(E-186) AGEs in the Diabetic Extracellular Matrix Drive Breast Cancer Cell Proliferation Ex Vivo

Introduction:: Diabetes has been implicated in promoting an aggressive breast cancer phenotype. Epidemiological evidence shows that breast cancer patients with diabetes mellitus tend to have poorer prognosis and higher risk of cancer-related death. One proposed mechanism to link these disease states is through the production of advanced glycation endproducts (AGEs) via hyperglycemia-induced crosslinks; we have previously shown that crosslinking of matrix proteins results in a stiffer tumor stroma and increased primary tumor growth. Separate work has shown that stiffening of the tumor stroma promotes tumor progression and cell proliferation. As such, we hypothesize that diabetes contributes to cancer progression through AGE production in the tumor matrix independent of glucose overabundance, insulin signaling, or secreted inflammatory cytokine effects via RAGE activation. To further study this, a model of recellularization to glycated tumor tissue substrates was developed to parse apart the change in proliferation of cancer cells seeded ex vivo due to glucose pre-conditioning of the extracellular matrix (ECM) separate from the effects of AGE-RAGE interactions and glucose uptake by cells.

Materials and Methods:: PyMT mice were injected with streptozotocin (STZ) to induce hyperglycemia and the crosslink-breaker ALT711 to break AGE crosslinks. A) PyMT (Ctrl, STZ, STZ+ALT711) tumors were sectioned, and AGEs were visualized with immunofluorescent staining. B) Samples of Ctrl, STZ, and STZ+ALT711 tumors were hydrolyzed via acid digestion, dried in vacuo, and filtered. Normalized autofluorescence (370/440 nm) of decellularized tumors were measured to detect AGEs. C) PyMT (shRAGE or shCTRL) cells were seeded into 3D collagen matrices following incubation with 100mM glucose to glycate the matrix. Cells were stained for RAGE and the proliferation marker Ki67. D) Quantification was performed for cells positive for Ki67/DAPI on glycated and unglycated substrates. E) PyMT tumors were decellularized using deoxycholate, glycated using 100 mM glucose or 0mM control, and recellularized with MDAMB231 cells. Confocal reflectance imaging was performed to visualize the tumor ECM. Cells were stained for Ki67, AGEs, and DAPI. F) Autofluorescence readings measuring AGEs in ex vivo glycated samples were performed and quantified (370/440 nm). G) Quantification was performed for cells positive for Ki67/DAPI on glycated and unglycated substrates.

Results, Conclusions, and Discussions:: Results PyMT tumors in diabetic mice contain greater levels of AGE crosslinking and treatment with ALT711 decreases AGE abundance (Fig 1A). AGEs in decellularized tumor matrices displayed similar trends when autofluorescence was measured (Fig 1B). Cell proliferation was greater on collagen matrices glycated with glucose, even when RAGE expression is silenced (Fig 1C,D). In a novel ex vivo model of the diabetic tumor ECM, tumor tissues may be decellularized and glycated ex vivo to tune AGE crosslinking (Fig 1E). By incubating these tissues with 0 and 100mM glucose, AGE levels were comparable to those seen in CTRL and STZ tumors, respectively (Fig 1F). Once the excess glucose is removed, MDAMB231 cells were seeded to these substrates, to expose cells to the abundant AGE crosslinks present in the substrate. When Ki67+ status of cells seeded to these substrates was quantified, we found that increasing glycation ex vivo is sufficient to increase proliferation as was seen vitro and has been observed in vivo (Fig 1G).

Conclusion We have developed a mouse model of diabetic breast cancer, as well as a complimentary model of ex vivo glycation to parse apart the effect that hyperglycemic conditioning exerts on the ECM of breast cancer cells. Incubation of matrix samples with glucose ex vivo promotes a similar increase in glycation, which can be assayed by measuring the natural autofluorescence of these AGE species. Seeding to glycated matrices is sufficient to induce proliferation in cancer cells, indicating that matrix modifications caused by hyperglycemia play a key role in worsened cancer progression, independent of other factors which have been studied such as glucose, insulin, or cytokine signaling.

Discussion Herewe show that diabetes results in glycation in the tumor microenvironment to drive tumor cell proliferation. AGEs are relevant in many disease states, including studies of diabetes, ageing, obesity, and wound healing, though previous literature has been focused on investigating the receptor for AGEs, RAGE, and the cell-signaling aspect of circulating AGEs. Use of models for ex vivo glycation in the ECM will be important to establish a mechanistic roll for aberrant crosslinking in disease progression.


Acknowledgements (Optional): : The work described was supported by funding from the Vanderbilt-Ingram Cancer Center and from NIH grant T32 DK101003 for Integrated Training in Engineering and Diabetes.

References (Optional): :