Role of Osteocytes in Cancer Progression in Bone and the Associated Skeletal Disease

Abstract

Purpose of review.

The goal of this manuscript is to review the current knowledge on the role of osteocytes in cancer in bone, discuss the potential of osteocytes as a therapeutic target, and propose future research needed to understand the crosstalk between cancer cells and osteocytes in the tumor niche.

Recent findings.

Numerous studies have established that cancer cells manipulate osteocytes to facilitate invasion and tumor progression in bone. Moreover, cancer cells dysregulate osteocyte function to disrupt physiological bone remodeling, leading to the development of bone disease. Targeting of osteocytes and their derived factors has proven to effectively interfere with the progression of cancer in bone and the associated bone disease.

Summary.

Osteocytes communicate with cancer cells and are also part of the vicious cycle of cancer in bone. Additional studies investigating the role of osteocytes on metastases to the bone and development of drug resistance are needed.

1. Introduction

Bone is a dynamic mineralized connective tissue. It provides support for body movement and protects vital internal organs. It is also considered a reservoir of calcium and phosphate, and provides an environment to the bone marrow. The bone marrow microenvironment contains mainly two different cell populations of stem cells: hematopoietic and mesenchymal. Hematopoietic stem cells differentiate into blood cells, as well as osteoclasts, the bone resorbing cells. Mesenchymal stem cells differentiate into osteoblasts, the bone forming cells, osteocytes, fat cells (adipocytes), chondrocytes, and myoblasts.

Bone is constantly being replaced by a process known as bone remodeling [1]. Bone remodeling is a complex cycle that requires the coordinated and balanced action of osteoclasts, osteoblasts, and osteocytes, which together form a temporary anatomical structure called a basic multicellular unit [2, 3, 1]. During physiological bone remodeling, old/damaged bone is replaced by the same amount of new bone. This process occurs in five phases: (1) activation, (2) bone resorption, (3) reversal, (4) formation, and (5) quiescence. Osteocytes are currently considered as key initiators and drivers of bone remodeling [4]. Apoptotic osteocytes might activate bone remodeling by sending signals to lining cells, which retract from the bone surface and to osteoclasts precursors to promote targeted recruitment. Once osteoclasts mature, they initiate the resorption of the old/damaged bone and undergo apoptosis. Osteoclasts are a known source of coupling factors that recruit osteoblast precursors to the resorption pit and enable spatial and temporal coordination of bone resorption with bone formation [5]. Osteoblasts are responsible for phase 4. They produce new bone made of collagen and other proteins, and control calcium and mineral deposition. After completion of matrix formation, osteoblasts have three possible fates: differentiate further and become osteocytes, die by programmed cell death, or transform into inactive quiescent bone lining cells covering the bone surface.

Alterations in the balance between the activity of osteoblasts or osteoclasts leads to bone mass changes, as seen in several genetic disorders favoring osteoblast formation resulting in bone gain, or in osteoporosis with a concomitant decline in osteoblasts and sustain activity of osteoclasts that results in bone loss. Different types of cancer cells grow in bone and markedly disrupt physiological bone remodeling. Primary bone tumors, such as osteosarcomas, chondrosarcomas, Ewing sarcomas, and chordomas, initiate from cells present in the bone tissue [6]. Other cancers start in bone but not from bone cells. This is the case of multiple myeloma, a hematological cancer of plasma cells that initiates in the bone marrow and causes bone lesions [7, 8]. Lastly, metastatic bone tumors develop from cancer cells that metastasize to the bone from other organs [9].

Bone is one of the most favorable sites for the growth of cancer cells [10]. Steven Paget originally proposed the seed and soil hypothesis, which stated that cancer cells metastasize to local favorable microenvironments to grow, just like seed needs a fertile soil to grow into a complete plant [11]. Bone is a highly vascularized tissue and contains multiple soluble factors. When released from the mineralized matrix, their factors promote tumor growth, initiating a vicious cycle that disrupts normal bone homeostasis and further stimulates tumor growth [12, 13]. In cancer-induced osteolytic bone disease (i.e. bone metastatic breast cancer and myeloma), tumor cells increase osteoclastogenesis and bone resorption to sustain tumor growth [12]. In osteoblastic bone lesions (i.e. osteosarcomas and prostate cancer metastasis), tumor-derived factors stimulate osteoblasts differentiation, leading to the formation of woven tissue with compromised mechanical properties [14]. In turn, osteoblasts produce growth factors that fuel tumor growth. The growth of cancer cells in bone has a strong negative impact on patients’ quality of life and represents a significant cause of morbidity and mortality.

Much is known about the contribution of communication between cancer cells and osteoblasts and osteoclasts to tumor progression in bone. However, the role of osteocytes, the most abundant cells in bone and key regulators of bone homeostasis, in cancer in bone is just starting to be revealed. In this review, we focus on the emerging data supporting that osteocytes contribute to the progression of cancer in bone.

II. Osteocytes in bone physiology

Osteocytes make up between 90–95% of the cellular component in bone and are considered permanent residents of the skeleton, where they can live for decades [15, 16]. Osteocytes belong to the mesenchymal lineage, and differentiate from osteoblasts that become entombed by the mineral they produce [17]. Mature osteocytes lived in lacunae within mineralized bone and are star-shaped looking bone cells, with lots of dendrite like structures protruding out from their cell bodies [15, 16]. Osteocytes communicate with neighboring osteocytes and other cells in the bone marrow via the lacuno-canalicular system, a large and complex network intercommunicating the lacunae and canaliculi containing both the osteocytes and their cytoplasmic processes [18]. Although initially thought to be passive cells, now we know that osteocytes play a key role in bone homeostasis. First, osteocytes are the main mechanosensing cells in bone, with a high ability to detect mechanical stimuli and send signals to other effector cells such as osteoclasts and osteoblasts [19, 20]. Second, osteocytes are endocrine cells that produce and secrete a number of factors with a strong influence on osteoblasts and osteoclasts biology. For instance, osteocytes secrete Sclerostin, a potent inhibitor of Wnt signaling that negatively regulates osteoblast differentiation and survival [21, 22]. Additionally, osteocytes are the major source of Receptor activator of nuclear factor kappa-β ligand (Rankl in adult bone, and thus regulate osteoclast differentiation and function [23, 24]. Lastly, as discussed above, osteocyte apoptosis is known to stimulate recruitment of osteoclast precursors and initiate targeted remodeling.

Accumulating evidence supports that alterations in osteocyte viability and function contributes to multiple skeletal pathologies. This is the case for osteoporosis, as well as the decline in bone mass with aging, both accompanied by increased apoptotic osteocytes [25]; hypophosphatemia, characterized by osteocyte-derived Fibroblast growth factor 23 (Fgf23) excess [26]; and several genetic high and low bone mass diseases [27, 28]. Give their importance in bone biology, osteocyte and their derived factors have become effective targets for the treatment of skeletal diseases (see section VII). As discussed below, emerging evidence shows that osteocytes also play an important role in the progression of cancer in bone, providing a microenvironment that is conducive to tumor growth and bone disease.

III. Osteocytes and hematological cancers: multiple myeloma

Multiple myeloma (MM) is a hematological cancer, caused by the uncontrolled proliferation of plasma cells in the bone marrow [7, 8]. The growth of MM cells in bone disrupts normal bone remodeling, leading to increased osteoclast formation and exacerbated bone resorption, as well prolonged suppression of osteoblast differentiation and new bone formation [29–31]. Consequently, MM patients present with osteolytic lesions that weaken the bones and favor pathological fractures. Interactions between MM cells and cells of the bone marrow microenvironment (osteoblasts, osteoclasts, stromal cells, adipocytes) lead to the formation of favorable niches that induce growth and survival of the MM cells and bone destruction [32]. The first evidence supporting the role of osteocytes in multiple myeloma was provided by Giuliani and colleagues. They showed in MM patients that osteocyte apoptosis is increased in areas of bones infiltrated with cancer cells [33]. Using a mouse model of MM-induced bone disease, our group also reported increased in osteocyte apoptosis in bones bearing MM cells [34]. Mechanistic studies demonstrated that MM cells physically interact with the osteocytes and activate Notch signaling, which together with MM-derived TNF, leads to osteocytes apoptosis [34]. Recent findings suggest that MM-induced osteocyte apoptosis could also be mediated by autophagy [35]. The increase in osteocyte apoptosis was sufficient to stimulate the recruitment of osteoclast precursors [34]. In addition, MM cells increased RANKL and IL-11 expression, and decreased OPG production in osteocytes [34, 33]. Together, these results support that both healthy and apoptotic osteocytes contribute to the exacerbated, local bone resorption seen in MM.

Accumulating evidence also supports that osteocytes contribute to osteoblasts suppression in the MM niche. Sclerostin, a potent inhibitor of bone formation [36, 37], is elevated in the circulation of MM patients [38]. Consistent with this clinical observation, Sclerostin production by osteocytes is increased in bones injected with MM cells [34]. Further, osteocytes are an important source of Dickkopf-related protein 1 (Dkk1), another Wnt signaling antagonist elevated in MM cells [39]. Genetic and pharmacologic approaches demonstrated that both Sclerostin and Dkkk-1 contribute to osteoblast suppression in MM [39–43]. More recently, research efforts from our group showed that MM cells stimulated Vascular endothelial growth factor-a (Vegf-A) production in osteocytes, and that osteocyte-derived Vegf-a contributes to the increased angiogenesis in MM [44]. Lastly, Fgf-23 in osteocytes is upregulated by MM cells, which in turn regulates the expression of osteocytic Vegf-a and heparanase, a pro-osteolytic factor, in MM cells [45].

Osteocytes are also able to communicate with MM cells and alter their function. The physical interaction between MM cells and osteocytes also results in Notch signaling activation in MM cells [34]. This activation increases cyclin D1 mRNA expression level and increases the proliferation of MM cells. Interestingly, osteocytes also altered the Notch receptor repertoire in MM cells by rapidly increases the expression of Notch receptors 3 and 4. Using genetic silencing tools, we recently showed that Notch receptor 3, but not Notch receptor 2, mediates the communication between osteocytes and MM cells via Notch [46]. Lastly, genetic ablation of osteocytes increased the homing of myeloma cells to particular areas of the bone and increased total tumor burden [47], suggesting that osteocytes may play a role in MM cell homing. Altogether, the results highlighted in this section provide compelling evidence supporting that osteocytes contribute to generate a microenvironment conducive to tumor progression, angiogenesis, and bone destruction in multiple myeloma.

IV. Osteocytes and cancer metastasis

Bone metastases are a common manifestation in patients with breast, prostate and lung cancer. Bone metastasis occurs when cancer cells migrate from a distant primary site and invade and colonize the bone [48, 49]. Metastasis to the bone is a complex process that begins with metastatic tumor cells undergoing epithelial to mesenchymal transition (EMT) and detaching from the primary tumor to enter into the circulation [50]. Once in bone, cancer cells undergo apoptosis, with only a few becoming dormant and surviving [51]. Interactions with resident bone cells such as osteoblasts, facilitate the homing and control the fate of metastatic cancer cells in bone [52, 32]. Although osteocytes have the potential to contribute to bone metastasis and homing of cancer cells to particular areas of bone, their role in these processes remains largely unknown. Below we summarize the main findings supporting a role of osteocytes in the establishment and progression of bone metastasis.

V. Osteocytes and breast cancer

Studies have suggested that the interaction between metastatic breast cancer cells and osteocytes not only contributes to bone destruction, but also to migration, invasion, EMT and its reversal, and tumor growth. For instance, treatment of breast cancer cells with conditioned medium from osteocytes induced EMT, increased migration, and stimulated proliferation of breast cancer cells [53, 54]. Mechanistically, these effects appear to be mediated by changes in Connexin 43, Snail, and Wisp signaling, as well as adenosine nucleotides released by osteocytes [55–59]. Interestingly, conditioned media from mechanically stimulated osteocytes decreased the expression of EMT-related genes and their invasion capacity [60]. These results suggest an important role of mechanical signals in the crosstalk between osteocytes and breast cancer cells. Although there is no direct evidence, osteocytes could also be involved in the early stages of breast cancer colonization of bone by releasing Cxcl12 [61], a chemokine involved in cancer cell dormancy [62, 63]. Similar to multiple myeloma cells, metastatic breast cancer cells also appear to alter osteocyte function. Conditioned media from breast cancer cells reduced osteocyte viability and increased their Rankl/Opg ratio [64]. Knowledge of the role of osteocytes on breast cancer metastasis remains limited. Thus, further studies are needed to clarify the role of osteocytes in breast cancer and the potential of mechanical signals for the treatment of breast cancer metastasis to bone (see section VII).

VI. Osteocytes and prostate cancer

In vitro experiments with conditioned media show that osteocytes also stimulate proliferation, migration, and invasion capacity of prostate cancer cells [53]. Mechanistically, Keller’s group demonstrated that osteocytes secrete Gdf15 to stimulate the growth and invasiveness of protate cancer cells [65]. Osteocyte-derived Fgf23 also enhances proliferation and invasion of prostate cancer cells, and the genetic knockdown of Fgf23 decreases tumor growth in vivo. Interestingly, FGF23 can also be produced by prostate cancer cells and could act in both paracrine and/or endocrine manner to promote prostate cancer progression. The growth of prostate cancer cells in bone also alters osteocyte biology [66]. Tumors growing in mouse tibiae increased intraosseous pressure and mechanically stimulated osteocytes, which in turn produced Ccl5 and matrix metalloproteinases to stimulate tumor proliferation and invasion. Intriguingly, the serum levels of the Wnt antagonists Sclerostin and Dkk-1 levels are elevated in patients with prostate bone metastasis [67]. This elevation might be a compensatory response to the increased osteoblasts presented at metastatic skeletal sites.

VII. Osteocytes and their derived factors as therapeutic targets

Osteocytes and their derived factors have become a valuable target for the treatment of bone disease. In this section, we will discuss the potential of targeting osteocytes for the treatment of cancer in bone and the effects of chemotherapy on osteocytes.

Sclerostin.

Sclerostin binds to Lipoprotein receptor-related proteins (Lrp) 4, 5, and 6 to downregulate Wnt/beta-catenin signaling and inhibit bone formation [68–70]. Several neutralizing antibodies against Sclerostin have been developed and have shown remarkable ability to stimulate new bone formation, and transiently decrease bone resorption, in osteoporotic patients. Thus, Sclerostin is currently recognized as a target for the treatment of osteoporotic bone loss [71]. Interestingly, Sclerostin circulating levels are elevated in patients with cancers that grow in bone [72, 38, 73, 74]. We and others demonstrated in preclinical models of MM that Sclerostin plays an important role in myeloma-induced suppression of bone formation. Inhibition of Sclerostin using genetic and pharmacologic approaches prevented the suppression of osteoblasts in bones bearing MM cells. More importantly, treatment with neutralizing antibodies stimulated osteoblast differentiation and promoted new bone formation in bones infiltrated with MM cells. Treatment with anti-Sclerostin had no effect on tumor burden across the different animal models used [42, 43, 41]. More recently, similar findings have been shown in a preclinical model of bone metastatic breast cancer [75]. Together, these results support that targeting osteocyte-derived Sclerostin is a novel therapeutic approach to combat cancer-induced bone disease. Despite the promising results obtained in preclinical models, to the best of our knowledge, there are no clinical trials studying the effects of anti-Sclerostin in cancer patients with bone involvement.

Dkk-1.

Dkk-1 is abundantly produced by osteocytes and osteoblasts, as well as by some cancer cells like MM cells, breast and prostate cancer cells. In MM, pharmacological inhibition of Dkk1 results in bone protection due to both increases in osteoblasts and decreases in osteoclasts [39, 40]. However, the in vivo effects of Dkk-1 inhibition on tumor were controversial between studies. Several clinical trials in different populations of MM patients were performed but, as of today, Dkk-1 is not an approved therapeutic target for the treatment of MM. Of note, a new bispecific antibody targeting both Sclerostin and Dkk-1 has been developed, which has more potent anabolic effects that targeting each factor alone. [76] However, the effects of this dual antibody in humans or cancer in bone remain to be determined.

Fgf-23.

Fgf-23 is an endocrine hormone secreted by the osteocytes that plays an important role in the regulation of phosphate levels [77]. Fgf-23 levels are elevated in several cancers, including prostate cancer and myeloma. [78, 45, 79, 80]. Importantly, low serum Fgf-23 levels correlate with survival in prostate cancer patients [78], suggesting that targeting it could be a therapeutic target to inhibit the progression of cancer. Earlier this year, the FDA approved the use of Burosumab, a neutralizing antibody against Fgf-23 to treat patients with treat X-linked hypophosphatemia (XLH), as well as tumor-induce osteomalacia [81]. Future research efforts are needed to define the effects of Burosumab in cancer that grows in bone.

Notch signaling.

The Notch signaling pathway mediates communication between adjacent cells and controls proliferation and programmed death programs. The dysregulation of Notch signaling leads to the progression of several cancers in bone [82, 83]. Pharmacologic inhibition of the Notch pathway with gamma secretase inhibitors (GSIs) in cancer results in multiple beneficial outcomes, including decreased tumor growth, maintenance of osteocyte viability, and inhibition of bone destruction [84, 85]. However, GSIs induce severe gut toxicity, limiting their use in the clinic [86]. Our group is developing a novel bone-targeted Notch inhibitor to bypass gut toxicity. Preliminary results showed that bone-targeted inhibition of Notch decreases tumor growth and bone destruction, without inducing gut toxicity. Although this approach is promising, identification of specific components of the Notch pathway that contribute to cancer progression in bone is paramount to avoid the off-target effects of GSIs [84]. In this line, neutralizing antibodies against Notch receptors and ligands Delta-1 and Jagged-1 are currently being studied in preclinical models [83, 87]. Future research efforts are needed to define the precise Notch receptors and ligands involved in the interactions between cancer cells and osteocytes.

Connexin 43 hemichannels.

Connexin 43 is a protein expressed in osteocytes that forms gap junctions and mediates communication among nearby osteocytes and between osteocytes other cells in the bone [88, 89]. Connexin 43 hemichannels are mechanosensitive and mediate osteocytic response to mechanical signals [88]. Osteocytic Connexin 43 hemichannels are required for the full inhibitory effects of conditioned media from mechanically stimulated osteocytes on migration, invasion, and growth of breast cancer cells [90, 59]. Further, the permeability of connexin 43 channels for small molecules makes them highly attractive for delivering drugs directly into the cytoplasm. Several small molecules targeting connexin 43 have been developed and are currently being tested for the treatment of osteosarcoma (i.e. ALMB-0168³; humanized Cx43 monoclonal antibody), which could be used in the future for the treatment of other cancers that infiltrate bones.

Mechanical stimulation.

Exercise promotes skeletal preservation by regulating bone remodeling via osteocytes and is often proposed as an intervention for cancer patients [91]. For instance, tibial compression decreases osteolysis and tumor formation in a human metastatic breast cancer animal model [60, 57]. Therefore, mechanical stimulation of osteocytes might be considered as a treatment for breast cancer patients. However, it is important to note that exercise and the associated mechanical stimulation in bone might not always be safe and possible during cancer treatment. Several studies have investigated the effects of delivering low-intensity vibration as a mean to deliver mechanical signals without exercise [91]. In preclinical models, low-intensity vibration signal decreased tumor progression and bone loss in several models of cancer, including MM [92, 93]. This promising approach is currently being tested in the clinic.

Rankl.

The pro-osteoclastogenic cytokine Rankl plays a key role in cancer-induced bone loss by promoting osteoclast differentiation and bone resorption [94]. New developments demonstrate that besides bone destruction, soluble Rankl, formed after cleavage of membrane-bound Rankl, could also be involved in bone cancer metastasis [95]. Denosumab is a monoclonal antibody against Rankl with strong anti-resorptive activity [96]. Currently, this agent is only approved to prevent skeletal-related events in cancer patients with solid tumors (breast and prostate) and bone metastases. However, a recent study showed that Denosumab has non-inferiority for the prevention of skeletal-related events in multiple myeloma patients when compared with zoledronic acid [97]. It is likely that osteocytes mediate some of the protective effects of Denosumab, as they are the main source of Rankl and cancer cells increase Rankl expression in osteocytes.

Effects of current anti-cancer therapies on osteocytes:

It is clear now that anti-cancer therapies have an impact on bone and osteocytes [98]. Bisphosphonates are potent anti-resorptive agents and the gold standard approach to treat/prevent skeletal-related events induced by cancer. Besides decreasing bone loss by inhibiting osteoclast activity, bisphosphonates prevent osteocyte apoptosis, which also contributes to better bone health [99]. The synthetic glucocorticoid dexamethasone is often used in combination with other chemotherapy due to its ability to induce cancer cell apoptosis. Glucocorticoids have deleterious effects in bone partially mediated by increasing apoptosis in osteocytes and osteoblasts [100]. Bortezomib is a first-in-class proteasome inhibitor that blocks the degradation of abnormal or misfolded proteins targeted for destruction [101]. These proteins accumulate and result in MM cell apoptosis. In addition to its anti-cancer effects, proteasome inhibitors also transiently increase bone formation by acting on osteoblasts [102]. In osteocytes, Bortezomib prevents the increase in osteocyte apoptosis induced by MM cells [35]. Another anti-cancer agent with effects in osteocytes is Aplidin [103, 104]. This agent, isolated from a marine truncate Aplidium albicans, exhibits anti-MM activity. Aplidin also induces osteocyte programmed cell death [105]. Interestingly, this effect is prevented when combined with Bortezomib [105]. Lastly, aromatase inhibitors are used to treat endocrine-responsive breast cancer [106]. Aromatase inhibitors induce bone loss, a process in part mediated by increases in osteocyte-derived Sclerostin [107]. For more effects of anti-cancer therapy in osteocytes and bone, see here [108].

VIII. Conclusion

Interactions between tumor cells and bone cells contribute to tumor progression in bone. Early studies demonstrated that osteoblasts and osteoclasts are important in the vicious cycle that fuels tumor progression in the skeleton. Advances in the last decade have provided compelling evidence that osteocytes communicate with cancer cells and thus are also part of the vicious cycle of cancer in bone. Now we know that cancer cells “educate” osteocytes to promote tumor progression and bone destruction (Figure 1). Further, preclinical work in the last years have proven that targeting osteocytes and their derived factors is an attractive strategy to prevent bone destruction and slow down tumor progression.

Figure 1. Bidirectional crosstalk signaling between cancer cells and osteocytes in the bone niche.

Osteocytes and cancer cells communicate via exchange of soluble factors and physical interactions. Cancer cells induced osteocyte apoptosis and increase their production of osteoclastogenic factors to further promote osteoclast formation and bone resorption. Further, cancer cells stimulate osteocyte production of Wnt antagonists, which in turn contribute to suppress osteoblast formation and new bone formation. Lastly, cancer cells also increase Vegf-a secretion by osteocytes to enhance angiogenesis in the tumor niche. Osteocytes communication with cancer cells results in increased proliferation and tumor growth, and increased invasiveness and migration in cancer cells, ultimately favoring bone metastasis.

Yet, several questions regarding the role of osteocytes in bone metastasis and the growth of cancer in the bone remain to be determined. For instance, it is hypothesized that primary tumors modify and prime the sites of future metastasis, creating a pre-metastatic niche that favors their growth and survival in bone [109]. However, if osteocytes contribute to generate a microenvironment conducive for bone metastasis is unknown. Moreover, although pharmacological targeting of osteocytes improves cancer-induced bone disease, it is unknown whether therapies targeting osteocytes can simultaneously 1) repair the osteolytic lesions and 2) preserve the anti-cancer effects when combined with chemotherapy. Osteoblasts are key regulators of cancer cell dormancy. Yet, whether direct effects of osteocytes or indirect, through their influence on osteoblasts, contribute to the regulation of dormancy demands further investigation. Another interesting underexplored area is the role of osteocytes in the development of resistance to chemotherapy. Identification of the specific mechanisms mediating the different functions that osteocytes perform in bones colonized by cancer cells will lead to a better understanding of cancer cell dissemination, survival, and proliferation in bone and guide the development of more efficient therapeutic approaches to treat cancer patients with bone lesions.