During the progression of cancer and formation of metastasis, tumor cells enter the circulation and are seeded to distant organs where they have to resist and overcome a non-permissive environment to survive. These events can occur early and may already have taken place long before diagnosis of the primary tumor [95
]. Increasing evidence suggests that, like normal stem cells, tumor-initiating cells, termed cancer stem cells, do not depend solely on cell-intrinsic events but instead rely heavily on the right microenvironment - or niche - to maintain activity and fitness [96
]. However, unlike normal stem cell niches, which have evolved for millions of years, resulting in a fine-tuned crosstalk between stem cells and their environment, the cancer - or metastatic - niche evolves in a remarkably short time, resulting in more disordered interactions. The location of metastatic niches is also more loosely defined and can change as the disease progresses. Hypoxic regions, invasive fronts, perivascular sides and normal stem cell niches are all possible locations where metastatic niches can form. Normal stem cell niches are influenced by the stem cells themselves, but the metastatic niche takes this to new heights. Recruitment of inflammatory cells, endothelial cells and myofibroblasts to the metastatic niche leads to a tremendously complex milieu of growth factors, chemokines, hormones, enzymes and ECM that can promote stem/progenitor cell traits [97
]. The niche that these components form may provide cancer stem cells with the necessary support to survive and grow into overt metastasis.
The qualities of metastatic niches are beginning to be resolved. Despite the somewhat chaotic nature of these niches, interesting parallels can be drawn between them and normal stem cell niches. Certain qualities and molecular interactions within the cancer niche are indeed directly adopted from normal niches. Many of these components are inducers and regulators of stem/progenitor pathways like the Wnt, Notch, Hedgehog, phosphoinositide 3-kinase (PI3K) and JAK-STAT pathways [99
]. Moreover, evidence is accumulating on the importance of stem cell features in cancer progression and these properties are associated with poor clinical outcome [99
]. Intriguingly, evidence supports not only a passive role of the niche maintaining already established stem/progenitor cell traits, but also that niche components can induce the cancer stem cell phenotype in already differentiated cancer cells. In colon carcinoma, myofibroblasts express hepatocyte growth factor (HGF), a ligand of c-Met receptor tyrosine kinase, leading to co-stimulation and enhancement of Wnt signaling in differentiated cancer cells and promoting their stem/progenitor properties [101
]. This underscores the importance of the niche and may be a key feature of the cancer niche since the cancer stem cell phenotype may be a rather unstable and context-dependent trait [102
The initial events upon entry into distant organs can be critical and most of the cancer cells die soon after extravasation or stay dormant indefinitely [105
]. Interesting studies have proposed that signaling molecules from the primary tumor may cause changes in distant sites, thereby facilitating metastatic colonization. The environment that this generates has been termed a pre-metastatic niche [107
]. Secretion of vascular endothelial growth factor (VEGF)A, placental growth factor (PlGF) and inflammatory cytokines leads to mobilization of VEGF receptor 1 (VEGFR1)-expressing bone marrow-derived cells (BMDCs) and recruitment to the lung where they form a niche that enhances metastatic outgrowth (Figure ) [107
]. The pre-metastatic niche has also been shown to be enriched for molecules like fibronectin, matrix metalloproteinase 1/2, S100A8/9 and lysyl oxidase (LOX), leading to further recruitment of supportive stromal cells and to ECM remodeling, which together promote the growth of cancer cells entering the niche [108
]. To resist the negative forces the cancer cells encounter at distant sites, they take advantage of the molecular interactions and signaling normally active in niches. Interestingly, in some cases cancer cells can even seek out and 'hijack' already established healthy stem cell niches. This has been demonstrated in prostate malignancies, where cancer cells were shown to form micrometastases within HSC niches in the bone marrow and compete with HSCs for the niche interactions (Figure ) [109
]. The chemokine CXCL12 and C-X-C chemokine receptor 4 (CXCR4), to which it binds, form an axis that is a key molecular interaction between HSCs and the bone niche [110
], and is also engaged in bone metastasis of prostate cancer [111
]. In addition, other cancers that metastasize to bone also take advantage of this axis. In breast cancer, the CXCL12-CXCR4 axis is enhanced by high Src activity, reinforcing PI3K signaling and promoting survival of cancer cells lodged in the bone [112
]. Whether a competition similar to the one seen in the bone marrow niche also occurs in other stem cell niches remains to be seen. However, while the CXCL12-CXCR4 axis is a very important mediator of bone metastasis in cancers like breast-, prostate- and small cell lung cancer, this interaction also mediates metastasis to liver, brain and lungs [113
]. Indeed, the chemokine CXCL12 is expressed by myofibroblasts and in hypoxic regions [114
], both found in various metastatic sites and both potential locations for a metastatic niche.
Figure 8 Examples of metastatic niches during early colonization of distant organs. (a) Systemic changes induced by the primary breast tumor: mobilization of VEGFR1+ bone marrow-derived cells (BMDCs), recruitment to the lungs, extracellular matrix (ECM) remodeling (more ...)
Important components of the metastatic niche can be expressed by the cancer cells themselves, thereby making cancer cells self-sufficient in this regard since they bring their own niche material to the distant site. The cancer cells that can produce components of a supportive niche on their own will gain a significant advantage upon their arrival in a non-permissive environment. These components can be various growth factors, chemokines or secreted enzymes. Moreover, the ECM can play a significant role in these events. It is increasingly appreciated that the ECM provides more than a structural scaffold for cancer cells and is actively involved in modulating cellular signaling [116
]. Indeed, the ECM protein tenascin C (TNC) expressed in normal stem cell niches [117
] was recently demonstrated to play an important role in metastatic breast cancer. Modulation of stem/progenitor signaling pathways as a result of TNC expression by the cancer cells was shown to be essential to 'jump-start' the growth of lung metastasis in breast cancer (Figure ) [119
]. The expression of TNC is frequently found in circulating cancer cells isolated from the pleural effusion of patients with systemic breast cancer, suggesting that cancer cell autonomy in TNC production may have a role in the broad and efficient spread of the disease [119
]. Moreover, upon activation of the microenvironment, TNC is produced by myofibroblasts and contributes further to metastatic progression [119
]. In addition to TNC, myofibroblasts produce periostin (POSTN), another ECM protein recently identified as a component of the metastatic niche (Figure ) [121
]. Interestingly, the role of POSTN in formation of lung metastasis shows a striking similarity to the role of TNC, tempting us to hypothesize that these molecules could be inter-connected or collaborative components of the same supportive system [122
]. TNC and POSTN were demonstrated to regulate key signaling pathways involved in the maintenance of cancer stem cell features and activity of Wnt and Notch pathways [119
]. Disseminated cancer stem cells engage these pathways to resist the inhospitable environment at distant sites.
Today, metastasis is essentially an incurable disease and there is a desperate need for new measures to target metastatic progression. The microenvironment that metastatic cells engage and take advantage of to form a niche is a significant contributor to metastatic outgrowth. Moreover, the niche may possibly also contribute to cancer stem cell resistance to therapeutic intervention. Future studies may lead to identification of niche components that could provide new targets against metastatic progression. Targeting the niche and disrupting the nurturing effect it provides could present us with new means to prevent or even treat metastatic disease.