Overexpression of leptin receptor predicts an unfavorable outcome in Middle Eastern ovarian cancer

Background Recent epidemiological studies have suggested that obesity is associated with ovarian cancer. Obesity hormone leptin and its receptor (Ob-R) contribute to tumor development by enhancing cell growth and survival. This study was design to investigate the prevalence of leptin and Ob-R in Middle Eastern epithelial ovarian cancer (EOC) and to analyze the role of leptin and the mechanisms under its action in EOC tissue sample and cell lines. Methods The expression of leptin and Ob-R was examined by immunohistochemistry in a tissue microarray of 156 EOC samples. Proliferation of EOC cells in response to leptin was assessed by MTT assays, and its anti-apoptotic effects were determined by flow cytometry. Effect of leptin on PI3K/AKT signaling pathway was further determined by western blotting. Results In clinical samples, Ob-R overexpression was seen in 59.2% EOCs and was significantly associated with poor progression free survival (p = 0.0032). Furthermore, Ob-R expression was associated with anti apoptotic proteins Bcl-XL (p = 0.0035) and XIAP (p = 0.0001). In vitro analysis using EOC cell lines showed that leptin stimulated cell proliferation and inhibits apoptosis via activation of PI3K/AKT signaling pathway. Inhibition of PI3K activity by LY294002, a specific inhibitor of PI3-kinase abrogated leptin mediated PI3K/AKT signaling. Gene silencing of Ob-R with Ob-R siRNA in EOC cells resulted in down regulation of phospho-AKT and its down stream targets. Conclusion Our findings have potential clinical implication for EOC development and progression.


Background
Despite rapid advances in understanding ovarian cancer etiology, epithelial ovarian cancer (EOC) remains the most lethal form of gynecologic cancers [1][2][3][4]. Malignant transformation of normal ovarian epithelial cells is caused by genetic alteration that disrupts proliferation, programmed cell death and senescence.
Leptin, the product of obesity gene (Ob) is suggested to be associated with cancer development and progression in many epithelial cancers including EOC [5][6][7][8] Leptin is 16KD adipokine produced predominantly by adipocytes with wide range of biological activities including appetite regulation, bone formation, reproductive function and angiogenesis [9][10][11]. Leptin mediated signaling pathways play an important role in cancer cell proliferation, invasion and metastasis [5]. Leptin exerts its activity through specific membrane receptor, the obesity receptor (Ob-R), which is assigned to class I cytokine receptor family [12]. Six splice variants of Ob-R have been identified up to now; a long isoform, four short isoforms discriminated by the different lengths of intracellular domain, and the secreted isoform, which modulates blood leptin [12,13]. According to the current knowledge, leptin signaling pathway is mainly transduced by JAK/ STAT, MAPK and PI3K signaling pathways [5]. Previous study [8] has suggested that leptin signaling pathway is transmitted via MAP kinase pathway. However, the interaction between leptin signaling and PI3K/AKT pathway in ovarian cancer remains unknown.
A recent epidemiological study [14] has found that among women who have never used menopausal hormone therapy, obese women are at increased risk of developing ovarian cancer compared with women of normal weight. Although a hormonal mechanism was suggested as a link between ovarian cancer and obesity, at present, a clear biological explanation for risk associated between obesity and EOC is not fully known. Therefore, the effects of obesity on ovarian cancer represent a critical intersection between these two important health problems. However, whether there is a direct relationship between leptin and ovarian cancer cannot be conclusively stated as increased leptin and ovarian cancer may both be secondary consequences of obesity. Considering the fundamental role of leptin and Ob-R in cancer development and progression, we sought to examine the expression pattern of leptin and Ob-R in large cohort of Middle Eastern EOC using TMA immunohistochemical analysis. We then examined the expression of leptin and Ob-R using EOC cell lines. Furthermore, we investigated the effect of leptin on malignant properties of EOC including proliferation and apoptosis. Finally we elucidated the PI3K/AKT signal transduction pathway regulating leptin-induced changes in the cancerous properties of EOC.

Immunohistochemical detection of Ob-R expression and its association with clinicopathological parameters
Ob-R expression was seen in 59.2% (90/152) of the EOCs analyzed ( Figure 1). No association was observed between Ob-R overexpression and age, FIGO Stage, Histology type and grade (Table 1). Ob-R expression was linked to PI3K/ AKT signaling pathway as evidenced by direct association of Ob-R expression with pGSK3 (p = 0.0009), PTEN (p = 0.0002) and end stream anti-apoptotic markers XIAP (p = 0.0001) and Bcl-XL (p = 0.0035) expression. However no association was seen with p-AKT (p = 0.2082).

Ob-R expression and progression overall survival
EOC patients with low expression of Ob-R had a poor progression free survival (PFS) of 13.1 months as compared to 21 months (p = 0.0032) with low Ob-R expression ( Figure 2). In the multivariate analysis using Cox Proportional Hazard model for multiple factors like age, FIGO stage, grade and Ob-R expression, the relative risk was 1.96 for high Ob-R expression (95% CI 1.28-3.06; p = 0.0020) and 1.81 for AJCC stage IV (95% CI 1.08-2.93; p = 0.0243). Thus Ob-R overexpression was an independent prognostic marker for poor survival in multivariate analysis (Table 2).

Immunohistochemical detection of leptin expression and its association with clinicopathological parameters
Leptin expression was noted in 89.5% (128/143) of the EOCs analyzed and leptin staining was seen in the nuclear as well as cytoplasmic compartment. As shown in Table 3 leptin expression was linked to PI3K/AKT signaling pathway as evidenced by direct significant association of leptin expression with p-AKT (p = 0.0344). However no association was seen with expression of PTEN (p = 0.3096) and end stream anti-apoptotic markers XIAP (p = 0.3500) and Bcl-XL (p = 0.1724). Also leptin overexpression was not associated with patient's age, histology type, tumor grade, FIGO stage and progression free survival (Table 3).

Leptin increase proliferation of EOC cells
The effects of leptin on growth rate of EOC cell lines were determined using MTT assay. MDAH2774 and SKOV3 cells were initially serum starved for 24 hours and then stimulated with various doses of recombinant leptin (10-200 ng/ml) for 48 hours compared to with cell serum free control cultures. As shown in Figure 3A, leptin induced significant cell growth of both EOC cell lines in a dose dependent manner.

Leptin prevent serum-starved apoptosis in EOC cells
EOC cell lines were seeded in six well plates and after 24 hours serum free medium alone or 100 ng/ml leptin was added. After 48 hours, apoptosis was measured by annexin/PI staining. As shown in Figure 3B, serum starvation resulted in apoptosis. Treatment of EOC cell line with leptin significantly prevented serum starved apoptosis suggesting that leptin counteract apoptosis in EOC cells (p = 0.0005).
Leptin activates the PI3-Kinase/AKTsignaling pathway PI3-kinase/AKT pathways have been implicated in playing crucial roles in regulating cell growth, cell proliferation prevention of apoptosis, which altogether attribute tumorigenesis [15]. In view of these findings we sought to determine whether PI3-kinase signaling is activated during leptin stimulated EOC cell line proliferation.
MDAH2774 cells were stimulated with 100 ng/ml leptin for various time periods. Cells were lysed and proteins were separated on SDS-PAGE and immunoblotted with p-AKT (activated AKT) and p-FOXO1 antibodies. As shown in Figure 4A, leptin treatment of MDAH2774 phosphorylated AKT and FOXO1 as early as 15 minutes and remained phosphorylated till 3 hours. Similar results were Tissue microarray based immunohistochemical analysis of Ob-R, pGSK3, Bcl-XL and XIAP in EOC patients obtained with other EOC cell lines (data not shown). These results suggest that leptin mediated cell proliferation occurs via PI3K/AKT signaling pathway.

Inhibition of PI3-kinase prevents leptin mediated AKT activation and its downstream effector FOXO1
Since our study suggesting that leptin stimulated PI3kinase signaling plays a role in EOC proliferation and promotes its anti-apoptotic effects. We sought to determine whether the inhibition of PI3'-kinase by its specific inhibitor, LY294002, abrogated leptin mediated PI3K/AKT signaling in EOC cell lines. Cells were seeded on culture plates for 24 hours. Starved EOC cell were pre-treated with 20 μM LY294002 for 2 hours and subsequently treated with and without 100 ng/ml leptin for 3 hours. Cells were lysed and proteins were separated on SDS-PAGE and immunoblotted by antibodies against p-AKT and p-FOXO1. As shown in Figure 4B, leptin phosphorylated AKT and FOXO1 in MDAH2774 cell line and pre-treatment with LY294002, prevented AKT and FOXO1 phosphorylation. In addition, pre-treatment of EOC cells with LY294002, abrogated leptin-induced cell proliferation ( Figure 4C) as well as prevented leptin-mediated antiapoptotic effects ( Figure 4D) on EOC cells suggesting that PI3-kinase/AKT pathway plays a critical role in leptininduced growth and proliferation of EOC cells. These data is also suggesting that leptin is acting upstream of PI3kinase/AKT pathway in modulating its anti-apoptotic response in EOC cells.

EOC cell lines express leptin receptors that mediate the PI3-kinase/AKT signaling pathways
To investigate whether leptin receptors are functional and linked to coordinate with PI3-kinase/AKT signaling pathway to regulate cell growth and proliferation of EOC cell lines, we utilized small interfering RNA (siRNA) strategies to transfect Ob-R specific siRNA as well as scrambled non specific siRNA in MDAH2774 cell line. After 48 hours transfection, cell were starved and then treated with and without 100 ng/ml leptin for 3 hours. As shown in Figure  5, MDAH2774 expressed functional leptin receptors, as shown previously [4] Treatment of scrambled siRNA harboring MDAH2774 cells with leptin showed activation of AKT, FOXO1 and elevated level of XIAP and Bcl-XL proteins that are involved in PI3-kinase/AKT pathway and play a critical role in cell survival. On the other hand expression of Ob-R specific siRNA in MDAH2774 cells knocked down the expression of Ob-R and prevented leptin-induced activation of AKT, FOXO1 as well as abrogated the expression of XIAP and Bcl-XL proteins. These data suggests that leptin utilizes PI3-kinase/AKT signal transduction pathway in mediation of EOC cell proliferation, further confirming that leptin is acting upstream of PI3K/AKT pathway.

Discussion
Recent reports suggest that leptin is overexpressed in various cancer cells and plays a role in the development and/  or progression of variety of variety of malignancies including colon, gastric, endometrial and breast cancers [16][17][18][19][20]. These findings were further supported by experimental evidence that leptin can stimulate growth and prevent apoptosis in different cellular cancer models [21][22][23]. A very recent epidemiological study has shown that obesity is a risk factor for ovarian cancer in postmenopausal women [14]. High levels of serum estrogen, derived from increased adipose tissue are considered to contribute to the pathogenesis of ovarian cancer in those obese patients. However, circulating leptin is an essential factor regulating fat metabolism, it can be hypothesized that leptin itself might be involved in the development of ovarian cancer. Therefore, we first examined the prevalence of leptin and its receptor Ob-R expression in 156 Saudi EOC samples. Our data shows that Ob-R protein was detected in 59.2% examined EOCs. Interestingly, patients with high Ob-R expression tumors showed a significant poor disease free survival (p = 0.0032) compared with reduced Ob-R expression which support the proposed role of leptin signaling in the EOC. Furthermore, Ob-R overexpression was an independent prognostic variable to predict poor pro-gression free survival. Interaction between leptin/Ob-R and other signaling pathways such as PI3K/AKT and MAP kinase have been reported in oncogenesis of various tumors [5,8,22,[24][25][26]. Therefore, we examined the relationship between the expression of Ob-R and the PI3K/ AKT pathway protein targets in EOC. Activated AKT protein expression was seen in 52% of our EOC examined. However, no correlation was observed with Ob-R expression and AKT activation by IHC staining which might sug-

Figure 3 Leptin increase proliferation of EOC cells and prevents serum starved apoptosis in EOC cells. (A) SKOV3 and
MDAH2774 cell lines were seeded in 96 well plates and after 24 hours serum free medium alone or with various doses of leptin as indicated was added. After 48 hours cell proliferation was measured MTT as described in materials and methods. The graph displays the mean +/-SD (standard deviation) of six independent experiments. (B) EOC cell lines were seeded in six well plates and after 24 hours serum free medium alone or 100 ng/ml was added. After 48 hours apoptosis was measured by annexin/PI staining. serum leptin levels are not associated with body mass index Furthermore, in vitro analysis using EOC cell lines was conducted to study the affect of leptin on EOC cell growth. Our experiments clearly showed that leptin has oncogenic affect on EOC cells and this oncogenic effect is due to a combination of cell proliferation and inhibition of apoptosis by leptin. To elucidate the signaling pathways involved in leptin mediated induction of cancerous properties of EOC cells, we examined the effect of leptin in activation of the PI3K/AKT pathway. Our data showed that leptin rapidly stimulates the PI3K/AKT pathway and induced phosphorylation of AKT thus activating this key signal transduction pathway associated with cell growth.
In addition, prevention of leptin-induced activation of PI3K/AKT with pharmacological inhibition in turn significantly reduced the activation of AKT. In addition our data showed that gene silencing of Ob-R in EOC cell lines abrogated the AKT activation as well as the expression of anti-apoptotic genes, XIAP and Bcl-XL suggesting that leptin mediated EOC growth is due to modulation of growth and anti-apoptotic genes.

Conclusion
Our data suggests that leptin pathway might play a major role in Middle Eastern EOC, and deciphered the molecular mechanisms responsible for leptin mediated EOC cell proliferation, establishing direct association between obesity and EOC carcinogenesis and presenting involvement of key molecules of oncogenic PI3K signaling pathway. Since the majority of patients are lost to follow-up as their disease reaches its terminal stages, it was impossible to determine overall survival in this specific patient population.

Tissue Microarray (TMA) Construction
Tissue microarrays were constructed from formalin-fixed, paraffin-embedded ovarian cancer specimens as described previously [28]. Tumor regions were mapped by a pathologist for coring. The tissue microarray was constructed with 0.6-mm diameter cores spaced 0.8 mm apart using a modified Tissue Microarrayer (Beecher Instruments, Sun Prairie, WI, USA). The tissue microarray block was cut into 5 mm sections, adhered to the slide by an adhesive tape-Leptin required functional leptin receptor (Ob-R) for activa-tion of PI3-kinase/AKT signaling in EOC cell lines transfer method (Instrumedics Inc., Hackensack, NJ, USA) and UV cross linked.

Antibodies and immunohistochemistry
Immunohistochemical studies on formalin-fixed, paraffin-embedded tissue sections were performed as describer in earlier studies [28,29]. Primary antibodies used, their dilutions, and other information is listed in Table 4. For antigen retrieval, Dako Target Retrieval Solution pH 6.0 (Catalogue number S1700) was used, and the slides were microwaved at 750W for 5 minutes and then at 250W for 30 minutes. The sections were incubated overnight with Ob-R) and the Dako Envision Plus System kit was used as the secondary detection system with DAB as chromogen.
We used a mouse monoclonal antibody from Santa Cruz Biotechnology; Clone B3 to detect Ob-R expression and this antibody binds to both short and long forms of Ob-R. Similarly, leptin expression was detected by using a rabbit monoclonal antibody Y20 from from Santa Cruz

Immunohistochemistry Assessment
In this study, Ob-R expression was categorized by doing an H score, which combines intensity of staining in each cell and percentage of stained cells. In brief, each TMA spot was assigned a staining intensity score from 0-3 [I0, I1-3], and a percent of stained tumor cells that was recorded in 5% increments from a range of 0-100 (P0, P1-3). For each spot analyzed, a score was generated from the product of intensity and percent of tumor cells stained. A final H score (range 0-300) was obtained by adding the sum of individual scores obtained for each tissue microarray spot. (H score = I1XP1+I2XP2+I3XP3). Ovarian tumors were categorized into 2 groups based on H score and using the X-tile plat as described below.
X-tile plots are constructed for assessment of biomarker and optimization of cut off points based on outcome [30,31]. The X-Tile plots allow determination of an optimal cut point while correcting for the use of minimum P statistics. Using the X-Tile program, an optimal cut point for Ob-R expression was determined at 20, with a Miller-Seigmund p value of 0.5950 as determined by X-Tile. Tumors with H score <20 were classified as low expressers (n = 62; 40.8%), and those with H score >20 were classified as high expressers (n = 90; 59.2%; Figure 1a and 1e). Similarly X-Tile was used to determine a cut point for leptin, Bcl-XL, XIAP, PTEN and pGSK3.
p-AKT & PTEN scoring was done as described earlier [32,33]. Briefly, p-AKT was scored as levels on an intensity scale ranging from 0 to 3. Scoring was performed as follows: 0, no appreciable staining in tumor cells; 1, barely detectable staining in tumor cells; 2, appreciable staining of moderate intensity, distinctly marking tumor cells and