Here, we present a protocol for the fabrication of gradient nanopattern plates via thermal nanoimprinting and the method of screening responses of human endothelial progenitor cells to the nanostructures. By using the described technology, it is possible to produce a scaffold that can manipulate cell behavior by physical stimuli.
Nanotopography can be found in various extracellular matrices (ECMs) around the body and is known to have important regulatory actions upon cellular reactions. However, it is difficult to determine the relation between the size of a nanostructure and the responses of cells owing to the lack of proper screening tools. Here, we show the development of reproducible and cost-effective gradient nanopattern plates for the manipulation of cellular responses. Using anodic aluminum oxide (AAO) as a master mold, gradient nanopattern plates with nanopillars of increasing diameter ranges [120-200 nm (GP 120/200), 200-280 nm (GP 200/280), and 280-360 nm (GP 280/360)] were fabricated by a thermal imprinting technique. These gradient nanopattern plates were designed to mimic the various sizes of nanotopography in the ECM and were used to screen the responses of human endothelial colony-forming cells (hECFCs). In this protocol, we describe the step-by-step procedure of fabricating gradient nanopattern plates for cell engineering, techniques of cultivating hECFCs from human peripheral blood, and culturing hECFCs on nanopattern plates.
Recently, the response of cells by the physical stimulation of surface topography has been spotlighted in the field of cell engineering1,2,3,4. Therefore, more attention has been focused on three-dimensional nanostructures at the cell attachment surface5. It has been reported that the integrin, which is the surface recognition device of the cell, transmits the physical stimulus driven by the micro-nano structures of ECM through mechano-transduction6. This mechanical stimulation regulates cell behavior through contact guidance7 and induces cytoskeletal reorganization to change shape, in addition to focal adhesions and stiffness of cells8.
Human endothelial progenitor cells (hEPCs) in the body closely interact with the microenvironment of the surrounding ECM9. This indicates that the physical state of the ECM acts as an important parameter for specific cell-matrix adhesion complex formation as much as shear stress derived from blood flow10. It is reported that surface nanotopography enhances the in vitro formation of extensive capillary tube networks of hEPCs11 and that an ECM/bio soluble factor combined system enables hEPCs to recognize dysfunctional substrates and promotes wound healing12,13. Nonetheless, the relationship between ECM and hEPCs is not clearly understood.
Although many researchers tried to clarify the relationship between cell responses and physical cues from different substrates14,15,16, these studies used only the fixed size of a nanostructure or nanopatterns with irregular arrangements that have a limitation to elucidate the relationship between the size of the nanostructure and cell behavior. The problem here is a lack of suitable tools for screening cellular responses that can replace existing tedious and iterative approaches to find the optimum size of the nanostructure. Therefore, a straightforward technique is required for screening cell reactions on physical stimulations without repetition.
Here, we describe a method used in our previous reports17,18,19 to produce a gradient nanopattern in which the diameter of the arranged nanopillars gradually increases. In addition, we also described how to cultivate and analyze the behavior of hECFCs on gradient nanopattern plates to determine the effect of physical stimuli on the cells. A mild anodization, gradual etching, and anti-sticking layer coating method were used to fabricate gradient AAO mold. By adopting a thermal imprinting lithography technique, identical polystyrene gradient nanopatterns were produced in a cost-effective and facile way. Using gradient nanopatterns, it is feasible to determine which size of nanostructure has a great effect on cell behavior in one set of experiment. We expect that this gradient nanopattern will be helpful in understanding the interaction mechanisms between blood-derived hECFC or other cells and various sizes of nanostructures.
This study was approved by the Institutional Review Board at Korea University Anam Hospital (IRB No. ED170495). All procedures were carried out in accordance with the Helsinki Declaration and its later amendments.
1. Preparation of Aluminum (Al) Substrate by Electropolishing
Caution: Electropolishing solution is corrosive and toxic. Wear personal protective equipment including nitrile gloves, goggles and lab coat. Perform this step in a fume hood.
2. Fabrication of Gradient AAO Mold with Phosphoric Acid Electrolyte
Caution: Methyl alcohol and its fume are ocular toxic. Continuous exposure to chromium can lead to serious chromium poisoning. Perform this step in a fume hood.
3. Deposition of Anti-Sticking Layer on Gradient AAO Mold with Self-Assembled Monolayer
Note: Perform steps 3.2.1 to 3.3.3 in a glove box. Connect a vacuum pump and dry nitrogen gas injector to the glove box. Place all samples, reagents, and apparatuses in the glove box prior to the dehumidification process. Repeat the evacuation and nitrogen gas injection cycle more than three times to adequately remove moisture from the glove box. Let the dry nitrogen flow through the experiment.
4. Fabrication of Gradient Nanopattern Plates by Thermal Imprinting
Note: Perform steps 4.2 to 4.7 in a clean room.
5. Sterilization and Hydrophilic Modification of Gradient Nanopattern Plates
6. Cultivation of hECFCs
Note: Conduct all centrifuging procedures at 4 °C unless otherwise noted.
7. Cell Seeding and Culture on the Gradient Nanopattern Plates
Note: Step 7 describes the culture of hECFCs on the gradient nanopattern plate, but other cell sources also can be used.
8. Observation and Analysis
Figure 1 shows SEM images of the fabricated gradient AAO molds according to their type and position. Figure 2 shows SEM images of gradient nanopattern plates with regular-rounded nanopillars, and Figure 3 is quantification data of the nanopillar diameter. Table 1 lists the characteristics of the fabricated nanopillars.
Figure 4A shows the scheme of cultivation of blood-derived hECFCs. Figure 4B shows the phase contrast image of thecultivated hECFCs. Figure 4C shows hECFC markers stained with CD144 (green), vWF (red), and nucleus (blue). Figure 5A shows representative SEM images of hECFCs after 2 days of culture on the flat, GP 120/200, GP 200/280, and GP 280/360 plates. Figure 5B and C depict quantitative data on cell area and perimeter as compared to the flat. The quantification data revealed that cells grown on nanopillar surfaces of smaller size showed reduced cell area and perimeter. Figure 5D is the representative SEM images of hECFCs which show the morphology of existing filopodia. Figure 5E depicts the quantitative data on filopodia number as compared to the flat. Significant filopodial outgrowth was observed in cells cultured on GP 120/200, while no significant change was observed in GP200/280 or GP280/360 in comparison with the flat. Figure 5F shows representative TEM images of hECFCs after 2 days of culture on flat and gradient nanopattern plates.
Figure 1. Gradient AAO mold. SEM images of gradient AAO molds for GP 120/200, GP 200/280, and GP 280/360 plates according to the position of molds. Please click here to view a larger version of this figure.
Figure 2. Gradient nanopattern plates. SEM images of pillar-type gradient nanopattern plates for GP 120/200, GP 200/280, and GP 280/360 plates according to position of plates. Please click here to view a larger version of this figure.
Figure 3. Size gradient of nanopillar diameter. Quantification data showing the gradient of nanopillar diameters for GP 120/200, GP 200/280, and GP 280/360 plates, respectively. Bars represent standard error. This figure was modified from [Acta Biomaterialia]20. Please click here to view a larger version of this figure.
Young’s modulus (GPa) | Nanopillar Diameter (nm) | Nanopillar Pitch (nm) | Center to Center Distance (nm) | Nanopillar height (nm) | Nanopillar stiffness (N/m) | |
Flat | 2.43±0.19 | – | – | – | – | – |
GP 120/200 | 1.74±0.11 | 120–200 | 320–240 | 440 | 276.9±21.9 | 9.35 |
GP 200/280 | 2.01±0.05 | 200–280 | 240–160 | 440 | 268.1±25.9 | 55.78 |
GP 280/360 | 2.1±0.13 | 280–360 | 160–80 | 440 | 293.6±23.6 | 134.15 |
Table 1. Characteristics of fabricated nanopillars. Quantified data showing Young's modulus, diameter, pitch, center to center distance, height, and stiffness of fabricated nanopillars. This table was modified from [Acta Biomaterialia]20.
Figure 4. Characterization of hECFCs. (A) Schematic diagram showing schedule of hECFCs derived from adult peripheral blood. (B) Phase contrast image of hECFC colony, derived from adult peripheral blood on day 14. (C) Immunofluorescent images showing CD144 (green), vWF (red), and cell nucleus stained with DAPI (blue). This figure was modified from [Acta Biomaterialia]20. Please click here to view a larger version of this figure.
Figure 5. hECFCs cultured on flat and gradient nanopattern plates for 2 days. (A) Representative SEM image of hECFCs cultured on Flat, GP 120/200, GP 200/280, and GP 280/360 plates. (B and C) Quantification data of the area and perimeter of hECFCs (n = 50). *p < .05 as compared to the Flat. (D) Representative SEM images of the leading edge of hECFCs on flat and gradient nanopattern plates. (E) Quantification data of the number of filopodia per one hECFC (n = 20). *p < .05 as compared to the Flat. (F) Representative TEM image of hECFC cultured on Flat and gradient nanopattern plates. White arrows indicate the interaction points between hECFCs and substrate surface. This figure was modified from [Acta Biomaterialia]20. Please click here to view a larger version of this figure.
Supplementary Figure 1. Water contact angle of pristine AAO and HDFS treated AAO. Contact angle measurements showing hydrophobic properties of HDFS self-assembled monolayer (A) before and, (B) after treatment. Please click here to download this figure.
Supplementary Figure 2. SEM images of fabricated gradient nanopattern plates with same AAO mold. Comparative SEM image sets for verifying the durability of AAO mold. (A) Gradient nanopattern plate produced with freshly made mold. (B) Gradient nanopattern plate produced with AAO mold after imprinting 15 times. Please click here to download this figure.
Supplementary Figure 3. Influence of protein coating. SEM images of hECFCs that were cultured on GP 120/200, GP 200/280, or GP 280/360 gradient nanopattern plates (A) without or (B) with the protein coating. This figure was modified from [Acta Biomaterialia]20 Please click here to download this figure.
Fabrication of an AAO often suffers from defects such as cracks, irregular shapes of pores, and burning. The main reason for these defects is called an electrolytic breakdown, which is strongly affected by the nature of the metal substrates being anodized and the resistivity of the electrolyte21. Since the resistivity of the electrolyte varies depending on its temperature, eliminating heat continuously from electrodes is the critical point to maintain the locational temperature of the electrolyte as stable in such a high-voltage anodizing condition. In this protocol, we fabricated AAO molds by modifying Masuda's two-step anodization22,23. Replacing almost half the electrolyte with methyl alcohol not only prevents the electrolyte from freezing at low temperatures but also deprives the electrode of heat by evaporation at the bottom of the pore24. The reason for using an overhead stirrer and impeller instead of a magnetic stirrer is also to ensure the temperature control. The typical magnetic stirrer has a high probability of idling owing to the accidental misalignment of the magnetic bar during long-term operation. This problem induces the improper circulation of the electrolyte, raising its temperature, and leading to burning.
To give a size gradient to the AAO mold, we gradually immersed the AAO in a phosphoric acid etching solution. The pores are widened in proportion to the time the AAO stays in the etching solution, as shown in Figure 1. Under this condition, the average pore-widening rate is 0.67 nm/min. Manipulating the immersing speed and time will change the slope of the size gradient and maximum pore diameter of the AAO. The maximum possible pore diameter is 400 nm. When the pore diameter exceeds 400 nm, the spacing wall between pores becomes so thin that it cannot tolerate the pressure during the thermal imprinting process.
An HDFS self-assembled monolayer is known to be able to add a hydrophobic property to the material surface25. The HDFS molecule makes a strong covalent bond with the hydroxyl groups on the AAO surface introduced by the Piranha treatment26. Therefore, as shown in Supplementary Figure 1, a solid self-assembled monolayer can be formed when a large number of hydroxyl groups are present on the substrate surface. For optimum quality with an HDFS self-assembled monolayer, we suggest conducting the reaction in a dehumidified atmosphere. When HDFS is exposed to moisture, a side reaction with water forms dimers, trimers, and even oligomers of silane molecules, and decreases the overall quality of the self-assembled monolayer.
Upon using the AAO mold with a size gradient in thermal nanoimprinting, cell culture plates with a gradient of nanopillars were obtained Figure 2 and Figure 3. The thermal nanoimprinting method has an advantage in that it can produce a large quantity of identical nanopattern plates by using one AAO master mold. Supplementary Figure 2 shows that there is no difference in the quality of the nanopattern produced, even after fifteen imprinting processes, compared to the nanopattern produced by the new AAO mold. The reason for choosing polystyrene as a material for transferring a nanopattern is that the polystyrene has an appropriate glass transition temperature (100 °C) for thermal imprinting, and it is widely used in commercial cell culture dishes owing to its transparent and nontoxic properties.
Fabricating nanostructures with high aspect ratio using thermal nanoimprinting method is difficult. Because, when process conditions, such as pressure and time increase, the mold becomes deeply embedded into the polymer substrate, making it hard to separate the mold from the substrate. Therefore, it is very challenging to produce nanopatterns having an aspect ratio of 1:3 or more using our nanoimprinting technique. However, according to TEM images we observed Figure 5C, we have found an interesting fact that the membrane of cells on nanopatterns was not in contact with the bottom of the nanopattern. In other words, all physical stimuli that can manipulate a response of cells were originated from the top parts of the nanopattern. Therefore, in this study, we did not have to consider the aspect ratio in studying responses of the cell to the nanopattern. As shown in Supplementary Figure 3, 0.1% protein coating solution had no influence on the attachment of hECFCs on the nanopillar surfaces. Furthermore, these nanoscaled stimuli from the gradient nanopattern plates decreased the cell area and perimeter of hECFCs and increased filopodial outgrowth Figure 5.
In summary, we established gradient nanopattern plates by manipulating a response of hECFCs in vitro. To know exactly what size of nanostructures the cells are greatly affected, a future work is required to design a cell-specific-sized nanopillar by adjusting the slope of the size gradient and minimum-maximum diameter of the nanopillars. We expect this gradient nanopattern to be a fascinating tool to screen the interaction between cells and nanostructures as well as provide advanced cell niches in which nanostructures can be freely tuned in size.
The authors have nothing to disclose.
This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (MEST) [NRF-2015R1D1A1A01060397] and Bio & Medical Technology Development Program of the NRF funded by the Ministry of Science, ICT & Future Planning [NRF-2017M3A9C6029563].
Perchloric acid 60% | Daejung Chemicals & Metals | 6512-4100 | |
Ethyl alcohol, absolute 99.9% | Daejung Chemicals & Metals | 4118-4100 | |
Phosphoric acid 85% | Daejung Chemicals & Metals | 6532-4400 | |
Methyl alcohol 99.5% | Daejung Chemicals & Metals | 5558-4400 | |
Chromium(VI) oxide | Daejung Chemicals & Metals | 2558-4400 | |
Sulfuric acid 95% | Daejung Chemicals & Metals | 7781-4100 | |
Hydrogen peroxide 30% | Daejung Chemicals & Metals | 4104-4400 | |
n-hexane 95% | Daejung Chemicals & Metals | 4081-4400 | |
Toluene 99.5% | Daejung Chemicals & Metals | 8541-4400 | |
(heptadecafluoro-1,1,2,2,-tetrahydrodecyl)dimethylchlorosilane | Gelest | SIH5840.4 | Moisture sensitive |
Methoxynonafluorobutane 99% | Sigma aldrich | 464309 | |
Collagen solution | Stemcell | #4902 | |
Gelatin | Sigma aldrich | G1890 | Protein coating solution |
Ficoll-Paque | GE Heathcare | 17-1440-03 | Hydrophilic polysaccharide solution |
EGM-2MV | Lonza | CC-3202 | Endothelial cell expansion medium |
Penicillin-Streptomycin | Gibco | 15140-122 | |
Phosphate buffered saline | Gibco | 10010031 | |
Fetal bovine serum | Gibco | 12483-020 | |
Paraformaldehyde | Sigma aldrich | P6148 | |
Glutaraldehyde | Sigma aldrich | G5882-100ML | |
Osmium tetroxide | Sigma aldrich | 201030-1G | |
Hexamethyldisilazane | Sigma aldrich | 440191 | |
Triton X-100 | Sigma aldrich | X100-100ML | Octylphenol ethoxylate |
Goat serum | Gibco | 26050-088 | |
anti-human vinculin primary antibody | Sigma aldrich | V9131 | |
F-actin probe | Molecular Probes | A12379 | Fluorescence-conjugated phalloidin |
Alexa Fluor 488-conjugated anti-mouse IgG antibody | Molecular Probes | A11001 | Fluorescence-conjugated secondary antibody |
4',6-diamidino-2-phenylindole | Sigma aldrich | D9542 | |
Mounting medium | DAKO | S3023 | |
Anti-human vWF primary antibody | DAKO | A0082 | |
Anti-human CD144 primary antibody | BD Biosciences | #555661 | |
Eponate 12™ Embedding Kit, with BDMA | Ted Pella | 18012 | Epoxy resin |
Uranyl Acetate, 25g | Ted Pella | 19481 | |
Lead Citrate, Trihydrate, 10g | Ted Pella | 19312 | |
Ultra pure aluminum plate | Goodfellow | 26050-088 | |
Polystyrene sheet | Goodfellow | ST313120 | |
8.0" silicon wafer | Siltron | 29-01024-03 | Single side polished, 725 µm thick |
Vacuum desiccator, 4.4 L | Kartell | KA.230 | |
Vacuum pump | Vacuumer | V3.VOP100 | |
Power supply | Unicorntech | UDP-3003 | |
Magnetic stirrer | Daihan scientific | SL.SMS03022 | |
Overhead stirrer | Daihan scientific | HT120DX | |
Circulator | Daihan scientific | WCR-P12 | |
Linear moving stage | Zaber | A-LSQ300A-E01-KT07 | |
Angle bracket, 90 degrees | Zaber | AB90M | Accessory of the linear moving stage |
PMP forcep, 145 mm | Vitlab | 67995 | Nonmetallic tweezer |
PTFE beaker, 250 mL | Cowie | CW007.25 | |
Ultrasonic cleaner | Branson | B2510MTH | |
PCB cutter | Hozan Tool Industrial | K-110 | |
Nanoimprint device | Nanonex | NX-2000 | |
Oxygen plasma generator | Femto Science | CUTE | |
Low temperature sterilizer | Lowtem | Crystal 50 | |
CO2 Incubator | Panasonic | MCO-18AC | |
Confoal laser scanning microscope | Carl Zeiss | LSM700 | |
Scanning electron microscope | JEOL | JSM6701 | |
Transmission electron microscope | Hitachi | H-7500 |