Flipping the Switch: Integrin Switching Provides Metastatic Competence

Abstract

Integrin switching plays a critical role in the progression to metastatic disease, but the mechanism by which it contributes remains poorly understood. In the 11 February 2014 issue of Science Signaling, Truong et al. identified a transforming growth factor–β–mediated, prometastatic switch that is activated by β1 integrin inhibition in triple-negative breast cancers (TNBCs). Their work provides insight into the complex signaling changes that arise from integrin switching. Further characterization of β-integrin switching will require elucidation of the distribution of specific α-β integrin heterodimers and the role of ligand binding. Identifying the nature of the molecular interactions and the influence of a specific oncogenic context, including the status of driver mutations such as those in Myc and p53, will define the next phase in integrin cancer biology.

Malignant progression of cancer is supported largely by cues received through altered cell-cell and cell-matrix interactions that promote the metastatic potential of malignant cells. Integrins, a class of cell-surface adhesive receptors that exist as obligate α-β heterodimers, mediate many of these interactions (1). Oncogenic transformation can directly or indirectly increase the abundance of certain integrin subunits and decrease that of other integrin subunits, thereby changing not only the absolute number of available integrins but also their α-β composition. Consequently, the integrin repertoire is changed or “switched” to support cancer initiation and progression. In many instances, integrins that secure adhesion to the basement membrane are lost, whereas the abundance of integrins that stimulate cell survival, migration, proliferation, and invasion are increased. Although the specific nature of integrin switching varies among cancers, α2β1 and α3β1 integrins suppress cancer progression and metastasis, whereas α5β1 and β3 and β4 integrins can promote metastatic progression (1, 2). Integrin switching is central to epithelialmesenchymal transition (EMT), a process in which polarized epithelial cells transition to a migratory mesenchymal-like state during malignant progression. However, the mechanism by which integrin switching contributes to metastatic progression remains poorly understood.

The β1 integrin subunit pairs with at least 11 α subunits to produce heterodimers that mediate adhesion to a wide variety of ligands, including matrix molecules (such as collagen, laminin, and fibronectin, among others) and cell surface proteins, such as E-cadherin and vascular cell adhesion molecules (1). Studies using in vitro and in vivo orthotopic or xenograft models indicate that the β1 integrins mediate drug resistance and stimulate metastasis in gastric, ovarian, and lung cancers (1). Similarly, animals with targeted deletion of the β1-encoding gene in mammary epithelium fail to develop cancer, suggesting that the β1 integrins are essential for cancer initiation (3). However, the loss of specific β1 integrin species may be important in progression to metastasis. Transgenic deletion of α2β1 integrin increases metastasis in the spontaneous MMTV-Neu tumor model, whereas the loss of α3β1 promotes metastasis in models of prostate cancer (2, 4). Decreased abundance of these β1 species in human patients affirms their clinical relevance in disease progression. Consequently, understanding how the loss of β1 integrin functions contributes to metastatic progression has both mechanistic and therapeutic implications.

In the 11 February 2014 issue of Science Signaling, Truong et al. reported that functional inhibition or knockdown of the β1 integrin enhanced metastatic lung colonization by promoting EMT through activation of the transforming growth factor–β (TGF-β) signaling network (Fig. 1A). This converted the migratory behavior of human and mouse triple-negative breast cancer cells (TNBCs) from collective to single-cell movement and enhanced metastatic lung colonization by decreasing the expression of the gene encoding E-cadherin through a pathway involving microRNA-200 (miR-200) and the transcription factor ZEB2 (5).

Fig. 1. Taking sides: integrin switching controls signaling output from the TGF-β network.

(A) β1 integrins provide proliferative and cell survival signals that are critical for tumor growth. Early in tumor progression, β1 integrins modulate TGF-β signaling to suppress its capacity for metastatic signaling. (B) In TNBCs, loss of β1 integrins and gain of β3 integrins enables a switch in TGF-β signaling, to inhibit the microRNA miR-200, and enable ZEB2 to suppress the expression of CDH1 that encodes E-cadherin. The integrin–TGF-β switch ultimately leads to EMT, individual cell migration, and metastasis.

These intriguing results are reminiscent of findings from Parvani et al. who observed a compensatory increase in β3 integrin expression upon β1 integrin inhibition in breast cancer (6) (Fig. 1B). However, whereas Parvani et al. suggested that β3 integrin overexpression alone was sufficient to mediate metastatic competence, Truong et al. found α3 integrin overexpression to be insufficient for TGF-α–induced metastasis in their model. Interestingly, Truong et al. observed that inhibition or knockdown of α1 integrin governed a trade-off between tumor cell proliferation and cell motility and dissemination, leading to reduced growth at both the primary and secondary locations while promoting systemic dissemination. Those findings reaffirm the well-established requirement of proliferative signals from α1 integrin and highlight the integration of proliferation and migration. The correlation to a α1/α3 switch to this behavior was also observed by Liu et al., who found that Myc overexpression supported tumor proliferation but not invasion and motility because of repressed expression of the gene encoding α3 integrin (7). Together, these findings indicate that both loss of α1 integrin and gain of α3 integrin are required to provide metastatic competence.

Observations by Truong et al. suggest that integrins are potent therapeutic targets in our attempts to counter the proliferative and invasive signals. However, it will be important to consider the balance of signal integration through α1 and α3 because simply inhibiting either alone could inadvertently drive dissemination or tumor growth.

These findings also offer a new perspective on the influence of integrin switching in EMT and tumor metastasis. TGF-α famously plays a dichotomous role in cancer, acting as a tumor suppressor in early carcinomas and a promoter of EMT and metastasis in more advanced cancers. Perhaps integrin switching from α1 to α3 integrin directs the biological response to activation of the TGF-α signaling network. Although conclusive evidence is not yet available, matrix interactions are clearly relevant as Leight et al. have recently demonstrated that matrix rigidity regulates the phosphoinositide 3 kinase/AKT-dependent switch between TGF-α–induced apoptosis and TGF-α–induced EMT (8). Additionally, direct interaction between TGF-α and both α1 and α3 integrins have been documented, indicating the potential for differential complex-specific signaling (6).

Truong et al. observed that α1 integrin inhibition potentiated TGF-α–induced EMT only in TNBC. As TNBCs present clinically with a high incidence of lymph node metastasis even among small tumors, the trade-off between proliferation and metastasis may explain an important aspect of TNBC biology. New TNBC subclassification schema proposed by Lehman et al. may help to further identify the specific oncogenic contexts where Truong’s integrin-TGF-α prometastatic switch may be relevant (9). Although there are several distinct properties that define the aggressive nature of TNBC, it is known to have particularly high p53 mutation status. Mutant p53 promotes both random cell motility and increased invasiveness by accelerating the recycling of α1 integrins to the plasma membrane, where the integrin complex may serve as a metastatic engine (10, 11). Upon phosphorylation, wildtype p53 binds Smads to support cytostatic TGF-α signaling. In contrast, complexes formed between mutant-p53 and Smads inhibit p63, a p53 family member that inhibits TGF-α–induced metastasis (12). Integration of the integrin–TGF-α prometastatic switch with p53 status may explain the divergent cellular response to TGF-α and thus warrants closer scrutiny.

In summary, the work of Truong et al. provides a mechanism by which loss of α1 integrin elicits a prometastatic switch in later stages of disease progression. Further characterization of α-integrin switching will require elucidation of the specific β-α integrin heterodimer distribution and the role of ligand binding. Identifying the nature of the molecular interactions and the influence of a specific oncogenic context, including the status of driver mutations, such as those in Myc and p53, will define the next phase in integrin cancer biology.