BIOARTIFICIAL TISSUE

The invention relates to a combined system for three-dimensional (3D) bioprinting, based on encapsulating of cells into an alginate hydrogel or alginate/gelatin hydrogel, followed by an overlay comprising a morphogenetically active polymer, polyphosphate-calcium complex or biosilica, providing a useful matrix for the cells, including bone-forming cells, to proliferate and be become functionally active, for example, bone cells to mineralize.

Background of Invention

The introduction of rapid prototyping (RP) or solid freeform fabrication (SFF) techniques has opened new possibilities not only in various technical areas, but also in the biomedical field, in particular in surgery and dentistry (Yeong WY, Chua CK, Leong KF, Chandrasekaran M (2004) Rapid prototyping in tissue engineering: challenges and potential. Trends Biotechnol 22:643-652; Butscher A, Bohner M, Hofmann S, Gauckler L, Muller R (2011) Structural and material approaches to bone tissue engineering in powder-based three-dimensional printing. Acta Biomater 7:907-920; Silva NR, Witek L, Coelho PG, Thompson VP, Rekow ED, Smay J (2011) Additive CAD/CAM process for dental prostheses J Prosthodont 20:93-96). These techniques allow the fabrication of custom-made scaffolds/implants for bone regeneration and repair, perfectly fitting into the cavity of the bone defect. The production of customized components using a bottom-up approach has advantages compared to subtractive manufacturing, e.g., cost effectiveness, saving of resources and avoidance of waste.

RP/SFF techniques used for biomedical applications can be grouped into (i) printer-based; (ii) extrusion-based; and (Hi) beam-based methods (Billiet T, Vandenhaute M, Schelfhout J, Van Vlierberghe S, Dubruel P (2012) A review of trends and limitations in hydrogel-rapid prototyping for tissue engineering. Biomaterials 33:6020-6041). In general, the fabrication of scaffolds using these techniques is based on a 3D design of the scaffold which is transferred into a STL file format. This is then split into a series of consecutive virtual slices to allow the 3D printer / RP machine to produce the scaffold in a layer-by- layer fashion.

However, these developments are still in a very preliminary phase and there are many challenges both in terms of the technologies to be applied and with respect to the material properties and the patient's response.

3D cell printing

It is generally agreed that the technique of three-dimensional (3D) printing of cells has an enormous potential for the fabrication of complex 3D tissues and organs (reviewed in: Wiist S, Muller R, Hofmann S (2011) Controlled positioning of cells in biomaterials - Approaches towards 3D tissue printing. J Funct Biomater 2: 119-154). However, there is still an unmet demand of a suitable printable scaffold material into which the cells can embedded. The desired scaffold should not only serve as a mechanical framework but should also be functionally active as a template that is able to elicit the expression of essential morphogens and cytokines by the cells in a spatially and temporally controlled way.

Polyphosphates

Polyphosphates (polyP) are naturally occurring linear polymers up to several hundreds of - - phosphate (Pi) residues (Kulaev IS, Vagabov V, Kulakovskaya T (2004) The Biochemistry of Inorganic Polyphosphates. New York: John Wiley & Sons Inc; Rao NN, Gomez-Garcia MR, Kornberg A (2009) Inorganic polyphosphate: Essential for growth and survival. Annu Rev Biochem 78:605-647). Their synthesis is catalyzed by specific polyphosphate kinases that polymerize polyP by transfer of the terminal Pi from ATP to the growing polyP chain (reviewed in: Schroder HC, Lorenz B, Kurz L, Muller WEG (1999) Inorganic polyP in eukary- otes: enzymes, metabolism and function. In: Inorganic Polyphosphates - Biochemistry, Biology, Biotechnology (Schroder HC, Muller WEG, eds). Prog Mol Subcell Biol 23:45-81). The recombinant enzymes can be used for the preparation of polyP under physiological, ambient conditions (Ahn K, Kornberg A (1990) Polyphosphate kinase from Escherichia coli. Purification and demonstration of a phosphoenzyme intermediate. J Biol Chem 265: 11734-11739). Besides the enzymatically produced bio-polymer (bio-polyP), polyP can be synthesized chemically. PolyP can be degraded by specific exopolyphosphatases and endopolyphospha- tases that have been isolated from both prokaryotic and eukarytic organisms (e.g., Lorenz B, Muller WEG, Kulaev IS, Schroder HC (1994) Purification and characterization of an exopol- yphosphatase activity from Saccharomyces cerevisiae. J Biol Chem 269:22198-22204).

The present inventors found already 15 years ago that polyP is present in bone tissue (Leyhausen G, Lorenz B, Zhu H, Geurtsen W, Bohnensack R, Muller WEG, Schroder HC (1998) Inorganic polyphosphate in human osteoblast-like cells. J Bone Mineral Res 13:803- 812; Schroder HC, Kurz L, Muller WEG, Lorenz B (2000) Polyphosphate in bone. Biochemistry (Moscow) 65:296-303) and is a substrate for the principle enzyme involved in bone formation, the bone specific alkaline phosphatase (Lorenz B, Schroder HC (2001) Mammalian intestinal alkaline phosphatase acts as highly active exopolyphosphatase. Biochim Biophys Acta 1547:254-261); this enzyme is assumed to provide the inorganic phosphate required for HA synthesis. These results have subsequently been confirmed by other groups (e.g., St- Pierre JP, Pilliar RM, Grynpas MD, Kandel RA (2010) Calcification of cartilage formed in vitro on calcium polyphosphate bone substitutes is regulated by inorganic polyphosphate. Acta Biomater 6:3302-3309; recent review: Kulakovskaya TV, Vagabov VM, Kulaev IS (2012) Inorganic polyphosphate in industry, agriculture and medicine: Modern state and outlook. Process Biochemistry 47: 1-10) and provided the basis for the development of scaffold materials or coatings for bone replacement/dentistry.

The morphogenetic activity of polyP used as a Ca ²⁺ complex has been demonstrated in osteoblast-like SaOS-2 cells. The polyP Ca ²⁺ complex (polyP chain length: 45 phosphate residues) was found not only to induce hydroxyapatite formation but also enhances the expression of the gene encoding BMP-2 in SaOS-2 cells (Wang XH, Schroder HC, Diehl-Seifert B, Kropf K, SchloBmacher U, Wiens M, Muller WEG (2013) Dual effect of inorganic polymeric phos- phate/polyphosphate on osteoblasts and osteoclasts in vitro. J Tissue Eng Regen Med 7:767- 776). In addition, this complex induces the alkaline phosphatase, both the activity of this enzyme and the expression of the gene encoding the bone alkaline phosphatase, the tissue nonspecific alkaline phosphatase (Muller WEG, Wang XH, Diehl-Seifert B, Kropf K, SchloBmacher U, Lieberwirth I, Glasser G, Wiens M, Schroder HC (2011) Inorganic polymeric phosphate/polyphosphate is an inducer of alkaline phosphatase and a modulator of intracellular Ca ²⁺ level in osteoblasts (SaOS-2 cells) in vitro. Acta Biomater 7:2661-2671). On the other hand, the polyP ^» Ca ²⁺ -complex inhibits the progression of RAW 264.7 cells into osteoclasts, most likely by impairment of the phosphorylation of ΙκΒα by the ΙκΒα kinase, a process that inhibits the differentiation of pre-osteoclasts into mature osteoclasts (Wang XH, Schroder HC, Diehl-Seifert B, Kropf K, SchloBmacher U, Wiens M, Muller WEG (2013) Dual effect of inorganic polymeric phosphate/polyphosphate on osteoblasts and osteoclasts in vitro. J Tissue Eng Regen Med 7:767-776). - -

Alginate hydrogel

The prerequisite for a matrix that can be used for cell printing is the biological inertness and the property to undergo a hardening process after printing. Such a polymer is alginate that meets the osmolar requirements of the cells and maintains their viability (Cohen DL, Tsavaris AM, Lo WM, Bonassar LJ, Lipson H (2008) Improved quality of 3D-printed tissue constructs through enhanced mixing of alginate hydrogels. Proc TERMIS -North Am 2008 Ann Conf Expos. San Diego, CA). Alginate is a polysaccharide from algae and bacteria, consisting of β- (l-4)-linked D-mannuronic acid and a-(l-4)-linked L-guluronic acid residues (Smidsrod O, Skak-Brask G (1990) Alginates as immobilization matrix for cells. Trends Biotechnol 8:71- 78). This plant anionic polysaccharide has the capacity to absorb water under formation of a hydrogel. Alginate can be hardened simply by a brief exposure to calcium chloride (Smrdel P, Bogataj M, Podlogar F, Planinsek O, Zajc N, Mazaj M, Kaucic V, Mrhar A (2006) Characterization of calcium alginate beads containing structurally similar drugs. Drug Dev Ind Pharm 32:623-633). This matrix has been used for embedding of bone-related SaOS-2 cells (SchloBmacher U, Schroder HC, Wang XH, Feng Q, Diehl-Seifert B, Neumann S, Trautwein A, Muller WEG (2013) Alginate/silica composite hydrogel as a potential morphogenetically active scaffold for three-dimensional tissue engineering. RSC Adv 3: 11185-11194; Muller WEG, Schroder HC, Feng QL, SchloBmacher U, Link T and Wang XH (2013) Development of a morphogenetically active scaffold for three-dimensional growth of bone cells: biosili- ca/alginate hydrogel for SaOS-2 Cell cultivation. J Tissue Engin Regener Med, doi: 10.1002/term. l745).

Summary of the invention

Sodium alginate hydrogel, stabilized with gelatin, is a biologically inert matrix that can be used for encapsulating and 3D bioprinting of bone-related SaOS-2 cells. However, the cells, embedded in this matrix, remain in a non-proliferating state.

The present invention is based on the finding of the present inventors that the addition of an agarose overlay onto the bioprinted alginate/gelatine/SaOS-2 cell scaffold, supplemented with the calcium salt of polyphosphate [polyP ^» Ca ²⁺ -complex], results in a marked increase in cell proliferation. In the presence of the polyP ^» Ca ²⁺ -complex (100 μΜ), the cells proliferate with a generation time of approximately 47 - 55 h. This effect was surprising and could not be expected from previous results showing that the polyP-Ca ² + -complex stimulates hydroxyap- atite/calcium phosphate deposition because the latter effect could be caused by the fact that the polyP-Ca ² + -complex (after enzymatic degradation, e.g. by alkaline phosphatase) is used as a source for calcium ions and phosphate ions needed for hydroxyapatite/calcium phosphate formation. It is also surprising that this effect was found despite the fact that the polyP-Ca ² + - complex is insoluble in water.

In addition, the inventors unexpectedly found that the hardness of the alginate/gelatin hydrogel substantially increases in the presence of the polymer. The reduced Young's modulus for the alginate/gelatin hydrogel is approximately 13 - 14 kPa, and this value drops to approximately 0.5 kPa after incubation of the cell containing scaffold for 5 d. In the presence of 100 μΜ polyP ^» Ca ²⁺ -complex, the reduced Young's modulus increases to about 22 kPa. The hardness of the polyP ^» Ca ²⁺ -complex containing hydrogel remains essentially constant if cells are absent in the matrix, but it drops to 3.2 kPa after a 5 d incubation period in the presence of SaOS-2 cells, indicating that polyP ^» Ca ²⁺ -complex becomes metabolized, degraded, by the cells. The initial increase in hardness is important for the stability of the implant. The later - - decrease in hardness is important because it facilitates the proliferation and migration of the incorporated cells, so that the initial implant can be replaced by the body's own bone mineral.

Another striking effect of the inventive technique, comprising the application of the pol- yP ^» Ca ²⁺ -complex in the agarose overlay onto the alginate/gelatin hydrogel matrix, is a significant increase in the mineralization of the cells, as demonstrated by staining with Alizarin Red. The morphology of the nodules, and their individual crystallites, formed on the surface of the cells embedded in the alginate/gelatin hydrogel does not significantly differ from the nodules on cells growing in monolayer cultures, as can be demonstrated by SEM analyses. The presence of calcium phosphate biomineral in the nodules can be confirmed by EDX two- dimensional element mapping analysis.

The inventive, newly developed technique, using cells encapsulated into an alginate/gelatin hydrogel and a secondary layer containing the morphogenetically active, growth promoting polymer polyP ^» Ca ²⁺ -complex opens new possibilities for the application of 3D bioprinting in bone tissue engineering.

Based on previous data, alginate is a candidate polymer for 3D bioprinting. For the described experiments, the inventors used the cell printer "Bioplotter" from Envisiontec that is considered to be suitable for rapid prototyping / 3D cell bioprinting (Wiist S, Muller R, Hofmann S (2011) Controlled positioning of cells in biomaterials - Approaches towards 3D tissue printing. J Funct Biomater 2: 119-154). The inventors demonstrated that SaOS-2 cells survive in the alginate matrix after printing; however, they lose their proliferating capacity.

The inventors added gelatin as first extracellular structural filamentous polymer to the hydrogel to stabilize the alginate hydrogel and to avoid mechanical shear forces. The alginate could be hardened by cross-linked with calcium chloride. The inventors showed that addition of gelatin to the alginate used in the present study as a hydrogel for embedding the cells provides this matrix with a significant higher stability.

However, after printing, the SaOS-2 cells, embedded in the alginate/gelatin hydrogel network, do not proliferate: The steady state number of cells residing in the alginate/gelatin hydrogel remains constant during incubation of the scaffolds, as revealed by the MTT assay.

The novel aspect of the inventive technique is the unexpected discovery that the cells, after addition of polyP ^» Ca ²⁺ -complex in an overlay, added as a secondary layer onto the bioprinted alginate/gelatine/cell scaffolds, respond with an unprecedented intense cell proliferation. As a result, a significant increase in number of cells occurs. In the absence of this polymer no steady-state change of the SaOS-2 cells is seen, while in the presence of polyP ^» Ca ²⁺ -complex, the cells proliferate with an average generation time of approximately 47 - 55 h during a 6 d incubation period.

The agarose, supplemented with polyP, was layered into the open interspace between the vascular channels. For these studies we used polyP together with CaCl ₂ [polyP ^» Ca ²⁺ -complex] in order to avoid any depletion of Ca ²⁺ ions required for mineral deposition onto SaOS-2. A schematic outline of the novel 3D bioprinting method according to the invention is given in Figure 1.

The reduced Young's modulus [RedYM], measured with an indenter device and using a cantilever on the top of a glass ferrule, for the alginate hydrogel in the absence of gelatin is 6.3 kPa, while together with gelatin the hardness, measured as RedYM, increases to ~ 12 kPa. - -

After an incubation period of 2 d, the RedYM drops to approximately 2.5 kPa, and after a 5 d incubation period, to values below 0.5 kPa.

Addition of 100 μΜ polyP ^» Ca ²⁺ -complex to the alginate/gelatin hydrogel additionally stabilizes the matrix which becomes substantially harder. The RedYM reaches values of around 22 kPa. However, after a 5 d incubation period the RedYM values for the gels containing SaOS- 2 cells and exposed to the polyP ^» Ca ²⁺ -complex drop down to 3.2 kPa. If the cells are absent from the hydrogel containing polyP ^» Ca ²⁺ -complex the hardness of the matrix remains high. This result indicates that polyP ^» Ca ²⁺ -complex undergoes metabolic degradation that is caused by the cells.

A further striking characteristics of the bioprinting method according to the invention, the alginate/gelatin hydrogel enriched with polyP ^» Ca ²⁺ -complex in the overlay, is its ability to enhance the mineralization of bone- forming cells, for example SaOS-2 cells. Addition of the osteogenic cocktail to SaOS-2 cells in the hydrogel matrix, growing either in the absence or presence of polyP ^» Ca ²⁺ -complex, causes as strong increase in mineralization, as visualized after staining with the fluorochrome Alizarin Red. SEM analyses revealed that the morphology of the nodules (diameters between 200 nm and 5 μιη) formed on the surface of SaOS-2 cells, embedded in the alginate/gelatin hydrogel, in the presence of the osteogenic cocktail does not differ from those formed onto cells growing as monolayers. EDX two-dimensional element mapping revealed within and in the closed neighborhood of the nodules strong signals for the elements Ca, P, and O (hydroxypapatite), but also C and to a smaller extent S (protein).

The chain lengths of the polyP molecules can be in the range 2 to 1000 phosphate units, preferentially in the range 4 to 100 phosphate units. Optimal results were achieved with polyP molecules with chain lengths of approximately 40 phosphate units.

A complex of polyP with divalent cations, preferably Ca ²⁺ ions (polyP ^» Ca ²⁺ -complex) is preferentially used to avoid a depletion of divalent cations, in particular calcium ions, by complex formation with polyP. Such a complex formation is expected to interfere with the results in vitro (cell culture experiments) but not or less with the effect in vivo.

The preparation of the polyP ^» Ca ²⁺ -complex is state-of-the-art and has previously been described by the inventors (Muller WEG, Wang XH, Diehl-Seifert B, Kropf K, SchloBmacher U, Lieberwirth I, Glasser G, Wiens M, Schroder HC. Inorganic polymeric phos- phate/polyphosphate as an inducer of alkaline phosphatase and a modulator of intracellular Ca ²⁺ level in osteoblasts (SaOS-2 cells) in vitro. Acta Biomater 7:2661-2671, 2011; Wang XH, Schroder HC, Diehl-Seifert B, Kropf K, SchloBmacher U, Wiens M, Muller WEG (2012) Dual effect of inorganic polymeric phosphate/polyphosphate on osteoblasts and osteoclasts in vitro. J Tissue Engin Regen Med 7:767-776; as well as patent application EP 11152208.2 Food supplement and injectable material for prophylaxis and therapy of osteoporosis and other bone disease. Inventors: Muller WEG, Wang X, Schroder HC).

A further aspect of this invention concerns a method for 3D bioprinting, based on encapsulating of the cells into an alginate/gelatin hydrogel, followed by an overlay comprising, instead of polyP or polyP ^» Ca ²⁺ -complex, another morphogenetically active polymer, such as polymeric silicic acid (silica) or one of its salts.

The counterion of the salt formed by polymeric silicic acid (silica) can be a monovalent cation such as lithium, sodium, or potassium. - -

The polymeric silicic acid can be formed by an enzyme or protein involved in bio silica (amorphous, hydrated silicon oxide) metabolism, such as silicatein or a silicatein fusion protein. The silicatein polypeptide or a silicatein fusion protein can be produced using a prokaryotic or eukaryotic expression system, or can be produced synthetically.

The silicatein or silicatein fusion protein can be present together with a suitable substrate (silica precursor) such as water glass, orthosilicic acid, orthosilicates, monoalkoxysilanetriols, dialkoxysilanediols, trialkoxysilanols, tetraalkoxysilanes, alkyl-silanetriols, alkyl-silanediols, alkyl-monoalkoxysilanediols, alkyl-monoalkoxysilanols, alkyl-dialkoxysilanols, or alkyl- trialkoxysilanes.

A further aspect of the invention concerns a method for 3D bioprinting, based on encapsulating of the cells into an alginate/gelatin hydrogel, followed by an overlay composed of agarose and a morphogenetically active polymer, wherein the morphogenetically active polymer is a combination of inorganic polyP or a complex of inorganic polyP and divalent metal ions with monomeric silicic acid (orthosilicic acid) or polymeric silicic acid (silica) or its salts.

A further aspect of the invention concerns a method for 3D bioprinting, based on encapsulating of the cells into an alginate/gelatin hydrogel, followed by an overlay consisting of agarose and inorganic polyP or a complex of inorganic polyP and divalent metal ions, combined with calcium carbonate or an enzyme forming calcium carbonate in the presence of calcium ions, such as carbonic anhydrase, as previously described by the inventors; e.g. Patent application GB1319416.2 (Modulator of bone mineralization based on a combination of polyphos- phate/carbonate and carbonic anhydrase activators; inventors: Miiller WEG, Schroder HC, Wang XH).

A further aspect of the invention concerns a material obtained by 3D bioprinting using one of the methods described above.

The invention will now be described further in the following preferred examples, nevertheless, without being limited thereto. For the purposes of the present invention, all references as cited herein are incorporated by reference in their entireties. In the Figures listing,

Figure 1 shows a sketch of the procedure of 3D cell printing of scaffolds subsequently covered with an agarose overlay using the 3D-Bioplotter from Envisiontec. The SaOS-2 cells are encapsulated into alginate/gelatin. This cell suspension is filled into a cartridge. Using a control element connected with the computer-guided printing apparatus the algi- nate/gelatin/SaOS-2 cells are passed through a needle into a CaCl ₂ bath. This scaffold is submersed into McCoy's medium/FCS and overlayed with an agarose layer containing pol- yP ^» Ca ²⁺ -complex as "differentiation medium".

Figure 2 shows the bioprinting of SaOS-2 cells in alginate/gelatin. The cells, at a density of 5x10 ⁵ per ml, were embedded into the alginate/gelatin hydrogel and then bioprinted. (A) The bioprinted stacks of 13 mm (diameter) to 1.5 mm (height) were bioprinted and then transferred into 12-well plates. (B) Subsequently they were overlayed with agarose. (C) The individual cells can be visualized by staining their nuclei with DRAQ5. (D to F) The meanderlike cylinders (cy) can be distinguished, like the interspace holes (h). In (F) the SaOS-2 cells (c) can be recognized on the basis of the bulges that protrude from the surface. (A and B: light microscopy; C: fluorescence microscopy; D to F: SEM). - -

Figure 3 shows the mechanical properties of the hydrogel, measured as reduced Young's modulus using a tissue nanoindenter. The bioprinted hydrogels were overlayed with agarose, containing either no polyP (minus polyP [open bars; hatched leftwards; hatched rightwards]) or were supplemented with 100 μΜ polyP ^» Ca ²⁺ -complex (plus polyP [filled bars; cross hatched white on black; cross hatched black on white]). Those two sets of bioprinted hydro- gels contained either no cells (open bars; filled bars), or 5xl0 ⁵ cells/ml (hatched leftwards; cross hatched white on black), or 2xl0 ⁶ cells/ml (hatched rightwards; cross hatched black on white). Data represent the means ± SD of ten independent experiments (* P < 0.01).

Figure 4 shows the growth of SaOS-2 cells in the alginate/gelatin hydrogel overlayed with agarose, containing no (minus polyP) or 100 μΜ polyP ^» Ca ²⁺ -complex (plus polyP), as measured by the colorimetric cell proliferation (MTT) assay. Following incubation with MTT, formazan crystals are formed and spectrophotometrically quantified. The increase in the OD reflects an increased overall activity of the mitochondrial dehydrogenases and directly correlates with the cell number. Data represent means ± SD of ten independent experiments (* P < 0.01).

Figure 5 shows the effect of polyP ^» Ca ²⁺ -complex on the cell density and, in parallel, effect of the osteogenic cocktail on the extent of mineralization. The cultures were incubated in the absence (minus polyP) (A, B, E and F) or presence of polyP ^» Ca ²⁺ -complex (plus polyP) (C, D, G and H) in the overlay for 3 d, followed by a period of 5 d in the absence (-OC; A, E, C and G) or presence of the osteogenic cocktail (+OC; B, F, D and H). Then the cells on a cover glass were stained with Alizarin Red S.

Figure 6 shows the quantitative assessment of the extent of mineralization using Alizarin Red S as an indicator reagent. The cells were incubated for 3 d in the presence (plus polyP) or absence of polyP ^» Ca ²⁺ -complex (minus polyP) in the overlay. Then the cultures remained incubated without the osteogenic cocktail (oc) or were exposed to the cocktail. After a 1 , 5 or 7 d incubation period the cells were collected, extracted, and reacted with Alizarin Red S, followed by a spectrophotometric determination. The extent of biomineralization is correlated with the DNA content in the assays in order to normalize for the cell number. Values represent the means (± SD) from 10 separate experiments each (* P < 0.01).

Figure 7 shows the fine structure of the mineral deposits on SaOS-2 cell surfaces; SEM images. (A to C) Mineral deposits, nodules (no), formed onto cells (c), growing as monolayer for 5 d in medium/serum, supplemented with osteogenic cocktail. The deposits can be readily distinguished, based on their characteristic structure. The individual crystallites appear as sharp-bordered platelets. (D to F) Nodules (no) formed onto SaOS-2 cells that had been encapsulated into alginate/gelatin hydrogel and then overlayed with 100 μΜ polyP ^» Ca ²⁺ - complex. The incubation in the presence of the cocktail was terminated after 5 d.

Figure 8 shows the element mapping of the surface of SaOS-2 cells comprising mineral nodules (no) within the cell layer. The analysis was performed by SEM, coupled to an EDX detector. (A) Secondary electron image showing the area that has been mapped. Element mappings for Ca (B), O (C), C (D), P (E) and S (F). The regions of brighter pseudocolor represent larger accumulations of the respective elements. (G and H) Individual EDX analysis of a nodule (no), using EDX spectroscopy. Some peaks, corresponding to an individual element are marked.

Examples - -

In the following examples, only the inventive 3D bioprinting technique is described, using the polyP ^» Ca ²⁺ -complex (with a preferred polyP chain length of 40 phosphate units and a preferred concentration of 100 μΜ) in the agarose overlay. Similar results can be obtained in the absence of Ca ²⁺ or in the presence of lower and higher concentrations of polyP or polyP'Ca ² - complex with lower and higher chain lengths.

Methods

Cells and their incubation conditions

Human osteogenic sarcoma cells, SaOS-2 can be used for the experiments. They can be cultivated in McCoy's medium containing 2 mM L-glutamine, and gentamycin (50 mg/ml), supplemented with 5% heat-inactivated FCS in 25 cm ² flasks in a humidified incubator at 37°C and 5% C0 ₂ . The cells are seeded at a starting concentration of 10 ⁵ cells/ml in 25 cm ² flasks. Culture medium/FCS is changed every 3 d.

The cells, growing in McCoy's medium/FCS, are exposed to the osteogenic cocktail (10 nM dexamethasone, 50 mM ascorbic acid, and 5 mM sodium β-glycerophosphate). Routinely, the mineralization activation osteogenic cocktail is added 3 d after starting the experiments. During each medium change new osteogenic cocktail is added.

The scaffolds prepared form alginate/gelatin/SaOS-2 cell are incubated first for 3 d in McCoy's medium/FCS and subsequently for 5 d and in one series of experiment for 7 d in McCoy's medium/FCS, supplemented with osteogenic cocktail.

Polyphosphate

Sodium polyphosphate (Na-polyP with an average chain of approximately 40 phosphate units) was used in the experiments described under "Examples". To compensate for any chelating effect, caused by polyP the polymer can be mixed together with CaCl ₂ in a stoichiometric ratio of 2: 1 (polyP:CaCl ₂ ); the salt is designated as "polyP ^» Ca ²⁺ -complex". In the described experiments, a concentration of 100 μΜ (14 μg/ml polyP ^» Ca ²⁺ -complex) is adjusted in the assays.

Biosilica

The preparation of the recombinant silicatein and silicatein fusion proteins, e.g (Glu)-tagged silicatein-a, is state-of-the-art; see, for example, European Patent No EP 1320624 and United States Patent No US 7,169,589 (Silicatein-mediated synthesis of amorphous silicates and si- loxanes and use thereof; inventors: Krasko A, Lorenz A, Miiller WEG, Schroder HC) and GB 1405994.3 (Osteogenic material to be used for treatment of bone defects; inventors Miiller WEG, Schroder HC, Wang XH). It is preferentially performed in E. coli, but the expression of the recombinant protein in yeast and mammalian cells is also possible.

Biosilica can be prepared using the recombinant silicatein or silicatein fusion protein following previously described methods, for example: Schroder HC, Wang XH, Manfrin A, Yu SH, Grebenjuk VA, Korzhev M, Wiens M, SchloBmacher U, Miiller WEG (2012) Silicatein: acquisition of structure-guiding and structure-forming properties during maturation from the pro-silicatein to the silicatein form. J Biol Chem 287:22196-22205; or: Wang XH, SchloBmacher U, Schroder HC, Miiller WEG (2013) Bio logically- induced transition of biosilica sol to mesoscopic gelatinous floes: a biomimetic approach to a controlled fabrication of bio-silica structures. Soft Matter 9:654-664). - -

In order to increase the yield of biosilica, the silicatein-mediated enzymatic reaction can be performed in the presence of recombinant silintaphin-1 (SchloBmacher U, Wiens M, Schroder HC, Wang XH, Jochum KP, Muller WEG (2011) Silintaphin-1 : Interaction with silicatein during structure-guiding biosilica formation. FEBS J 278: 1145-1155). In addition, the formed biosilica material can be hardened in the presence of poly(ethylene glycol) [PEG] (Schroder HC, Wang XH, Manfrin A, Yu SH, Grebenjuk VA, Korzhev M, Wiens M, SchloBmacher U, Muller WEG (2012) Silicatein: acquisition of structure-guiding and structure-forming properties during maturation from the pro-silicatein to the silicatein form. J Biol Chem 287:22196- 22205).

Alternatively, the biosilica can be produced during the incubation of the bioprinted scaffolds; in that case, 200 μΜ prehydrolyzed TEOS (tetraethyl orthosilicate) and 20 μg/mL of recombinant silicatein (final concentrations) are added to the agarose overlay. Prehydrolyzed TEOS 200 μΜ is also added to the incubation medium and is present throughout the experiment. Lower or higher concentrations of TEOS or another silicic acid precursor, and lower or higher concentrations of the recombinant silicatein or silicatein fusion protein can also be used.

As a final alternative, also ortho-silicate, preferably at a concentration of between 100 μΜ and 1 mM, but also lower and higher concentrations can be applied instead of polymeric biosilica.

Hydrogel preparation

Sodium alginate powder can be exposed/sterilized by ultraviolet light for overnight. Then a solution of 50 mg/ml sodium alginate is prepared in physiological saline which is supplemented with 50 mg/ml of low-melting gelatin. Those concentrations of alginate and gelatin give a suitable viscosity for 3D printing, but also lower and higher concentrations can be used. For a better visualization during the printing process the solution can be supplemented with 10 μΐ/ml of a phenol red [0.5%] solution. SaOS-2 cells being close to the end of the logarithmically growing phase are centrifuged (2,000xg; 3 min; 37°C); the sell sediment is added to the alginate/gelatin solution to give a final concentration of, for example, 5x10 ⁵ cell/ml. After freeing from air bubbles by a short application of vacuum the alginate/gelatin/SaOS-2 cell suspension is used for 3D cell printing.

3D-cell printing

The freshly prepared alginate/gelatin/SaOS-2 cell suspension is prewarmed to 37°C and filled air bubble-free into sterile 30 ml printing cartridges (e.g., Nordson EFD, Pforzheim; Germany), connected with, for example, a 1.98 x 0.41 mm steel needles (e.g., Nordson EFD) and placed into the preheated (30°C) printing head of, for example, a 3D-Bioplotter™ (4 ^th generation blotter; Envisiontec).

As an example, using the settings of 0.9 bar pressure, a speed of 26 mm/sec and 30°C the round gel cylinders are printed to a scaffold, measuring 13 mm in diameter and 1.5 mm in height. The cylinders are injected into 94 mm Petri dishes (e.g., Greiner Bio-One), filled with 0.4% CaCl ₂ as cross-linking solution, as described (SchloBmacher U, Schroder HC, Wang XH, Feng Q, Diehl-Seifert B, Neumann S, Trautwein A, Muller WEG (2013) Alginate/silica composite hydrogel as a potential morphogenetically active scaffold for three-dimensional tissue engineering. RSC Adv 3: 11185-11194). Sterile 3 mm thick chromatography paper (GEHealthcare) can be used as a support for the printing of the scaffold. The cylinders, diameter of 400 μιη, can be arranged in a meander-like pattern by changing the directions of the consecutive layers in a rectangular way. The dimensions of the stacks are predetermined, e.g., using the computer program Bioplotter RP 2.9 CAD software (Envisiontec). Using the same software, the cylinders can be sliced to individual layers corresponding to the diameter of the - - printing needle and subsequently transferred to the VisualMachines 3.0.193 printer software (Envisiontec). The complete printing process for one scaffold lasts approximately 1 min. The scaffolds remain for one more min in 0.4% CaCl ₂ solution to allow completion of the cross- linking process. Then the stacks/scaffolds are transferred with a spatula into 12-well Multi- well Plates (Greiner) and submersed in 1 ml of McCoy's medium/FCS and incubated at 37°C.

In order to accelerate the proliferation rate the 3D printed scaffolds are overlayed with 500 μΐ of 0.3% agarose, dissolved in McCoy's medium/FCS and supplemented which 100 μΜ pol- yP ^» Ca ²⁺ -complex. Prior to the addition to the cells the agarose is prewarmed to 37°C; after overlaying onto the cylinders, containing the cells the cultures are placed for 10 min at 4°C in order to reach a gelatinous consistence. A schematic outline is given in Figure 1.

Determination of the mechanical properties of the alginate/gelatin/SaOS-2 cell scaffold

The mechanical properties of the alginate scaffolds can be determined, for example, by a nanoindenter, using a cantilever that had been fabricated on the top of a ferruled optical fiber (Chavan D, Andres D, Iannuzzi D (2011) Note: ferrule-top atomic force microscope. II. Imaging in tapping mode and at low temperature. Rev Sci Instrum 82:046107, doi: 10.1063/1.3579496; Chavan D, van de Watering TC, Gruca G, Rector JH, Heeck K, Slaman M, Iannuzzi D (2012) Ferrule-top nanoindenter: an optomechanical fiber sensor for nanoindentation. Rev Sci Instrum 83: 115110). In the experiments shown, a ferrule-top nanoindenter setup together with the PIUMA controller/drive have been used (Optics 11, Amsterdam). For each measurement 10 single indents are performed on the same spot. The indents are depth controlled (10 μιη) and the loading and unloading period is set to 2 s. Prior to the loading and after the unloading step the data are collected. Based on the load-displacement curves the reduced Young's modulus [RedYM] can be calculated (Fischer-Cripps AC (2011) Nanoindentation testing. In: Nanoindentation, Mechanical Engineering Series 1. Berlin: Springer Science+Business Media, pp 21-37).

Cell staining

In order to visualize the cells in the hydrogel the nuclei can be stained, for example, with DRAQ5. The fluorescence of the cells is excited at 635 nm, and the emission is recorded at 705 nm in an Olympus 1X71 fluorescence microscope.

Cell proliferation assay

The growth of SaOS-2 cells is determined in the hydrogels after different incubation periods; 3 d in McCoy's medium/FCS or 3 d in McCoy's medium/FCS and additional 5 d in the presence of osteogenic cocktail. For this series a cell density of 5xl0 ⁵ cells/ml is chosen. After termination of the experiments the cells in the different scaffolds (either in the absence or in the presence of 100 μΜ polyP ^» Ca ²⁺ -complex) are released by treating the samples with 55 mM Na-citrate, as described (SchloBmacher U, Schroder HC, Wang XH, Feng Q, Diehl- Seifert B, Neumann S, Trautwein A, Muller WEG (2013) Alginate/silica composite hydrogel as a potential morphogenetically active scaffold for three-dimensional tissue engineering. RSC Adv 3: 11185-11194). Then the cell suspension is incubated with fresh medium containing 100 μΐ of MTT for 12 h in the dark. Subsequently the remaining MTT dye is removed and 200 μΐ of dimethylsulfoxide [DMSO] are added to solubilize the formazan crystals. Finally the optical densities (OD) at 570 nm are measured using the ELISA reader/ spectrophotometer.

Mineralization by SaOS-2 cells in vitro - -

The mineral deposits onto the cells can be visualized by Alizarin Red S staining. The SaOS-2 cells are incubated onto plastic cover slips (Nunc) that are placed into 24-well plates (Greiner) and incubated in McCoy's medium/FCS and antibiotics for 3 day. Subsequently the medium/serum is removed and fresh McCoy's medium FCS is added which is supplemented with the osteogenic cocktail. As indicated, polyP ^» Ca ²⁺ -complex is added at a concentration of 100 μΜ to the assays. After an incubation period of up to seven d the slides are removed and stained with 10% Alizarin Red S (staining for mineral deposition). In parallel, the assays are quantitatively assessed, also using the fluorochrome Alizarin Red S as an indicator, and applying a spectrophotometric assay (Wiens M, Wang XH, Schroder HC, Kolb U, SchloBmacher U, Ushijima H, Miiller WEG (2010) The role of biosilica in the osteopro- tegerin/RANKL ratio in human osteoblast like cells. Biomaterials 31 :7716-7725). The amount of bound Alizarin Red S is given in μιηοΐεβ. Values are normalized to total DNA in the samples. The DNA content can be determined by using the PicoGreen method (Schroder HC, Borejko A, Krasko A, Reiber A, Schwertner H, Miiller WEG (2005) Mineralization of SaOS- 2 cells on enzymatically (silicatein) modified bioactive osteoblast-stimulating surfaces. J Bi- omed Mat Res Part B - Appl Biomater 75B:387-392), with calf thymus DNA as a standard.

Microscopic analyses

Scanning electron microscopy (SEM) can be performed, for example, with a HITACHI SU 8000 (Hitachi High-Technologies Europe GmbH), equipped with at low voltage (<1 kV; analysis of near-surface organic surfaces) detector. Digital light microscopic studies can be performed, for example, using a VHX-600 Digital Microscope (Keyence) equipped with a VH-Z25 zoom lens.

Energy-dispersive X-ray spectroscopy

For example, the SEM (HITACHI SU 8000) coupled to an XFlash 5010 detector - an X-ray detector that allows simultaneous energy-dispersive X-ray (EDX)-based elemental analyses, can be used. This can be coupled at higher voltage (4 kV) to the XFlash 5010 detector that is used for element mapping of the surfaces of SaOS-2 cells. HyperMap databases are collected, as described (Salge T, Terborg R (2009). EDS microanalysis with the silicon drift detector (CDD): innovative analysis options for mineralogical and material science application. Anadolu Univ J Sci Technol 10:45-55).

Statistical analysis

The results are statistically evaluated using paired Student's t-test. The generation time of SaOS-2 cells is calculated according to (Powell EO (1956) Growth rate and generation time of bacteria, with special reference to continuous culture. J Gen Microbiol 15:492-511).

Encapsulation of cells into alginate/gelatin and overlay

As outlined under "Methods" the SaOS-2 cells were suspended into an alginate/gelatin hy- drogel and then bioprinted into stacks of 13 mm in diameter and 1.5 mm in height (Figure 2A). In the standard bioprinting assay 5xl0 ⁵ cells/ml were added. Subsequently, the hydrogels were overlayed with agarose (Figure 2B) and then incubated. The cells within the gels could be visualized by staining their nuclei with DRAQ5 (Figure 2C).

The individual cylinders, 200 - 400 μιη thick, containing the cells, can be visualized by SEM analysis (Figure 2D and E). In this network of cylinders the interspace can be seen that become filled by the agarose overlay. The embedded SaOS-2 cells can be recognized on the bulges of the surfaces of the cells (Figure 2F). - -

Mechanical properties, reduced Young's modulus of the scaffold

The hardness of the alginate/gelatin/SaOS-2 cell scaffold was measured with an indenter device and using a cantilever on the top of a glass ferrule. The scaffolds were printed with algi- nate/gelatin and overlayed with agarose. The agarose remained either free of polyP ^» Ca ²⁺ - complex, or was supplemented with 100 μΜ of this polymer. In all assays lacking pol- yP ^» Ca ²⁺ -complex the reduced Young's modulus [RedYM] was found to lower than in those hydrogels that contain that polymer (Figure 3). At time zero in those assays without pol- yP ^» Ca ²⁺ -complex, the RedYM varies around 13 - 14 kPa, irrespectively on the presence of cells. After an incubation period of 2 d the values for all three experiments with algi- nate/gelatin, containing no cells, 5xl0 ⁵ or 2xl0 ⁶ cells/ml, the RedYM drops to approximately 2.5 kPa, without showing any significant differences within this group. After the longer, 5 d incubation period, the gels without cells decrease to values below 0.8 kPa.

If 100 μΜ polyP ^» Ca ²⁺ -complex is added to the hydrogels they become substantially harder. The RedYM increase for the hydrogels reaches values of around 22 kPa (Figure 3); no significant differences are seen between the three experimental groups at day 0. However, after a 1 d incubation period the RedYM values for the gels without cells did not change significantly (from 21.7±3.8 to 18.3 ± 3.4), while those values for the gels containing 5xl0 ⁵ or 2xl0 ⁶ cells/ml drop down to 4.8±0.53 kPa and 3.2±0.41 kPa. After the 5 d incubation period the values of RedYM of gels, not containing cells, remain still high with 12.8±1.8 kPa, while the one for the gels with cells, containing 5xl0 ⁵ or 2xl0 ⁶ cells/ml decrease to 2.4±0.34 kPa and 1.2±0.31 kPa, respectively.

Effect of polyP on growth of SaOS-2 cells in the alginate/gelatin hydrogel

After bioprinting the cylinders, composed of alginate/gelatin into which the cells were encapsulated, the scaffold was overlayed with agarose in the absence or presence of polyP ^» Ca ²⁺ - complex. If the cell density/viability was assessed immediately after measuring the mitochondrial dehydrogenases with MTT no difference in the OD is seen between the assays without (0.42±0.07) or with 100 μΜ polyP (0.49±0.09); Figure 4. Those values remain unchanged in the assays without polyP even after a total incubation period of 6 d in McCoy's medium/FCS. At day 3 the medium/serum was replaced. However, if 100 μΜ polyP is added a significant increase in the OD values from 0.49±0.09 (time 0), to 1.42±0.19 (3 d of incubation) and 2.98±0.41 (6 d) occurs. Based on these values, the number of doublings during the 3 d incubation period amounts to 1.53, and during the 6 d incubation it amounts to 2.61. From these numbers a generation time can be calculated to be approximately 47 - 55 h.

The increase in cell density in the alginate/gelatin hydrogel after addition of 100 μΜ pol- yP ^» Ca ²⁺ -complex into the overlay can also be followed light microscopically. After an initial 3 d incubation period and in the absence of polyP ^» Ca ²⁺ -complex in the overlay and then a further incubation for 5 d in the absence or presence of the osteogenic cocktail, the cell density is low (Figure 5A and B). However, if the cells were incubated in the presence of 100 μΜ polyP ^» Ca ²⁺ -complex during the total incubation period the number of cells considerably increases, irrespectively of the absence or presence of the osteogenic cocktail (Figure 5C and D).

Effect of biosilica on growth of SaOS-2 cells in the alginate/gelatin hydrogel

The experiments were performed as described before ("Effect of polyP on growth of SaOS-2 cells in the alginate/gelatin hydrogel") but in the presence of biosilica instead of polyP ^» Ca ²⁺ - complex. Again, no difference in the OD was seen between the assays without (0.42±0.07) or with 100 μΜ biosilica (0.46±0.1), if the cell density/viability was assessed immediately after measuring the mitochondrial dehydrogenases with MTT. The values remain unchanged in the - - assays without biosilica after an incubation period of 6 d. However, if 100 μΜ biosilica was added, the OD values significantly increased from 0.44±0.08 (time 0), to 1.15±0.24 (3 d of incubation) and 1.87±0.30 (6 d).

Response of the cells to the osteogenic cocktail

Addition of the osteogenic cocktail to cultures, growing either in the absence or presence of polyP ^» Ca ²⁺ -complex, causes as strong increase of the mineralization, as microscopically visualized after staining with Alizarin Red S (Figure 5F and H). In the absence of the cocktail no bright red staining is seen (Figure 5E and G).

A quantitative assessment of the mineralization was achieved by a spectrophotometric determination, again by using the fluorochrome Alizarin Red S as a reagent, after a pre-incubation period for 3 d, followed by a 1 , 5 and 7 d incubation period in the absence or presence of pol- yP ^» Ca ²⁺ -complex. The cultures either remained without the osteogenic cocktail or were exposed to this cocktail (Figure 6). After a one d incubation period, the cells did not show any significant staining for Alizarin Red S after, irrespectively on the presence of polyP ^» Ca ²⁺ - complex or of the osteogenic cocktail. However, after a 5 d incubation with the cocktail the cultures showed a strong and significant increase of the red color, if compared to the reaction seen in the absence of the cocktail. After a 7 d incubation period those cells in the assays that were exposed to the osteogenic cocktail and polyP ^» Ca ²⁺ -complex reacted strongly with Alizarin Red S; the value measured was 0.95±0.13 nmoles per μg of DNA. This extent is significantly higher than the color reaction seen for the assays of cells not exposed to polyP ^» Ca ²⁺ - complex, but treated with the osteogenic cocktail (Figure 6).

Addition of the osteogenic cocktail to cultures also caused as strong increase in mineralization of cells, growing in the presence of biosilica, compared to controls without this polymer.

The spectrophotometric determination, using the fluorochrome Alizarin Red S, revealed, after a 7 d incubation period of cells exposed to the osteogenic cocktail and biosilica, a value of 0.81 ±0.20 nmoles per μg of DNA, which is significantly higher than the color reaction in controls (assays of cells not exposed to biosilica, but treated with the osteogenic cocktail; see Figure 6).

In the presence of ortho-silicate, instead of polymeric silica, at a concentration of 100 μΜ, a value of 0.72±0.15 nmoles per μg of DNA has been determined (7 day incubation period).

Mineral crystallite formation

In order to verify that the Alizarin Red S-positive staining is due to mineral deposition, SEM analyses were performed. In the absence of the osteogenic cocktail no crystallites could be identified onto the SaOS-2 cells during the 10 days incubation period (not shown here). If the cocktail is added to the cells, nodules started to grow after an incubation of 3 d. SEM images from crystallites, onto the SaOS-2 cells, after an incubation for 5 d in the presence of the cocktail are seen (Figure 7). The diameters of the nodules, formed on SaOS-2 cells that grew under monolayer conditions and not in a hydrogel vary between 200 nm and 5 μιη (Figure 7 A and B). At higher magnification the individual crystallites with dimensions of approximately 10-20 nm can be distinguished (Figure 7C).

The morphology of the nodules on SaOS-2 cells, and their individual crystallites forming them, and which are growing in the alginate/gelatin hydrogel (Figure 7D and E) are not to distinguish from those formed onto cells growing as monolayers. The size is likewise in the - - range of approximately 2-5 nm. The edges of the crystallites are somehow less sharp (Figure 7F).

Element analysis of the nodules formed

The element distribution of a surface of SaOS-2 cells, growing in hydrogel, was performed by EDX two-dimensional element mapping (Figure 8A to F). Cells growing in the hydrogel which was overlayed with agarose, containing 100 μΜ polyP ^» Ca ²⁺ -complex, were used for the analysis. The cells were exposed for the last 5 d prior to harvesting with the osteogenic cocktail. In Figure 8A a secondary electron image of a series of nodules formed onto the cells are visualized. The mapping data, recorded from the same area, show that within and in the close neighborhood of the nodules, strong signals for the bone-prevalent elements Ca (Figure 8B), O (Figure 8C), C (Figure 8D), P (Figure 8E) and then to a smaller extent also for S (Figure 8F) could be collected.

Using EDX - spot - characterization reveals that strong signals for the characteristic bone- related elements P and C highlight (Figure 8H) if a crystallite deposit (Figure 8G) is analyzed.