FluidFM: Precise fluidic positioning and delivery platform with applications in cell biology and soft matter

Article image: 
Patrick Frederix1, Paul Werten1, Dalia Yablon2
1. Nanosurf AG; Liestal, Switzerland
2. SurfaceChar LLC; Sharon MA USA
Patrick Frederix is Dutch citizen and studied applied physics at the Eindhoven university of technology. His PhD he did under supervision of Hans Gerritsen at the University of Utrecht where he worked with fluorescence microscopy. In 2001 he moved to Switzerland and learned AFM in the group of Andreas Engel at the Biozentrum of the university of Basel. Here he worked on the combination of AFM and scanning electrochemical microscopy (SECM) and automation of force spectroscopy analysis. In 2010 he joined Nanosurf in Liestal as application scientist and is now product manager for the FluidFM products at Nanosurf.
 The platform of fluid force microscopy (FluidFM) offers researchers unique capabilities in precise and well-controlled fluid delivery in small quantities down to femtoliters to the surface.  It is based on a microfluidic controller coupled with an innovative AFM probe where a microchannel is hollowed out so that the probe essentially functions as a nanopipette.  Applications of FluidFM are described including enhanced manipulation of cells, improved force measurements on biological materials including bacteria, and fluidic writing on hydrogels.
Corresponding author
Dr. Patrick Frederix
Head of Applications & Service
Phone +41 61 927 47 86
Fax +41 61 927 47 00
Atomic force microscopy is primarily known for its imaging capabilities. Invented 30 years ago, imaging topography together with mechanical or electrical properties at impressive nanoscale resolution is among its most popular uses. An additional key advantage of AFM over other microscopy methods is its ability to work in any environment – vacuum, ambient, or liquid.
This flexibility enabled AFM to grow from a tool for traditional surface science into a more versatile and powerful method that can operate in various environments, a flexibility that is critical for fields such as biology. The ability to characterize biological systems such as bacteria, DNA, and cells in situ under physiological conditions makes AFM a unique method to study such systems. 
In addition to imaging, single point measurements with AFM that “poke” or “touch” a sample have also become popular in the field of force spectroscopy. This measures specific mechanical properties of a material such as stiffness or adhesion. Additionally, the AFM functions as a high resolution mechanical probe that can touch, move, or squeeze localized features on a surface. This ability can be taken advantage of to conduct lithography and manipulation measurements. 
Here we review the technology of fluidic force microscopy (FluidFM®), a platform developed in 20091 where a specialized microchanneled AFM cantilever is combined with a microfluidic controller to perform a variety of novel fluidic measurements that are very useful for cell biology and soft material research. The innovation in this AFM-based technology is coupling a novel type of probe that has force control with a highly accurate fluid delivery system. A schematic of this platform is show in Figure 1a. The specialized probes function as nanopipettes delivering tiny, femtoliter volumes of solution and are compatible with any solvents. 
Figure 1. (A) Schematic of fluidic force microscopy showing the microfluidics control system that manages the fluid level in the reservoir which flows to the microchanneled cantilever. (B) SEM images of the apex of the probe. 
The microfluidic controller provides a well-controlled liquid delivery system. The final component of this platform is the AFM technology based on an optical feedback system with a laser, which provides exact positioning both laterally and vertically for the probe in addition to sensitive force feedback on the probe for well-controlled interaction with the substrate. 
This platform offers significant improvement in manipulation and liquid delivery capabilities with an AFM probe. First, by having the ability to overpressurize and underpressurize the cantilever, the cantilever can more gently and controllably pick up and drop objects such as cells or particles, resulting in increased manipulation and dexterity on the surface.
Another significant improvement is the ability to controllably and precisely deliver biomolecules such as dyes, enzymes, and viruses in physiological conditions through the microchanneled cantilever, which truly functions as a nanopipette. All biomarkers are possible as long as they fit through the micron-high channel in the lever and the sub-micrometer large opening.
With the pressurizing mechanism offered by FluidFM, cell immobilization onto the probe is now possible through a fast and reversible process for the first time. Finally, these nanopipettes enable insertion and extraction measurements into target cells. The tip can actually be inserted into the cell and inject or extract biomolecules of interest through overpressurizing or underpressurizing. We review FluidFM’s recent, novel contributions in measurements of cell adhesion, single cell manipulation, single cell extraction as well as soft matter lithography. 
Materials and Methods
The classical AFM probe is composed of a cantilever with a tip hanging off the end of the cantilever. Typically such levers are manufactured from either silicon or silicon nitride. FluidFM involves a novel design of a silicon nitride probe where a channel approximately 30 µm wide and 1 µm high is etched through the cantilever all the way through the apex of the tip. SEM images of various probes that serve different applications are shown in Figure 1b.
Tipless probes shown on the left are used for mammalian cell adhesion measurements. An example of a probe with an opening at the end of the pyramidal tip is shown in the middle image. These kinds of probes are primarily for experiments in bacterial adhesion, spotting, and nano-lithography. Finally, pyramidal tips with a hole on the side as shown in the right image are primarily used for injection and extraction experiments. The opening in the injection-type probe on the right has been FIB-milled, though batch microfabricated probes have been produced as well.
This specialized cantilever is connected to a fluidic pressure control system creating a continuous flow from the control system through the apex of the cantilever. The opening at the apex of the probe vary from 300 nm to 8 µm, but can also be made smaller with FIB processing. The volumes handled with these levers are tiny — femtoliters of solution.
These probes are coupled with the force feedback of conventional AFM probes so that the approach and interaction of the tip with substrate can be controlled with great sensitivity of piconewtons of force. The stiffness of the probes can vary in a range of 0.3–4 N/m, depending on the application. The Flex-FPM system (Nanosurf, Switzerland) was the first such platform commercially available and was used for many of the available protocols developed to date, from cantilever preparation2,3 to experimental workflows as detailed below.
The principle measurement in most FluidFM experiments is single point force spectroscopy as mentioned above. In this measurement, the force exerted by the cantilever on the surface is tracked as a function of z piezo movement (which can be subsequently converted to tip-sample separation) as the cantilever approaches and then retracts from the surface.
A sample force curve is shown in Figure 2 showing the different segments of interaction: (A) cantilever approach, (B) snap-into contact, (C) repulsive portion where tip and sample are in contact, (D) the repulsive portion on the withdrawal of the cantilever from the surface, (E) the pull-out segment where the tip gets “stuck” in the adhesive dip before it emerges from the adhesion at the interface, and (F) the cantilever returns to its unperturbed state. Alternatively, the probe can be moved over the surface in an imaging or lithographic way, by moving either the AFM scanner or the stage for even larger modification areas.4
Figure 2. Sample force curve measuring cantilever deflection vs. z piezo movement at a single point on the surface.  
New immobilization protocols for improved cell adhesion measurements
FluidFM offers simplified single cell immobilization protocols and new capabilities in cell adhesion measurements. Typically, adhering cells to conventional AFM probes is both irreversible and time-consuming in terms of probe prep and calibration. As a result, multiple biological measurements with a single cantilever are not possible, resulting in poor statistics and low throughput. Furthermore, adhering cells onto cantilevers involves chemically specific coatings, electrostatic or hydrophobic interactions, or using some type of glue.
Unfortunately, these methods are typically not strong enough to overcome cell-substrate interactions and so the cell often detaches from the cantilever during the retraction portion of the measurement, which is used to measure adhesion. The platform of FluidFM introduces reversible cell immobilization by applying underpressure through the microchanneled cantilever as it “picks up” a cell with forces into the µN-range for the largest openings. 
Adhesion measurements were first shown to work for yeast and mammalian cells on differently coated or textured surfaces5,6. Cells were reversibly immobilized onto the FluidFM probe with pressures down to 80 kPa below ambient pressure. During the retraction portion of the force curve, the underpressure was maintained until the cells were fully detached from the substrate. 
With an overpressure of 100 kPa, cells can then be detached from the cantilever, providing the cantilever was appropriately coated with an anti-fouling agent.2 The cantilever can then either be used immediately to pick up a new cell, or it can be cleaned within minutes with enzymatic6 or oxidizing solutions7 before re-using it. Such a cleaning procedure can be carried out without removing the cantilever holder from the AFM. In another, more recent example of adhesion measurements. intercellular adhesion was demonstrated for mouse fibroblasts and human endothelial umbilical artery cells (HUAECS).8
Measurements were conducted both with individual cells and cells surrounded by other cells in a monolayer. The mouse fibroblast cells were found to show low adhesion force to glass, both for isolated cells and for cells in a monolayer. Isolated HUAECS revealed an adhesion force of 805 nN, while a significant adhesion increase to 1170 nN was observed for HUAECS in a monolayer, ascribed to intercellular interactions.
The strong, reversible cell immobilization offered by FluidFM was key to measuring such strong intercellular adhesion forces, and would not have been possible with conventional AFM force spectroscopy measurements.
The FluidFM platform was also used to explore the adhesion force of bacteria9,10. Polydopamine is a common adhesive used for cell immobilization onto conventional cantilevers. Unfortunately, bacteria typically fall off the conventional AFM probe during the retraction portion of the curve because the bacteria-AFM probe attachment is not sufficiently strong. A FluidFM study of the adhesion of E. coli bacteria to polydopamine-coated surfaces demonstrated that the FluidFM immobilization mechanism was stronger than chemical bacteria–substrate interactions, with bacteria remaining attached instead of falling off the probe during the measurement. Additionally, cell survival was confirmed after the bacteria was “dropped” from the cantilever after the measurement.
Using this method, adhesion forces of 4–8 nN were measured between E. coli cells and polydopamine-coated surfaces. Adhesion forces of 3–14 nN were also measured between glass and another bacteria, the smaller and spherical S. pyogenes. This bacteria naturally occurs in cell chains. Upon retraction, up to 4 bacterial cells were sequentially detached from the glass, as can be seen in the sawtooth like behavior recorded in the retraction force curve in Figure 3a, corresponding to a total detachment force of 35 nN. Detachment of E. coli from a poly-L-lysine surface [PLL] (due to electrostatic interactions, PLL is also commonly used for cell immobilization) also reveals structure in the retraction curve resolved with a force resolution in the low pN-range, as seen in Figure 3b and 3c, typical for tethers between cell and substrate during a pulling experiment11 For the E. coli–PLL measurement, this pattern is attributable to pili-mediated adhesion. FluidFM also provides the abilities to do force measurements after very long (greater than 2h) contact time between the cell and substrate, compared to tens of seconds with conventional force spectroscopy. This enables phenomena such as contact time-dependent bond strengthening to be demonstrated for the E. coli system explored here. 
Figure 3. (A) Sample retraction curve of FluidFM adhesion measurement of S. pyogenes on glass. (B) sample retraction curve of E. coli on PLL where different colored curves correspond to different contact times between probe and cell. (C) Zoom-in on retraction portion of curve showing force plateaus and jumps. Reprinted from[9]
The immobilization protocol offered by FluidFM enables additional adhesion measurements that take advantage of the cantilever functionalization in different ways. For example, cells were adhered to the cantilever to probe their adhesive strength to surfaces functionalized with non-convalently immobilized peptide ligands12. Additionally, two species of protozoa were attached to the microchanneled cantilevers to measure their deformability as a function of thermal treatment through force spectroscopy13. Demonstrating further flexibility, the FluidFM microchanneled cantilevers have also been functionalized with colloidal particles for various adhesion and stiffness studies14, 15,16.
Controlled nanopipettes for biological extraction and controlled delivery
The microchanneled probes in FluidFM can also be used as nanopipettes for injection into and sample extraction from the cytoplasm or even the cell nucleus. Typically, this process is accomplished with microcapillaries, however their aperture sizes are usually bigger and they can also induce membrane damage as the force cannot be controlled.
Using the cantilevered silicon nitride probes with apertures of a few hundred nm, FluidFM provides a well-controlled and forgiving manner to insert probes into mammalian cells with minimal cell perturbation to quantitatively extract intracellular molecules without harming the cell. The sharp tip requires a force below the µN-range to enter the cell. Hence a relatively soft cantilever can be used, preventing excessive force on the cell and thus reducing the risk of damaging the cell.
During injection, a lesion in the cell membrane of up to a few µm is created. It was shown empirically that the leakage via the tip-cell surface contact area is minimal and that the lesion closes immediately upon controlled retraction with the AFM17
The lack of leakage was shown by injection of fluorescently labelled dextran into the cell nucleus, which cannot penetrate an intact nuclear envelope. Fluorescent dye did not diffuse into the cytoplasm either during or after injection and retraction. The membrane recovery likely relates to the pyramidally shaped tip, where the lesion gradually changes during entering and removal, allowing the membrane to reorganize.
Similar to the injection into the cell described above, metabolite extraction with sub-picoliter volumes from single live cells has also been performed, both from the cytoplasm as well as from the cell nucleus18, 19.
A range of 0.8 to 2.7 picoliters of cytoplasm was successfully extracted from HeLa cells in a manner that did not harm cell viability. The targeted cells produced green fluorescent protein (GFP) in their cytoplasm so that fluorescence microscopy could be used to track the extracted material in the cantilever and in the cell.
Not only was the cytoplasm successfully extracted, it could then also be controllably handled further by specially coated cantilevers. These levers released the tiny amounts of cytoplasmic material into a microarray of spots for further analysis with mass spectrometry20. Similarly, it has been shown that these extracts can be used for TEM analysis, enzymatic assays or single cell transcriptional readout19.
In another demonstration of its nanomanipulation capability, FluidFM technology was used to write miniaturized DNA spots with a diameter of less than 2 µm for DNA sequencing. The microchanneled cantilever was filled with DNA solution, which could then be spotted at well-controlled positions with diameters of 1.4–1.9 µm. Conventional force spectroscopy measurements were then performed between a DNA-coated probe and the previously deposited DNA spots to detect DNA biomarkers present at very low copy numbers21
A final research example that takes advantage of the positioning accuracy and sensitive force control of fluidic force microscopy is chemical stimulation of neurons18.
In this study, the neurotransmitter glutamate was introduced very precisely onto the cell membranes of neurons for local stimulation. The sensitive force feedback on the cantilever allowed accurate positioning both laterally and in the critical z dimension, and then a gentle approach to the cell surface without hurting its viability. An optical microscopy image of the FFM probe approaching the target neuron (artificially colored in green) is shown in Figure 4.
Figure 4. FluidFM probe approaching target neuron. Reprinted from[18]
Once the cantilever was positioned above the cell surface, the stimulation solution inside the microchannel cantilever was then applied to the target cell with constant flow rates. This chemical stimulation platform, which operates under physiological conditions and allows control of position and dose, provides significant advantages over the more traditional and less physiological electrical stimulation methods. 
Soft Material manufacturing
A final demonstration of the versatility and power of the FluidFM platform is the use of microchanneled cantilevers to conduct lithography on soft materials. In this case, “chemical writing” was achieved on soft hydrogel films where the variables of pressure and time were explored22.
This is a subtractive technique where the hydrogel was selectively removed (dissolved) through application of an alkaline solution by FluidFM for manufacturing of desired structures and patterns on the surface. As the cantilever makes contact with the surface, an overpressure is applied for a given time interval in order for the alkaline solution to dissolve the hydrogel and “write” the feature of interest. By varying the pulse duration or amount of overpressure applied, different size structures can be written. The smallest spot size that could be written was approximately 2.5 µm, similar to the 1.8-µm aperture diameter of the probes used in this study.
A variety of structures such as stripes or lines can be written with the FluidFM platform. These structures require controlled lateral movement of the cantilever during the writing process, which is easily achieved with the AFM at the heart of the platform. By varying the velocity during lateral motion, different line thicknesses are possible, as shown in the AFM images in Figure 5.
Figure 5. Stripes written into hydrogels through dissolution by alkaline solution, delivered by FluidFM cantilevers in subtractive lithography. Reprinted from[22]
The ultimate resolution of the written features is on the micron-length scale, while the time to write these features is very fast, on the timescale of seconds. A final advantage of using AFM as central part of the FluidFM platform is that the accurate and sensitive force feedback enables very gentle interactions between the probe and the surface thereby preventing mechanical destruction of the hydrogel present in other techniques used for hydrogel structuring.
At the cost of some resolution, structures that extend the scan range of the AFM can be written on a surface by moving a motorized stage under the cantilever, but still maintaining low interaction forces using the AFM force feedback. An example of this was demonstrated with neuron cell patterning on a 530-µm length scale as shown in Figure 6. In this measurement, the surface was first coated with PLL-PEG which prevents cell growth. FluidFM then locally replaced the PLL-PEG with PLL (a cell-adhesive) in a smiley pattern, with co-deposition of fluorescent dye to enable better visualization of this pattern in Figure 6a.
Figure 6. Large-scale surface patterning (a) and cell growth (b). Reprinted from[4]
The neuronal cells were then “seeded” onto the surface and grown. Ultimately, the green fluorescently labelled cells grew only on the smiley part of the surface as shown in Figure 6b since the rest of the surface was unsuitable for cellular growth due to the PLL-PEG coating. This shows the ability to both pattern and steer cell growth on large length scales4.
FluidFM provides an innovative platform for precise and gentle fluid delivery on the nanoscale. This technology couples innovations in AFM probes, together with a microfluidic control system and AFM technology, for accurate positioning and gentle interaction with the surface. The specialized probes have microchannels drilled through them so that they can function as nanopipettes delivering femtoliters of solution able to write micron-sized features on a surface. Applications of FluidFM are primarily for biological sciences and soft matter research. Recent examples are highlighted including measurements of bacterial cell adhesion through a new cell immobilization protocol, cell manipulation and extraction, chemical stimulation of neurons, and fluidic writing of features in soft hydrogel material. In all of these examples, the soft materials such as cells and hydrogels were left unperturbed and viable thanks to the control and high sensitivity of the platform.
1. Meister, A., et al., FluidFM: Combining Atomic Force Microscopy and Nanofluidics in a Universal Liquid Delivery System for Single Cell Applications and Beyond. Nano Lett, 2009. 9(6): p. 2501-2507.
2. Guillaume-Gentil, O., T. Zambelli, and J. Vorholt, Isolation of single mammalian cells from adherent cultures by fluidic force microscopy. Lab Chip, 2013. 14: p. 402-414.
3. Roder, P. and C. Hille, A Multifunctional Frontloading Approach for Repeated Recycling of a Pressure-Controlled AFM Micropipette. PLOS ONE, 2015: p. https://doi.org/10.1371/journal.pone.0144157.
4. Vincent Martinez, a.C.F., a  Serge Weydert,a  Mathias J. Aebersold,a  Harald Dermutz,a  Orane Guillaume-Gentil,b  Tomaso Zambelli,a  János Vörösa  and  László Demkó, Controlled single-cell deposition and patterning by highly flexible hollow cantilevers. Lab Chip, 2016. 16: p. 1663-1674.
5. Eva Potthoff, O.G.-G., Dario Ossola, Jérôme Polesel-Maris, Salomé LeibundGut-Landmann, Tomaso Zambelli , Julia A. Vorholt, Rapid and Serial Quantification of Adhesion Forces of Yeast and Mammalian Cells. PLOS ONE, 2012: p. https://doi.org/10.1371/journal.pone.0052712.
6. Potthoff, E., et al., Toward a Rational Design of Surface Textures Promoting Endothelialization. Nano Letters, 2014. 14: p. 1069.
7. Jaatinen, L., Quantifying the effect of electric current on cell adhesion studied by single-cell force spectroscopy. Biointerphases, 2016. 11: p. 011004.
8. Sancho, A., et al., A new strategy to measure intercellular adhesion forces in mature cell-cell contacts" Scientific Reports. Scientific Reports, 2017. 7: p. 46152.
9. Potthoff, E., et al., Bacterial adhesion force quantification by fluidic force microscopy. Nanoscale, 2015. 7: p. 4070-4079.
10. Kathrin S. Sprechera, I.H., Jutta Nespera†, Eva Potthoffb*, Mohamed-Ali Mahic, Matteo Sangermania, Volkhard Kaeverd, Torsten Schwedec, Julia Vorholtb, Urs Jenala, Cohesive Properties of the Caulobacter crescentus Holdfast Adhesin Are Regulated by a Novel c-di-GMP Effector Protein. mBio, 2017. 8e: p.:e00294-17.
11. Helenius, J., et al., Single cell force spectroscopy. Journal of Cell Science, 2008. 121: p. 1785-1791.
12. Sankaran, S., et al., Cell Adhesion on Dynamic Supramolecular Surfaces Probed by Fluid Force Microscopy-Based Single-Cell Force Spectroscopy. ACS Nano, 2017. 11(4): p. 3867-3874.
13. McGrath, J.S., et al., Deformability Assessment of Waterborne Protozoa Using a Microfluidic-Enabled Force Microscopy Probe. PLOS ONE, 2016: p. https://doi.org/10.1371/journal.pone.0150438.
14. Dorig, P. and e. al., Exchangeable Colloidal AFM Probes for the Quantification of Irreversible and Long-Term Interactions. Biophysical Journal, 2013. 105(2): p. 463-472.
15. Simon, B.R., et al., Density gradients at hydrogel interfaces for enhanced cell penetration. Biomater Sci, 2015. 3: p. 586-591.
16. Helfricht, N., et al., Colloidal Properties of Recombinant Spider Silk Protein Particles. J. Phys. Chem. C, 2016. 120(32): p. 18015-18027.
17. Orane Guillaume-Gentil, et al., Force-Controlled Fluidic Injection into Single Cell Nuclei. Small, 2012. 9(11): p. 1904-1907.
18. Abersold, M., et al., Local Chemical Stimulation of Neurons with the Fluidic Force Microscope (FluidFM). Chemphyschem, 2017: p. doi: 10.1002/cphc.201700780. .
19. Guillaume-Gentil, O., et al., Tunable Single-Cell Extraction for Molecular Analyses. Cell, 2016. 166(2): p. 506-516.
20. Guillaume-Gentil, O., et al., Single-Cell Mass Spectrometry of Metabolites Extracted from Live Cells by Fluidic Force Microscopy. ACS analytical Chemistry, 2017. 89(9): p. 5017-5023.
21. Lee, Y., et al., Quantification of Fewer than Ten Copies of a DNA Biomarker without Amplification or Labeling. JACS, 2016. 138(22): p. 7075-7081.
22. Nicolas Helfricht, et al., Writing with Fluid: Structuring Hydrogels with Micrometer Precision by AFM in Combination with Nanofluidics. Small, 2017. 13(31): p. 1613.
Website developed by S8080 Digital Media