A zwitterionic surface with general cell-adhesive and protein-resistant properties

Abstract

Choline phosphate (CP) contains amino and phosphate groups in the reverse order in which they are present in phosphate choline (PC), which is the headgroup of phospholipids in all eukaryotic cell membranes. Here we used HeLa and L929 cells to study the general interaction between CP-modified surfaces and cells/proteins. The results of cell counting kit-8 (CCK-8), confocal laser scanning (CLSM) and scanning electron microscopy (SEM) indicated that CP-modified surfaces could improve cell adhesion by almost three times than unmodified surfaces through a unique CP–PC interaction with PC headgroups of the cell membrane. The mechanism for this interaction was the formation of a quadrupole from two quaternary nitrogen–phosphorus pairs. Moreover, the results of fibrinogen and fetal bovine serum protein adsorption experiments indicated that CP-modified surfaces could also resist nonspecific protein adsorption due to the zwitterionic property of the CP group. Therefore, this surface offers a general strategy for the preparation of biomaterials with both cell-adhesive and protein-resistant properties.

---

1. Introduction

In many biomedical applications of materials, such as biomaterials for an artificial extracellular matrix (ECM), the surface chemical property of biomaterials can be modified to control cell adhesion and proliferation. To improve cell adhesion, the most common strategy used is the introduction of cell adhesion-mediating proteins or peptides on to the surface of biomaterials, which can mimic the environment of the ECM.¹ Cell adhesion-mediating proteins are mainly ECM proteins, which contain specific functional domains that can interact with integrin receptors present on the cell surface. Cell adhesion peptides, i.e., arginine–glycine–aspartic acid (RGD) peptides, which are present in many ECM proteins, are widely used as a ligand for integrin-mediated cell adhesion. For example, poly(ε-caprolactone) (PCL) film surfaces have been modified with collagen or RGDs, which demonstrated excellent cell-adhesion characteristics.^1a However, when biomaterials are implanted in vivo, a variety of proteins from the blood quickly adsorb on to the surface, which may lead to a series of complications, such as promoting platelet adhesion and even thrombosis, and ultimately to implantation failure. For example, Horbett et al. have shown that platelet adhesion occurs on fibrinogen-preadsorbed surfaces and that platelet adhesion correlates with the amount and conformation of adsorbed fibrinogen.² Since proteins are typically polyelectrolytes with zwitterionic charge, surface modification with a zwitterionic or hydrophilic group could inhibit nonspecific protein adsorption through enrichment of the hydrated layer or steric effects, and in general, reduce the interaction of proteins with materials. Thus, considerable efforts have been devoted to modifying material surfaces with functional groups such as zwitterionic and poly(ethylene glycol) (PEG) groups to prevent nonspecific protein adsorption.³ For instance, Liu et al. have fabricated three zwitterionic surfaces and a PEG surface on cellulose membranes, which all reduced the nonspecific adsorption of proteins, platelet adhesion and cell attachment, indicating generally and significantly improved anti-biofouling properties.^3h,i

However, most materials surfaces with nonspecific protein-resistant properties may also resist cell adhesion-mediating proteins, such as fibronectin and vitronectin, which subsequently inhibits cell adhesion. Hence, there is an urgent need to design materials with surfaces that have both protein-resistant and cell-adhesive properties. To this end, various strategies have been developed based on the combination of protein-resistant molecules (e.g., PEG) and cell-adhesive molecules (e.g., RGD), copolymer modifications and other specific molecules.^1c–e,4 For example, Ding's group has modified persistent non-fouling PEG hydrogels with RGD to improve a fundamental understanding of cell–material interactions, which indicated that different nanospaces or microarrays could influence cell adhesion, spreading, proliferation, and differentiation.^1c–e

Recently, Brooks et al. reported the synthesis of choline phosphate (CP) containing amino and phosphate groups in the reverse order of phosphate choline (PC), which is the headgroup of phospholipids in all eukaryotic cell membranes.⁵ They synthesized polymers with CP groups and found that these molecules can bind to a variety of cell membranes driven by a unique CP–PC interaction between the molecules and cell membranes.^5a Some other research groups have introduced CP groups on to polymers or micelles, which have also exhibited favorable biocompatibility and interactions with cells.⁶ Based on these interesting findings, we also introduced CP groups on to a surface by ‘click reaction’ and investigated protein (bovine serum albumin, collagen and serum protein) adsorption and human umbilical vein endothelial cell adhesion to test whether CP-modified surfaces have potential for tissue engineering.⁷

In this work, we attempted to study the general interactions between CP-modified surfaces (Glass-CP) and cells/proteins. Two cell lines, HeLa cells, as a model cancer cell line, and L929 cells, as a model normal cell line, were used to study the general cell adhesion properties of the Glass-CP surface. Furthermore, fibrinogen, as a model protein and a key protein for biomaterials in blood-contacting applications, was also used to study the general protein-resistance properties of the Glass-CP surface. Serum-containing culture medium was used to further study the cell-adhesive mechanism of Glass-CP surfaces. We demonstrate that CP-modified surfaces have both cell-adhesive and protein-resistant properties for potential biomedical applications (Scheme 1).

Scheme 1 Schematic illustration of the protein-resistant and cell-adhesive properties of CP-modified surface.

2. Materials and methods

2.1 Materials

Materials: glass slides (18 × 18 mm, CITOGLAS®) were modified with prop-2-ynyle choline phosphate (p-CP).⁷ Cell counting kit-8 (CCK-8) assay kit, Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), sodium dodecyl sulfate (SDS), DAPI, penicillin and streptomycin were purchased from Baoxin Biotechnology Co. Ltd (Chengdu, China). LysoTracker Green DND-26 was purchased from Shanghai Qcbio Science & Technologies co. Ltd. Ultrapure water with a resistivity of 18.2 MΩ cm was used throughout.

Characterizations: scanning electron microscopy (SEM) was performed on a Hitachi S-450 microscope operated at 20 kV. Confocal laser scanning microscopy (CLSM) was performed on a TCRS SP5 microscope (Leica, Germany). A microplate reader (Spectra Plus, Tecan, Zurich, Switzerland) was used to record the adsorption intensity at 578 nm.

2.2 Protein adsorption

Protein adsorption experiments were conducted with a bovine serum fibrinogen solution (1 mg mL⁻¹ in PBS, pH 7.4) and DMEM culture medium with 20% FBS. Glass slides were immersed in the protein solution for 3 h at 37 °C or serum-containing cell culture medium for 3 h at 37 °C with an atmosphere of 5% CO₂ and 95% relative humidity. After protein adsorption, glass slides were carefully washed with PBS three times. Then, 1 mL of 2 wt% sodium dodecyl sulfate (SDS) solution was added for 30 min at 37 °C to remove proteins adsorbed on to the glass slides. The amount of protein eluted into the SDS solution was quantified with commercial BCA Protein Assay Kits (Beyotime, Shanghai, China) at 578 nm.

2.3 Cell culture

2.3.1 Cell adhesion assay. In order to study the interaction between cells and CP-modified glass slides, SEM and CLSM images were captured to observe cell adhesion. Slides were disinfected with 75% ethanol for 1 h and sterilized by UV irradiation for 1 h before use. HeLa or L929 cells were allowed to adhere for 3 h on to glass slides in 6-well plates at a density of 2 × 10⁵ cells per well and incubated in 2.5 mL serum-free DMEM at 37 °C with an atmosphere of 5% CO₂ and 95% relative humidity. After 3 h, slides were washed three times with PBS to remove loosely attached cells, and then transferred to fresh 6-well plates and incubated in 2.5 mL DMEM supplemented with 10% fetal bovine serum (FBS), 100 units per mL of penicillin and 100 μg mL⁻¹ of streptomycin at 37 °C for 24–48 h with an atmosphere of 5% CO₂ and 95% relative humidity. After 3 h of HeLa cell adhesion, the culture media was removed and cells were washed three times with PBS (pH 7.4) to remove loosely attached cells, then fixed with 4% paraformaldehyde for 4 h and dehydrated in a series of ethanol solutions (50–100%) for SEM imaging. For CLSM imaging, at different time intervals, the culture media was removed and cells were washed three times with PBS. Then cell lysosomes were stained with LysoTracker Green DND-26 (500 nm) for 20 min and cells were fixed with 4% paraformaldehyde at 4 °C for 20 min. Afterwards, cell nuclei were stained with DAPI (5 μg mL⁻¹) for 10 min. Finally, glass slides were placed on a glass microscope slide and the prepared samples were imaged by fluorescence microscopy using the CLSM. The number of attached HeLa cells on the surface was counted from six randomly selected fields (4×).

2.3.2 Cell proliferation assay. Cell counting kit-8 (CCK-8) assay was used to analyze cell adhesion and proliferation. The optical density (OD) values of CCK-8 containing culture media at 450 nm were measured after cultivating for 3 h at 37 °C in serum-free or 10% FBS-containing cell culture medium. After the CCK-8 was incubated with cells for 3 h, the medium was exchanged with DMEM supplemented with 10% FBS, 100 units per mL of penicillin and 100 μg mL⁻¹ of streptomycin for CCK-8 assays done at 24 and 48 h.

2.4 Statistical analysis

Data are presented as mean ± SD. Analysis of variance and t-test were used for statistical comparison. A p value < 0.01 was considered significant.

3. Results and discussion

Choline phosphate, which contains amino and phosphate groups in the reverse order of phosphate choline, was synthesized by a two-step method and used to modify a glass surface by a surface ‘click reaction’.⁶ Fig. 1 shows the adsorption of fibrinogen proteins on to Glass-OH and Glass-CP surfaces. Fibrinogen was chosen as a model protein to examine the protein-resistant properties of Glass-CP. Unmodified glass surfaces adsorbed more fibrinogen protein than Glass-CP surfaces (p < 0.01). The amount of fibrinogen adsorbed on to the unmodified glass surface was 1.2429 μg cm⁻², however, for the Glass-CP surface, the amount of fibrinogen adsorbed was 0.5549 μg cm⁻²; i.e., a reduction of 55.35%. In our previous work, we showed that the Glass-CP surface could also significantly reduce the adsorption of BSA and collagen.⁷ Since fibrinogen is a key protein for biomaterials in blood-contacting applications, the reduction of fibrinogen adsorption has important implications for blood compatibility. The protein-resistant property of CP-modified surfaces could be due to the zwitterionic property of CP groups, which can restrain nonspecific protein adsorption through the enrichment of the hydrated layer. It is noted that fibrinogen is merely a model protein, and it is not contained in the serum medium in the following cell culture studies; thus, its amount does not have a direct correlation with cellular adhesion in this work.

Fig. 1 Amounts of fibrinogen adsorbed on to Glass-OH and Glass-CP for 3 h. Protein concentration is 1 mg mL⁻¹. (**p < 0.01; data is represented as mean ± SD, n = 3).

HeLa cells were used as a model of adherent cells to study the general interaction of cells with CP-modified surfaces. Cell counting kit-8 (CCK-8) assays were used to test the adhesion and proliferation of HeLa cells on Glass-OH and Glass-CP (Fig. 2). During the initial 3 h, the glass substrates were incubated in serum-free or serum-containing cell culture medium to explore the mechanism. HeLa cells significantly adhered better on to Glass-CP than on to Glass-OH, irrespective of whether serum-free or serum-containing medium was added (p < 0.01). In addition, the optical density (OD) value of the Glass-OH group in serum-containing cell culture medium was significantly higher than that of the Glass-OH group in serum-free medium (p < 0.01), indicating that more cells adhered on to glass when serum was added. It is interesting to note that, for the Glass-CP group, no significant difference in the number of adhered cells was observed between the serum-free and serum-containing conditions in the first 3 h (p > 0.05), which will be discussed later.

Fig. 2 CCK-8 assay indicating the proliferation profiles of HeLa cells, which were cultured without or with serum in the initial 3 h. (**p < 0.01).

To further study the cell adhesion behavior on CP-modified surfaces, HeLa cells were imaged using a CLSM (Fig. 3). HeLa cells were allowed to adhere on to glass slides in serum-free cell culture medium for 3 h. Then, the medium was changed to 10% serum-containing cell culture medium. The results indicated that initially more cells can adhere on to the Glass-CP surface than the Glass-OH surface and more cells can grow with time. It is obvious that more cells adhered on to the Glass-CP surface, which had almost three times the number of cells than that on the Glass-OH surface (1403.0 ± 94.6 vs. 476.0 ± 63.8 per view) at the initial stage of cell adhesion (Fig. 4). We also investigated what happened in serum-free cell culture medium within the initial 3 h of cell adhesion by SEM images (Fig. 5). By SEM, we noted that numerous cells had a flattened well spread out morphology on Glass-CP surfaces, which benefits subsequent cell proliferation. Meanwhile, the CCK-8 and CLSM results also indicated that cell adhesion at the initial stage influences cell proliferation at 24 and 48 h in a similar manner. These results indicated that CP-modified surfaces improve cell adhesion in serum-free cell culture medium. Cell adhesion is enhanced due to strong CP interactions with the PC headgroups of cell membranes, that form as a result of a quadrupole from two quaternary nitrogen–phosphorus pairs, which was illustrated previously in Brooks' work.^5a

Fig. 3 CLSM images of HeLa cells on Glass-OH and Glass-CP at different time intervals (scale bar: 50 μm).

Fig. 4 Adhesion of HeLa cells on different surfaces in serum free medium after 3 h. The number was counted from six randomly selected fields (4×, **p < 0.01).

Fig. 5 SEM images of HeLa cells on Glass-OH and Glass-CP surfaces after adhesion for 3 h (scale bar: 100 μm). All images are of cells in serum-free conditions.

Although HeLa cells are an adherent cell, it is also a cancer cell. Therefore, we further tested L929 cells, i.e., mouse fibroblast cells, as a model of a normal cell, to investigate the general interaction between cells and CP-modified surfaces. Fig. 6 shows the adhesion and proliferation profiles of L929 cells on Glass-OH and Glass-CP surfaces (as assayed by CCK-8 test). In the initial 3 h, glass substrates were incubated in serum-free or serum-containing cell culture medium. L929 cells, similar to HeLa cells, also significantly adhered better on to Glass-CP than on to Glass-OH, irrespective of whether serum-free or serum-containing medium was added (p < 0.01). The optical density (OD) values of the Glass-OH group in serum-containing cell culture medium are significantly higher than those in serum-free medium (p < 0.01), indicating that more cells adhered on to glass when serum was present. As for the Glass-CP group, no significant difference in the number of adherent cells was observed between serum-free and serum-containing conditions in the first 3 h (p > 0.05). The CLSM images of L929 cells also indicated that more cells adhered on to the Glass-CP surface initially stage and more cells can grow with time (Fig. 7). It is of interest to note that adhesion on CP-modified surfaces is better for HeLa cells than for L929 cells, which may be due to an intrinsic ability of HeLa cells to adhere to surfaces. These results suggest that although CP-modified surfaces can promote the adhesion of various cells, such as cancer cells and fibroblast cells, their cell adhesion profiles may differ depending on intrinsic properties of the different cells.

Fig. 6 CCK-8 assay indicating the proliferation profiles of L929 cells, which were cultured without or with serum in the initial 3 h (**p < 0.01).

Fig. 7 CLSM images of L929 cells on Glass-OH and Glass-CP at different time intervals (scale bar: 50 μm).

To further understand the mechanism underlying the special cell-adhesive and protein-resistant properties of CP-modified surfaces, we undertook additional serum protein adsorption tests using DMEM culture medium with 20% FBS. Glass slides were immersed in serum-containing cell culture medium for 3 h at 37 °C with an atmosphere of 5% CO₂ and 95% relative humidity. The results are shown in Fig. 8: the amounts of serum protein adsorbed on to Glass-CP and Glass-OH surfaces are 0.2527 and 0.5716 μg cm⁻², respectively. That is, serum protein adsorption on to Glass-CP is reduced by 55.79% compared to adsorption on to Glass-OH. This result is consistent with cell adhesion data shown in Fig. 2, 3, 6, and 7. In serum-containing culture medium, many serum proteins, such as ECM proteins, can be adsorbed on to the Glass-OH surface, which can promote cell adhesion. It has been well recognized that reduced protein adsorption on to a surface is correlated with reduced cell adhesion. However, the cell-adhesive mechanism of Glass-CP surfaces is different, since it promotes significantly higher cell adhesion (both in serum-free and serum-containing condition) than Glass-OH surfaces, but adsorbs significantly less serum protein than Glass-OH surfaces. Thus, cell adhesion on to Glass-OH surfaces is due to the adsorption of ECM proteins, but the general cell adhesive property of Glass-CP surfaces is likely due to the unique CP–PC interaction between Glass-CP and cell membranes,^5a which is not discernably affected by ECM proteins adsorption.

Fig. 8 Amounts of serum proteins adsorbed on to Glass-OH and Glass-CP surfaces after 3 h. Serum proteins are from 20% FBS in DMEM culture medium. (**p < 0.01; data is represented as mean ± SD, n = 3).

4. Conclusions

In this study, fibrinogen and serum protein adsorption, and cell adhesion experiments were conducted to investigate the general properties of a zwitterionic CP-modified surface. The CP-modified surface can reduce the adsorption of fibrinogen (55.35%) and serum protein (55.79%), due to the zwitterionic property of CP groups, but can also promote cell adhesion by almost three fold through the unique CP–PC interaction. Our study provides a general strategy to prepare cell-adhesive surfaces with protein-resistance for biomedical applications.

Acknowledgements

The authors thank the financial support by the National Natural Science Foundation of China (51322303, 21534008, 51573110) and the State Key Laboratory of Polymer Materials Engineering (sklpme2014-3-01).

References

1. (a) F. J. Xu, Z. H. Wang and W. T. Yang, Biomaterials, 2010, 31, 3139; (b) F. J. Xu, Y. Q. Zheng, W. J. Zhen and W. T. Yang, Colloids Surf., B, 2011, 85, 40; (c) Z. Zhang, Y. Lai, L. Yu and J. Ding, Biomaterials, 2010, 31, 7873; (d) X. Wang, C. Yan, K. Ye, Y. He, Z. Li and J. Ding, Biomaterials, 2013, 34, 2865; (e) B. Cao, R. Peng, Z. Li and J. Ding, Biomaterials, 2014, 35, 6871.
2. (a) J. M. G. WeiBor and T. A. Horbett, J. Biomed. Mater. Res., 1999, 44, 130; (b) M. C. Shen and T. A. Horbett, J. Biomed. Mater. Res., 2001, 57, 336.
3. (a) I. Banerjee, R. C. Pangule and R. S. Kane, Adv. Mater., 2011, 23, 690; (b) J. R. Lu, E. F. Murphy, T. J. Su, A. L. Lewis, P. W. Stratford and S. K. Satija, Langmuir, 2001, 17, 3382; (c) H. Chen, X. Hu, Y. Zhang, D. Li, Z. Wu and T. Zhang, Colloids Surf., B, 2008, 61, 237; (d) H. Chen, M. A. Brook, Y. Chen and H. Sheardown, J. Biomater. Sci., Polym. Ed., 2005, 16, 531; (e) H. Kitano, A. Kawasaki, H. Kawasaki and S. Morokoshi, J. Colloid Interface Sci., 2005, 282, 340; (f) Y. Xu, M. Takai and K. Ishihara, Biomaterials, 2009, 30, 4930; (g) K. Jang, K. Sato, K. Mawatari, T. Konno, K. Ishihara and T. Kitamori, Biomaterials, 2009, 30, 1413; (h) P. Liu, Q. Chen, L. Li, S. Lin and J. Shen, J. Mater. Chem. B, 2014, 2, 7222; (i) S. K. Schmitt, W. L. Murphy and P. Gopalan, J. Mater. Chem. B, 2013, 1, 1349; (j) C. J. Huang, Y. Li, J. B. Krause, N. D. Brault and S. Jiang, Macromol. Rapid Comm., 2012, 33, 1003; (k) Y. Chang, Y. J. Shih, C. J. Lai, H. H. Kung and S. Jiang, Adv. Funct. Mater., 2013, 23, 1100.
4. (a) X. Liu, Z. Wu, F. Zhou, D. Li and H. Chen, Colloids Surf., B, 2010, 79, 452; (b) D. Grafahrend, K. H. Heffels, M. V. Beer, P. Gasteier, M. Moller, G. Boehm, P. D. Dalton and J. Groll, Nat. Mater., 2011, 10, 67; (c) Y. Wei, Y. Ji, L. L. Xiao, Q. K. Lin, J. P. Xu, K. F. Ren and J. Ji, Biomaterials, 2013, 34, 2588.
5. (a) X. Yu, Z. Liu, J. Janzen, I. Chafeeva, S. Horte, W. Chen, R. K. Kainthan, J. N. Kizhakkedathu and D. E. Brooks, Nat. Mater., 2012, 11, 468; (b) X. Yu, Y. Zou, S. Horte, J. Janzen, J. N. Kizhakkedathu and D. E. Brooks, Biomacromolecules, 2013, 14, 2611; (c) X. Yu, X. Yang, S. Horte, J. N. Kizhakkedathu and D. E. Brooks, Chem. Commun., 2013, 49, 6831; (d) X. Yu, X. Yang, S. Horte, J. N. Kizhakkedathu and D. E. Brook, Macromol. Biosci., 2014, 14, 334; (e) X. Yu, X. Yang, S. Horte, J. N. Kizhakkedathu and D. E. Brooks, Biomaterials, 2014, 35, 278.
6. (a) S. Fujii, K. Nishina, S. Yamada, S. Mochizuki, N. Ohta, A. Takahara and K. Sakurai, Soft Matter, 2014, 10, 8216; (b) G. Hu, S. S. Parelkar and T. Emrick, Polym. Chem., 2015, 6, 525.
7. X. Chen, T. Chen, Z. Lin, X. Li, W. Wu and J. Li, Chem. Commun., 2015, 51, 487.

---