Collagen IV preparation
5 mg of type IV human placenta collagen (Sigma-Aldrich, C7521) was reconstituted in 5 mL of sterile 1 X phosphate buffered saline (PBS) and let mix on a roller shaker for 3 hours. The solution was then split into five aliquots of 1 mL and stored at −20 °C.
Copolymerization of collagen IV on a polyacrylamide gel
Flat polyacrylamide covered with collagen (FPC) and grooved collagen-patterned polyacrylamide (GPC) were compared to demonstrate that the technique showed in a previous work19 can be applied to mimic in the vitro cardiac extracellular environment and promote NRVM alignment. In this work, we describe only the fabrication of FPC since the fabrication of GPC was explained in19. The spacing value of GPC (width of the ridges and grooves and depth of the grooves) was chosen as the optimum pattern for NRVMs based on the literature (i.e., 10 µm width and 1 µm deep grooves)20.
Briefly, this method relies on the copolymerization of collagen IV into the gel during the direct contact of the acrylamide precursor mix with a protein-activated Parylene C films. Glass coverslips (13 mm in diameter) were washed in acetone, isopropanol, and deionised water and then dried using nitrogen gun. Parylene C films (6–7 µm thick) were then deposited on the coverslips (Fig. 1a) by chemical vapour deposition using a commercially available parylene coater (PDS2010). The silane 3-trimethoxysilylpropyl methacrylate (A174) was used to improve the adhesion of the Parylene C film on the glass. Samples were treated with O2 plasma for 1 min and 40 s using an inductively coupled plasma reactor (ICP, OPT 100 ICP 380, Oxford Instruments Plasma Technology) (Fig. 1b). Plasma was generated at a pressure of 1.33 Pa, flow 100 sccm, 1000 W source power, and 20 W bias generator power. The time of exposure to plasma was identical to the one used for preparing the patterned Parylene C masks with hydrophobic/hydrophilic regions19. 10 µl of protein solution was then spread on the activated Parylene C films and masks were let dry under a fume hood for 30 min (Fig. 1c).
5 mL polyacrylamide hydrogel precursor with a total polymer content of 8.48% (w/v) and crosslinker concentration of 5.66% (w/w) was prepared. The solution was degassed for 1 h to remove all the dissolved gas that could limit the free radical polymerisation. To initiate gelation, 3 μL of 10% w/v ammonium persulfate (APS, Sigma-Aldrich, A3678) was added to 300 μL of gel precursor solution followed by 0.3 μL of N, N, N′, N′-Tetramethylethylenediamine accelerator (TEMED, Sigma-Aldrich, T9281). A 60 μL of gel precursor mix was gently pipetted on activated glass coverslips5 and sandwiched with protein-functionalized Parylene C masks (Fig. 1e). Gels were then left to polymerise at room temperature for one h (Fig. 1f). After polymerisation, masks were then removed and the gels were stored in PBS at 4 °C (Fig. 1g). Constructs were directly transferred in sterile tissue culture petri dish covered with cell culturing medium (Fig. 1h,i).
Contact angle measurements
Parylene C (6–7 µm) was placed as described previously on glass coverslips of 13 mm of diameter. PDMS was instead spin coated at 6000 rpm (~ 8 µm) on glass coverslip of 13 mm of diameter and cured in an oven at 75 °C for 1 hour. Sessile deionised water 10 µl droplets were gradually engaged on the surface of Parylene C and PDMS to measure the static contact angle. A Drop Shape Analysis System (DSA 30 Kruss Co., Germany) was employed for the calculation of the contact angle. A polynomial function was fitted to the two-3-phase sections of the profile in the region of the baseline. One measurement per three samples per each category was performed, and the average values were extrapolated. Samples were blown with nitrogen gun before each measurement. Samples were stored at room temperature (22 °C) and low humidity level (35%).
The topography of flat polyacrylamide was visualised using environmental variable-pressure (VP) scanning electron microscopy (SEM) (Zeiss EVO 50XVP). Almost 100 µm thick PAm was fixed on glass coverslips as described above and it was hydrated overnight in deionised water. Samples were speckled gently with Kimwipe and loaded atop SEM stubs using carbon tape. VPSE detector at 20 kV acceleration voltage was used. Measurements were taken in a reasonable time to preserve the gradual loss of water of the gel over time. The topography of Parylene C mask and PAm was evaluated using a stylus profiler (KLA-Tencor) and by using an optical microscope (Zeiss Axio Lab) at 50x magnification. Three measurements on three samples were taken for the statistical analysis.
Mechanical characterisation at the nano-scale
Flat polyacrylamide samples were prepared using the polymerisation between two activated coverslips as described elsewhere5. They were stored either in PBS or water at 4 °C before the nanoindentation tests. Our experimental setup for measuring the mechanical properties is a hybrid system made of a commercial head (SMENA, NT-MDT) with home-built electronics. Hydrogels were tested both in water and in PBS.
A spherical tip of 5 µm of diameter (sQUBE, CP-CONT-BSG-A) was used to minimise the indentation. Cantilever stiffness calibration was done following the work of Sader et al.21, and it was estimated to be 0.2 N/m with an error of 10%. Samples were placed in a petri dish of 35 mm of diameter filled with liquid. The approaching/retraction rate used was 400 nm/s.
where FN is the maximum force applied, E is Young’s modulus, R is the radius of the spherical tip, the indentation depth is δ = (z − d) where z is the piezo position and d is the cantilever deflection. ν is the Poisson’s ratio equal to 0.4823.
NRVM isolation and culture
The study with NRVMs was exempted from formal ethics review. NRVMs were isolated from Sprague-Dawley rats two days after birth in compliance with Schedule 1 methods (Scientific Procedures) Act 1986. The isolation technique has been described previously11,24. Briefly, hexempteart ventricles were minced and enzymatically digested. After the digestion, the tissue was centrifuged to produce a cell suspension which was filtered to remove any non-digested tissue. Cells were then suspended again in 25 ml NRVM medium (67% Dulbecco’s modified Eagle medium (DMEM), 16% Medium 199, 10% Horse serum (Gibco), 4% foetal bovine serum (FBS) (Gibco), 2% HEPES (4-(2-hydroxyethyl)-1- piperazineethanesulfonic acid) buffer and 1% penicillin-streptomycin). The suspension was then plated in T75 flasks for 1 hour at 37 °C to remove fibroblasts. NRVMs were collected and counted using a hemocytometer. NRVMs were plated on FPC and GPC substrates, and they were successively incubated (37 °C, 5% CO2). The NRVM medium was replaced every 2–3 days. All experiments were performed 3–4 days post seeding.
Monoclonal anti-α-actinin antibody (Sigma, A7811), deoxyribonucleic acid labeled with fluorescent 4′,6-diamidino-2-phenylindole (DAPI) (Invitrogen) staining were used to quantify cell alignment. Fluorescent images were obtained using an inverted Zeiss LSM-780 confocal microscope with x40 oil-immersion lens (Carl Zeiss).
Quantification of cellular alignment, morphology, and orientation
Each cell was approximately fitted with an ellipse whose minor (Dmin) and the major axis (Dmax) was defined as the length and width of individual myocyte. Dmax of each cell was selected relative to the horizontal axis of the image field. Elongation was described through the elongation factor (Dmax/Dmin − 1) that describes the extent of the equimomental ellipse lengthened or stretched out25. The values of the elongation factor were determined from each scaffold on almost n = 15 cells randomly selected from each sample.
where P is the perimeter and A is the cell area, quantified using ImageJ. Circularity index equal to 1 indicates a non-elongated cell.
The cell orientation angle (θ) was measured manually using an ImageJ software package (http://rsb.info.nih.gov/ij) as the lack of deviation between the major elliptical axis and the reference axis of a single cell. For alignment of cells on grooved substrates, the direction of the pattern (preferential mean axis) was considered as a reference axis, whereas for alignment of cells on the flat substrates an arbitrary axis of alignment was chosen. The minimum alignment value close to 10° denotes parallel alignment with the direction of the pattern, and the maximum value close to 90° represents perpendicular alignment. Cells with an orientation angle of less than 10° were counted as aligned cells27. Orientation angles are reported in polar plots obtained using ready-made routines in MATLAB.
Nuclei orientation was quantified as described previously via conversion of the DAPI channel images into binary images using ImageJ11 to recognise any ellipse (nuclei) present with the size set to 10–50 µm. Alignment was defined as the lack of deviation in the axis of an individual nucleus from the mean axis of all individual nuclei11. The ellipses were only counted if they were between the set range to exclude non-nuclei or composite structures from the analysis. At least 90 cells were randomly selected per each category of scaffolds to determine cell shape and orientation. Each analysis was performed on fluorescence images (three images per each scaffold) at the same magnification.
Ca2+ transient measurements
Fluo-4-acetoxymethyl ester (fluo-4 AM) (Invitrogen, ThermoFisher Scientific) was used to visualise the intracellular calcium transients. NRVMs were loaded with 4 μL fluo-4 AM (prepared as 50 µg Fluo-4 AM dissolved in 50 µL DMSO) in 1 mL DMEM and placed in the incubator at 37 °C for 20 min11,20. The DMEM was refreshed, and the cells were returned to the incubator for a further 20 minutes for de-esterification of the dye. The constructs were mounted on the stage of an upright Nikon Eclipse FN1 microscope or an inverted Nikon Eclipse TE2000 microscope in a glass bottom dish (MatTek Corporation) and observed through a 40x water immersion or 40x oil objective respectively. Ca2 transients in cardiomyocytes were studied by field stimulating the cells at 1 Hz to induce rhythmic depolarisation, and line scans were recording. A custom made MATLAB code was used to calculate the normalized amplitude as f/f0, time to peak (Tp), times to 50% (T50) and 90% declines (T90) in the transients28.
All the described experiments were repeated at least three times. With regards to topography, nonparametric analysis was carried out among samples to statistically evaluate the significant difference between ridges and grooves of the mask and the gel, respectively. A Mann-Whitney test was also performed to compare the two groups, and the significance was fixed at 1%.
For statistical analysis of cell elongation, morphology and orientation, three different images were acquired per each sample to analyse at least 80 cells for statistical analysis.
Mann-Whitney test was employed for evaluating cell elongation, circularity, and area whereas nuclei alignment and calcium transient were performed using an unpaired t-test. Cell area and circularity are presented as box plots, and each box is composed of whiskers indicating the lower extreme and upper extreme of the data, while the top part and the bottom part of the box are the first and third quartiles. The line inside the box is the median. In the plots, *indicates p < 0.05 and **indicates p < 0.01. The statistical analysis and graphic presentation were performed using OriginPro v.8 SR2 software and Prism 7 software (GraphPad Software Inc.). All data are expressed as a mean ± standard error of the mean.
The data that support the findings of this study are available from the University of Southampton institutional repository at: https://doi.org/10.5258/SOTON/D0605.