Development of Three Alternative Strategies for the Binding of Cells to Functionalized DeepTip^TM AFM Probes

Abstract

The efficient design of biohybrid materials requires controlling the interaction between the cell and the material for a wide range of possible combinations. Single cell force spectroscopy (SCFS), an atomic force microscopy (AFM) experimental procedure based on the binding of an individual cell to an AFM cantilever and the assessment of the adhesion force between the cell and a target substrate, represents one of the most promising alternatives to characterize the interaction between cell and material. However, SCFS relies on the efficient binding of the cell to the AFM in order to avoid drawbacks, such as the detachment of the cell. In this work, three different versatile and robust procedures are presented that allow for the binding of either non-adherent (CD4⁺ T-lymphocytes) or adherent (mesenchymal stem cells, MSC) cells to the AFM probe. The three crosslinking strategies comprise (1) the streptavidin/biotin system, (2) sulfhydryl group-based crosslinkers, and (3) “click” (bioorthogonal) chemistry. Additionally, three decoration schemes of the functionalized AFM probes are explored: a specific antibody, concanavalin A, and direct binding of the cell through azide-derivatized membrane proteins. Differences are observed between these alternatives and it is found that the strength of the interaction (in decreasing order) is as follows: specific antibody, concanavalin A, and binding through azide-derivatized proteins.

1. Introduction

Biohybrid materials constitute a thriving research field that arises from the combination of living cells and inert materials [1,2]. This combination, on the one hand, may serve to create a protective environment for the cells [3], while the material may take advantage of the biological constituent in order to respond to external stimuli [4]. The opportunities opened by biohybrid materials are reflected in the large number of intended applications, including, for instance, Materials Science [5] and robotics [6]. In this context, it is worth highlighting the attempts to produce novel biomaterials and tissue engineering scaffolds from this type of composite material [2,7,8].

The expected extension of the field is clearly perceived by considering the enormous amount of alternatives that arise from combining various cell lineages [9,10] and different materials [11,12]. Consequently, it is necessary to rely on experimental techniques that allow for an efficient characterization of the interaction established between cells and materials in order to have reliable guidance for the design and implementation of new biohybrid materials [13].

Among the increasing number of applications of atomic force microscopy to biological systems [14,15,16], its usage for the characterization of the interaction between cells and material surfaces allows us to exploit the excellent performance of AFM, both in terms of spatial and force resolution [17]. These experiments usually require the binding of a cell that must be adhered to the AFM cantilever. Subsequently, the attached cell is taken to contact the substrate while controlling both the force and the time of contact [18,19]. This experimental procedure leads to the so-called single cell force spectroscopy (SCFS) experiments [20].

However, the whole experimental scheme outlined above depends critically on the robustness and reliability of the cell-AFM cantilever adhesion. Several binding approaches have been implemented, including the usage of glutaraldehyde [17] and of biological molecules, such as concanavalin A, fibronectin, and poly-L-lysine, that are adsorbed on the surface of the AFM cantilever [21,22]. However, all these approaches tend to be affected by problems such as the uncontrolled detachment of the cell from the cantilever.

In this context, we present here three crosslinking strategies that allow for the efficient and reliable binding of cells to a functionalized AFM probe. Amine-functionalized DeepTip^TM AFM probes are used and decorated with each one of the crosslinking strategies, which comprise (i) the biotin–streptavidin system, (ii) thiol crosslinking chemistry, and (iii) bioorthogonal (“click”) chemistry. When required, antibodies or concanavalin A were covalently bound to the AFM cantilever through these chemistries, and the successful decoration of the probe was initially assessed through specific fluorophores. The binding of cells to the decorated cantilevers was verified with two different cell lineages: non-adherent CD4⁺ lymphocytes and adherent mesenchymal stem cells (MSCs). Although all strategies were observed to lead to the binding of the cell to the probe, it was found that the strongest adhesion corresponded to the antibody-decorated probes.

2. Materials and Methods

This section provides a detailed description of the experimental procedures followed for the following: (i) the extraction and maintenance in culture of CD4⁺ T cells and of the cell culture of mesenchymal stem cells (MSCs); (ii) the functionalization of MSCs with azide groups; (iii) the decoration of AFM probes with the biomolecules of interest for subsequent cell attachment, using three different crosslinking chemistries (biotin/streptavidin, thiol chemistry, and bioorthogonal chemistry), including the corresponding assessment of each probe decoration through fluorescence microscopy; and (iv) verification of the cell binding to the functionalized cantilevers.

2.1. T Cells Extraction and Culture

T lymphocytes were obtained from Mus musculus CD1 mice. The mice were housed with ad libitum access to food and water, under controlled temperature and humidity conditions, as well as regulated light–dark cycles. All animals used were healthy mice of both sexes, aged between 2 and 4 months. No animals were sacrificed for the isolation of this cell type, and all procedures described below were conducted in accordance with Spanish legislation on animal experimentation and welfare.

T cells were collected from the peritoneal cavity, following a standardized laboratory protocol. Briefly, 1 to 3 mL of sterile 1X phosphate-buffered saline (PBS) were injected into the peritoneal cavity, the abdomen of the animal was gently massaged, and approximately 80% of the injected volume was recovered. The recovered volume contained various types of immune cell, including T lymphocytes.

The peritoneal cavity cells suspended in PBS were centrifuged at 396× g for 10 min at 4 °C (Rotina 380R, Hettich, Kirnlengern, Germany). The supernatant was then discarded, and the cell pellet was resuspended in complete RPMI medium, consisting of RPMI 1640 medium (Roswell Park Memorial Institute 1640, Gibco) supplemented with 10% fetal bovine serum (FBS, Cytiva), 200 mM L-glutamine (Gibco), and 1% penicillin–streptomycin (P/S, Sigma-Aldrich). Subsequently, to eliminate any erythrocytes that were potentially present due to the extraction procedure, red blood cell lysis was performed using BS PharmLyse^TM solution (BD Biosciences), following the manufacturer’s instructions.

To separate the non-adherent T lymphocyte population from other adherent cells in the sample, the cell suspension was placed into wells containing a migration inhibitory factor (Kartell) for 30 min at room temperature. After this incubation, the supernatant—expected to contain only CD4⁺ and CD8⁺ T lymphocytes—was carefully collected, using a pipette. T lymphocytes were visually identified by light microscopy and counted using a Neubauer chamber prior(Sigma Aldrich, Inc. Missouri, USA) to culture.

For T cell culture, TexMACS™ medium (Miltenyi Biotec) supplemented with 10% FBS, 1% L-glutamine, 1% P/S, and 0.01 mM 2-mercaptoethanol (Sigma-Aldrich) was used, and cells were seeded at a minimum density of 14,000 cells/cm². Cultures were maintained in an incubator under controlled conditions at 37.5 °C and 5% CO₂.

2.2. MSCs Culture

Bone-marrow-derived mesenchymal cells from CD1 mice were used (hereafter MSCs). MSCs were isolated from the spinal bone marrow of mice and expanded, following established protocols [23]. For their culture, DMEM medium supplemented with 10% FBS, 1% L-glutamine, and 1% P/S was used. Cultures were maintained in an incubator under controlled conditions at 37.5 °C and 5% CO₂; expansions were made once the cells reached approximately 80% confluence. Expansions were performed according to a standard protocol under sterile conditions. Briefly, the medium was removed, and the Petri dish was washed with 1X PBS. After discarding the PBS buffer, 0.05% trypsin-EDTA (Gibco) was added and allowed to act in the incubator for 3 to 5 min. Subsequently, trypsin was inactivated by adding DMEM, and the cell suspension was collected into a centrifuge tube and centrifuged at 423× g for 5 min. Supernatant was discarded and the pellet was resuspended in complete DMEM. Cells were counted using a Neubauer chamber and seeded at an approximate density of 5000 cells/cm².

2.3. MSCs Functionalization via Bioorthogonal Chemistry

MSCs were functionalized with an azide-labeled sugar (mannose), 1,3,4,6-tetra-O-acetyl-N-azidoacetylmannosamine (Ac₄ManNAz, hereafter) (Sigma-Aldrich). For this purpose, MSCs obtained following the previous protocol were expanded at a seeding density of 10,000 cells/cm² and incubated for 24 h. Afterwards, Ac₄ManNAz was added at 40 µM and cells were incubated for an additional 48 h.

2.4. Assessment of MSCs Azide Functionalization

To assess the functionalization of the MSCs, cells were labeled with a fluorescent molecule coupled to a bioorthogonal chemistry reagent, BP fluor 488–dibenzocyclooctyne-N-hydroxysuccinimidyl ester, isomer 5 (BP fluor 488–DBCO, BroadPharm). The culture medium was removed, and the cells were washed twice with 1X PBS and fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. Two additional washes with PBS were performed for 15 min each. BP fluor 488–DBCO was then added at 25 µM for 20 min in the dark at room temperature. Samples were washed three times with PBS for 5 min each. Cells were subsequently stained with DAPI at a 1:1000 dilution for 5 min in the dark at room temperature, followed by three washes with PBS for 5 min each. Negative controls included cells not treated with Ac₄ManNAz, but incubated with BP fluor 488–DBCO, and cells not treated with either Ac₄ManNAz or the fluorescent DBCO (

Table 1. Conditions for the negative controls used for the characterization of MSCs functionalized with Ac₄ManNAz.

	Sample	Control 1	Control 2
Cells	MSCs + Ac4ManNAz	MSCs	MSCs
BP fluor 488–DBCO

). Fluorescence microscopy was used to capture at least three images in random areas of each culture well, to obtain a representative sample. All fluorescence images were acquired at 10× magnification with identical imaging settings (DAPI: exposure 300 ms, gain 2×, gamma 1.4; BP fluor 488–DBCO: exposure 446.8 s, gain 2×, gamma 1).

2.5. Functionalization of AFM Probes

All functionalizations were performed using AFM DeepTip™ SiN R67 AFM probes (chips), kindly provided by, Bioactive Surfaces, Galapagar, Madrid, Spain.

2.6. Biotin-NHS Ester/Streptavidin Functionalization

AFM probe biotinylation began by briefly immersing the chips in 2-propanol (Scharlab) to remove impurities, followed by incubation in a solution of Sulfo-NHS-LC-biotin EZ-Link (biotin-NHS, Thermo Fisher Scientific) at 5 mg/mL in 1X PBS for 30 min at room temperature. The chips were washed with PBS and then incubated in a 0.5 mg/mL streptavidin solution in PBS for 30 min at room temperature. After incubation, the chips were washed with PBS and stored at 4 °C for up to 7 days prior to use

2.7. Assessment of Biotin-NHS/Streptavidin Functionalization

Biotin-NHS functionalization was verified using streptavidin conjugated to fluorescein isothiocyanate (FITC) (excitation: 498 nm, emission: 517 nm) (streptavidin-FITC, Molecular Probes). After biotin-NHS incubation, chips were incubated in a 0.5 mg/mL streptavidin-FITC solution in PBS for 30 min at room temperature in the dark. Samples were washed, and fluorescence images were acquired at 10x magnification (exposure: 1 s, gain: 2.1×, gamma: 0.83) and analyzed using ImageJ. Negative controls were prepared by incubating samples with PBS only, instead of a biotin-NHS solution.

2.8. Decoration with Concanavalin A

Concanavalin A (ConA) is a lectin extracted from Canavalia ensiformis that is known for binding specifically to glycoprotein receptors that are present in cell membranes. It was used for MSC assays due to its ease of storage, low cost compared to antibodies, and efficient MSC binding. Chips were functionalized with biotin-NHS and streptavidin, as described above, and subsequently incubated with biotinylated ConA (Sigma-Aldrich) at 0.4 mg/mL in PBS for 30 min at room temperature. Chips were washed three times with PBS and were ready for use.

2.9. Thiol Modification and Verification

To functionalize the AFM chips via thiol chemistry, both molecule and substrate must contain either thiol (–SH) or pyridyl disulfide groups. Modifications were applied to both of them to introduce these groups for covalent attachment.

2.10. Derivation of the Biomolecule with Thiol Groups

Streptavidin, lacking native thiol groups, was chemically modified to enable surface attachment through thiol groups, as shown in Figure 1. Streptavidin from Streptomyces avidinii (Sigma-Aldrich) was dissolved at 2 mg/mL in PBS and reacted with 20 mM sulfosuccinimidyl 6-(3′-(2-pyridyldithio)propionamide)hexanoate (sulfo-LC-SPDP, Thermo Scientific) for 50 min. The resulting modified streptavidin with a pyridyldisulfide group (streptavidin-PDP from now on) was purified using size-exclusion chromatography columns (MWCO 7.000, Thermo Scientific).

2.11. Functionalization of the AFM Probes with Thiol Groups

The AFM chips were cleaned with 2-propanol and incubated with 2.5 mg/mL sulfo-LC-SPDP for 50 min at room temperature. Afterwards, they were washed five times with 1 mM ethylenediaminetetraacetic acid (EDTA, Sigma-Aldrich) in PBS 1X (PBS-EDTA) and then reduced with 3 mg/mL tris(2-carboxyethyl)phosphine (TCEP, Sigma-Aldrich) in PBS-EDTA for 15 min at room temperature, followed by three washes with PBS-EDTA. Then, 40 µg/mL of streptavidin-PDP in PBS-EDTA was incubated with the functionalized chips for at least 18 h at 4 °C. The chips were washed with PBS 1X containing 0.2% TWEEN^® 20 (Sigma-Aldrich) and 0.1% sodium dodecyl sulfate (SDS, Fisher Bioreagents) for 30 min under gentle stirring at room temperature. Samples were finally rinsed with PBS 1X. The whole process is summarized in Figure 2.

2.12. Assessment of the Thiol Streptavidin Decoration

The decoration of the chips with streptavidin was assessed by using biotinylated horseradish peroxidase (biotin-HRP, Invitrogen). The chips were incubated with a solution of biotin-HRP at 100 ng/mL in PBS-EDTA for 30 min under gentle stirring at room temperature. The chips were then washed with a solution of PBS 1X containing 0.2% TWEEN^® 20 (Sigma-Aldrich) and 0.1% sodium dodecyl sulfate (SDS, Fisher Bioreagents) for 30 min under stirring. Subsequently, chips were incubated with 1-Step^TM Turbo TMB-Elisa (TMB, Thermo Scientific) for 5 min under stirring, and the enzymatic reaction was stopped by adding 2 M sulfuric acid. TMB (3,3′,5,5′-tetramethylbenzidine) undergoes a reaction catalyzed by HRP that leads to an increase in the absorbance of the solution at a wavelength of 450 nm.

Solutions were collected and absorbance was measured at 450 nm. Controls included AFM probes without sulfo-LC-SPDP and without streptavidin-PDP, adding PBS-EDTA instead (

Table 2. Conditions for the negative controls used for verification of streptavidin-PDP immobilization on the AFM probe surface.

	Chips	Control 1	Control 2
Sulfo-LC-SPDP
Streptavidin-PDP

). A calibration curve was prepared by using serial dilutions of biotin-HRP (102.4 ng/mL to 0.2 ng/mL), adding to each solution TMB for 5 min and stopping the reaction with 2 M sulfuric acid. Subsequently, absorbance was measured at 450 nm.

2.13. Decoration with Anti-CD4 Antibodies

Streptavidin-PDP-decorated chips were blocked with 2% BSA for 1 h at room temperature, incubated with biotinylated anti-CD4 antibody at 10 µg/mL for 1 h, and rinsed twice with PBS 1X. Correct decoration was confirmed using Alexa Fluor^® 488-conjugated secondary antibody (1:800) for 1 h at room temperature in the dark. Negative controls lacked the primary antibody. As a final step, every sample was rinsed with PBS 1X. Fluorescence images were captured at 20× magnification (exposure: 1 s, gain: 2.1×, gamma: 0.83) and analyzed using ImageJ.

2.14. Functionalization Through Bioorthogonal Chemistry

Chips were washed in 2-propanol and incubated overnight with dibenzocyclooctyne-N-hydroxysuccinimidyl ester (DBCO ester, Sigma-Aldrich) at a concentration of 250 µg/mL in dimethyl sulfoxide (DMSO, PanReac Applichem) with 0.5% diisopropylethylamine (DIPEA, Thermo Scientific) at room temperature, in the dark, under gentle agitation. Chips were washed with PBS 1X containing 0.2% TWEEN^® 20 and 0.1% SDS for 15 min and subsequently rinsed with PBS 1X. Samples could be stored at 4 °C for up to one week, although they were preferably used the same day as prepared. A schematic drawing of this process is shown in Figure 3.

2.15. Procedure fo the Assessment of the Functionalization Through Bioorthogonal Chemistry

DBCO ester functionalized chips were incubated with 200 µg/mL FAM-azide, isomer 6 (excitation: 492 nm, emission: 517 nm) (6-FAM-azide, Lumiprobe) in DMSO for 2 h at room temperature in the dark with gentle agitation. Chips were washed with 10% SDS for 5 min under gentle agitation and three times with PBS 1X for 5 min. Negative controls were incubated in the DMSO/DIPEA 0.05% solution. Fluorescence imaging was performed at 20× magnification (exposure: 1 s, gain: 2.1×, gamma: 0.96). Images were analyzed using ImageJ.

2.16. AFM Probe Fluorescence Image Analysis

All fluorescence images were analyzed using ImageJ (v1.52n, NIH, Maryland, USA). Each image was analyzed individually. RGB color channels were separated to isolate the specific fluorophore, and fluorescence intensity was specifically measured in areas corresponding to the cantilevers. Data obtained in arbitrary units was adjusted to a scale of 10 for each experiment independently, with 10 being the highest value among all replicates.

2.17. Cell Binding to the AFM Probes

Measurements were performed using a Nanolife AFM (Nanotec Electrónica, Spain) and WSxM 5.0 software [24]. The Nanolife AFM is equipped with an inverted optical microscope to monitor cell positioning, relative to the cantilever on transparent sample holders. Experiments were conducted nominally at 25 °C. DeepTip™ SiN R67 chips were functionalized and decorated according to the protocols described above.

Cells were expanded according to their respective protocols, counted with a Neubauer chamber, and separated for each assay. For T lymphocytes, 30,000 cells from the day of extraction or the following day were used. For MSCs, 10,000 cells from passages 10–16 were used. For Ac₄ManNAz-functionalized MSCs, 10,000 cells from passage 16 after 48 h of incubation with Ac₄ManNAz were used. At least three independent samples were prepared with 0.45 µm-filtered culture medium.

One sample was added to the AFM sample holder, ensuring cells were sufficiently separated to bind just one cell to the AFM probe at a time. Using the inverted microscope, an intact, spherical cell was selected. The cantilever tip was carefully placed over the cell without direct pressure.

Attachment was performed via a slow, controlled approach until contact, maintaining 100 pN force during the contact time. Contact times depended on the chemistry used: CD4⁺ T cells required 5–15 s to bind to anti-CD4-decorated chips, pristine MSCs required 15–30 s to bind to ConA-decorated chips, and Ac₄ManNAz-functionalized MSCs required up to 2 min to bind to DBCO-decorated chips.

After contact, the cantilever-cell probe was retracted from the surface and allowed to rest for about 5 min to stabilize binding. Successful attachment was confirmed visually through the horizontal movement of the AFM cantilever, while verifying that the cell moved together with the cantilever.

3. Results

3.1. Assessment of Biotin-NHS Ester Functionalization

Figure 4 shows the results obtained from the fluorescence analysis of AFM probes functionalized with biotin–NHS and decorated with streptavidin–FITC (n = 3), as well as the corresponding controls (n = 2). In the control image (Figure 4A, bottom), the cantilevers are silhouetted to allow for visualization. Significant differences are observed between the control group and the streptavidin-FITC-decorated samples, as indicated by Student’s t-test.

3.2. Assessment of Streptavidin-PDP Immobilization Using Biotinylated Peroxidase

Absorbance measurements of the solutions obtained after the enzymatic reaction of biotin–HRP are shown in Figure 5. The controls indicate the presence of possible residual nonspecific interactions between sulfo-LC-SPDP molecules on the substrate and biotin–HRP (control without streptavidin, n = 2), as well as possible nonspecific interactions between streptavidin–PDP and the nonfunctionalized surface (control without sulfo-LC-SPDP, n = 2). However, as confirmed with Student’s t-test, significant differences are observed between the samples (n = 3) and the corresponding controls, indicating that this type of sample is properly functionalized and decorated.

3.3. Assessment of Anti-CD4 Decoration of the AFM Probes

Fluorescence images of anti-CD4-decorated chips are shown in Figure 6, as well as the fluorescence intensity analysis of the cantilevers. Due to the sample size (n = 2 for every group) and since the assumption of normal data distribution was not met, a Mann–Whitney test was performed instead of Student’s t-test, as it is considered to be more robust under these conditions. As shown in Figure 6B, significant differences were found between samples and controls, indicating that the chips were successfully decorated with anti-CD4 antibodies. After fluorescence intensity analysis, the brightness and contrast of the images were slightly adjusted to facilitate visualization.

3.4. Assessment of MSCs Functionalization Through Bioorthogonal Chemistry

In Figure 7A, representative images of Ac₄ManNAz-functionalized MSCs are shown. The images correspond to both the (negative) control group (cells not cultured with Ac₄ManNAz) and to the functionalized cells obtained after incubation with DBCO–BP Fluor 488. As observed, control cells display a certain residual level of fluorescence, as quantified in Figure 7B. However, the fluorescence observed in the functionalized cells is significantly higher than the residual fluorescence found in the control, as established quantitatively through the comparison of the mean value through Student’s t-test (Figure 7B). Means correspond to three biological replicates for each group.

3.5. Assessment of the Functionalization Through Bioorthogonal Chemistry

Figure 8A shows representative images of chips functionalized using the DBCO ester chemistry, and the measured intensities are quantified in Figure 8B. After performing Student’s t-test, significant differences were observed between the control group (n = 2) and the sample (n = 3), indicating that the surface of the chips were successfully functionalized using this chemistry.

3.6. Assessment of the Cell Binding to the AFM Chips

Optical microscopy images obtained after confirming cell binding to the cantilever through each corresponding crosslinking chemistry are shown in Figure 9 for each of the cell types and crosslinking chemistries. The saturation and brightness of the images have been modified and cells are silhouetted to facilitate visualization. The number of cells tested were four CD4⁺ T cells, six pristine MSCs, and three azide-functionalized MSCs. Non-silhouetted images are shown in the Supplementary Materials.

In order to further assess the efficiency of cell binding to the functionalized cantilevers, as well as to verify the stability of the resulting attachment, a procedure based on the variation in the resonance frequency after cell binding was developed and applied. After waiting 5 min to stabilize the binding of the cell to the cantilever, as indicated above, the cantilever resonance frequency was measured five times. Each resonance frequency measurement is performed by oscillating the cantilever and a typical experiment consists of five independent frequency screenings. Consequently, the cell must remain bound to the cantilever for a duration of at least five minutes, during which the cantilever oscillates.

An example of the amplitude vs. frequency plot obtained from the frequency screening is illustrated in Figure 10A. The black line corresponds to the initial frequency screening and shows a clear peak at the resonance frequency, f₀, of the cantilever. When a cell is bound to the cantilever, the added mass leads to a decrease in the resonance frequency (blue line) and the displacement of the resonance frequency is calculated as Δf = (f₀ − f₁)/f₀.

The result of applying this methodology to the various decoration strategies and cell lineages is shown in Figure 10B. Figure 10B shows the variation in the resonance frequency resulting from the binding of pristine MSCs (n = 7) to concanavalin A-decorated cantilevers, and of the binding of CD4⁺ T lymphocytes (n = 4) to anti-CD4+decorated cantilevers. Each experiment was performed using an independent cell–cantilever pair, since once the cell was attached, it was not possible to detach it from the cantilever. No data are included on the azide-functionalized MSCs binding to DBCO-decorated cantilevers because the binding was not sufficiently stable to maintain the cell attached during the measurements, resulting in cell detachment during frequency screening. This experiment was repeated three times with three different cell-cantilever pairs.

4. Discussion

The use of AFM to investigate the interaction between different cell types and materials through SCFS enables the exploration of these interactions from a perspective that exploits the potential of AFM in terms of both spatial and force resolution. The diversity of approaches available for attaching a cell to the cantilever requires a systematic procedure that not only allows for assessing the advantages and limitations of each binding strategy, but also its optimization. In this work, different strategies that allow for the binding of a cell to the AFM probe are amply explored. Since the main objective of this study is to provide a useful toolkit that can be broadly applied, the proposed crosslinking chemistries are combined with different decorating molecules and cell lineages. The three proposed crosslinking chemistries comprise the biotin/streptavidin system, a thiol-based crosslinking chemistry, and a “click” bioorthogonal crosslinking chemistry.

It may be worth indicating the characteristic features of each chemistry. To begin with, the biotin/streptavidin is a very versatile approach, as it enables the usage of any biotinylated biomolecule, such as commercially available biotinylated antibodies. In case of necessity, there exist protocols that allow biotinylating of practically any required antibody.

The thiol-based crosslinking chemistry offers the unique possibility of reusing decorated probes by reducing the disulfide bond that maintains the binding molecule that is attached to the surface. This reduction can be achieved through reducing agents, such as TCEP or dithiothreitol (DTT). Although no attempt was made within the context of this work on developing a detailed protocol for the reuse of the probes, its feasibility is demonstrated by the preparation of the SH-functionalized probes that result from the reduction in the original pyridyldisulfide group with TCEP. With respect to the thiol-based crosslinking strategy, we also show how it can be applied even if the decorating molecule lacks a free sulfhydryl group, as shown by the decoration of the SH-functionalized surface with the SPDP-derivatized streptavidin. In this latter case, however, some extra time is required for the derivation of the molecule, although, as illustrated with the derivation of streptavidin, this additional time may be in the range of one hour.

Finally, the last crosslinking chemistry, the biorthogonal (“click”) chemistry is of particular interest in the biomedical field because it involves relatively small, highly specific molecules that do not occur naturally in living organisms. This allows for its use in biological environments without causing cellular damage and with minimal risk of interfering with cellular biochemistry or metabolism, as the reagents involved are not present endogenously [25,26,27,28]. In order to attach MSCs to the AFM cantilever using this chemistry, cells were successfully functionalized with Ac₄ManNAz through metabolic glycan engineering. The fluorescence images indicate that the cells incorporated azide groups on their surface. However, a significant background signal was detected in the control group, which had not been reported in other studies [27,29,30]. This background signal might result from relatively intense nonspecific interactions between the fluorophore and the non-functionalized cells. In parallel, the cantilevers were decorated with DBCO molecules: a process that was successfully achieved, as shown from the comparison of the fluorescence intensity between controls and functionalized samples. The main advantage of the click chemistry lies in the ability to perform metabolic glycan engineering to incorporate bioorthogonal chemical groups on the cell surface without affecting other cellular functions. This implies that, in principle, the click chemistry might be the most direct technique for attaching cells to the surface of AFM cantilevers of all the three strategies presented in this work. However, as indicated below, this strategy leads to a relatively labile interaction between cell and material.

In those cases in which the cells could not be bound directly to the functionalized probe, it is necessary to use a decorating molecule. Following the rationale of creating a toolkit that is as versatile as possible, two types of biomolecules were employed: a specific antibody and concanavalin A. In addition, the assessment of the decoration of the surface with the SPDP-derivatized streptavidin was performed through the interaction of the streptavidin with a biotinylated enzyme (HRP) which validates not only the presence of functional streptavidin on the surface, but also the preservation of the enzymatic activity after the binding of the molecule to the streptavidin.

Finally, the versatility of the developed protocols was assessed by its application to two different cell lineages, both in terms of size and of its adherent character towards materials. CD4⁺ lymphocytes are non-adherent cells of a moderate size, while MSC are adherent cells of a relatively large size. It is shown that both cell lineages can be efficiently bound to the AFM probes. In this regard, optical microscopy images showed that CD4⁺ T cells, pristine MSCs, and functionalized MSCs were successfully bound to the cantilever through the three different chemistries: biotin-NHS with concanavalin A was used to bind pristine MSCs, thiol chemistry was used to decorate the cantilever with an anti-CD4 antibody in order to bind CD4⁺ T cells and, lastly, bioorthogonal chemistry was used to bind azide-functionalized MSCs to alkyne-functionalized cantilevers. It was observed, however, that despite applying the same AFM observation procedure and the same force (≈100 pN) to attach the cell to the cantilever in all cases, significant differences were observed in the contact time required for the successful attachment of the cell to the probe. Thus, whereas initially it was assumed that the bioorthogonal chemistry would provide the fastest and most effective attachment [31,32], it was found that longer contact time was required for the cell to adhere (around a minute). Even though the Ac₄ManNAz concentrations used have been shown to be optimal for covering the cell surface, avoiding cell viability issues [29,33] (which means azide density on the cell surface should be sufficient), this crosslinking chemistry also produced the least stable binding, as the cell would detach from the cantilever after approximately 10–15 min. In contrast, both decoration of the cantilevers with anti-CD4 antibodies through the thiol chemistry and with concanavalin A through the biotin/streptavidin system were found to lead to a stable binding of the cell to the AFM probe. It was established, however, that the required contact time to bind the cell to the antibody-decorated cantilever was significantly shorter (≈10 s) than that required for the binding of pristine MSCs to the concanavalin A-decorated AFM probe (≈30 s). In both cases, cells remained stably adhered to the cantilever for a period of at least two hours. It should also be stressed that the stability of the binding of pristine MSCs and CD4⁺ T lymphocytes to concanavalin A- and antibody-decorated cantilevers, respectively, allowed for the performance of frequency screening measurements. It was not possible to perform those experiments on azide-functionalized MSCs, since cells become detached from the DBCO cantilevers before completing the frequency screening.

Since the objective of this work was the development of a toolkit for the binding of cells to AFM probes that includes a broad range of crosslinking chemistries and decorating biomolecules, and that can be applied to the largest possible number of cell lineages, the interesting question of the interaction between the cells and the substrate is not pursued further here. An example of the power of this type of analysis is offered, for instance, by the comparison of the different adhesion forces of MSCs on titanium, depending on the peptides used for the decoration of the material [20]. In this case, it was possible to correlate the strength of the adhesion with the proliferation of the cells on the material, which was found to be dependent on the particular peptide used for the decoration.

5. Conclusions

In this work, a preparatory experimental study of the performance of three different crosslinking strategies, biotin/streptavidin, thiol-based crosslinking chemistry, and bioorthogonal chemistry, for attaching cells to AFM probes, is presented. Two types of cells, adherent MSCs and non-adherent CD4⁺ T lymphocytes, were used as model systems to prove the versatility of the procedures. Successful functionalization of AFM probes was demonstrated for all three chemistries through fluorescence microscopy. In turn, MSCs were also successfully functionalized with Ac₄ManNAz, as was also confirmed by fluorescence microscopy. Both MSCs and CD4⁺ T lymphocytes were successfully attached to cantilevers functionalized with the corresponding chemistries: pristine MSCs were attached to concanavalin A-decorated cantilevers; CD4⁺ T lymphocytes were bound to anti-CD4-decorated cantilevers, and Ac₄ManNAz-functionalized MSCs were attached via bioorthogonal chemistry to DBCO-decorated chips. It was found, however, that the binding of the functionalized MSCs to the DBCO-decorated cantilever required a longer contact time and led to a more labile binding compared with the other two attachment strategies. Although within the framework of biohybrid materials, it will be necessary to perform specific studies for each cell lineage–substrate pair of interest, the proposed protocols are shown to constitute a practical toolkit for this type of study. In this context, these protocols may be taken as a convenient starting point for assessing the interaction between materials and cells obtained from different species including, most conspicuously, those of human origin.