Bone voyage—Osteoblasts remotely control tumors

See allHide authors and affiliations

Science  01 Dec 2017:
Vol. 358, Issue 6367, pp. 1127-1128
DOI: 10.1126/science.aar2640

Cancer is a systemic disease. Tumor growth and malignant progression rely not only on the intrinsic aberrant genetic and epigenetic makeup of tumor cells but also on the tumor-induced systemic factors that affect cells in the primary tumor as well as distant microenvironments (1). Notably, bone marrow-derived cells (BMDCs) have been shown to contribute to primary tumor progression by promoting hallmark processes such as inflammation, immunosuppression, vasculogenesis, and extracellular matrix remodeling. BMDCs are also involved in establishing tumor-permissive microenvironments that form before the arrival of disseminated tumor cells at future metastatic sites (known as premetastatic niches) and promote metastatic outgrowth (25). In addition to the direct effects of tumor-secreted factors on BMDC recruitment to tumors, on page eaal5081 of this issue Engblom et al. (6) report that osteoblasts, which reside in the bone, can be remotely activated by secreted factors from lung adenocarcinoma, which in turn mobilize a specific subset of BMDCs—neutrophils—to foster tumor growth.

Using mouse models of lung adenocarcinoma, the authors showed that in the absence of metastasis, primary lung tumors promote the production of osteoblasts expressing osteocalcin (OCN+), increasing bone mass and density (osteopetrosis) and altering osteoblast gene expression. Remarkably, deletion of OCN+ cells reduced mature osteoblast numbers and led to significant tumor suppression, indicating that these OCN+ cells are functionally required for lung adenocarcinoma progression. Interestingly, Engblom et al. showed that this tumor suppressive phenotype can be rescued through parabiosis (a surgical process to join two animals that allows the sharing of blood circulation) of OCN+ osteoblast-deficient mice with OCN+ osteoblast-sufficient mice, which suggests that the phenotype is mediated by circulating factors stimulated by these immobilized bone-residing osteoblasts.

Indeed, the authors found that a specific subset of neutrophils that express CD11b and Ly6G was recruited to the primary tumor in an OCN+ osteoblast-dependent manner. Consistently, antibody depletion of CD11b+ Ly6G+ neutrophils in tumor-bearing mice harboring increased numbers of OCN+ osteoblasts significantly suppressed tumor growth. Further analysis revealed that OCN+ osteoblasts induced a particular subset of neutrophils that highly express the cell surface adhesion receptor sialic acid-binding immunoglobulin-like lectin F (SiglecFhigh neutrophils). This subset of neutrophils exhibited protumorigenic properties, promoting tumor growth in vivo.

Engblom et al. provide new insight on the systemic effects of osteoblasts during lung adenocarcinoma progression (see the figure). Osteoblasts and osteoclasts cooperate to control bone homeostasis, which is typically disrupted by cancer metastasis. The role of osteoblasts in establishing a favorable microenvironment for bone metastasis is documented in many cancers (7). Negative regulation of osteoblasts (that is, loss of proliferative phenotype and increased terminal differentiation) in osteolytic premetastatic lesion formation has also been reported (8). Here, Engblom et al. reveal a mechanism by which tumor-driven osteoblasts support primary lung tumor growth by supplying a specific subset of neutrophils remotely to the primary tumor. A similar mechanism of reciprocal interaction between primary tumor and osteoblasts was described in prostate cancer (9).

Future work is needed to understand the mechanism underlying osteoblast-induced neutrophil migration. For example, do these activated osteoblasts elevate the production of neutrophils, accelerate the transendothelial migration of neutrophils across the bone marrow endothelium, or up-regulate the expression of receptors on neutrophils that respond to the recruiting signals from the tumor? Engblom et al. demonstrate that osteopetrosis is a systemic effect of lung cancer, which also correlates with increased bone densities observed in some lung cancer patients. By contrast, osteoporosis is also a feature of many patients with cancer (7). Could this discrepancy be due to the tumor type-specific molecular signals that target osteoblasts and osteoclasts (which break down bone)? Or do other systemic effects [for example, those that induce cachexia (muscle wasting)] that indirectly lead to bone loss play a role in osteoporosis? Furthermore, it is tempting to speculate that other cell populations derived from the osteoblastic niches could also be influenced through osteoblast modulation and systemically contribute to tumor progression, a hypothesis that can be supported by the parabiosis experiment and antibody depletion of CD11b+ Ly6G+ cells carried out by Engblom et al. Addressing these questions will help elucidate the mechanisms underlying the systemic regulation of cancer progression.

Cancer is a systemic disease

Primary tumor mediates cross-talk with other organs via systemic factors to promote primary tumor growth and metastasis, and elicits systemic inflammation and a variety of other nonmetastatic systemic complications such as cachexia, osteopetrosis, osteoporosis, and paraneoplastic syndromes.


Neutrophils present in the tumor microenvironment contribute to tumor invasion, angiogenesis, and progression; however, whether neutrophils play strictly pro- or antitumor roles remains controversial (10). The dichotomous roles of neutrophils on tumors have thus limited our ability to therapeutically tackle tumor-associated neutrophils. Engblom et al. identified a specific neutrophil subset that is mobilized by tumor-instigated osteoblasts and recruited to the primary lung tumor, where they acquired a full protumorigenic gene expression profile. The gene expression profile of SiglecFhigh neutrophils has been associated with poor prognosis in lung cancer patients (6), suggesting that this transcriptomic signature could serve as a biomarker of poor prognosis.

Importantly, the molecular switch inducing the protumorigenic polarization of SiglecFhigh neutrophils upon encountering the primary tumor microenvironment remains unknown. Therefore, a better understanding of the mechanisms driving neutrophil polarization is required to allow manipulation of neutrophils for therapeutic purposes. Furthermore, SiglecFhigh neutrophils share similar phenotypic and functional profiles with granulocytic myeloid-derived suppressor cells (MDSCs), a major component of the systemic regulation of tumor progression via immunosuppression (11, 12). Because of the similar cell-surface markers that have been used to identify neutrophils and MDSCs, clarifying the molecular profiles that distinguish these two subpopulations is required for guiding precise therapeutic strategies.

Engblom et al. highlights the importance of identifying messengers that mediate the systemic cross-talk between tumor cells and distant organs to open up new avenues for therapeutic blockade of tumor progression. Notably, they showed that soluble receptor for advanced glycation end products (sRAGE) was elevated in the serum of lung tumor-bearing mice and that sRAGE accounts, at least partially, for the osteoblast activation and subsequent egress of neutrophils from bone marrow. The source of sRAGE, its effects on osteoblasts, and the identification of other potential factors mediating such cross-talk warrant further investigation. Indeed, to convey messages to distant organs, tumor cells can use different “messengers,” including chemokines, cytokines, cell-free nucleic acids, and extracellular vesicles, all of which are either secreted by or in response to tumors. Exosomes in particular can educate bone marrow cells to promote their mobilization to premetastatic sites (13). Thus, whether exosomes play critical roles in mediating the systemic cross-talk between tumor cells, osteoblasts, and neutrophils is worth exploring.

Engblom et al. add important insights to our understanding of cancer as a systemic disease. Do these neutrophils also contribute to metastatic disease (premetastatic niche establishment and metastatic colonization and outgrowth) of lung cancer in other organs? Or are independent mechanisms, or different subsets of neutrophils, involved in lung cancer metastasis? Are other organs affected by the systemic regulation of tumor progression? Cancer has adverse nonmetastatic effects, including cachexia and paraneoplastic syndrome (when tumor-induced factors cause damage in the central and peripheral nervous system) (14, 15). Does mobilization of neutrophils promote functional and metabolic alterations in other organs besides the lung to contribute to such nonmetastatic complications? A better understanding of the systemic cross-talk between tumor cells and remote cells—such as osteoblasts, which can influence tumorigenic and nontumorigenic complications—should aid in developing novel therapeutic approaches.


View Abstract

Stay Connected to Science

Navigate This Article