A Novel Unidirectional Cross-Talk from the Insulin-Like Growth Factor-I Receptor to Leptin Receptor in Human Breast Cancer Cells

Obesity is an important and manageable risk factor for the development and progression of postmenopausal breast cancer (1). Increased body weight and body mass index are associated with reduced disease-free and overall survival and poorer therapeutic response rates in breast cancer patients, regardless of menopausal status or age (2). Although the exact molecular mechanisms by which obesity influences cancer biology are unknown, there is evidence suggesting that increased production and secretion of adipocyte-derived growth factors and hormones contributes to cellular transformation and tumorigenesis (3, 4). The obesity-associated hormones leptin and insulin-like growth factor-I (IGF-I) have been independently implicated in the connection between obesity and breast cancer (5).

Leptin, a product of the obese (ob) gene, is an adipocytokine that regulates appetite, bone formation, reproduction, cellular proliferation, and angiogenesis (6). Because of the strong association between human obesity and elevated levels of circulating leptin, this hormone has been widely studied in the fields of nutrition and weight management (7). More recently, however, leptin has emerged as a potential factor contributing to mammary tumorigenesis. In vitro studies showed that leptin stimulates the growth, survival, and transformation of breast cancer cells (5), primarily by activating the Janus-activated kinase (JAK)/signal transducers and activators of transcription (STAT) signaling pathway (8, 9) and the phosphoinositol-3-kinase/Akt and mitogen-activated protein kinase (MAPK) pathways (10). Leptin induces cell cycle progression by up-regulating cyclin D1 expression and cyclin-dependent kinase 2 activity, as well as by inactivating the retinoblastoma growth suppressing protein (11). Importantly, leptin and its receptor (Ob-R) were found to be overexpressed in a majority of breast cancer tissues, especially in high-grade tumors, but absent or expressed at very low levels in normal mammary epithelium or benign tumors (5, 12). In addition, leptin-deficient mice have a decreased incidence of spontaneous and oncogene-induced mammary tumors (13). Thus, leptin signaling seems to play an important role in breast cancer biology.

Similar to leptin, increased levels of IGF-I and its receptor are detected in sera and primary tumors of breast cancer patients (14, 15), and transgenic overexpression of IGF-I receptor (IGF-IR) has been shown to induce mammary tumor formation (16). IGF-I is an important endocrine, paracrine, and autocrine regulator of breast epithelial cell growth. Increased signaling through the IGF-IR results in increased cellular proliferation, mitogenesis, and survival and decreased apoptosis, causing resistance to numerous antineoplastic agents (14, 17). For these reasons, the IGF-IR has become an important therapeutic target for drug discovery in breast oncology (17).

Cross-talk between different growth factor receptor families is frequently observed in tumors. This mechanism allows cancer cells to enhance downstream signaling resulting in greatly increased proliferation, mitogenesis, and cell survival. The IGF-IR has been shown to interact and cross-talk with multiple receptors, including the epidermal growth factor receptor (EGFR; ref. 18), HER2 (19), platelet-derived growth factor receptor (20), and the estrogen receptor (14). Due to the correlations between elevated levels of leptin, IGF-I, and their associated receptors with obesity and breast cancer, we hypothesized that interactions and/or cross-talk may occur between these two signaling pathways.

The human breast cancer lines MDA-MB-231 (MDA231), MCF7, BT474, and SKBR3 were examined for expression of the IGF-IR and leptin receptor (Ob-R). Immunoblotting of total protein lysates (Fig. 1A) showed that the two major isoforms of Ob-R, called Ob-Rb (longer isoform) and Ob-Rt (shorter isoform), are expressed at similar levels in all cell lines (Fig. 1B). IGF-IR is expressed at higher levels in MCF7 and BT474 cells versus SKBR3 and MDA231 cells, with highest levels observed in MCF7 cells (Fig. 1B).

Immunoprecipitation of Ob-R with subsequent immunoblotting for IGF-IR showed that Ob-Rb and Ob-Rt are both pulled down with IGF-IR in all four cell lines (Fig. 2A). Conversely, IGF-IR immunoprecipitation pulled down Ob-Rb and Ob-Rt in each cell line, with preferential interaction observed with the shorter isoform of Ob-R in MCF7, BT474, and SKBR3 cells (Fig. 2B). Quantitation showed that IGF-IR was pulled down with Ob-R to a similar extent in all four lines (Fig. 2C). Total Ob-R was pulled down with IGF-IR in all four lines; however, higher levels of Ob-R interacting with IGF-IR was observed in MCF7 cells (Fig. 2C), likely due to the higher expression level of total IGF-IR in these cells (Fig. 1B). Negative controls in which cell lysates were immunoprecipitated with rabbit IgG confirmed that IGF-IR and Ob-R were not pulled down (Fig. 2D). In addition, because IGF-IR has been shown to interact with insulin receptor (21), we blotted IGF-IR immunoprecipitates for insulin receptor as a positive control (Fig. 2D). Insulin receptor was pulled down with IGF-IR in all four lines. Finally, another tyrosine kinase receptor, EGFR, was immunoprecipitated and blotted for Ob-R in all lines (Fig. 2D). Collectively, the results of these immunoprecipitation experiments indicate that the IGF-IR and leptin receptor interact in human breast cancer cells.

To determine the effect of IGF-IR/leptin receptor interaction on receptor signaling, MCF7 cells were serum-starved overnight and then stimulated with IGF-I (100 ng/mL) for up to 1 hour. IGF-IR phosphorylation was induced within 5 minutes (Fig. 3A), while total IGF-IR levels were unaltered. Importantly, phosphorylation of Ob-R was also induced within 5 minutes of IGF-I exposure, suggesting potential cross-signaling from IGF-IR to leptin receptor. Similarly, in BT474 cells (Fig. 3B) and MDA231 cells (Fig. 3C), IGF-I stimulation induced phosphorylation of both IGF-IR and Ob-R within 5 minutes, without affecting total levels of either receptor. To determine if IGF-I stimulates phosphorylation of the leptin receptor via the IGF-IR kinase, MCF7 cells were treated with the IGF-IR kinase inhibitor I-OMe-AG538 and stimulated with IGF-I (Fig. 3D). Immunoblotting showed that inhibition of IGF-IR kinase blocked IGF-I–stimulated phosphorylation of leptin receptor. Thus, IGF-I cross-signals to the leptin receptor via the IGF-IR kinase.

Having established that IGF-IR stimulates phosphorylation of the leptin receptor, we examined IGF-I–mediated effects on downstream receptor signaling. MCF7 cells were stimulated with IGF-I and immunoblotted for phosphorylated and total JAK2 and STAT3 (Fig. 4A) and for phosphorylated and total Akt, extracellular signal-regulated kinase 1/2 (ERK1/2), and p38 MAPK (Fig. 4B). Significant phosphorylation of JAK2 and STAT3 was observed in response to IGF-I within 5 minutes. IGF-I also activated the phosphoinositol-3-kinase pathway, as shown by phosphorylation of Akt. Phosphorylation of ERK1/2 and p38 MAPK was rapidly activated by transient versus other signaling pathways. Collectively, these results support the concept that IGF-I cross-activates the leptin receptor signaling pathway, although the signaling molecules examined are downstream of multiple growth factor receptors and, thus, do not strictly confirm activation of leptin receptor signaling. However, as leptin receptor phosphorylation was induced by IGF-I and blocked by IGF-IR kinase inhibitor on Tyr¹¹⁴¹, which is the phosphorylation site that binds STAT3 and activates downstream signaling, our results strongly suggest that IGF-IR induces activation of the leptin receptor.

We next examined whether cross-talk occurs in the opposite direction, i.e., from the leptin receptor to IGF-IR. MCF7 cells were serum starved and stimulated with leptin (1,000 ng/mL) for up to 6 hours. Leptin induced phosphorylation of leptin receptor within 5 minutes (Fig. 5A). However, phosphorylation of IGF-IR at either Tyr¹¹³¹ or Tyr^1135/1136 was not stimulated by leptin at these time points of up to 6 hours nor was it stimulated at shorter time point increments or longer time points of up to 24 hours or with lower doses of leptin (not shown). As a positive control, IGF-I stimulated phosphorylation of IGF-IR as expected and also induced phosphorylation of leptin receptor as previously observed (Fig. 3). Similarly, BT474 cells stimulated with leptin showed phosphorylation of leptin receptor but not of IGF-IR at either of the three sites examined (Tyr¹¹³¹, Tyr¹¹³⁵, and Tyr¹¹³⁶; Fig. 5B). Thus, our results suggest a unidirectional cross-talk from the IGF-IR to the leptin receptor in breast cancer cells.

Epidemiologic studies estimate that obesity increases the risk of breast cancer by up to 50% (3). The molecular mechanisms guiding obesity-associated breast cancer are not well understood, but are likely to involve an increased production and secretion of obesity-associated hormones (22). IGF-I and leptin are capable of regulating mammary tissue growth at multiple levels (5). Both hormones are secreted by abdominal adipocytes, resulting in endocrine effects on various tissues, including the breast. Paracrine growth stimulatory effects occur via IGF-I and leptin released by the adipocyte component of stroma surrounding breast epithelial cells or existing breast tumor cells. In addition, an autocrine signaling component is present as breast cancer cells themselves produce and secrete IGF-I and leptin and express cell surface receptors for both ligands. Thus, IGF-I and leptin represent a molecular link between adipose tissue and mammary tissue.

The IGF-IR and Ob-R signaling pathways have each been independently implicated in the development and progression of breast cancer. High circulating levels of IGF-I have been associated with an increased risk of developing breast cancer, and patients with existing breast cancer expressed high serum levels of IGF-I (17). In addition, transgenic mouse models overexpressing IGF-I, IGF-II, or IGF-IR showed an increased incidence of mammary tumor formation (16, 17, 23, 24). Conversely, liver-specific depletion of IGF-I caused reduced circulating levels of IGF-I in mice, resulting in diminished IGF-I endocrine effects on mammary tissue and, ultimately, reduced incidence of breast tumors (25). Similar to the IGF-I signaling pathway, leptin signaling has been associated with breast cancer. Leptin and its receptor were shown by immunohistochemistry to be overexpressed in primary and metastatic breast cancers relative to noncancer tissues (5). Expression of both leptin and Ob-R was most abundant among high-grade tumors, supporting a role for this pathway in breast cancer progression. In addition, in vivo models showed that whereas mice that overexpress transforming growth factor-α developed mammary tumors, leptin-deficient transforming growth factor-α mice were resistant to mammary tumor development (13), illustrating the important contribution of the leptin signaling pathway to some forms of breast cancer. Hence, because IGF-I and leptin are frequently detected in the serum of breast cancer patients and both receptors are overexpressed in a majority of breast tumors, we sought to determine whether molecular interactions occur between IGF-IR and leptin receptor in breast cancer.

We showed the following novel findings (Fig. 6):

(a) The IGF-I and leptin receptors interact in human breast cancer cells. Of potential interest, IGF-IR may preferably associate with Ob-Rt versus Ob-Rb in MCF7, BT474, and SKBR3 cells, as more of this isoform was pulled down in the IGF-IR immunoprecipitates (Fig. 2B); total levels of both Ob-R isoforms were similar in each line (Fig. 1A).

(b) Cross-signaling occurs from IGF-IR to Ob-R in breast cancer. IGF-I stimulation induces phosphorylation and activation of Ob-R, whereas IGF-IR kinase inhibition blocks IGF-I–mediated Ob-R activation. Downstream signaling molecules examined included JAK2, STAT3, Akt, and ERK1/2, all of which are functional in the leptin and IGF-IR pathways, as well as in multiple other signaling pathways. Thus, the IGF-I signaling experiments do not strictly indicate that IGF-I induces activation of one particular pathway. However, our results clearly indicate that IGF-I induces phosphorylation of Ob-R on Tyr¹¹⁴¹. Phosphorylation of Tyr¹¹⁴¹ is required for Ob-R to bind to the STAT3 transcription factor, which is then activated by JAK2 and translocated to the nucleus to stimulate transcription of downstream target genes (26). Thus, our results indicate that IGF-I activates Ob-R via the IGF-IR kinase.

(c) Cross-talk is unidirectional, as leptin does not activate IGF-IR. Whereas it is feasible that other phosphorylation sites on IGF-IR may be affected by leptin stimulation, the three sites examined here (Tyr¹¹³¹ and Tyr^1135/1136) were not affected by leptin. These three phosphorylation sites are the critical sites known to be required for IGF-IR mitogenicity and transforming activity (27). Thus, the inability of leptin to induce phosphorylation at these sites suggests that the leptin hormone alone is not likely to affect IGF-IR oncogenic function in breast cancer. However, because IGF-IR cross-talks to Ob-R, it is feasible that Ob-R may contribute to IGF-IR molecular or biological effects and is worthy of further study.

Thus, we have identified a novel receptor interaction and unidirectional cross-talk involving the IGF-IR and leptin receptor, which has not been previously described. Interestingly, Garofalo et al. (5) showed that IGF-I can induce leptin transcript levels in MCF7 cells. Our results further support this concept of IGF-I–mediated positive regulation of the leptin pathway.

Cross-talk from IGF-IR to other signaling pathways seems to be a potentially common mechanism used by cancer cells to enhance tumor growth and supports the significance of the IGF-I system to the biology of breast cancer, as well as the relevance of IGF-IR as a therapeutic target. We previously showed that IGF-IR cross-talks to the HER2 cell surface receptor in breast cancer cells that have become resistant to the HER2-targeted agent trastuzumab (19). Others have also shown that IGF-IR is capable of cross-signaling to the EGFR (18) and to the estrogen receptor (14). Thus, understanding the mechanisms by which IGF-IR mediates activation of other growth factor signaling pathways is important to breast cancer research. We have examined the role of the Src kinase family in mediating IGF-IR cross-talk to leptin receptor and have found that Src kinase inhibition does not inhibit IGF-IR/Ob-R cross-talk (not shown). Future studies will examine the molecular mechanisms mediating this receptor cross-talk. In addition, cotargeting leptin receptor and IGF-IR as a strategy to inhibit breast cancer progression, as well as the contribution of leptin receptor to IGF-I–mediated promitogenic and antiapoptotic effects, will be examined in breast cancer cells.

In summary, our results show, for the first time, that the IGF-I and leptin receptors physically form a protein complex in breast cancer cell lines and, further, that there exists a one-way cross-talk whereby IGF-IR induces phosphorylation and activation of the leptin receptor in breast cancer.

Human recombinant IGF-I (Sigma) was dissolved at 100 μg/mL in PBS and used at 100 ng/mL in culture. Human recombinant leptin (EMD Biosciences) was dissolved at 1 mg/mL in PBS and used at 100 or 1,000 ng/mL. I-OMe-AG538 IGF-IR kinase inhibitor (Sigma) was dissolved at 1 mmol/L in PBS and used at 10 μmol/L in culture.

MDA-MB-231 (MDA231), MCF7, BT474, and SKBR3 breast cancer cells were purchased from the American Type Culture Collection and maintained in DMEM supplemented with 10% FCS.

Cells were serum starved overnight, and then stimulated with IGF-I (100 ng/mL) for 1, 5, 15, 30, or 60 min or leptin (1,000 ng/mL) for 5 min, 0.5 h, 1 h, 3 h, or 6 h. In addition, a subset of cells were serum starved, treated with the IGF-IR kinase inhibitor I-OMe-AG538 (10 μmol/L overnight), and stimulated with IGF-I (100 ng/mL).

Total protein lysates (200 μg) were incubated with 1 μg of Ob-R or IGF-IR antibody or 1 μg rabbit IgG, rotating for 4 h, followed by addition of protein A/G-agarose (Cell Signaling) and rotating overnight. Beads were then washed thrice in PBS containing 0.1% Tween 20 and immunoblotted to detect Ob-R (H-300, Santa Cruz), IGF-IR (polyclonal, Cell Signaling), EGFR (monoclonal 1F4, Cell Signaling), or insulin receptor β (polyclonal, Cell Signaling). Blots of immunoprecipitations were quantitated using NIH imaging software ImageJ.

Cells were lysed in buffer containing 10 mmol/L Tris (pH 7.5), 100 mmol/L NaCl, 1 mmol/L EDTA, 1% NP40, and protease and phosphatase inhibitor cocktails (Sigma). Total protein extracts (50 μg) were immunoblotted using the following antibodies at the indicated dilutions: IGF-IRβ (polyclonal at 1:1,000; Cell Signaling); p-Tyr¹¹³¹-IGF-IR/Tyr¹¹⁴⁶-IR (polyclonal at 1:200; Cell Signaling); p-Tyr^1135/1136-IGF-IR/Tyr^1150/1151-IR (polyclonal at 1:200; Cell Signaling); leptin receptor (Ob-R; H-300 polyclonal at 1:200; Santa Cruz Biotechnology); p-Y¹¹⁴¹-Ob-R (polyclonal at 1:200; Santa Cruz); actin (monoclonal AC-15 at 1:5,000; Sigma Chemical); from Cell Signaling, polyclonal antibodies against p-STAT3 (Tyr⁷⁰⁵), total STAT3, p-JAK2 (Tyr²²¹), total JAK2 (24B11), total Akt, p-Thr²⁰²/Tyr²⁰⁴ p42/p44 MAPK (ERK1/2), total p42/p44 MAPK (ERK1/2), p-pThr¹⁸⁰/Tyr¹⁸² p38 MAPK, and total p38 MAPK, monoclonal 587F11 against p-Ser⁴⁷³-Akt, each used at 1:1,000 dilution, and monoclonal 1F4 anti-EGFR used at 1:200 dilution. Secondary antibodies were chosen according to the species of origin of the primary antibody. Protein bands were detected using the Odyssey Imaging System (Li-Cor Biosciences). Bands were quantitated using NIH imaging software ImageJ.

No potential conflicts of interest were disclosed.