The possibility of controlling cell populations on biointerfaces has recently been put forward as a fundamental key within the “biologization of materials research” strategy, in order to link biological systems with man-made materials and to investigate the interaction between life and matter, in parallel to the development of biomedical and biotechnological applications [1
]. In paper [1
], Niemeyer and colleagues also highlight the role of biointerfaces that are capable of precisely interacting with cells as fundamental elements, in a global plan towards modular biohybrid systems that may be used as “biofactories of the future” for more sustainable and efficient production processes.
Apart from their promising future, polymeric biointerfaces are already being used extensively in a wide set of biomedical devices and systems, in which controlling cells and cell cultures, for studying or promoting desired cell–cell, cell–material and cell–drug interactions, in conditions as biomimetic and physiological as possible, is necessary. These applications include: cell culture devices, labs- and organs-on-chips, varied microfluidic systems, tissue engineering scaffolds, active devices and even implants with improved biological performance [2
Among the different micro and nanomanufacturing processes that can be employed for the generation of biointerfaces capable of interacting at a cellular level and with the ability of controlling cell populations and thereby driving their behavior, dynamics, gene expression, differentiation or fate, it is important to mention the following: techniques derived from the electronic industry, such as photolithography, which allow for the generation of selective patterns using light polymerizing photoresins through masks [5
], can be employed for creating controlled surface topographies in a wide set of materials. The controlled topographies can be achieved either by patterning the photoresins or by subsequent chemical etchings of the substrates, upon which the pattern of photoresin is created. Such patterns or textures can be transferred to softer, biomimetic and biologically adequate materials (i.e., PDMS or poly dimethyl siloxane) by using soft-lithography related processes [6
The mass production of microtextured biodevices with polymeric biointerfaces for studying cell-material interactions can be also achieved by linking rapid prototyping and rapid tooling with microinjection molding [7
]. Hot embossing or compression molding provides a remarkable alternative for a wide set of biological and biomedical micro electromechanical systems (bioMEMS) [8
]. Electrospinning is not only capable of creating ultrathin fibers for functionalizing 2D surfaces and enhancing their viability as probes for cell culture, but can also be applied to the complex geometries of 3D biodevices, such as tissue engineering scaffolds [9
]. Micromolded electrospun hydrogel mats have been also employed recently for creating multiscale constructs, with micro and nanometric features, for interacting with cells and tissues [10
Processes enabling micro and nanopatterning of biomolecules, as functionalizations upon materials surfaces, provide improved ways of fixing/anchoring cells upon surfaces for studying a wide set of cell-material interactions. Among these, dip-pen nanolithography is a very singular nanofabrication technology that can directly write a variety of molecular patterns upon surfaces with excellent resolution, as recently reviewed [11
]. Parallel dip-pen nanolithography constitutes an evolution capable of integrating functional biomolecules on subcellular length scales, as a consequence of its constructive nature, high resolution and high throughput [12
Different techniques of recent development, enabling the creation of design-controlled, super-hydrophilic and super-hydrophobic micropatterns and transitions upon surfaces [13
], are promoting many innovative applications in the biomedical and biotechnological fields, especially in connection with high-throughput screening [14
]. Additive manufacturing of 2D½ bioMEMS and of 3D microstructures upon surfaces are also promoting the almost free-form fabrication of biointerfaces in a wide set of materials, in many cases polymers. Multi-scale combinations of additive manufacturing technologies (AMT) have been reported [16
], as well as 3D written structures capable of interacting at the single-cell level [17
]. Functionally graded microporous scaffolds upon surfaces, manufactured with biodegradable polymers combining additive processes with surface polymerizations, as biointerfaces for controlled cell cultures have been described [18
In spite of the varied processes available, in the creation of successful polymeric biointerfaces, researching synergies between them can prove a promising strategy for increasing the level of productivity, throughput and complexity of the biointerfaces created, as required for interacting with the complex human physiologies and for systematically studying cellular mechanisms in disease and cell–material interactions in biomedical devices.
In this study we present and analyze synergies between an innovative approach for surface microstructuring, aimed at the mass production of polymeric biointerfaces, and a recently developed molecular nanopatterning procedure for the straightforward or “plug-and-play” functionalization of material surfaces.
Eukaryotic cell cultures upon the biointerfaces developed, both before and after molecular patterning, help to validate the proposal and to discuss the synergies between the surface microstructuring and molecular nanopatterning techniques described. Their potential in the production of versatile polymeric biointerfaces, for being integrated into lab- and organ-on-a-chip biodevices or for being used as components of complex and biomimetic co-culture systems and cell cultivation set-ups, are also examined.
The proposed design and manufacturing processes produced polymeric substrates (thermoplastic PMMA) with microstructured surfaces, as can be observed in Figure 1
, Figure 2
and Figure 3
. First, Figure 1
a presents the computer-aided design of microstructured surfaces as an object of study, showing both positive and negative alternatives, (that is, a textured vascular-like region upon a plane and a planar vascular-like region surrounded by texture), as well as detailed views of the topographies. Figure 1
b presents the previous design converted into checkerboard-like pattern, for higher throughput and more systematic testing, ready for direct laser writing.
then presents the mold insert and Figure 3
a shows a microscopy of the hot embossed PMMA template or substrate, obtained after the direct laser writing of master models, metallization and pattern transfer by compression molding. Some details of the hot embossed microstructures, which present remarkable similarity with the original CAD geometries, are included in Figure 3
In consequence, it is important to highlight that the proposed techniques and processes allow for the creation of microstructures with overall sizes in the order of magnitude of common cell types (e.g., 10 × 10 × 10 μm3). The generated microstructures, which mimic the typical multi-scale bumpy features common of the epidermis of different types of plants and leaves, incorporate finer details (in the 2 × 2 × 2 μm3 range) with the potential to promote even subcellular interactions. A close look at the “H-like” vascular patterns helps realize that the different branches of the “H” vascular patterns have different widths: the central channels have a width of 30 μm for letting groups of three cells interact, the lateral channels have a width of 20 μm for trapping cells in couples and the more external channels of the “H” figures have a width of 10 μm, within which only a row of single cells can be arranged.
The gradients of width in the template have a twofold purpose: on one hand, changing the thickness systematically allows for the verification of the precision and viability of the manufacturing processes for achieving multi-scale and complex surface topographies, which is confirmed. On the other hand, it may support the topography-guided organization of cells, forming rows or columns, duplets and triplets, for a wide set of potential uses for labs- and organs-on-chips.
Moreover, these positive and negative features, in the form of checkerboard-like pattern, are aimed at having a versatile cell culture platform with 3D microtextures as, depending on the cell type (s) cultured, some cells may adhere to the textures, while some others may be repelled by them.
Ideally, this could lead to viable co-cultures of endothelial cells (for the vascular structures) and of other cell types (for the parenchymal tissues surrounding the vasculatures), in which the positions of the different cells may be controlled by the designed surface textures, as to achieve biomimetic physiological structures on chips.
In addition, by using a couple of microtextured surfaces or chips, culturing two different cell types upon them and placing the chips face-to-face (with the parenchymal regions facing the vascular structures), it may be possible to achieve sandwiched microenvironments to study cell–cell and cell–material interactions in depth [25
]. Furthermore, the defined transitions of planar and microtextured regions can be applied to the creation of microfluidic devices “on chips”, in which wet and dry zones may be defined just by design-controlled surface texturing upon a single functional layer or surface. This may support the development of a new generation of labs- and organs-on-chips, in which the number of components (layers, tubes, membranes, clamping elements) may be reduced, once fluids and cells are controlled by surface texturing, for which the employment of additional surface functionalizations, such as molecular nanopatterning to more precisely fix cell positions, will surely be needed, as we complete further analysis.
Once manufacturability is confirmed, cell culture experiments, both upon non-functionalized and functionalized microstructured surfaces, as detailed in Section 2.3
and Section 2.4
, lead to interesting results and discussions regarding cell–(material, topography, molecular functionalization) interactions. As first experimental results, Figure 4
a presents a set of 3T3 cells, imaged after 40 min of culture upon a microtextured sample, and shows how cells are organized due to the presence of topographical transitions (rows, duplets, triplets and clusters can be observed). Figure 4
b shows the same sample, after washing with PBS, which removes almost all cells.
We conclude that, although the use of controlled microtopographies upon PMMA substrates can be applied for organizing cellular suspensions on chips, it is not enough for fixing the cells to desired positions for long-term studies. In order to fix the cells to desired (and design-controlled) positions for longer cell culture studies, in which long-term viability, gene-expression processes, progress of disease or effects of drugs for healing (among other options) can be studied, it is necessary to post-process the hot embossed samples and to convert them into more active biointerfaces. In this study, we resorted to molecular patterning of fibronectin using PPL. Fibronectin is a high-molecular weight glycoprotein of the extracellular matrix that binds to membrane-spanning receptor proteins called integrins, as well as to other extracellular matrix proteins, including collagen, fibrin and heparan sulfate proteoglycans [26
Details of the functionalization and subsequent cell culture protocols for impact assessment are given in Section 2.3
and Section 2.4
. As an example of the post-processing performed to some hot embossed samples, Figure 5
a shows the fibronectin functionalization obtained by PPL, highlighted in fluorescent microscopy as a pattern of green dots/squares. Existing as an equally spaced pattern upon a checkerboard design, a similar number of functionalized dots/squares was applied upon the planar regions (the pale gray zones of the checkerboard) and upon the microtextured or microstructured ones (the dark gray zones of the checkerboard).
The results of the cell culture upon fibronectin-functionalized PMMA replicas can be observed in Figure 5
b–d, while cell culture result upon non-functionalized PMMA substrates is presented in Figure 5
e, showing just one image as an example, due to the lack of cells. Subsequently, with the support of Figure 5
b–d, the influence of microtexturing and molecular patterning on the cellular attachment to desired positions of the PMMA substrates is studied and presented in the summarizing graph of Figure 6
. The analysis is performed simply by counting the number of living cells (per visual field) attached to planar and to textured regions of the functionalized and non-functionalized substrates.
According to results presented in Figure 5
and summarized in Figure 6
, the microstructured PMMA samples are not capable of fixing the fibroblasts to desired positions, unless a molecular functionalization is applied. Without the functionalization with fibronectin, the topographies do not have any special impact on cell attachment, only on the organization of cells in suspension, as also shown in the first experiment presented in Figure 4
. However, once functionalized, an interesting synergy between surface microstructuring and molecular patterning can be observed: cells seem to attach in a much better way to regions, in which microtextures and fibronectin co-exist.
4. Discussion: Potentials, Limitations and Continuation Proposals
The main novelty of the presented study, as compared with previous studies from our team and with the introduced state-of-the-art methods, relies on: (1) the degree of multiplexation, achieved by 3D direct laser writing of large fields, in which designs are repeated in a checkerboard pattern; (2) the precision of mass production tools obtained by electroplating and employed for the systematic replication of polymeric test probes by hot embossing; and (3) the combined utilization of surface texturing and molecular nanopatterning for improving the degree of control upon cells on chips. Such synergies are, to our knowledge, presented and analyzed for the first time here.
In previous research [27
] fibronectin was found to be appropriate for controlling cell positions upon cell culture plates and devices, but here we show a reinforced ability, which we attribute to the adjuvant effect of the design-defined microstructures. It is likely that the topographies created, which include details in the subcellular scale range, provide anchorage elements for the cells, in which the pseudopodia of the eukaryotic cell membrane can grasp in an enhanced way, as compared to the situation upon completely planar regions. These results are in agreement with varied reported effects of surface topography on cell adhesion and motility, which were described in recent research papers [28
]. Apparently, the microtextures do not only provide anchorage elements for the cells, but may also serve as reservoirs for the patterned fibronectin, which may explain the special synergy found between texturing and molecular patterning. Although in this study the PPL process has been applied to create a periodic pattern of fibronectin spots, innovative modifications to the traditional PPL technique allow for more customizable and flexible surface functionalizations, including the creation of complex paths and of both periodic and aperiodic patterns, even with multiple functionalizing inks, in ways as versatile as those achievable with ink upon paper when using a desktop ink-jet printer, but at the micro and nanoscale [31
]. Future studies using more flexible printing methods will progress in these directions.
Regarding Figure 6
, cells are counted after fixing and nuclei staining. In consequence, there is no relation between the initial concentration (1 million cells) of living cells and the fixed ones that survive the whole experiment. Unfixed cells are washed away, even if they might have been still searching for anchorage regions. The figure legend denotes that the number of fixed cells per visual field (800 µm × 800 µm square area) is presented. Although we are not able to observe cell division, the distribution of adhered cells over different regions, due to the presence of textures and fibronectin, can be observed clearly.
It is important to consider that cell adhesion involves the interaction of cells with other cells or with the extracellular matrix (ECM), like fibronectin, which is expressed by fibroblasts and many other cells. Hence, fibroblast cells are chosen as the best cell model for attaching to the printed fibronectin pattern. Fibronectin is an important constituent of the extracellular matrix and it interacts with specific domains on the fibroblast surface. Namely, it binds to membrane-spanning receptor proteins-integrins. Integrins are proteins that function mechanically, by attaching the cell cytoskeleton to the extracellular matrix, and biochemically, by sensing whether adhesion has occurred. In any case, since fibronectin is a major component of the extracellular matrix, these observations suggest that it may provide at least one of the signals, by which the matrix conveys the competence that permits fibroblasts to replicate in the presence of an appropriate progression signal, as previous studies also suggest [35
We would like to point out that the technological aspect was developed and demonstrated as a type of pilot production plant, which combines bioinspired design approaches, 3D direct laser writing, electroplating (for the creation of mass-production tools), hot embossing (for creating replicas of polymeric sheets that enable systematic testing) and molecular nanopatterning for improving the degree of control upon cultured cells. The hot embossed polymeric sheets or biointerfaces prove interesting in terms of usability, as compared with thicker injection-molded, especially regarding manipulation and visualization. The presented strategy has been preliminarily validated with a robust cell line, and it was also chosen for its affinity with the fibronectin spots employed as molecular pattern, although we expect to progress towards the application of this set of tools for developing complete organ-on-chips and for answering specific biological questions.
In regards to more relevant industrial applications, the manufacture of technically relevant samples, which may require functional areas of some cm2, is still a limitation of presented chain of processes, as the writing field in available two-photon polymerization systems is still limited to just some hundreds of μm2. Nevertheless, by stitching writing fields, it is possible to manufacture samples of several mm2, at least very adequate for in vitro testing, as has been demonstrated. The attainable accuracy is, in any case, outstanding and remarkable for conceptual proofs of different micro and nanosystems, but the writing speed and attainable part size are still limited for mass production.
Once photopolymerization with micrometric precision improves in terms of achievable part sizes, the mass-production of a wide set of surface microstructured devices and components will be relatively straightforward, by just directly converting polymeric master models into hot embossing tools by electroplating and subsequent compression molding, as shown in this study. Polymer pen lithography also demonstrates precise and versatile molecular functionalizations, as post-process for enhancing the performance of mass-produced polymeric components by hot embossing.
Alternatively, a purposely designed microcontact stamp that matches the layout of the microstructures may provide even more control over the distribution of the functionalization on the positive and negative areas. It may also synergize interestingly with mass-produced substrates by micro-injection molding, as an alternative to hot embossing when three-dimensional shape complexity is needed.