Immunohistochemical expression of mTOR negatively correlates with PTEN expression in gastric carcinoma

Introduction

Despite a declining incidence rate in the United States and a number of other Western countries, gastric cancer continues to be a worldwide health problem with more than 600,000 cases reported annually, far higher than pancreatic cancer with 125,000 cases (1). Gastric cancer is the most common gastrointestinal malignancy in East Asia, Eastern Europe and parts of Central and South America and is the second leading cause of cancer-related mortality (2). Despite improvements in surgery, radiotherapy and cytotoxic chemotherapy, survival rates for advanced gastric cancer are poor. Five years after multimodal treatment, less than 40% of Western patients with stage II or III disease are likely to be alive. At metastatic stage IV disease, the mean survival is only 10 months (3).

The phosphoinositide-3 kinase (PI3K)-AKT-mammalian target of rapamycin (mTOR) pathway is an important cellular pathway involved in cell growth, tumorigenesis, cell invasion and drug response (4,5). PI3K is a heterodimer of 85- and 110-kDa subunits and has a tyrosine kinase activity, the activation of which stimulates the production of phosphatidylinositol 3,4,5-triphosphate resulting in activation of the kinases PDK1 and Akt. Akt is a kinase that phosphorylates a variety of target molecules to mediate signals, including mTOR, while mTOR phosphorylates and activates p70 S6 kinase (S6K)-1 and also inhibits eukaryotic translation initiation factor 4E-binding protein (4E-BP), resulting in enhanced protein synthesis and cell proliferation (6–8). This pathway is frequently activated in numerous types of cancer and uncontrolled PI3K-AKT-mTOR signaling may also result in a poor clinical outcome in lung, cervical, ovarian and esophageal cancers (4,6).

mTOR was identified in 1994 by several groups of investigators as the kinase targeted by rapamycin linked to the cellular protein FKBP12 (FK506-binding protein) (9). It was therefore also named FKBP-RAP-associated protein (FRAP), RAP FKBP12 target (RAFT) 1 and RAP target (RAPT) 1. mTOR is a 289-kDa, ubiquitously expressed, evolutionarily conserved serine/threonine protein kinase (9) which is important in cellular protein synthesis and energy balance, affecting numerous aspects of cell growth and proliferation, including differentiation, cell-cycle progression, angiogenesis, protein degradation and apoptosis (10). mTOR is also instrumental in protein translation initiation (the rate-limiting step of protein synthesis) by enabling the recruitment of ribosomes to mRNA by eukaryotic initiation factor. Consequently, mTOR activates its downstream mediator ribosomal S6K and is responsible for the progression of the cell from G0/G1 to S phase (11). Consistent with its essential role in cell growth, aberrant activity of the mTOR pathway is frequently observed in a number of types of cancer (12).

Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is critical in cell growth, migration and death. It is mutated or deleted at a high frequency in various human cancer tissues to promote tumorigenesis (13). The PI3K-AKT-mTOR pathway is one of the most upregulated pathways in neoplastic cells through mechanisms such as PTEN loss of function or PI3K activating mutations (14). PTEN antagonizes PI3K pathways by dephosphorylating phosphatidylinositol 3,4,5-triphosphate to convert it back to phosphatidylinositol 4,5-biphosphate. Thus, PTEN is considered to be a negative regulator of the PI3K-AKT-mTOR pathway (15).

The aim of the present study was to explore the involvement of the PI3K-AKT-mTOR signaling pathway in the progression of human gastric cancer. The expression levels of mTOR and PTEN in human gastric cancer tissues were determined using immunohistochemical study using biopsies from 33 patients and correlations with pathological parameters and prognoses were evaluated.

Materials and methods

Patients and tumor samples

Paraffin-embedded sections were obtained from patients with gastric cancer who had undergone surgery at Renmin Hospital of Wuhan University (Wuhan, China) between 2005 and 2008. Tissues obtained from 30 cases of chronic superficial gastritis diagnosed by gastroscopic biopsy were used as control samples for the immunohistochemical staining. General informed consent with regard to the use of the patients’ tissue specimens and clinical information was obtained from all patients. None of the patients recruited in this study had undergone chemotherapy or radiotherapy prior to surgery. The patients’ clinicopathological data are summarized in Table I. The histological diagnosis was determined using hematoxylin and eosin staining according to the WHO criteria (16). Pathological staging was performed according to the American Joint Committee on Cancer (AJCC) Cancer Staging Manual revised in 2010 (17). This study was approved by the Institutional Ethics Board of the Renmin Hospital of Wuhan University.

Table I. Clinicopathological parameters of the patients in the study.

Table I.

Clinicopathological parameters of the patients in the study.

Factors	Values
Age, years, mean (range)	51.6 ±11.6 (24–74)
Gender, n (%)
Male	15 (45.5)
Female	18 (54.5)
Invasive depth, n (%)
Early stage	6 (18.2)
Advanced stage	27 (81.8)
Differentiation, n (%)
Well and moderate	15 (45.5)
Poor	18 (54.5)
Lymph node metastasis, n (%)
Positive	16 (48.5)
Negative	17 (51.5)
Pathological stage, n (%)
I+II	18 (54.5)
III+IV	15 (45.5)

Immunohistochemical staining

For malin-fixed, paraffin-embedded tissue blocks obtained from human tissue were cut into 4-μm thick sections and mounted on adhesive-coated glass slides. mTOR was detected with a rabbit monoclonal anti-mTOR antibody (Cell Signaling Technology, Inc., Danvers, MA, USA) and PTEN was observed with a mouse monoclonal anti-PTEN antibody (Maxim Biotechnology Development Co. Ltd, Fuzhou, China). Sections were dewaxed in xylene and rehydrated using graded ethanol and were then incubated in 3% solution of hydrogen peroxide in methanol for 10 min to inactivate endogenous peroxidase. This was followed by an antigen retrieval step. The slides were placed in 0.01 mol/l citrate buffer solution (pH 6.0) and heated (92–100°C) for 10 min in a microwave oven. Following a wash with phosphate-buffered saline (PBS; pH 7.2) buffer, the sections were covered with 2% normal serum for 20 min at room temperature to eliminate non-specific binding of the antibody and were then incubated overnight at 4°C with the primary antibodies diluted in PBS. After washing with PBS, the secondary biotinylated antibody was added for a 20 min incubation at 37°C. Slides were then rinsed with PBS and treated with streptavidin-peroxidase solution for 10 min. Tissue sections were washed once in PBS buffer and covered with 3,3′-diaminobenzidine solution for 10 min. Finally, the specimens were counterstained with hematoxylin. Normal gastric sections served as positive controls while negative control slides were incubated with the antibody diluents instead of the primary antibody.

Evaluation of slides

Immunohistochemical staining was evaluated by 3 independent experienced pathologists who were blinded to the clinicopathological parameters and clinical outcomes of the patients. In cases of disagreement between the observers slides were re-evaluated until a consensus was achieved. The sections were examined at x200 magnification using light microscopy. The immunostaining was considered to be positive when the neoplastic cells exhibited specific immunoreactivity in the cytoplasm for mTOR or in the nucleus for PTEN. The immunostaining results were assessed semiquantitatively. For each sample the positive rate was calculated according to the percentage of positive cells of all counted cells from 5 randomly selected representative fields. Additionally, the expression was classified according to the percentage of stained tumor cells as low expression (−, <10% positive carcinoma cells), intermediate expression (+, ≥10% and <50% positive carcinoma cells) and high expression (++, ≥50% positive carcinoma cells) (18).

Statistical analysis

Statistical analysis was performed using the PASW 18.0 software program for Windows. The results for the correlation between mTOR and PTEN were evaluated using the Chi-squared test. P<0.05 was considered to indicate statistically significant differences.

Results

Differential expression of mTOR and PTEN in human gastric cancer

The expression levels and cellular distribution of mTOR and PTEN in the 33 specimens of human gastric cancer and 30 normal gastric tissues were examined by immunohistochemical staining. mTOR was distributed mainly in the cytoplasm. Staining was weaker in low-grade tumors. Positive mTOR expression was observed in tumor cells in 51.5% (17/33) of the gastric cancer patients. By contrast, little or no expression of mTOR was observed in normal gastric tissues (Fig. 1). PTEN was distributed mainly in the nuclei. Staining was stronger in low-grade tumors. Positive PTEN expression was observed in tumor cells in 54.5% (18/33) of the gastric cancer patients (Fig. 2).

Figure 1. mTOR was mainly distributed in the cytoplasm. (A) Staining was hardly detected in normal gastric tissues and (B) was weaker in well-differentiated gastric cancer but was stronger in (C) moderately and (D) poorly differentiated tumors. Original magnification, x400. mTOR, mammalian target of rapamycin..

Figure 2. PTEN was mainly distributed in the nucleus. (A) Staining was stronger in normal gastric tissue and (B) well-differentiated gastric cancer and was weaker in (C) moderately and (D) poorly differentiated tumors. Original magnification, x400. PTEN, phosphate and tensin homolog.

The expression of the mTOR and PTEN antigens was assessed by immunohistochemical staining in sections obtained from 33 gastric cancer patients with various histological diagnoses and pathological staging according to the AJCC handbook (Table II). In early and advanced cases, the respective positive expression rates were 16.7% (1/6) and 59.3% (16/27) for mTOR and 100.0% (6/6) and 44.4% (12/27) for PTEN. In well- and moderately differentiated tissues and in poorly differentiated specimens, the respective positive expression rates were 26.7% (4/15) and 72.2% (13/18) for mTOR and 73.3% (11/15) and 38.9% (7/18) for PTEN. In patients with or without lymph node metastasis, the respective positive expression rates were 81.3% (13/16) and 23.5% (4/17) for mTOR and 18.8% (3/16) and 88.2% (15/17) for PTEN. In stage I+II and stage III+IV, the respective positive expression rates were 16.7% (3/18) and 93.3% (14/15) for mTOR and 94.4% (17/18) and 6.7% (1/15) for PTEN.

Table II. Correlations between mTOR and PTEN expression and clinicopathological characteristics in gastric carcinoma cases.

Table II.

Correlations between mTOR and PTEN expression and clinicopathological characteristics in gastric carcinoma cases.

	mTOR expression			PTEN expression
Factors	Positive (%)	Negative (%)	P-value	Positive (%)	Negative (%)	P-value
Gender			0.849			0.407
Male	8 (53.3)	7 (46.7)		7 (46.7)	8 (53.3)
Female	9 (50.0)	9 (50.0)		11 (61.1)	7 (38.9)
Age (years)			0.221			0.009
<54	10 (62.5)	6 (37.5)		5 (31.3)	11 (68.8)
≥54	7 (41.2)	10 (58.8)		13 (76.5)	4 (23.5)
Invasive depth			0.085			0.021
Early stage	1 (16.7)	5 (83.3)		6 (100.0)	0 (0)
Advanced stage	16 (59.3)	11 (40.7)		12 (44.4)	15 (55.6)
Differentiation			0.009			0.048
Well and moderate	4 (26.7)	11 (73.3)		11 (73.3)	4 (26.7)
Poor	13 (72.2)	5 (27.8)		7 (38.9)	11 (61.1)
Lymph node metastasis			0.001			0.000
Positive	13 (81.3)	3 (18.8)		3 (18.8)	13 (81.3)
Negative	4 (23.5)	13 (76.5)		15 (88.2)	2 (11.8)
Pathological stage			0.000			0.000
I+II	3 (16.7)	15 (83.3)		17 (94.4)	1 (5.6)
III+IV	14 (93.3)	1 (6.7)		1 (6.7)	14 (93.3)

[i] mTOR, mammalian target of rapamycin; PTEN, phosphate and tensin homolog.

The expression of mTOR had a significant positive correlation with differentiation, lymph node metastasis and clinical staging (P<0.01), but was independent of gender, age and invasive depth (P>0.05). The expression of PTEN was negatively correlated with invasive depth and differentiation (P<0.05) and significantly negatively correlated with age, lymph node metastasis and clinical pathological staging (P<0.01), but was not associated with gender (P>0.05).

Correlation between mTOR and PTEN in human gastric cancer

When the expression of mTOR was analyzed with regard to PTEN expression, the staining pattern was divided into 4 groups: mTOR−/PTEN−, mTOR+/PTEN+, mTOR+/PTEN− and mTOR−/ PTEN+. A comparison of the mTOR−/PTEN+ and mTOR+/PTEN− groups (Table III) revealed that that in early gastric cancer, the size of the former group was larger than the that of latter and that the differences between the 2 groups were statistically significant with regard to invasive depth (P=0.041). In well- and moderately differentiated gastric cancer, the mTOR−/PTEN+ group was larger than the mTOR+/PTEN− group and the difference between the 2 groups was statistically significant with regard to differentiation (P=0.012). In patients without lymph node metastasis, the mTOR−/PTEN+ group was larger than the mTOR+/PTEN− group and the difference between the 2 groups was statistically significant with regard to lymph node metastasis (P=0.000). In stage I+II, the mTOR−/PTEN+ group was larger than the mTOR+/PTEN− group and the difference between two groups was statistically significant with regard to pathological stage (P=0.000).

Table III. Co-expression of mTOR and PTEN and clinicopathological characteristics.

Table III.

Co-expression of mTOR and PTEN and clinicopathological characteristics.

Factors	mTOR−/PTEN+	mTOR+/PTEN−	P-value
Invasive depth
Early stage	5	0	0.041
Advanced stage	9	13
Differentiation
Well and moderate	10	3	0.012
Poor	4	10
Lymph node metastasis
Positive	2	12	0.000
Negative	12	1
Pathological stage
I+II	14	0	0.000
III+IV	0	13

[i] mTOR, mammalian target of rapamycin; PTEN, phosphate and tensin homolog.

Discussion

Previous studies have suggested that the PI3K-AKT-mTOR pathway is frequently activated in various types of cancer and that this pathway is considered to be important for cancer cell survival, proliferation, angiogenesis and resistance to chemotherapy (19–21). However, the activated molecule of the PI3K-AKT-mTOR pathway in gastric cancer has not yet been studied. In the present study, 33 cases of gastric cancer were investigated and statistical analyses were performed concerning the correlation between the clinicopathological parameters in gastric cancer and the immunohistochemical expression levels of mTOR and PTEN. The results indicated that mTOR and PTEN were negatively correlated in the pathogenesis of gastric cancer. The overexpression of mTOR and low expression of PTEN proteins were strongly correlated with the pathological staging. These results suggested that mTOR and PTEN may be clinically useful prognostic markers and may provide additional information for the histological diagnosis and pathological staging of gastric cancer.

Possible correlations between mTOR expression in gastric cancer and pathological parameters were investigated. The present study demonstrated that mTOR was activated in human gastric cancer and was significantly correlated with invasive depth, differentiation and lymph node metastasis, suggesting that the high expression of mTOR contributes to the progression and metastasis of gastric cancer. Positive mTOR expression was detected in tumor cells in 51.5% (17/33) of gastric cancer patients, while little or no expression was observed in normal gastric tissues. Furthermore, 93.3% (14/15) of gastric cancer patients had positive expression of mTOR at stage III+IV suggesting that the hyperactivation of mTOR kinase was a late event in the development of gastric cancer. Similarly, the positive expression rate of mTOR was high for those patients with lymph node metastasis (81.3%, 13/16). mTOR immunoreactivity intensity data revealed that the high expression levels of mTOR significantly increased with the tumor progression.

It has been reported that mTOR is a powerful oncoprotein overexpressed in numerous types of cancer, including hepatocellular carcinoma (22), lung cancer (23), esophageal squamous cell carcinoma (24) and breast cancer (25,26). Furthermore, patients with breast cancer and mTOR overexpression had a risk of recurrence 3 times greater than that of patients without mTOR overexpression (25,27). The mTOR inhibitor everolimus has demonstrated promising clinical efficacy in a phase III randomized and double-blind trial in patients with metastatic renal cell cancer (28). The exact mechanisms for the overexpression of mTOR in carcinoma remain unclear. As a highly conserved, ubiquitously expressed signaling molecule, mTOR is activated downstream of multiple distinct growth factor receptors and is crucial for mediating cell proliferation and survival (29). Several mTOR inhibitors, including rapamycin/sirolimus (Wyeth) and derivatives, such as RAD001/everolimus (Novartis), CCI-779/temsirolimus (Wyeth) and AP23573 (Ariad), are being developed as anti-cancer agents against various types of malignancies (30).

By contrast, PTEN was shown by immunohistochemistry to be expressed in normal gastric tissues and almost all the early gastric cancer cases. The underexpression of PTEN was significantly correlated with invasive depth, differentiation and lymph node metastasis. Approximately one-half of the gastric cancer patients exhibited a loss of PTEN expression. Additionally, the loss of PTEN expression was correlated with invasive depth and differentiation (P<0.05) and closely correlated with lymph node metastasis and clinical pathological staging (P<0.01). PTEN immunoreactivity intensity data were also analyzed and the results revealed that the high expression levels of PTEN significantly decreased with the tumor progression.

PTEN, which is located at human chromosome 10q23, has been identified as a tumor suppressor gene and an important negative regulator of the PI3K-AKT-mTOR signaling pathway that promotes cell proliferation and inhibits apoptosis (31). Inactivation of PTEN by mutation, deletion and promoter hypermethylation has been demonstrated in a range of cancer types, including lung, breast, prostate and esophageal carcinomas (31–35). Langlois et al observed that PTEN controls the cellular polarity, establishment of cell-cell junctions, paracellular permeability, migration and tumorigenic potential of human colorectal cancer cells (36). Varied analysis of colorectal carcinomas suggested that the patients without PTEN expression had shorter survival times than the patients with PTEN expression (P=0.003) (37). Abnormal expression of PTEN may predict the metastasis and prognosis of gastric cancer (21,38,39).

We concluded that mTOR facilitated the development of gastric cancer while PTEN, a tumor suppressor gene, was able to inhibit tumor invasion and metastasis. mTOR and PTEN co-regulate the progression of tumors and participate in proliferation, invasion and metastasis in gastric cancer. Bakarakos et al discovered that the loss of PTEN and activation of mTOR was closely correlated with breast cancer (40). In vitro studies suggested that PTEN is capable of inhibiting cell proliferation and promoting apoptosis via inhibition of the activity of the PI3K-Akt-mTOR pathway (41). The combined deletion of PTEN and Lkb1 in the mouse bladder significantly activated the mTOR pathway and increased bladder epithelial cell proliferation and tumorigenesis (42). When we compared the mTOR-/PTEN+ and mTOR+/PTEN− groups, the differences between them were statistically significant with regard to invasive depth, histological type, lymph node metastasis and pathological stage. Consequently, collaborative detection of mTOR and PTEN expression may be more useful in the diagnosis of gastric cancer.

In summary, upregulated expression of mTOR and downregulated expression of PTEN were involved in carcinogenesis and progression of gastric cancer. A negative correlation between mTOR and PTEN expression implied that their modified expression may be important in the pathogenesis, invasion and metastasis of carcinoma tissue. Combined detection of mTOR and PTEN expression may be used to evaluate the degree of malignancy in gastric cancer, which may be a useful marker for the early diagnosis of gastric cancer. Further studies with more patients, including follow-up and different molecular biomarkers in addition to these two molecules, would aid the clarification of the disease pathogenesis and identification of potential therapeutic approaches.

Acknowledgements

The author Min Li gratefully acknowledges the assistance of his elder sister Li Li for her critical reading of the manuscript before its submission. We also gratefully acknowledge the assistance of Xinyu Qin, Huawen Sun and Lujun Song in the preparation of this study.

References