Balanced Effects of Surface Reactivity and Self-Association of Bifunctional Polyaspartamide on Stem Cell Adhesion.

Extensive efforts have been made to regulate surface wettability using bivalent polymers composed of hydrophobic surface-reactive groups and hydrophilic groups. To further enhance the controllability, this study demonstrates that the balance between the surface reactivity and self-aggregation of bivalent poly(hydroxyethyl-co-methacryloxyethyl aspartamide) (PHMAA) is crucial in controlling the wettability of methacrylated glass and thus the adhesion of stem cells. In particular, the wettability of the glass and the subsequent cell spreading became maximal with PHMAA that led to the largest and most uniform coverage of hydroxyl groups. In summary, this study would be useful in advancing various molecules used for surface engineering.

Introduction

The technologies developed to regulate surface wettability have garnered much attention due to its usefulness in innumerable applications, such as printing,[1] liquid separation,[2] self-cleaning,[3] and biomolecular and cell manufacturing.[4] In particular, the surface energy of materials used for cell biology studies and large-scale cell manufacturing influences protein adsorption[5] and conformation,[6] and, in turn, regulates cell adhesion. For example, hydrophilic materials can promote the deposition of endogenous cell adhesion proteins and hence stimulate cell adhesion.[7] In contrast, hydrophobic materials show competition between albumin and cell adhesion proteins toward adsorption,[8] thereby hindering cell adhesion to the substrate.[9]

Prior studies commonly controlled surface energy by altering the chemical compositions of materials used in cell culture, which include glass, plastic, and silicon. However, this approach can influence other properties that can potentially affect the cellular response to materials, which include mechanical properties and permeability. This challenge prompted efforts to control surface wettability of materials without altering their bulk properties. One popular approach is to introduce desired chemical groups exclusively on the surface using chemical treatments, chemical vapor deposition,[10] and corona[11] or plasma[12] treatments. Alternatively, a series of fabrication techniques developed to control surface topology can control the surface energy of materials. However, these methods often require sophisticated and expensive processing units and high-level skills, thus making it difficult for industrial-scale manufacturing of materials.

To this end, it is popular to tailor the surface energy of cell adhesion substrates by coating the substrates with a controlled number of hydrophilic groups. As success in this approach relies on the sustained presence of the hydrophilic polymers on the substrate, it is common to use amphiphilic polymers, which are widely being used for surface coating and functionalization of various materials, with chemical groups reactive to the substrate.[13−15] In particular, hydrophobic units, such as methacrylate (MA) groups, are often coupled to the polymer so as to create covalent bonds with substrates that are pretreated to present MA groups. However, self-aggregation between these hydrophobic moieties may hamper the controllability of the substrate surface energy and subsequent cell adhesion, depending on the degree of substitution (DS) of the hydrophobic moieties. No systematic study has been made to address the potential effect of hydrophobic and chemically reactive groups on cell adhesion.

In this study, we hypothesized that the balance between surface reactivity and self-aggregation of the surface modifier would play a significant role in regulating the surface energy and ultimately cell adhesion. We examined this hypothesis by using poly(hydroxyethyl-co-methacryloxyethyl aspartamide) (PHMAA) as a model surface modifier because of minimal toxicity and easy chemical modifications. The hydroxyl (OH) group of PHMAA was used as a hydrophilic epitope of PHMAA, whereas the MA group was used as a hydrophobic one that is reactive to the substrates with MA groups. We used glass coverslips pretreated to present MA groups as a model substrate. The role of polymers was evaluated (1) by measuring the contact angle, (2) imaging OH groups of the substrate with Raman microscopy, (3) characterizing intermolecular association with pyrene probes, and (4) quantifying the number of cells adhered to the substrate. Previously, PHMAA was used as a protein linker that allows conjugations of extracellular matrix proteins to various substrates. Apart from the previous study, this study would focus on regulating the surface energy of the substrate by using PHMAA with controlled DS of MA groups.[16]

Results and Discussion

PHMAA was synthesized by a sequential reaction of 2-aminoethyl methacrylate (2-AEMA) and ethanolamine with polysuccinimide (PSI) (Figure ). The molar ratio between 2-AEMA and the succinimide of PSI was regulated to control the DS of the MA (DSMA) group from 0 to 10%, as characterized by the 1H NMR spectra (Figure S1, Table ). Separately, the molar ratio between ethanolamine and succinimide of PSI was stoichiometrically altered to vary the DS of OH groups (DSOH) from 95 to 85% to have the number of unopened succinimide rings in PHMAA maintained constant in all groups. The resulting DSOH ranged from 92 to 81% based on 1H NMR spectra (Figure S1, Table ). The molar ratio between MA and OH groups ranged from 0 to 11% (Table ).

Figure 1

Chemical synthesis scheme of PHMAA with controlled DSMA and DSOH. The PHMAA was synthesized by nucleophilic substitution of 2-AEMA (1) and ethanolamine (2) to PSI.

Table 1

DS of MA and OH Based on NMR

	stoichiometric DSMA
	0%	2%	5%	10%
DSMA	0%	2%	5%	10%
DSOH	92%	89%	86%	81%
DSMA/DSOH	0%	2%	6%	12%

The resulting PHMAAs with varied DSMA were used to coat the glass pretreated to present MA groups using a silane reagent. The polymers mixed with a photoinitiator, Irgacure 2959, were placed on the glass and exposed to ultraviolet (UV) light for immobilization. The immobilized PHMAA significantly changed the surface energy of the glass, which was analyzed by measuring the contact angle of a liquid water droplet (Figure ). The contact angle of a water droplet on the methacrylated glass was around 70°. Increasing DSMA from 0 to 5% reduced the contact angle of a water droplet from 68 to approximately 55°. According to the Young equation,[17] increasing DSMA from 0 to 5% made 1.7-fold increase of the glass/liquid interface energy. However, further increase of DSMA to 10% rather increased the contact angle of a water droplet to 60°.

Figure 2

Contact angle measurement of PHMAA-coated glass. Values and error bars represent average values and standard deviation of three different samples per condition, respectively. *Statistical significance between DSMA of 2 and 5% for the difference in contact angle (*p < 0.05). #Statistical significance between DSMA of 5 and 10% for the difference in contact angle.

Using Raman spectroscopy coupled with a microscope, the spatial distribution of OH groups on the glass surface was mapped to underlie the mechanism by which the DSMA of PHMAA modulates the surface energy of the glass. Raman microscopic imaging was carried out with peaks at 1600–1700 cm–1, as OH bending is shown near 1640 cm–1.[18] According to the Raman spectra and image, both bare glass surface treated to present MA groups and glass treated with poly(hydroxyethyl aspartamide) (PHA) groups presented a minimal number of OH groups. No peak at 1600–1700 cm–1 that represents the OH group was observed under these conditions (Figure A). Accordingly, the Raman image of the glass surface did not display the area that marked the presence of OH groups, indicated by the red color intensity (Figure A–C). This result confirms that PHA minimally adheres to the glass surface due to the absence of MA groups.

Figure 3

Raman microscopic imaging of OH groups on glass substrates coated with PHMAA with controlled DSMA. (A) Raman spectroscopy and (B) Raman intensity mapping of chemical peaks at 1600–1700 cm–1 assigned to the OH bending of PHMAA. The scale bar represents 100 μm. (C) Quantifications of (I) the relative area of the OH-coated surface (yellow- and red-colored areas in Raman mapping), (II) the number of OH-free islands, and (III) the area of OH-free islands (green- and blue-colored areas in Raman mapping). Values and error bars represent average values and standard deviation of three different samples per condition, respectively. *Statistical significance between DSMA of 5 and 10% for the difference in the relative area of the OH-coated surface (C-I), the number of OH-free islands (C-II), and the area of OH-free islands (C-III) (*p < 0.05).

In contrast, glass coated with PHMAA with 2% DSMA presented OH groups. The surface area covered by OH groups reached 50% of the entire surface. A further increase in DSMA to 5% raised the area percentage to present OH groups to more than 95% (Figure ). This enhanced coverage of OH groups is well related to the increase in surface energy evaluated by contact angle measurements. This result is owing to the larger number of PHMAA that stably adhered to the glass surface, which led to the larger number of OH groups on the surface.

Interestingly, glass surfaces coated with PHMAA with 5 and 10% DSMA presented some circular island areas free of OH groups (Figure ). In particular, PHMAA with DSMA of 10% created a larger number of OH group-free islands on the glass surface than PHMAA with DSMA of 5% (Figure C). The average diameter of islands was almost 2-fold greater with PHMAA with DSMA of 10% (Figure C). The area percentage of OH groups was slightly decreased by increasing DSMA from 5 to 10% (Figure C).

The larger OH-free islands made by PHMAA with DSMA of 10% implicated that self-aggregation between PHMAA molecules would be another key factor to influence the surface coverage of OH groups and the subsequent contact angle of the water droplet. Therefore, the self-association between PHMAA with varying DSMA was evaluated by examining the intermolecular organization of water-soluble 8-aminopyrene-1,3,6-trisulfonic acid trisodium salt (denoted as pyrene trisulfonates) in the polymer solution. The pyrene trisulfonates dissolved in water display an active “excimer” peak at approximately 490 nm and a “monomer” peak at approximately 420 nm when excited at 330 nm.[19,20] Note that the “excimer” represents pyrene probes stacked together. The water-soluble pyrene trisulfonates form excimers in aqueous media because the intermolecular arrangement is thermodynamically favorable. We suggest that the amphiphilic PHMAA polymer hinders this excimer formation by intercalating MA groups into the space between pyrene trisulfonates and, in turn, weakening the π–π bonds between pyrene molecules. We propose that effect of OH groups on the change in pyrene excimer fraction is not significant because the hydrophilic OH groups mainly interact with sulfonate groups substituted to the pyrene ring.

Interestingly, the pyrene trisulfonates introduced into the aqueous solution of PHA free from MA groups exhibited excimer peaks with the same height as that of pyrene trisulfonates introduced into polymer-free aqueous media (Figure A). In contrast, pyrene trisulfonates added into the aqueous solutions of PHMAA with DSMA being 2 and 5% exhibited a significant decrease in the excimer emission intensity (Iex). A further increase of DSMA to 10% resulted in another decrease in Iex. Therefore, the ratio of emission intensity between the excimer and monomer (Iex/Imo) was almost inversely proportional to the DSMA (Figure B5). The decrease in Iex/Imo of pyrene trisulfonates in the polymer solution indicates the extent of self-aggregation between amphiphilic polymers such as PHMAA. In particular, the result implicates that MA groups of PHMAA actively self-associate at DSMA of 10%, thus significantly inhibiting the excimer formation of pyrene probes in the media. It is highly likely that self-association of the MA groups of PHMAA with DSMA of 10% prevents the crosslinking between MA groups of PHMAA and the coverslip, which results in larger OH-free islands and decreased surface wettability (Figure ).

Figure 4

Pyrene-based analysis of the intermolecular association of PHMAA. (A) The emission spectra of pyrene suspended in PHMAA solutions. (B) The ratio of emission intensity between the excimer and monomer (Iex/Imo). Values and error bars represent average values and standard deviation of three different samples per condition, respectively. *Statistical significance between DSMA of 0 and 2% for the difference in Iex/Imo (*p < 0.05). #Statistical significance between DSMA of 5 and 10% for the difference in Iex/Imo The difference of Iex/Imo between 2 and 5% is not statistically significant.

Figure 5

Schematic description of the PHMAA conformation on the glass substrate. In PHMAA, the blue segment represents OH groups, the violet segment represents polymer backbone, and the red chain represents MA groups.

This controlled surface energy of the glass influenced the adhesion of cells to the substrate. This study used bone marrow-derived mesenchymal stem cells (MSCs) as a model cell. MSCs have recently emerged as a new regenerative medicine because of their potential to differentiate multiple tissue-forming cells and secrete a series of therapeutic growth factors and cytokines.[21,22] As the successful use of these cells relies on the capability to produce them on a large scale, extensive efforts have been made to optimize the extracellular microenvironment, including the surface energy of substrates used for in vitro cell manufacturing. MSCs were placed on the glass surface coated by PHMAA with varying DSMA. After incubation for 3 h, MSCs showed two different states of adhesion: some cells stably adhered to the substrate and started to spread, whereas others remained round due to poor adhesion (Figure A). The glass coated with PHMAA with DSMA of 5% displayed the maximal number of cells that stably adhered to the glass (Figure A,C).

Figure 6

Analysis of MSC adhesion on glass substrates. (A) Optical microscopy images of cells adhered to the PHMAA-coated glass. The scale bar represents 200 μm. (B) Fluorescence images of actin filaments in cells adhered to the PHMAA-coated glass. Green color represents actin filaments, and blue color represents cell nuclei. The scale bar represents 50 μm. Quantification of (C) the relative number and (D) the area of cells adhered to the PHMAA-coated glass. Values and error bars represent average values and standard deviation of three different samples per condition. *Statistical significance between DSMA of 2 and 5% for the difference in the relative number of cells (*p < 0.05). #Statistical significance between DSMA of 5 and 10% for the difference in the area of cells.

The long-term stability of the PHMAA coating was evaluated by examining the adhesion and proliferation of MSCs plated on the substrate over a week. MSCs were seeded at either a lower concentration of 250 cells/cm2 or a higher concentration of 2500 cells/cm2. The number of cells initially adhered was maximal with the glass coated with PHMAA with DSMA of 5%, regardless of the cell seeding density (Figure S2). The cells proliferated and became confluent throughout 5–7 days of cell culture (Figure S2). This result concludes that PHMAA coupled to the glass surface can stably support cell adhesion and proliferation.

Consistent with the analysis of the cell attachment, the extent of cell spreading was also regulated by DSMA of PHMAA used to coat the glass surface. In this analysis, intracellular actin filaments were stained with fluorescently labeled phalloidin. MSCs cultured on the methacrylated glass and those on the glass treated with PHA displayed a round morphology. Most actin filaments were localized along the edge of a cell membrane (Figure B). PHMAA with DSMA of 5% led to the maximum spreading of cells (Figure B,D). In particular, PHMAA with DSMA of 5% made almost 2-fold increase of the cell spreading area than PHA free of MA groups. In contrast, the glass coated with PHMAA with 10% DSMA decreased the extent of cell spreading compared with the glass covered with PHMAA with 5% DSMA.

These cell adhesion properties regarding the cell attachment rate and spreading were well related to the contact angle of water droplets placed on the glass surface. In particular, the number of adhered cells and the extent of cell spreading became maximal on the glass surface, which led to the lowest contact angle of the water droplets. As the contact angle of the water droplet represents the surface energy of a substrate, we propose that cell adhesion is solely dependent on the surface energy controlled with PHMAA engineered to present controlled DSMA.

Conclusions

In conclusion, the results of this study demonstrate that a balance between the reactivity and self-association of PHMAA plays a significant role in regulating the surface energy of and, in turn, cell adhesion to a cell culture substrate. Either PHA or PHMAA with a low DSMA (e.g., 2%) could not sufficiently increase the surface energy of glass used for cell culture substrates because of low reactivity to the methacrylated glass. Therefore, cells could not properly adhere and spread on the substrate. In contrast, PHMAAs with a high DSMA (e.g., 10%) self-associate in aqueous solution and subsequently form hydrophobic islands on the substrate, which negatively affects cell attachment and spreading. Overall, the cell attachment and spreading were maximal on the glass coated with PHMAA with 5% DSMA due to the balance between surface reactivity and intermolecular association. We propose that this new finding will be highly useful in advancing the coating quality of a wide array of molecules used to modulate the surface properties of cell culture and manufacturing substrates.

Materials and Methods

Materials

All chemicals were purchased from Sigma unless specified otherwise.

PHMAA Synthesis

PSI was synthesized by acid-catalyzed polycondensation of l-aspartic acid (Sigma, MO).[1] Designated amounts of 2-AEMA (Sigma) and ethanolamine were added sequentially to PSI dissolved in dimethylformamide. The molar ratio of 2-AEMA was adjusted to regulate the DS of MA groups (DSMA) from 0 to 10%. The molar ratio of ethanolamine was adjusted to regulate the DS of hydroxyl groups from 95 to 85%, respectively. The molar amounts of the reagents used for the reactions are listed in Table .

Table 2

Molar Amounts of the Reagents Used for the Reactions

	moles
PSI	0.02	0.02	0.02	0.02
ethanolamine	0.019	0.0186	0.018	0.017
AEMA	0	0.0004	0.001	0.002

The resulting polymer was dissolved in DMSO-d6 (Cambridge Isotope Laboratories, Inc.) and analyzed by 1H NMR (Varian Unity 500 MHz). DSMA and DSOH were calculated from the integrated peaks of the 1H NMR spectra using the following equations[2]

Coating of Glass with PHMAA

To coat glass with PHMAA, glass was first modified to present MA groups. Briefly, coverslips were treated with Piranha (25% H2SO4, 15% H2O2) for 1 h, then rinsed with H2O and ethanol, and dried. The treated glass was submerged in a solution of 2% (v/v) 3-(trimethoxysilyl)propyl methacrylate (Sigma) dissolved in ethanol for 30 min. The solution pH was adjusted to 5.0 with glacial acetic acid. Then, the MA-group-coated glass was washed with ethanol and dried. To conjugate PHMAA on the MA-modified glass, the modified coverslips were submerged in PHMAA (20 mg/mL) mixed with 0.01% Irgacure 2959 (Ciba), and the polymer solution was exposed to UV light for 20 min to activate the radical crosslinking reaction.

Hydroxyl Group Mapping Using the Raman Microscope

Raman spectra of OH groups on the glass surface were mapped using the Nanophoton Raman-11 Laser Raman Microscope system equipped with a 532 nm wavelength laser. For each spectrum, a grating (600 nm–1) scan was taken over the range 1400–3600 cm–1. The spectral intensity was normalized to the hydroxyl peak at 1600–1800 cm–1.

Analysis of Pyrene-Based Intermolecular Association

The hydrophobic interactions between PHMAA in the solution were assessed by the monomer and excimer emission of pyrene probes. Stock solution (1 μL) of 8-aminopyrene-1,3,6-trisulfonic acid (pyrene, 100 μg/mL in cyclohexane) was added to 1 mL of PHMAA solution (20 mg/mL) with varying DSMA. The mixtures were sonicated for 1 min to completely dissolve the pyrene and incubated for 1 day. The mixture was excited at a wavelength of 330 nm, and the resulting emission spectrum of pyrene in each solution ranging from 350 to 600 nm was acquired using a spectrofluorometer (FluoroMax-4, HORIBA Jobin Yvon).

Cell Culture and Analyses

Mouse bone marrow-derived MSCs (D1, ATCC) between passages 21 and 25 were cultured in Dulbecco’s modified Eagle’s medium (Gibco-BRL), containing 10% (v/v) fetal bovine serum (Gibco-BRL) and 1% (v/v) penicillin–streptomycin (Gibco-BRL). MSCs were plated at a cell density of 1 × 105 cells/cm2 and examined after 3 h. Cells were fixed, permeabilized, and sequentially incubated with Alexa Fluor 488 phalloidin (Molecular Probes) and 4′,6-diamidino-2-phenylindole (Sigma). Then, cell actins and the nucleus were imaged using a laser scanning confocal microscope (LSM 710, Zeiss).

Statistical Analysis

The results were expressed as mean ± standard deviation and the data were analyzed using the Graph pad prism 6.0 software. Statistical analysis was performed using one-way analysis of variance with the Tukey significant difference post hoc test. The p-value <0.05 (95% confidence interval) was considered to be significantly different.