TISSUE ENGINEERING: Part A Volume 20, Numbers 11 and 12, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/ten.tea.2013.0350

Subtle Changes in Surface Chemistry Affect Embryoid Body Cell Differentiation: Lessons Learnt from Surface-Bound Amine Density Gradients Bahman Delalat, PhD, Renee V. Goreham, BSc, Krasimir Vasilev, PhD, Frances J. Harding, PhD, and Nicolas H. Voelcker, Dr rer nat

Advanced approaches to direct the differentiation of embryonic stem cells are highly sought after. The surfacebound chemical gradient format is a powerful screening approach that can be deployed to study changes in stem cell behavior as a function of subtle changes in surface chemistry. Here, we investigate the spontaneous differentiation of cells derived from differentiating mouse embryoid body (mEB) cells into endoderm, mesoderm, and ectoderm following culture on surface-bound gradients of chemical functional groups in the absence of differentiation-biasing bioactive factors. Gradients were created using a diffusion-controlled plasma polymerization technique. The generated coating ranged from hydrophobic 1,7-octadiene (OD) plasma polymer at one end of the gradient to a more hydrophilic allylamine (AA) plasma polymer on the opposite end. The gradient surface was divided into seven equal regions of progressively increasing AA plasma polymer content and mEB cell response within these regions was compared. Cells adhered preferentially to the central regions of the gradient; however, cell proliferation increased toward AA-plasma-polymer-rich end of the gradient. Variation in the expression of germ layer markers was noted across the gradient surface. High AA:OD plasma polymer ratios triggered cell differentiation toward both mesoderm and ectoderm. Expression of tissue-specific markers, in particular, KRT18, AFP, and TNNT2, was strikingly responsive to subtle changes in surface chemistry, exhibiting vastly different expression levels between adjacent regions. Our results suggest that the surface-bound gradient platform is well suited to screening surface chemistries for use in the field of stem cell technologies and regenerative medicine.

Introduction

T

he repair of damaged or diseased tissue is a major focus of research in which biomaterials research plays an increasingly important role.1 Over the last two decades, stem cell technology has shown promise as a way to prepare cost-effective and functional substitutes for damaged tissue.2,3 One key challenge in this field is the design of optimal biomedical scaffolds with chemical properties that can direct stem cell fate, so that sufficient numbers of functional cells can be produced to replace human tissue.4 More and more embryonic stem (ES) cell–derived therapies are being brought into clinical trials.5 These rely heavily on the availability of materials with precise surface functionality, which can regulate cellular responses from initial attachment and migration through to differentiation and formation of new tissue.6 Efficient screening of the cell response to material surface chemistry underpins the discovery effort

of optimal conditions to support targeted cell differentiation and survival. High-throughput screening (HTS) technologies including cell microarrays have contributed significantly to modern biological discoveries pertaining to the control of cell behavior.7–9 Among the currently available HTS methods are surface-bound gradients, which allow deeper insight into biological processes at the interface between natural and synthetic materials.10,11 Plasma polymer deposition is a very attractive surface modification technique that can be applied to any surface without changing its topography.12 Plasma polymers provide smooth and pinhole-free coatings and allow control of surface chemistry independent of the substrate.13–15 This technique can also be used to prepare a large variety of surface chemistries and has recently been shown to be suitable for clinical applications.16,17 Keratinocytes and human limbal epithelial cells have been grown on plasma-polymerized acrylic acid and both successfully transplanted to wound beds.18,19

Mawson Institute, University of South Australia, Mawson Lakes, Australia.

1715

1716

The early differentiation of murine ES (mES) cells can be initiated by embryoid body (EB) formation.20,21 The scaledup production of differentiated cell populations from EBs has been demonstrated.22,23 However, the heterogeneity of differentiated cell populations derived from EB is a limitation for its application in cell therapies.24,25 Identifying a suitable surface to direct differentiation and produce more homogeneous cell populations would be a key enabler to progress ES-cell-derived therapies to the clinic.26 Addressing this problem, we employed a surface-bound gradient format to screen for surface properties that supported the differentiation of cell populations derived from mouse EB (mEB) cells to specific lineages. An amine plasma polymer gradient was produced by changing the composition of the plasma gas feed from 1,7-octadiene (OD) to allylamine (AA) as a glass coverslip was progressively moved under a slot in a mask. Subsequently, we analyzed mEB cell response to the gradient surfaces by monitoring cell attachment and expression of specific germ layer markers: nestin (NES) and beta III tubulin (TUBB3) as markers of ectodermal differentiation; kinase insert domain protein receptor (KDR), cardiac troponin T (TNNT2), and brachyury (T) as markers of mesoderm formation; and alpha-fetoprotein (AFP) and cytokeratin 18 (KRT18) as markers of endodermal differentiation. We investigated cell behavior as a function of surface chemistry in terms of cell attachment, proliferation, and differentiation, as a precursor to more complex biological assessment. Materials and Methods Gradient generation via plasma polymerization (preparation of OD-AA plasma polymer gradient surface)

Surface gradients progressing from OD plasma polymer to AA plasma polymer were prepared on 13-mm glass coverslips by plasma copolymerization of the two monomers (obtained from Sigma-Aldrich) through a 1-mm slot in a mask passed over the moving substrate. The plasma-deposition apparatus has been described previously.15,27 The slope and shape of the gradient was controlled by the rate at which the OD:AA monomer ratio was changed. The initial flow rate of OD was 10 standard cubic centimeters per minute (sccm) that was linearly reduced to 0 sccm between the 6- and 12-mm positions along the gradient. The flow rate of AA was linearly increased from 0 sccm at 6 mm to 12 sccm at 10 mm. The plasma was excited using a 13.56 MHz radiofrequency generator. The power of deposition was 10 W and remained constant during the entire deposition process. Surface characterization

X-ray photoelectron spectroscopy (XPS) analysis was used to determine the surface composition of the OD-AA plasma polymer gradients. XPS spectra were recorded on a Specs SAGE XPS spectrometer using a Mg Ka radiation source (hm = 1253.6 eV) operated at 10 kV and 20 mA. Elements present in a sample surface were identified from the survey spectrum recorded over the energy range 0–1000 eV at a pass energy of 100.0 eV and a resolution of 0.5 eV. The areas under selected photoelectron peaks in a wide-scan spectrum

DELALAT ET AL.

were used to calculate percentage atomic concentrations (excluding hydrogen). High-energy resolution (0.1 eV) spectra were then recorded for pertinent photoelectron peaks at a pass energy of 20.0 eV to identify the possible chemical binding environments for each element. All binding energies were referenced to the C1s neutral carbon peak at 285.0 eV to compensate for the effect of surface charging. The XPS analysis area was circular with a diameter of 0.7 mm. The processing and curve-fitting of the high-energy resolution spectra were performed using CasaXPS software. The spectrometer was calibrated using the Ag3d 7/2 peak position at 368.0 eV for a Ag sample after Ar sputtering for at least 15 min until the surface carbon contamination level was < 5% (measured from the C1s core-level peak). Cell culture

Cell responses of both D3 (ATCC CRL1934) and ESE14TG2a (ATCC CRL1821) mES cells were investigated. E14TG2a cells are derived from genetic background (129/ ola mouse strain) distinct from that of D3 (127/Sv), and were employed to further verify conclusions drawn from observations of the D3 cell line. mES cells were cultured on a feeder layer of gamma-irradiated STO1 cells and incubated at 37C with 5% CO2 in advanced DMEM supplemented with 10% ES cell qualified fetal bovine serum (Invitrogen), 10 ng/mL leukemia inhibitory factor (Millipore), 0.1 mM b-mercaptoethanol (Sigma), 2 mM L-glutamine, 100 U/mL penicillin, and 100 g/mL streptomycin (Invitrogen), for 2–3 days until they were 70–80% confluent. For mEB cell formation, mES cells were dissociated with trypsin/EDTA solution and plated at density of 2.5 · 104 cells/cm2 onto 10-cm nonadhesive petri dishes with EB medium that was composed of advanced DMEM, 15% fetal bovine serum (Invitrogen), 0.1 mM b-mercaptoethanol, 2 mM L-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin. After 2 days in culture, mES cells aggregated into mEBs.28 Culture of mEBs continued for a further 6 days before single-cell suspensions of mEB cells were generated using 0.05% trypsin–0.53 mM EDTA (Sigma) (Fig. 1). The cells were passed through a nylonmesh filter with pore size of 100 mm (BD Falcon) to obtain a single-cell suspension. The viability of mEB cells was assessed before seeding onto the OD-AA plasma polymer gradient using Trypan blue staining. mEB cell immobilization on gradient surfaces. Prior to incubation of mEB cells with the OD-AA plasma polymer gradient surface, the OD-AA plasma polymer gradient coverslip glasses were washed with copious amounts of sterile Dulbecco’s phosphate-buffered saline solution (PBS) (Sigma) to remove any excess residuals. Then, the coverslips were sterilized with 200 U/mL penicillin, 200 mg/mL streptomycin, and 500 ng/mL amphotericin B (Invitrogen) in sterile PBS for 4 h and were washed three times in sterile PBS. Each coverslip was placed in a 24-well plate (Nunc) and seeded with cells at density of 1 · 104 cells/mL in fresh cell culture medium. As a control, cells were also plated onto sterile, round, 13-mm glass coverslips at the same density. All mEB cells were cultured for 7 days in EB medium at 37C in a humidified atmosphere with 5% CO2. To study the effects of the OD-AA plasma polymer gradient on EB cell behavior, no

SCREENING OF EB CELL BEHAVIOR ON PLASMA POLYMER GRADIENTS

FIG. 1. Schematic of studying mouse embryoid body (mEB) cell differentiation on 1,7-octadiene–allylamine (ODAA) plasma polymer gradient surfaces: formation of EB from murine embryonic stem (mES) cells (8 days of culture); trypsinization of mEBs to obtain individual mEB cells; cell attachment to gradient substrates; proliferation of mEB cells on gradient substrates for 7 days, followed by immunofluorescence analysis of germ layer marker expression. Color images available online at www.liebertpub.com/tea other bioactive factors were added to the culture medium. The growth medium was replaced every second day. Cytotoxicity assays of mEB cells on gradient surfaces were performed by live-dead staining using a final concentration of 15 mg/mL fluorescein diacetate (FDA; Invitrogen) and 5 mM propidium iodide (PI; Sigma) for 3 min at 37C. Unattached cells were removed by rinsing with PBS for *12 h after seeding. Gradients were divided into seven equally broad regions of slowly increasing AA plasma polymer deposition (positions from OD-plasma-polymer-rich end of the gradient: 0.0–1.85 mm = region 1, 1.85–3.7 mm = region 2, 3.7–5.55 mm = region 3, 5.55–7.4 mm = region 4, 7.4– 9.25 mm = region 5, 9.25–11.1 mm = region 6, and finally 11.1–13.0 mm = region 7). Brightfield images of each region were acquired using an inverted light microscope (Olympus CK2) equipped with a Nikon Digital Sight DS-SM digital camera (Nikon). Cell adhesion on the gradient surfaces was evaluated by counting the number of attached cells in each region on each day for the first week of culture. Experiments monitoring cell proliferation were repeated six times in independent experiments. mEB cell morphology and differentiation

Cell morphology was investigated in each region across the gradient after 7 days. Differentiation of mEB cells was evaluated by immunocytochemistry techniques; cells were

1717

assayed for their capacity to differentiate into endoderm, mesoderm, or ectoderm by expression of specific markers via immunostaining as described later. Cultured cells were gently washed with PBS to remove culture media and fixed in 4% paraformaldehyde solution (Electron Microscopy Science) for 10 min. Cells were permeabilized with 0.1% Triton X-100 in PBS at room temperature for 5 min and then blocked with 10% serum in PBS—from the species in which the secondary antibody was raised—for 1 h. Primary antibodies used for immunocytochemical staining were as follows: KRT18 (Santa Cruz; diluted 1:200) against a cytoskeleton protein present in hepatocyte and epithelial cells; AFP (Santa Cruz; diluted 1:200) as an early hepatocyte marker; T (Santa Cruz; diluted 1:100) as an early indicator of mesoderm; KDR (Santa Cruz; diluted 1:200) as an early hemangioblast marker; TNNT2 (Santa Cruz; diluted 1:200) as a cardiomyocyte marker; NES (Santa Cruz; diluted 1:200) as a neural stem cell marker; TUBB3 (Abcam; diluted 1:100) as a neuron-specific marker, and POU5F1 (Santa Cruz; diluted 1:200) as a pluripotency marker. All antibodies were diluted in 1% bovine serum albumin (BSA; Sigma) in PBS. After incubation with the primary antibodies overnight at 4C and washing three times with PBS, the corresponding fluorescence-labeled secondary antibody (Santa Cruz; diluted 1:100 in PBS) was added for 1 h at room temperature. Negative controls were performed by omitting the primaryantibody-labeling step from the procedure, which in all cases resulted in a complete loss of signal from fluorescencelabeled secondary antibodies. Finally, the nuclei were counterstained with 0.2 mg/mL Hoechst 33342 (Invitrogen) in PBS for 10 min and rinsed with PBS and mounted. Gradients were divided into seven equal zones of gradually increasing AA plasma polymer deposition as per ‘‘mEB Cell Morphology and Differentiation’’ section, and cell behavior was compared between each zone. Images were obtained on a Nikon Eclipse Ti-S inverted fluorescence microscope equipped with a Nikon Digital Sight DS-2MBWc digital camera and NIS-Elements imaging software. Marker expression levels (% positive cells) were assessed by identifying each cell nuclei and scoring the colocalized marker expression (nuclear or cytoplasmic as appropriate) as positive or negative. Analyses were performed for each marker independently. Immunofluorescence analysis was repeated in three independent experiments for each marker. The same experiment was performed to assess the identity of the mEB cell population before seeding on OD-AA plasma polymer gradient surfaces. Marker expression scores of a subset of differentiation markers (NES, T, KRT18, and KDR) were verified using the Operetta High Content Imaging System (PerkinElmer), which combines fluorescence microscopy with automated image acquisition and quantitative analysis. Analysis was performed at timepoints specified in the text (after 0.5, 4, and 7 days of culture on gradient surfaces). To count total cell numbers, nuclei were counterstained with Hoechst 33342. Cells were washed once with PBS and mounted for imaging. Images of the different dyes were acquired on using a 10 · LWD objective in wide-field mode in combination with standard filters for Hoechst 33342 (excitation filter: 360–400 nm; emission filter: 410–480 nm), FITC (excitation filter: 460–490 nm; emission filter: 500–550 nm), and PE (excitation filter: 560–580 nm; emission filter: 590–640 nm).

1718

Both gamma and contrast settings were optimized to achieve optimal images, omitting any background and autofluorescence. The laser autofocus was applied and images were acquired across the entire gradient. Data analysis for percentage of differentiated cells was performed by using the Harmony High-Content Imaging Software (Perkin Elmer; version 3.1). For quantitative analyses, individual cells were segmented based on the Hoechst 33342 nuclear stain using the ‘‘Find Nuclei’’ building block in the Harmony high content imaging and analysis software (PerkinElmer) and the intensity of each marker was quantified within the Hoechst-defined boundaries for each cell. The ‘‘Select Population’’ module of Harmony was used to identify the subpopulation of differentiated cells based on fluorescence intensity thresholds and thus determine the percentage of marker-positive cells within the total cell population within each region (see Supplementary Fig. S5B, C for an example; Supplementary Data are available online at www.liebertpub.com/tea). The average and standard error mean were calculated from independent experimental triplicates. To investigate the relative extent of differentiation toward ectoderm and mesoderm germ layers, cells were costained for NES together with KDR and together with T after 4 and 7 days of culture on the plasma polymer surfaces.

DELALAT ET AL.

FIG. 2. Atomic concentration of nitrogen across the gradient as determined from X-ray photoelectron spectroscopy measurements (n = 3).

Statistical analysis

One-way ANOVA analysis was carried out to quantify cell adhesion differences between the seven gradient regions specified in ‘‘Cell Culture’’ section, followed by post hoc analysis of means using the Tukey test. A p-value of < 0.05 was considered statistically significant. Statistical analysis was performed using KaleidaGraph software (version 4.0; Synergy Software). Linear regression of KDR expression was performed using Excel (2010; Microsoft). Results Gradient generation and characterization

Analysis of the chemical composition of the gradient was conducted using XPS. Figure 2 shows the atomic concentration of nitrogen across the surface quantified from the survey spectra. At position 1, 0.9 mm from the leading edge of the coverslip, there was no detectable nitrogen signal, consistent with surface modification deposition carried out in an atmosphere of pure OD monomer. The nitrogen concentration linearly increased across the gradient and reached 12% in region 7. At this end, surface deposition is carried out in an atmosphere of pure AA monomer. Our previously published analysis of these gradient surfaces has shown that changes in amine group density translate into a charge density on the surface, creating a surface with a potential spanning from 0 to 80 mV.30 Characterization of wettability indicates a close to linear decrease in water contact angle across the gradient surface, from 76 to 56.15 Cell response to OD-AA plasma polymer gradients

mEB cell attachment at 12 h postseeding on the OD-AA plasma polymer gradient is shown in Figure 3A. The density of adherent cells was highest on region 4 (corresponding to the center of the gradient) and lowest on region 1. Cell attachment increased from region 1 toward region 4, and

then decreased toward the AA-plasma-polymer-rich end of the gradients (region 7). In contrast, only a few cells were loosely attached to a glass coverslip without any plasma polymer, and after 2–3 days, these cells began detaching from the glass surface. We investigated mEB cell proliferation on the gradient surfaces over 1 week under cell culture conditions (Fig. 3B). Across the gradient surface, cell density increased over the course of 7 days. The greatest increase in cell density was observed on region 7. Cells proliferated more rapidly on the AA-plasma-polymer-rich regions (6 and 7) compared with the regions with lower density of AA plasma polymer; the cell doubling time reduced from 3.6 days in region 4 to 1.7 days in region 7. Cell density in region 7 was significantly greater than observed on the other regions after 7 days in culture ( p £ 0.05). On regions 6 and 7, cell density increased seven- and ninefold, respectively, over the course of the experiment. Cell proliferation increased with nitrogen content in the underlying plasma polymer. To assess the biocompatibility of the surface, cell viability and morphology were investigated using FDA and PI staining. In general, extensive cell spreading indicates perfect cellsurface contacting properties, while rounding up of cells on a surface reflects weak cell-surface interaction, often followed by cell cycle arrest, cell detachment, and death. Cells within region 1 (OD plasma polymer rich) were mostly rounded after 7 days in culture. In comparison, cell spreading was more extensive over the plasma polymer containing a high AA plasma polymer density. Cell viability across all regions was high after 7 days of culture; only a few dead cells were observed, confined to regions 1 and 2 (Fig. 3C). Differentiation of EB cells into three germinal layers

Next, we studied mEB cell differentiation trajectories on the OD-AA plasma polymer gradients. Cells cultured on the

SCREENING OF EB CELL BEHAVIOR ON PLASMA POLYMER GRADIENTS

1719

FIG. 3. (A) Total cell attachment (averaged per sample) determined by a cell count in each region of the OD-AA plasma polymer gradient surface after 12 h of incubation (n = 6). (B) Cell proliferation (averaged per mm2) determined by cell counts at each region every 24 h over the course of 7 days. Cell seeding density was 1 · 104 cells/mL. Error bars correspond to standard error (n = 6). (C) mEB cell viability on the OD-AA plasma polymer gradient after 7 days in culture. Live cells were stained with fluorescein diacetate (green) and dead cells were stained with propidium iodide (red). One representative fluorescence micrograph per region is shown (scale bar = 100 mm). Color images available online at www.liebertpub.com/tea

OD-AA plasma polymer gradient gradually developed various distinctive characteristics: fibroblast-like, triangular, polygonal, cuboidal and squamous shapes and dendritic morphologies were observed at 7 days. To investigate the formation of endoderm, mesoderm and ectoderm germ layers from mEB cells on the OD-AA plasma polymer gradients, expression of differentiation markers was monitored by immunofluorescence over a period of 7 days culture on the plasma polymer gradients. First, the expression of marker proteins AFP and KRT18 (endoderm); KDR, TNNT2, and T (mesoderm); NES and TUBB3 (ectoderm); and POU5F1 (OCT4, pluripotency associated) was determined in mEB single-cell suspensions before seeding on the gradient surface at day 0 (Fig. 4A, B). About 69% – 1.5%, 25% – 2.1%, and 17% – 1.3% stained positive for T, POU5F1, and KDR, respectively. AFP, KRT18, TNNT2, NES, and TUBB3 were not expressed in individual mEB cells at day 0. Twelve hours postseeding, when the cell adhesion process would be complete, investigation of cell expression of T across the gradient was revealed to be slightly polarized toward amine-rich regions of the gradient (Supplementary Fig. S1). No KDR staining was observed after cell attachment to the test substrates (data not shown). The majority of cells observed at this timepoint were POU5F1 negative. However, a few POU5F1-positive cells were detectable toward the OD end of the gradient, particularly in region 1 (Fig. 5). Loss of POU5F1 expression in regions 2–7 indicated that increasing AA plasma polymer content supported further differentiation of the mEB cells. Attempts to

characterize the mEB cell commitment to endodermal layer by immunofluorescence techniques showed that cells sporadically expressed AFP and KRT18 on the surfaces (Fig. 6 and Supplementary Fig. S2), seemingly unaligned with the gradient in amine functionality. We observed that cells within regions 1, 5, and 6 were positive for AFP, and cells within regions 2 and 4 were positive for KRT18 consistently across independent experiments using the D3 line. There were slight differences in the pattern of KRT18 expression observed on the surfaces using the E14TG2a cell line, but endodermassociated gene expression attained on the gradient surfaces for both cell lines was low (Supplementary Fig. S2). Immunofluorescence analysis of mesoderm markers was used to further characterize the mEB cell differentiation pattern. KDR-positive cells were observed across the gradient on all regions (Fig. 7A, D and Supplementary Fig. S3) at the end of the 7-day culture period. The level of KDR was dependent on the position of EB cell attachment on the plasma polymer gradient. For instance, in region 1 where AA plasma polymer content was almost zero, there were 10% – 5.8% KDR-positive cells using the D3 cell line, increasing to 52% – 7.7% on region 7. An analysis of the relationship between % KDR-expressing cells and the nitrogen signal in the underlying plasma polymer revealed an almost linear correlation (R2 = 0.96), suggesting that KDR upregulation is mediated by the surface density of amine groups on the substrate. A high number of cells expressing TNNT2, 80% – 2.8% and 87% – 2.5%, was detected in regions 3 and 6 of the gradient, respectively, but on

1720

DELALAT ET AL.

FIG. 4. Immunofluorescence staining revealed the expression of proteins associated with three early germ layers in individual mEB cells before seeding on gradient surfaces. (A) Cells were labeled for kinase insert domain protein receptor (KDR; green) and brachyury (T; green), indicating mesodermal layer markers, and POU5F1 (green) as a pluripotency marker. Blue signals indicate nuclei of mEB cells that were stained using Hoechst 33342 (scale bar = 100 mm). (B) Quantification of marker protein expression in mEB cells before seeding on OD-AA plasma polymer gradient surfaces. The error bars represent the standard error (n = 3). Color images available online at www.liebertpub.com/tea

the other regions TNNT2 expression was notably absent (Fig. 7B, D). T-positive cells rose region 7 of the gradient (Fig. 7B, C and Supplementary Fig. S4D). To probe the onset of neural differentiation in the culture (as a measure of ectoderm differentiation), expression of NES activity and TUBB3 was examined (Fig. 8 and Supplementary Figs. S3–S7). Immunofluorescence analysis showed that the percentage of NES-positive cells increased from *10% in region 1 to > 70% in region 7 using the D3 cell line (Fig. 8C). Similarly dominant populations of NESexpressing cells were observed in region 7 using the E14TG2a line (Supplementary Fig. S6). For both NES and TUBB3, a positive correlation with amine functional group surface density was observed in regions 3–7. Interestingly, staining for either neurectodermal marker was absent in region 2 of the gradient for the D3 line. The response to AA plasma polymer content appears to be generalizable across ES cell lines, since similar trends in surface marker expression were observed between the E14TG2a and D3 cell lines during differentiation on the gradient substrates. There are some features of the marker expression pattern, such as the distinct drop in NESexpressing population in region 2 observed using the D3 cell line, which did not occur in E14-TG2a cells (Supplementary Figs. S3–S7). In both cell lines, the differentiation pattern

across the gradient surfaces for early mesodermal (T) and ectodermal (NES) genes fully emerged after 4 days on the gradient surfaces (Supplementary Figs. S4–S7). However, cells that express markers associated with later-differentiation stages (KDR and KRT18) were not observed in appreciable numbers until the 7 days of time point. An obvious candidate to link cell response to underlying substrate chemistry is through protein adsorption, for example, from the culture medium or as component of the extracellular matrix secreted by the cells. Protein adsorption is implicated in mediating cell adhesion in nitrogencontaining plasma polymers.27 Precedent exists for significant variation between the adsorption profiles of proteins on plasma polymer surface chemistry gradients.1 Investigation of the adsorption of serum proteins on the gradient substrate was carried out alongside BSA and IgG. Each protein (or protein mixture) exhibited a different trend in adsorption across the gradient (Supplementary Fig. S8). However, in all cases, protein adsorption was greatest in region 7, at the amine-rich end of the gradient. Discussion

Cell culture substrates can be readily modified to display specific chemical functional groups using an easy one-step

FIG. 5. Immunofluorescence staining (green) corresponding to POU5F1 expression in mEB cell cultures 7 days postseeding on OD-AA plasma polymer gradient surface (scale bar = 100 mm). One representative fluorescence micrograph per region is shown. Cells were counterstained with Hoechst 33342 (blue) to reveal nuclei. Color images available online at www.liebertpub.com/tea

SCREENING OF EB CELL BEHAVIOR ON PLASMA POLYMER GRADIENTS

1721

FIG. 6. Immunofluorescence staining for the expression of proteins associated with endodermal layer. (A) Alpha-fetoprotein (AFP; green) and (B) cytokeratin 18 (KRT18; green) representative fluorescence micrographs of mEB cells on the seven regions of the OD-AA plasma polymer gradient surfaces are shown 7 days after seeding (scale bar = 100 mm). Blue signals indicate nuclei of mEB cells (stained using Hoechst 33342). Cell seeding density was 1 · 104 cells/mL. (C) Quantification of endodermal layer differentiation marker protein expression in mEB cells at each region of the gradient. The error bars represent the standard error (n = 3). Color images available online at www.liebertpub.com/tea

plasma polymerization process. Surface-bound chemical gradients are valuable tools for studying and guiding cellular responses, such as attachment, proliferation, and differentiation. In the present study, an OD-AA plasma polymer gradient was generated on coverslip glass by our previously reported method of changing the composition of the monomer gas feed as the surface is progressively moved under a slot in a mask.15,28,29 XPS measurements taken along the OD-AA plasma polymer gradient surface confirmed that the percentage of nitrogen across the gradient (a surrogate for the amine group density30) increased linearly toward the AA plasma polymer end of the gradient. The gradient surface was then used as a cell culture substrate. Having successfully fabricated OD-AA plasma polymer gradients, mEB cell behavior was then evaluated from the viewpoint of attachment, proliferation, and expression of markers associated with differentiation. Protocols for the targeted differentiation of ES cells to specific cell types often commence with EB formation.20,21 In addition, past analyses of the influence of surface chemistry on ES cell differentiation have been limited by the formation of cell aggregates on the test substrate, making it difficult to parse out the effects of the underlying surface after the formation of multilayered cell colonies.29 Here, mES cell differentiation was initiated in mEB, and mEBs were subsequently trypsinized to obtain a single-cell suspension. The partly

differentiated mEB cells formed a monolayer on the test substrate, allowing cell attachment, proliferation, and spontaneous differentiation in response to surface properties to be examined without the confounding influence of colony formation. After 1 day of culture, mEB cells attached best to region 4 at the center of the gradient. On region 1, cell attachment was very low, as expected on a hydrocarbon-rich surface.31,32 Unexpected, cell density decreased also toward the AA-plasma-polymer-rich end of the gradient. The results of mEB cell attachment on OD-AA plasma polymer gradients contrast with those previously reported by us using undifferentiated ES cells, where regions of high amine functional group surface density supported greater ES cell attachment.29 However, decreased cell attachment was observed for both undifferentiated D3 ES cells and partially differentiated mEB cells at the amine-rich end of the plasma polymer gradients. In contrast to cell attachment, the overall trend of cell proliferation correlated with increasing AA plasma polymer density across the gradient, such that after 7 days of culture, region 7 exhibited the highest cell density. Others have also found that amine-functionalized surfaces support cell proliferation.33 Enhanced proliferation may be mediated by surface charge, surface energy,34 and/or protein adsorption to the surface.35,36

1722

DELALAT ET AL.

FIG. 7. Immunofluorescence staining for mesoderm-related markers. (A) KDR (green), (B) TNNT2 (red), and (C) T (green) representative fluorescence micrographs of mEB cells on the seven regions of the OD-AA plasma polymer gradient surfaces are shown 7 days after seeding (scale bar = 100 mm). Blue signals indicate nuclei of mEB cells that were stained using Hoechst 33342. Cell seeding density was 1 · 104 cells/mL. (D) Quantification of mesodermal layer differentiation marker protein expression in mEB cells at each region on the gradient. The error bars represent the standard error (n = 3). Color images available online at www.liebertpub.com/tea Methods to facilitate targeted differentiation of EB cells are highly sought after. In recent years, there have been many reports on ES cell differentiation on defined surface chemistries.37–40 Immunofluorescence staining for specific lineage markers after 7 days of culture on the OD-AA plasma polymer gradients confirmed that changes in surface chemical properties affected mEB cell differentiation, in the absence of a bias through biological growth factors in the cell culture medium. As a general trend, increased AA plasma polymer content was associated with the promotion of differentiation; POU5F1-positive cells were only detected within regions 1 and 2. Levels of expression of both ectoderm (TUBB3 and NES) and mesodermal markers (T and KDR) increased along the axis of the amine gradient, reaching a maximum in region 7. Expression of KDR, in particular, correlated directly with AA plasma polymer content on the gradient. Surprisingly, the expression of endoderm-associated markers KRT18 and AFP and the cardiac-specific marker TNNT2 was found to occur within specific isolated regions of the gradient. For these genes, no

correlation with the amine content in the plasma polymer and gene expression was seen. Our results are consistent with the observations made by Liu et al.,41 who noted a delay in differentiation and simultaneous upregulation of AFP transcription in EB-derived cells cultured on uncharged polyacrylamide hydrogels. In contrast, higher expression of T was found on negatively charged polymer substrates. A report by Ren and coworkers42 indicates that positively charged (amine-terminated) surfaces sustain cell adhesion, viability, migration, and neuronal differentiation of neural stem cells compared with other surface chemistries associated with negative or neutral charge. Intriguingly, we observed that the cell population arising in region 7 of the plasma polymer gradient expressed the primitive mesodermal marker T well beyond the time window in mES differentiation reported previously,20,43 simultaneously expressing high levels of NES. Noting both high cell proliferation in this region of the gradient and an association of NES expression with ‘‘stem cell status’’ both within and outside the ectoderm germ layer,44 we suspect

SCREENING OF EB CELL BEHAVIOR ON PLASMA POLYMER GRADIENTS

1723

FIG. 8. Immunofluorescence staining for ectoderm-related markers. (A) Nestin (NES; red) and (B) beta III tubulin (TUBB3; green) representative fluorescence micrographs of mEB cells on the seven regions of the OD-AA plasma polymer gradient surfaces are shown 7 days after seeding (scale bar = 100 mm). Blue signals indicate nuclei of mEB cells, which were stained using Hoechst 33342. Cell seeding density was 1 · 104 cells/mL. (C) Quantification of ectodermal layer differentiation marker protein expression in mEB cells at each region on the gradient. The error bars represent the standard error (n = 3). Color images available online at www.liebertpub.com/tea

that the amine-rich AA plasma polymer may inhibit terminal differentiation of EB cells. The coincident high expression of TUBB3 in region 7, primarily cited as a neuron-specific marker,45 is more difficult to account for, but has also been reported to be expressed in other cells and tissues.46,47 Further analysis on uniform plasma polymer samples is required to define this cell population. Variation in mEB cell response across the gradient surface may result from the differential adsorption of proteins from culture media to the surface,35,36,48,49 resulting from changes in surface charge across the plasma polymer gradient. Preliminary studies that examine the adsorption of fluorescently labeled serum and model proteins BSA and IgG indicate that each protein (or mixture) generates a unique adsorption profile on plasma polymer gradient surfaces. There are a number of possible mechanisms that can be postulated to explain how different differentiation trajectories could develop on the gradient substrates as a result of protein adsorption. First, growth factors and morphogens that are able to influence differentiation may be pulled down on the substrate from the cell milieu in different ratios.27 Differential protein adsorption may also influence cell behavior via mechanotransduction pathways.50 Noting also that cell adhesion is affected by protein adsorption onto the surface,48,49,51 protein adsorption may also mediate the numbers of cells adhering to a surface, which in turn in-

fluence differentiation through modulating cell–cell communication52 or autocrine factor secretion.53 Localization of specific populations from the initial heterogeneous cell population seeded on the gradient surface may begin the formation of the distinct patterns of gene expression observed, as suggested by the small changes in the proportion of T-expressing cells attaching to each region of the gradient. The present screening study provides evidence that the density of amine functional groups present on culture substrates influence the differentiation trajectory of mES cells. OD-AA plasma polymer gradients were used to assess the response of differentiating mES cells to changes in amine group surface density on the culture substrate. mEB cells derived from differentiating mEB preferentially adhered to central regions of the OD-AA plasma polymer gradient test substrate. Cell proliferation increased with AA plasma polymer density over a 7-day culture period. A positive correlation was found between amine surface density and cell differentiation toward ectodermal and mesodermal fates. In contrast, expression of KRT18, AFP, and TNNT2 did not correlate with amine group surface density, and was observed to be upregulated sporadically across the range of surface group densities tested. Our platform offers the flexibility of assessing stem cell–surface interactions with higher throughput. We believe that such screening platforms

1724

may enable the definition of culture substrates to support the production of highly specific ES-cell-derived populations required for cell therapies, toxicology, and drug screening. They may also aid in the design of advanced biomaterials as scaffolds for tissue engineering. Acknowledgments

Financial support from the South Australian Premier’s Science and Research Fund is kindly acknowledged. The authors thank Marc Cirena and Soraya Rasi Ghaemi for their assistance with constructing figures. Disclosure Statement

No competing financial interests exist. References

1. Zelzer, M., Majani, R., Bradley, J.W., Rose, F.R.A.J., Davies, M.C., and Alexander, M.R. Investigation of cell– surface interactions using chemical gradients formed from plasma polymers. Biomaterials 29, 172, 2008. 2. Godbey, W., and Atala, A. In vitro systems for tissue engineering. Ann N Y Acad Sci 961, 10, 2002. 3. Chen, C., Hu, Z., Liu, S., and Tseng, H. Emerging trends in regenerative medicine: a scientometric analysis in citespace. Expert Opin Biol Ther 12, 593, 2012. 4. Woodfield, T.B.F., Moroni, L., and Malda, J. Combinatorial approaches to controlling cell behaviour and tissue formation in 3D via rapid-prototyping and smart scaffold design. Comb Chem High Throughput Screen 12, 562, 2009. 5. Schwartz, S.D., Hubschman, J.-P., Heilwell, G., FrancoCardenas, V., Pan, C.K., Ostrick, R.M., Mickunas, E., Gay, R., Klimanskaya, I., and Lanza, R. Embryonic stem cell trials for macular degeneration: a preliminary report. Lancet 379, 713, 2012. 6. Genzer, J. Surface-bound gradients for studies of soft materials behavior. Annu Rev Mater Res 42, 435, 2012. 7. Mei, Y., Saha, K., Bogatyrev, S.R., Yang, J., Hook, A.L., Kalcioglu, Z.I., Cho, S.-W., Mitalipova, M., Pyzocha, N., Rojas, F., Van Villet, K.J., Davies, M.C., Alexander, M., Langer, R., Jaenisch, R., and Anderson, D. Combinatorial development of biomaterials for clonal growth of human pluripotent stem cells. Nat Mater 9, 768, 2010. 8. Hook, A.L., Thissen, H., and Voelcker, N.H. Advanced substrate fabrication for cell microarrays. Biomacromolecules 10, 573, 2009. 9. Kurkuri, M.D., Driever, C., Johnson, G., McFarland, G., Thissen, H., and Voelcker, N.H. Multifunctional polymer coatings for cell microarray applications. Biomacromolecules 10, 1163, 2009. 10. Michelmore, A., Clements, L., Steele, D.A., Voelcker, N.H., and Szili, E.J. Gradient technology for highthroughput screening of interactions between cells and nanostructured materials. J Nanomater 2012, 8, 2012. 11. Khung, Y., Barritt, G., and Voelcker, N. Using continuous porous silicon gradients to study the influence of surface topography on the behaviour of neuroblastoma cells. Exp Cell Res 314, 789, 2008. 12. Vasilev, K., Sah, V.R., Goreham, R.V., Ndi, C., Short, R.D., and Griesser, H.J. Antibacterial surfaces by adsorptive binding of polyvinyl-sulphonate-stabilized silver nanoparticles. Nanotechnology 21, 215102, 2010.

DELALAT ET AL.

13. Cole, M.A., Voelcker, N.H., Thissen, H., and Griesser, H.J. Stimuli-responsive interfaces and systems for the control of protein–surface and cell–surface interactions. Biomaterials 30, 1827, 2009. 14. Hook, A.L., Thissen, H., Hayes, J.P., and Voelcker, N.H. Spatially controlled electro-stimulated DNA adsorption and desorption for biochip applications. Biosens Bioelectron 21, 2137, 2006. 15. Vasilev, K., Mierczynska, A., Hook, A.L., Chan, J., Voelcker, N.H., and Short, R.D. Creating gradients of two proteins by differential passive adsorption onto a PEGdensity gradient. Biomaterials 31, 392, 2010. 16. Barry, J.J.A., Silva, M.M.C.G, Shakesheff, K.M., Howdle, S.M., and Alexander, M.R. Using plasma deposits to promote cell population of the porous interior of three-dimensional poly (d, l-lactic acid) tissue-engineering scaffolds. Adv Funct Mater 15, 1134, 2005. 17. Low, S.P., Voelcker, N.H., Canham, L.T., and Williams, K.A. The biocompatibility of porous silicon in tissues of the eye. Biomaterials 30, 2873, 2009. 18. Notara, M., Bullett, N., Deshpande, P., Haddow, D.B., MacNeil, S., and Daniels, J.T. Plasma polymer coated surfaces for serum-free culture of limbal epithelium for ocular surface disease. J Mater Sci Mater Med 18, 329, 2007. 19. Haddow, D.B., Steele, D., Short, R.D., Dawson, R.A., and MacNeil, S. Plasma-polymerized surfaces for culture of human keratinocytes and transfer of cells to an in vitro wound-bed model. J Biomed Mater Res A 64, 80, 2002. 20. Ng, E.S., Azzola, L., Sourris, K., Robb, L., Stanley, E.G., and Elefanty, A.G. The primitive streak gene Mixl1 is required for efficient haematopoiesis and BMP4-induced ventral mesoderm patterning in differentiating ES cells. Development 132, 873, 2005. 21. Koike, M., Sakaki, S., Amano, Y., and Kurosawa, H. Characterization of embryoid bodies of mouse embryonic stem cells formed under various culture conditions and estimation of differentiation status of such bodies. J Biosci Bioeng 104, 294, 2007. 22. Zandstra, P.W., Bauwens, C., Yin, T., Liu, Q., Schiller, H., Zweigerdt, R., Pasumarthi, K.B., and Field, L.J. Scalable production of embryonic stem cell-derived cardiomyocytes. Tissue Eng 9, 767, 2003. 23. Niebruegge, S., Bauwens, C.L., Peerani, R., Thavandiran, N., Masse, S., Sevaptisidis, E., Nanthakumar, K., Woodhouse, K., Husain, M., Kumacheva, E., and Zandstra, P.W. Generation of human embryonic stem cell-derived mesoderm and cardiac cells using size-specified aggregates in an oxygen-controlled bioreactor. Biotechnol Bioeng 102, 493, 2009. 24. Sachlos, E., and Auguste, D.T. Embryoid body morphology influences diffusive transport of inductive biochemicals: a strategy for stem cell differentiation. Biomaterials 29, 4471, 2008. 25. Levenberg, S., Huang, N.F., Lavik, E., Rogers, A.B., Itskovitz-Eldor, J., and Langer, R. Differentiation of human embryonic stem cells on three-dimensional polymer scaffolds. Proc Natl Acad Sci U S A 100, 12741, 2003. 26. Chien, K.R. Regenerative medicine and human models of human disease. Nature 453, 302, 2008. 27. Steele, J.G., Johnson, G., McFarland, C., Dalton, B.A., Gengenbach, T.R., Chatelier, R.C., Underwood, P.A. and Griesser, H.J. Roles of serum vitronectin and fibronectin in initial attachment of human vein endothelial cells and

SCREENING OF EB CELL BEHAVIOR ON PLASMA POLYMER GRADIENTS

28. 29.

30.

31.

32.

33. 34.

35.

36.

37.

38.

39.

40.

41.

dermal fibroblasts on oxygen- and nitrogen-containing surfaces made by radiofrequency plasmas. J Biomater Sci Polym Ed 6, 511, 1995. Goreham, R.V., Short, R.D., and Vasilev, K. Method for the generation of surface-bound nanoparticle density gradients. J Phys Chem C 115, 3429, 2011. Harding, F., Goreham, R., Short, R., Vasilev, K., and Voelcker, N.H. Surface bound amine functional group density influences embryonic stem cell maintenance. Adv Healthc Mater 8, 1739, 2012. Michelmore, A., Mierczynska, A., Poh, Z., Goreham, R.V., Losic, D., Short, R.D., and Vasilev, K. Versatile gradients of chemistry, bound ligands and nanoparticles on alumina nanopore arrays. Nanotechnology 22, 415601, 2011. France, R.M., Short, R.D., Dawson, R.A., and Macneil, S. Attachment of human keratinocytes to plasma co-polymers of acrylic acid/octa-1, 7-diene and allyl amine/octa-1, 7diene. J Mater Chem 8, 37, 1998. Daw, R., Candan, S., Beck, A.J., Devlin, A.J., Brook, I.M., MacNeil, S., Dawson, R.A., and Short, R.D. Plasma copolymer surfaces of acrylic acid/1, 7 octadiene: Surface characterisation and the attachment of ROS 17/2.8 osteoblast-like cells. Biomaterials 19, 1717, 1998. Lee, J.H., Jung, H.W., Kang, I.-K., and Lee, H.B. Cell behaviour on polymer surfaces with different functional groups. Biomaterials 15, 705, 1994. Hamerli, P., Weigel, T., Groth, T., and Paul, D. Surface properties of and cell adhesion onto allylamine-plasmacoated polyethylenterephtalat membranes. Biomaterials 24, 3989, 2003. Inoue, S., Imamura, M., Umezawa, A., and Tabata, Y. Attachment, proliferation and adipogenic differentiation of adipo-stromal cells on self-assembled monolayers of different chemical compositions. J Biomater Sci Polym Ed 19, 893, 2008. Calonder, C., Matthew, H.W.T., and Van Tassel, P.R. Adsorbed layers of oriented fibronectin: a strategy to control cell–surface interactions. J Biomed Mater Res A 75, 316, 2005. Karp, J.M., Yeh, J., Eng, G., Fukuda, J., Blumling, J., Suh, K.-Y., Cheng, J., Mahdavi, A., Borenstein, J., Langer, R., and Khademhosseini, A. Controlling size, shape and homogeneity of embryoid bodies using poly (ethylene glycol) microwells. Lab Chip 7, 786, 2007. Gerecht, S., Burdick, J.A., Ferreira, L.S., Townsend, S.A., Langer, R., and Vunjak-Novakovic, G. Hyaluronic acid hydrogel for controlled self-renewal and differentiation of human embryonic stem cells. Proc Natl Acad Sci U S A 104, 11298, 2007. Park, J., Cho, C.H., Parashurama, N., Li, Y., Berthiaume, F., Toner, M., Tilles, A.W., and Yarmush, M.L. Microfabrication-based modulation of embryonic stem cell differentiation. Lab Chip 7, 1018, 2007. Melkoumian, Z., Weber, J.L., Weber, D.M., Fadeev, A.G., Zhou, Y., Dolley-Sonneville, P., Yang, J., Qiu, L., Priest, C.A., Shogbon, C., Martin, A.W., Nelson, J., West, P., Beltzer, J.P., Pal, S., and Brandenberger, R. Synthetic peptide-acrylate surfaces for long-term self-renewal and cardiomyocyte differentiation of human embryonic stem cells. Nat Biotechnol 28, 606, 2010. Liu, J.F., Chen, Y.M., Yang, J.J., Kurokawa, T., Kakugo, A., Yamamoto, K., and Gong J.P. Dynamic behavior and spontaneous differentiation of mouse embryoid bodies on

42.

43.

44.

45. 46.

47.

48. 49.

50.

51. 52. 53.

1725

hydrogel substrates of different surface charge and chemical structures. Tissue Eng Part A 17, 2343, 2011. Ren, Y.-J., Zhang, H., Huang, H., Wang, X.-M., Zhou, Z.Y., Cui, F.-Z., and An, Y.-H. In vitro behavior of neural stem cells in response to different chemical functional groups. Biomaterials 30, 1036, 2009. Fehling, H.J., Lacaud, G., Kubo, A., Kennedy, M., Robertson, S., Keller, G., and Kouskoff, V. Tracking mesoderm induction and its specification to the hemangioblast during embryonic stem cell differentiation. Development 130, 4217, 2003. Wiese, C., Rolletschek, A., Kania, G., Blyszczuk, P., Tarasov, K.V., Tarasova, Y., Wersto, R.P., Boheler, K.R., and Wobus, A.M. Nestin expression—a property of multi-lineage progenitor cells? Cell Mol Life Sci 61, 2510, 2004. Easter, S., Ross, L., and Frankfurter, A. Initial tract formation in the mouse brain. Neurosci 13, 285, 1993. Locher, H., Frijns, J.H., Huisman, M.A., and de Sousa Lopes, S.M. TUBB3: neuronal marker or melanocyte mimic? Cell Transplant 2013. DOI:10.3727/096368913x 674099. Carre´, M., Andre´, N., Carles, G., Borghi, H., Brichese, L., Briand, C., and Braguer, D. Tubulin is an inherent component of mitochondrial membranes that interacts with the voltage-dependent anion channel. J Biol Chem 277, 33664, 2002. Arima, Y., and Iwata, H. Effects of surface functional groups on protein adsorption and subsequent cell adhesion using self-assembled monolayers. J MaterChem 17, 4079, 2007. Arima, Y., and Iwata, H. Effect of wettability and surface functional groups on protein adsorption and cell adhesion using well-defined mixed self-assembled monolayers. Biomaterials 28, 3074, 2007. Wells, N., Baxter, M.A., Turnbull, J.E., Murray, P., Edgar, D., Parry, K.L., Steele, D.A., and Short, R.D. The geometric control of E14 and R1 mouse embryonic stem cell pluripotency by plasma polymer surface chemical gradients. Biomaterials 30, 1066, 2009. Curtis, A.S., and Forrester, J.V. The competitive effects of serum proteins on cell adhesion. J Cell Sci 71, 17, 1984. Harb, N., Archer, T.K., and Sato, N. The rho-rock-myosin signaling axis determines cell-cell integrity of self-renewing pluripotent stem cells. PLoS One 3, e3001, 2008. Peerani, R., Onishi, K., Mahdavi, A., Kumacheva, E., and Zandstra, P.W. Manipulation of signaling thresholds in ‘‘engineered stem cell niches’’ identifies design criteria for pluripotent stem cell screens. PLoS One 4, e6438, 2009.

Address correspondence to: Nicolas H. Voelcker, Dr rer nat Mawson Institute University of South Australia Mawson Lakes SA 5095 Adelaide Australia E-mail: [email protected] Received: June 10, 2013 Accepted: December 18, 2013 Online Publication Date: February 5, 2014