Background
Breast cancer is the second leading cause of cancer deaths in women in the United States and this tumor frequently metastasizes to bone. Upon arriving in bone, breast cancer cells disrupt normal bone remodeling by increasing bone resorption, leading to several serious clinical complications including life-threatening hypercalcemia, spinal cord compression, fractures, and extreme bone pain, which result in a significantly decreased quality of life [
1,
2]. Bone metastases have also been hypothesized to serve as reservoirs for breast cancer to metastasize to other tissues, such as the lung, liver, lymph node, or brain [
3]. Thus, breast cancer patients with bone metastases often have a poor prognosis [
4].
Breast cancer cells have been shown to promote bone resorption by enhancing osteoclast formation and function via a number of factors derived from the tumor including M-CSF, transforming growth factor (TGF)-β, tumor necrosis factor α, insulin-like growth factor II, parathyroid hormone related peptide, IL-1, IL-6 and IL-11 [
2,
5‐
7]. IL-11 is a member of the IL-6 family that recruits a homodimer of gp130, a promiscuous 130 kDa β subunit, after binding to their own non-signaling ligand-specific receptor, IL11R [
8,
9]. IL-11 is produced by a variety of stromal cells, including fibroblasts, epithelial cells, and osteoblasts and has a variety of functions, including being involved in multiple aspects of hematopoiesis, inhibition of adipocytogenesis, altering neural phenotype, stimulating tissue fibrosis, minimizing tissue injury, and regulating function of chondrocytes, synoviocytes and B cells [
10]. Apart from contributing to inflammation, gp130 signaling cytokines also function in the maintenance of bone homeostasis.
Cancer cells have been shown to directly produce IL-11 and to stimulate osteoblasts to secrete IL-11 [
11], which in turn is known to suppress the activity of osteoblasts [
12]. It has been shown that breast cancer cell lines produce IL-11 [
13] and that forced over-expression in cell lines increases tumor burden and osteolytic lesions in an
in vivo bone metastasis model [
5]. Moreover, human breast cancer tumors expressing IL-11 have higher rates of bone metastasis occurrences [
3]. Taken together, these observations support the notion that IL-11 plays an important role in breast cancer-induced osteolysis.
Using a knockout mouse model for IL-11, the cytokine was determined to be required for normal bone turnover, with the knockout mice exhibiting increased bone mass as a result of a reduction in osteoclast differentiation [
14]. IL-11 has been proposed to stimulate osteoclastogenesis independent of RANKL in one study [
15], whereas another study showed that IL-11 did not induce osteoclastogenesis unless marrow cells were co-cultured with calvaria cells [
16]. Similarly, other groups argue that IL-11 stimulates osteoblasts to secrete RANKL and/or proteinases [
17,
18]. Thus, while a functional role of IL-11 in the osteoclastogenic process has been well established, the molecular and cellular mechanisms by which IL-11 promotes osteoclast differentiation and function warrant further investigation. Given the known role of IL-11 in hematopoiesis [
10], we hypothesize that IL-11 may exert effects on osteoclast progenitor cells.
In the current study, we further characterize the role of IL-11 in supporting osteoclast formation, function and survival. Our data indicate that IL-11 promotes osteoclastogenesis primarily by increasing the pool of osteoclast progenitor cells. Consistently, we have also found that MDA-MB-231 conditioned media were able to support a population of bone marrow cells that are capable of differentiating into osteoclasts. These findings provide a better understanding of the mechanism by which IL-11 exerts its impact on osteoclast biology, and also suggest a new concept that breast cancer may also promote osteoclast formation by targeting osteoclast progenitor cells.
Methods
Chemicals and reagents
Chemicals were purchased from Sigma (St. Louis, MO) unless indicated otherwise. Recombinant GST-RANKL was purified as described previously [
19]. Recombinant mouse M-CSF (rM-CSF) (416-ML-010) and IL-11 (418-ML-005) were obtained from R&D Systems (Minneapolis, MN). Neutralizing anti-human IL-11 antibody (AB-218-NA) and normal goat IgG control antibody (AB-108-C) were also obtained from R&D Systems.
Animals
C57BL/6 mice were purchased from Harlan Industries (Indianapolis, IN). Mice were maintained, and the experiments performed in accordance with the regulations of the University of Alabama at Birmingham (UAB) institutional animal care and use committee (IACUC).
In vitroosteoclastogenesis assays
Breast cancer conditioned α-MEM was prepared by growing the human breast cancer line MDA-MB-231 to confluence, changing media to α-MEM plus 10% inactivated fetal bovine serum (iFBS), and collecting conditioned media after 24 hours. To generate osteoclasts from breast cancer conditioned media- dependent precursors, cells from the bone marrow cavities of the femur and tibia from C56BL/6 mice less than eight weeks of age were used. The bone marrow flushes were maintained in α-MEM for 24 hours at 37°C, in 7% CO2, and then cultured in breast cancer conditioned α-MEM or regular α-MEM supplemented with 10% iFBS. Media were changed every 3 days, and after 6 days, cells from the breast cancer conditioned α-MEM pretreated bone marrow flushes were plated in tissue-culture treated dishes at varying densities as indicated specifically in each experiment and treated with rM-CSF (10 ng/ml) and RANKL (100 ng/ml) for 4–6 days to form osteoclasts. Separately, IL-11 neutralizing antibody (5 ug/ml) was added to bone marrow flushes in 20% MDA-MB-231 breast cancer conditioned media, and surviving cells counted at days 3, 4, 5, and 6.
To generate osteoclasts from IL-11-dependant precursors, bone marrow flushes were maintained in α-MEM for 24 hours at 37°C, in 7% CO2, and then cultured in α-MEM with the presence of IL-11 (10 ng/ml) or equal volumes of PBS containing 0.01% bovine serum albumin. After 6 days, cells from the IL-11 pretreated bone marrow flushes were plated in tissue-culture treated dishes at varying densities as indicated specifically in each experiment and treated with rM-CSF (10 ng/ml) and RANKL (100 ng/ml) for 4–6 days to form osteoclasts. Separately, IL-11 neutralizing antibody (2 ug/ml) was added to bone marrow flushes in α-MEM with the presence of IL-11 (10 ng/ml), and surviving cells counted at days 3, 4, 5, and 6.
For IL-11 mechanistic studies of osteoclastogenesis, BMMs were isolated from marrow flushes of the long bones of 4–8-week-old C57BL/6 mice and were maintained in α-minimal essential medium (α-MEM) for 24 hours at 37°C, in 7% CO2, before separation with Ficoll gradient. To generate osteoclasts from BMMs, following Ficoll gradient separation, 1 × 105 or 5 × 104 cells, respectively, were plated in either 24-well or 48-well tissue culture plates. Cells were cultured in the presence of different concentrations and combinations of rM-CSF, RANKL, and IL-11 as indicated in individual experiments. The osteoclastogenesis cultures were stained for tartrate resistant acid phosphatase (TRAP) activity with a Leukocyte Acid Phosphatase kit (387-A) from Sigma. All assays were performed in triplicate and repeated at least three times. A representative view from each condition is shown.
In vitrobone resorption assays
5 × 104 BMMs were plated on bovine cortical bone slices in 24-well plates, and the cultures were treated with rM-CSF (10 ng/ml) and RANKL (100 ng/ml), or with IL-11, rM-CSF, or RANKL as detailed in each experiment. Cultures were maintained for 9 days to allow for bone resorption and then cells were removed from the bone slices with 0.25 M ammonium hydroxide and mechanical agitation. Bone slices were then subjected to scanning electron microscopy (SEM) using a Philips 515 SEM (Materials Engineering Department, University of Alabama at Birmingham). The percentage of the resorbed area was determined using ImageJ analysis software obtained from the National Institutes of Health.
Statistical analysis
Osteoclastogenesis data are expressed as mean ± standard error (SE) of numbers of TRAP-positive cells. Cell viability assays are expressed as mean ± SE of numbers of viable cells on each day counted. Statistical significance was determined using Student’s t test, and p values less than 0.05 were considered significant.
Discussion
Since the initial study showing the expression of IL-11 in breast tumor tissues more than 25 years ago [
27], numerous investigations have been subsequently undertaken to address the regulation and pathological significance of IL-11 expression in breast cancer and, in particular, in the tumor-induced osteolysis [
5,
13,
28‐
34]. Collectively, these studies have led to two important observations: a) IL-11 is not only expressed in a significant number of breast cancers but also has the potential to serve as a prognostic factor in human breast cancer, and b) IL-11 plays an important role in breast cancer-mediated osteolysis by promoting osteoclastogenesis and bone resorption. Notably, several studies have demonstrated that breast tumor cells can also target osteoblasts to stimulate their production of IL-11 [
11,
17], further increasing IL-11 concentrations in the bone microenvironment. Therefore, elucidation of the molecular mechanism by which IL-11 increases osteoclastogenesis and bone resorption in breast cancer bone metastasis may help guide development of effective drugs and/or therapeutic regimens for preventing and treating breast cancer-induced osteolysis.
Early studies on the role of IL-11 in osteoclast formation and function involved the use of the co-culture system containing bone marrow cells and calvarial osteoblasts [
16,
35]; the key finding of these early investigations was that IL-11-mediated osteoclastogenesis requires the presence of osteoblasts, but the precise reason for the dependence of IL-11-mediated osteoclastogenesis on osteoblasts was not fully understood. After the discovery of the RANKL/RANK/OPG system in the late 1990s, it then became clear that osteoblasts in the co-culture system primarily serve as a source of RANKL and IL-11 stimulates osteoblasts to produce RANKL [
36,
37]. This led to the notion that IL-11 can promote osteoclastogenesis indirectly by stimulation osteoblast production of RANKL. On the other hand, it was shown that osteoclasts express IL-11R [
35], suggesting that IL-11 may also directly target osteoclasts and/or its precursors to regulate osteoclast formation and/or function. Intriguingly, one study demonstrated that IL-11 directly target osteoclast precursors to stimulate osteoclastogenesis and it does so independent of RANKL [
15]. However, this finding is inconsistent with the early studies showing that IL-11-mediated osteoclastogenesis requires the presence of osteoblasts, which is a known source of RANKL.
In this work, we independently carried out a series of
in vitro studies to further address the role of IL-11 in osteoclastogenesis. First we determined that the conditioned media of MDA-MB-231, a breast cancer cell line expressing IL-11 [
13,
20], gave rise to a population of cells which can form osteoclasts in response to RANKL and M-CSF treatment (Figure
1), indicating that IL-11 may play an important role in osteoclastogenesis by regulating the development and/or survival of osteoclast progenitor cells. Because the MDA-MB-231 also secrete other factors that play a role in osteoclastogenesis it was necessary to look specifically at IL-11 function. Importantly, the ability of the breast cancer conditioned media to generate a population of osteoclast progenitor cells was significantly inhibited by a neutralizing anti-IL-11 antibody (Figure
3). These findings suggest that tumor-derived IL-11 may increase the extent of osteoclastogenesis by promoting the development of a population of osteoclast progenitor cells. To verify the specificity of IL-11, we found that culturing of murine bone marrow cells with IL-11 for 6 days is able to give rise to a pool of osteoclast progenitor cells (Figure
2).
We then investigated other ways that IL-11 may play a role in osteoclastogenesis. We found that IL-11 does not exert any effect on osteoclast survival (Figure
4). We then examined if IL-11 is able to promote osteoclast formation in the absence of RANKL and our data demonstrate that IL-11 cannot induce osteoclastogenesis in tissue culture dishes or on bone slices in the absence of RANKL (Figure
5). We and others have demonstrated that while IL-1 and TNF-α cannot promote osteoclastogenesis in the absence of RANKL, they can do so with suboptimal levels of RANKL or from RANKL-pretreated BMMs [
21‐
26]. As such, we then investigated whether IL-11 can act in a similar manner. Our data show that IL-11 is not able to promote osteoclastogenesis in the presence of suboptimal levels of RANKL (Figure
6) or from RANKL-pretreated BMMs (Figure
7).
Based on these new findings and those reported previously [
16,
34,
36,
37], we propose that IL-11-expressing breast cancer cells cause increased osteoclast formation and bone resorption by two distinct mechanisms: a) the tumor cells produce IL-11 which in turn stimulate the production of RANKL by stromal cells/osteoblasts in the bone microenvironment, and b) tumor cell-derived IL-11 also augments the pool of osteoclast progenitor cells to increase the extent of osteoclastogenesis. Therefore, our work has led to a better understanding of the action of IL-11 in breast cancer-induced osteolysis. However, the precise mechanism by which IL-11 promotes the development of a population of osteoclast progenitor cells remains unclear. While it is possible that IL-11 does so by stimulating the differentiation, proliferation and/or survival of osteoclast progenitor cells, this cytokine may exert the impact on osteoclast progenitor cell population indirectly through other cell types in the bone marrow. Further studies are needed to elucidate how exactly IL-11 promote the development of a pool of osteoclast progenitor cells.
Moreover, our new data may help guide the development of better therapeutic regimens for preventing and treating breast cancer-induced osteolysis. Particularly, denosumab, a humanized anti-RANKL developed by Amgen Inc, has been approved by the FDA to treat breast cancer-mediated osteolysis. For IL-11 positive tumors, denosumab may be effective only in blocking the RANKL-dependent action of IL-11. In contrast, it is likely that an efficient inhibition of IL-11 can block the IL-11-mediated increase of the pool of osteoclast progenitor cells as well as the RANKL-dependent pathway, thus having the potential to give rise to better efficacy. Future animal model studies need to be undertaken to address the therapeutic potential of targeting IL-11.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors read and approved the final manuscript. EM developed the idea, performed the experiments, analyzed the data, and prepared the manuscript. HH and HCP provided technical assistance. XF initially conceived the idea, and participated in the experimental design and manuscript preparation.