Regulation of Signaling Pathways Involved in the Anti-proliferative and Apoptosis-inducing Effects of M22 against Non-small Cell Lung Adenocarcinoma A549 Cells

The compound 29-(4-methylpiperazine)-luepol (M22), a novel derivative of lupeol has shown anti-proliferative effects against the human non-small cell lung cancer A549 cell line. M22 showed significant anti-proliferative activity at 6.80 μM and increased accumulation of G1 cells and effectively suppressed expression of the G1 arrest-related genes cyclins D1 and E1, CDK2 and CDC25A. This was further confirmed by Western blotting demonstrating decreased cyclin D1 and CDC25A protein levels. Furthermore, M22 caused induction of apoptosis that downregulated the anti-apoptotic BCL-2 gene and increased expression of BAX, CASP3 and CASP9 as well as the APAF1 gene. The effect of caspase-induced apoptosis was confirmed by an increase in reactive oxygen species (ROS), loss of mitochondrial membrane potential (MMP). Taken together, our findings indicated that M22 possessed potent anti-proliferative and apoptotic activities.

We further tested the long-term effect of M22 to inhibit cell viability in A549 cells using colony formation assay. The results revealed that the potent inhibitory activity of M22 in concentration-dependent manner. (Fig. 1E).

Effect of M22 on cell cycle progression.
To determine the effect of M22 on cell cycle arrest and cell death in lung cancer cells, A549 cells were treated with various concentrations of M22 for 24 h. The cells were analyzed by flow cytometry and stained with propidium iodide (PI). As shown in Fig. 2, M22 addition also resulted in a G1/G1S block reaching almost 75%. In addition, the sub G1 population was increased to 4.94% at 13 μM M22 at 24 h, indicating the drug might induce apoptosis in A549 cells (Fig. 2).
Effect of M22 on genes and proteins involved in G1 to S transition. Consistently, M22 treatment significantly decreased CCND1 and CDK4 expression (Fig. 3). This was consistent with the role of cyclin D1 to bind and activate cyclin-dependent kinases 4 and 6 (CDK4 and CDK6) which regulated the G1/S transition 13 .
In conjunction with these findings, mRNA levels of the genes for G1-related cyclin E1, PCNA and CDC25A were dramatically decreased by M22 treatment compared to the untreated control. M22 also induced down regulation of the CDC25A and cyclin D1 proteins (Fig. 3).

M22 induces apoptosis in A549 cells.
We also found a significant increase of early apoptosis and a progressive increase of late apoptosis with increasing concentrations of M22 at 48 h. Apoptotic cells increased from 23% to 57% following M22 treatment in increasing dosages (Fig. 4A, Quadrant 2 and 4). Hoechst 33258 staining result recorded that most cells showed typical apoptosis characters in M22 treated group. This change was accompanied by DNA fragmentation that was an indicator of apoptosis (Fig. 4B).
M22 also increased reactive oxygen species (ROS) in a dose dependent manner compared with vehicle-only treated cells (Fig. 5A,B). This was accompanied by a significant loss of mitochondrial membrane potential (MMP) (Fig. 5E,F). These changes were accompanied by DNA degradation that was a major indicator of apoptosis (Fig. 5C,D).

Effect of M22 on genes and proteins involved in apoptosis.
Expression of genes involved in the control of programmed cell death was evaluated in A549 cells after exposure to M22 for 48 h. We measured the steady state mRNA levels of BCL-2, BAX, APAF1, CASP9 and CASP3. All four of the apoptotic genes were upregulated while expression of the anti-apoptotic gene BCL-2 was lowered (Fig. 6A). In support of this, M22 increased cleavage of pro-Caspases 9 (46 kDa) and 3 (37 kDa) to their active forms. The cleaved caspases 9 (10 kDa) and 3 (17 kDa) are the characteristic hallmarks of apoptosis, which were detected in this study (Fig. 6B). Together these data indicated that M22 induces apoptosis.

Discussion
Lupeol treatment of cells previously showed a certain degree of cytotoxic, anti-inflammatory and tumor pro-apoptotic activities 14,15 . Lupeol could induce apoptosis in A431 and LNCaP cells through the mitochondrial cell death pathway 16,17 . The derivative M22 was reported to be about 5 times more active 12 . In the present study, we explored the underlying anti-tumor mechanism of M22 on two principal aspects, cell cycle, apoptosis, related genes and proteins. We found that M22 lowered CDC25A mRNA and protein levels were consistent with a G1 block. The CDC25A protein prevents dephosphorylation of Cdk2 and leads to the inactivation of the Cdk2/Cyclin E complex that triggers the G1 to S transition 18 . A decrease in CDC25A is consistent with our findings of a G1 block. Cdk2 activity was further regulated by p21 waf1/cip1 although we found no significant changes in the mRNA levels of p21 waf1/cip1 . However, cyclin D1 mRNA and protein levels were decreased by M22. The release of the p21 waf1/cip1 CDK2 inhibitor from the disrupted cyclin D1/Cdk4/6 complexes would result in the inactivation of the S-phase-promoting cyclin E /CDK2 19,20 complex. This was also consistent with our results with a G1 block as the result of inactivation of the S-phase promoting complex. Besides, M22 treatment reduced expression of PCNA genes that would interfere with DNA replication and repair resulting in a G1 /S arrest [21][22][23] .
We also found an increase in BAX mRNA expression with a concomitant decrease in BCL2 expression following M22 treatment, indicative of apoptosis. M22 significantly increased ROS as MMP gradually decreased. This would result in cytochrome c release from mitochondria leading to decreased levels of ATP 24 . Cytochrome c binding to Apaf-1 would activate Caspase 9 [25][26][27] in the presence of sufficient ATP or dATP. Caspase 9 was involved in the activation cascade responsible for the execution of apoptosis and Caspase 9 binding to Apaf-1 enables the latter to cleave and activate Caspase 3. Caspase 3 activation was critical for apoptosis in A549 cells because this led to nuclear DNA fragmentation. Here we showed that up-modulation of Caspases 9 and 3, Apaf1 and Bax with the down-regulation of Bcl-2. This may represent a novel pathway through which M22 induces apoptosis.

Conclusions
M22 treatment of cell cultures showed significant cytotoxicity, increased the population of sub-G1 cells and attenuated the expression of Bcl-2, Bax, pro-Caspase 9 and pro-Caspase 3. The drug also suppressed the mRNA expression of G1 arrest-related genes of cyclin D1, CDC25A and PCNA and significantly changed the expression of ROS and MMP levels in A549 cells (Fig. 7).

Methods
Reagents and Cells. M22 was synthesized in our own laboratory with a purity of 99% as assessed by HPLC.
M22 (C 34 H 58 N 2 O, MW 510.84) with a purity over 98% by HPLC was dissolved in DMSO and stored as small aliquots Cell apoptosis was determined by flow cytometry using annexin-V/PI staining (Biotine, Shanghai, China). Briefly, A549 cells were seeded in 6-well plates with density of 3 × 10 5 cells/well for 24 h and treated with M22 for 48 h. The cells were harvested and labeled with annexin V -fluorescein isothiocyanate solution and PI in binding buffer 28 . Cell fluorescence intensity was measured by flow cytometry as above. The analysis was repeated three times and the apoptosis rate was (%) for each M22 treatment was obtained.  Protein and mRNA Expression. Western blot analysis used cells lysed in RIPA buffer (Biotine, Shanghai, China) and total protein was quantified using a BCA assay (Biomed, Beijing, China). Proteins were separated in an 8-10% denaturing polyacrylamide gradient gel. Proteins were transferred to PVDF membranes   29 . The membrane was probed with antibodies for Cyclin D1, CDC25A, Caspase 3, Caspase 9 and GAPDH (Affinity Biosciences, Cincinnati, OH, USA) followed by exposure to horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies as appropriate (Earth Ox, San Francisco, CA, USA). Protein expression was determined using an Enhanced Chemiluminescence kit (Bio-Rad, Hercules, CA, USA). Total RNA was isolated using an RNeasy mini kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. RNA was reverse transcribed using TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China) and qPCR was performed using primers listed in Table 2 with the BAX System Q7 instrument (DuPont Nutrition & Health, Wilmington, DE, USA) according to the manufacturer's protocol. The mRNA level of GAPDH was used to normalize the expression of genes of interest.

ROS and MMP and TUNEL assays.
Briefly, A549 cells were suspended in PBS supplemented with 50 mM glucose for ROS detection and incubated with 10 mM DCFH-DA (Biotine, Shanghai, China) at 37 °C for 1 h 30 . The signal was assessed from fluorescence by flow cytometry (FACS Calibur, Becton Dickinson, Franklin Lakes, NJ, USA) and determined in terms of mean fluorescence intensity (MFI). The mitochondrial membrane potential was analyzed by incubating A549 cells with 5, 5′,6,6′-tetrachloro-1,1′,3,3′ tetraethylbenzimidazolylcarbocya-nine iodide/chloride (JC-1), a cationic dye that exhibits potential-dependent accumulation in mitochondria, indicated by fluorescence emission shift from red (590 nm) to green (525 nm) (Biotine, Shanghai, China). After M22 treatment, cells were stained with JC-1 and analyzed by a flow cytometry (FACS Calibur, Becton Dickinson, Franklin Lakes, NJ, USA). The ratio of mean fluorescence intensity (MFI) of red to green fluorescence was calculated for each treatment and plotted 31 .Values were given in terms of mean fluorescence intensity (MFI). DNA degradation was measured using a terminal deoxynucleotidyl transferase isothiocyanate-dUTP nick-end labeling (TUNEL) (TransGen Biotech, Beijing, China). Single cell suspension of treated and untreated A549 cells was washed with PBS and the cells (1 × 10 6 ) were fixed overnight in ethanol (90%). The cells were washed and labeled with fluorescein-dUTP for 1h 32 . The signal were assessed from fluorescence by flow cytometry (FACS Calibur, Becton Dickinson, Franklin Lakes, NJ, USA) and determined in terms of mean fluorescence intensity (MFI).
Statistical analyses. Data were presented as means ± standard deviation (SD) of three or more replicates.
Statistical significance was verified by Dunnett's multiple comparison tests using GraphPad Prism 5 software (GraphPad, San Diego, CA, USA).