Tumor progression and aggressiveness are associated with altered expression of glycans and glycoproteins in the pericellular microenvironment.  Hyaluronan (HA), a naturally occurring glycosaminoglycan, is an integral component of the extracellular matrix (ECM).  HA exerts its biological functions by binding to its specific cell surface receptor, CD44, triggering cell signaling pathways associated with the ability of malignant cells to migrate and invade basement membranes ⁄ ECM, thus leading to formation of metastasis.[3,4] The HA activity in cancer is dependent on its molecular weight (Mw); while this association has been demonstrated, it remains unclear which Mw range exerts the highest malignancy. The lack of coherence in the reported results is associated with the use of different models and different presentation of HA to the cells. Studies are typically done with soluble HA that is not relevant in the ECM context. In order to address this hurdle, we developed substrates using layer-by-layer (LbL) assembly of HA (polyanion) with different Mws (6.4, 752 and 1500 kDa) and Poly-L-Lysine (polycation). These substrates resemble ECM as they are soft, highly hydrated and provide spatial freedom to the HA to dynamically alter its conformation upon recognition by proteins and cells. In this work, we tested the developed substrates with MKN45 gastric cancer cell line (known for overexpressing CD44) and evaluated the impact of the HA Mw on cell behavior. The LbL assembly was confirmed by QCM-D: an increased film thickness was observed for HA with higher Mw. The presence of the HA in the outermost surface of the assemblies was confirmed by electrokinetic analysis, which showed negative zeta potential values for all the assemblies regardless the HA Mw. Substrate-cell interactions were also probed using QCM-D: while MKN45 cells adhered to all studied substrates, only in the case of HA of higher Mws (i.e. 752 and 1500 kDa) cells remodel the substrate surface. Scanning electron microscopy and immunostaining (Paxilin) was consistent with the QCM-D results showing more contact sites between MKN45 cells and the surface assembled with HA of 1500 kDa. Overall, our results show that the tested LbL assemblies are a valid ECM-mimicking model to study the impact of HA in tumor cell behavior.
Acknowledgements: The authors would like to thank to Fundação para a Ciência e Tecnologia (FCT), for the Ph.D. Grant SFRH/BD/112075/2015. This work was also supported by the European Commission: H2020-TWINN-2015-692333 – CHEM2NATURE project and H2020-WIDESPREAD-2014-2-668983 – FORECASTproject.
 M. J. Paszek, et. al., Nature 2014, 511, 319-325;  T. Chanmee, P. Ontong and N. Itano, Cancer Lett 2016, 375, 20-30;  R. L. Wu, L. Huang, H. C. Zhao and X. P. Geng, J Cancer Res Clin Oncol 2017, 143, 1-16.;  S. Misra, P. Heldin, et al., FEBS J 2011, 278, 1429-1443.