Hydrogels with Lotus Leaf Topography: Investigating Surface


Hydrogels with Lotus Leaf Topography: Investigating Surface...

0 downloads 149 Views 28MB Size

Subscriber access provided by UNIV OF REGINA

Article

Hydrogels with Lotus Leaf Topography: Investigating Surface Properties and Cell Adhesion Miriem Santander-Borrego, Elena Taran, Audra M. A. Shadforth, Andrew Keith Whittaker, Traian V Chirila, and Idriss Blakey Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b03547 • Publication Date (Web): 05 Dec 2016 Downloaded from http://pubs.acs.org on December 8, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Hydrogels with Lotus Leaf Topography: Investigating Surface Properties and Cell Adhesion Miriem Santander-Borrego,† Elena Taran ,ǁ,† Audra M. A. Shadforth,§ Andrew K. Whittaker,†,‡,§ Traian V. Chirila,†,§,ß,#,¥ and Idriss Blakey*,†,‡,§



The University of Queensland, Australian Institute for Bioengineering and Nanotechnology, St Lucia,

Queensland 4072, Australia. ‡

The University of Queensland, Centre for Advanced Imaging, St Lucia, Queensland 4072, Australia.

ǁ

Australian National Fabrication Facility-Queensland Node, St Lucia, Queensland 4072, Australia.

β

The University of Queensland, Faculty of Medicine and Biomedical Sciences, Herston Road, Herston,

Queensland 4029, Australia. §

Queensland Eye Institute, 140 Melbourne Street, South Brisbane, Queensland 4101, Australia.

#

Queensland University of Technology, Science & Engineering Faculty, 2 George Street Brisbane,

Queensland, 4001, Australia. ¥

The University of Western Australia, Faculty of Science, Crawley, Western Australia 6009, Australia.

ABSTRACT: The interactions of cells with the surface of materials is known to be influenced by a range of factors that include chemistry and roughness, however it is often difficult to probe these factors

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

individually without also changing the others. Here we investigate the role of roughness on cell adhesion while maintaining the same underlying chemistry. This was achieved by using a polymerization in mold technique to prepare poly(hydroxymethyl methacrylate) hydrogels with either a flat topography, or a topography that replicated the microscale features of lotus leaves. These materials were then assess for cell adhesion and then atomic force microscopy and contact angle analysis were then used to probe the physical reasons for the differing behavior in relation to cell adhesion.

KEYWORDS: hydrogels, lotus leaf, cell adhesion, atomic force microscopy, replica molding

INTRODUCTION The remarkable physical properties of natural materials have been of significant interest to scientists. For instance, spider silk has exceptional mechanical properties,1,2 while mussel proteins3 or Gecko feet4 inspired development of tissue adhesives. The leaves of lotus and some other plants exhibit non-stick behavior, while their surface is superhydrophobic and displays self-cleaning properties.5,6 These properties have been attributed to the topography of the leaf surface, which exhibits hierarchical roughness at the meso- and nanoscales. When an artificial hierarchical structure is produced using a hydrophobic substratum the surfaces have been shown to become superhydrophobic.7-9 Such surfaces exhibit increased protein adsorption, but can also exhibit lower cell adhesion when compared to flat surfaces with the same chemistry.10 On the other hand, the hydrogels with a flat topography are wettable, promote low non-specific protein binding and generally exhibit low cell adhesion.

ACS Paragon Plus Environment

2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The attachment, spreading and growth of cells on solid substrata, and the cells’ response to the topography and chemistry of the substratum surfaces has been a field of intense activity since the early stages of cell culture methodology,11-14 leading to the concept of “contact guidance”.15,16 Modern research confirmed that the presence of topographic patterns on the surface of substrata influence – generally in a positive way – the process of cells’ adhesion and growth and is of crucial importance for the anchorage-dependent cells; that the serum proteins deposited at the substratum-cell interface mediate the process; and that the chemical modification of surfaces can affect significantly the process.17-30 With the advent of the ability to tailor polymer structure, such materials have become the preferred substrata for growing cells in laboratory. The literature is replete with successful in vitro cell adhesion experiments reported on polymer surfaces of a great diversity, such as rigid or flexible, hydrophilic or hydrophobic, nonbiodegradable or biodegradable, neutral or ionic, positively charged or negatively charged, porous or non-porous. There have been also many reports on polymer surfaces not allowing the attachment of cells, or allowing attachment but not the subsequent growth. The results were sometimes contradictory, and the reasons for cell growth, or lack of it, have not been satisfactorily explained in most of the cases. At the same time, the reported enhancement of cell adhesion induced by increased roughness of the substratum surface is based on results that are often confounded by the changes in surface chemistry that accompany the methods used to alter the surface topography. We investigated poly(2-hydroxyethyl methacrylate), henceforth PHEMA, as a hydrogel substratum with hierarchical roughness over the meso- and nanoscales that was generated by taking replica-molds of lotus leaves. There have been previous attempts to render the surface of

ACS Paragon Plus Environment

3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

PHEMA31-38 or other hydrogels39 to be superhydrophobic, but none of them took the lotus leaf as a model, and both procedures and aims were different from those reported here. In our procedure, Teflon AF was used to form a mold of the lotus leaf. Being a perfluorinated polymer, Teflon AF is effective for generating molds, because it is incompatible with both hydrophilic and hydrophobic materials, which facilitates easy processing during the generation of the mold. The properties of Teflon AF are also favourable for the replication process, where here we have used a polymerization in mold-nanoimprint lithography (PIM-NIL) technique to prepare replicas of the lotus leaf topography in PHEMA. Hence, we can prepare hydrogels with the same surface chemistry that are either flat, or have a hierarchical surface topography. This allows us to evaluate the effect of surface topography on cell adhesion and then investigate the factors that may contribute to this by studying the effect of topography on contact angle and the adhesion of cells to the surface. EXPERIMENTAL SECTION Materials. Commercial dried lotus leaves were obtained from a Chinese market in Brisbane, Australia. Teflon AF 2400, a copolymer of 2,2-bis(trifluoromethyl)-4,5-difluoro-1,3dioxole and tetrafluoroethylene, was a gift from DuPont. 2,2,3,3,4,4,5-Heptafluoro-5(1,1,2,2,3,3,4,4,4-nonafluorobutyl)tetrahydrofuran was obtained as Fluorinert FC-75 solvent from Fluorochem, UK. Ammonium persulfate, ethylene glycol dimethacrylate (EGDMA), HEMA, and N,N,N′,N′-tetramethylethylenediamine (TEMED) were all supplied by SigmaAldrich, St Louis, MO, USA. Fabrication of a Negative Mold of the Lotus Leaf. The lotus leaves were rinsed with deionized water and air dried before use. The leaves were cut into pieces of 40 x 45 mm. Teflon AF powder (35.3 mg) was dissolved in FC-75 (1 mL) with constant stirring for 5 h at room

ACS Paragon Plus Environment

4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

temperature. To prepare the mold, the fluoropolymer solution was dispensed on the top of a piece of lotus leaf, and then a second piece of lotus leaf was placed on the top of this solution. A weight was applied to this stack until the solvent had evaporated. The dried negative mold was then peeled off the lotus leaf master templates. Fabrication of PHEMA Hydrogels Replicating the Topography of the Lotus Leaf. The Teflon negative mold of the lotus leaf was fixed onto a glass plate using double-sided tape. An O-ring was then placed on the plate surrounding the negative mold. To make a fluid-tight reactor, a second glass plate was added to the assembly, which was then held together with 4 clips. HEMA (28 g, 0.125 mol), water (12 g), EGDMA (132 µL, 0.7 mmol) as a crosslinking agent, ammonium persulfate (420 µL, 3.7 mmol) and TEMED (54 µL, 0.36 mmol) as components of the initiator system were mixed in a vial. This polymerization mixture was then injected into the reactor assembly containing the Teflon negative mold through the O-ring using a needle and syringe, with the aid of a second needle to facilitate exit of gas. The polymerization was allowed to proceed at room temperature for 16 h. Fabrication of PHEMA Hydrogels with a Flat Topography. The fabrication of flat PHEMA surface was conducted using the same procedure as for the PHEMA replica, except that no negative mold was used. Scanning Electron Microscopy (SEM). The surface topography of the lotus leaves and Teflon negative molds was investigated using a Philips XL30 scanning electronic microscope. The samples were coated with gold before analysis. The images were acquired at acceleration voltages of 5 kV and 15 kV. Atomic Force Microscopy (AFM) Measurements. A MFP-3D (Asylum Research, Santa Barbara, CA, USA) atomic force microscope (AFM) was used to measure the pull-off

ACS Paragon Plus Environment

5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

forces between an unmodified tip, or a tip modified with a colloidal silica particle and the polymeric film. The AFM was mounted on an anti-vibrational table (Herzan, Laguna Hills, CA, USA) and operated within an acoustic isolation enclosure (TMC, Peabody, MA, USA). All the measurements were done in pure water at room temperature. The cantilevers used were silicon nitride from BudgetSensors® (Sofia, Bulgaria) having a nominal spring constant of 0.27 N/m and nominal resonant frequency of 30 kHz. The AFM probes were calibrated, before the measurements, by employing the thermal vibration method embedded in the Asylum Research AFM software. To characterize the adhesion force and modulus, arrays of 20 × 20 force curves were recorded over an area of 50 × 50 µm. To eliminate capillary forces, all the measurements were performed in deionized water. An adhesion histogram was generated from the results. By fitting a Gaussian distribution to the histogram, the mean and standard deviation of the pull-off force were calculated. All measurements were performed at an applied load of 10 nN and using a scan velocity of 1.5 µm/s. The surface topography of the films was examined in air by employing the contact mode of the AFM using HA_NC (Etalon series cantilevers from NT-MDT (Zelenograd, Russia)) having a nominal resonant frequency of 110 kHz. The root-mean-square (RMS) roughness on the micro- and nanoscale topography was measured in an area of 2 × 2 µm. All surfaces were examined in triplicate. Static Water Contact Angle and Hysteresis Contact Angle. Contact angle measurements were acquired using a goniometer (model OCA 20, DataPhysics Instruments GmbH, Filderstadt, Germany). Static contact angles were measured using the sessile drop method,40 where a 5-µL drop of water was used. Advancing (θa) and receding (θr) contact angles

ACS Paragon Plus Environment

6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

were measured by initially placing a 2-µL drop of water on the surface, and then the volume was slowly increased to 10 µL. Subsequently, the water drop was gradually decreased. Five replicates were taken for each sample. SCA-20 (version 3.7.4) software (DataPhysics) was used to process the data. In vitro Cell Attachment Assays. Sample disc preparation. Circular samples, 4 mm in diameter, were cut from films of the flat PHEMA control and PHEMA lotus replicas using a hole biopsy punch. These were then placed in O-rings with a 6-mm external diameter and 4-mm internal diameter. This configuration allowed for easy transfer and correct orientation of the discs, while minimizing handling. The O-rings also fit snugly within the wells of 96-well plates. The samples were sterilized by complete immersion in 70% ethanol for 2 h and then aseptically washed in phosphate buffered saline (PBS) and stored in PBS until the cell attachment assay was carried out. Cell culture. The human corneal epithelial cell line, HCE-T (Riken Cell Bank, Japan),

was used to assess cell attachment and viability. This cell line was established by immortalizing human corneal epithelial cells with the SV40-adenovirus vector.41 HCE-T has been extensively used as a standard model for drug permeation, bioavailability, pre-screening and toxicity assessment.42-46 The cell line was routinely cultured in DMEM/F12, GlutaMAXTM medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA). Prior to setting up the cell attachment assay, HCE-T cells were serum-starved overnight. This step involved washing the cells with Hank’s buffered saline solution (HBSS) (Invitrogen, Carlsbad, CA) and adding fresh serum-free medium into the flask (DMEM/F12, GlutaMAXTM medium + 1% Penicillin/Streptomycin).

ACS Paragon Plus Environment

7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

Cell Adhesion to Different PHEMA Surfaces. Cell adhesion on the flat PHEMA controls and the PHEMA lotus replicas were analysed at 4, 24 and 48 h following cell seeding. Three replicates per sample were placed in separate wells of three 96-well plates, one plate for each time point. Serum-free medium was added to each well and ~12,000 cells were seeded per disc (this equates to ~100,000 cells/cm2, as each disc had an area of ~0.12 cm2). A standard curve was set up on each plate using a defined percentage of seeding density under the same serum-free conditions on tissue culture plastic. The number of cells on each flat PHEMA control and PHEMA replicas were determined using the standard curve that was prepared for each time point. The 96-well plates were incubated at 37 °C in 5% CO2. At each time point (4, 24 and 48 h), the respective plates were washed gently with HBSS, and representative areas were photographed. Subsequently, cell numbers (illustrated by the DNA content) were measured and quantified using the Quant-iT™ PicoGreen® dsDNA assay kit following the manufacturer’s instructions (Invitrogen, Carlsbad, CA). The blank wells contained 100 µL of Tris EDTA buffer and 100 µL of 1:200 Quant-iT™ PicoGreen® dye diluted 1:200. All samples were tested in triplicate and the fluorescence was read with a fluorescence plate reader (FLUOstar OPTIMA, BMG Labtech, Ortenberg, Germany) using and excitation wavelength of 480 nm, and an emission wavelength of 520 nm. RESULTS AND DISCUSSION Polymerization in Mold-Nanoimprint Lithography (PIM-NIL). Figure 1a shows a schematic of the PIM-NIL process for preparing PHEMA replicas that mimic the topography of the lotus leaf. To prepare the mold, a solution of the perfluoropolymer Teflon AF in FC-75 was poured onto a lotus leaf master and the solvent was allowed to evaporate. A perfluorinated

ACS Paragon Plus Environment

8

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

solvent such as FC-75 is a non-solvent for non-fluorinated compounds, so is an ideal agent for casting molds to capture surface topography because the swelling or dissolution of the master is avoided. For example, Khang and coworkers were able to fabricate molds of patterned polymeric photoresist, poly(methyl methacrylate) and polystyrene by casting solutions of Teflon AF in a perfluorinated solvent.47 Removal of the mold from the lotus leaf master was facile due to the very low surface energy of the perfluoropolymer (15.6 dyn/cm).48 Teflon AF molds prepared in this manner have previously been used in low pressure nanoimprint lithography, but here we test whether the molds can be used to carry out PIM-NIL. Teflon AF 2400 is neither soluble nor swollen by conventional organic solvents, has a glass transition temperature of 240 °C, a dynamic modulus of approximately 1 GPa in the temperature range 0–100 °C, and is chemically inert, therefore should be well suited for this application.48-51 Unlike other methods which use PDMS to capture the features of the lotus leaf,7 the method reported here does not require the negative mold to be coated with an anti-stick release agent prior to generating the replica. To prepare the PHEMA lotus replicas a solution of HEMA, EGDMA, redox initiator and co-initiator were poured into the mold, which was mounted between two glass plates separated by an O-ring. After deoxygenation by sparging with nitrogen, the mixture was allowed to polymerize. After polymerization, the PHEMA replica was peeled off the mold, which again occurred without sticking due to the low surface energy of the perfluoropolymer. While not the focus of this study, we were also able to produce lotus replicas from hydrophobic monomers such as methyl methacrylate.

ACS Paragon Plus Environment

9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

Figure 1. (a) Schematic representation of the replica-molding technique: (i) a solution of Teflon AF in FC-75 was added to the lotus leaf master and the FC-75 was allowed to evaporate; (ii) the remaining Teflon AF mold was peeled off the lotus leaf master; (iii) A mixture of HEMA, water, EGDMA and redox initiator system was added to the Teflon AF mold and allowed to polymerize; (iv) the PHEMA replica was peeled off the Teflon AF mold. (b) Representative SEM micrographs of the lotus leaf master and (c) the Teflon AF mold. AFM topographical maps of (d) the lotus leaf master, (e) the Teflon AF mold, (f) the PHEMA replica, and (g) the flat PHEMA control. Characterization of Surface Topography. The surface topography of the lotus leaf and the Teflon AF lotus negative molds were examined using SEM (Figure 1b and 1c, respectively).

ACS Paragon Plus Environment

10

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The lotus leaf surface has a hierarchy of structures, which are comprised of microscale protrusions and nanoscale features that cover the entire surface, which is consistent with literature reports.52,53 The surface of the negative molds had cavities with an average width of 9±1 µm across the entire surface, which was consistent with a negative imprint of the characteristic protrusions of the lotus leaf topography. Negative imprints of other features present at the surface of the lotus leaves were also evident. These included the stomata structures from the lotus leaf surface, which are pores that occur in the epidermis of leaves, stems and other organs of plants. The presence of these complex structures demonstrated the high fidelity of the replication process at a micron scale. AFM was then used to examine the surface of the lotus leaf, Teflon AF lotus-leaf mold, PHEMA lotus replica and the flat PHEMA (Figure 1 d-g). The topographical features are analyzed in Table 1. As expected, the flat PHEMA control exhibited no mesoscale features and the RMS roughness was determined to be 0.003±0.001 µm.

When the dimensions of the

protrusions of the lotus leaf (10±2 µm wide, 3±1 µm high) and PHEMA lotus replica (9±2 µm wide, 3±1 µm high) are compared, it can be seen that they match closely within the experimental error. These values also correspond well with the width of the holes in the Teflon AF mold (9±1 µm wide). Another way to compare the master and the replicas was to measure the microscale RMS roughness, which was determined from statistical analysis of the AFM images. Again, within the experimental error, the lotus leaf (2.5±0.4 µm) and the PHEMA lotus replica (2.2±0.2 µm) have comparable values. On the other hand, the nanoscale roughness, which was measured between the protrusions, had slightly decreased going from the lotus leaf (0.37±0.06 µm) to the PHEMA lotus replica (0.30±0.01 µm). This small decrease may be attributed to distortions induced by changes in swelling of the hydrogel, or an inability of the molding process to capture

ACS Paragon Plus Environment

11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

more complex nanostructures such as co-continuous cavities. Nonetheless, based on these measurements it can be concluded that using a PIM-NIL process, we have been successful in generating a hydrogel with a surface topography that is hierarchical in nature with both mesoand nanoscale roughness, which is a close approximation of the lotus leaf.

Table 1. Summary of topographical features of lotus leaf master, Teflon AF negative mold, PHEMA replica and flat PHEMA control. Samples

Protrusion dimensions (µm)

RMS roughness (µm)

Height

Width

Microscale b

Nanoscale c

Lotus leaf

3±1

10 ± 2

2.5 ± 0.4

0.37 ± 0.06

Teflon AF negative lotus mold

3 ± 1d

9 ± 1b

-

0.19 ± 0.03

PHEMA lotus replica

3±1

9±2

2.2 ± 0.2

0.30 ± 0.01

Flat PHEMA control

-

-

-

0.003 ± 0.001

d

a

measured from SEM images.

b

calculated from AFM scans.

c

calculated from regions between the protrusions in the AFM scans.

This corresponds to the depth of the negative mold of the protrusions

Assessment of Cell Adhesion. To study the effect of surface topography on cell adhesion, the flat PHEMA control and the PHEMA lotus replica were incubated with the HCE-T cells. PHEMA is a well known biomaterial with many real or putative applications, and it is considered the classic epitome of the biomedical hydrogels. However, it has a major drawback: PHEMA surface displays poor cell-adhesive properties. In this study we investigated a dual

ACS Paragon Plus Environment

12

Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

strategy to influence these properties: (a) biomimetically structuring the surface topography in order to generate roughness on meso- and nanoscales, and (b) changing the hydrophilicity of the surface by imparting superhydrophobic properties. These two strategic components are intrinsically related. Lydon, Tighe and co-workers demonstrated19,54,55 with a significant number of cell types that flat and smooth PHEMA surfaces poorly promote cell attachment, and do not support spreading and growth, a finding soon confirmed in other studies.55,57 They compared PHEMA to copolymers of HEMA and other hydrogels, and noticed that if the equilibrium water content (EWC) is taken as a criterion, the hydrogels with EWC between 2 and 35% and, with some exceptions, between 60 and 90% supported cell adhesion.19,55 Regarding those hydrogels with EWC between 35 and 60%, including PHEMA (EWC 40%), cells did attach to their surface but eventually they became rounded and no further proliferation could be seen. In the case of hydrogels,

a

number

of

factors

must

be

considered

besides

EWC,

such

as

hydrophilicity/hydrophobicity, wettability (gauged by the contact angle), hydroxyl group expression, and surface polarity. For PHEMA, it has been shown that EWC is related to surface polarity,55 but not to its wettability.58 Presumably, water absorption to hydrogel surfaces may affect the macromolecular conformation in a way that can be relevant to cell behavior. Hydroxyl group expression is rather too deceptive to consider. For instance, polystyrene as such is a noncell-adhesive material, but treating its surface with sulfuric acid renders it into the ideal substratum for any eukaryotic cell, the traditional “tissue culture plastic” used worldwide in tissue culture laboratories. It was demonstrated that the acid treatment generated hydroxyl groups on the surface of polystyrene, and was assumed that this factor alone led to cell adhesivity.59 However, when we consider that a clear correlation between the chemical nature of functional

ACS Paragon Plus Environment

13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

groups expressed at the polymer surface and the cell adhesion has yet to be established,21 it should be perhaps not so intriguing that the abundant inherent availability of hydroxyl groups expressed on the PHEMA’s surface does not have the same beneficial effect on cell proliferation as it had in polystyrene. After all, the groups that mediate the attachment of cells are not necessarily involved in their spreading and growth. There have been few studies regarding cell behavior in contact with superhydrophobic surfaces of synthetic polymers, and the findings are sometimes hard to reconcile. An early study60 on polydimethylsiloxane (PDMS) rubber as such and on that rendered either superhydrophobic (by laser treatment) or superhydrophilic (by chemical modification) found that the adhesion on blood platelets to both superhydrophobic (rough) and superhydrophilic (smooth) PDMS surfaces was about ten times less than to virgin PDMS. In another study,61 nanostructures created on the surface of a fluorinated poly(carbonate urethane) by using carbon nanotubes templates led to superhydrophobic surfaces that prevented completely the adhesion of platelets. In a series of studies62-64 based on poly(L-lactic acid ) (PLLA) where the superhydrophobicity was induced by creating micro- and nanoscale geometric features, it has been shown that cell adhesion on these surfaces was significantly reduced. It was surmised that the cells prefer to grow on larger, rather than small, adhesive islands, where they are allowed to distend properly. While they may also prefer to anchor to topographic asperities, this can be negatively affected by the existence of air pockets entrapped within the cavity-type topographic features. Similar findings have been reported64,65 on superhydrophobic polystyrene with some cell types; however, a cell line (L929) was able to even proliferate.65 Other investigators reported enhanced cell attachment and growth on the superhydrophobic surfaces of certain synthetic polymeric materials, such as fluorinated Teflon-coated poly(ethylene terephthalate),66,67 or perfluorinated

ACS Paragon Plus Environment

14

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

poly(3,4-ethylenedioxythiophene),68 where the surfaces have been nano-structured by various methods. Two recent comprehensive reviews69,70 have summarized the information gathered on the issues of cell adhesion and protein adsorption on superhydrophobic surfaces and their relevance to biomedical applications, and have attempted to explain the variability of reported results. Essentially, the two processes depend on the roughness of the surface and the stability of the air-pockets barrier. These factors can have opposite effects: while rough topographies may enhance protein adsorption and/or cell adhesion, the presence of air pockets can inhibit the same processes. In the case of hydrogels, the material and its limiting surface are already saturated with water, and fully hydrated PHEMA is a typical example. In regards to superhydrophobic PHEMA surfaces, their application as arrays32-35,38 was based on creating superhydrophilic spots separated by superhydrophobic barriers, where the former support cell adhesion while the latter inhibit it. However, in the present study we found that PHEMA with a flat topography exhibited cell counts that were effectively static over this period and significantly lower than those on the PHEMA lotus replica (Figure 2). The morphology of the cells was assessed after 48 h using optical microscopy. The cells adhering to the PHEMA with a flat topography exhibited a rounded morphology (Fig. 2 b), while those adhered to the rough PHEMA surfaces exhibited a cobblestone morphology that is typical of HCE-T cells (Fig. 2 c). A rounded morphology indicates poor cell-substratum affinity.

ACS Paragon Plus Environment

15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

Figure 2. (a) Analysis of cell attachment on flat and microstructured PHEMA surfaces. Representative optical micrographs of HCE-T cells adhered on (b) flat and (c) microstructured PHEMA surfaces at 48 h. A possible explanation for our results is that the microstructured PHEMA surface provides physical cues that facilitate HCE-T cell adhesion which would not be expected to occur on the flat and smooth surface. To gain an insight into the increased cell adhesion to the hierarchically structured PHEMA lotus replicas, contact angle experiments as well as AFM evaluation of Young’s modulus were carried out. Contact Angle Measurements. The static and advancing/receding contact angles for the flat and lotus replica PHEMA are listed in Table 2. The static contact angle of the flat PHEMA control was 65±3°, which is similar to reported in the literature values of 61°.21,71 On the other hand, the contact angle of the PHEMA lotus replica was significantly higher (86±2°). This result is in contrast to poly(methyl methacrylate) substrata that replicate the microstructure of the lotus

ACS Paragon Plus Environment

16

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

leaf but not the nanostructure between the microscale asperities, which exhibits a lower contact angle (70°) compared to flat films (75°).72 To gain insight into the reasons for the differences, the contact angles were considered in terms of both the Wenzel and Cassie-Baxter models.8,73-75 Wenzel’s model takes into consideration the roughness of the surface r (r ≥ 1) and an assumption is made that water displaces air that is present between any asperities. Hence, the Wenzel contact angle (θ W) of a rough surface can be modelled in terms of the Young’s contact angle of a flat surface (θ Y) using the following equation:  =   However, for the PHEMA lotus replica this gives an r value of less than 1, because θ

W

is

larger than  . This result then suggests that the Cassie-Baxter model may be more appropriate and that air may be trapped between the micron sized asperities. For the Cassie-Baxter model the droplet is assumed to be in a fakir state, where the water droplet only touches the apexes of the asperities, leaving air trapped beneath the droplet. The Cassie-Baxter contact angle (θC-B) can be modeled using the following equation below.  =  + 1 − 1 In this equation, f is the fractional area between the substratum and droplet. Using the contact angles obtained in this study, the Cassie-Baxter model gives an f value of 0.75. This value is significantly higher than that obtained for a lotus leaf, which has an f value of 0.063 (using θ Y of 103° for carnauba wax and an apparent contact angle of 162° for a lotus leaf). The difference between these values suggests that water is in contact with a greater fraction of the asperities than for the natural leaf. This higher degree of contact is probably due to PHEMA being more

ACS Paragon Plus Environment

17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

hydrophilic than carnauba wax, but the deviation from unity also suggests that air pockets are present.

Table 2. Summary of the contact angle measurements. Young’s contact angle (°)

Advancing contact angle (°)

Receding contact angle (°)

Contact angle hysteresis (°)

Flat PHEMA control

65 ± 3

70 ± 2

25± 2

45 ± 4

Flat PHEMA (lit) 57,75

61

58-84

9-21

PHEMA lotus replica

86 ± 2

95 ± 1

80 ± 3

Sample

15 ± 4

This situation was further investigated by undertaking dynamic contact angle measurements. The advancing, receding and hysteresis contact angles of the flat PHEMA were 69.8±2°, 25±2°, and 45±1°, respectively. This leads to a hysteresis of 45±4°. These values are consistent with values reported elsewhere for PHEMA.58,76 The significant hysteresis can be attributed to interactions between PHEMA and water and to structural rearrangement of functional groups at the air-hydrogel and water-hydrogel interfaces in the presence and absence of water. This phenomenon has also been observed for other hydrogels such as gelatin.77 The advancing and receding contact angles for the PHEMA lotus replica were 95±1° and 80±3° respectively, which corresponds to a much smaller hysteresis of 15±4°. This smaller value is also consistent with the presence of trapped air. The hysteresis of the PHEMA lotus replica is higher than reports for lotus leaves (3°), however, it is significantly lower than the hysteresis of surfaces that replicate both the microstructure and surface chemistry of the lotus leaf, but have no inherent nanostructure between the micron sized asperities (29±2°).8 Hence, it appears that the

ACS Paragon Plus Environment

18

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

hierarchical structure of the PHEMA lotus replica is playing a key role in dictating the interactions of water with the surface, despite the chemistry of the substratum being relatively hydrophillic. Whilst the hysteresis observed for the PHEMA lotus replica is higher than what is arbitrarily considered superhydrophobic (< 10°),78 a value of 15±4° is still quite remarkable for a hydrogel such as PHEMA. Probing Adhesion and Young’s Modulus with AFM. Figure 3 shows the results of the AFM studies of the Young’s modulus and adhesion of the flat PHEMA control and the PHEMA lotus replica which were carried out in distilled water to eliminate capillary forces.

The

frequency distribution plot of the Young’s modulus for the flat PHEMA control sample (Figure 3 a) shows a unimodal distribution with an average of 1.5 MPa. The 2D map (Figure 3 b) reflects this result, showing a homogeneous distribution of the modulus across the surface. On the other hand, the PHEMA replica of the lotus exhibits a trimodal population distribution, with populations centered at