EMMQ-induced DNA damage in wild-type p53 cell lines

23

To determine whether EMMQ induced DNA damage in NSCLC cells in vitro using comet assay. The results showed that EMMQ induced DNA damage in A549 and H460 cells with wild-type p53 status. At 5 μM concentrations of EMMQ led to A549 cells a long DNA migration smear (comet tail), and these effects occurred in a dose-dependent manner (Fig.

3A), but the similar effect on H460 cells occurred at the highest dosage (10 μM). The comet tails indicating DNA lesions in A549 and H460 cells were detected after 24 h EMMQ treatment and the appearance of the excluded tail length was dose-dependent (Fig. 3B).

3. EMMQ increased sub-G1 population cells, G2/M arrest and apoptosis in both A549 and H460 cells

To gain more knowledge on cell cycle distribution following treatment, EMMQ induced cell cycle disturbance in NSCLC cells were examined. After 48 h EMMQ treatment, the cells were stained by PI and the cell cycle distribution was detected by flow cytometry. The sub-G1

cell population significantly increased in A549 and H460 cells than in DMSO vehicle controls and the effects were dose-dependent, while no effect shown in H1299 cells (Fig. 4A). In addition, our data showed that EMMQ induced A549 and H460 cells cell cycle arrest in G2/M in 8 μM at 24 h. The results indicated that EMMQ induced cell cycle arrest in G2/M at 24 h and later apoptosis at 48 h in wild type p53 A549 and H460 cells (Fig. 4B).

To examine whether EMMQ treatment induces apoptosis, NSCLC

24

cells are treated with various concentrations of EMMQ (0, 5, 8, and 10 µM) and then conducted the flow cytometry-based Annexin V and PI double staining assay. Starting at a concentration of 5 μM after 48 h treatment of EMMQ, the early and late apoptotic phase cell population rose to 13% and 16% for A549 cells as well as 6% and 5% for H460 cells, respectively. However, no apoptotic cell death was shown in null p53 H1299 cells (Fig. 5A and 5B). The results implied that EMMQ-induced cell viability reduction in NSCLC cells carrying wild-type p53 was caused by apoptotic cell death following DNA damage.

4. EMMQ-induced apoptosis through intrinsic pathway

The apoptosis attributed to DNA damage can proceed through intrinsic pathway or extrinsic pathway [12, 29]. To clarify whether the intrinsic pathway is involved in EMMQ-induced apoptosis, we firstly examined the protein expression of p53, anti-apoptotic Bcl-2 family protein (Bcl-2), caspase-3 and cleavage of poly(ADP ribose) polymerase (PARP) by western bolt. Fig. 6A shows the increased concentrations of EMMQ activated p53 and down-stream effector p21, reduced p-Akt^S473, Bcl-2 and procaspase-3 levels, and increased cleavage of poly(ADP ribose) polymerase (PARP) in A549 and H460 cells after 48 h treatment.

On the other hand, in the presence of 5 μM of EMMQ, activation of p53, reduction of p-Akt^S473 and Bcl-2 intensities as well as procaspase-3 dissipation and increased PARP cleavage in A549 and H460 cells were detected in time-dependent manners. No change was shown in Akt and p-Akt^S473, Bcl-2 levels and procaspase-3 as well as cleavage PARP in

25

p53-null H1299 cells within the time intervals and drug concentration ranges as studied (Fig. 6B). The results of western blots implied that EMMQ induced apoptosis through p53 activation and cleavage of PARP and diminished procaspase-3 is related to intrinsic pathway.

Mitochondria-related apoptotic pathway was linked to signal cascade following the ΔΨm disruption, signaling mitochondrial

dysfunction involvement [46]. It’s well known that the attenuated ΔΨm, outer membrane regulator Bcl-2 and release of downstream modulator mitochondrial cytochrome c indicated mitochondria-mediated pathway leading to apoptosis [47, 48]. Fig. 7A shows ΔΨm loss in A549 cells suggested that the mitochondria-mediated apoptosis by low dosage of EMMQ was initiated starting at 4 h, but the effect was absent in H1299 cells. The impaired mitochondrial functions were further accentuated by cytochrome c release in both A549 and H460 cells with increasing drug concentrations after 24 h treatment (Fig. 7B).

5. The suppressed growth of xenograft tumors in EMMQ-treated A549 cells

To assess the effect of EMMQ in vivo, a tumor xenograft study (6 animals per group) was carried out. Subcutaneous injection of EMMQ twice a week at a concentration of 1 mg/kg per mouse resulted in a

significant decrease in tumor volumes. There was also a notable decrease in the size and excised volume of tumors from EMMQ-treated mice as compared with control (Fig. 8A, 8C and 8D). The size and the excised

26

volume of tumors decreased to more than 50% following 4 week treatment, while no significant difference of body weight in

EMMQ-treated nude mice relative to vehicle control as shown in Fig. 8B.

The lysates from the dissected tumors were further analyzed by western blot analysis. The levels of survival genes Akt, p-Akt^S473 and cell

proliferation marker PCNA as well as mitochondrial modulator Bcl-2 were noticeably reduced. The intensities of p53 and cleaved PARP were elevated, whereas those of procaspase-3 decreased in EMMQ-treated group as compared with those of control treatment (Fig. 8E).

6. The extent of EMMQ-induced apoptosis is dependent on p53 status To examine the role of p53 played during the process of EMMQ induced apoptosis. H1299 cells with stable expression of ectopic p53 (H1299/p53) (positive control) or mutant p53R267P (H1299/p53^R267P) (negative control) were established [49]. EMMQ induced cell viabilities significantly decrease in H1299/p53 cells as compared with

H1299/p53^R267P clone (Fig. 9A). Fig. 9B shows the loss of ΔΨm in H1299/p53 cells suggested that the mitochondria-mediated apoptosis occurred in low dosage of EMMQ, but the effect was absent in

H1299/p53^R267P clone.

Furthermore, the induced apoptotic sub-G1 cell populations by EMMQ were more apparent in H1299/p53 cells (Fig. 10). A substantial amount of cells remained at G2/M phase in H1299/p53^R267P clone.

The apoptotic cells induced by EMMQ were more distinct in H1299/p53

27

than those of H1299/p53^R267P clone as determined by

annexin-V/PI-staining (Fig. 11A). The release of cytochrome c by EMMQ in H1299/p53^R267P was comparably less than that of H1299/p53 cells (Fig. 11B).

The apoptotic cells induced by EMMQ were more distinct in H1299/p53 than those of H1299/p53^R267P clone as determined by annexin-V/PI-staining (Fig. 11A). The release of cytochrome c by EMMQ in H1299/p53^R267P was comparably less than that of H1299/p53 cells (Fig. 11B). The results demonstrated that no significant effects of sub-G1 population, apoptosis ration and cytochrome c release level were detected during EMMQ concentration (0, 5, 8 and 10 μM) in

H1299/p53^R267P cells (Fig. 9, 10 and 11).

7. Down-regulated p53 abolished the onset of EMMQ-induced cell death in NSCLC cells

To ensure that p53 was indeed necessary in drug-mediated cell death, experiments by transfecting shRNA targeting exon 7 of p53 to cells prior to drug treatment were carried out along with those of NS control. The result of cell viability indicated that A549 and H460 cells were

transfected with p53 shRNA led to the sensitivity toward EMMQ was eliminated as compared with cells transfected with NS control (Fig. 13).

Western blot analysis of A549 cells showed that cells introduced with p53 shRNA exhibited significant reduction of p53 as compared with

28

those transfected with NS control alone. In addition, the mitochondria modulator Bcl-2 and pro-survival gene p-Akt^S473 were unaffected by EMMQ by knocking down p53 (Fig. 14A). Similar results were also observed with reduced viabilities and Bcl-2 attenuation (Fig. 14B) in p53 shRNA-transfected H460. The results altogether suggested that p53 was needed during mitochondrial pathway activation that predates the

effectiveness of EMMQ in motivating apoptotic cell death of NSCLC cells.

8. The effect of EMMQ inhibited metastasis is dependent on Akt and β-catenin status

To gain more knowledge on cell metastasis following treatment, we examined whether EMMQ inhibited NSCLC cell migration. The

extensive evidence suggests that Akt-transduced signals directly influence cell motility in normal development and in disease [50]. These evidences demonstrated that the Akt inhibited cell motility and metastasis in cancer cells [51, 52]. In addition, β-catenin regulates the expression of genes involved in cell migration and adhesion, such as cadherins, catenins, ADAMs, and integrins, and this gene expression regulation is likely the mechanism by which β-catenin regulates cell motility and adhesion [53].

The results by western blot analysis demonstrated that EMMQ reduced the levels of Akt, p-Akt^S473 and β-catenin in A549 and H460 cells, but no change of protein expression level was detected in H1299 cells (Fig.

15A).

29

The wound healing assay is measure cell migration in vitro. The basic steps involve creating a ‘‘scratch’’ in a cell monolayer, capturing the images at the beginning and at regular intervals during cell migration to close the scratch, and comparing the images to quantify the migration rate of the cells [54] . Fig. 14B showed that low dosage of EMMQ reduced A549 cells wound closure, however, the H1299 cells migrated into the wound and filled with the area at EMMQ treatment for 48 h. The result suggested that EMMQ may inhibit A549 cells migration at 5 µM.

MMPs are a group of proteolytic enzymes which are important in many physiological processes such as embryogenesis, development, and wound healing. Dysregulated MMP activity has long been implicated in diseases associated with uncontrolled proteolysis of connective tissue matrices such as arthritis, tumourgenesis and tissue ulceration. The results indicated that EMMQ may inhibit A549 and H460 cell migration at 5 µM.

In addition, EMMQ reduced MMP-2 and MMP-9 activities of A549 cells in a dose and time manner (Fig. 14D). This implied that EMMQ may inhibit cells metastasis or possibly invasion in NSCLC.

30

VI. DISCUSSION

Chemotherapy adverse side effects and resistance to current

anticancer agents have been the pressing problems in the success of lung cancer therapy [55]. As in the case of other chemotherapeutic agents, cisplatin therapy usually results in the second option, leading to

programmed cell death, also known as apoptosis. Cisplatin is the most widely used anticancer drug and causes cell death by inducing apoptosis.

Nevertheless, the high rate of resistance observed during therapy with cisplatin, as well as the occurrence of non-sensitive cancer cells, prompt the quest for agents endowed with apoptosis-independent mechanisms of action [56] . New chemotherapy strategies are urgently needed for lung cancer treatment.

Previous study showed that the IC50 value of a quinoline derivative, 6-methoxy-8-[(2-furanylmethyl)amino]-4-methyl-5-(3-trifluoromethylphe nyloxy)quinolone, is 16 ± 3 nM in breast cancer cells [57]. Another novel derivative of N-amidino-substituted benzimidazo[1,2-α]quinoline

produced differential antiproliferative mechanisms in two human colorectal cancer cell lines that differ in p53 gene status [58]. PQ1,

6-methoxy-8-[(3-aminopropyl)amino]-4-methyl-5-(3-trifluoromethylphen yloxy)quinoline induced apoptosis in T47D breast cancer cells through activation of caspase-8 and caspase-9 [59]. PQ15,

6-methoxy-4-methyl-8-[(4-quinolinylmethyl)amino]-5-(3-trifluoromethyl phenyloxy)-quinoline affected the survival pathway of T47D breast cancer cells [60].

31

Among the indole derivatives, indole-3-carbinol mainly exists in cruciferous vegetables such as broccoli, cabbage, cauliflower, brussels sprouts, collard greens and kale was reported with inhibitory effects on prostate tumors through inhibiting proliferation, cell cycle progression and cell survival, thus induced cell death [61]. The combined

indole-3-carbinol and genistein induced apoptosis in human colon cancer HT-29 cells [62]. Therefore, the possibility of combining indole with quinoline subunits with synergistic anticancer effect is of interest.

The reports showed that IC50 values of cisplatin for the A549 and H460 cells were 30 and 16 μM, respectively[63, 64]. Both Fig. 1B and 1C showed that IC50 values of cisplatin and EMMQ were 50 and 9 μM, respectively. In addition, EMMQ reduced cell viability by more than 50%

at 9μM in A549 cells (Fig. 1C). Our data showed that, compared to cisplatin, EMMQ is more potent in inhibiting the growth of A549 and H460 cells. In many types of cancer, DNA repair systems have been entirely or partially switched off but not in the normal tissue cells [65].

Gene regulation concerning translation, transcription, cell cycle, and the functions were regulates in MRC-5 cells by UVB irradiation [66]. The study further validated that DNA repair system is maintained.in MRC-5 and H1299 cells, but not in A549 and H460 cells as indicated by comet assay (Fig 3A and 3B). The deficient DNA integrity accounts for the sensitivity and the effectiveness of EMMQ.

32

The results showed that the sensitivity of the EMMQ in A549 and H460 cells is dependent on status of p53 that led to reduction of cell viability and tumor growth through apoptotic cell death (Fig. 1B, 8A and 9B). The EMMQ induced apoptosis in A549 and H460 cells at low concentrations was attributed to transient p53 elevation and PARP cleavage because of mitochondria perturbation (Fig. 6 and 7). Put altogether, the results suggest that EMMQ has a more powerful growth inhibitory effect than cisplatin by activating intrinsic pathway cell death in wild-type p53 NSCLC cells.

In this work using NSCLC cells, we showed that the interference of ΔΨm in EMMQ-treated cells appeared not only in A549 cells, but also in H1299 cells with stable expression of ectopic p53 (Fig.7A and 9B).

However, ΔΨm was not affected in p53-null H1299 and H1299/p53^R267P cells (Fig.7A and 9B). The results showed that EMMQ-induced cell apoptosis are depends on the status of p53 through permeabilization of the mitochondria and release of cytochrome c into the cytoplasm (Fig. 7B and 11B). However, release of cytochrome c was not affected in p53-null H1299 and H1299/p53^R267P cells (Fig. 11B). The results suggest that p53 may trigger permeabilization of the outer mitochondrial membrane through direct activation of proapoptotic Bcl-2 proteins Bax or Bak or through binding and inactivating of antiapoptotic Bcl-2 proteins such as Bcl-2 or Bcl-XL [67-69]. The interference of ΔΨm and inactivating of antiapoptotic Bcl-2 proteins led to release of cytochrome c [70-73].

33

Activation of p53 can occur in response to a number of cellular stresses, including DNA damage, hypoxia and nucleotide deprivation.

Several forms of DNA damage have been shown to activate p53,

including those generated by ionizing radiation (IR), radio-mimetic drugs, ultraviolet light (UV) and chemicals such as methyl methane sulfonate [29]. In the present study, we confirmed that EMMQ has cytotoxic effects on A549 and H460 cells, and we also demonstrated that EMMQ induced DNA damage (Fig. 3) and activation of p53 in vitro and vivo (Fig. 6, and 8) and all the effects were dose-dependent. Our data also showed that EMMQ induced cell cycle arrest in G2/M at 24 h and apoptosis at 48 h in A549 and H460 cells (Fig. 4 and 5). In the present study, we confirmed that the sensitivities of EMMQ can be offset by knocking-down p53 (Fig.

12) and the final apoptotic cell death reduced (Fig. 13). On the other hand, activate of p53 reduced expression levels of Bcl-2 and Akt [74]. Our

study confirmed that EMMQ reduced levels of Aktthat were reverted by knocking down p53 (Fig. 13). These results suggested EMMQ-mediated apoptosis in the presence of p53.

The PI3K/Akt pathway is a key signaling pathway involved in cell proliferation and differentiation, and is activated in many cancers [75].

More accumulating evidences suggested that signaling through PI3K/Akt is critically involved in trophoblast invasion and migration [76, 77].

Activation of PI3K/Akt pathways regulated of cell adhesion molecules E-cadherin and β-catenin, contributing to the attenuation of cell-cell adhesion and promoting the enhanced motility and migration [78]. The

34

induction of EMT by nuclear β-catenin has been explored during

development in many cell lines and tumors [79]. Our data showed that the Akt and β-catenin level were decreased as the concentrations of EMMQ were increased in A549 cells (Fig. 14A).Fig. 14A demonstrated that EMMQ inhibited cell-cell adhesion and EMT by decreasing β-catenin expression.

MMP plays an important role in tumorigenesis, including early carcinogenesis events, tumor growth and tumor invasion and metastasis.

Cell surface localization and activation of MMP is essential for cells to migrate, through rearrangement of ECM to suit cell migration. Certain MMP, such as gelatinases and membrane type 1 MMP, have special mechanisms to localize at leading edges in both types of cell migration [80, 81]. The growth inhibition in H460 and A549 cells was between 8 and 10µM. The data demonstrated that EMMQ inhibited metastasis with concentration of less than 5 µM in in A549 and H460 cells (Fig 14) by migration assay. The results suggested that EMMQ-inhibited metastasis may not be caused by apoptosis. More experiments are needed to confirm the drug effectiveness in metastasis

35

VII. CONCLUSION

This study supports the role of a small molecule indolylquinoline as a novel regulator against NSCLC. The sensitivity and specificity of EMMQ against NCSLC dells depend on the status of p53by activating intrinsic mitochondrial pathway (Fig. 15). This study extends the feasibility of the use of EMMQ as cancer therapeutics specific against malignant cells, and more extensive exploration on theunderlying

mechanisms that control the function of the targeted p53 pathway should provide broadened scope prior to clinical application of the drug. This therapeutic approach with EMMQ may have the potential implication and new alternative to overcome limits of current available therapies.

36

VIII. FIGURE AND LEGENDS (A)

(B)

37

(C)

Figure 1 The structure of a new synthetic compound EMMQ and its inhibitory effect on cell growth of wild type p53 NSCLC cell lines. (A) The chemical structure of EMMQ. (B) Sensitivity of H460 cells to

cisplatin. The cells were cultured in a 96-well plate for 48 h, and then were treated with different concentrations of cisplatin as indicated. After 48 h of treatment, MTT cell viability assay was performed and the blue formazan dye produced by the viable cells was quantified by measuring the absorbance of 540 nm after elution by DMSO. The DMSO treated cells were served as 100% control. Data are represented as the mean values ± standard deviation (SD). The asterisk ( ) p< 0.05 and ( ) p <

0.01 indicated a significant difference from the vehicle control. (C) Treatment with various concentrations of EMMQ in A549, H1299, H460 and MRC-5 cells.Symbol (D) indicates DMSO.

38

(A)

(B)

39

Figure 2 EMMQ treatment inhibits growth of wild type p53 NSCLC cell lines. (A) Colony formation assay. The seeded cells in 6-well plates were treated with 5, 8 and 10 μM of EMMQ, respectively, for 48 h. The media were replaced with fresh media and incubated at 37 °C. After 24 days, the cells were fixed and stained with 10% methylene blue in 70%

ethanol for counting colony numbers. (B) Statistical analysis of colony formation. The numbers of colonies of DMSO treated cells were served as 100% control. Data are presented as the mean values ± standard

deviation (SD). The asterisk ( ) p< 0.05 indicates a significant difference from the vehicle control.Symbol (D) indicates DMSO.

40

(A)

(B)

41

Figure 3 EMMQ treatment induces DNA damage in A549 and H460 cells. (A) Representative profiles of comet assay. EMMQ- induced DNA damage in human NSCLC cells were examined by comet assay. The A549, H1299 and H460 cells were seeded in 12-well plates and incubated with various concentrations of EMMQ for 24 h. The cells were collected and DNA damage was determined by comet assay as described in the Materials and Methods. (B) Statistical analysis of comet assay. The asterisk ( ) p< 0.01 and ( ) p< 0.001 indicates a significant difference from the vehicle control. Symbol (D) indicates DMSO.

42

(A)

43

(B)

44

Figure 4 EMMQ treatment induces change in cell cycle distribution in A549 and H460 cells. (A) A549, H1299 and H460 cells were treated with various concentrations of EMMQ for 48 h. Cells were stained with PI and detected the cell cycle distribution by flow cytometry. The

populations of cells in sub-G1 phase increased on EMMQ treatment and the populations were at the expenses of G1 cells. (B) A549 cells treated with EMMQ resulted in a significant increase in the proportion of cells at the G2 phase at 8 μM for 24 h.Symbol (D) indicates DMSO.

45

(A)

(B)

46

Figure 5 Apoptosis induced by EMMQ in NSCLC cell line. (A)

Apoptosis assay: A549, H1299 and H460 cells were treated with EMMQ for 48 h and analyzed by flow cytometry after double staining with Annexin V-FITC/PI. The top right quadrant indicates late apoptosis and the bottom right early apoptosis. (B) The quantitative determination of apoptosis cell population. Early (dark) and late (light) apoptotic cell populations are shown for different cell lines. Symbol (D) indicates DMSO.

47

(A)

(B)

Figure 6 EMMQ treatment induces apoptotic cell death in intrinsic pathway related proteins in A549 and H460 cells. (A) The protein lysates of the collected A549, H1299 and H460 cells treated with 5, 8 and 10 μM of EMMQ for 48 h were analyzed by Western blot as specified in Materials and Methods. (B) Cells treated with 5 μM EMMQ were

collected at different time points (12, 24 and 48 h) and were analyzed by Western blot.Symbol (D) indicates DMSO.

48

(A)

(B)

49

Figure 7 EMMQ treatment causes apoptosis through reduced ΔΨm and enhanced cytochrome c release in A549 and H460 cells. (A) Evaluation of ΔΨm following EMMQ treatment. Cells treated with 5, 8 and 10 μM of EMMQ were evaluated for changes in ΔΨm in both A549 (8 and 12 h) and H1299 cells (12 h). (B) The induced release of

cytochrome c from mitochondria to cytosol in A549 and H460 cells treated with 2 or 5 μM EMMQ or DMSO vehicle control for 24 h. The cells were stained with Mitotracker Green (mitochondrial staining), DAPI (blue, nuclear staining) and antibody conjugated with TRITC for

cytochrome c from mitochondria to cytosol in A549 and H460 cells treated with 2 or 5 μM EMMQ or DMSO vehicle control for 24 h. The cells were stained with Mitotracker Green (mitochondrial staining), DAPI (blue, nuclear staining) and antibody conjugated with TRITC for