoxib at different doses for 24 hours and NO readings were measured using chemiluminescence reader. Results are shown in fig-9(c), increasing concentrations of celecoxib inhibit the NO release in BCG treated Raw cells. It is also seen that celecoxib treated cells do not exhibit NO in the absence of BCG. Thus, celecoxib inhibits BCG induced COX-2, thereby inhibiting NO production.

However, on the contrary, in another experiment with the same experimental set-up, we found the opposite where celecoxib increased NO levels in BCG treated cells [Fig-10(a)].

2.5.4 NOS2 and COX2 protein expression of Raw cells co-treated with BCG and celecoxib

The expression of extracted protein lysates from NO measured cells were also demonstrated with western blot technique, confirming the inhibitory effect of celecoxib (at 40µM) and BCG on COX-2 and NOS-2 [Fig-9(d)]. Different observation was seen in the immunoblot analysis of proteins extracted from another individual experiment [Fig-10(a)]. BCG induced COX-2 protein was expressed at all doses of celecoxib and NOS-2 which was reprobed on the same membrane showed induction even at high dose of celecoxib (30 and 40 µM) [fig-10(b)].

Eventhough the NO data in fig-10(a) showing no inhibition by celecoxib on NO release of BCG-treated Raw cells is contradictory to fig-9(c) where celecoxib inhibits BCG-induced NO levels, and the western blot in fig-10(b) shows the opposite NOS2 and COX2 protein expression to the western blot in fig-9(d), the western blot results in fig-9(c) [where NOS-2 and COX-2 protein expression are inhibited at highest celecoxib dose-40µM combined with BCG] correlates with its corresponding inhibited NO measures in fig-9(d) and the results in fig-10(b) corresponds to results in fig-10(a) [where celecoxib fails to show inhibition].


Figure 9: Effect of BCG and celecoxib on Raw cells. (a) and (b) straight line indicates the effect of celecoxib(µM) at different doses and dashed line indicates the combined effect of BCG (5*106 CFU/ml) and celecoxib(µM) on Raw 264.7 mouse macrophages measured by XTT and Trypan blue assay respectively. Celecoxib with/without BCG shows a reduction in cell absorbance in both tests. XTT test was set up in duplicates and was repeated three times. Error bars show standard error of mean (SEM). (c) shows the NO measurements of Raw cells with/without treatment of BCG (5*106 CFU/ml) combined with celecoxib(µM) as analysed by chemiluminiscence reader. There was inhibition on NO levels of BCG-treated cells by celecoxib. (d) Protein expression of NOS2 and COX2 in raw cells treated with BCG and celecoxib by immunoblot analysis, β-actin was used as the loading control.

Figure 10: Celecoxib (µM) doesn't show any inhibition on BCG induced NO in Raw cells. (a) shows the NO measurements of Raw cells with/without treatment of BCG (5*106 CFU/ml) combined with celecoxib (µM) as analysed by chemiluminiscence reader. (b) Protein expression of NOS2 and COX2 in raw cells treated with BCG and celecoxib by immunoblot analysis, β-actin was used as the loading control.

2.5.5 Immunofluorescence staining of mouse macrophages mitochondria after co-treatment with BCG and celecoxib

Effect of celecoxib with/without BCG on mitochondria of Raw cells was also examined and since celecoxib shows toxicity at 30 and 40µM, those doses were considered with/without the combination of BCG at 5*106 CFU/ml (fig-11). Untreated cells were set as control and BCG treated cells were set as BCG control to compare the observation. Fig-11, shows the overlay of both immunofluorescently stained images of mitochondria and nuclei of Raw cells. Significant toxic effect of celecoxib can be seen at 40µM dose when compared to 30µM dose exhibiting reduction in cell number. However, prominent shrinkage in the nuclei size was seen after addition of celecoxib in BCG treated cells compared to BCG treatment alone.

Figure 11: Effect of celecoxib and BCG on mitochondrial staining on Raw cells. Immunofluorescent images of Raw cells after addition of celecoxib (30 and 40µM) with/without BCG (5*106 CFU/ml). Red fluorescence shows the mitochondrial staining using mitotracker red CMXRos probe and blue fluorescence represents the nuclei staining using DAPI mounting medium.

Furthermore, immunofluorescent staining was also performed to examine whether BCG induces COX-2 or not. Cells with/without BCG (5*106 CFU/ml) were initially stained with mitochondria red CMXros probe followed by addition of COX-2 antibody and secondary antibody. As shown in fig-12, prominent expression of COX-2 is seen in the BCG treated raw cells when compared to untreated cells.

Untreated cells

Figure 12: BCG-treatment induces COX-2 in Raw cells. Red fluorescence indicates the mitochondrial staining of raw cells using mitotracker red CMXros. Green fluorescence indicates the COX-2 expression in mitochondria after addition of BCG and yellow denotes the co-staining in overlay.

3. Discussion

In this study, the effect of COX-2 inhibition on NO release of BCG-treated macrophages was investigated. COX-2 inhibition shows a dual response on the NO release of BCG treated raw cells and this result suggest that COX-2 inhibition might follow both NO dependent/independent pathway in BCG-treated raw cells. Despite the fact that manufacturers call XTT assay as cell proliferation kit, our result shows that XTT assay only measures cell viability depending on mitochondrial activity of cells. Notably, BCG triggers mitochondrial activity in macrophages which appears as an increase in cell viability in the XTT assay even though we know that cells die. Interestingly even though BCG induces COX-2, when combined with celecoxib, a dose-dependent decrease in raw cell viability was observed which indicates high toxicity due to co-treatment. Celecoxib might also inhibit the formation of tumour growth in both COX-2 dependent and independent pathways since its action does not correlate with COX-2 inhibition always.

This study suggests that BCG-induced COX-2 inhibition follows NO dependent as well as independent pathways. Our findings show inhibition of COX-2 activity (by celecoxib) inhibiting BCG induced NO activity in Raw cells [fig-9(c)]. Immunoblot analysis also shows inhibition of NOS-2 and COX-2 expression by celecoxib at 40µM dose [fig-9(d)]. These datas demonstrates the previous results which suggests that BCG-induced COX-2 expression follows NO-dependent pathways in macrophages (Bansal et al., 2009). On the other hand, NOS2 was unexpressed in western blot at BCG dose (5*106 CFU/ml) [lane-1 in fig-8(d)] as well as in [lane-5 in fig-7(d)] in western blot even though increased NO levels from Raw cells at same BCG dose were observed [fig 7(c)]. The discrepancy in the above mentioned result would be due to methodological/laborative errors since high NO levels were measured at BCG dose (5*106 CFU/ml) [fig-7(c)].Conversely, we also have results showing no inhibition of BCG triggered NO levels in Raw cells by celecoxib [fig-10(a)] and even BCG induced NOS2 expression is seen at high celecoxib doses-30&40µM [fig-10(d)] which is contradictory to fig-9(d). These results correlates with the idea that BCG induced COX-2 expression not only involves NOS2 dependent pathways, but that can also act through some NOS2 independent pathways (Bansal et al., 2009).

As reported in a study, BCG induces COX-2 expression in macrophages through both NO-dependent and independent pathways by activation of NF-kβ by Notch1-PI-3K signaling cascades and ERK1/2, P38 MAPKs respectively (Bansal et al., 2009). Apart from this, the interaction between COX-2 and NOS2 differs in each cell line, tissues and pathophysiological conditions (Xuan et al., 2003). Co-expression between these two genes and its byproduct are also identified in several human cancers like Hepatocellular carcinoma (HCC), non-small cell lung cancer and colorectal cancer (Marrogi et al., 2000),(Rahman et al., 2001), (Hong et al., 2004). Thus, the cross-talk between NOS2 and COX-2 demands further investigation of these genes to identify the additional factors associated for NOS2 upregulation in BCG-treated macrophages.

3.1 BCG triggers mitochondrial activity in macrophages

High BCG doses have a cytotoxic effect on cells (Gontero et al., 2010), however increased doses of BCG failed to show a notable decrease in cell viability when we performed the XTT viability assay[fig-7(a)]. Accordingly, the live raw cells counted manually from same XTT plate, using trypan blue inclusion [fig-7(b)] shows a significant decrease in cell viability at BCG dose-5*10^6 CFU/ml clearly breaking the proliferative trend of cells that were treated with low BCG doses (0.1-1*10^6 CFU/ml). A sharp decline in cell viability was also seen at highest BCG dose-10*10^6 CFU/ml. Theory behind XTT assay suggests that the increased number of live cells correlates with increased mitochondria activity by formation of more orange-coloured formazan. Since BCG at high doses is anti-proliferative and observed results were not the expected results in XTT viability test which relies on mitochondrial activity, staining was performed for further investigation. The increased viability suggested by XTT test in fact is increased mitochondrial activity represented by larger and brighter cells in immunofluorescent staining (fig-8). Also with DAPI, we can clearly see that there are only fewer cells at highest BCG dose (10*10^6 CFU/ml) which indicates BCG toxicity (fig-8). Hence this result indicates that BCG, at appropriate dose triggers the mitochondrial activity which ultimately shows a proliferative effect in XTT results, eventhough truly there is not an increase in cell viability as seen in trypan blue test results [fig-7(a-b)].

3.2 BCG induces NO in macrophages

It is generally believed that NO is a critical signalling and effector molecule in macrophage cytotoxicity post BCG immunotherapy and several studies prove that BCG induces NO activity in macrophages. Our results from the Chemiluminescence analysis [fig-7(c)] replicate the previously published data that NO levels increase with increased concentrations of BCG (Bansal et al., 2009). Our data in fig-7 (b-c) suggest that low NO levels promotes cell growth whereas high concentrations exhibit cytotoxic effects. This hypothesis supports an earlier study which proposes that low concentrations of NO might stimulate NF-kβ leading to NOS2 transcription and that high NO levels result in activation of apoptosis inducing factors (Umansky et al., 1998). These results thereby elucidate the involvement of NO in anti- tumour mechanism of BCG in macrophages.

3.3 BCG induces COX-2 in macrophages

Several researchers have demonstrated the association of COX-2 and its byproduct PGE2 with procarcinogenic effects in the progression of bladder carcinomas (Chen et al., 2002). COX-2, one of the major inflammatory mediators in TCC is believed to promote tumour progression by supporting angiogenesis (Chen et al., 2009). Therefore, elevated COX-2 levels can be used as an effective biomarker in bladder cancer to improve targeted therapy. Immunostaining (fig-12) and Western blot results [fig-7(d)] clearly indicate that BCG induces COX-2 expression in macrophages (fig-12) and our result goes in line with the previous published data (Bansal et al., 2009). Inhibition of COX-2 after BCG stimulation could reduce the cytotoxicity induced by macrophage and its secreted factors.

3.4 Celecoxib decreases Raw cell viability in a dose-dependent manner Mechanistically, celecoxib is known to downregulate PI-3K, a positive regulator of COX-2, resulting in activation of caspase-9 finally leading to apoptosis (Liu et al., 2008). Celecoxib, also initiates apoptosis by cell cycle arrest on a dose-dependent manner (Mohammed et al., 2006) which matches our toxicity measures of the drug on cell viability (fig-5). The drug was also found to be very sensitive. If frozen, celecoxib appears to lose its toxicity and it was also noted that it is not stable in DMSO for more than a couple of weeks.

3.5 Mode of action of Celecoxib

Mechanism of action of celecoxib is complex, involving different inhibitory pathways and it has been suggested to act as an anti-proliferative, anti-apoptotic and anti-angiogenic agent in a number of different cell types (Davies et al., 2000). COX-2 expression always doesn't correlate with the response of celecoxib. We had results in which COX-2 was not always inhibited by celecoxib [fig-9(d), 10(d)], although we also had experiments showing that celecoxib had some effect in cell viability [fig-5, 9(a-b)]. Other study suggests that, even though celecoxib is known as a selective COX-2 inhibitor, it is capable of inducing toxic activity in tumour cells and tissues in the absence of COX-2 (Grosch et al., 2001). These results reveal that celecoxib induces action by both COX-2 dependent and independent pathways in urinary bladder cancer cells (Dhawan et al., 2008). Another study, suggests that celecoxib induces apoptosis through activation of a novel mitochondrial pathway, where the members of the Bcl-2 family are inhibited (Jendrossek et al., 2003). As shown in fig-9(a-b), the "only CXC" does not show the same results with XTT and trypan blue tests, which might be because of the alteration in mitochondrial activity of Raw cells (in XTT test) after addition of celecoxib (fig-11). These results support the previously published data that celecoxib induces apoptosis by breakdown of mitochondria membrane potential illustrating the role of celecoxib in mitochondria-mediated apoptosis pathways (Jendrossek et al., 2003).

3.6 Co-treatment of BCG with celecoxib shows high toxicity

Our results suggest that celecoxib in addition with BCG, accelerates toxicity by decreasing cell viability as seen in both XTT and trypan blue exclusion viability tests [fig-7(a-b)]. This supports the results from a recent study which suggests that, combination of BCG and celecoxib increases the release of tumour infiltrating lymphocytes (TILs) in macrophages when compared to BCG alone in UCC mouse model (Dovedi et al., 2008). Hence, combination therapy of celecoxib and BCG which shows an increase in tumour efficacy turns out to be a promising therapeutic strategy for invasive bladder cancer patients.

3.7 Implications for future research

In future, since, celecoxib is known to have a half life of about 11 hours (Davies et al., 2000), a time course response of celecoxib in cell viability should be performed. The effect of NO inhibition on COX-2 using L-NAME should be performed since a positive feedback loop exists between these two (Bansal et al., 2009). We also need to investigate whether supernatant from BCG treated raw cells, where COX-2 is inhibited, can induce NOS-2 up-regulation in urinary bladder cancer cells (MBT2 cells).

3.8 Conclusion

In conclusions, instillation of BCG in the urinary bladder induces a pro-inflammatory response which activates immune cells releasing various types of cytokines. Both COX-2 and NOS-2 stimulated by these cytokines seems to be co-regulated with each other. Another study shows that NF-kβ binding sites-regulators of inflammation and cell survival, are also present in the promoter regions of both NOS2 and COX-2 suggesting their role in immune response (Posadas et al., 2003). NO mediated regulation of COX-2 takes place through several pathways in macrophages (Bansal et al., 2009) which ultimately results in different macrophageal responses to varied NO levels caused in the inflammatory response (Xuan et al., 2003). In order to enhance the quality of life of bladder cancer patients, determining whether high NO levels are either advantageous to BCG treatment or not is essential. Also, identifying the molecular markers or regulators that modulate tumour microenvironment improves the overall efficacy of BCG therapeutic response. Thus, a better understanding of the BCG-induced tumour remission mechanism in macrophages is crucial for enhancing the treatment of bladder cancer by specific targeting of tumour microenvironment and adaptive responses.

4. Materials and Methods

4.1 Culturing of Raw cells (264.7)

"Mouse leukemic monocyte macrophages" Raw Cells (264.7- cell line) were maintained in Dulbecco's Modified Eagle Medium (DMEM, high glucose) (Invitrogen, Eugene, Oregon USA) supplemented with 10% Fetal Bovine Serum (FBS), 45units/ml Pencillin Streptomycin (PEST), 1mM L-glutamine, 100nM β-mercaptoethanol, 50ug/liter of pyruvate (Invitrogen, Eugene, Oregon USA) and was incubated at 37°C in a humidified incubator with 5% CO2. Initially, cells were washed twice using calcium and magnesium free Dulbecco's Phosphate Buffered Saline (DPBS 1X) (Invitrogen, Eugene, Oregon USA) and were detached from the tissue culture flask by addition of 2-3ml 0.25% Trypsin-EDTA (Invitrogen, Eugene, Oregon USA). The trypsinated cells were transferred to a centrifuge tube and after gentle mixing, the suspended cells underwent 5min centrifugation at 1200xg. The supernatant was discarded followed by resuspension of the pellet in appropriate quantity of DMEM media. Cells were seeded to new flasks and/or 96-well microtitre plates. Cell suspension (10µl) was added to a simplified version of a Bürkitt chamber (covalently Glasstic Slide 10, Hycor, California, USA) to determine the cell concentration.

4.2 Cell viability assay

A cell viability assay was carried out as described below, to determine two things, namely optimal cell density for the assay, and toxicity of both Celecoxib (Sigma-Aldrich, Sweden,- COX-2 inhibitor) and BCG (RIVM strain derived from strain 1173-P2, Medac, Hamburg, Germany) or the combination of both. Each experiment was repeated at least twice, in at least duplicates. Celecoxib was prepared by dissolving the drug in DMSO and BCG was prepared by mixing a vial of BCG with 5ml DMEM.

4.2.1 Density test:

Raw cells (100µl) were seeded in 96- well microtitre plates, at concentrations increasing from 0.5 to 5 million cells, in duplicates for each experiment and were incubated for 48 hours at 37°C with 5% CO2. The culture medium (DMEM) without cells was set for background. Cell proliferation kit II (XTT) (Roche, Mannheim, Germany) was used to measure the cell viability as described by (Scudiero and et al., 1988) with slight modifications. In brief, XTT solution was prepared by mixing 2.5ml of XTT labeling reagent with 0.05ml of electron coupling reagent, so that a final volume of 50µl solution was added to each culture well followed by a 2 hour incubation at 37°C. After the incubation period, absorbance was measured spectrophotometrically using an ELISA plate reader at a wavelength of 450nm with reference wavelength of 650nm. The absorbance of background (media only) was subtracted from the other absorbance results.

4.2.2 Toxicity test

To determine how toxic the celecoxib drug and BCG are, raw cells (optimum cell concentration 10,000 cells/100µl) were seeded in 96-well microtitre plate and were incubated overnight at 37°C with 5% CO2. Next day, cells were treated with 1-100µM of celecoxib, 0.1-10*10^6 CFU/ml of BCG or the combination of both and the plate was again incubated for another 24 hours. Cells were treated with celecoxib one hour prior to addition of BCG. Also to confirm that DMSO (Dimethyl- sulfoxide), in which celecoxib is dissolved, is not toxic to cells, a concentration of 0.4% DMSO was used as control. This-DMSO dose corresponds to the treatment of cells at 40µM celecoxib treatment. XTT assay was performed as described above.

4.3 Trypan Blue exclusion test of cell viability

Cells were counted using the trypan blue exclusion assay on the same tissue culture plates which were already analysed with XTT. Cells were washed with 100µl (per well) of magnesium and calcium free PBS followed by addition of 10µl of 0.25% trypsin to detach them from wells. DMEM (90µl per well) was added and 50µl of carefully resuspended cells were transferred to another plate containing 50µl 0.4% trypan blue (Sigma, Ateinhemm, Germany). Finally, 10µl cell suspension from each well was counted twice in a simplified version of a Burkitt chamber.

4.4 Determination of NO levels

Raw cells were seeded in T25 cell culture flasks and were incubated at 37°C with 5% CO2 in a humidified incubator. Next day, the sub-confluent flasks were treated with varying concentrations of celecoxib and BCG. Celecoxib (10-40µM) was added one hour before BCG (5*10^6 CFU/ml) treatment and untreated raw cells were set as control whereas only BCG added to sub-confluent flasks was set as BCG-control. During the day of analysis of NO levels, caps of tissue culture flasks were screwed tightly and flasks were incubated for 4 hours before NO level measurements. Using a syringe, 30ml of air from the head space of each flask was injected into a CLD77 AM (Eco Physics, Durnten, Switzerland) chemiluminescense reader and peak levels of NO measures were recorded.

4.5 Protein Extraction

For Protein extraction, all the confluent flasks were placed on ice and the following steps were carried out in ice cold temperature. All the flasks were washed twice with 1xPBS after removal of media. After complete aspiration of all wash buffer from the flasks, 300µl ice-cold EBC lysis buffer (50mM Tris (pH 8.0), 120mM NaCl, 0.5% NP40, 100mM NaF, 0.2mM Sodium Orthovanadate, 10µg/ml PMSF, 10µg/ml Aprotinin, 50µg/ml β-glycerophosphate ) was added to each flask. Immediately, the cell lysates were scraped off the flasks and were transferred to fresh eppendorf tubes. The lysates were vortexed gently followed by a 20 min centrifugation at 14000rpm/10937xg in 4°C. Supernatants (300µl) were transferred carefully to new tubes leaving the cell debris behind. The protein concentration in the whole cell lysates was measured using Bradford's Protein Assay (Bio-Rad Laboratories) according to the manufacturer's instructions, by vortexing 2µl cell lysates, 498µl water and 500µl Bradford dye in cuvettes followed by an incubation of 5 min at room temperature. The spectrophotometer (Pharmacia Gene Qant ProAmersham) was adjusted to a wavelength of 595nm by setting the reference using a blank (1000µl water) and the absorbance was recorded for all the cell lysates at 595nm. The proteins were denatured by addition of SDS loading buffer (1:4) (0.25M Tris (pH 6.8), 40% glycerol, 8% SDS, 4% β-mercaptoethanol, 1% Bromophenol blue) and stored in -20°C.

4.6 Western Blot

Western blot was performed in order to assess the protein expression of the extracted lysates. The protein cell lysates diluted in SDS loading buffer were defrosted and denatured at 95°C for 5 minutes. 10-50µg of protein (depending on the experiment) were loaded on to pre-cast 4-8% SDS Pierce protein gels (ThermoScientific), and were separated by electrophoresis. Full range Rainbow (GE-Healthcare, Biosciences) was used as a ladder to verify protein size. Gels were run at 185V for approximately 30 minutes depending on choice of gel and protein size. Transfer of proteins to PVDF membrane was done by dry blot technique using an iBLOT (Invitrogen, California, USA), according to the manufacturer's instructions and the program was set for 9 minutes. To prevent non-specific binding of biomolecules, the PVDF membrane was blocked in 0.5% BSA (BioRad) dissolved in T- PBS (0.1% Tween 20, Phosphate Buffered Saline) for 1hour in room temperature. The membrane was incubated for 1 hour at room temperature in primary antibody, diluted in blocking solution (different dilution for different antibody) followed by overnight incubation at 4°C. Next day, the membranes were washed with 3x 10' TPBS, to remove the excess non-specific binding of antibody. The membranes then underwent 1hour incubation at room temperature with Horse Radish Peroxidase (HRP) conjugated secondary antibody diluted in TPBS. Again, membranes were washed with 3x 10' TPBS followed by 1x PBS wash lasting for minimum 10 minutes. Supersignal West Femto Chemiluminescent Substrate (Thermo Fisher Scientific, Massachusetts) was used to detect protein signal from HRP-conjugated antibody- probed membranes on autoradiographs (Kodak X-Omat 2000 processor, Kodak, New York, USA).

Antibodies used:

β-actin (Sigma-Aldrich), COX-2 (BD Biosciences, New Jersey, USA), NOS-2 (BD Biosciences, New Jersey, USA).

4.7 Immunofluorescence Staining

Mitochondria of the live raw cells were stained using Mitotracker Red CMXRos probe (Invitrogen, Eugene, Oregon USA) following the manufacture's instructions with slight modifications. The mitotracker stock solution was prepared by diluting the probe in DMSO to a concentration of 1mM (Invitrogen, Eugene, Oregon USA). Raw cells cultured in DMEM were seeded on glass coverslips in 6-well culture dishes at a density of 1*106 cells/ml. Next day, cells were treated with 20, 30 or 40µM of celecoxib and/or 0.1-10*106 CFU/ml BCG and the plates were incubated for 24 hours. The media was discarded and pre-warmed mitochondria staining solution (0.1µl of 1mM mitotracker stock solution per 1ml of DMEM, final concentration 0.5µM) was added to all wells, and was incubated for 10 minutes at 37°C with 5% CO2. The cells were washed twice with PBS followed by fixation with ice cold methanol (1ml/well) and the plates were incubated in the freezer for 5 minutes followed by PBS wash three times. Post fix solution (3:1, 100% Ethanol:Acetic acid, stored at -20°C) was added to permeabilise cells and the plates were incubated for 5 minutes at -20°C followed by PBS wash three times (2ml/well). The cells were blocked in 0.5% BSA (BioRad) dissolved in T- PBS (0.1% Tween 20, Phosphate Buffered Saline) for 30 minutes at room temperature followed by a PBS wash. The cells were incubated in 200-400µl primary antibody- COX-2 (1:200 dilution, BD Biosciences, New Jersey, USA), diluted in PBS- 0.5% BSA for 1 hour at room temperature on wet towels. Cells were then rinsed in PBS three times and incubated for 30 minutes at room temperature with 200-400µl goat anti mouse antibody labeled with Alexa Fluor 488 (1:400 dilution, Invitrogen) diluted in PBS-0.5% BSA. The plates were covered in foil since secondary antibody is light-sensitive. After three times rinse in PBS, the stained cells were mounted with DAPI containing mounting medium (Vector Laboratories) on glass slides (76Ã-26mm, Menzel Glaser) and the images were viewed under fluorescence microscope (Nikon microscope Eclipse E800 and Nikon camera DXM 1200) using the appropriate filter.

5. Acknowledgements

I extend my sincere and deepest gratitude to Ass. Prof. Petra de Verdier for her excellent supervision, invaluable guidance and constant support throughout the project and also for giving me the opportunity to work in her laboratory. Special thanks to Nasrin, Lotta, Mirjana and Emmie for their kind hospitality and great working environment. Its been a pleasure to get to know you all. I am also indebted to thank my course co-ordinator Dr. Lena Aslund.