The effects of macroporosity and stiffness of poly[(methyl vinyl ether)-alt-(maleic acid)] cross-linked egg white simulations of an aged extracellular matrix on the proliferation of ovarian cancer cells

Abstract

Ageing and remodelling of the extracellular matrix (ECM) leads to enzyme activation, resulting in a fibrotic microenvironment and fluctuating ECM stiffness. In this study, prepared egg white (EW) cross-linked with poly[(methyl vinyl ether)-alt-(maleic acid)] (P(MVE-alt-MA)) with macroporous structures were used to simulate an aged extracellular matrix that affected the malignant behaviour of cancer cells at different stages. Increased macroporosity and stiffness properties clearly enhanced the proliferation of cancer cells. In other words, different levels of EW cross-linking had different effects on cell malignancy, thereby determining the ability and speed of cell migration in scaffolds. The study showed that porosity and stiffness changes in the matrix were possible mechanisms for cancer cell invasion and metastasis besides blood vessels and lymphatic invasion in human body. P(MVE-alt-MA)-cross-linked EW contained a major component of ECM and provided a useful model to evaluate the proliferation and metastasis of cancer cells in vitro. This has important significance when it comes to exploring the effects of the material microenvironment on tumour prevention and treatment.

---

1. Introduction

One of the main roles of the extracellular matrix (ECM) is to provide scaffolding for tissues and cells, serving as a three-dimensional (3D) structure for cell adhesion and movement and as a signal for morphogenesis and differentiation. Related literature provides compelling evidence that biological cells can sensitively detect not only biochemical stimuli, but also the mechanical properties of their microenvironment.^1–4 The physical and chemical properties of the ECM surrounding cells have an effect on cell activity.⁵ Biophysical factors mainly include the hardness, density and porosity of substrate material, as well as other mechanical forces acting on cells, such as fluid shear stress, tensile stress and so on.^6,7

The cell mechanical response is principally manifested via the induction of basal hardness changes in adhesion, which are adjusted through biological behaviours such as cell morphology, proliferation, differentiation, migration orientation changes and so on.^8–11 In other words, various cell behaviours during development and disease are correlated with dynamic changes in the stiffness of cellular microenvironments.^12–14

Ideally, the scaffold should mimic the characteristics of the tissue's ECM with similar structure and physical properties.¹⁶ The scaffold must be able to support cell attachment and proliferation, as well as differentiation of the specific cell phenotype.¹⁵ Additionally, the porosity of tissue engineering scaffolds should remain above 90%, as a larger specific surface area is favourable for cell adhesion. Meanwhile, the interconnected pore structures in a scaffold should help cells to form a distribution network inside the material to simultaneously disburse cytokines and nutrients throughout the entire scaffold.¹⁷ Maintaining the viability and functioning of ovarian cancer cells in vitro remains challenging; as such, the construction of various scaffolding materials and the establishment of cancer cell culture systems are essential for cancer research.

As is known, the mechanical properties of the ECM vary widely across various tissues, individuals and age groups. Histologically, solid tumours also demonstrate different degrees of brittle lesions at different stages. In contrast, benign tumours tend to show good flexibility and elasticity. Designing different culture scaffolds to simulate the ECM, which influences the behaviour of tumour cells, is an important aspect of preventing the proliferation of tumour cells and of effectively treating tumours. Natural materials with strong compatibility, such as collagen and chitosan, were commonly used as tissue engineering scaffolds. The main disadvantage of natural materials was rapid degradation and the loss of mechanical stability. However, the structure can be improved and tuned to a particular purpose through chemical cross-linking and/or blending with other components.^18–20

Egg white (EW) contains all the nutrition and signal components required for cell growth and its mechanical properties can be modulated across a significant range. Thus, the cross-linked EW can be prepared to study the effects of stiffness and the impact of various material morphologies present in the microenvironment on cell behaviours.^21–27

Chemical cross-linking is one of most important modification methods for EW use.^28,29 P(MVE-alt-MA) can be used to prepare tissue engineering scaffolds with macroporous structures due to its natural and non-toxic side effects (Fig. 1).^30–34

Fig. 1 Schematic demonstration: (A) chemical cross-linking reaction between EW and poly[(methyl vinyl ether)-alt-(maleic acid)] (P(MVE-alt-MA)). (B) The SKOV-3 cancer cells were cultured in cross-linked EW.

Ageing is the main contributing factor to a person's vulnerability to cancer.³⁵ The cross-linking theory of ageing suggests that macromolecules in the body, such as proteins and nucleic acids, interfere with the functioning of normal cells through the development of covalent cross-links. The cross-linking reaction exists in the nucleic DNA and the extracellular collagen fibres, which suggests that the cross-linking of polypeptide chains in collagen emerges and becomes more frequent in old age.^36,37 Accordingly, this study aimed to simulate the ageing of the ECM through the preparation of P(MVE-alt-MA) cross-linked EW of varying macroporosity and stiffness, and assessed its effects on the proliferation of cancer cells.³⁸

As a result, the cross-linked porous EW scaffolds provided a platform to maintain the viability and functioning of SKOV-3 cancer cells for the reconstruction of engineered cancer tissue. The cross-linked EW provided the required microenvironment of simulated cell ageing and cell behaviours further revealed the relationship between ageing and tumour progression.³⁹

2. Experimental method

2.1 Materials

P(MVE-alt-MAH) (M_w = 1 080 000, M_n = 311 000, MDW = 3.47) was purchased from Sigma-Aldrich in Shanghai, China and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were purchased from Suzhou Highfine Biotech. Co., Ltd. Chicken eggs were purchased from the local supermarket.

2.2 Preparation of P(MVE-alt-MA) cross-linked EW

The general preparation procedure of P(MVE-alt-MA) cross-linked EW was as follows:

Twenty grams of P(MVE-alt-MAH) was added to 80 mL (namely 80 g) of ultrapure water. The mixture was heated to 90 °C and was stirred at a rate of 1000 rpm for two hours to obtain a 20% aqueous solution of P(MVE-alt-MA). Aqueous solutions of 2.5%, 5%, 10% and 15% P(MVE-alt-MA) were prepared by similar methods.

Sodium hydrocarbonate dry powder was first added to a given amount of P(MVE-alt-MA) solution to maintain a pH of 5.5–6.0 with a reaction time of 10–15 minutes. EDC was then added and left to react for 15–20 minutes. Following this, NHS was added and allowed to react for an additional 15–20 minutes. Finally, a given amount of EW was added, left to react at 37 °C for 24 hours, and stirred at a rate of 200 rpm. At the end of the reaction, P(MVE-alt-MA) cross-linked EW was obtained and pH was further regulated to 7.2 for cell cultures.

All prepared P(MVE-alt-MA) cross-linked EW samples were divided into two groups, Group A and Group B. Group A included S1-1, S2-1, S3-1, S4-1 and S5-1. In Group A, the total reaction volume was maintained at 3.0 mL. The ratio of P(MVE-alt-MA) to EW increased gradually in the samples, with S1-1 having the lowest ratio and S5-1 having the greatest.

Group B included S1-2, S2-2, S3-2, S4-2 and S5-2. In Group B, the ratio of P(MVE-alt-MA) to EW was kept at a constant level.

The prepared formulas of all samples are listed in Table 1.

Table 1 The preparation formulas of P(MVE-alt-MA) cross-linked EW

Group A	Formula	Group B	Formula
S0	2 mL EW	S0	2 mL EW
S1-1	1 mL 2.5% P(MVE-alt-MA)	S1-2	1 mL 2.5% P(MVE-alt-MA)
	2 mL EW		0.25 mL EW
S2-1	1 mL 5% P(MVE-alt-MA)	S2-2	1 mL 5% P(MVE-alt-MA)
	2 mL EW		0.5 mL EW
S3-1	1 mL 10% P(MVE-alt-MA)	S3-2	1 mL 10% P(MVE-alt-MA)
	2 mL EW		1 mL EW
S4-1	1 mL 15% P(MVE-alt-MA)	S4-2	1 mL 15% P(MVE-alt-MA)
	2 mL EW		1.5 mL EW
S5-1	1 mL 20% P(MVE-alt-MA)	S5-2	1 mL 20% P(MVE-alt-MA)
	2 mL EW		2 mL EW

---

In a 6-well plate, 1.0 mL P(MVE-alt-MA) cross-linked EW was added to each well, while 50 μL P(MVE-alt-MA) cross-linked EW was added to each well in the 96-well plates.

Native EW was kept as control samples S0-1 and S0-2.

2.3 Rheological characterisation

The elastic modulus G′ and viscous modulus G′′ of cross-linked EW samples were incrementally measured at 37 °C using an Anton Paar Rheometer (MCR302 Anton Paar; Austria) with a heating rate of 1 °C min⁻¹. A parallel plate geometry with a gap of 1 mm (at 10% compression to alleviate slippage) was employed. Strain sweeps were performed to ensure operation in the linear range and a frequency sweep (0.1–100 rad s⁻¹) at a constant strain of 1% was used to conduct modulus measurements.

2.4 Morphology observations

All cross-linked EW samples were quickly frozen at −80 °C before being further freeze-dried in a Freezone 6 freeze drier (Labconco Corporation; Kansas City, MO, USA) under vacuum conditions at −42 °C for at least 48 hours until the sublimation of the solvent was complete. For scanning electron microscopy (SEM) measurement, freeze-dried samples of the native EW and cross-linked EW freeze-dried powders were sectioned carefully and coated with copper; their morphology was studied using an Ultra Plus scanning electron microscope (Zeiss, Oberkochen, Germany).

2.5 ATR-FTIR spectra analysis

The attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectra were obtained using a Nicolet 5700 spectrometer (Thermo Scientific; Philadelphia, PA, United States) with a Wilks model 10 ATR accessory at an angle of 45° using a potassium bromide (KBR) crystal. The samples of freeze-dried powder were compressed into thin slices following the addition of a dry KBR mixture; the samples were then tested using the infrared testing platform. Spectra were recorded at a 4 cm⁻¹ resolution between 4000 and 500 cm⁻¹ and results were summed from 256 individual scans.

2.6 Cell cultures

The cross-linked EW samples were placed in cells of an ultra-clean workbench which had been illuminated by ultraviolet light for 30 minutes to ensure that the support material was aseptic. A previously prepared SKOV-3 ovarian cancer cell suspension (1 × 10⁴ cells per well) was added to 96-well plates. During this process, half of the culture liquid was renewed every three days. The cancer cell aggregates were observed after five days of culture and twenty diameter values of the aggregates in each cross-linked EW were averaged. Cells in the 96-well plates were stained by using Hochest33342 and calcein-AM. Cell-loaded scaffolds in each well were supplemented with 20 μL Hochest33342 30 minutes prior to observation. For the detection of SKOV-3 cancer cell viability, cell-loaded scaffolds in each well were incubated with 20 μL calcein-AM for one hour. All SKOV-3 cells were observed with an inverted fluorescence microscope and the excitation and emission wavelengths of Hochest33342 staining were maintained at 350 and 460 nm. The excitation and emission wavelengths of calcein-AM staining were kept at 490 and 515 nm.

2.7 CCK-8 assay

The cell proliferation and biocompatibility of scaffolds were analysed using the Cell Counting Kit-8 (CCK-8) set. Previously prepared macroporous scaffolds were added to wells of the 96-well plates (50 μL per well). Cells (1 × 10⁴ cells per well) seeded in P(MVE-alt-MA) cross-linked EW with the same porosity were cultured for five days. After each day of culture, a 15 μL CCK-8 solution was added to each well (n = 4). To facilitate the conversion of CCK-8 into formazan crystals through the mitochondrial dehydrogenases of living cells, each plate was incubated at 37 °C and 5% CO₂ for an additional two hours. The optical density was read in a Multimode Reader (Synergy HT, BioTek, Winooski, VT, USA) at 450 nm. The same procedure was performed for the cultured cells in culture plates without cells as control. The statistical data was treated using GraphPad Prism 5.0 software.

3. Results and discussion

In order to observe how changes in the microstructure of P(MVE-alt-MA) cross-linked EW would affect its biological properties, we focused on the following two aspects: the stiffness and the micro-scale macroporosity of the scaffolds. These two important factors of the microenvironment affected the behaviour of cells and tissue, eventually leading to regeneration and remodelling of the latter.

3.1 Rheological modulus of cross-linked EW samples

Elastic modulus G′ and viscous modulus G′′ of cross-linked EW samples S1-1, S2-1, S3-1, S4-1 and S5-1 of differing P(MVE-alt-MA) degrees were tested to understand the rheological characteristics of EW cross-linked with P(MVE-alt-MA) (Fig. 2). Elastic modulus G′ results were 3.77 Pa, 4.92 Pa, 5.58 Pa, 6.83 Pa and 29.64 Pa (S0-1 was 2.65 Pa for control) at 5 rad s⁻¹. Viscous modulus G′′ results showed 2.36 Pa, 1.66 Pa, 2.11 Pa, 1.74 Pa and 13.62 Pa (S0-1 was 1.32 Pa). In the same manner, S1-2, S2-2, S3-2, S4-2 and S5-2 were also tested. Elastic modulus G′ results were 5.57 Pa, 6.43 Pa, 12.76 Pa, 26.99 Pa and 63.28 Pa (S0-2 was 1.21 Pa for control) at 5 rad s⁻¹. Viscous modulus G′′ results showed 1.22 Pa, 1.61 Pa, 1.64 Pa, 6.38 Pa and 7.28 Pa (S0-2 was 0.43 Pa). The cross-linking of P(MVE-alt-MA) with EW samples increased the stiffness of materials and contributed to the formation of large pores. Increased stiffness and porosity, which in turn boost material support, can be utilised in senile patients with cancer and ECM variations, as shown in this simulation study. With an increase in cross-linking degrees, rheological property changes of the cross-linked EW would affect the behaviour of the cells. Changes in macroporosity and elasticity are potentially important characteristics of the ECM in cancer patients, as these conditions are likely conducive to the invasion and metastasis of cancer cells. Meanwhile, differences in stiffness between S0-1 and S0-2 implied the existence of individual differences in patients with similar cancer diagnoses.

Fig. 2 Elastic modulus G′ and viscous modulus G′′ of cross-linked EW samples (A-1) S0-1 as control, (A-2) S1-1, (A-3) S2-1, (A-4) S3-1, (A-5) S4-1, (A-6) S5-1, (B-1) S0-2 as control, (B-2) S1-2, (B-3) S2-2, (B-4) S3-2, (B-5) S4-2, (B-6) S5-2 with different degree of P(MVE-alt-MA) cross-linking were tested.

3.2 SEM morphologies of cross-linked EW samples

The freeze-dried EW after cross-linking reaction was observed by SEM (Fig. 3). Results showed complete morphological change after being cross-linked with P(MVE-alt-MA): (A-2) S1-1, (A-3) S2-1, (A-4) S3-1, (A-5) S4-1 and (A-6) S5-1. With the increase in cross-linking degree, the porosity also increased gradually from 40–70 μm, 60–90 μm, 100–120 μm, 110–140 μm, to 120–160 μm. It was sufficient to allow for cancer cell growth in the deep locations of these pores on the basis of average diameter of cells was between 10 and 20 μm in human body. Increased amounts of both components (EW and the polymer) at identical ratios of P(MVE-alt-MA) to EW resulted in a stiffer gel of the same porosity. The average porosity of (B-2) S1-2, (B-3) S2-2, (B-4) S3-2, (B-5) S4-2 and (B-6) S5-2 was roughly 115–160 μm. This study aimed to further differentiate the effects of porosity and/or stiffness on cell viability.

Fig. 3 Scanning electron microscopy (SEM) images of freeze-dried EW after cross-linking. The figure shows complete morphological change after cross-linking of P(MVE-alt-MA) with EW included native EW S0-1 and S0-2 as control. (A-1) S0-1, (A-2) S1-1, (A-3) S-12, (A-4) S3-1, (A-5) S4-1, (A-6) S5-1, (B-1) S0-2, (B-2) S1-2, (B-3) S2-2, (B-4) S3-2, (B-5) S4-2, (B-6) S5-2.

3.3 ATR-FTIR characterisation of cross-linked EW samples

Fig. 4 illustrates the ATR-FTIR transmission spectrum of native and cross-linked EW (S0-1, S1-1, S2-1, S3-1, S4-1, S5-1). With increases in the amount of added P(MVE-alt-MA), the intensity of peaks at 1690–1630 cm⁻¹ rose, which was attributed to the C=O of amide groups produced from the reaction of carboxylic acid in P(MVE-alt-MA) and amine in EW.

Fig. 4 The attenuated total reflectance Fourier transform infrared (ATR-FTIR) transmission spectrum of P(MVE-alt-MA) cross-linked EW samples (S1-1, S2-1, S3-1, S4-1, S5-1), S0-1 as control.

3.4 Distribution of SKOV-3 cell aggregates at different depth levels in cross-linked EW samples

Growth of the SKOV-3 ovarian cancer cells in P(MVE-alt-MA) cross-linked EW scaffolds was observed with an inverted fluorescence microscope after five days. SKOV-3 cells were cultured alive and distributed at different depth levels in P(MVE-alt-MA) cross-linked EW samples S1-1, S2-1, S3-1, S4-1, S5-1, S1-2, S2-2, S3-2, S4-2 and S5-2, including control groups S0-1 and S0-2 (Fig. 5). To achieve bright field imaging, nuclei were stained with Hochest33342 for blue (Fig. 6), and cytoplasms were stained with calcein-AM for green (Fig. 7). The green cytoplasm of SKOV-3 cells aggregates stained by calcein-AM was observed at different depth levels in the P(MVE-alt-MA) cross-linked EW samples S1-1, S2-1, S3-1, S4-1 and S5-1, including control group S0-1. In native egg whites, only some SKOV-3 ovarian cancer cells formed aggregates of approximately 50 μm. In cross-linked EW, the majority of SKOV-3 ovarian cancer cells formed aggregates with an average diameter of 140 μm at different depth levels of the scaffolds (Fig. 8).

Fig. 5 The SKOV-3 cancer cells grew in different P(MVE-alt-MA) cross-linked EW. (A-1) S0-1 as control, (A-2) S1-1, (A-3) S2-1, (A-4) S3-1, (A-5) S4-1, (A-6) S5-1, (B-1) S0-2 as control, (B-2) S1-2, (B-3) S2-2, (B-4) S3-2, (B-5) S4-2, (B-6) S5-2.

Fig. 6 The SKOV-3 cancer cells grew in different P(MVE-alt-MA) cross-linked EW. (A-1) S0-1 as control, (A-2) S1-1, (A-3) S2-1, (A-4) S3-1, (A-5) S4-1, (A-6) S5-1, (B-1) S0-2 as control, (B-2) S1-2, (B-3) S2-2, (B-4) S3-2, (B-5) S4-2, (B-6) S5-2 (Hochest33342 staining cell nucleus blue).

Fig. 7 The SKOV-3 cancer cells grew in different P(MVE-alt-MA) cross-linked EW. (A-1) S0-1 as control, (A-2) S1-1, (A-3) S2-1, (A-4) S3-1, (A-5) S4-1, (A-6) S5-1 (calcein-AM staining cytoplasm green).

Fig. 8 The aggregation of the ovarian cancer SKOV-3 cells changed with the stiffness and macroporosity increasing in cross-linked EW (S1-1, S2-1, S3-1, S4-1, S5-1) at the same time (S0-1 as control).

3.5 The effects of macroporosity and stiffness on the proliferation of cancer cells

It was noted that matrix connectivity and elasticity were coupled with P(MVE-alt-MA) cross-linked EW. Therefore, it was determined that increased cross-linking density led to higher G′ and porosity, which may have promoted nutrient exchange within the P(MVE-alt-MA) cross-linked EW scaffolds. While macroporosity and stiffness both contributed to the malignant proliferation and aggregation behaviours of ovarian cancer cells, it was believed that stiffness provided the main driving force. Enhanced proliferation malignancy caused by higher stiffness of P(MVE-alt-MA) cross-linked EW was observed (Fig. 9).^40,41 This result indicated that more malignant cancer cells should be found in the scaffold with higher cross-linking degrees.^42,43 On one hand, scaffold porosity mainly provided exchange and transformation space for nutrients, oxygen and signalling molecules to facilitate the growth and proliferation of cells. At the same time, porosity provided a discharge channel for metabolic waste. Alternatively, scaffold porosity provided space support for cell aggregation and distal metastasis. This culture system may have served as a useful in vitro platform to maintain the phenotype of SKOV-3 ovarian cancer cells and realise their proper function.

Fig. 9 The absorbance of ovarian cancer SKOV-3 cells changed with the increased stiffness G′ of cross-linked EW S1-2, S2-2, S3-2, S4-2, S5-2 with the same porosity at 450 nm (stained with CCK-8).

In the framework of larger porosity, cells were likely to grow in deep pores following the culture period such that additional cells could not be observed within the focal length of the materials surface. This indicated that after the onset of ageing, which was demonstrated by an increased cross-linking level, cell proliferation first appeared and was followed by cell growth at deeper levels with the emergence of many large pores.^44–46

The SKOV-3 cancer cell aggregates were formed in P(MVE-alt-MA) cross-linked EW scaffolds with varying degrees of macroporosity and stiffness.⁴⁷ In SKOV-3 cancer cell aggregates, tight cell–cell junctions were meant to sustain cell viability for extended culture periods and maintain high level tissue-specific functions. Previous work had revealed that aggregate size was crucial to SKOV-3 cancer cell functions and that oxygen diffusion would not be limited. It had also shown that central SKOV-3 cancer cells would not become hypoxic when the aggregate diameter was less than the order of 100 μm. Cell-substrate adhesion strength was a major factor governing the formation and morphology of multicellular aggregates. Ideally, cells should initially adhere to the substrate for retention, but adherence should not be so tight as to discourage their migration toward one another or prevent their eventual disassociation from the substrate in order to form multi-layered aggregates.⁴⁸

The ECM can not only control the speed and direction of cell migration but can also provide scaffolding to facilitate cell migration.⁴⁹ For example, laminin can promote migration in a variety of tumour cells.⁵⁰ Cell migration depends on cell adhesion and is subject to the cytoskeleton assembly. In contrast, cell adhesion is influenced by certain ECM-induced adhesion plaque formations, which are adhesive contact spots between the ECM and cytoskeletal rivets.⁵¹ Molecular and protease synergies engage in a dynamic process that constantly constructs the tumour matrix; this process may be closely related to tumour invasion and metastasis.⁵²

It was reported that cells could feel the stiffness of the substrate and respond accordingly, which could be applied to the treatment of cancer, where cell proliferation might vary depending on the stiffness of the matrix.⁵³ Matrix stiffness caused an increase in the proliferation rate of SKOV-3 cancer cells. This result was a positive contribution to the architecture of the cellular microenvironment surface, which controlled the fate of the cell wherein the cross-linked EW structure contained native ECM. Cancer cells in 3D porous scaffolds displayed high malignancies of proliferation due to cross-linking, fibrosis and hypoxia.⁵⁴ It was suggested that cross-linked porous EW scaffolds provided a platform to maintain the viability and functioning of SKOV-3 cancer cells for the reconstruction of engineered cancer tissue and enabled investigations of complex physiological cell–ECM interactions.^55–57

4. Conclusions

In summary, the prepared P(MVE-alt-MA) cross-linked EW with macroporous structures was used to simulate an aged extracellular matrix at different stages. Increased stiffness properties enhanced the malignancy of tumour cells in processes such as proliferation. The constant migration and proliferation of cancer cells was directed by the macroporous structure of the materials. Thus, the ability and speed of cell migration was also influenced by different degrees of cross-linked EW.

The study showed that porosity and stiffness changes in the matrix were possible mechanisms for cancer cell invasion and metastasis besides blood vessels and lymphatic invasion in human body. Compared with native EW, increases in the stiffness and macroporosity of P(MVE-alt-MA) cross-linked EW did indeed promote the proliferation of SKOV-3 cancer cells. P(MVE-alt-MA) cross-linked EW contained a major component of the ECM and provided a useful model to evaluate the proliferation and metastasis of SKOV-3 ovarian cancer cells as well as other cancer cells in vitro. This study holds significance for future exploration into the effects of the material microenvironment on tumour cells during the prevention and treatment of cancer.

Acknowledgements

This research is supported by the Special Project on Development of National Key Scientific Instruments and Equipment of China (2011YQ03013403), the National Basic Research Program of China (2011CB933503), the Science and Technology Project in Suzhou (ZXY201440), the National Natural Science Foundation of China (21503004), the Fundamental Research Funds for the Central Universities (KYLX_0191), the Large Instrument Analysis Test Fund Project of Southeast University, the Natural Science Key Project of Bengbu University (2015ZR03zd).

Notes and references

1. J. Jiang and E. T. Papoutsakis, Adv. Healthcare Mater., 2013, 21, 25–42.
2. R. A. Mathias, Y.-S. Chen, B. Wang, H. Ji, E. A. Kapp, R. L. Moritz, H.-J. Zhu and R. J. Simpson, J. Proteome Res., 2010, 9, 1007–1019.
3. A. Bergamaschi, E. Tagliabue, T. Sørlie, B. Naume, T. Triulzi, R. Orlandi, H. G. Russnes, J. M. Nesland, R. Tammi, P. Auvinen, V.-M. Kosma, S. Ménard and A.-L. Børresen-Dale, J. Pathol., 2008, 214, 357–367.
4. L. Dao, C. Gonnermann and C. M. Franz, J. Mol. Recognit., 2013, 26, 578–589.
5. R. Podlipec, S. Gorgieva, D. Jurašin, I. Urbančič, V. Kokol and J. Strancar, ACS Appl. Mater. Interfaces, 2014, 6, 15980–15990.
6. J. Haier and G. L. Nicolson, APMIS, 2001, 109, 241–262.
7. Q. Kong and M. Vazquez, J. Biomed. Mater. Res., Part A, 2009, 89, 969–979.
8. L. Dafik, M. Alarcao and K. Kumar, J. Med. Chem., 2010, 53, 4277–4284.
9. F. Ruffini, G. Graziani, L. Levati, L. Tentori, S. D'Atri and P. M. Lacal, Int. J. Cancer, 2015, 136, E545–E558.
10. C. Pelillo, A. Bergamo, H. Mollica, M. Bestagno and G. Sava, J. Cell. Biochem., 2015, 116, 2385–2396.
11. A. M. Piccinini and K. S. Midwood, Int. J. Exp. Pathol., 2014, 99, 158–180.
12. B. O. Guillame-Gentil, O. Semenov, A. S. Roca, T. Groth, R. Zahn, J. Vörös and M. Zenobi-Wong, Adv. Mater., 2010, 22, 5443–5462.
13. W. M. Petroll and L. Ma, Dev. Dyn., 2008, 237, 2726–2736.
14. G. Wang, T. Zhao, X. Song, W. Zhong, L. Yu, W. Hua, M. M. Q. Xing and X. Qiu, Polym. Chem., 2015, 6, 283–293.
15. J. J. Rice, M. M. Martino, L. D. Laporte, F. Tortelli, P. S. Briquez and J. A. Hubbell, Adv. Healthcare Mater., 2013, 2, 57–71.
16. K. Nam, Y. Sakai, S. Funamoto, T. Kimura and A. Kishida, J. Biomater. Sci., Polym. Ed., 2011, 22, 1963–1982.
17. R. Bareille, F. Villars, M. Remy, P. Fernandez and L. Bordenave, Cell Biol. Int., 2003, 27, 439–443.
18. S. W. Liao, T.-B. Yu and Z. Guan, J. Am. Chem. Soc., 2009, 131, 17638–17646.
19. J. P. Jung, A. J. Sprangers, J. R. Byce, J. Su, J. M. Squirrell, P. B. Messersmith, K. W. Eliceiri and B. M. Ogle, Biomacromolecules, 2013, 14, 3102–3111.
20. X. Yang, H. Shang, C. Ding and J. Li, Polym. Chem., 2015, 6, 668–680.
21. B. A. Kaipparettu, I. Kuiatse, B. T.-Y. Chan, M. B. Kaipparettu, A. V. Lee and S. Oesterreich, BioTechniques, 2008, 45, 165–171.
22. Y. Mousseau, S. Mollard, H. Qiu, L. Richard, R. Cazal, A. Nizou, N. Vedrenne, S. Rémi, Y. Baaj, L. Fourcade, B. Funalot and F. G. Sturtz, Lab. Invest., 2014, 94, 340–349.
23. F. Ferranti, F. D'Anselmi, M. Caruso, V. Lei, S. Dinicola, A. Pasqualato, A. Cucina, A. Palombo, G. Ricci, A. Catizone and M. Bizzarri, PLoS One, 2013, 8, e76192.
24. F. D'Anselmi, M. G. Masiello, A. Cucina, S. Proietti, S. Dinicola, A. Pasqualato, G. Ricci, G. Dobrowolny, A. Catizone, A. Palombo and M. Bizzarri, PLoS One, 2013, 8, e83770.
25. S. Jalili-Firoozinezhad, S. Rajabi-Zeleti, P. Mohammadi, E. Gaudiello, S. Bonakdar, M. Solati-Hashjin, A. Marsano, N. Aghdami, A. Scherberich, H. Baharvand and I. Martin, Adv. Healthcare Mater., 2015, 4, 2281–2290.
26. I. A. Janson and A. J. Putnam, J. Biomed. Mater. Res., Part A, 2015, 3, 1246–1259.
27. J. Wlodarczyk, I. Mukhina, L. Kaczmarek and A. Dityatev, Dev. Neurobiol., 2011, 71, 1040–1053.
28. A. Mamun, C. G. Bazuin and R. E. Prud’homme, Macromolecules, 2015, 5, 1412–1417.
29. D. Gu, K. Ladewig, M. Klimak, D. Haylock, K. M. McLean, A. J. O'Connor and G. G. Qiao, Polym. Chem., 2015, 6, 6475–6487.
30. T. Zhang, E. Nies, G. Todorova, T. Li, H. Berghmans and L. Ge, J. Phys. Chem. B, 2008, 112, 5611–5615.
31. P. Viswanathan, S. Chirasatitsin, K. Ngamkham, A. J. Engler and G. Battaglia, J. Am. Chem. Soc., 2012, 134, 20103–20109.
32. T. Xia, Y. Huang, X. Jiang, Y. Lv, Q. Yang and G. Li, Macromolecules, 2013, 20, 8323–8333.
33. N. Ayres, Polym. Chem., 2010, 1, 769–777.
34. D. J. Phillips and M. I. Gibson, Polym. Chem., 2015, 6, 1033–1043.
35. D. E. Ingber, Differentiation, 2002, 70, 547–560.
36. T. W. Kragstrup, M. Kjaer and A. L. Mackey, Scand. J. Med. Sci. Sports, 2011, 21, 749–757.
37. I. Stamenkovic, J. Pathol., 2003, 200, 448–464.
38. B. L. Ekerdt, R. A. Segalman and D. V. Schaffer, Biotechnol. J., 2013, 8, 1411–1423.
39. A. Cretu and P. C. Brooks, J. Cell. Physiol., 2007, 213, 391–402.
40. S. Mathew, L. Fu, T. Hasebe, A. Ishizuya-Oka and Y.-B. Shi, Birth Defects Res., Part C, 2010, 90, 55–66.
41. J. Kim, Y. J. Kim, W.-G. Bae, K.-J. Jang, K. T. Lim, P.-H. Choung, Y.-H. Choung and J. H. Chung, J. Biomed. Mater. Res., Part B, 2015, 103, 1320–1327.
42. G. L. Cromar, X. Xiong, E. Chautard, S. Ricard-Blum and J. Parkinson, Proteins, 2012, 80, 1522–1544.
43. T.-W. Lin, H.-T. Chang, C.-H. Chen, C.-H. Chen, S.-W. Lin, T.-L. Hsu and C.-H. Wong, J. Am. Chem. Soc., 2015, 137, 9685–9693.
44. L. Liu, G. Duclos, B. Sun, J. Lee, A. Wu, Y. Kam, E. D. Sontag, H. A. Stone, J. C. Sturm, R. A. Gatenby and R. H. Austin, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 1686–1691.
45. A. Wu, K. Loutherback, G. Lambert, L. Estévez-Salmerón, T. D. Tlsty, R. H. Austin and J. C. Sturm, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 16103–16108.
46. K.-V. Nguyen-Ngoc, K. J. Cheung, A. Brenot, E. R. Shamir, R. S. Gray, W. C. Hines, P. Yaswen, Z. Werb and A. J. Ewald, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, E2595–E2604.
47. A. Luzardo-Álvarez, J. Blanco-Méndez, P. Varela-Patiño and B. Martín Biedma, J. Biomater. Sci., Polym. Ed., 2011, 22, 329–342.
48. Y. Liu, L. Zhang, J. Wei, S. Yan, J. Yu and X. Li, J. Mater. Chem. B, 2014, 2, 3029–3040.
49. J. J. Grzesiak and M. Bouvet, Cancer Sci., 2008, 99, 1553–1563.
50. P. Rørth, EMBO Rep., 2012, 13, 984–992.
51. J. M. Jang, S.-H.-T. Tran, S. C. Na and N. L. Jeon, ACS Appl. Mater. Interfaces, 2015, 7, 2183–2188.
52. M. Kato and M. Mrksich, J. Am. Chem. Soc., 2004, 126, 6504–6505.
53. K. Nam, Y. Sakai, S. Funamoto, T. Kimura and A. Kishida, J. Biomater. Sci., Polym. Ed., 2011, 22, 1963–1982.
54. L. Chen, Z. Xiao, Y. Meng, Y. Zhao, J. Han, G. Su, B. Chen and J. Dai, Biomaterials, 2012, 33, 1437–1444.
55. P. G. Gritsenko, O. Ilina and P. Friedl, J. Pathol., 2012, 226, 185–199.
56. R. Harisi, J. Dudás, F. Timár, G. Pogány, J. Timár, I. Kovalszky, T. M. Szendr and A. Jeney, Cell Biol. Int., 2005, 29, 959–967.
57. J. Leroy-Dudal, C. Demeilliers, O. Gallet, E. Pauthe, S. Dutoit, R. Agniel, P. Gauduchon and F. Carreiras, Int. J. Cancer, 2005, 114, 531–543.

---