Self-assembling Fmoc dipeptide hydrogel for in situ 3D cell culturing
© Liebmann et al. 2007
Received: 31 August 2007
Accepted: 10 December 2007
Published: 10 December 2007
Skip to main content
© Liebmann et al. 2007
Received: 31 August 2007
Accepted: 10 December 2007
Published: 10 December 2007
Conventional cell culture studies have been performed on 2D surfaces, resulting in flat, extended cell growth. More relevant studies are desired to better mimic 3D in vivo tissue growth. Such realistic environments should be the aim of any cell growth study, requiring new methods for culturing cells in vitro. Cell biology is also tending toward miniaturization for increased efficiency and specificity. This paper discusses the application of a self-assembling peptide-derived hydrogel for use as a 3D cell culture scaffold at the microscale.
Phenylalanine derivative hydrogel formation was seen to occur in multiple dispersion media. Cells were immobilized in situ within microchambers designed for cell analysis. Use of the highly biocompatible hydrogel components and simplistic procedures significantly reduced the cytotoxic effects seen with alternate 3D culture materials and microstructure loading methods. Cells were easily immobilized, sustained and removed from microchambers. Differences in growth morphology were seen in the cultured cells, owing to the 3-dimentional character of the gel structure. Degradation improved the removal of hydrogel from the microstructures, permitting reuse of the analysis platforms.
Self-assembling diphenylalanine derivative hydrogel provided a method to dramatically reduce the typical difficulties of microculture formation. Effective generation of patterned 3D cultures will lead to improved cell study results by better modeling in vivo growth environments and increasing efficiency and specificity of cell studies. Use of simplified growth scaffolds such as peptide-derived hydrogel should be seen as highly advantageous and will likely become more commonplace in cell culture methodology.
Cell culture techniques are an integral part of cell analysis. Culture methods are often not a primary concern in growth studies, but rather a means to experimental results. As more understanding of its influences is revealed, the significance of cell growth methodology is receiving considerably more attention. For example, it is increasingly common to see a shift in growth studies from 2D surface cultures to 3D suspension cultures [1–3]. This clear trend is a product of growth methodology improvements arising leading to more realistic and relevant analysis results. In order to best mimic natural cell growth, in vitro cultures attempt to closely model the extracellular support structure naturally occurring in tissues. In vivo tissue growth involves cell production of extracellular matrix material, composed primarily of various proteins. Artificially produced extracellular matrices commonly used consist of mixtures of proteins such as collagen, fibrin, elastin, fibronectins, and laminins. Such components are understood to play a role in cell anchoring and separation, intercellular communication, and even nutrient feed and waste displacement. Recent studies reveal the possibility for more simplified extracellular matrix materials.
Along with the trend of 3D growth, modern cell laboratories strive to improve culture analysis platforms through size reduction and design specificity. Miniaturization dramatically increases the efficiency of studies by allowing higher throughput studies and reducing unnecessary and excessive material use. It also allows the experimental designer to develop patterns with significantly increased complexity, making concurrent parallel studies possible. This added complexity often requires formation of well defined culture volumes. Micropatterned cell analysis platforms enable the use of microfluidics for cell treatment and analysis, providing new methods for cell study with increased possibilities and improved productivity. However, this concept has proven rather problematic with previously used artificial extracellular matrices. The use of microstructures introduces significant barriers for creating stable cell suspensions confined within small compartments. Some studies have employed rapid gel formation of a continuous laminar flow of cells in solution through microchannels . Others involve pumping of a temperature sensitive sol-gel, followed by induced gel formation via temperature change . The many complexities associated with gelling laminar flows and precision volume pumping can be eliminated by incorporating dipeptide derivative hydrogels. The combined use of 3D growth and culture patterning incorporates the advantages of both methods for higher efficacy and increased productivity.
Earlier studies have incorporated simplistic 9-fluorenylmethyl carbamate (Fmoc) peptide hydrogel, first introduced by Xu et al. , as a growth substrate, adding ease of gel formation and experimental flexibility. The earliest growth study of Fmoc peptide hydrogel presented both surface (2D) and 3D growth , followed almost immediately by a second publication on a similar dipeptide-derived hydrogel used in 2D culturing . The simplicity of the latter Fmoc hydrogel featured by Mahler et al. is a product of two dominant features. First, the hydrogel is a superabsorbent material, primarily consisting of water; stable gel formation was consistently seen with as low as 0.5 wt% peptide derivative. Additionally, the seemingly complex structure of this fibrous matrix is entirely self-assembling. Introduction of appropriate Fmoc peptide concentrations to a suitable environment induces rapid self-assembly of hollow nanotubes. This low concentration Fmoc dipeptide solution is able to support a gel conformation because of its stable network of interweaving nanotubes, forming a rigid scaffold within the aqueous dispersion medium.
Here we present a new approach and application of self-assembling hydrogels for 3D cell cultures in various structures, including but not limited to microscale structures. The novelty is in the method of formation of the cell culture, enabling simple and rapid self-assembly of a gel matrix in a wide range of cell analysis structures. Spontaneous gel formation was used to immobilize liquid cell suspensions in situ by simply adding a concentrated Fmoc peptide solution to the cell dispersion. Rapid formation of the fibrous network led to hydrogel formation that could adapt to the size, shape and even complexity of the patterned container. This method eliminates the problematic loading of gels or highly viscous precursors into the designed container. With increasing need for specificity of culture patterns, whether large or microscale, a flexible method for cell culture formation will prove useful, with possibilities to adapt to a plethora of structural designs.
An additional feature of the hydrogel makes it increasingly favorable for use within complex structures. Despite the stability of the assembled hydrogel samples, multiple stimuli were seen to initiate reversing of the sample and breakdown of the hydrogel structure. Previous studies with peptide nanotubes and peptide derivative hydrogels have shown response to various types of treatment, whether changing stability or degrading entirely [8–10]. The Fmoc diphenylalanine gel demonstrated sensitivity to mechanical forces, environmental changes, and enzyme digestion. Use of one or more of these treatments enabled degradation of the gel and facilitated removal from the culture system. This can be highly favorable for systems with confined or inaccessible culture compartments. Such is often the case in microscale gel cultures, where microfluidic treatment requires compartmentalization of the immobilized cells.
Hydrogel formation results
Gelling Agent Concentration (mg/ml)
Final Peptide Concentration (wt%)
Hydrogel Assembly Result
1. PBS 10×
2. PBS 1×
3. PBS 1×
4. PBS 1×
5. PBS 1×
In the case of microscale gel formation, loading a microfluidic device with a highly viscous gel can be rather complex and difficult to consistently reproduce. Some traditional gels used for cell immobilization are highly temperature sensitive and require thermal control to prevent early gel formation. The peptide-derived hydrogel presented here requires no temperature regulation and can be designed to form entirely within the structure designed for microscopy cell analysis (detail provided in methods section).
In addition to simplified handling, in situ formation allows for more flexible experimental design. Since gel formation is initiated within the assay structure, the precursor is an aqueous solution that can be dispensed into a wide variety of patterned platforms. In this way, the hydrogel can be confined within an assay chamber designed with high experimental specificity. This allows for more complex design possibilities without the concern of difficult gel loading. The useful feature of adapting to various confining structures was demonstrated by removing a rigid gel sample formed within a conical mold.
As seen in naturally occurring human tissues, cells often grow in a 3D network. Given this spatial orientation, cells interact with both their surroundings and each other in a way fitting to their environment. Cellular function is likely to be significantly different when provided different stresses and orientation. The morphology difference demonstrated here is a closer to realistic tissue conformations. More realistic cell function, as affected by realistic cell morphology, should give rise to more relevant experimental results. For better understanding of the cell immobilization and morphology changes, the cell-hydrogel interaction should be further examined. As cell attachment is considered significant to metabolic activity and general cell function, examination of cell-hydrogel interaction would improve understanding of the cell's level of tolerance for the hydrogel environment.
Introduction of Fmoc peptide hydrogels to cell culture methodology has shown potential to both simplify experimental procedures and increase protocol flexibility. In situ hydrogel formation eliminates the complexities associated with gel handling. With gelling, the samples become more difficult to handle and apply to the actual observation/analysis platform. The peptide-derivative hydrogel simplified both handling and loading of the gel to the microstructures. With easier gel formation, it is plausible to design and construct more complex culture systems. This opens the door to new possibilities and new approaches to cell biology.
One intriguing possibility that this hydrogel enables is the study of cells at an interface. With the ability to easily control the cell type, cell density and composition of each given hydrogel layer, interactions can be studied at a well defined interface. This culture method allows one to study cellular interaction in a controlled manner that can be adapted to a variety of scales and designs.
Use of the hydrogel as a cell culture medium showed many additional benefits. Living cells suspended within the dispersion medium were easily immobilized with addition of the gelling agent. Immobilization of the cells allows for easy monitoring over extended period of time. Microscopic studies demonstrated that the immobilized cells were not restricted from migration or possible proliferation, while maintaining high viability.
Another significant benefit of the hydrogel as a culture method is the improved relevance. The goal of mimicking in vivo cell growth required a more realistic spatial growth arrangement than conventional 2D surface growth. Immobilization of suspended cells allows for a 3D dispersion of cells throughout the gel. This arrangement more closely models the in vivo growth of cells seen in tissues. Microcopy also revealed that cells suspended in a hydrogel sample tend to adopt a 3D structure, rather than the elongated conformation seen in surface cultures. The support provided by a 3D gel results in more spherical cell growth. This may have significant implications in cell studies as more relevant and realistic results are desired.
With a trend toward miniaturization of cell studies, microstructures become more common culture platforms. As micropatterning can carry a significant expense, the ability to reuse the microstructures can be economically preferable. Traditional gels used in 3D cell culture can be extremely difficult to remove from microstructures due to the increased surface interactions. Hydrogel reversibility by means of applied force, environmental change, or enzymatic digestion facilitates breakdown of the rigid gel structure. Removal of the hydrogel culture can then allow for repeated experiments within a single device. Hydrogel reversing was observed, but could be improved with further studies. Increased specificity of enzyme activity could dramatically increase the rate and extent of Fmoc peptide digestion. With improved reversal, total removal could be achieved with non-intrusive methods, possibly facilitating downstream recovery of cells for additional testing.
The self-assembling hydrogel samples were made by first creating a gelling agent. Lyophilized Fmoc-Phe-Phe dipeptide (Bachem) was weighed and dissolved in dimethyl sulfoxide (DMSO). Deviating from the preliminary methods provided by Mahler et al., DMSO was selected to replace 1,1,1,3,3,3-hexafluoro-2-propanol because it is less detrimental to cell viability. The gelling agent was used in concentrations of 25 and 100 mg/ml. A small stock solution was created to run experiments for each day so as to avoid any pre-aggregation of dipeptide derivatives. The gelling agent was applied to the desired dispersion medium by pipette addition to a final concentration of either 5 or 10 mg/ml; the higher concentration corresponded to increased gel stability. The gel was created in situ within the desired analysis structure. All structures were fabricated via silicon etching/masking procedures by the Microsystems Technology Group in the School of Electrical Engineering, Royal Institute of Technology, Sweden. Transparent glass cover slides were bonded to the etched surface to create channels and chambers for microscopic observation.
Layer formation was performed by repetition of the gel forming procedure. The initial layer was first generated in a generic mold and allowed to stabilize. Upon stabile hydrogel formation, the additional dispersion media was added to the mold, forming a fluid layer above the stable hydrogel. An appropriate amount of gelling agent was directly added to the fluid layer. Layers were only formed in purified water, but it can be assumed that all successful dispersion media can also be used in layer formation. We only examined the use of 25 mg/ml gelling agents for a final hydrogel concentration of 10 mg/ml. Diffusion of dispersion medium in layers was observed by adding a colored dye in one layer of hydrogel. For general diffusion observation, a drop of dye was applied to a stable get and allowed to diffuse into the sample.
To confirm gel formation, two protocols were employed. The first was a macroscopic observation of the gel stability. In the presence of a rigid conformation, the hydrogel should be sufficiently stable to undergo and inversion test without conformational changes. Inversion testing was performed on 2 ml samples in a 15 ml Falcon Tube. After appropriate stabilization time, the sample was inverted and monitored for deformation. A sample undergoing incomplete gel transition shows deformation in the form of flow toward the lid (lowest point of the inverted tube). A complete and stable hydrogel transition was characterized by an inversion test resulting in zero visible deformation. A successful inversion test confirmed stable macroscopic hydrogel formation.
The second protocol utilized to verify gel stability required observation at the microscopic level. Optically distinguishable foreign material was added to the gel and monitored by confocal microscopy. The 'impurity' behavior was then used to infer hydrogel characteristics. Fluorescent microspheres (15 μm Focal Check microspheres, Molecular Probes) were suspended in the dispersion medium of choice just prior to addition of the gelling agent. The microspheres were then induced with 488 nm laser excitation on a Zeiss LSM 5 Pascal confocal microscope over a 10 minute detection time interval. Confocal tracking of the microspheres presented the level of stability of the hydrogel matrix in the form of sphere immobility. Immobilization of the suspended microspheres over an extended time period was a sufficient indication of a stable hydrogel matrix.
Stable hydrogel samples were monitored for sensitivity to multiple reversing methods. The above gel transition methods were utilized to demonstrate reverse transitions from stable a stable hydrogel matrix to a less stable liquid solution. If a stable hydrogel sample can be modified such that it no longer satisfies the discussed requirements for a stable gel, the sample has reverted from a gel to a liquid phase solution. More explicitly, a sample that fails the inversion test or shows a dramatic increase in microscopic movement upon modification is no longer considered a stable hydrogel.
The cell lines examined in hydrogel cultures include COS-7, MDCK, and two strains of Astrocytes (CTX TNA2 and DI TNC1), all obtained from ECACC. Each cell line was cultured according to the ECACC recommendations. Astrocytes were grown to confluency in a liquid culture of MEM (Gibco) and the remaining cell lines were grown in DMEM (Sigma). Cells were grown with an addition of 10% fetal bovine serum (Gibco), 1.0% penicillin/streptomycin (Sigma) and 1.0% L-glutamine (Sigma). MDCK cells were grown with an additional treatment of 1.0% non-essential amino acids (Sigma). Incubation was performed at 37°C with 5% CO2 and 100% humidity. The cells were harvested from the culturing petri dishes by trypsinization. The harvested cells were spun down to a pellet by centrifugation for 5 minutes at 1700 rpm. Cells were then washed twice before a final spin down. Staining treatments and washing were performed with a standard phosphate buffered saline solution (1× PBS). The final cell pellet was then immobilized in the peptide derivative extracellular matrix by first suspending in PBS and adding the gelling agent to initiate hydrogel self-assembly. A typical cell number density used was between 5 and 12 million cells/ml of final gel volume. Nutrients were supplied to the immobilized cell suspensions by perfusion as soon as the hydrogel was stabilized. For long-term culturing, the microstructures were removed from the microchip platform and submerged in culture media under 37°C. Cells were viewed by staining with 1.0 μM calcein-AM (Invitrogen). If the cells were to be stained before in situ culture formation, calcein-AM was added to the dispersion medium. For in situ staining within the microchamber, calcein-AM was added to a perfusion solution of growth medium and perfused through the microchannels. Cells were maintained in stain solution for 30 minutes to 1 hour at 37°C before confocal observation.
We would like thank Thomas Frisk (Microsystems Technology Group in the School of Electrical Engineering, Royal Institute of Technology, Sweden) for providing microstructures. Additional thanks to Gunaratna Rajarao (Applied Environmental Microbiology, Royal Institute of Technology, Sweden) for her assistance and support. Funding for this project was provided by the Swedish Research Council and the European Community-funded CellPROM project under the 6th Framework Program, contract no. NMP4-CT-2004-500039.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.