Particle aggregates formed during furfuryl methacrylate plasma polymerization affect human mesenchymal stem cell behaviour

Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Full Length Article

Particle aggregates formed during furfuryl methacrylate plasma polymerization affect human mesenchymal stem cell behaviour Hanieh Safizadeh Shirazi a,b,∗ , Nicholas Rogers a,b , Andrew Michelmore a,b,c , Jason D. Whittle a,b,c a

Future Industries Institute, University of South Australia, Mawson Lakes, Adelaide, SA 5095, Australia Cooperative Research Centre for Cell Therapy Manufacturing (CRC-CTM), Mawson Lakes, SA 5095, Australia c School of Engineering, University of South Australia, Mawson Lakes, Adelaide, SA 5095, Australia b

a r t i c l e

i n f o

Article history: Received 27 June 2017 Received in revised form 22 September 2017 Accepted 26 October 2017 Available online 26 October 2017 Keywords: Human mesenchymal stem cells Furfuryl methacrylate coatings Plasma polymerization

a b s t r a c t Human Mesenchymal Stem cells (hMSCs) are becoming a major focus in biomedical fields. Application of in vitro expanded hMSCs to treat numerous ailments has led to a commercial emphasis on improving hMSC growth ex vivo. Production of substrate independent, novel thin films is one potential tool for production of commercial viable hMSC expansion. Plasma polymerization allow controlled chemical optimisation of large scale surface areas in a substrate independent manner. Previous study shown that plasma polymerized Furfuryl Methacrylate (ppFMA) surfaces allowed primary fibroblast cells adhesion and proliferation. However, under some deposition conditions, particle aggregates formation was observed. These aggregates had the effect of disrupting cell attachment, despite being chemically indistinguishable from the underlying surface. Herein, hMSCs were cultured on ppFMA surfaces to determine their suitability for stem cell culture and observe the effect of particle aggregates on hMSC attachment and growth. Both metabolic and DNA quantification assays showed that surfaces with particle aggregates had lower numbers of attached cells and slower growth. Uniform surfaces without aggregates showed higher cell attachment and growth levels, which were comparable to Thermanox. Phenotypic analysis showed that there was no change to hMSCs phenotype after 7 & 14 days of culture on uniform ppFMA surface. Further investigation using time-lapse image analysis indicated that particle aggregates reduced cell attachment by presenting a physically weak boundary layer, which was damaged by intracellular tension during cell spreading. ppFMA surface can provide a stable substrate independent hMSCs expansion interface that could be applied to larger scale bioreactors, beads or scaffolds. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Cell therapies are viewed as a potential cure for a range of human diseases and health conditions, where damaged cells can be replaced by healthy functional cells that have been grown ex vivo [1,2]. Efficient expansion of patient derived stem cells, within a laboratory, is a major focus for the cell therapies fields. Mesenchymal stem cells are of particular interest as they can be both expanded ex vivo and differentiated into specific mesodermal lineages by osteogenesis, adipogenesis or chondrogenesis [3–7]. To provide commercial quantities of stem cells for medical applications, a culture surface which provides good adhesion,

∗ Corresponding author at: Future Industries Institute, University of South Australia, Mawson Lakes, Adelaide, SA 5095, Australia. E-mail address: hanieh.safizadeh [email protected] (H.S. Shirazi). https://doi.org/10.1016/j.colsurfb.2017.10.065 0927-7765/© 2017 Elsevier B.V. All rights reserved.

proliferation together with suitable release and maintenance properties needs to be utilised. Previous studies, using traditional polymerisation techniques, have identified key chemical functional groups suitable for application in the tissue engineering and biomedical fields [8,9]. Furfuryl methacrylate (FMA) has been identified as a hit monomer, and its use for coatings and biomedical applications is the subject of current investigations [10]. FMA has received particular attention as it is a stable molecule that is not polymerized at room temperature, exhibits low shrinkage and has been used in biomedical and biomaterial applications as coating and adhesives [10–15]. It has been shown that poly(FMA) surfaces are suitable for human embryonic stem cell adhesion, proliferation and maintenance [8]. However, polymer based coatings of FMA have distinct limitations when scaling the technology into commercially viable settings. Thus, alternative surface modification methods are being explored in order to facilitate industrial uptake. One of the surface

262

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

modification methods is plasma polymerization, a well-developed technique that has been widely used for the last 20 years [16–22]. Plasma polymerization has been shown to be a suitable way to produce thin film coatings on various planar or textured substrates with retention of desired functional properties. Plasma polymerization does not require surface preparation or any initiator in the process and is a single step, fast, easy, dry and scalable process [16,20,23–26]. Previous work has shown that plasma polymerized FMA (ppFMA) surfaces, produced under suitable conditions, promoted fibroblast cell adhesion and proliferation, but the conditions needed to be carefully controlled to avoid the confounding effect of particle aggregate formation on cell attachment [27]. Mesenchymal stem cells are particularly sensitive to environmental biological cues which can result in differentiation. As such, cell therapy applications using ppFMA coatings must both be sufficient to grow large numbers of cells while simultaneously maintaining pluripotent phenotype. In this study, the interaction of human mesenchymal stem cells (hMSCs) with ppFMA surfaces is investigated during the attachment and proliferation stages. Metabolic and DNA quantification assays were performed to assess the expansion of hMSCs. Concurrent experiments using confocal time-lapse imaging was performed to explore the dynamics of cell attachment and growth in the presence of ppFMA particle aggregates. In addition, flow cytometry was performed on hit ppFMA surfaces to assess hMSCs phenotype after 7 & 14 days of culture. 2. Experimental 2.1. Materials FMA 97%, containing 200 ppm monomethyl ether hydroquinone as inhibitor, Sigma−Aldrich was used as received. Thermanox coverslips (13 mm dia) (ProSciTech), silicon wafers (M.M.R.C), 35 mm optical petri dishes (Ibidi) and 24 well plates (Corning Costar) were used as substrates for deposited plasma polymer coatings. Silicon wafers were cut into 1 cm × 1 cm squares and cleaned with acetone before being rinsed with MilliQ water and dried with nitrogen gas. 2.2. FMA plasma polymerization A steel capacitively coupled reactor, 25 cm long and 30 cm in diameter that encloses an RF electrode of 28 cm in diameter, was used for plasma polymer coating deposition. RF power at 13.56 MHz was applied to the internal electrode through an impedance matching network (AMN 150R). A rotary pump was used to evacuate the plasma chamber, to a base pressure of <3 × 10−3 mbar [28]. After degassing the monomer with freeze-pump-thaw cycling, it was allowed to flow into the reactor and controlled with a 13 mm ball valve. The glass coverslips and 24- well plates were placed on the bottom (earthed) electrode. During the plasma polymerisation experiment the working gas pressure was either 2.3 × 10−2 mbar or 4.5 × 10−2 mbar and the flow rate was either 2.6 sccm or 4.9 sccm. After setting the pressure and flow rate, the RF power of 2 and 4 W were used to deposit the plasma polymers for a duration of 1 and 10 min. 2.3. Surface Characterizations: XPS, ToF-SIMS and SEM The ppFMA surfaces were characterized with X-ray Photoelectron Spectroscopy (XPS) and Time-of-Flight Secondary Mass Spectrometry (ToF-SIMS) for their chemistry and Scanning Electron Microscopy (SEM) to assess morphology. In a previous study on ppFMA, the surfaces were analysed and discussed in detail [27]. In

brief, for XPS, a SPECS SAGE instrument was used to record high resolution (20 eV pass energy) spectra for C1 s and O1 s peaks where all binding energies were referenced to the aliphatic C1 s carbon peak at 285 eV [29]. Casa XPS software was used to process and analyse the spectra with typical full-width-at-half-maximum (FWHM) of 1.6 eV. For ToF-SIMS analysis a PHI TRIFT V nanoTOF was used. The system was operated at 30 kV energy under vacuum of 5 × 10−6 and armed with a pulsed liquid metal 79+ Au primary ion gun (LMIG). Positive SIMS spectra were collected for duration of 1 min using 100 × 100 ␮m raster with bunched Au1 beam for better mass resolution [27]. SEM micrographs were obtained with a ZEISS Gemini 2 (crossbeam 540). The images were taken with working distance of 1–2 mm and magnification of 1–10 ␮m with the choice of in-lens secondary electron imaging. The resolution was 1024 × 768 with a scan speed of zero for image focusing and a scan speed of 4 or 5 for capturing the image. 2.4. Human mesenchymal stem cells (hMSCs) and ppFMA surfaces Human Mesenchymal stem cells (Roosterbio) were cultured in ␣MEM (Lonza) supplemented with 10% foetal calf serum (FCS, Lifetech (Gibco), Glutamax (1X, Invitrogen) and Pen/Strep (10000 U/ml, Invitrogen)). Cells were incubated in a humid atmosphere with 5% CO2 at 37 ◦ C. Three sets of plasma polymer samples were seeded with three different human mesenchymal stem cell populations (p4-7) to investigate the surface attachment and proliferation properties. 24-well plates coated with ppFMA surfaces were seeded at 3 × 104 cells/well for 1 h attachment assays and 1 × 104 cells in 500 ␮L for proliferation and growth assays (over 5 days). To measure the hMSCs attachment and proliferation, resazurin assays were conducted. A Nikon Eclipse TE 2000-U Microscope with plan fluor 4x/0.3 objective lens was used to observe and image the attachment and proliferation of hMSCs. 2.5. Resazurin assay A FLUOstar OPTIMA plate reader from BMG LABTECH was used to perform the resazurin assay. After 1 h from initial seeding, the 30 × 103 seeded well plates were washed twice with PBS to remove any unattached cells and 500 ␮L of 10% resazurin medium added to each well and incubated for another hour together with bare Thermanox as the control. After 1 h, 100 ␮L of the solution was transferred to a fresh 96-well plate for reading at an excitation and emission wavelength of 544 and 590 nm respectively. The same procedure was followed with 10 × 103 seeded well plates for the proliferation assays after 1st and 5th day. The data from three biological repeats were analysed and compared to bare thermanox as a control. 2.6. Confocal microscopy live cell imaging Nikon’s A1R+ confocal laser scanning microscope- (Coherent Scientific, Australia), was used to capture images of live hMSCs. The cells were stained with Cell Tracer, Oregon Green 488 (Invitrogen), 24 h prior to seeding. The stain solution was prepared by adding a mixture of 1 vial of cell tracer plus 20 ␮L of DMSO into 5 mL PBS. Cells were washed thrice with PBS and cell tracer solution added to them. After adding the stain solution, cells were incubated for 20 min. Subsequently, cells were washed with PBS to remove excess stain and incubated with fresh complete media overnight. Stained cells were counted the following day and seeded at 50 K density onto 35 mm ibidi optical dishes which coated with ppFMA

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

(A and C). The dishes were then placed into a pre heated Okolab H301-ECMCLTi environmental chamber attached to a Nikon A1R confocal laser scanning microscope with a motorized x-y-z piezo stage. The environmental chamber was humidified and maintained at 37 ◦ C, 5% CO2 . Transmitted light images were taken of two points every 5 min for 1 h followed by every 15 min for the next 4 h using NIS Elements software to image.

2.7. Flow cytometry analysis Image Stream (amnis) FACS instrument with ISX program which separates the cells base on their florescence properties was used to check the cell phenotype. hMSCs were stained using the antibodies and reagents provided in the Human Mesenchymal Stem Cell Marker Verification Multi-Colour Flow Cytometry Kit from R&D systems (Catalogue # FMC020). Coated wells seeded with hMSCs were washed with PBS, trypsinized and the media was removed after spinning down the cells. The hMSCs were resuspended in 1 mL media for counting. The cell suspension was transferred to a 1.5 mL micro centrifuge tube and spun down using centrifuge 5424- Eppendorf with 300 x rcf for 5 min. The media were discarded carefully and cells washed with 1 mL of Staining Buffer and resuspended in another ml of staining buffer. The cell density of 1 × 105 in 100 ␮L were transferred into a 5 mL Flow Cytometry tube for two positive and two negative controls. 10 ␮L of each positive antibody were added into 2 positive control tubes (2.6sccm-2w–1 min & bare thermanox) and 10 ␮L of the negative isotype control cocktail into 2 negative control tubes (2.6 sccm-2w–1 min & bare thermanox). Samples were incubated for 30–45 min at room temperature in the dark. After incubation, hMSCs were washed twice with 1 mL of Staining Buffer. Finally, cell pellets were resuspended in 100 ␮L of Staining Buffer and ready to analyse for marker expression. Two laser wave lengths of 488 and 672 with 50 mW were chosen in the ISX program and 10000 cell counts for both positive and isotype controls.

3. Results 3.1. Surface Characterization, XPS, ToF-SIMS and SEM Plasma polymer surfaces were prepared for testing following Shirazi, 2016 with conditions 2.6 sccm, 2w, 1 min (A), 4.9 sccm, 4w, 1 min (B) and 4.9 sccm, 10w, 10 min (C) [27]. The XPS results indicates no difference in the chemical composition of the ppFMA surfaces. A sample spectrum with the fitted peaks at C C/CH at 285 eV, C O C at 286.4 eV, O C O at 289.2 eV and C COOH at 285.7 eV and C’ component which is C O at 287.9 eV can be seen in Fig. 1 I. The other spectra can be found in the supplementary information (SI1). Additionally, FMA monomer theoretical value of each C component (C C (44.44), C O C (33.33), C COOH (11.11), O C O (11.11), C O (0.00) showed similar values with the results obtained from plasma polymerization which were 62.3, 17.1, 6.7, 6.7 and 7.2 for each C component respectively. These data can also be found in the supplementary information (SI2). ToF-SIMS data shows not a significant difference in the intensity of C5 H5 O (81) which is related to the furan ring as an important structure for cell attachment and proliferation. However, it shows the retention of the intact monomer itself C10 H9 O2 (166) in all the polymerization conditions. Therefore, the chemistry is constant for all samples within error. Furthermore, the SEM images presents different surface topography on the ppFMA surfaces (Fig. 1-II) where ‘A’ has no particle aggregates and is a uniform surface but ‘B’ and ‘C’ shows particle aggregates with different size and concentration.

263

3.2. hMSCs attachment to ppFMA surfaces The metabolic activity of the hMSCs on ppFMA surfaces was investigated after 1 h of attachment using a resazurin assay. Fig. 2 shows a significant increase of hMSC attachment to surface A (uniform and flat coating) when compared to Surface B and C (with small and large particle aggregates, respectively). The trend in resazurin assay data was also observed using a cyQuant (DNA quantification) assay (SI3). hMSCs attachment on uniform ppFMA (A) is not statistically different from thermanox. In addition to the metabolic and DNA quantification assays, images were taken using bright field microscopy (Fig. 2 part II). The hMSCs were then counted using Image J software and confirm the same trend previously seen (Fig. 2 part I). The cell counts data corroborate that uniform ppFMA surfaces are at least as suitable for hMSCs attachment as Thermanox plastic. This combined data also shows the detrimental effects of ppFMA particles on hMSCs attachment. 3.3. Confocal live cell imaging to assess the effect of particles on hMSCs To better visualise the effects that ppFMA particles have on hMSCs attachment, confocal live cell imaging was carried out on surfaces ‘A’ and ‘C’. hMSCs behaviour was studied for two time lapse periods of 1 and 5 h (Fig. 3). Live cell imaging showed that hMSCs settled and attached quickly on surface ‘A’. hMSCs attachment to particle containing surfaces indicated two distinct behaviours. During the 1 h time lapse on an area with fewer aggregates, it can be seen that hMSCs initially attached and spread on the surface (points a, b &c) but shrank and adopted a round, punctate morphology after a few hours (small bright cells, Fig. 4). This can also be seen in the bright field images of Fig. 5-I where hMSCs left a halo behind after they receded. Areas with more particle aggregates monitored for 5 h that displays maximum number of cells were not able to attach to the surface, even for the cells which attempted to do so (point d), they went through blebbing and detached (Fig. 4). Both cell shrinkage and blebbing are types of apoptosis which indicate cell death and matches our metabolic and DNA quantification data sets. We hypothesize that the particles are weakly adhered to the surface, and are pulled from the surface as the focal complex matures into a stable focal adhesion. As a result, the cell rounds up and becomes detached leading to cell death (Fig. 5-I) and leads to a ‘halo’ effect that can be seen in the brightfield images (Fig. 5-II). 3.4. Proliferation of hMSCs on ppFMA Resazurin assays were conducted on day 1 and 5 to determine the metabolic activity of the cells. Significantly more hMSCs were grown on ppFMA surfaces when compared to surfaces without particles. Importantly uniform ppFMA surfaces had similar growth to Thermanox after 5 days (Figs. 6I and 7 I). Resazurin results were supported by cyQuant assay results (SI3) which showed the same trend. The proliferation and morphology was monitored by bright field microscopy for 1st (Fig. 6II) and 5th (Fig. 7II) day of proliferation. Both stages indicated higher proliferation on surface ‘A’ which was comparable to Thermanox whereas surfaces ‘B’ and ‘C’ indicated lower rates of cell proliferation. These results were confirmed by cell counting (using Image J) which also followed the same trend as the resazurin assay results (Figs. 6 and 7I). 3.5. Maintenance of hMSCs phenotype after culture on ppFMA surfaces Phenotypic analysis of hMSCs grown on ppFMA surfaces was conducted by flow cytometry. hMSCs cultured on uniform ppFMA

264

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

Fig. 1. Part I shows typical XPS spectrum for surface ‘C, 4W-10min-4.9sccm’ with 5 fitted peaks and part II presents SEM images showing different surface morphology; ‘A’ as a uniform surface, and ‘B’ & ‘C’ with sizes and densities of particle aggregates Scale bar: 10 ␮m (A,B,C) 1 ␮m (B’, C’).

Fig. 2. Part I presents Resazurin Assay and cell counts of hMSCs after 1 h attachment (3 × 104 cells seeded), shows ‘A’ and Thermanox are similar but significantly different from ‘B’ where “p < 0.05” and more significant different from ‘C’ where “p < 0.01” (n = 18) and part II shows Bright field images of hMSCs on ppFMA surfaces after 1hr attachment (4X objective).

surfaces after 7 and 14 days show clear surface expression of mesenchymal markers CD90, CD73 and CD105 with minimal

expression of antigens of CD45,CD34 CD11b CD79A and HLADR(negative marker cocktail, Table 1) (SI 4). This data shows that

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

265

Fig. 3. Confocal live cell imaging during time lapses of hMSCs on Surface ‘A’ for 1 h and 5 h with 100 ␮m and 50 ␮m respectively.

Fig. 4. Confocal live imaging during the time lapses of hMSCs on Surface ‘C’ for 1 h and 5 h.

Fig. 5. (I) Schematic showing the hypothesized interaction between hMSCs and ppFMA particle aggregates and (II) Bright Field images a) 4X and b) 20X of dead hMSCs on surface ‘C’.

266

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

Fig. 6. Part I presents Resazurin Assay and cell counts of hMSCs proliferation (1 × 104 ), presents ‘A’ and thermanox are comparable but different from ‘B’ where “p < 0.05” and more significant with ‘C’ where “p < 0.001” for 1st proliferation day (n = 18) and part II Bright field images of hMSCs with objective lens 4X of 1st proliferation day.

Fig. 7. Part I presents Resazurin Assay and cell counts of hMSCs proliferation (1 × 104 ), presents ‘A’ and thermanox are comparable but different from ‘B’ where “p < 0.05” and more significant with ‘C’ where “p < 0.001” for 1st proliferation day (n = 18) and part II demonstrates the hMSCs Bright field images with objective lens 4X of 5th proliferation day.

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268 Table 1 Flow cytometry of hMSCs grown on ppFMA surfaces. Antigen

Day 7

Day 14

CD73 CD90 CD105 Negative cocktail

>99% Positive >99% Positive >99% Positive <1% Positive

>99% Positive >99% Positive >95% Positive <4% Positive

hMSCs grown on uniform ppFMA surfaces maintain their phenotype. 4. Discussion XPS analysis was utilised on plasma polymer surfaces which revealed that there was no noticeable differences among all the surfaces with or without the particle aggregates, therefore more sensitive ToF-SIMS analysis was conducted to investigate these surfaces. SIMS analysis was unable to differentiate the particle aggregates from the plasma polymer surface. This suggests that the chemistry of particle aggregates is the same as the plasma polymer deposited on the surface, although the density of these aggregates was altered by varying the plasma polymerization parameters. In addition, ToF-SIMS results illustrated the furan ring stability and retention on all surfaces which also correlated with MS analysis of the plasma polymer gas phase from previous studies [30]. Surfaces were characterised via SEM to observe their morphology and topography, which showed the formation of particle aggregates on the ppFMA surfaces under certain conditions. We propose that the particle aggregate formation can be described in a similar way to DLVO theory [31,32]. DLVO theory describes the interaction between colloidal particles and their aggregation behaviour, and was developed in the 1940s. Electrostatic repulsive forces between two particles produce an energy barrier which can prevent them from aggregating. However, particles which overcome the barrier energy can form aggregates through strong attractive van der Waals forces[33]. All surfaces in contact with plasma acquire a negative potential relative to the plasma phase [34], including particles which grow in the plasma phase [35]. Therefore, particles which grow in the plasma phase will be electrostatically repelled from the surface and held in the plasma phase. Aggregates continue to grow until their mass overcomes this repulsive force and they deposit on the surface. Also, increasing the deposition time produces additional particles which leads to the formation of more aggregates. This higher particle formation at RF power 4 W can be due to intact FMA which can help grow the particles via radical propagation, but if the power is increased too much, the monomer becomes too fragmented and the double bond functionality may be lost, resulting in fewer particles being formed [27]. Human MSC attachment after 1 h was significantly higher on surface ‘A’ which was a uniform surface with no particle aggregates and comparable to thermanox control surface. Surfaces ‘B’ and ‘C’ with low and high density of particle aggregates had lower cell attachment which was more noticeable on surface ‘C’. Proliferation data on 1 day and 5 days post seeding for surfaces ‘A’, ‘B’ and ‘C’ followed the same trend as attachment which suggest the particle aggregates inhibit normal cell attachment, growth and proliferation on the ppFMA surfaces. Flow cytometry analysis was conducted to observe hMSCs maintain their phenotype on ppFMA surfaces. Hence, hMSCs were grown on surface ‘A’ as an optimal surface and ‘C’ as a representative for surfaces with aggregates. After both 7 and 14 days, surface ‘C’ did not produce enough cells to run flow cytometry analysis in a 10 cm dish, due to the existence of large particle aggregates on this ppFMA surface. On the other hand, cells were harvested from surface ‘A’

267

and after analysis showed no change in hMSCs phenotype for 7 and 14 days post seeding. The morphology of dead cells on surface ‘C’ was observed under the bright field microscopy where a halo around them was seen. Hence, it was suggested that cell death could be due to separation of particle aggregates from the surface as a result of lose focal attachment to substrate. The mechanism of cell attachment on surfaces with particle aggregates were investigated via confocal live microscopy. It was noticed when hMSCs contacted with particles, they struggled to attach to the surface and at the end went through cell shrinkage and blebbing as forms of apoptosis which showed cell death and matches metabolic and DNA quantification data sets. The effect of nanotopography has been observed previously [36–38]. It has been shown that moderate levels of topographical features can enhance cell attachment and proliferation. However this was shown for smaller topographical features, which were stable on the surface. In this study, particle aggregates formed during the surface coating process were only loosely adhered to the surface. The results above show that hMSCs that attempt to attach to the surface are able to move these particle aggregates due to intracellular tension, resulting in the halo appearance of the surface. It must be noted that the surface analysis showed that the chemistry of the particles and the underlying surface were identical, and so changes in cell behaviour must be due to particle morphology, not chemical differences. Importantly, ppFMA surfaces without particle aggregates showed good performance in terms of hMSC attachment and proliferation, demonstrating their potential in biomaterial applications. However, these results show that when coating biomaterials with thin film coatings, the formation of particle aggregates should either be minimised during the fabrication process, or they must be strongly adhered to the surface to maximise cell attachment and proliferation. 5. Conclusion Deposition of ppFMA thin films and resultant effects on hMSCs attachment and expansion have been studied. Flat, uniform ppFMA depositions resulted in hMSCs adhesion and proliferation comparable to Thermanox controls. It is important to note that chemically identical surfaces (B) that produced both smaller and fewer particle or larger aggregates (B, C) showed inhibition of hMSCs attachment. As a result, particle containing surfaces, showed significantly reduced hMSCs after 1 and 5 days of culture. hMSCs attachment and expansion on uniform ppFMA was not significantly different from thermanox controls by both metabolic and DNA quantification assays. Flow cytometry analysis showed that hMSCs, cultured on uniform ppFMA surfaces maintained their phenotype after 7 and 14 days of culture. This data indicates that the flat and uniform surface and the resultant ppFMA chemistry has no adverse effects on hMSCs fate. These data give substantial evidence of ppFMA as a substrate independent thin film coating for the ex vivo expansion of hMSCs. The coating techniques used within the study has a distinct advantage as it can potentially be applied to larger scale tissue culture surfaces or cheap materials including beads, scaffolds or lattices for commercial scale up of hMSCs expansion. Acknowledgments The authors would like to acknowledge financial support from the Cooperative Research Centre for Cell Therapy Manufacturing. HSS would like to thank the University of South Australia financial support in the form of an IPS scholarship. A special thanks to Louise Smith for training and providing information on confocal live imaging.

268

H.S. Shirazi et al. / Colloids and Surfaces B: Biointerfaces 161 (2018) 261–268

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.colsurfb.2017.10. 065. References [1] H. Lin, H. Ouyang, J. Zhu, S. Huang, Z. Liu, S. Chen, G. Cao, G. Li, R.A. Signer, Y. Xu, Lens regeneration using endogenous stem cells with gain of visual function, Nature 531 (2016) 323–328. [2] S.A. Fisher, S.J. Brunskill, C. Doree, A. Mathur, D.P. Taggart, E. Martin-Rendon, Stem cell therapy for chronic ischaemic heart disease and congestive heart failure, Cochrane Libr. 4 (2014) CD007888, http://dx.doi.org/10.1002/ 14651858.CD007888.pub2. [3] B. Abdallah, M. Kassem, Human mesenchymal stem cells: from basic biology to clinical applications, Gene Ther. 15 (2007) 109–116. [4] A.I. Caplan, Adult mesenchymal stem cells for tissue engineering versus regenerative medicine, J. Cell. Physiol. 213 (2007) 341–347. [5] G. De Wert, C. Mummery, Human embryonic stem cells: research, ethics and policy, Hum. Reprod. 18 (2003) 672–682. [6] H.J. Jin, Y.K. Bae, M. Kim, S.-J. Kwon, H.B. Jeon, S.J. Choi, S.W. Kim, Y.S. Yang, W. Oh, J.W. Chang, Comparative analysis of human mesenchymal stem cells from bone marrow, adipose tissue, and umbilical cord blood as sources of cell therapy, Int. J. Mol. Sci. 14 (2013) 17986–18001. [7] J.F. Welter, K.J. Penick, L.A. Solchaga, Assessing adipogenic potential of mesenchymal stem cells: a rapid three-dimensional culture screening technique, Stem Cells Int. 8 (2013) 806525. [8] A. Celiz, J. Smith, A. Patel, R. Langer, D. Anderson, D. Barrett, L. Young, M. Davies, C. Denning, M. Alexander, Chemically diverse polymer microarrays and high throughput surface characterisation: a method for discovery of materials for stem cell culture, Biomater. Sci. 2 (2014) 1604–1611. [9] D. Schmaljohann, Thermo-and pH-responsive polymers in drug delivery, Adv. Drug Deliv. Rev. 58 (2006) 1655–1670. [10] A.A. Kavitha, A. Choudhury, N.K. Singha, Controlled radical polymerization of furfuryl methacrylate Macromolecular Symposia, vol. 240, Wiley Online Library, 2006, pp. 232–237. [11] X. Cheng, H.E. Canavan, M.J. Stein, J.R. Hull, S.J. Kweskin, M.S. Wagner, G.A. Somorjai, D.G. Castner, B.D. Ratner, Surface chemical and mechanical properties of plasma-polymerized N-isopropylacrylamide, Langmuir 21 (2005) 7833–7841. [12] J. He, Y. Zhang, E.Y. Chen, Anionic polymerization of biomass-derived furfuryl methacrylate: controlling polymer tacticity and thermoreversibility, J. Polym. Sci. Part A: Polym. Chem. 51 (2013) 2793–2803. [13] J. Lange, J. Rieumont, N. Davidenko, R. Sastre, Photoinitiated bulk polymerization of furfuryl methacrylate. Experimental and kinetic modelling results obtained at different temperatures, Polymer 39 (1998) 2537–2542. [14] J. Lange, J. Rieumont, N. Davidenko, R. Sastre, Kinetics modelling of the crosslinking in the photopolymerization of furfuryl methacrylate in bulk, Comput. Theor. Polym. Sci. 9 (1999) 63–72. [15] S.K. Feng, M. Schmitt, E.Y.X. Chen, Organocatalytic polymerization of furfuryl methacrylate and post-Diels–Alder click reaction to cross-linked materials, Macromol. Chem. Phys. 216 (2015) 1421–1430. [16] H. Biederman, D. Slavınská, Plasma polymer films and their future prospects, Surf. Coat. Technol. 125 (2000) 371–376. [17] D. Hegemann, Plasma polymerization and its applications in textiles, Indian J. Fiber Textile Res. 31 (2006) 99–115. [18] A. Michelmore, C. Charles, R.W. Boswell, R.D. Short, J.D. Whittle, Defining

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29] [30]

[31]

[32] [33] [34] [35] [36]

[37]

[38]

plasma polymerization: new insight into what we should be measuring, ACS Appl. Mater. Interfaces 5 (2013) 5387–5391. A. Michelmore, D.A. Steele, J.D. Whittle, J.W. Bradley, R.D. Short, Nanoscale deposition of chemically functionalised films via plasma polymerisation, RSC Adv. 3 (2013) 13540–13557. F.F. Shi, Recent advances in polymer thin films prepared by plasma polymerization synthesis structural characterization, properties and applications, Surf. Coat. Technol. 82 (1996) 1–15. K.S. Siow, L. Britcher, S. Kumar, H.J. Griesser, Plasma methods for the generation of chemically reactive surfaces for biomolecule immobilization and cell colonization-a review, Plasma Process. Polym. 3 (2006) 392–418. P.-Y. Wang, L.R. Clements, H. Thissen, W.-B. Tsai, N.H. Voelcker, Screening rat mesenchymal stem cell attachment and differentiation on surface chemistries using plasma polymer gradients, Acta Biomater. 11 (2015) 58–67. K.D. Anderson, R.B. Weber, M.E. McConney, H. Jiang, T.J. Bunning, V.V. Tsukruk, Responsive plasma polymerized ultrathin nanocomposite films, Polymer 53 (2012) 4686–4693. D.T. Tran, S. Mori, M. Suzuki, Characteristics of polyimide-based composite membranes fabricated by low-temperature plasma polymerization, Thin Solid Films 516 (2008) 4384–4390. R.M. France, R.D. Short, Plasma treatment of polymers: the effects of energy transfer from an argon plasma on the surface chemistry of polystyrene, and polypropylene. A high-energy resolution X-ray photoelectron spectroscopy study, Langmuir 14 (1998) 4827–4835. A.J. Friedenstein, K.V. Petrakova, A.I. Kurolesova, G.P. Frolova, Heterotopic transplants of bone marrow, Transplantation 6 (1968) 230–247. H.S. Shirazi, N. Rogers, A. Michelmore, J.D. Whittle, Furfuryl methacrylate plasma polymers for biomedical applications, Biointerphases 11 (2016) 031014. A. Michelmore, J.D. Whittle, R.D. Short, R.W. Boswell, C. Charles, An experimental and analytical study of an asymmetric capacitively coupled plasma used for plasma polymerization, Plasma Process. Polym. 11 (2014) 833–841. G. Beamson, D. Briggs, High Resolution of Organic Polymers, The Scienta ESCA 300 Database, John Wiley and Sons, New York, 1992. S. Saboohi, S.A. Al-Bataineh, H. Safizadeh Shirazi, A. Michelmore, J.D. Whittle, Continuous-wave RF plasma polymerization of furfuryl methacrylate: correlation between plasma and surface chemistry, Plasma Process. Polym. 14 (2016) 1600054. B. Derjaguin, Theory of the stability of strongly charged lyophobic sols and the adhesion of strongly charged particles in solutions of electrolytes, Acta Physicochim. USSR 14 (1941) 633–662. E. Verwey, J.T. Overbeek, Theory of the Stability of Colloidal Dispersions, Elsevier, Amsterdam, 1948. M. Elimelech, J. Gregory, X. Jia, Particle Deposition and Aggregation: Measurement Modelling and Simulation, Butterworth-Heinemann, 2013. B.N. Chapman, Glow Discharge Processes: Sputtering and Plasma Etching, Wiley, 1980. M.A. Lieberman, A.J. Lichtenberg, Principles of Plasma Discharges and Materials Processing, John Wiley & Sons, 2005. R.V. Goreham, A. Mierczynska, L.E. Smith, R. Sedev, K. Vasilev, Small surface nanotopography encourages fibroblast and osteoblast cell adhesion, RSC Adv. 3 (2013) 10309–10317. J. Yang, F.R. Rose, N. Gadegaard, M.R. Alexander, A high-throughput assay of cell-surface interactions using topographical and chemical gradients, Adv. Mater. 21 (2009) 300–304. R.J. McMurray, N. Gadegaard, P.M. Tsimbouri, K.V. Burgess, L.E. McNamara, R. Tare, K. Murawski, E. Kingham, R.O.C. Oreffo, M.J. Dalby, Nanoscale surfaces for the long-term maintenance of mesenchymal stem cell phenotype and multipotency, Nat. Mater. 10 (2011) 637–644.