Bioscience orbiotechnology scientists artificially grow cells and use them as research materials [1–5]. Because, among in vitro cultured cells, several types of cells such as embryonic stem cells (ESCs) are hard to culture without feeder cells or a specific extracellular matrix, natural proteins are needed for the culturing and attachment [6–8]. Thus, when cells need to be cultured, some proteins are coated on a culture dish or glass template to help cell attachment and growth. Gelatin is often used as a coating material because it is cheaper than other natural proteins [9,10]. However, although the density or coating structure of gelatin used in cell culture is known to have an important effect on cell growth, research on the mixing ratio and density of gelatin has been insufficient. Because it is expected that the rate of attachment and growth of cells will vary depending on the mixing ratio and rigidity of gelatin, in this study we used atomic force microscopy (AFM) to measure the Young’s modulus [11–13] of gelatin samples with different mixing ratios.

AFM is useful for measuring the structural and mechanical properties of materials at nanoscale [14–17]. AFM has several advantages such as high-resolution imaging, ease of use, and compatibility with samples of various characteristics. AFM can perform not only sample surface imaging, but also measurement of mechanical properties such as modulus, deformation, adhesion, stiffness, and energy dissipation of materials at nano scale [18,19]. In this study, we measured the Young’s modulus, which corresponds to the stiffness of the sample and surface roughness based on the surface morphology, at nanoscale.

We measured the Young’s modulus of gelatin-coated samples using the Pinpoint^TM nanomechanical mode in AFM to confirm that each gelatin-coated glass template with different mixing ratio has a different Young’s modulus. AFM can measure nanomechanical properties including Young’s modulus by repeatedly measuring force-distance (FD) curves at each pixel in the scan area, allowing us to simultaneously obtain topographic images and Young’s modulus values for the sample surface [20,21].

Materialssuch as Gelatin (Duksan, Korea), DMEM (Welgene, Korea), FBS (Gibco), MEM NEAA (Gibco), Glutamax (Gibco), 2-Mercaptoethanol (MP Biomedicals), Penicillin-Streptomycin (Gibco), CHIR9- 9021 (MCE), PD0325901 (MCE), Mouse LIF Protein (Sigma-Aldrich), and culture plate (SPL, Korea) were purchased from the indicated company.

For ESC culturing [22–25] in feeder-free conditions, culture plates were coated with autoclaved gelatin solutions of various concentrations (0, 0.1, 0.5, 1.0, and 2.0 %). After confirming the coating methods, gelatin solution was added enough to each culture plate to submerge the surface, and the plates were incubated for 30 min on a clean bench, followed by removal of the gelatin solution and drying.

Mouse ESCs (E14 cell line, 6.25 × 10⁴ cell/ plate) were seeded on culture plates coated with specific gelatin concentrations and maintained using DMEM-based complete media supplemented with 15 % FBS, MEM NEAA, Glutamax, 55 µM 2-Mercaptoethanol, Penicillin- Streptomycin, 3 µM CHIR99021, 1 µM PD0325901, and 10³ U LIF for 2 days in a humified culture incubator (37°C with 5 % CO₂). Morphologies of ESCs were observed and analyzed with Inverted Laboratory Microscope (Leica) in bright field mode.

To verify ESC attachment, the cells were seeded on wells of plates coated with 0, 0.1, 0.5, 1.0, and 2.0 % gelatin. After 2 days, the adhesion rate was observed. As a result, it was confirmed that an ESC colonies had adhered successfully to the gelatin coating concentration of 0.1%, while ESC was found not to attach to the non-coated plate (Fig. 1).

Figure 1. Growth of mouse ESC colonies cultured on gelatin coated plates.

We used the Pinpoint^TM nanomechanical mode of the AFM system (Park Systems Co., NX10) to measure the Young’s modulus values of all samples. The Pinpoint^TM nanomechanical mode can measure quantitative nanomechanical properties such as modulus, adhesion, deformation, stiffness, and energy dissipation. All measurements were conducted in ambient air with the same experimental parameters of image size, 20 µm × 20 µm; pixels, 256 × 256; and z‐scanner speed, 25 µm/s. Appropriate cantilevers were used for accurate measurement. Considering the Young’s modulus values of the samples, we choose the cantilever model AC160TS (Olympus Co.).

The Pinpoint^TM nanomechanical mode is a measurement method that determines mechanical properties by deriving F-D curves at all pixels in the scan area. F-D curves show the force acting according to the distance between the tip and the sample surface; the slope of the F-D curve obtained through this process indicates the stiffness. Young’s modulus can be obtained using the slope of the retraction curve: the harder the sample, the steeper the slope of the F-D curve and the higher the Young’s modulus. In general, because the rigidity of gelatin is lower than that of plastic or glass, a large Young’s modulus was found for samples with low mixing ratio of gelatin. Conversely, in samples with large gelatin mixing ratio, low Young’s modulus values were found. For identical samples, it was confirmed that a constant Young’s modulus was measured for uniform part of the gelatin coating, but a modulus similar to that of a non-coated culture dish (or glass template) was measured at positions where the gelatin coating was peeled off.

3.1. Photodetector sensitivity (A-B sensitivity)

Three items need to be calibrated to correctly derive F-D curves. The first is the A-B sensitivity. When the tip is sufficiently enough to the sample, it bends due to interaction between tip and sample. As a result, the position at which the laser irradiates the end of the cantilever is reflected and the position sensitive photo diode (PSPD) changes. At this point, the positional displacement of the reflected laser corresponds to the z scanner displacement, and the A-B displacedisplacement in voltage in meters is converted to meters.

The slope of the deflection versus the z scanner disposition curve indicates the A-B sensitivity. The slope deflection measurement process must be performed using a rigid substrate such as silicon. When a rigid substrate is not used, bias due to sample deformation may occur, and accurate measurement may not be possible. Therefore, we adjusted the ratio of deflection (z-axis height change) (µm) and the deflection (A-B value) measured by PSPD in situations in which only deflections by force occur, without sample deformation. In this measurement, the sensitivity was corrected to about 33 V/µm.

3.2. Force slope correction

The next step is force slope correction. During this process, the force slope, which is the change in the A-B value according to the position of the z-axis scanner, is corrected. Because the z-axis scanner moves, the signal changes even if the cantilever does not bend. Therefore, the slope baseline of the F-D curve was corrected using force slope correction. If the force slope correction is inaccurate, an inclination may occur in the baseline of the F-D curve. If the force slope correction is not performed, the baseline of the F-D curve may have an inclination. If the baseline of the F-D curve is tilted, the Young’s modulus cannot be measured, so the baseline must be modified. In this experiment, the force slope was corrected to about 37 mV/µm.

3.3. Spring constant (k) calibration

When converting the cantilever deflection into force, the cantilever’s spring constant value must be determined. Considering the characteristics of the sample, AC160TS, with a spring constant of 22 ± 3 N/m, was used. After completing the three calibrations, the detailed parameters of the Pinpoint^TM nanomechanical mode must be adjusted. The speed, set point, and baseline were adjusted to suit the characteristics of the sample. If not set to an appropriate parameter, the Young’s modulus may not be valid.

By adjusting the details to suit each sample, reasonable measurement data can be obtained. In this experiment, we set the following parameters: speed: 25.0 µm/s; set point: 112 nN; and min Length: 40 nm.

3.4. Measurement of gelatin samples by AFM

We prepared 5 different samples (0, 0.1, 0.5, 1.0, and 2.0 % gelatincoated culture dishes) to be measured by AFM Pinpoint^TM nanomechanical mode [Fig. 2(a)]. Figure 2(b) shows the measurement of the cantilever used to measure Young’s modulus placed on the gelatin samples; cantilever has a spring constant of 22 ± 3 N/m (AC160TS, Olympus Co.). When cantilever approaches surface of gelatin samples, signals suddenly decrease via contact with sample and cantilever experiences repulsive force from the surface. The cantilever starts to retract after reaching a certain force setpoint value; next there is a decrement of repulsive force. Finally, the cantilever detaches from the surface by further cantilever retraction. With this F-D curve, one can derive the Young’s modulus value at the contact spot. By moving the x- and y-axes (scanning), one can obtain mapping images from the Young’ modulus values of the target samples [20].

Figure 2. Samples and measurements. (a) Pictures of samples used for measurement. 0, 0.1, 0.5, 1.0, and 2.0 % gelatin coated culture dishes. (b) Optical microscope image of a cantilever approached to a gelatin sample. (c) Representative F-D curve on gelatin sample (0.5 %) after calibration of A-B sensitivity, force slope correction, and spring constant (k) calibration.

3.5. Roughness of gelatin samples

We first examined the roughness of the gelatin surface, one of the factors of the cell culturing environment that determines whether cells can grow well or not; for this, we measured the surface roughness using the height information obtained from the AFM data according to the order of samples, from low mixing ratio to high mixing ratio. Figure 3 shows the surface morphology of gelatin-coated culture dishes of 0 % [Fig. 3(a)], 0.1 % [Fig. 3(b)], 0.5 % [Fig. 3(c)], 1.0 % [Fig. 3(d)], and 2.0 % [Fig. 3(e)], obtained using AFM Pinpoint^TM nanomechanical mode. All measurements were conducted in ambient air with identical experimental parameters of image size: 20 µm × 20 µm; pixels: 256 × 256; and z-axis scanner speed: 25 µm/s. We can define the roughness of the sample surface to a uniform degree according to the height measured at each pixel point. When the z-axis height data of all samples were collected and displayed in a normal distribution table, the z-axis height distributions of all samples were shown to be similar [Fig. 3(f)]. It was determined that the gelatin mixing ratio did not clearly affect the roughness of the gelatin-coated sample surfaces; thus, we need to define another factor for the cell culturing environment, such as the Young’s modulus.

Figure 3. Surface morphologies (z-axis height) of (a) 0 %, (b) 0.1 %, (c) 0.5 %, (d) 1.0 %, and (e) 2.0 % gelatin coated culture dishes for surface roughness obtained by AFM Pinpoint^TM nanomechanical mode. (f) Normal distribution table of heights (z-axis) of samples. The surface roughness is defined according to the height of the sample in the

Next, we examined Young’s modulus values obtained by AFM according to mixing ratios, in the order of low to high mixing ratio. We used the Derjaguin-Muller-Toporov model because the modulus values of the samples were greater than 1 GPa and the samples were adhesive [20]. Figure 4 shows Young’s modulus mapping images of gelatin-coated culture dishes of 0 % [Fig. 4(a)], 0.1 % [Fig. 4(b)], 0.5 % [Fig. 4(c)], 1.0 % [Fig. 4(d)], and 2 % [Fig. 4(e)], obtained by AFM Pinpoint^TM nanomechanical mode. As the density of gelatin coated on the culture dish increased, the brightness decreased, indicating that the surface properties of the samples changed due to covering with soft materials, leading to good ESC culture condition. The value of the scale bar on the left was fixed to allow a clear comparison of the Young’s modulus values; it can be seen that, the higher the modulus value, the brighter the image (0 %), and the lower the modulus value, the darker the image (2 %). Through the overall data, we found that Young’s modulus gradually decreases as mixing ratio of gelatin increases. We determined that when Young’s modulus values of measured sample surfaces were expressed in a normal distribution table and compared, changes in modulus between 0 and 0.1 % were the largest and a clear difference could be seen [Fig. 4(f)]; this indicates that the cell culturing environment (gelatin coated culture dish) is sufficient to be grow well after the coating density of 0.1 %. Therefore, we confirmed a gelatin coating density on the culture dish of over 0.1 % should be sufficient.

As can be seen in Fig. 1, the shapes of the cells in the culture dish that was not coated with gelatin (0 %) were round. This indicates that the cells are not stably attached to the culture dish. On the other hand, cells in the culture dishes coated with gelatin at 0.1 % or more are observed as extending tentacles. This morphology shows that the cells are stably attached and growing on the culture dish. This result confirms that cells cannot easily grow on a hard surface. Therefore, to study how soft a cell surface should be, the Young’s modulus must be considered. In general, cells are microscopic in size; therefore, the depth at which they penetrate the coating material when attached is very shallow, ranging from a few micrometers to a few hundreds of nanometers; the surface characteristics of sample are thus very important. Therefore, we judged that the Young’s modulus of the surface to which cells attached was important when determining the appropriate cell culture conditions. In other words, cell adhesion changed greatly between 0 % gelatin and 0.1 % and higher gelatin-coated samples; it was confirmed that cells grow well from a gelatin coating value of 0.1 % (Fig. 1). Correspondingly, when AFM was used to measure the Young’s modulus according to the gelatin concentration, the greatest difference in Young’s modulus was confirmed between 0 and 0.1 %; this confirms that the Young’s modulus decreased slightly beyond 0.1 % coating. It was confirmed that there is a correlation with the data shown in Fig. 4. From this, we determined that Young’s modulus values of coating materials used for cell culture dish growth environments affect cell attachment and growth.

Although we already knew, empirically, the conditions for optimizing the cell culture environment (good condition is above 0.1 % gelatin), we thought it would be meaningful to analyze the optimized culture environment and produce hard data. Data obtained by analyzing the cell culture environment at mechanical scale can also help determine the appropriate density range for all other samples. It can also help researchers who want to introduce new coating materials to find the material density that will optimize the cell culture. Through this experiment, it was confirmed that cell adhesion increased on surfaces with Young’s modulus values less than about 4 GPa, coated with gelatin having concentration of 0.1 % or more rather than a hard surface, which had Young’s modulus values of about 6 GPa. Overall, we determined that cells began to attach at the level of gelatin coating of 0.1 %; the cells were very well attached in the good environment produced from gelatin coating of 0.5 % and higher.

Using AFM Pinpoint^TM nanomechanical mode, we measured the surface roughness and Young’s modulus of gelatin coated culture dishes with different densities. Valid results were obtained only when the F-D curve correction process was performed correctly. Through measurements, it was confirmed that Young’s modulus of samples decreased as mixing ratio of gelatin increased; this shows that cell culturing environment correlates closely with Young’s modulus values. We found that, above 0.1 % of gelatin coating on culture dishes, Young’s modulus showed decreasing values, corresponding to better cell culturing environments. Using AFM, we measured Young’s modulus values of surfaces of culture dishes coated with gelatin of various concentrations. We have a plan to extend this experiment to show how well cells can grow when 2D materials, such as graphene or transition metal dichalcogenides (TMDC), are covered on top of the gelatin, and further analyse the correlation of these samples with their measured Young’s modulus values. However, when coating a 2D material such as graphene or TMDC on top of gelatin, it is expected that cell will not easily attach to the surface because it is too smooth; the resulting change in Young’s modulus is thus expected to be small. This study might be a good next research step. Young’s modulus measurement is thought to be a very useful method for observing microscale or nanoscale sample surfaces. Our results can be applied to gelatin coating templates with respective Young’s modulus values to study the cell growth environment. Finding a range of densities of gelatin that can maximize the rate and conditions of ESC growth will help producing organoids in good condition for 3-dimensional tissue culturing using gelatin-coated culture matrixes.

This research was supported by the National University Promotion Program at Jeonbuk National University in 2021.

The authors declare no conflict of interest.