In the last years, the in vitro cell culture fields have undergone significant developments, especially in their application to a wide range of research areas, such as screening of molecules to be used in the treatments of pathologies, toxicology and disease studies [1
]. A number of concerns has been raised regarding the quality of cell cultures, with particular focus on the surfaces used to propagate cells. While plastic flasks and dishes have been available since the 1960s, to perform in vitro cell culture, nowadays, most of these vessels are in polystyrene, a long carbon chain polymer with benzene rings attached to carbons. This material can be easily bent to form any kind of container or support, which has excellent optical clarity and simple sterilization processes. The main disadvantage is the hydrophobicity of polystyrene, which does not allow cell attaching [2
]. For this reason, these vessels are modified to have a more hydrophilic surface by either corona discharge under atmospheric conditions or gas-plasma under vacuum.
Treated-polystyrene still has limitations, especially in serum-free culture conditions and to maintain differentiated cell functions in primary cultures. In order to bypass these problems, different polystyrene coatings have been proposed, involving both synthetic polymers, i.e., poly-d
-lysine, which confers a positive charge to surface and biological materials, such as extracellular matrix components (ECM) like collagens, fibronectin, vitronectin or laminin [3
]. The poly-d
-lysine is more suitable for cell culture but also time-consuming and expensive [5
]. The biological coating with ECM components from specific tissues could provide a suitable microenvironment to cultivate cells, avoiding their dedifferentiation. Unfortunately, the most common biological coating strategies employ purified matrix proteins, which produce a microenvironment for the cell growth that is different from the in vivo conditions, and this can lead to modifications of the cellular behavior. Other strategies include the use of tissue-derived ECM, but this implies that the cells are grown in 3D in a scaffold and not in monolayer (2D) [6
Another weakness of cell-culture models is the use of cell lines, which have been in vitro immortalized or obtained by cancers. These cells are only slightly representative of tissue cells, while primary cells allow greater predictability of the in vivo behavior in the presence of putative drugs or in the study of diseases progression. Unfortunately, the primary cells are much more sensitive to their environment and could respond differently if cultured on particular surfaces.
Actin is the most representative protein of the cytoskeleton of eukaryotic cells [7
]. The cytoskeleton, by integrating the activity of cytoplasmic proteins and organelles, coordinates the physical and biochemical connection of cells to the external environment [8
]. Thus, the study of actin filaments is a good method to analyze the cell behavior and the interaction between the cells and the growth environment.
Moreover, to analyze the phenotype of cells during in vitro culture, several growth factors can be measured. The Osteocalcin (OC) production is typical of osteoblasts (hOBs) [9
], Fibroblast Growth Factor-2 (FGF-2) is typical of fibroblasts (FBs) [10
] and the Bone Morphogenetic Protein-2 (BMP-2) is involved in chondrocyte (HPCs) metabolic activity [11
Fibroblasts, chondrocytes and osteoblasts are cells involved in production of extracellular matrix components in their in vivo tissues, skin, cartilage and bone, respectively. Among the ECM components, they produce the collagens, mainly Collagen type I (Coll I) in skin and bone and Collagen type II (Coll II) in cartilage [12
]. Prolonged 2D cell culture can lead to the downregulation of the production of these ECM components, and the decrease of collagen production indicates that cells are losing their differentiated state; thus, a good practice to verify the suitability of the cell culture conditions is to measure the synthesis of collagens.
Several instruments are available to characterize specific physiologically related properties of biologically related microscopic samples [15
]. Among these, Atomic Force Microscopy (AFM) is nowadays a tool commonly employed to characterize at high resolution a wide range of samples at nanometer level, with negligible perturbations, minimal sample preparation and providing a truly three-dimensional quantitative mapping at the nanoscale of the studied material [16
]. In particular, the use of this quantitative high-resolution technique has demonstrated to be very appropriate to monitor the properties of the cell wall and cytoskeleton in different cell types and various conditions [18
] and is the ideal label-free tool to monitor the actin cytoskeleton in cells [21
]. The high resolution of AFM imaging allows highlighting of the fine details, i.e., roughness of the cell membrane [23
] or stiffer structures present near the cellular membrane, such as large actin filaments, which are typical of good cellular spreading on the substrate and indicative of a good cell–substrate interaction and of a stronger attachment [18
Recently, in our laboratory, nanostructured titanium surfaces have been analyzed, with the aim to find a surface with the best osseointegration features [24
]. Titanium is used for most types of implants; moreover, modified titanium surfaces are more suitable for osseointegration, due to the increased wettability or to their microstructural features, such as the surface micro- and nano-roughness, which is very beneficial to cell growth and differentiation [26
]. The coating of titanium with hard and highly biocompatible titanium carbide layer (TiC), by ion plating plasma-assisted deposition (IPPA), has been shown with a wide range of characterization techniques, ranging from XPS to AFM and SEM, to be resistant, reliable and with a well-defined chemical structure [27
]. One more positive aspect of IPPA-deposited TiC is that it is inexpensive and can easily be introduced in the substrate preparation industrial procedures.
Taking into account these findings, we studied the adhesion, proliferation and morphology of cells on nanostructured TiC surfaces, comparing them to both cell-culture-treated polystyrene dishes and poly-d-lysinated glass slides (poly-d-Lys). For a more reliable investigation, we chose to use human primary cells, isolated from patients undergoing surgical treatment. Three different types of cells were studied, fibroblasts (FBs), osteoblasts (hOBs) and chondrocytes (HPCs), in order to verify whether nanostructured TiC could stimulate cells adhesion and thus be considered a good cell culture surface.
In the last decades, the in vitro cell culture methods have undergone a strong development, due to their employment in many research areas, such as drug development, toxicology and disease studies [31
]. However, several concerns have been raised, regarding in particular the environment in which cells are cultivated [1
]. Usually, in the field of drug discovery, two-dimensional (2D) monolayer cell culture conditions are used and they have proven to be an effective and convenient method to perform analyses. Nevertheless, 2D cell cultures show disadvantages regarding the loss of tissue-specific architecture, particularly in those tissues where the interactions between cells and matrix is very desirable, such as skin [32
], cartilage [33
] and bone [34
]. Nowadays, 2D cell cultures are performed by using treated-polystyrene or coated-glass, but a certain number of limitations remain. Cell cultures on a treated-polystyrene surface can lead to loss of differentiated status and/or cell functions, those on coated surfaces, as with poly-d
-lysine, are time-consuming and expensive, and the biological coating with ECM components could provide a microenvironment for the cell growth that is different from the in vivo conditions, leading to modifications of the cellular behavior; [3
]. In the last years, we studied nanostructured TiC material, and considering all its features [24
], we decided to evaluate the 2D cell culture on this surface in comparison with cell cultures on treated-polystyrene and poly-d
-Lysine-coated glass. The titanium carbide layer was produced by IPPA deposition [27
] directly on glass slides, obtaining surfaces with 25% light transmittance ability. This feature allows to perform several kinds of experiments on cells, as immunofluorescence, while retaining the good characteristics of nanostructured titanium. Moreover, the IPPA deposition method is not expensive and can be easily introduced in the industrial procedures for glass or plastic cell substrates.
Actin is the most representative protein of the cytoskeleton of eukaryotic cells [7
] and it is crucial for the maintenance of cell structure, polarity and mobility [35
]. Most mammalian cells grow in adhesion and need, for their survival, to be attached to a surface. To analyze the effects of the supports on cell growth, we used human primary cells, which are much more sensitive to the underlying environment compared to immortalized cells.
The analysis of the actin organization is a method to evaluate the adhesion of cells to a substrate. Very interestingly, the cells cultivated on TiC showed an actin structure more similar to the organization they have in the tissues, with widespread cells bearing abundance of stress fibers (Figure 1
and Figure 4
). The FBs were arranged in 3D structure, showing filaments disposed on different planes (Figure 1
). HPCs formed a particular structure with the nucleus on one side and the cytoplasm on the other side (Figure 1
and Figure 3
), resembling the same structure previously observed in chondrocytes cultured in 3D systems [36
]. Finally, the hOBs showed on TiC an arrangement which is much more resembling to that commonly found in bone: an increased production of collagens and other ECM components and subsequently an improved differentiation in osteocytes, that have several filaments used to stay in contact with other cells and with the bone microenvironment. Both of these features were more present in the hOBs grown on TiC, compared to the cells cultivated on poly-d
-Lysine and polystyrene (Figure 2
and Figure 4
). This is confirmed by the AFM images, where the cells grown on TiC substrates show well-defined actin filaments which are not evident on the membrane of cells grown on polystyrene and poly-d
-Lys. Indeed, the presence of these stress fibers and the overall height of the cells over the substrate indicate that the cells have a better attachment and suggest a higher affinity with TiC (Figure 4
). This is also suggested by the cell form and spreading over the substrates and by the average height of the cells, which is much smaller on TiC compared to the other substrates (Table 2
). This confirms the results obtained on hOBs grown on the very same substrate [24
] and substantiates that are found with the FBs and HPCs cells.
FBs, belonging to a dermis cell type with mesenchymal origin, play a critical role in epithelial–mesenchymal interactions, wound healing and synthesis and secretion of extracellular matrix components, including collagens [37
]. In culture medium, FBs secrete several factors, such as basic fibroblast growth factor (FGF-2). This factor is a heparin-binding protein, which interacts with cell-surface-associated heparan sulfate proteoglycans, and it is involved in many biological processes, such as cell growth and wound healing [38
]. Its production is indicative of the ability of cells to adhere to surface and to proliferate; nevertheless, FGFs are secreted in a very low amount from cell cultures [39
]. We found that FBs cultivated on TiC produced a higher amount of FGF-2 compared to polystyrene and poly-d
-Lys at 24 h, whereas after 72 h, only on poly-d
-Lys, a residual production was observed (Figure 2
). It is probable that the cells on TiC had reached, after 72 h, a good level of maturation and adhesion to the growth surface; thus, they did not need to further secrete FGF-2. Moreover, cells on TiC reached confluence, as shown in Figure 3
. In order to allow cells to produce a detectable number of collagens, cells were cultivated for seven days before analysis, and, at this time point, the different confluence of cells on TiC compared to those on polystyrene and poly-d
-lys was much more evident. Finally, considering that cells were producing more collagen type I compared to the other two substrates (Table 1
), we can conclude that they preserved the differentiated status.
HPCs are the only cell type present in cartilage, and their role is to produce and renew the extracellular matrix. BMP-2 is a growth factor used in cell culture, to stimulate mesenchymal cells to differentiate in chondrocytes [35
]. HPCs cultured on poly-d
-Lys and TiC produced a higher amount of BMP-2 compared to polystyrene; considering that the chondrocytes we used are from adult tissue, the production of BMP-2 by these cells is very promising. Very interesting is also the ability of HPCs cultured on TiC to produce collagen type II (Figure 3
), which is almost undetectable on the other two surfaces. Usually, these cells when cultured in 2D dedifferentiate and lose their ability to produce Coll II. TiC, on the other hand, appears to allow the chondrocytes to be maintained in a differentiated state.
Osteoblasts originate by mesenchymal cells and are primarily responsible for the synthesis and deposition of the mineralized, collagen-rich matrix present in the bone tissue [40
]. Another very abundant factor present in the bone is osteocalcin, which is produced exclusively by osteoblasts. Osteocalcin is used as a serum marker of bone formation, and is also an indicator of the proliferative and differentiated state of osteoblasts in cell culture. So far, it has been believed to act on bone to regulate the mineralization, but recent studies suggested a broader role for osteocalcin, extending its role to the regulation of the whole body metabolism [41
]. Osteoblasts cultivated on TiC produced a statistically significant higher amount of OC compared to polystyrene and poly-d
-Lys, and moreover, a higher amount of Coll I, a collagen very abundant in bone matrix (Figure 2
and Figure 3
4. Materials and Methods
Human primary cell isolation. Human primary cells were isolated from patients that underwent surgical treatment, and full ethical consent was obtained from all donors and the Research Ethics Committee, Sapienza University of Roma, approved the study.
Human primary diploid dermal fibroblasts (FBs) were obtained from young adult male patients who underwent circumcision surgery for phimosis. FBs were isolated as previously described [37
]; briefly, after scraping away excess fat and subcutaneous tissue, the foreskins were floated in 0.25% trypsin solution (Sigma Aldrich, Co. Saint Louis, MO, USA), at room temperature, for 1 h, under gentle agitation. Trypsin was neutralized by soybean trypsin inhibitor, and the cells were cultivated in High Glucose Dulbecco’s modified Eagle’s medium (DMEM) (Sigma Aldrich), supplemented with penicillin/streptomycin, L-glutamine, Na-pyruvate, non-essential amino acids (Sigma Aldrich) and 10% Fetal Bovine Serum (FBS) (Sigma Aldrich).
Human Primary Chondrocytes (HPCs) were isolated from femoral and tibial condyles and from femoral heads, obtained from patients who underwent a total knee and hip arthroplasty. HPCs were isolated, as previously described, from articular cartilages aseptically dissected from patients, selecting areas of macroscopically normal cartilage [36
]. The isolated chondrocytes were grown to 80% confluence in DMEM (HyClone, Logan, UT, USA) supplemented with L-glutamine, penicillin/streptomycin (Sigma Aldrich) and 10% Fetal Bovine Serum (FBS).
Human primary osteoblasts (hOBs) were isolated from bone fragments, as previously described [29
]. In brief, the bone fragments were washed in sterile phosphate-buffer saline (PBS), minced and treated with 1 mg/mL collagenase type IV and 0.25% trypsin for 1 h, at 37 °C, with gentle agitation. Then, hOBs were grown to 80% confluence in McCoy’s medium (Sigma Aldrich) supplemented with L-glutamine, penicillin/streptomycin and 15% FBS.
To avoid the bias of senescence modification in cell metabolism, mycoplasma-free FB clones between the 5th and 20th in vitro passage were used, whereas HPCs and hOBs were used at first passage (p1).
All types of cells were seeded on test substrates, polystyrene, poly-d-Lys and TiC, at a density of 8 × 103/cm2, and cultured for the requested times, at 37 °C, in 5% CO2.
Substrate preparation. Cells were cultured on commercially available cell-culture-treated polystyrene dishes, on home-prepared poly-d-lysinated glass and on nanostructured TiC.
Glass slides were poly-d-lysinated by adding 50 μg/mL of poly-d-lysine in sterile water, left 10 min at room temperature and then rinsed with PBS.
Nanostructured TiC film was prepared as previously described [24
]; in brief, glass slides were coated with thin film of TiC (10.5 nm) by Ion Plating Plasma Assisted deposition method. The layers allowed the transmission of 25% of the light, which was enough to perform fluorescence microscope analyses. The TiC slides were previously characterized by XPS, to determine the composition [30
Immunofluorescence. Cells plated at a density of 8 × 103/cm2 were analyzed by immunofluorescence, to visualize the actin filaments. Cells were cultured for 48 h and then washed in PBS, fixed in 4% paraformaldehyde in PBS for 15 min, at 4 °C, and permeabilized with 0.5% Triton-X 100 in PBS, for 10 min, at room temperature. After blocking with 3% bovine serum albumin (BSA) in PBS for 30 min, at room temperature, cells were incubated with Phalloidin Alexa Fluor 488 (Immunological Sciences, Rome, Italy) 1:40, for 20 min, at room temperature. Cells were ultimately washed in PBS and incubated with DAPI (Invitrogen, ThermoFisher Scientific, Waltham, MA, USA), to visualize the nuclei. To analyze the collagens, cells, after blocking, were incubated at 1 h, at room temperature, with mouse monoclonal anti-Collagen I antibody (Santa Cruz Biotechnology, Inc, Dallas, TE, USA) or with mouse monoclonal anti-Collagen II antibody (Abcam, Cambridge, UK), washed with PBS and then incubated for 1 h, at room temperature, with Alexa Fluor 488 or Alexa Fluor 568 goat anti-mouse (Immunological Sciences), respectively. Slides were washed and then incubated with DAPI (Invitrogen, ThermoFisher Scientific), to visualize the nuclei. The images were captured by optical microscope Leica DM IL LED, using AF6000 modular Microscope, and analyzed with Leica DM microscope (Leica Microsystem, Milan, Italy).
. The free software ImageJ (https://imagej.nih.gov/ij/
) was used to perform the densitometric analysis of collagen production. For each cell culture condition, the integrated density values of fluorescence were considered. The sum of all pixel intensities in the region of interest is listed in Table 1
ELISA. Concentration of Bone Morphogenetic Protein (BMP-2), Osteocalcin (OC) and Fibroblast Growth Factor (FGF-2) in culture medium was measured by BMP-2, OC and FGF-2 ELISA kit Fine Test (Fine Biotech Co., Ltd., Wuhan, China), respectively, according to the manufacturers’ instructions.
Atomic Force Microscope
. The experiments were conducted, using a commercial FlexAFM (Nanosurf, Liestal, Switzerland) installed over an inverted optical microscope Olympus IX 70 (Olympus, Rungis, France) equipped with a high-resolution camera (F-View-II—Soft Imaging System GmbH, Münster, Germany). We used silicon nitride BrukerDNP-10 probes (Bruker probes, Manning Park Billerica, MA, USA), choosing the cantilever with nominal elastic constant of 0.06 N/m and a 4-sided pyramidal tip shape with apical radius of 10 nm. Prior to each day of measurements, the cantilevers were calibrated, using a built-in thermal calibration protocol [41
The cells on the different substrates were fixed, using 0.5% glutaraldehyde in PBS, for 3 h, at 4 °C, and then washed in PBS, kept in ultra-pure water, and finally air-dried prior to imaging. The AFM images were collected in contact mode, with an applied force of 10 nN. All images were acquired with a minimum of 512 points per line and with an acquisition speed of 2 lines per second for all the presented 30 × 30 µm images. The images are presented without any filter; only a background subtraction and plane alignment were performed. A minimum of 10 cells per sample were analyzed.