This work is carried out at the Advanced Medical and Dental Institute, Universiti Sains Malaysia and supported by Sciencefund, MOSTI, Malaysia.

The mammalian target of rapamycin, commonly known as mTOR, is a serine/threonine kinase which belongs to the family of phosphatidylinositol kinase-related kinase (PIKK). TOR was discovered in Saccharomyces cerevisiae as product of a gene whose mutation confers resistance to growth inhibition by the macrolide antibiotic, rapamycin (Kunz et al 1993). mTOR plays a critical role in coordinating cell growth in response to mitogens, amino acid and energy sufficiency (Schmelzle & Hall 2000). The pathway leading to mTOR activation by these factors, usually results in an increased protein biosynthesis, cell growth and G1 cell cycle progression (Seeliger et al 2007). The cell growth promoting effects of mTOR are mediated by two proteins which regulate protein translation, namely p70S6K and 4E-BP1 (Proud 2004). mTOR phosphorylates p70S6K which leads to the phosphorylation of ribosomal protein S6 and subsequently to the translation of mRNA which encodes for ribosomal proteins and elongation factors (Jacinto & Hall 2003; Proud 2004). 4E-BP1, a binding protein and translational repressor, upon phosphorylation by mTOR, will release the initiation factor eIF4E. Phosphorylation of 4E-BP1 disrupts its interaction with the eIF4E translation initiation factor, allowing eIF4E to bind to the cap structure at the 5’end of mRNAs, which subsequently promotes ribosome recruitment and initiation of translation (Gera et al 2004). In a recent study, p-4E-BP1 was found to be the main factor in signaling pathways which was associated with prognosis and grade of malignancy in breast tumors. Expression of p-4E-BP1 was correlated with a high tumor proliferation rate (Rojo et al 2007).

mTOR activation by mitogens is mediated by upstream phosphatidylinositol 3-kinase-AKT pathway (Chiang & Abraham 2005). The first event in the activation process of mTOR is the binding of growth factors (eg. insulin) to their receptors. In response to that, class I PI3Ks promote the conversion of the membrane lipid phosphatidylinositol 4, 5 biphosphate (PIP2) into phosphatidyl-inositol 3,4,5-triphosphate (PIP3) (Seeliger et al 2007). Activation of AKT by PIP3 is a central event in the signaling cascade. However PIP3 can be reconverted to PIP2 by the lipid phosphatase activity of the tumor suppressor PTEN (Neufeld 2003). PTEN is frequently mutated in many cancers and in a group of cancer-like syndromes that are characterized by the emergence of hamartomas (Inoki et al 2005). A critical outcome of PTEN inactivation is an increase in mTOR activity, resulting in the phosphorylation of 4E-BP1 and S6K, and subsequently uncontrolled cell growth. The Akt pathway is found to be activated early in breast cancer cases (Bose et al 2006).

Interestingly, mTOR kinase is also a key regulatory component that controls the induction of autophagy. Inhibition of mTOR (by nutrient-depletion, starvation and rapamycin) leads to cell cycle arrest, inhibition of cell proliferation, immunosuppression and induction of autophagy. Increased levels of the mTOR kinase is found to inhibit the autophagy process resulting in an increased in cell growth and tumor development. The first potent inhibitor of mTOR, rapamycin, is known to promote autophagy and inhibits the growth of malignant glioma cells (Takeuchi et al 2005). Autophagy, an intra-lysosomal degradation pathway that controls cell growth by allowing the turnover of damaged long-lived proteins and organelles, is an important process for normal development and are implicated in various pathophysiological conditions, including breast cancers (Cuervo 2004; Levine 2007; Levine & Kroemer 2008; Liang et al 1999).

Deregulation of the mTOR signaling pathway and aberrant autophagy is associated with oncogenesis (Bialik & Kimchi 2008; Chiang & Abraham 2007; Cully et al 2006). Targeting the mTOR pathway using mTOR inhibitors in cancer have become widely appreciated only recently. It is suggested that by pharmacologically inhibiting mTOR, cells will be conditioned to halt cell cycle progression and results in G1 arrest and subsequently autophagy (Gera et al 2004). Patients with tumors expressing high level of mTOR kinase expression are expected to benefit from treatment from the mTOR kinase inhibitors. Rapamycin is an example of a prototype of mTOR inhibitor and many mTOR inhibitors are currently in development. The mTOR inhibitor, CCI-779 (temsirolimus) ^{is a recently Food and Drug Administration (FDA), USA–approved anticancer drug with efficacy in certain solid tumors and hematologic malignancies (Shor et al 2008). In addition, another mTOR inhibitor, RAD001 (everolimus) is currently in Phase II/III clinical trials for various cancers, including breast cancers (Campone et al 2009; Seeliger et al 2007).}

The aims of this study were to determine the mTOR protein expression in breast cancer and normal breast tissues by standard immunohistochemistry method. In addition, this study attempts to determine the relationship between the expression of this protein with type of tissues and clinicopathological characteristics of the breast tumors. Finally, by determining the proportion of breast cancer which expresses mTOR protein, it may provide us some clue on the proportion of breast cancer patients whom may be benefit from mTOR inhibitors.

Materials And Methods

Tissue microarrays and antibodies

Immunohistochemistry detection of mTOR protein in tissues

Assessments of mTOR immunostaining and statistical analysis

Results

In the present study, mTOR protein expression was successfully evaluated using immunohistochemistry technique in 78 (98%) of 80 cases of breast cancer tissues (missing data for 2 cases) and in 53 (76%) of 70 cases of normal breast tissues (missing data for 17 cases, no normal breast cells were found in the tissue cores). The mTOR protein was observed in the cytoplasm of tissues that were stained positive. The percentage of cells stained positive was found to be more than 90% in all tissues. The tissue microarray core details, clinicopathological data and results of the mTOR protein expression in breast cancer and normal breast tissues are shown in Tables 1 and 2.

Figure 1

Table 1 Clinicopathological variables and mTOR protein expression in breast cancer tissues (BR701 and BRN801)

+, weak stained; ++, moderate stained; +++, strong stained cytoplasm; -, negative stain; ND, not supplied by the manufacturer;

IDC,invasive ductal carcinoma; DCIS, ductal carcinoma in situ;

1-68 (BR701); 69-78 (BRN801); 2 cores in BR701 missing;

The clinical data presented in the table were provided by Biomax (http:// www.biomax.us/)

Figure 2

Table 2. mTOR protein expression in normal breast tissues (BRN801)

+, weak stained; ++, moderate stained; +++, strong stained cytoplasm; -, negative stain

The age distribution of the breast cancer cases ranged from 24 to 81 years (mean age, 46.9 years). Majority of the cases (63%) were categorized in the 30-50 years age group. However, not all cases were supplemented with complete clinicopathological data (Table 1). Only 67 of 78 cases were provided data on their disease stages (TNM). There were 10 cases (15.0%) with stage 1 cancer, 40 cases (60.0%) with stage II, 10 cases (15.0%) with stage III and 7 cases (10.0%) with stage IV cancer (Table 3). Majority (77 cores) were cases of invasive ductal carcinoma (98.7%), and 1 case (1.3%) of ductal carcinoma in situ (DCIS) (Table 1). Data on estrogen (ER), progesterone (PR) and C-erb2 receptor status were only provided for 64 cases (42 cases were positive, 22 were negative), 64 cases (39 were positive, 25 were negative) and 60 cases (34 were positive, 26 were negative) respectively (Table 3).

Figure 3

Table 3. Summary of the results obtained following analysis of breast cancer and normal breast tissues for the expression of mTOR protein

The clinical data presented in the table were provided by Biomax (http:// www.biomax.us/. However some cases were not supplemented with complete TNM classification and receptor status

The immunohistochemistry results showed that mTOR protein expression in breast cancer tissues were present (positive staining) in majority of cases (75 of 78 cores) (96.2%) (Tables 1, 3). However, mTOR protein was not detected in 3 cases (3.8%). Of the positive cases, 17 cases (21.8%) showed weak staining (+), 34 cases (43.6%) showed moderate staining (++), 24 cases (30.8%) showed strong staining intensity (+++) (Table 3). As for the normal breast tissues, mTOR expression was observed in only 19 of 53 cases (35.8%) and was absent in 34 cases (64.2%) (Tables 2, 3). Of the positively stained normal tissues, 12 cases (22.6%) showed weak expression, 7 cases (13.2%) showed moderate expression and none (0.0%) showed strong expression. Chi-Square Test showed that there was a significant association between mTOR expression and the type of tissues (p < 0.000) (Table 3). All negative controls produced no staining while all positive controls (breast cancer slides) produced staining for mTOR protein. Replicate arrays produced similar results for all tissue cores. Representative images of mTOR protein expression in breast cancer and normal breast tissues are shown in Figure 1 and 2.

Figure 4

Figure 1. Representative immunohistochemical staining of negative and positive mTOR

protein expression in breast cancer tissues (BR701). Immunohistochemical staining showing an absence of mTOR protein expression in breast cancer tissues (A); weak cytoplasmic expression (B) moderate cytoplasmic expression (C) and strong cytoplasmic expression (D) (original magnification x 200).

Figure 5

Figure 2 Representative immunohistochemical staining of negative and positive mTOR protein expression in normal breast tissues (BRN801). Immunohistochemical staining showing an absence of mTOR protein expression in normal breast tissues (A); weak cytoplasmic expression (B) and moderate cytoplasmic expression (C). None of the normal tissues showed strong mTOR expression (original magnification x 200).

Majority of the breast cancer cases were seen in patients between 30-50 years of age (49 cases). 25 cases were seen in patients above 50 years old whereas only 4 cases were below the age of 30 years (Table 3). A high number of cases in each age group exhibited moderate to strong mTOR expression (4 cases in < 30 years; 34 cases in 30-50 years and 20 cases in > 50 years age group). However, there was no statistically significant association between age group and mTOR expression in breast cancer tissues (p=0.723, p>0.05).

A high proportion of stage I (10 cases of 10 cases), stage II (30 cases of 40 cases), stage III cancers (9 cases of 10 cases) produced moderate to strong expression of mTOR protein. However, stage IV cancers produced dissimilar observations and Chi-Square Test showed no significant relationship between disease stages and mTOR expression in breast cancer tissues (p=0.316, p>0.05)

42 cases of the breast cancers were positive for estrogen receptor (ER) while 22 cases were negative. Of the positive and negative ER cases, majority were positive for mTOR expression. 72.7% of the negative and 76.2% of the positive ER cases showed moderate to strong mTOR expression. There was no significant association between ER status and mTOR expression (p=0.983, p>0.05). 39 cases of the breast cancers were positive for progesterone receptor (PR) while 25 cases were negative. Majority of both negative and positive cases showed mTOR expression, with 84.0% of the negative and 69.2% of the positive cases showed moderate to strong immunoreactivity towards mTOR protein, respectively. However, there was no statistically significant association between this variable and mTOR expression. 34 cases of the breast cancers were positive for C-erb2 receptor while 26 cases were negative for this receptor. Majority of both negative and positive cases showed mTOR expression, with 84.6% of the negative and 67.6% of the positive cores showed moderate to strong immunoreactivity towards mTOR protein, respectively. Similarly, there was no significant association between C-erb2 status and mTOR expression.

Discussion

The pathogenesis of breast cancer is multi-factorial. The variability in gene and protein expression patterns and signaling pathways activated in tumors supported the need for tumor markers identification. Studying molecular changes in tumors helps generate better understanding of tumor progression and foster development of targeted and individualized pharmacotherapy against cancer. mTOR is a protein kinase that is centrally involved in the control of cancer cell metabolism, growth, proliferation and autophagy. Deregulated signaling of mTOR and aberrant autophagy process has been linked to tumor progression (Bialik & Kimchi 2008; Chiang & Abraham 2007). The mTOR pathway has attracted broad scientific and clinical interest, particularly in light of the ongoing clinical cancer trials with mTOR inhibitors.

Previous studies have implicated the role of phosphorylated mTOR (p-mTOR) in breast cancer pathogenesis. Zhou and co-workers have examined 165 breast cancers with specific ^{antibody for p- mTOR using immunohistochemistry, cell culture and western blot techniques (Zhou et al 2004). They found that the expression of p-mTOR protein is increased from normal breast epithelium to hyperplasia and abnormal hyperplasia to tumor invasion. Higher levels of p-mTOR were associated with poor disease-free survival. The overall results supported the trend that phosphorylation of mTOR increases with the progression of tumor. In another study, Lin and co-workers have also examined mTOR phosphorylation status in tissue microarray slides containing 89 invasive breast cancer tissues and 6 normal mammary tissues by using immunohistochemistry staining. Elevated phosphorylation of mTOR was highly associated with invasive breast tumors (Lin et al 2005).}

Bose and co-workers have studied the expression levels of PTEN and phosphorylated forms of the constituent proteins such as Akt and mTOR. These proteins were evaluated by immunohistochemistry, on consecutive sections from a tissue microarray containing 145 invasive breast cancers and 140 pure ductal carcinomas in-situ. The Akt pathway was found to be activated early in breast cancer, in the in-situ stage and cancers with mTOR overexpression have showed a three times greater risk for disease recurrence (Bose et al 2006). In another recent study, which examined the relationship between Akt (upstream of mTOR) and 4E-BP1 (downstream of mTOR) in primary breast tumors and their distant metastasis, it was revealed that most primary breast tumors and metastatic tumors expressed p-Akt (76%). Similarly, most of the primary and metastatic tumors were also positive for p-4EBP1 (75% and 74% respectively) (Akcakanat et al 2008).

In general, most of the studies looked at the phosphorylated forms of mTOR and have indicated that p-mTOR expression is linked to tumor progression and cancer pathogenesis. In this study, the non-phosphorylated form of mTOR was analyzed and a greater number of normal breast tissues were used in addition to the cancer tissues. The results of the present study showed that 75 cases of the 78 breast cancer tissues expressed the mTOR protein as compared to the 19 cases of 53 normal breast tissues. By contrast, a majority of the normal breast tissues (34 of 53 cases) produced negative immunoreactivity towards mTOR protein. Chi-Square Test showed significant association between mTOR expression and type of tissues. A significant majority of breast cancer tissues were found to express mTOR protein as compared to the normal tissues. A significantly higher number of breast cancer tissues were also found to express the mTOR protein in various grades of intensity as compared to the positively stained normal tissues, which further suggests that the over-expression of mTOR protein could have played a significant role in the pathogenesis of breast cancers. These results are consistent with previous studies which implicated the role of phosphorylated-mTOR in cancer pathogenesis. This study has also suggested that 96% of the breast cancers cases which were positive for mTOR protein could be potential candidates for mTOR inhibitors treatment.

The relationship between clinicopathological variables of the patients (based on tissue cores) and mTOR protein were analyzed in the present study. However, there were no significant relationship between the variables and the mTOR expression. Age groups, disease stages and receptor status of the subjects were not significantly associated with mTOR protein expression. Currently, there was no recent publication on the direct relationship between mTOR expression and these variables. However, studies have linked the various receptor statuses of the breast tumors with some of the molecules along the PI3K-Akt-mTOR signaling pathways.

In a recent analysis, a gene signature specific to ER/PR tumors were identified among breast tumors using 5 published studies of human breast cancers with clinically assigned hormone receptor status. ER+/PR- breast cancers defined by RNA profiling were associated with poor patient outcome, worse than those with pure ER+/PR+ patterns. Targets of transcriptional up-regulation by specific oncogenic pathways, including PI3 K/Akt/mTOR, were enriched in both ER+/PR- and ER-/PR- compared to ER+/PR+ tumors. It was also observed that ER+/PR- tumors as defined by RNA profiling represent a distinct subset of breast cancer with aggressive features and poor outcome, despite being clinically ER+ (Creighton et al 2008).

Previous studies have also associated the activation of the Akt-mTOR pathway in breast tumors with poor patient outcome. In one study, a set of genes induced by Akt in a transgenic mouse model, a subset of which were sensitive to mammalian target of rapamycin (mTOR) inhibitor RAD001, was examined in five public gene expression profile data sets of clinical breast tumor specimens (representing >1000 samples). In this study, the gene signatures of Akt-mTOR-dependent (RAD001-sensitive) and Akt-mTOR-independent (RAD001-insensitive) pathways were each examined for correlation to patient outcome. It was observed that tumors with high average expression of the Akt-induced, RAD001-sensitive genes tended to have ER-negative status. However, no consistent correlation patterns were observed between the Akt-associated genes and PR expression. In three of the data sets, the Akt and RAD001-sensitive genes were positively correlated with HER2 (c-erb) expression. Akt and RAD001-sensitive genes were also positively correlated with increasing grade and highly correlated with increasing size of the tumors (Creighton 2007). However, there were no significant observations found between mTOR status and the variables examined in this study.

In conclusion, mTOR protein was found to be expressed in a significantly larger number of breast cancer tumors as compared to the normal breast tissues. A significantly higher number of breast cancer tissues were also found to express the mTOR protein in various grades of intensity as compared to the positively stained normal tissues. The results in this study further strengthen the current hypothesis that over-expression of mTOR protein may play a role in the development of breast cancers.

Acknowledgements