Construction of a biocompatible MWCNTs–chitosan composite interface and its application to impedance cytosensing of osteoblastic MC3T3-E1 cells

In this work a carboxylated MWCNTs–chitosan composite sol–gel material was developed via one-step electrodeposition on a glassy carbon electrode as the cytosensing interface of a novel impedance cytosensor. SEM verified the formation of a three-dimensional hierarchical and porous microstructure favorable for the adhesion and spreading of osteoblastic MC3T3-E1 cells. By correlating impedance measurements with fluorescence microscopic characterization results, the cytosensor was demonstrated to have the ability to determine the MC3T3-E1 cell concentration ranging from 5 × 103 to 5 × 108 cell per mL with a detection limit of 1.8 × 103 cell per mL. The impedance cytosensor also enabled monitoring of the cell behavior regarding the processes of cell attachment, spreading, and proliferation in a label-free and quantitative manner. By taking advantage of this cytosensing method, investigating the effect of the C-terminal pentapeptide of osteogenic growth peptide (OGP(10–14)) on MC3T3-E1 cells was accomplished, demonstrating the potential for the application of OGP(10–14) in bone repair and regeneration. Therefore, this work afforded a convenient impedimetric strategy for osteoblastic cell counting and response monitoring that would be useful in evaluating the interactions between osteoblastic cells and specified drugs.


Introduction
Cell impedance biosensors are a type of electrochemical device 1 that employs living cells on the electrode surface as a sensing component to detect cell responses to various stimuli by measuring the impedance variation. 2 With the ability to analyze the biological events of cell adhesion, spreading, proliferation, and apoptosis and other processes in a label-free and quantitative fashion, cell impedance biosensors have the potential to be used as powerful tools in an array of elds such as drug screening, toxicology testing, and cytophysiological and pathological mechanism research. 3 The characteristics of electrode interface play a crucial role in successful applications since the interface needs to provide not only a microenvironment for accommodating the living cells but also an electrochemical connection to the electrode for electric signal transduction. 2 Therefore, developing new electrode interfacial construction strategies by using optimal materials or material combination have been the research focus of cell impedance biosensors to achieve desired performance.
To date, various materials have been reported for the interface construction of cell impedance biosensors, such as metallic and carbon-based nanoparticles, 4 synthetic 5 and bionic polymers, 6 quantum dots, 7 etc. Chitosan and its modied versions 8 and multi-walled carbon nanotubes (MWCNTs) 9 are among the most widely employed materials for cytosensing interface construction. Chitosan is a biocompatible and biodegradable natural polymer with a exible molecular backbone and numerous amino and hydroxyl groups that render chitosan soluble in acidic solution and can be exploited for further functionalization. While MWCNTs are electricity-conductive, biocompatible, relatively rigid, and under acidic conditions can be dispersed in the solution of chitosan to form a well homogeneous state. These characteristics of chitosan and MWCNTs make them suitable partners for cytosensing interface construction. 10 The carboxylated version, cMWCNTs, which is more hydrophilic and dispersible, has been also employed for this purpose. 11 In addition, silica sol-gels, which are formed by in situ hydrolysis of various silicates such as tetraethyl orthosilicate under acidic or basic conditions, has been considered a kind of glue for building stable biosensing interfaces due to their ability to form covalent three-dimensional networks with other components, 12,13 showing promising potential in constructing various biosensor interfaces. Nevertheless, the combination of the specied three materials has not been reported thus far for constructing a cytosensor interface.
On the other hand, murine osteoblastic MC3T3-E1 cells are a type of model cell line that has been utilized for investigating its interactions with a variety of scaffold materials and evaluating the role of the interfacial morphology and osteogenic growth factors. 14 Osteogenic growth peptide (OGP), along with its active form C-terminal pentapeptide OGP (10)(11)(12)(13)(14), is a class of biologically active peptide hormones with promising effects on the proliferation and differentiation activity of osteoblastic MC3T3-E1 cells. 15 Recently, using OGPbased supramolecular hydrogels to promote hBMSC osteogenesis differentiation 16 and OGP-functionalized highly porous scaffolds to stimulate the proliferation and growth of osteoblastic cells 17 were also disclosed, suggesting these OGPbased scaffolds have promising potential as biomaterials for bone regeneration. Despite these important results, there are no reports regarding cytosensing investigation of the MC3T3-E1 cell behavior and the effect of (OGP)/OGP (10)(11)(12)(13)(14) on the cell adhesion, spreading, and proliferation, and these knowledges would be helpful to comprehension of osteoblastic cellmaterial interactions and development of novel cytosensing interface materials.
Given the advances of impedance cytosensors and their interface materials, in the present work, a novel impedance cytosensor for osteoblastic MC3T3-E1 cells was constructed by a one-step electrodeposition of three sol-gel components, chitosan, carboxylated MWCNTs, and a silica precursor (3-aminopropyl)triethoxysilane (APTES), on glassy carbon electrode (GCE). This type of composite sol-gel cytosensing interface was highly porous and cytocompatible, hence suitable for MC3T3-E1 cell growth and electrochemical impedance monitoring. By correlating impedance measurements with uorescence microscopic characterizations, the impedance cytosensor was demonstrated to be very useful in counting MC3T3-E1 cells and monitoring cell growth and stimulus response to OGP(10-14) (Scheme 1).

GCE modication with cMWCNTs-chitosan composit sol-gel
Glassy carbon electrodes (GCE) with 3 mm in diameter were polished successively with 1.0, 0.3, and 0.05 mm alumina slurry, followed by rinsing and sonicating with double-distilled water and alcohol, and dried at room temperature. To prepare the cMWCNTs-chitosan composite sol-gel solution, a chitosan acetic acid solution (0.5% chitosan at pH 5) was rst made; then to this solution proper amounts of cMWCNTs and water were added to give a cMWCNTs-dispersed solution with the cMWCNTs content of 1.0 mg mL −1 and chitosan concentration of 0.05%; and nally the composite sol-gel solution was obtained by mixing 2300 mL of the chitosan acetic acid solution, 600 mL of cMWCNTs-dispersed solution, 60 mL of APTES, and 40 mL of 30% H 2 O 2 solution. Electrodeposition was performed using a CHI660D electrochemical workstation (Shanghai Chenhua, China) and a three-electrode system with a saturated calomel electrode (SCE) as the reference electrode, a platinum foil as the counter electrode, and the cleaned GCE as the working electrode. The composite sol-gel solution was used as the electrolyte, and the electrodeposition was conducted using a constant potential at −0.4 V vs. SCE and room temperature for 300 s. The modied electrodes were carefully rinsed with water, then dried for 60 min in air and stored at 4°C for use. The morphology of the modied electrode interface was characterized by SEM (JEOL-6700F, Japan).

Cell culture
The osteoblastic MC3T3-E1 cells were cultured at 37°C in DMEM medium (Gibco, Life Technologies) in a humidied atmosphere containing 5% CO 2 . The DMEM medium was complemented with 10% fetal calf serum (FBS) (by volume, Tianjin Haoyang Biological Corporation, Ltd China), 100 mg mL −1 penicillin, and 100 mg mL −1 streptomycin (Gibco, Life Technologies). Then the cells were digested with 1.5 mL of 0.25% trypsinase (Gibco, Life Technologies), separated from the medium by centrifugation at 1000 rpm for 5 min, and washed twice with a sterile PBS at pH 7.4. The cells were diluted with serum-contained medium to yield a cell suspension as stock solution with a nal concentration of 5 × 10 8 cells mL −1 . Cell suspensions with various concentrations were prepared from this stock. For the cell culture on the electrode surface, 5 mL of the cell suspension was dropped on the electrode and incubated at 37°C for a certain time. OGP (10)(11)(12)(13)(14) immobilized GCE was prepared by dropping 5 mL of the OGP(10-14) solution on the cMWNTs-chitosan composite sol-gel surface, maintained at a humidied atmosphere overnight for cell culture. The cell immobilized GCE was rinsed thoroughly with PBS three times for electrochemical impedance measurements.

Electrochemical measurements
Cyclic voltametric measurements were performed in pH 7.4 PBS on the CHI660D electrochemical workstation and the threeelectrode system as described in electrodeposition. Electrochemical impedance spectroscopy (EIS) was carried out on an Autolab PGSTAT30 workstation (The Netherlands) using a 0.01 M pH 7.4 PBS solution containing 5 mM Fe(CN) 6 3−/4− (1 : 1 in molar ratio) with 1.0 M KCl as the supporting electrolyte, The amplitude of the applied sine wave potential was set at 5.0 mV within the frequency range of 0.1 Hz-100 kHz. All measurements were carried out at 37 ± 2°C.

Inverted uorescence microscopy
For inverted uorescence microscopy characterization, glassy carbon sheets instead of GCE electrodes were used, and the surface modication and MC3T3-E1 cell culture were conducted using the same procedures mentioned above. 3 Results and discussion

Preparation and characterization of cMWCNTs-chitosan composite sol-gel interface
Organic-inorganic sol-gels are extensively used for constructing biosensors to improve stability and biocompatibility of the interface structure and to enhance the capability of accommodating biological species, 18 owing to their desired advantages such as tunable porosity, chemical inertness, and lm-forming ability. 19 In the present work, three materials, chitosan, cMWCNTs, and the silicate precursor APTES, were combined to form the sol-gel solution. To accelerate the formation of the cytosensing interface lm on the electrode surface, a proper amount of H 2 O 2 was added and electrochemically reduced at a potential of −0.4 V vs. SCE to hydroxide, which raises the local pH, leading chitosan from soluble to insoluble state 19 and to crosslinking with APTES to give a stable and rigid composite sol-gel lm. SEM images in Fig. 1a-c show plainly the difference of the morphologies of cMWCNTs, cMWCNTs-chitosan mixture, and cMWCNTs-chitosan composite sol-gel lm. The pristine cMWCNTs with diameter of 20-30 nm are worm-like discrete nanotubes (Fig. 1a), which become a coagulate state aer treatment with a 0.05% chitosan aqueous solution with acidity adjusted to pH 5.1 by adding a concentrated NaOH solution (Fig. 1b), due to the insolubility of chitosan aer deprotonation. 20 Under the electrodeposition conditions, the raised local pH at the electrode interface resulted in the formation of chitosan-wrapped cMWCNTs and thus a highly porous morphology of cMWCNTs-chitosan composite sol-gel with cMWCNTs in diameter ranging from 20 to 30 nm is presented. The porous microstructure is of vital importance for cell survival because it can serve to mimic the actual in vivo microenvironment where cells interact and growth and to facilitate nutrient and oxygen diffusion and waste removal. 21 Osteoblastic cell responses to the nano-micro multiscale cMWCNTs-chitosan composite sol-gel lm was also assessed using SEM to characterize the cell morphology when cultured aer 3 days on the lm surface. As shown in Fig. 1d, the cells spread out and cover the surface well. Two cells, one with at round shape and another with elongated shape, form connection with each other. These facts demonstrate that the cells response to and interact with the composite sol-gel lm properly, 22 reecting the as-prepared lm is cytocompatible and favorable to the attachment and growth of osteoblastic MC3T3-E1 cells.
producing detectable electrochemical signals on a suitably modied electrode for characterizing cell activity. 23 The cyclic voltammograms were acquired in pH 7.4 PBS for bare GCE, cMWCNTs-chitosan composite lm/GCE, MC3T3-E1 cells/GCE, and MC3T3-E1 cells/composite lm/GCE as shown in Fig. 2A.
The bare GCE has a minimum background current, which is raised in the high potential region of 0.6-1.3 V on the other three electrodes, attributable to the presence of chitosan and MC3T3-E1 cells that form larger double layer capacitances in this region on the respective surfaces ( Fig. 2A, curves a-d).  Compared with the direct attachment of the cells with GCE, which displays only a raised capacitance current (curve c), the cells on the electrode with the cMWCNTs-chitosan composite lm present an irreversible oxidation peak at +0.78 V vs. SCE in the voltammogram (curve d). A similar observation was reported on a gold nanoparticles-chitosan nanocomposite modi-ed GCE where an oxidation peak at nearly the same potential position was produced from the K562 leukemia cell response. The oxidation peak was ascribed to the detectable electron transfer in the intracellular conversion of guanine to 8-oxoguanine in the living cells. 24,25 In our case, the oxidation peak at +0.78 V vs. SCE might also come from the same intracellular reaction because like gold nanoparticles the cMWCNT conductivity could mediate the charge transfer within the cells to the electrode surface, making the intracellular reaction detectable. The output of the electrochemical signal suggests that the cells are maintained in a living state and the composite interface provides a compatible environment for the cells.
The signicant inuence of the cells on the electrode electrochemical process comes from the impedance characterization, from which the electron transfer resistance (R et ) of the cell adhesion on GCE was measured using [Fe(CN) 6 ] 3−/4− as the redox probe. In addition, these interfaces show negligible solution contact resistance R s , which was also observed in the graphene oxide quantum dots/cMWCTs composite modied electrodes, 11 favorable to the interface charge transfer.
The impedance response of live cells on electrode surface has been exploited for constructing sensitive and label-free cytosensors for many applications such as drug screening, toxicology testing, and cytophysiological and pathological mechanism research. 3 In the present work, the relationship between the cell concentration and the impedance R et value was also investigated.  (Fig. 2D). According to this linear relationship the cell density in a MC3T3-E1 cell culturing suspension could be easily determined, affording a simple and straightforward way for the cell number quantication.

Monitoring of MC3T3-E1 cell behaviour on cMWCNTschitosan composite lm/GCE
On the basis of the established cMWCNTs-chitosan composite lm/GCE cytobiosensor and the linear relationship between R et and log C, forthcoming investigation was focused on the MC3T3-E1 cell behavior on the composite lm/GCE surface, and understanding the cell growth process would be helpful to the application in investigating the interactions of cell factors and drugs with the cells. 26 The cells were incubated on the cMWCNTs-chitosan composite modied electrodes at 37°C and the impedance response was monitored. Fig. 3a displays the time course of the R et value variation with incubating time. The R et value increases rapidly before the incubating time of 12 h, and then the increase rate becomes slow but steady till the time reaches 72 h, aer which the R et value drops speedily. To correlate the impedance response with cell behavior, inverted uorescence microscopy was applied, and the uorescence images corresponding to the incubating times are given in Fig. 3b-f. It can be seen that, at the incubating time 2 h, the cells began to attach to the surface; at 6 h, more cells inhabited on the surface with cell occupying area slightly enlarged; at 12 h, the cells became larger and the cell number somewhat also increased; at 72 h, the cells spread substantially and accompanied with the still increase in cell number; and nally at 96 h, the cell number on the surface was notably decreased and cell contraction was observed for the attached cells.
Careful analysis of the variation trends of the R et value and the cell behavior on the composite/GCE surface, a preliminarily correlation could be drawn between the two variation trends, that is, the R et value is closed related to the number and occupying area of the attached cells. The cell number provides a main contribution to the R et value, which is signicant before the incubating time 12 h, resulting from the cell attachment plus probably the cell proliferation. The MC3T3 cell proliferation in this time scale was also observed for in vitro cultured on polyaniline doped titania nanotubes biointerface. 27 The cell spreading process contributes less to the R et value, which mainly occurs in the incubating time ranging from 12 h to 72 h, because in this stage the cell spreading is predominant.

Role of OGP(10-14) on MC3T3-E1 cells
As another potential application, the fabricated cMWCNTschitosan composite lm/GCE cytosensor was used to investigate the role of OGP (10)(11)(12)(13)(14) by impedance monitoring the behavior of MC3T3-E1 cells. OGP (10)(11)(12)(13)(14), as well as OGP, has the potential in medical applications for bone repair and regeneration through stimulating the proliferation and growth of osteoblast cells, as demonstrated by administration or incorporation into scaffolds composed of various materials such as poly(lactic-coglycolic acid), 28 poly(ester urea), 29 and chitosan-coated poly(lactic acid). 17 However, to our best knowledge, there are no reports on the role of OGP and OGP(10-14) using cytosensing methodologies. Fig. 4 shows the impedance responses of MC3T3-E1 cells on the sensor surface aer incubation in solutions containing 1 × 10 5 cell per mL of the cells with the presence (red curve) and absence (blue curve) of 10 −12 M OGP (10)(11)(12)(13)(14) in a period of 72 h. Again, the R et value increases rapidly before the incubating time of 12 h, and then the increase rate becomes slow but steady, and at any inspection time point, the presence of OGP (10)(11)(12)(13)(14) results in a larger R et value, corresponding a higher density of cell, than that without OGP (10)(11)(12)(13)(14), suggesting the promoting role of OGP (10)(11)(12)(13)(14) on the cell attachment, spreading, and proliferation on the cMWCNTschitosan composite lm/GCE cytosensing surface.

Conclusions
Cell impedance biosensors have become a powerful tool to analyze the biological events of cell adhesion, spreading, proliferation, apoptosis, and other processes in a label-free and quantitative fashion, nding potential applications in the elds of drug screening, toxicology testing, etc. However, for fabricating an osteoblastic cell cytosensor, developing a suitable electrode interface material is still challenging and remains to be explored due to the propensity of osteoblastic cells and the associated environmental requirements.
In this work an osteoblastic cell-compatible and highly porous cMWCNTs-chitosan composite sol-gel material was developed and used as the cytosensing interface material via a one-step electrodeposition on the GCE surface as a novel cytosensor. SEM veried the three-dimensional hierarchical and porous microstructure favorable to the adhesion and spreading of cells. The cytosensor was used for determining osteoblastic MC3T3-E1 cell concentration and thus a linear relationship was established in the range from 5 × 10 3 to 5 × 10 8 cell per mL with a detection limit of 1.8 × 10 3 cell per mL. The impedance cytosensor was able to reveal convincingly the cell behavior regarding the cell attachment, spreading. And proliferation. Furthermore, the effect of OGP(10-14) on the osteoblastic MC3T3-E1 cells was investigated using this impedance cytosensor, showing the potential of OGP (10)(11)(12)(13)(14) in bone repair and regeneration. Therefore, this work established a novel impedimetric cytosensing strategy for osteoblastic cell concentration determination and stimulus response monitoring that would be very useful in estimating the interactions between osteoblastic cells and interfacial materials or relevant drugs.

Conflicts of interest
There are no conicts to declare.