Date Published: July 10, 2017
Publisher: Public Library of Science
Author(s): Brittany N. Balhouse, Logan Patterson, Eva M. Schmelz, Daniel J. Slade, Scott S. Verbridge, Pankaj K. Singh.
It is well documented that the tumor microenvironment profoundly impacts the etiology and progression of breast cancer, yet the contribution of the resident microbiome within breast tissue remains poorly understood. Tumor microenvironmental conditions, such as hypoxia and dense tumor stroma, predispose progressive phenotypes and therapy resistance, however the role of bacteria in this interplay remains uncharacterized. We hypothesized that the effect of individual bacterial secreted molecules on breast cancer viability and proliferation would be modulated by these tumor-relevant stressors differentially for cells at varying stages of progression. To test this, we incubated human breast adenocarcinoma cells (MDA-MB-231, MCF-DCIS.com) and non-malignant breast epithelial cells (MCF-10A) with N-(3-oxododecanoyl)-L-homoserine lactone (OdDHL), a quorum-sensing molecule from Pseudomonas aeruginosa that regulates bacterial stress responses. This molecule was selected because Pseudomonas was recently characterized as a significant fraction of the breast tissue microbiome and OdDHL is documented to impact mammalian cell viability. After OdDHL treatment, we demonstrated the greatest decrease in viability with the more malignant MDA-MB-231 cells and an intermediate MCF-DCIS.com (ductal carcinoma in situ) response. The responses were also culture condition (i.e. microenvironment) dependent. These results contrast the MCF-10A response, which demonstrated no change in viability in any culture condition. We further determined that the observed trends in breast cancer viability were due to modulation of proliferation for both cell types, as well as the induction of necrosis for MDA-MB-231 cells in all conditions. Our results provide evidence that bacterial quorum-sensing molecules interact with the host tissue environment to modulate breast cancer viability and proliferation, and that the effect of OdDHL is dependent on both cell type as well as microenvironment. Understanding the interactions between bacterial signaling molecules and the host tissue environment will allow for future studies that determine the contribution of bacteria to the onset, progression, and therapy response of breast cancer.
The tumor microenvironment is now a widely recognized and well-studied contributor to cancer dynamics, particularly for breast cancer. While increased matrix density, programming of cancer-associated stromal cells, evolving gradients of oxygen and nutrients, and leaky vasculature have all been implicated as key players in breast cancer progression [1–4], the impact of the recently identified breast tissue resident microbiotic niche has received little attention. Beyond the effects of pathogenic or tumorigenic bacteria such as Chlamydophila pneumonia, Salmonella typhi, Streptococcus gallolyticus , Helicobacter pylori  and Fusobacterium nucleatum , the majority of analyses of tumor-microbiome interactions have centered on local cell-cell interactions within the gut microenvironment, or more systemic immune effects influenced by gut microbiota . Only a handful of studies have been conducted to investigate the influences of tissue-resident bacteria in other tumor sites, such as for breast cancer [9–11]. Even fewer have investigated how small molecules released from resident bacteria may interact with cells in the presence of other critical microenvironmental factors, e.g. tumor hypoxia, to regulate cancer progression. In an effort to address these questions, we investigated interactions between the quorum-sensing molecule N-(3-oxododecanoyl)-L-homoserine lactone (OdDHL) and the breast tumor relevant microenvironmental cues of a stiff collagen-derived tissue mimic and hypoxia. This representative study will aid in our understanding of how the understudied breast tissue microbiome may contribute to disease phenotypes, patient-to-patient variability, and cancer progression.
We first compared highly malignant MDA-MB-231 cells with non-malignant MCF-10A cells. We found that the response to OdDHL was dependent not only on cell-type, but also the culture condition. As compared to the control, the malignant MDA-MB-231 cells showed significantly different responses to 400 μM of OdDHL in all culture conditions (Fig 1A–1E). In the 2D/normoxia condition, the 400 μM OdDHL treatment corresponded to approximately 52.2% (± 2.2%) viability, relative to the control. In the 2D/hypoxia condition, that relative viability was increased to 60.6% (± 2.2%) and in 3D/normoxia condition, that viability was increased to 81.9% (± 2.2%) (Fig 1C and 1D, respectively). There were significant decreases in the MDA-MB-231 viability at the 100 and 200 μM levels of OdDHL, as compared to the control, for the 2D conditions as well (Fig 1A and 1B). In contrast, the non-malignant MCF-10A cells exposed to OdDHL showed no significant change in viability relative to the control across all concentrations in all culture conditions (Fig 1F–1H).
The analysis of the viability of MDA-MB-231and MCF-10A cells in response to OdDHL confirmed our hypotheses that there would be reduced MDA-MB-231 response to OdDHL treatment in both hypoxia and in 3D while MCF-10A cells would show no such changes. Our 2D/normoxia results are consistent with published literature , while our other conditions provide further insight into the interaction of OdDHL with other tumor microenvironment factors. As we anticipated, malignant cells cultured in a 3D environment had the highest relative viability (Fig 1) with OdDHL treatment as compared to cells in 2D. Cells in hypoxia also had increased relative viability with OdDHL treatment as compared to those in normoxia (Fig 1). Thus, we found that OdDHL preferentially affects MDA-MB-231 cells as compared to MCF-10A cells and this effect is blunted by both hypoxia and 3D culture. This result parallels the common finding that tumor cells are more resistant to chemotherapies when studied in 3D and/or hypoxic conditions [27, 31–34], and suggests that these conditions also represent a more physiologically relevant context for the testing of the impact of microbial factors in the tumor microenvironment.
We showed that the selective effect of OdDHL on the viability of breast cancer cells is significantly mitigated by two key hallmarks of the tumor microenvironment: hypoxia and stiff ECM. It is important to note that, although there was a blunted response to OdDHL treatment for MDA-MB-231 cells in hypoxia and in 3D, there was still a significant decrease in proliferation and viability, and a significant increase in necrosis with treatment. The intermediary response of MCF-DCIS.com may demonstrate a differential role of microenvironmental bacterial factors over the course of breast cancer progression. Thus, further exploration of microbiome-tumor interactions and OdDHL as a potential therapy is merited.