Akt/protein kinase B (PKB), which is included in phosphatidyl inositol-3-OH kinase (PI3K) signaling, controls many intracellular processes, such as the suppression of apoptosis and the promotion of the cell cycle. Therefore, the authors investigated phosphorylated Akt (Ser473) in colorectal carcinomas to reveal the role of PI3K signaling during the development of colorectal carcinoma.
Expression of phosphorylated Akt (Ser473) in two colon carcinoma cell lines (DLD-1 and Colo205) and 65 human colorectal carcinomas was analyzed using western blotting and immunohistochemistry, respectively. Growth inhibition and induction of apoptosis caused by LY294002, a specific inhibitor of PI3K, were also examined in these cell lines. In tumor samples, the level of cell proliferation activity (Ki-67) and number of apoptotic bodies (single stranded DNA) were determined by counting positive cells.
LY294002 significantly affected the proliferation and apoptosis of Colo205 cells, suggesting an association with the low phosphorylation level of Akt protein. Immunohistochemic analysis showed that 46% of the tumors had a high level of expression of phosphorylated Akt with a close association with Ki-67 proliferative activity (P < 0.001) and the number of apoptotic bodies (P < 0.001). Akt phosphorylation was also correlated with some clinicopathologic parameters of the malignancies, including depth of invasion, infiltration to venous vessels, lymph node metastasis, and clinicopathologic stage.
Akt/protein kinase B (PKB) plays a critical role in controlling the balance between cell survival and apoptosis.1 Phosphorylation of Akt is promoted by phosphatidylinositides converted by phosphatidylinositol-3-OH kinase (PI3K) products, which subsequently facilitate transmembrane signaling by serving as membrane-localization elements to recruit target proteins to specific sites in response to mitogens, especially in Ras signaling and adhesion,1, 2 as mutant Ras can transform fibroblasts via PI3K/Akt signaling.
Akt is activated by various growth/survival factors, such as epidermal growth factor (EGF), platelet-derived growth factor, and interleukins.1–3 Three members of the Akt family (Akt1, 2, and 3) have been identified, each of which contains a Src homology 2 or aminoterminal pleckstrin homology (PH) domain.4, 5 The PH domain can mediate lipid-protein and/or protein-protein interactions as well as hydrophobic and proline-rich domains at the carboxyl terminus, resulting in the activation of signal transmission for cell proliferation and escape from apoptosis.6–8 Four phosphorylation sites have been identified at the PH domain of Akt: Ser124, Thr308, Thr450, and Ser473.6, 9 Thr308 and Ser473 are inducibly phosphorylated after treatment of cells using extracellular stimuli, whereas Ser124 and Thr450 appear to be basally phosphorylated. Additionally, mutagenesis studies have shown that phosphorylation of Thr308 and Ser473 is required for Akt activation and that mimicking phosphorylation partially activates Akt.6
Previous reports have shown that Akt delivers antiapoptotic survival signals by phosphorylating Bad and activating caspase-9.10, 11 In addition, inhibition of Akt signaling can induce apoptosis in some human cancer cell lines.12, 13 It has also been reported that Akt/PKB plays an important role in cell survival when cells are exposed to different apoptotic stimuli, such as growth factor withdrawal, ultraviolet radiation, matrix detachment, cell cycle discordance, and DNA damage.14–18 Moreover, inactivation of PI3K using a specific inhibitor has led to dephosphorylation of Akt at Ser473, consequently causing translocation of Akt to the nucleus, where it was believed to regulate the transcription of genes mediating cell survival or apoptosis.14 Although these events have been observed in some human cancer cells, including nonsmall cell lung carcinoma, prostate carcinoma, and pancreatic carcinoma cells,15–19 the function of Akt in human colorectal carcinoma has not been clarified.
In the current study, we focused on the role of Akt in cell growth and apoptosis in colon carcinoma cell lines using a specific PI3K inhibitor, LY294002. We also analyzed the association between Akt phosphorylation and clinicopathologic findings as well as cell-proliferative activity and apoptotic bodies to evaluate the importance of Akt in the development and progression of human colorectal carcinoma.
MATERIALS AND METHODS
Cell Culture and Tissue Samples
Two colon carcinoma cell lines, DLD-1 and Colo205, were used in the current study. The cells were maintained in RPMI1640 (Life Technologies, Inc., Grand Island, NY) supplemented with 1 mM L-glutamine, 10% fetal bovine serum (FBS; Life Technologies, Inc.), and 12.5 μg/mL gentamicin (Sigma Chemical Co., St. Louis, MO). The cells were then incubated at 37 °C in a humidified atmosphere of 95% air and 5% CO2.
Prior to analysis, DLD-1 and Colo205 cells were seeded at a density of 1 × 107 cells, incubated in low-nutrient medium (RPMI1640 supplemented with 0.5% FBS) for 48 hours, and subsequently treated with LY294002 (Sigma Chemical Co.), EGF (Upstate Biotechnology, Lake Placid, NY), and tumor necrosis factor-α (TNF-α; Upstate Biotechnology). LY294002 was dissolved in dimethyl sulfoxide (DMSO) at a stock concentration of 10 mM, stored at −20 °C, and added to cell cultures; we found that the final concentration of DMSO used did not affect cell survival or protein phosphorylation. Additionally, EGF and TNF-α were dissolved in distilled water at a concentration of 100 ng/μL.
For immunohistochemic examination, a total of 65 colorectal tumors removed at Yamagata University Hospital were analyzed. Informed consent was obtained from all patients before surgery. The clinicopathologic findings are summarized in Table 1. In particular, there were 38 male and 27 female patients having a mean age of 67.3 years (range, 43–84 years). The mean tumor size was 32.8 mm in diameter (range, 5–183 mm). Formalin-fixed, and paraffin-embedded tissue samples were used for histologic diagnosis and immunohistochemical analyses. Histologic classification and clinicopathologic staging were performed according to the General Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus along with the International Union Against Cancer classification (Table 1).20 This study was carried out in accordance with the ethics code for Human Experimentation, Yamagata University School of Medicine.
Table 1. Expression of Phosphorylated Akt (Ser 473) in Human Colorectal Carcinoma. Correlation with the Status of bcl-2, p53, Survivin, and Clinicopathologic Findings
Cell growth in the presence or absence of experimental agents was determined using the Premix WST-1 Cell Proliferation Assay System (Takara Biochemicals, Tokyo, Japan). The new tetrazolium salt WST-1 was evaluated for use in a colorimetric assay of cell viability and proliferation, as were other tetrazolium salts, such as MTT and XTT. Briefly, 1 × 105 cells (100 μL volume/well) maintained in phenolred-free medium for 24 hours were inoculated into 96-well microtitier plates. Next, LY294002 (20 μM), EGF (25–100 ng/mL), and TNF-α (100 ng/mL) were added under various conditions to triplicate wells and cultured at 37 °C for 24 hours. Following treatment, 10 μL of Premix WST-1 was added to each microculture well, and plates were incubated for 30 minutes at 37 °C, after which absorbance at 450 nm was measured using a microplate reader.
Cells were lysed in a buffer containing 50 mM Tris-HCl (pH 7.4), 125 mM NaCl, 0.1% Triton X-100 (Wako Pure Chemical Industries, LTD., Osaka, Japan), and 5 mM ethylenediamine-tetraaetic acid (EDTA) containing both 1% protease inhibitor and 1% phosphatase inhibitor cocktail II (Sigma Chemical Co.). For western blotting, 50 μg of protein extracted from cells that were treated with LY294002, EGF, and TNF-α underwent WST-1 assay performed on a sodium dodecyl sulfate-polyacrylamide gel followed by electrotransferral onto a Sequi-Blot PVDF Membrane (BioRad Laboratories, Hercules, CA). Rabbit polyclonal antibodies for phosphorylated Akt (Ser473) were purchased from New England BioLabs (Beverly, MA). Also, horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG) (Jackson Immuno-Research Laboratories Inc., West Grove, PA) was used as a secondary antibody for enhanced chemiluminescence (NEN Life Science Products, Inc., Boston, MA).
Apoptotic Body, DNA Fragmentation, and Caspase-3 Activity
Apoptotic (sub-G1 phase) DLD-1 and Colo205 cells were counted after staining with propidiumiodide (Molecular Probes, Eugene, OR) and fixation with 4% paraformaldehyde. To detect apoptotic DNA fragmentation, genomic DNA was extracted from members of both cell lines. Treated cells were electrophoresed in 1% agarose gels as described elsewhere.21 The cells were then lysed in a buffer containing 50 mM Tris-HCl, pH 7.8, 10 mM EDTA, and 0.5% sodium N-lauroylsarcosinate. Next, the lysates were incubated with 1 μg RNase A and then 1 μg of proteinase K at 50 °C for 30 minutes each.
In addition, we used the Caspase Colorimetric Substrate/Inhibitor (BioMol Research Laboratories, Inc., Plymouth Meeting, PA) for the detection of caspase-3 activity. Briefly, 5 × 106 cells were exposed to LY294002, EGF, or TNF-α for 24 hours and then suspended in cell lysis buffer (50 mM HEPES [pH 7.4] 0.1% CHAPS, 1 mM DTT, 1 mM EDTA; the cell lysate was then centrifuged at 10,000 × g at 4 °C for 10 minutes. A total of 50 μL of assay buffer (100 mM HEPES [pH 7.4] 200 mM NaCl, 0.8% CHAPS, 20 mM DTT, 2 mM EDTA, 20% glycerol) and 10 μL of Ac-DEVD-pNA substrate (BioMol) were added to triplicate culture wells in 96-microwell plates. Ac-DEVD-CHO (BioMol), an aldehyde caspase-3-specific inhibitor, was also used to test the specificity of caspase-3. The plates were then incubated at 37 °C for 6 hours, and the level of caspase-3 activity in the lysate was determined according to the absorbance at 405 nm measured using a microwell plate reader.
Immunohistochemistry was performed to study apoptotic bodies and altered protein expression in colon carcinoma tissues. Rabbit polyclonal antibodies against phospho-Akt (Ser473; Cell Signaling Technology, Beverly, MA), single-stranded DNA (ssDNA; DAKO, Glostrup, Denmark), mouse monoclonal antibodies against bcl-2 (DAKO), p53 DO-7 (Immunotech, Marseille, France), Ki-67 (Immunotech), and a goat polyclonal antibody against survivin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were used as primary antibodies. A modified version of the immunoglobulin enzyme bridge technique was used as described elsewhere.22 Deparaffinized tissue sections were immersed in 10 mM sodium citrate buffer, and pretreatment autoclaving was performed to retrieve antigenicity in all but the ssDNA sections; endogenous peroxidase activity was blocked using methanol containing 0.03% hydrogen peroxide. Following incubation with a blocking buffer, the sections were incubated with a primary antibody overnight at 4 °C. EnVision peroxidase anti-rabbit IgG (DAKO) and avidin DH-biotinylated horseradish peroxide complex (Histofine kit; Nichirei, Tokyo, Japan) were used as secondary antibodies. Peroxide staining was performed for 2–3 minutes using a solution of 3,3-diaminobenzidine tetrahydrochloride in 50 mM Tris-HCl (pH 7.5) containing 0.001% hydrogen peroxide. The sections were counterstained with hematoxylin.
The degree of immunostaining of Akt (Ser473) was evaluated as follows; −, nearly no positive cells; +, 5%–25% of tumor cells showing weak to moderate immunoreactivity; ++, 25%–50% of cells showing moderate to intense immunoreactivity; and +++, > 50% of tumor cells showing intense immunoreactivity. The mean percentage of positive tumor cells was determined in at least five areas at ×400 magnification (high power field [HPF]). Furthermore, the Ki-67 labeling index (growth index [GI]) was determined by counting the positive cells in a total of 1000 tumor cells observed in ≥ 10 representative HPFs, while the apoptotic body index (AI) was determined by counting the ssDNA-positive cells in ≥ 10 HPFs. Immunostaining was evaluated independently.
The relationship between the results of the immunohistochemic study and the clinicopathologic parameters was determined using the Spearman rank correlation coefficient. Also, multivariate analysis was carried out using the logistic model of the StatView-J 5.0 software program (SAS Institute, Cary, NC). A P value < 0.05 was considered to be statistically significant. Values were presented as the mean ± standard deviation.
Growth Inhibitory Effect of LY294002 and Expression of Phosphorylated Akt in Colon Carcinoma Cell Lines
We first examined the role of PI3K in cell proliferation in colon carcinoma cell lines. When 20 μM LY294002 was added to the DLD-1 and Colo205 cell lines after serum-free conditioning for 24 hours, cell proliferation was significantly decreased in both lines (Fig. 1A). Also, the mean decrease in WST-1 absorbance was 57% in DLD-1 cells, while a remarkable decrease of 89% was detected in Colo205 cells. This reduction in cell proliferation was not recovered by EGF. Additionally, the growth suppression by TNF-α in these colon carcinoma cells was more than that by LY294002 (Fig. 1A). The mean decrease in WST-1 absorbance due to TNF-α was 3% in DLD-1 cells and 45% in Colo205 cells.
Next, Akt activity was assessed using western blotting with phosphorylation-specific antibodies against Akt at Ser473. We found that although the Akt phosphorylation level was constitutively active in DLD-1 cells, Colo205 cells showed fundamentally low levels of Akt phosphorylation (Fig. 1B).
Induction of Apoptosis and Activation of Caspase-3 by LY294002 in Colon Carcinoma Cells
We counted the apoptotic bodies in the current series to show the association between growth suppression and apoptosis caused by LY294002 (Fig. 2A). We found that LY294002 did not influence the induction of apoptosis in DLD-1 cells (< 5%), whereas apoptosis was significant in Colo205 cells (71%). In general, the ratio of apoptotic cells was good when correlated with the growth-inhibition rate observed above (Fig. 1A). Additionally, EGF could not prevent apoptosis in either cell line under the conditions caused by LY294002. Interestingly, the apoptotic effect of TNF-α was a little greater than that of LY294002 in DLD-1 cells, but the growth-inhibitory function was not. Furthermore, an apoptotic DNA ladder was detected in Colo205 cells but not in DLD-1 cells (Fig. 2B).
We also performed a caspase-3 activity assay to reveal the participation of this proapoptotic factor in cell apotosis. We detected activation of caspase-3 in Colo205 cells treated with LY294002 in the presence or absence of EGF (Fig. 3). However, DLD-1 cells showed significantly low caspase-3 activity both with and without LY294002 treatment. TNF-α did not influence caspase-3 activity in either DLD-1 or Colo205 cells.
Phosphorylation of Akt in Human Colorectal Carcinoma Tissues
Phosphorylation of Akt in human colorectal carcinoma tissues was examined immunohistochemically. The results of this examination are shown in Table 1. In normal colorectal mucosa, phosphorylated Akt was expressed in the nucleus and cytoplasm of epithelial cells, especially those on the surface as opposed to in the deep layer of the mucosa (Fig. 4A). No significant expression was detected in fibroblasts, smooth muscle cells, or endothelial cells except for in some inflammatory cells.
In colorectal carcinoma samples of the 65 cases, 30 cases (46%) showed a high level of phosphorylated Akt expression (+++, 19 cases [29 %]; ++, 11 cases [17%]), while 35 had a low level of Akt expression (54%; +, 22 cases [34%]; −, 13 cases [20%]) (Fig. 4B). The expression of phosphorylated Akt in these cases was statistically correlated with some clinicopathologic parameters, especially venous vessel invasion (P = 0.048), depth of invasion (P = 0.0007), clinicopathologic stage (P = 0.0203), and lymph node metastasis (P = 0.0216). Also, associations between Ki-67 proliferative activity and apoptosis in colorectal carcinomas were analyzed (Fig. 4C). We found that Akt phosphorylation statistically correlated with the Ki-67 GI (P < 0.001) and AI (P < 0.001) (Fig. 5).
We investigated the status of bcl-2, p53, and survivin in this series of carcinomas to show the association with Akt phosphorylation (Fig. 4D). However, there was no significant relationship among these molecules (bcl-2, P = 0.0714; p53, P = 0.3563; survivin, P = 0.2765).
Phosphatidylinositol–3–OH kinase (PI3K) is the core of intracellular signaling in the stimulation of growth factors. Additionally, it is well known that Akt/PKB is one of the kinases that are phosphorylated and activated by PI3K.4 Activated Akt stimulates intracellular signaling in various human malignancies, which especially blocks apoptosis and/or cell proliferation.6 Previous studies have reported that the PI3K/Akt pathway is a survival factor for various human cancer cells, inducing those of carcinomas of the lung, prostate, pancreas, and breast.17–19, 23, 24 In the current study, we examined phoshorylation in cells in normal colorectal mucosa. We observed that cells on the surface of the mucosa had high phosphorylated Akt expression, while those in deep layers had low expression. These findings suggest that phosphorylated Akt may participate in cell senescence and differentiation in the colorectum.25, 26
The growth-inhibitory effect of LY294002 was observed in both DLD-1 and Colo205 cells. However, both this effect and the increase of the AI were more significant in Colo205 cells than in DLD-1 cells when they were treated with LY294002. What causes these differences in the sensitivity of cells to LY294002? First, Akt was constitutively active in DLD-1 cells in both the presence and absence of LY294002, while the expression of phosphorylated Akt was significantly lower in Colo205 cells. The expression of phosphorylated Akt was decreased a little by the effect of LY294002, and the DLD-1 cells seemed to be resistant to apoptosis. Further investigation is needed to clarify the mechanism that keeps Akt in active form in DLD-1 cells and Akt/PKB-related proteins, such as Bad, caspase-9, and glycogen synthase kinase-3.1, 2
Additionally, caspase-3 works downstream of Akt in the cell signaling pathway, subsequently leading cells to undergo apoptosis.19 In the current study, the activity of caspase-3 correlated with the growth-inhibitory and apoptosis-inducing effects of LY294002. Interestingly, the activity of caspase-3 in Colo205 cells was higher than in DLD-1 cells. Therefore, we determined that Colo205 cells have high sensitivity to LY294002, which leads to apoptosis.
To our knowledge, this is the first report to show frequent phosphorylation of Akt protein in human colon carcinomas. In addition, the high level of expression of phosphorylated Akt correlated with Ki-67 cell proliferation activity and inhibitors of apoptosis. These results indicate the requirement of Akt activation for cancer growth and the blocking of apoptosis. Moreover, activated Akt showed significant relationships with several clinicopathologic findings, suggesting that this protein may contribute to tumor progression and development in human colorectal carcinomas. Although we could not follow the prognosis of all the patients, there was a tendency for a high level of phosphorylated Akt expression to influence a poor prognosis (data not shown). Therefore, Akt status in carcinoma cases can monitor aspects of malignancies, such as proliferation rate, resistance to apoptosis, and invasion/metastasis. However, further investigation is needed to reveal the importance of Akt in other human malignancies.
Furthermore, the status of p53, bcl-2, and survivin did not correlate with Akt phosphorylation. Nuclear accumulation of mutant p53 is detected in more than half of all colon carcinomas, as well as other human malignancies.27, 28 It has been believed that this is because p53 acts as a gatekeeper when DNA is damaged by ultraviolet radiation, subsequently causing G1 arrest, DNA repair, and/or cell apoptosis. Disruption of p53 activity is led by a somatic mutation of the p53 gene and loss of heterozygosity.27–29 Although we believe that p53 inactivation is not associated with Akt phosphorylation, induction of apoptosis was detected more often in Colo205 (wild-type p53) cells than in DLD-1 (mutant p53) cells. This may be one cause of the difference between these two cell lines in sensitivity to LY294002 and induction of apoptosis. Survivin and bcl-2 also were not shown to have a relationship with phosphorylated Akt. Survivin is expressed in most cancers and lymphomas, blocking apoptosis downstream of cell death signaling.30, 31 Also, the function of the antiapoptotic factor bcl-2 is inhibited by the p53-inducible gene Bax, which inhibits bcl-2 homodimerization.32 The mechanisms that regulate the expression and function of both survivin and bcl-2 are still unknown.
In the current study, EGF (25 and 100 ng/mL) did not influence cell proliferation or apoptosis when cells were treated with 20 μM LY294002, suggesting that LY294002 effectively suppresses extracellular stimulation of growth factors. Interestingly, in both cell lines, LY294002 was more effective than TNF-α was in inhibiting growth and upregulating apoptosis as well as caspase-3 activity.33, 34 Consequently, LY294002 effectively suppressed Akt function and led cells to undergo apoptosis.
Previous studies showed the usefulness of LY294002 in nutrient-deprived pancreatic carcinoma cells.35 For example, Brongnard et al.17 have shown that modulation of Akt/PKB activity via pharmacologic or genetic approaches alters cellular responsiveness to the therapeutic modalities typically used to treat nonsmall cell lung carcinoma. The current findings also suggest the possibility of treating colon carcinoma using LY294002. We further studied the therapeutic effect of LY294002 in a mouse xenograft using a colon carcinoma cell line. We found that administration of LY294002 influenced tumor growth and induction of apoptosis both in vivo and in vitro.36 Additionally, although accumulation of genetic alterations in oncogenes and tumor suppressor genes has been reported in studies of colon multistep carcinogenesis, response to stimuli from growth factors may be a common characteristic in all carcinoma cells. The current results indicate that LY294002 may block tumor progression in spite of the patient's genetic background. They also suggest that LY294002, when combined with other anticancer agents, may be a therapeutic agent for colorectal carcinomas.
The authors thank Emile M. Youssef, Department of Thoracic/Head and Neck Medical Oncology, The University of Texas, M.D. Anderson Cancer Center, for help in preparing the article.