microRNAs, Gap Junctional Intercellular Communication and Mesenchymal Stem Cells in Breast Cancer Metastasis

Abstract

The failed outcome of autologous bone marrow transplantation for breast cancer opens the field for investigations. This is particularly important because the bone marrow could be a major source of cancer cells during tertiary metastasis. This review discusses subsets of breast cancer cells, including those that enter the bone marrow at an early period of disease development, perhaps prior to clinical detection. This population of cells evades chemotherapeutic damage even at high doses. An understanding of this population might be crucial for the success of bone marrow transplants for metastatic breast cancer and for the eradication of cancer cells in bone marrow. In vivo and in vitro studies have demonstrated gap junctional intercellular communication (GJIC) between bone marrow stroma and breast cancer cells. This review discusses GJIC in cancer metastasis, facilitating roles of mesenchymal stem cells (MSCs). In addition, the review addresses potential roles for miRNAs, including those already linked to cancer biology. The literature on MSCs is growing and their links to metastasis are beginning to be significant leads for the development of new drug targets for breast cancer. In summary, this review discusses interactions among GJIC, miRNAs and MSCs as future consideration for the development of cancer therapies.

Introduction

Breast cancer cell (BCC) entry into the bone marrow (BM) could lead to establishment of dormancy by the cancer cells. This type of quiescence could explain tertiary metastasis and/or resurgence of breast cancer from the BM [1]. Two major questions are: I) Is there a population of dormant BCCs in the BM ready for resurgence? II) Can inflammation transition the dormant BCCs in bone marrow to highly metastatic cells? An understanding of the mechanism that facilitates dormancy holds the potential for therapy. This could prevent cancer cells quiescence and/or reverse dormancy. If so, this could be a key point in the development of novel strategies to target cancer cells while in they are in BM. This will have a tremendous impact on metastatic breast cancer. Gap junctional intercellular communications (GJIC) between BCCs and cells of the BM microenvironment appear to have roles in the dormancy of cancer cells in BM [2].

BCCs could enter BM through the lymphatic system and also by hematogenous route. Experimental studies were conducted to understand the entry of BCCs in BM and the data suggest that the efficiency could be improved if the cancer cells are in contact with mesenchymal stem cells (MSCs) [3]. This efficiency could be partly explained by enhanced transmigration of BCCs across BM endothelial cells in the presence of MSCs [3]. The role of MSCs in the migration of BCCs across endothelial cells involves cytokine production from both cells types [4]. These are studies that need to be explored since the cytokines could be important in the fine-tuned regulation between BCCs and MSCs in the survival of the cancer cells in BM.

BCCs are found in cycling quiescence phase close the endosteum [1]. Furthermore, the same area shows BCCs in contact with stroma through the establishment of GJIC [2]. It is highly possible that a dormant phenotype of the BCCs is acquired via microRNA (miRNA) exchanges between the two cell types. These are ongoing studies, including investigations in our laboratory. However, speculation could be made and tested since the process of GJIC-dependent miRNA transfer between BCCs and stromal cells could lead to future cancer therapies.

Inflammation has been linked to cancer resurgence in patients [5–9]. This strongly suggests that inflammatory mediators might transition dormant cancer cells into more metastatic cells with high proliferative abilities [5]. On the other hand, if the dormant cancer cells are cancer stem cells, it would be important to determine if cytokines can induce maturation of the stem cells to highly proliferating cancer progenitors. Markers of chronic inflammation are associated with increased breast cancer invasion and reduced survival in primary and resurgent breast cancers [6,7]. A proposed mechanism by which inflammation promotes cancer recurrence might be explained by enhanced extravasation of metastatic cells and angiogenesis through changes in the microenvironment [5,8,9]. This association between inflammatory markers and breast cancer recurrence may be valuable in the development of a novel diagnostic tool to monitor the recurrence of disease in previously treated patients [6,8]. While there are several investigations on the mechanisms by which dormant cancer cells exit quiescence, little is known about the method by which cancer cells enter dormancy and which organs are important to maintain such states. The latter mechanism is equally important because it could lead to the development of new therapies to prevent dormancy or subsequently remove cancer cells from quiescence for targeting by current therapies.

Bone Marrow (BM)

The adult BM is home to a finite number of self-renewing hematopoietic stem cells (HSCs), which replenish the immune system throughout life [10]. Hematopoiesis is regulated by complex network of soluble and insoluble mediators [11]. Although recent controversies question the major cellular support of hematopoiesis, the evidence, nonetheless, lies in favor of both stroma and osteoblasts, found close to the endosteal region with low oxygen levels [12]. The BM cavity is complicated by gradient changes in cytokines and oxygen [12]. Regions close to the endosteum show low oxygen levels as compared with areas within the central sinus [12]. Similar observations have been reported for the chemokine, CXCL12 [12]. It has been suggested that low oxygen is important for protecting HSCs from radicals and other insults [12]. Thus, it is logical that dormant BCCs would also prefer the same region for protection, similar to those afforded to HSCs. Given the above brief description, it is easy to appreciate the difficulty in modeling bone by bioengineering methods.

Gap Junctional Intercellular Communication (GJIC)

Cellular contacts form communication partly through gap junctions, referred as GJIC. Gap junctions are expressed on stroma cells, which are differentiated from MSCs and also on some BCCs [2]. GJIC are formed by members of the connexin (Cx) family, which comprises >20 members [13]. Generally, six Cx subunits form a hemichannel [13]. Oligomerizations could occur with homomeric or heteromeric Cxs [14]. Hemichannels could be formed by combinations of Cxs with pairs forming GJIC [13]. GJIC mediates exchanges of small molecules (<1000 Da), second messengers and ions. Human breast epithelial and BCCs express Cx43 and Cx26 [15,16]. Cx43 knockdown in BCCs resulted in an aggressive phenotype [15]. On the other hand, ectopic expression in aggressive BCCs results in reduced metastatic phenotype [13]. The significance of gap junctions in breast cancer has just begun to emerge.

While the Cxs have been established regulators of hematopoiesis, their role in cancer biology has just begun. Although this subject is at its infant stages of investigation, the role of Cxs in breast cancer is now beyond speculation. The field has emerged as an important area of investigation. Moharita et al. has reported on a significant relationship between BCCs and stroma, via GJIC [2]. In other studies, Donahue et al. show that MSC-derived osteoblasts form gap junctions with BCCs [16]. Upon the formation of these gap junctions, cytosolic calcium is mobilized, which activates the osteoblasts to retract away from one another to allow for BCC migration. [16] Roles for Cx32 have been investigated with primary breast tumors and metastases. Cx32 is not expressed in normal breast epithelium or myoepithelium. However, in ductal carcinoma, Cx32 has been shown to be increased in both the primary tumor and in the lymph node metastasis with the greatest expression on the cancer cells found in the lymph nodes [17]. Li et al. reported on decreased expression of Cx43 in a highly metastatic BCC line when compared with non-metastasizing breast epithelium and that this decrease was even greater in a bone homing cancer cell line [13]. The bone homing variant also showed a greater adherence to an osteoblast cell line [13]. Metastatic breast cancer expresses high levels of OB-cadherin, which is decreased after Cx43 is expressed [13]. This suggests that Cx43 may regulate adherence by interfering with the expression of OB-cadherin, and indicate that decrease in Cx43 could cause a decreased in the metastatic potential of cancer cells, especially to bone. These are significant findings due to the relevance of future therapy to decrease metastasis. Conflicting data showed Cx43 expression on a non-metastatic, GJIC-deficient mammary epithelial tumor cell line resulted in the formation of gap junctions and an increase in diapedesis [18]. The carcinogenic compound, organochlorine, was used to decrease GJIC in MCF-7 and also in human mammary epithelial cells [18]. Organochlorine appears to function by inhibiting the phosphorylation of proteins that are required to form GJIC [19].

Mesenchymal Stem Cells (MSCs)

MSCs are often referred as BM stromal stem cells [20]. However, this designation could be confusing since stromal cells are the differentiated cells of MSCs [21]. Therefore, we will use the designation MSC to avoid confusion since the discussion also focuses on BM stromal cells. MSCs are ubiquitously present in adult and fetal tissues [20,22]. In adults, the BM is the major site of MSCs. These stem cells are also present in umbilical cord blood, although at lower frequency [22]. Morphologically, MSCs are symmetrical cells with fibroblastoid appearance [23]. Phenotypically, MSCs express CD44, CD29, CD105, CD73, and CD166 and lack markers of hematopoietic lineage, in particular CD45 [23]. MSCs also express neural-associated markers, such as neural ganglioside, GD2, supporting the current evidence on the transdifferentiation potential of these cells [24,25].

MSCs show functional plasticity with regards to their immune properties by exerting both immune suppressor and enhancer functions [26]. MSCs can express varied cytokines and also express cytokine receptors, providing them with the ability to regulate their functions through autocrine and paracrine mechanisms [26]. MSCs show promise in cell therapy. However adjuvant treatment might be most efficient with anti-rejection drugs and/or with other drugs to facilitate tissue repair and/or replacement [26]. This review argues for adjuvant therapy because it is highly possible that efficient treatment with MSCs might be gained if the microenvironment is regulated by drug interventions. The argument is for particular drugs to preconditioning the microenvironment and then to direct the MSCs to act accordingly. These types of interventions would be available with ongoing investigations to identify cues that induce lineage-specific differentiation of MSCs [27].

The immunomodulatory properties of MSCs have received much attention in recent years. To reiterate, MSCs show therapeutic potential for inflammatory disorders. This benefit is mostly due to the immune suppressor functions of MSCs as well as their utility to be used for ‘off the shelf’ delivery [28]. The use of MSCs in Crohn’s disease, for example, has been suggested as a method to relieve widespread inflammation in the gastrointestinal tract [28]. There are global investigations to explore on mechanisms by which MSCs mediate immune suppression through their plasticity. MSCs suppress T-cell activation and dendritic cell maturation, and also inhibit their migration to secondary lymphoid organs [13,28]. Contact-dependent and -independent factors are involved in immunosuppression by MSCs, indicating the complexity in the biology of MSCs with regards to the immune system [29]. Clinical trials for graft-versus-host disease have used MSCs have shown delays in disease progression [30]. The mechanisms, although unclear, have implicated interferon-γ [30]. Despite the immune suppressive properties of MSCs, it appears that MSCs are safe, due to their use in clinical trials, and also other experimental studies that showed viral clearance, even in their presence [31].

Despite the advantages of MSC therapy for immunological and inflammatory diseases, the prospect of cancer initiation by MSCs has emerged [32]. The discovery of two key properties of MSCs has led investigators to explore MSCs as possible mediators of malignancy. Firstly, MSCs are themselves likely to undergo spontaneous transformation ex vivo in long-term culture [32]. MSCs have been shown to form dermatofibrosarcoma protuberans, fibrous histiocytoma, and other sarcomas [32–35,]. The Wnt pathway has been show to exert tumor suppression and its inhibition has led to the formation of sarcomagenesis [35]. This observation however, is complex since other studies with Kaposi sarcoma reported on a decrease in tumor formation when the Wnt pathway was inhibited [36]. A second, and arguably more important, function of MSCs with regards to tumorigenesis is the consistency in reports showing a supporting role by MSCs not only in BM but at distant sites [37]. Postulated reasons for MSC-mediated effects on tumorigenesis include an advantage for their survival by cytokines, perhaps from the cancer cells [38]. In this regard, there could be a ‘symbiotic’ relationship between MSCs and the cancer cells. Recent studies have shown that the interplay between mesenchymal and epithelial cells may be more dynamic than previously thought. Bidirectional conversion between mesenchymal and epithelial phenotypes may explain metastatic dissemination of epithelial cancers [39]. Current speculation is that breast cancer cells with mesenchymal morphology acquire stem-like characteristics after losing apical-basal polarity [40]. This phenomenon, known as epithelial-to-mesenchymal transition (EMT), is remarkable because, during cancer progression, EMT recapitulates critical events that occur during embryogenesis and wound healing [39]. These types of characteristics have been demonstrated in Wnt-dependent colorectal carcinomas [39].

The predilection for MSCs to enter tumor site has not gone unnoticed. The thought is that MSCs could be a cellular vehicle for the delivery of genes to the sites of tumors. In fact, studies have been done to determine if MSCs can deliver tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) to tumors. MSCs transduced with TRAIL were able to decrease tumor burden in breast, squamous, and cervical cancer in subcutaneous xenograft experiments [41]. In other studies, engineered MSCs show similar homing with anti-tumor effects on both malignant and invasive primary gliomas [42]. Interestingly, the invading MSCs remained undifferentiated without evidence of proliferation[42].

The immunosuppressive properties of MSCs have led to speculation that MSCs can also offer immune-mediated protection of BCCs in the BM, and in other organs. Although the evidence has not gained a strong foothold at this point, the theory is likely to be supported by new discoveries. Ongoing studies are being conducted to determine the nature of the interactions between cytotoxic T cells, natural killer cells, MSCs, and BCCs.

Cytokines and Cancer Metastasis

Contact-dependent and -independent effects of MSCs have thus far been shown to contribute to the immunomodulatory properties of MSCs. Similar mechanisms might be involved in the immune biology of breast cancer metastasis not only in the BM, but in other organs. Contact between MSCs and BCCs appears to be likely mechanism for the migration of BCCs in BM, and perhaps toward the endosteal regions of the cavity [3]. However, a discussion of the interactions between BCCs and MSCs is incomplete without considering contact-independent effects, such as the roles of soluble factors from both cell types on the behavior of the cancer cells [3]. Cytokines secreted by immune and other cells can facilitate cellular communications and might also facilitate BCC dormancy in BM [1]. Studies have partly identified cytokines that facilitate communication between MSCs and BCCs [2,3]. The chemokine CXCL12, also known as stromal cell-derived factor 1 (SDF-1), is critical for the interaction between BCCs and MSCs [2,3]. This coupling is important and might reveal critical roles of breast cancer metastasis and/or dormancy.

CXCL12 is produced by MSCs [2]. It exists as membrane-bound in MSCs and BCCs where it is able to interact with its receptor (CXCR4) on both cells types, thereby leading to double interactions [2,3]. Experimental studies with endothelial cells from BM aspirates showed that CXCL12-CXCR4 coupling is relevant to the transmigration of BCCs across endothelial cells [2,43]. Once in the BM, the cancer cells are likely to interact with MSCs in a 3-D configuration (Figure 1). This CXCR4-CXCL12 coupling is consistent with an increase in CXCR4 expression in highly metastatic BCC lines [1].

Figure 1.

Shown is a central role for mesenchymal stem cells (MSCs) and microRNAs (miRNAs) in the promotion of breast cancer as well as during dormancy. The cancer cells are shown entering bone marrow cavity. This process is facilitated by cytokines following by interactions between breast cancer cells and MSCs. The interactions can be accounted for by CXCL12 and CXCR4 interactions, both expressed on MSCs and the cancer cells. MicroRNA exchange is proposed to occur through gap junction between the stroma and breast cancer cells. The proposed hypothesis is that dormancy is attained by mechanisms involving miRNAs. The role of MSCs and miRNAs is not limited to the bone marrow since similar mechanisms could occur at sites, distant from the primary tumor.

To appreciate roles for cytokines in the migration and adherence of BCCs, it is important to examine the role of the TAC1 gene, whose peptides induce the production of multiple cytokines [3]. Peptides derived from the TAC1 gene regulate the production of several cytokines, which appear to regulate cancer metastasis [3]. Expression of the TAC1 gene correlates with the induction of CXCR4 [3]. TAC1 knockout BCCs caused a decrease in the expression of CXCR4 [3]. This decrease correlated with reduction in metastasis to BM [3]. Ectopic expression of CXCR4 in the TAC1 knockout cells reversed the migration of BCCs to BM in an in vivo model [3]. Substance P, the major peptide derived from the TAC1 gene, can induce the expression of its receptor, neurokinin-1 (NK1), including the truncated NK1 [44]. Truncated NK1 causes non-tumorigenic breast cells to be transformed [44]. The transformation mediated by truncated NK1 seems to be partly explained by the induction of growth-promoting cytokines and substance P [44].

Although the above focuses on defined mechanism by which BCCs and MSCs interact, the mechanism involves networks of interconnected pathways. This adds to the complexity in metastasis of cancer cells to BM, which is a major organ of MSCs [2–4]. BCCs express vascular endothelial growth factor (VEGF) and fibroblast growth factor 2 (FGF-2) [2]. Since MSCs express receptors for these factors, it is likely that they will respond to VEGF, FGF-2 and perhaps other angiogenic factors to establish a supporting microenvironment for BCCs [2]. Wu et al. reported on the expression of high levels of VEGF and angiopoietin-1 in MSCs, thereby supporting an angiogenic function of MSCs [45].

Other members of the chemokine family are also important in breast cancer metastasis. CCL5 and CCL5R expression is higher in breast cancer tissue than in cells of normal surrounding tissue. By acting on the chemokine receptor CCR5 on BCCs, CCL5 enhances metastasis through increased motility and invasion [46]. The expression of CCL2 and CCL5 on MSCs, as well as another significant cell subset in breast cancer metastasis, osteoblasts, support an potential role for MSCs in bone metastasis [47,48].

The production of IL-6 by MSCs could interact with its cognate receptor in BCCs to activate STAT3 [49]. This has been demonstrated in the estrogen-receptor (ER)-positive breast cancers, although similar responses might occur in other cancer cells [49]. Osteoprotegerin, a member of the TNF family has been shown to protect from apoptosis via neutralization of TRAIL [50,51].

Although MSCs are likely to confer immune-protective effects on BCCs through the production of cytokines, they can be engineered as vehicles of anti-cancer therapy. MSCs engineered to express IL-12 resulted in decreased breast tumor burden and angiogenesis [52]. Although genetically engineered MSCs express anti-tumor cytokines as a method for cancer therapy, it should be noted that the MSCs themselves can also support cancer growth. As research progresses, it would be important to study how gene delivery of MSCs could occur without invoking the tumor-supporting role of MSCs.

MicroRNAs and Breast Cancer

MicroRNAs (miRNAs) are highly conserved class of 19–23 nucleotides, small, non-coding RNA molecules that are able to regulate gene expression at the post-transcriptional level by binding to the 3′UTR of target mRNAs and resulting in mRNA cleavage or inhibition of protein synthesis [53–55]. There are increasing reports showing roles for miRNAs in the biology of cancer. miRNA expressions and functions are dysregulated in human tumors with evidence of their role as oncogenes [56,57]. One of the most well studied ‘oncomirs,’ miR-21, is up-regulated in breast cancer when compared with normal breast tissue [58–60]. Both in vitro and in vivo studies have demonstrated dose-dependent growth suppression of low-invasive breast cancers following decreased expression of miR-21 [60,61]. Not surprisingly, there is an increase in apoptosis when miR-21 was inhibited; further supporting that miR-21 is a tumor promoting molecule [60]. Consistent with these findings are decreased expression of Bcl-2 when miR-21 expression was suppressed in BCCs. Interestingly the decrease in Bcl-2 was more pronounced in vivo than in vitro, suggesting that the microenvironment could have a role in the biology of miR-21 [60]. Other studies have shown that genes involved in the p53 pathway are affected by miR-21 [61]. Knockdown of p53 showed no difference in proliferation between LNA-miR-21 line and cells with a functional miR-21, suggesting that inhibition p53 may have a role in enhanced proliferation via an increase in miR-21 [61]. Despite these findings, the mechanism by which miR-21 supports tumor growth is unclear. Investigation into possible targets of miR-21 may lead to a greater understanding of the mechanism by which this particular miRNA and also others could be involved in protecting cancer cells from undergoing apoptosis.

Consistent with its oncogenic functions, miR-21 affects the expression of tumor suppressor genes, PTEN, TPM 1, and PDCD 4 [7,58,61]. A decrease in PTEN in breast cancer is associated with lymph node metastasis, estrogen expression, tumor grade, stage, and microvessel density [62,63]. Expression of miR-21 was significantly different between reduced PTEN expressing and high PTEN expressing subgroups of breast cancer [58]. miR-21 can inhibit the translation of PDCD4 [61].

Although miR-21 has received considerable attention in recent years, the role of miR-205 has been elucidated in the context of oncogenesis. miR-205 expression is significantly decreased in BCCs when compared with matched normal breast tissue. Ectopic expression of miR-205 significantly reduced cell proliferation, anchorage-independent growth and cell invasion. The targets of miR-205 were found to be erbB3 and VEGF-A. These results are indicative of miR-205 functioning as a tumor suppressor gene. [57,64] The HER/EGFR family is widely known to be involved in breast tumorigenesis when their signaling function is dysregulated. HER2 overexpression is associated with aggressive tumors. In the absence of ligand-binding, HER2 activity depends on the interaction with other members of the HER family, mainly HER3 phosphorylation. miR-205 can regulate HER3 expression [65]. There is an inverse correlation of miR-205 and HER3 in breast tumors when compared with normal breast tissue. A direct interaction between miR-205 and HER3 3′UTR is the likely mechanism for the decreased HER3 [64]. The primary oncogenic signaling by HER2-HER3, mediated through HER3, is through the activation of the PI3K/Akt survival pathway. An increase in miR-205 leads to a decrease in phosphorylated Akt [64]. This suggests that HER3 inhibition could increase the effectiveness of tyrosine kinase inhibitor therapy [60]. In fact, the proportion of apoptotic cells by treatment with gefitinib or lapatinib were increased in cells transfected with miR-205 [64].

Investigations were done to link miRNA and cancer drug resistance. Studies with doxorubicin resistant cells showed >100 differentially expressed miRNAs, some upregulated and others, downregulated [66]. These findings support the theory that the multifactorial mechanisms of drug-resistance involve several miRNAs with complex functions.

An example of the link between miRNA and cancer drug resistance is demonstrated by studies in which down-regulation of miR-27b and up-regulation of cytochrome P450 CYP1B1 resulted in increased metabolism of doxorubicin [66]. The increase in miR-28 lead to the loss of BRCA1 expression and subsequent increase in the resistance to doxorubicin [66]. The epigenetic abnormality leading to resistance appeared to be linked to increases in the expressions of miR-22, miR-29a, miR-194, and miR-132, which target DNA methyltransferases 3A and 3B and methyl CpG binding protein 2 [66,67]. The level of miR-451 is inversely correlated with the multiple drug related gene (MDR1). Therefore, it is understandable that an increase in miR-451 expression would lead to an increase in the sensitivity to doxycyclin treatment [66].

Steroid receptor coactivator 3 (SRC-3) is downregulated by miR-17-5p and therefore plays a role in the regulation of breast tissue growth [68]. An increase in miR-17-5p correlated with a decrease in estrogen stimulated growth of breast tissue. Conversely, a decrease in miR-17-5p correlated with an increase in estrogen stimulated growth. This suggested that a crosstalk exists between an estrogen receptor co-activator and miRNA for a more effective therapy in the estrogen-expressing BCC subset. Further investigation is needed into interactions of other miRNAs with estrogen receptor co-activators [69].

In addition to these miRNAs, miR-206 has been implicated in breast cancer, especially in the context of hormonal regulation. miR-206 overexpression resulted in a decrease in estrogen-mediated responses in BCCs. This phenomenon occurs even in the absence of the 3′ UTR sequence in the endogenous estrogen receptor-αmRNA, suggesting that the mechanism of action by miR-206 could be at the level of co-activators. EGF treatments increase the level of miR-206 suggesting that the repression of estrogen response caused by EGFR signaling may occur through miR-206 [70].

The ability of miRNAs to target many genes may signify that they could be involved in conferring self-renewal of cancer stem cells. The ability of stem cells to bypass the G1/S checkpoint is in part accomplished via miRNAs [71]. Therefore, miRNA may play an important role in the population of BCCs that could be breast cancer stem cells. Investigation into this role holds a strong potential for the designs of future therapies in the treatment of breast cancer.

Networks Linking Bone Marrow and Breast Cancer

Breast and other cancers enter the bone marrow and/or surrounding region as the preferred organ of metastasis and/or invasion [3,72,73]. Metastasis of BCCs to BM correlates with poor prognosis [74]. Although markers to detect breast cancer cells in the BM during the early stages of diagnosis have been identified, further work must be done towards prevention [75]. The exchange of miRNA through GJIC between the bone marrow stroma and BCCs may be an important factor that contributes to the dormancy of the cancer cells to impart a quiescent phenotype in BM [2]. The GJIC identified between BM stroma and BCCs in the endosteal region is consistent with the failure of high-dose chemotherapy in autologous hematopoietic stem cell transplantation [2]. Breast cancer can resurge from BM after ten or more years of remission [76]. At this time it is unclear what causes the dormant cancer cells in BM to resurge. However, since inflammation has been strongly linked to breast cancer recurrence, inflammation-mediated mechanisms in BM could account for relapse with cancer in BM [77].

The TAC1 gene partly mediates BCC entry to BM [3]. In vitro studies suggest that the entry of BCCs in BM could be facilitated by interactions with MSCs through CXCL12-CXCR4 interactions [3]. One of the current methods used to determine the stage of breast cancer is the tumor-nodes-metastasis (TNM) system, but many women have poor prognosis due to the system’s inability to detect the early micrometastasis in the BM and peripheral blood [78,79]. Certain markers have been found that may identify a subset of patients more likely to develop metastasis [72]. One study found increased bone marrow stromal protein 2 (BST2) expression as well as increased serum BST2 when comparing bone metastatic breast cancer to primary human breast cancer, signifying that this may be a potential marker for bone metastasis [72]. The level of urokinase plasminogen activator receptor (u-PAR) in the peripheral blood correlates to increased risk of breast cancer metastasis and poorer survival [80]. In another study it was concluded that u-PAR might be an essential molecule in bone marrow disseminated tumor cells for long-term survival during dormancy, and/or reactivation [81]. In the developed world, breast cancer is the most common cancer in women and the bone marrow is perhaps the most common site of distant metastasis [37]. Bone-specific therapies have become available to patients with breast cancer restricted to the skeleton [37].

Conclusion

Understanding the molecular and genetic basis of the subset of BCCs that are able to evade chemotherapy will be invaluable to the future treatments for breast cancer. Current therapy, such as treatment with anti-inflammatory medications like COX-2 inhibitors, is targeted at decreasing breast cancer metastasis into the bone marrow [82]. However, as further evidence shows the presence of a dormant population of BCCs within the BM, it is becoming increasingly important to understand the mechanisms behind the dormancy and quiescence of these cancer cells. This will hopefully lead to more specific therapy aimed at stopping resurgence by preventing those cancer cells from hiding within the bone marrow.