Entangled Polymer Surface Confinement, an Alternative Method to Control Stem Cell Differentiation in the Absence of Chemical Mediators

Research Article

Ann J Materials Sci Eng. 2014;1(3): 7.

Entangled Polymer Surface Confinement, an Alternative Method to Control Stem Cell Differentiation in the Absence of Chemical Mediators

Chang C1, Bherwani A2, Simon M2, Rafailovich M1 and Jurukovski V1,3*

1Department of Materials Science & Engineering, Stony Brook University, USA

2Department of Oral and Pathology, Stony Brook University, USA

3Department of Biology, Suffolk County Community College, USA

*Corresponding author: Jurukovski V, Department of Materials Science & Engineering, Stony Brook University, 314 Engineering Bldg., Stony Brook, NY 11794-2275, USA

Received: July 15, 2014; Accepted: September 01, 2014; Published: September 24, 2014


We have demonstrated that the moduli of spun cast and annealed polybutadiene films can be enhanced by nearly an order of magnitude, due to confinement near an attractive HF etched Si substrate interface. The region of enhancement scaled with the polymer radius of gyration and persisted for distances greater than twenty times the radius of gyration of the polymer, Rg, from the Si interface. We also showed that dental pulp stem cells (DPSC) can be plated directly on these films without any additional coatings and that the DPSCs were able to adjust their moduli in a continuous manner in response to that of the substrate. A critical value of the substrate modulus, Mc= 2.3MPa was found which was independent of substrate molecular weight or thickness, such that for substrates with moduli M>Mc, the DPSC were induced to produce hydroxyapatite mineral deposits and upregulate gene expression of osteocalcin (OCN) and alkaline phosphatase (ALP), in the absence of any other soluble factors.

Keywords: Biomineralization; Hydroxylapatite; Mechanical properties; Odontogenesis/Osteogenesis; Stem cells; Elastomers; Surface interactions


PB: Polybutadiene; DPSC: Dental Pulp Stem Cells; SEM: Scanning Electron Microscopy; EDX: Energy-dispersive X-ray Spectroscopy; GIDX: Grazing Incidence X-ray Diffraction; RT-PCR: Real Time- Polymerase Chain Reaction; OCN: Osteocalcin; ALP: Alkaline Phosphatase; LSCM: Laser Scanning Confocal Microscopy; Dex: Dexamethasone; RMS: Root Mean Squared; AFM: Atomic Force Microcopy; SMFM: Shear Modulation Force Microscopy


Recently, considerable advances have been made our understanding of the interaction between stem cells and materials scaffolds. In particular much effort has been expended on materials for the delivery of stem cells for regenerative medicine [1]. While the central focus is the healing of damaged tissues and organs, there is concern that the transplantation niche may not support cell proliferation or may promote unpredicted and undesired changes in cell differentiation. Numerous groups have experimented with hydrogels where the mechanical properties are controlled via the cross linking density. Even though these systems were able to demonstrate successfully that differentiation could be controlled via the substrate modulus, most of the gels required additional coatings to allow cell adhesion or mask the effect of excess cross linker which can segregate to the free surface [2,3]. In order to accomplish differentiation in the absence of any additional chemical factors or coatings, we propose an alternative method to cross linked hydrogel scaffolds. It has been previously established that confinement of long chain entangled polymers near an interactive surface, can pin the chains to the surface, forming an effective “strangulated” gel [4]. It was shown that the influence of the confinement on the rheological properties of the films persisted for more than 10 polymer radii of gyration from the surface. Hence, through variation of the film thickness it was shown that the viscosity of the film can be changed by more than an order magnitude [5]. We therefore postulated that the mechanical response of the films can also be controlled simply by changing the film thickness of a polymer on a surface which interacted favorably with the polymer. Here we demonstrate that this is indeed the case when long chain, entangled elastomers are used. Furthermore, we show that this method can be used to determine the functional form, which, as expected, is continuous and differential, of the change in modulus with film thickness, or distance from the interacting surface. This system provides a convenient platform which allows us to ask several fundamental questions regarding stem cell response to changes in the substrate mechanics which cannot otherwise be addressed in a hydrogel system. First, since these polymers are hydrophobic, we must establish the ability of these substrates to promote cell adhesion in the absence of additional chemical factors or coatings. Second, assuming that the surfaces are adhesive, we can determine whether the cells adapt to changes in mechanics in a continuous manner, where even small differential changes can be measured. Finally, we can determine whether changes in the cell modulus induced by the surface can promote differentiation in the absence of additional soluble mediators.

Undifferentiated mesenchymal stem cells isolated from dental pulp (dental pulp stem cells-DPSCs) when grown in specific inducing media can differentiate and express markers of odontoblasts, osteoblasts, adipocytes or neuronal cells [6-10]. In addition, DPSCs have been shown to produce dentin when implanted in immuno-compromised mice [6,8,11]. cDNA microarray analyses show that DPSCs have gene expression pattern similar to that of bone marrow mesenchymal stem cells [12]. Induction of DPSCs with dexamethasone (Dex), an active glucocorticoid analog, in presence of inorganic phosphate and ascorbic acid results in the production of a mineralized calcified matrix with sparsely distributed condensed nodules containing high levels of calcium [6]. DPSCs are located in the relatively soft environment of the pulp and differentiate into odontoblasts which produce mineralized matrix-dentin. Changes in the pulp as result of tooth damage or bacterial infection can trigger calcification of the pulp and effectively render the tooth dead. This transformation correlates with the influx of inflammatory cytokines and the change in the mechanical properties of the pulp [13,14]. Obviously changes in the chemical and physical properties of the environment induce signal transduction events leading to differentiation.

The use of materials whose mechanical properties can regulate or direct stem cell proliferation and differentiation offers exciting applications. Engler et al [2,3] used substrates made from polyacrylamide gels overlaid with collagen. In these experiments the mechanical properties were controlled by varying the acrylamide crosslinker and hence the cross-linking density. Although the mechanical properties of the collagen coating was not directly measured, cell differentiation was shown to correlate with the mechanical properties of the acrylamide gels obtained from bulk measurements, which were not sensitive to the actual moduli of the sample surface. In this model, the differences of moduli required to affect differentiation were large (1kPa- 100kPa), and may have been much larger than the surface deviations. In addition, both chemical and mechanical induction of differentiation was used.

In this paper, we introduce the use of polybutadiene (PB), which is a biocompatible, synthetic rubber material similar to the natural rubber gutta-percha, material used in dentistry to fill root canals. PB can be synthesized in the monodisperse form, with high molecular weight chains, (Mw= 205K, Mw/Mn= 1.49) such that Rg = 135Å. Since PB is a rubber under ambient condition (Tg= -95°C), the effect of surface interactions on PB films of varying thickness can be probed without the high temperature conditions described in the previous literature [4,5]. Furthermore, when annealed under these conditions, X-ray reflectivity in conjunction with atomic force microscopy, have shown that the films produced are extremely uniform with an RMS roughness less 6Å [4]. Hence PB provides the ideal substrate for probing only the effects of variable substrate mechanics, without any chemical or topographical variations, on biological systems where the operational temperature range is restricted to 37°C. Here we describe the use of specially processed PB films whose surface properties and thickness can be characterized with nanometer precision, to produce substrates with controlled mechanical response, in order to investigate to what degree DPSC can sense the mechanics of the underlying substrates and determine whether a critical value exists for the modulus above which the cells will spontaneously differentiate.

Materials and Methods

Sample preparation

To prepare various thicknesses of PB spun-cast films, monodisperse PB (Mw=205,800, Mw/Mn = 1.49; Scientific Polymer Products, Inc.) was dissolved in toluene (certified A.C.S.; ACROS) at different concentrations (w/v); 3, 5, 10, 20 and 25 mg/mL and then spun-cast at 2,500 rpm onto HF-etched wafers (Wafer World Corporation, West Palm Beach, FL) to produce of PB films with thicknesses of 200, 500, 1500, 2000 and 3000 Å as measured by ellipsometry. PB films were then annealed in the ultra-high vacuum oven at 10-8 torr at 150°C for 24 h to remove the residual solvent and relax the polymer structure.

Cell isolation and culture

Dental pulp stem cells (DPSCs) strain AX3 were isolated from the third molar teeth (IRB #20076778) as previously described [11] and were grown in α–MEM media (Invitrogen) supplemented with 10% fetal bovine serum, 200 μM L-ascorbic acid 2-phosphate, 2 mM L-glutamine, 100 units/ml penicillin/ 100μg/ml streptomycin, and 10mM β-glycerophosphate. For chemical induction of osteogenesis/ odontogenesis, medium was supplemented with 10-8 M Dex (Sigma- Aldrich). The cells were grown at 37°C in 5% CO2.

Cell plating and proliferation

DPSCs were harvested, counted and then plated on tissue culture polystyrene plastics (TCP) and PB films at a density of 5,000 cells cm2. For the study of cell proliferation, the cell number was determined on day 3, 5 and 7 using a Hemacytometer (Hausser Scientific).

Shear modulation force microscopy (SMFM)

The moduli of PB films and DPSCs cultured on PB films were measured by atomic force microscopy (AFM, Dimension 3000; Digital Instruments, Santa Barbara, CA) using shear modulation force microscopy (SMFM). The experimental setup of the SMFM method was described in earlier articles [15,16]. Briefly, the AFM tip, applied to indent the surface, was laterally modulated and the mechanical response was fed into a lock-in amplifier and recorded. During the measurement, a normal indenting force of 25nN was exerted by the cantilever to maintain tip-surface contact, while a sinusoidal drive signal was applied to the piezo controlling the cantilever to induce a small oscillatory motion. The lateral deflection (response) amplitude of the cantilever was measured against the drive amplitude, both in mV, the response amplitude, therefore, being proportional to drive amplitude.

The moduli of films with different molecular weight were obtained using Bruker ICON AFM on the Peak Force QNM mode. Peak Force QNM (Quantitative NanoMechanics), an extension of Peak Force Tapping mode, enables quantitative measurement of nano-scale material properties such as modulus, adhesion, deformation and dissipation. Peak Force Tapping mode, the core technology behind Peak Force QNM performs a very fast force curve at pixel in the image. The peak interaction force of each of these force curves is then used as the imaging feedback signal. Peak Force Tapping mode modulates the Z-piezo at ~2kHz with a default Peak Force Amplitude of 150 nm. Analysis of force curve data is done on the fly, providing a map of multiple mechanical properties (Adhesion, Modulus, Deformation, and Dissipation) that has the same resolution as the height image.


Instrumented indentation was performed using a nanoindenter (Micro Materials Nano Test) with diamond spherical indenters. Depth controlled indentations were performed with the maximum depth set from 35 to 60 nm depending on the sample, and represented no more than 30% of the total film thickness. A 5μm radius of curvature conical indenter was used on the film samples, and a 200μm radius spherical indenter used on the bulk sample. Film substrates were mounted with an epoxy adhesive to aluminum sample holders for testing. Load-displacement data was obtained using a constant loading and unloading rate of 1μN/s. Typical maximum loads reached were on the order of 50-200μN. A 5 second hold segment was taken at the maximum displacement before unloading to take into account creep effects. On each sample, ten identical indentations were performed with a minimum spacing interval of 50μm between successive indents to ensure no interaction between residual plastic deformations. To determine elastic modulus of the films, the slope (S) of the initial unloading data in conjunction with the spherical contact area is used in the conventional Oliver and Pharr method [17].

Staining and laser scanning confocal microscopy (LSCM)

Immunocytochemistry was used to stain actin filaments and nuclei. Cell s were washed with Ca, Mg-free phosphate buffered saline, pH 7.4 (PBS), fixed with 3.7% formaldehyde for 15 minutes, and permeabilized with 0.4% Triton X-100 in PBS. To visualize actin filaments the cells were incubated with 1:200 dilutions of Alexa Fluor 488 Phalloidin (Invitrogen, Carlsbad, CA). After 2 x 5 minute washes in 0.05% Tween 20 in PBS propidium iodide (Sigma Chemical Co., St. Louis, CA) was added to visualize nuclei. Samples were analyzed and images captured using a Leica TCS SP2 LSCM (Leica micro-system Inc., Bannockburn, IL).

Scanning electron microscopy and energy dispersive X-ray (SEM/EDX)

Surface morphologies of PB films were characterized using scanning electron microscopy (SEM, LEO1550) at 20 kV acceleration voltage and 10 mm working distance. Samples were prepared by gently washing with deionized water to remove soluble salts from media and air-drying for 1 day. Samples were sputter-coated with gold for 15seconds and loaded on aluminum stubs for SEM imaging. The elemental compositions of deposits on PB films were determined by using Energy Dispersive X-ray spectroscopy (EDX) in conjunction with SEM.

Real time-polymerase chain reaction

Total RNA was isolated using the RNeasy Mini kit (Qiagen, Valencia, CA) as described by the manufacturer. RNA concentrations were determined by absorbance at 260 nm; only samples with a ratio of OD 260:280>1.9 were used. For mRNA quantitation, 5 μg of total RNA was separately reverse transcribed with 200 U Superscript II and 250 ng random primers (Invitrogen, Carlsbad, CA) as recommended by the manufacturer. Real-time PCR was performed in triplicate with 4 pmol amplifying primers and 8 μlc DNA in a final volume of 20 μL using QuantiTect SYBR Green PCR Kit (Qiagen, Valencia, CA). Real-time PCR was carried out and analyzed using MJ Research Opticon System (MJ Research, Waltham, MA), a service provided by the DNA sequencing facility of Stony Brook University. The expression levels of bone alkaline phosphatase (ALP) and osteocalcin (OCN) were determined using gene specific primers (ALP- forward primer: 5’-GTACTGGCGAGACCAAGCGCAA-3’ and reverse primer: 5’-ACCCCACACAGGTAGGCGGT-3’; OCN- forward primer: 5’-ATGAGAGCCCTCACACTCCTCG-3’ and reverse primer: 5’-GTCAGCCAACTCGTCACAGTCC-3’). 18S served as control using forward primer: 5’-GTAACCCGTTGAACCCCATT-3’ and reverse primer: 5’-CCATCCAATCGGTAGTAGCG -3’.

Results and Discussion

Control of thin film moduli via surface confinement

In Figure 1A we plot the relative moduli of the annealed PB (Mw=205K) films for different film thicknesses using the SMFM method where we can see that the moduli decrease with increasing film thickness. The same samples were also measured with a Hysitron nanoindentor (Figure 1B), and in Figure 1C we plot the SMFM vs the nanoindentation measurements where we find a linear relationship, indicating that the relative values between samples are similar, despite differences in the techniques. The accuracy of the slope, m=1.20(5), allows us to calibrate the SMFM measurements relative to those obtained with the Hysitron calibrated nanoindentor, where we find that the moduli of the PB films, of Mw=205K, are in the range of 1-8 MPa for films ranging from 20nm-300 nm. It has previously been demonstrated by Li et al. [18] that the effective viscosity of a polymer film at a temperature above its glass transition, Tg, increases exponentially with decreasing film thickness. The phenomena was explained in terms of the formation of an adsorbed brushy state at the interface with the substrate, which in turn can trap other chains, not in contact with the surface, slowing down, but not stopping their ability to diffuse or flow in a manner similar to a “strangulated” gel, where, the thickness of this gel layer had been shown by Zheng et al [5] to scale with Rg, the radius of gyration of the polymer. In order to determine whether the decrease in modulus with thickness is correlated to the polymer structure, as opposed to the proximity of the hard Si interface, we also tested films of two other molecular weights, whose properties are listed in Table 1. In Figure 2A we plot the surface modulus, M, vs film thickness, t, for the film of three molecular weights, where the solid lines in each case are fits to an exponential function of the form;