Compartmentalized bioencapsulated liquefied 3D macro-construct by perfusion-based layer-by-layer technique

Abstract

Self-supporting, millimeter length 3D constructs consisting of individualized liquefied compartments, were produced using cell encapsulated hydrogel beads as building blocks. A perfusion-based layer-by-layer approach was used that allowed bioencapsulated beads to assemble, pattern, hold and attach to produce non-liquefied 3D constructs with controlled shape, displaying the binding feature of a continuous nanometric multilayer coating. No binders or crosslinking strategies were used. The internal microenvironment of this 3D construct was modified from solid to liquefied state by chelation. MTS and live–dead assays showed enhanced L929 cell viability with liquefied 3D constructs, compared to non-liquefied ones. The proposed technique opens new prospects to create complex 3D polyelectrolyte based constructs for tissue engineering applications.

---

Tissue engineering (TE) approaches present us with technologies that are helpful to restore, repair or replace a specific damaged tissue in a patient. The main aim is to produce and control the 3D design, which closely mimics the native tissue, but also to help in defining structure-to-function relationships to model cellular events and diseases. Among the various materials used in TE strategies, hydrogels are of great interest owing to their characteristics that resemble several features of the natural extracellular matrix along with efficient and homogeneous cell encapsulation. Cell encapsulation by various methodologies has shown their great potential in the field of TE. Since their inception which produced solid spherical beads/microspheres by inotropic gelation^1–3 they are currently being used as building blocks to boost the engineering of new tissue and organ regenerative strategies.^4–9

Cell encapsulated hydrogel microspheres/beads having a solid core display good viability due to efficient diffusion of nutrients, oxygen, and cell metabolites.^10–12 A two-step process was introduced for post-production of beads, first by using a layer-by-layer (LbL) technique with conventional dipping method and followed by a chelation process, to transform them into polyelectrolyte multilayered capsules (PMC).^13–17 LbL is an adequate technique to process nanostructured multilayerd films through spontaneous sequential adsorption of distinct materials being frequently used in the biomedical field.^18,19 PMC's are characterized by the presence of liquefied environments enclosed within a permselective membrane with increased mass transport capacity than that of solid cores, resulting in higher cell viability rates.^13–17 Customization of such liquefied environments by adding/modifying new components or materials used for LbL process could further provide suitable microenvironments to control different biological functions, such as cell proliferation, differentiation, and migration.

Despite the potential benefits of polyelectrolyte multilayered capsules, their use in cell-based bottom-up strategies has been severely hampered because of the difficulty to assemble or pattern them over length scales for mimicking the complex 3D architecture and organization of native tissues. This is primarily due to physical characteristics of PMC which are soft and can be easily deformed due to frail mechanical features.²⁰ The difficulty is even more compounded if using conventional techniques of LbL techniques, where prefabrication of strong 3D core structure is essential that can withstand the stress/strain experienced during the coating process. Other bottom-up approaches using cell-laden hydrogels, currently, propose use of nano/micro scale technologies, such as microdroplet technologies based on bioprinting, microfluidics, acoustic/magnetic fields, and surface tension.^21–28 Most of these technologies based on complex steps require advanced and expensive machines to fabricate the desired constructs. Additionally, in these strategies cells are confined inside a solid/elastic polymeric matrix. An interesting alternative would be to change solid matrix to liquefied or non-adhesive 3D environments.

Recently we had reported the new perfusion-based LbL technique using drop-wise addition of polyelectrolyte solutions, instead of conventional dipping or spraying, that apart from coating also display real time binding potential of nanometeric multilayer to produce organized porous 3D macrostructures entirely made of nanometric membranes.^29–31 Performed over a perforated base this technique primarily initiates the assembly or stacking of freeform spherical objects, acting as sacrificial templates, which stick together into easily modulated but frail 3D core structure that consolidates during the multilayer process while using significantly small volume of polyelectrolyte solutions. These nanometric multilayer coatings produce 3D porous macrostruture which is basically the negative replica of the assembled 3D core material. Binding/sticking feature adds versatility to the conventional LbL methodology by directly organizing building blocks to produce 3D core structures, thereby, eliminating the use of binders or crosslinking strategy which is otherwise necessary for conventional LbL approach. This is an important outcome of this technique that eliminates the process of prefabricating strong 3D core objects. We anticipate using beads encapsulating cells as templates to produce 3D constructs via a perfusion-based LbL that is potentially more advantageous than the above mentioned techniques.^1,2,21–28 This strategy is a biological-friendly technique that is a suitable alternative to other methods that rely in crosslinking agents, ionization, UV-Visible light, or elevated temperatures during processing, which, although cytocompatible, may also be regarded as potentially toxic or denaturing.³² In addition to this it also can aid in interchanging the internal microenvironment also by cell friendly liquefaction process.

Considering these arguments, we suggest a novel approach that will allow an active spatiotemporal control over cell-encapsulated spheres to produce 3D structures with well-defined geometrical features to fabricate liquefied 3D bio-construct as seen in Scheme 1. We show how these cell-laden beads (Scheme 1A) may be used as building blocks to produce 3D structures of solid microspheres (Scheme 1B). The templates can then be chelated, leading to a liquefied 3D construct held solely by the assembled multilayers (Scheme 1C).

Scheme 1 Schematic representation of the process to fabricate liquefied 3D structures using cell-encapsulated hydrogel beads assembled by a perfusion-based LbL technique. (A) Cells are encapsulated in hydrogel beads, produced, for example, using alginate-based formulations. The beads are packed into defined shapes and coated with multilayer films; (B) 3D patterned constructs are obtained. This intermediate product can be referred to as non-liquefied construct. (C) Liquefied 3D construct after dissolution of the template beads.

As a proof-of-concept, alginate beads (around 2 mm in diameter) were first obtained by ionotropic gelation with calcium ions (henceforth referred to as Ca-Alg) without cells. Simple 3D structures were obtained by placing random spheres next to each other over a perforated base, which led to spontaneous aggregation until they form a freeform subtle 3D core structure. We observed that the aggregation of such elementary beads was spontaneous, which can be attributed to the effect of lateral capillary forces due to the hydrophilicity of the bead surface and the multilayer interfaces that in turn hold the structure together.³³ With the aid of appropriate molds, it was possible to assemble such core structures with distinct lengths, heights and more complex shapes, having minimum distortion and good uniformity of size and volume. Finally the initially fragile structure of Ca-Alg agglomerated beads was stabilized by coating with 6 pairs of alginate/chitosan multilayers which remained intact throughout the process. Some examples of the simple and complex structures (with molds) are shown in Fig. 1A and B, respectively.

Fig. 1 3D assembly of Ca-Alg beads with different shapes and sizes, bound by 6 pairs of alginate/chitosan multilayers. (A) Simple geometries with different external shapes and height scales. (B) Hollow cylinders with different internal diameters.

To show the effective coating and binding capability endured by the multilayers, the Ca-Alg beads were liquefied by immersing the constructs in EDTA solution. When observed by optical microscopy, transformation of the dense and opaque Ca-Alg core (Fig. 2A) to a liquefied and transparent one (Fig. 2B) could be easily assessed by the improved visualization of a scale placed in the background of the construct. At higher magnification, the multilayer coatings assembled around the individual hydrogel spheres and contact points could be easily distinguished from each other – see Fig. 2C. It is noteworthy to mention that the final 3D constructs even though liquefied displayed fair structural integrity. This shows that such apparently fragile structures are in fact robust organizations that withstand every stage of their synthesis and posterior handling without disrupting or puncturing and more importantly maintaining their initial configuration. These observations confirm the formation of stable and continuous multilayer coating along the external and internal surface outlines/contours of the whole construct. This is only possible by following this LbL technique where in the 3D assembly of hydrogel beads once produced is not constrained by any substrate/surface/mold to hold the shape, which can be a case using permanent mold, thereby do not interfere with the LbL process. These results validate the process and application of the perfusion-based LbL technique to produce liquefied 3D constructs using hydrogel microspheres. We observed that most of the characterization techniques employed for single bead were not satisfactory to evaluate the final 3D constructs because of their morphology.

Fig. 2 Images of the 3D construct taken by optical microscopy during the progression of liquefaction process in 0.05 M EDTA solution. (A) Initial aspect before core liquefaction. Note that the scale underneath the construct is not visible. (B) Aspect after 20 min of immersion in EDTA. The scale underneath is clearly visible. (C) Higher magnification of (B), showing the individual multilayer coatings and connections. Scale bars: 1 mm.

At this stage, we hypothesized that cell-encapsulating hydrogel beads could also produce same results. With this strategy the constructs could be adapted from, one having the solid beads or non-liquefied, (Scheme 1B), which has a more rigid microenvironment and would come at the expense of reduced cell mobility or the other with liquefied beads (Scheme 1C) with increased cell mobility. We need to mention here, that construct with rigid environment (Scheme 1B) can also be exploited for the cell types that need such specific microenvironment. Ca-Alg beads encapsulating L929 fibroblast-like cells were used as building blocks. Fig. 3A shows an optical microscopy micrograph of non-liquefied constructs maintained in cell culture medium with encapsulated cells. No signs of blockage or complexation were found in the open spaces between spheres, suggesting a uniform multilayer perfusion coating – see Fig. 3B. This is advantageous for cell encapsulated constructs or assemblies, since it increases the reach of cell culture medium that will ensure perfusion of nutrients even to the inside of whole construct. The black dots in Fig. 3B also evidence the presence of cells inside the bound beads, spatially confined and protected within the permeable multilayer wall.

Fig. 3 Optical images of the 3D constructs with cells encapsulated. (A) Honeycomb arrangement of cell-encapsulating beads non-liquefied constructs after 3 days of culture. (B) Focus on the gap between beads within the construct after perfusion-LbL coating. (C) Liquefied 3D construct after 3 days of culture. (D) Liquefied 3D construct after 7 days of culture. Scale bars: 1 mm.

In Fig. 3C and D, cell-encapsulating liquefied constructs are shown after 3 or 7 days of incubation with cell culture medium, respectively. While in both cases cells were detected, in the latter, cells had proliferated and filled the beads as suggested by the increased optical density. This whole arrangement looks like the assembly of cell-laden hydrogel beads with individual multilayer walls that behave as individual compartments. The versatility of this fabrication process along with choice of customizing building block features may provide ways to address different tissue engineering requirements: the complexity of the proposed 3D construct could be tailored and enhanced by producing hierarchical constructs using different sizes of microspheres or encapsulating single or multiple cell types in a same construct or in a same hydrogel bead to promote directed tissue assembly, organ engineering or develop soft matter models.^34–37 Furthermore, it may be speculated that the permselective properties of the multilayer boundaries could be tuned by selecting different functional materials or vary the number of layers to modulate the flow of nutrients and cellular waste.

The biological response of L929 cells encapsulated in individual Ca-Alg beads, non-liquefied 3D constructs, liquefied 3D constructs and was assessed by live–dead and MTS assays. Ca-Alg beads were selected as reference as reference group using an equal number of unbound beads to that in 3D constructs. The green fluorescence detected in live–dead assays revealed that encapsulated cells remained viable up to 7 days of culture both in non-liquefied (Fig. 4A) and liquefied (Fig. 4B) constructs.

Fig. 4 L929 cell viability up to 7 days of culture. (A) Live–dead fluorescence assay at day 7 of cell culture of non-liquefied and (B) liquefied 3D constructs. Living cells were stained green by calcein-AM and dead cells red by propidium iodide. Scale bars: 200 μm. (C) MTS colorimetric assay of control, non-liquefied and liquefied 3D constructs after 3 and 7 days of culture (* for same sample groups, # for different sample groups, p < 0.05). The error bar for liquefied sample after 3 days is too small to be seen.

These observations suggest the process of making these constructs did not jeopardize cell viability, and the flow of culture medium was efficient in providing nutrients to the encapsulated cells. The quantitative MTS assay when comparing the values obtained for the 3^rd and 7^th day showed that whereas in both reference group and non-liquefied constructs cell metabolic activity decreased, it increased significantly in liquefied constructs with increasing cell culture times – see Fig. 4C. This is a key result because it shows and supports the positive effect of having cells encapsulated within a liquefied environment, rather than within a solid one. Comparing to solid core structures, once can envisage selecting and modifying some variables within the system such as the addition of application specific bioactive molecule (e.g., proteins, growth factor, crosslinking agents, polymer concentration, number of layers). Furthermore, characteristics like increased gas/nutrient diffusion, which is a main concern in 3D macro-constructs, and adaptability to various shapes and patterns make of these liquefied architectures an alternative technique to other crosslinked structures that aim to mimic ECM or develop tissues.

Conclusion

Liquefied cell-encapsulating beads were assembled for the first time with geometrical control in 3D using perfusion-based LbL technique. This approach aims to provide control over size, shape, and other features of bottom-up assemblies of cell-based devices in a scalable manner. Three dimensional constructs were produced using cell-laden beads acting as building blocks that could be aggregated/assembled into various geometrical features, bound together by LbL coatings. The hydrogel beads behaved as individual compartments and the liquefied structure showed significant improved cell viability. The improvement of this platform technology may be done by customizing the selected biomaterials to mimic the extracellular matrix within the liquefied space and combining different cell types encapsulated individually. In the future, beads with different diameters may be used as building blocks by modifying the extrusion and ionotropic gelation conditions of alginate droplets. Selecting different materials for the LbL coating may lead to the development of structures exhibiting selective permeability. We envision that such technologies will provide hydrogel-based 3D constructs that can support long-term cell encapsulation to create complex tissue and organ substitutes with 3D organization for tissue engineering applications.

Experimental section

Production of 3D constructs

Water soluble polyelectrolytes were selected for this work: chitosan (Protasan UP CL 213, viscosity 107 mPa s, NovaMatrix, Norway) and low viscosity alginate from brown algae (250 cP, SigmaAldrich, USA). Sodium chloride (NaCl), glacial acetic acid, sodium hydroxide (Panreac Spain), calcium chloride (CaCl₂, Merck, Germany), ethylenediaminetetraacetic acid (EDTA, Sigma-Aldrich, USA) were used as received.

Alginate was dissolved in 0.15 M NaCl at a concentration of 1.2% w/v. The pH of the alginate solution was adjusted to 7.0. When encapsulating cells, L929 fibroblast-like cells were added to this solution at a concentration of 1 × 10⁶ cells per mL. The alginate solution for ionotropic gelation was added drop-wise to a 0.1 M CaCl₂ solution using a 21G needle and stirred for 20 min to form Ca-Alg beads. Then, the hydrogels beads were collected and rinsed with 0.15 M NaCl.

Molds were used to produce uniform constructs with a fixed height of 5 mm and transferred to a perforated plate. The beads were directly transferred to the same plate and arranged in predetermined organizations. This was followed by perfusion-LbL with a sequential drop-wise addition of chitosan and alginate over the assembled 3D template until 6 bilayers were assembled. The concentration of alginate and chitosan was 1 mg mL⁻¹ at a pH value of 7.0. Intermediate rinsing steps were performed using 0.15 M NaCl. The obtained non-liquefied constructs were then immersed in 0.05 M EDTA at pH 7.0 for 7 min to liquefy the alginate cores. The formation of multilayers coating was demonstrated by optical microscope only. Additional microscopic observations were not possible because of the size and characteristics of the construct.

Live–dead assays

Ca-Alg beads with cells (reference samples), and both liquefied and non-liquefied 3D constructs encapsulating L929 cells were incubated for 3 and 7 days at 37 °C in a humidified 5% CO₂ atmosphere. Calcein-AM and propidium iodide (PI) dyes (1 mg mL⁻¹, Molecular Probes, Invitrogen, USA) were used to perform the live–dead assay at each time-point. Briefly, samples were incubated with the dyes at 37 °C protected from light. After 15 min, samples were washed with phosphate-buffered saline (PBS). Samples were immediately visualized by fluorescence microscopy (Axioimage RZ1M, Zeiss, Germany) in the dark. All pictures were obtained in Z-stack mode using the AxioVision software.

MTS viability assay

Metabolic activity of encapsulated cells in sample groups was assessed using the MTS colorimetric assay (Cell Titer 96 AQueousOne Solution Cell Proliferation Assay, Promega, USA). Ca-Alg beads with cells (reference samples), and both liquefied and non-liquefied 3D constructs encapsulating L929 cells were placed in cell culture plates (n = 1 construct per well, performed in triplicate) and incubated at 37 °C, 5% CO₂. Briefly, the culture medium was removed and 1 mL of serum-free Dulbecco's Modified Eagle Medium (DMEM) containing the MTS solution (1 : 5 dilution ratio) was added to each well. Samples were incubated in the dark at 37 °C and 5% CO₂. After 3 h, 100 mL of each well were transferred to a 96-well plate (in triplicate). The amount of formazan product was measured by absorbance at a wavelength of 490 nm using a microplate spectrophotometer (Synergy HT, Bio-TEK). The background was corrected by subtracting the absorbance obtained from samples without cells to those with cells encapsulated. All sample-related effects were considered to be statistically significant if p-value was less than 0.05, using t-Student test. All data was analyzed using OriginPro 8 (OriginLab, Northampton, MA, USA).

Acknowledgements

The authors acknowledge the financial support by the Portuguese Foundation for Science and Technology (FCT) for grants SFRH/BPD/48948/2008, SFRH/BD/69529/2010 and SFRH/BPD/95446/2013, and project PTDC/CTM-BIO/1814/2012, co-financed by the Operational Human Potential Program (POPH) developed under the scope of the National Strategic Reference Framework (QREN) from the European Social Fund (FSE).

References

1. U. A. Gurkan, S. Tasoglu, D. Kavaz, M. C. Demirel and U. Demirci, Adv. Healthcare Mater., 2012, 1, 149–158.
2. K. Y. Lee and D. J. Mooney, Chem. Rev., 2001, 101, 1869–1880.
3. J. Malda, J. Visser, F. P. Melchels, T. Jüngst, W. E. Hennink, W. J. A. Dhert, J. Groll and D. W. Hutmacher, Adv. Mater., 2013, 25, 5011–5028.
4. Y. Du, E. Lo, S. Ali and A. Khademhosseini, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 9522–9527.
5. D. L. Elbert, Curr. Opin. Biotechnol., 2011, 22, 674–680.
6. G. Eng, B. W. Lee, H. Parsa, C. D. Chin, J. Schneider, G. Linkov, S. K. Sia and G. Vunjak-Novakovic, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 4551–4556.
7. P. M. Kharkar, K. L. Kiick and A. M. Kloxin, Chem. Soc. Rev., 2013, 42, 7335–7372.
8. K. H. Lee, D. Y. No, S.-H. Kim, J. H. Ryoo, S. F. Wong and S.-H. Lee, Lab Chip, 2011, 11, 1168–1173.
9. J. S. Liu and Z. J. Gartner, Trends Cell Biol., 2012, 22, 683–691.
10. A. D. Augst, H. J. Kong and D. J. Mooney, Macromol. Biosci., 2006, 6, 623–633.
11. B. B. Lee, P. Ravindra and E. S. Chan, Chem. Eng. Technol., 2013, 36, 1627–1642.
12. G. Orive, R. M. Hernandez, A. R. Gascon, R. Calafiore, T. M. S. Chang, P. D. Vos, G. Hortelano, D. Hunkeler, I. Lacik, A. M. J. Shapiro and J. L. Pedraz, Nat. Med., 2003, 9, 104–107.
13. Y. Wang, A. Angelatos and F. Caruso, Chem. Mater., 2008, 20, 848–858.
14. A. G. Skirtach, A. M. Yashchenok and H. Mohwald, Chem. Commun., 2011, 47, 12736–12746.
15. S. De Koker, R. Hoogenboom and B. G. De Geest, Chem. Soc. Rev., 2012, 41, 2867–2884.
16. N. L. Costa, P. Sher and J. F. Mano, Adv. Energy Mater., 2011, 13, B218–B224.
17. C. R. Correia, P. Sher, R. L. Reis and J. F. Mano, Soft Matter, 2013, 9, 2125–2130.
18. J. Borges and J. F. Mano, Chem. Rev., 2014, 114, 8883–8942.
19. R. R. Costa and J. F. Mano, Chem. Soc. Rev., 2014, 43, 3453–3479.
20. R. Mercadé-Prieto and Z. Zhang, J. Microencapsulation, 2012, 29, 277–285.
21. B. Derby, Science, 2012, 338, 921–926.
22. J. Karoly, N. Cyrille, M. Francoise, M. Keith, V.-N. Gordana and F. Gabor, Biofabrication, 2010, 2, 022001.
23. F. P. W. Melchels, M. A. N. Domingos, T. J. Klein, J. Malda, P. J. Bartolo and D. W. Hutmacher, Prog. Polym. Sci., 2012, 37, 1079–1104.
24. A. Blaeser, D. F. Duarte Campos, M. Weber, S. Neuss, B. Theek, H. Fischer and W. Jahnen-Dechent, BioRes. Open Access, 2013, 2, 374–384.
25. B. Guillotin and F. Guillemot, Trends Biotechnol., 2011, 29, 183–190.
26. K. Jakab, C. Norotte, F. Marga, K. Murphy, G. Vunjak-Novakovic and G. Forgacs, Biofabrication, 2010, 2, 022001.
27. A. Schober, U. Fernekorn, S. Singh, G. Schlingloff, M. Gebinoga, J. Hampl and A. Williamson, Eng. Life Sci., 2013, 13, 352–367.
28. K. Jakab, A. Neagu, V. Mironov, R. R. Markwald and G. Forgacs, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 2864–2869.
29. P. Sher, C. A. Custódio and J. F. Mano, Small, 2010, 6, 2644–2648.
30. P. Sher and J. F. Mano, Bioinspired, Biomimetic Nanobiomater., 2012, 1, 245–251.
31. J. M. Silva, N. Georgi, R. Costa, P. Sher, R. L. Reis, C. A. Van Blitterswijk, M. Karperien and J. F. Mano, PLoS One, 2013, 8, e55451.
32. S. R. Van Tomme, G. Storm and W. E. Hennink, Int. J. Pharm., 2008, 355, 1–18.
33. N. Bowden, A. Terfort, J. Carbeck and G. M. Whitesides, Science, 1997, 276, 233–235.
34. D. Gonzalez-Rodriguez, K. Guevorkian, S. Douezan and F. Brochard-Wyart, Science, 2012, 338, 910–917.
35. V. Mironov, R. P. Visconti, V. Kasyanov, G. Forgacs, C. J. Drake and R. R. Markwald, Biomaterials, 2009, 30, 2164–2174.
36. G. Villar, A. D. Graham and H. Bayley, Science, 2013, 340, 48–52.
37. T. Xu, W. Zhao, J.-M. Zhu, M. Z. Albanna, J. J. Yoo and A. Atala, Biomaterials, 2013, 34, 130–139.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra11674g

---