Refine
Has Fulltext
- yes (8)
Is part of the Bibliography
- yes (8)
Document Type
- Doctoral Thesis (8)
Keywords
- Cement (2)
- Struvit (2)
- Zement (2)
- Adipose Tissue (1)
- Baghdadite (1)
- Biomaterial (1)
- Bone-replacement (1)
- Brustkrebs (1)
- Calcium Phosphate (1)
- Calciumphosphat (1)
Institute
- Abteilung für Funktionswerkstoffe der Medizin und der Zahnheilkunde (8) (remove)
The implantation of any foreign material into the body automatically starts an immune reaction that serves as the first, mandatory step to regenerate tissue. The course of this initial immune reaction decides on the fate of the implant: either the biomaterial will be integrated into the host tissue to subsequently fulfill its intended function (e.g., tissue regeneration), or it will be repelled by fibrous encapsulation that determines the implant failure. Especially neutrophils and macrophages play major roles during this inflammatory response and hence mainly decide on the biomaterial's fate. For clinically relevant tissue engineering approaches, biomaterials may be designed in shape and morphology as well as in their surface functionality to improve the healing outcome, but also to trigger stem cell responses during the subsequent tissue regeneration phase.
The main focus of this thesis was to unravel the influence of scaffold characteristics, including scaffold morphology and surface functionality, on primary human innate immune cells (neutrophils and macrophages) and human mesenchymal stromal cells (hMSCs) to assess their in vitro immune response and tissue regeneration capacity, respectively. The fiber-based constructs were produced either via melt electrowriting (MEW), when the precise control over scaffold morphology was required, or via solution electrospinning (ES), when the scaffold design could be neglected. All the fiber-based scaffolds used throughout this thesis were composed of the polymer poly(ε caprolactone) (PCL).
A novel strategy to model and alleviate the first direct cell contact of the immune system with a peptide-bioactived fibrous material was presented in chapter 3 by treating the material with human neutrophil elastase (HNE) to imitate the neutrophil attack. The main focus of this study was put on the effect of HNE towards an RGDS-based peptide that was immobilized on the surface of a fibrous material to improve subsequent L929 cell adhesion. The elastase efficiently degraded the peptide-functionality, as evidenced by a decreased L929 cell adhesion, since the peptide integrated a specific HNE-cleavage site (AAPV-motif). A sacrificial hydrogel coating based on primary oxidized hyaluronic acid (proxHA), which dissolved within a few days after the neutrophil attack, provided an optimal protection of the peptide-bioactivated fibrous mesh, i.e, the hydrogel alleviated the neutrophil attack and largely ensured the biomaterial's integrity. Thus, according to these results, a means to protect the biomaterial is required to overcome the neutrophil attack.
Chapter 4 was based on the advancement of melt electrowriting (MEW) to improve the printing resolution of MEW scaffolds in terms of minimal inter-fiber distances and a concomitant high stacking precision. Initially, to gain a better MEW understanding, the influence of several parameters, including spinneret diameter, applied pressure, and collector velocity on mechanical properties, crystallinity, fiber diameter and fiber surface morphology was analyzed. Afterward, innovative MEW designs (e.g., box-, triangle-, round , and wall-shaped scaffolds) have been established by pushing the printing parameters to their physical limits. Further, the inter-fiber distance within a standardized box-structured scaffold was successfully reduced to 40 µm, while simultaneously a high stacking precision was maintained. In collaboration with a co-worker of my department (Tina Tylek, who performed all cell-based experiments in this study), these novel MEW scaffolds have been proven to facilitate human monocyte-derived macrophage polarization towards the regenerative M2 type in an elongation-driven manner with a more pronounced effect with decreasing pore sizes.
Finally, a pro-adipogenic platform for hMSCs was developed in chapter 5 using MEW scaffolds with immobilized, complex ECM proteins (e.g., human decellularized adipose tissue (DAT), laminin (LN), and fibronectin (FN)) to test for the adipogenic differentiation potential in vitro. Within this thesis, a special short-term adipogenic induction regime enabled to more thoroughly assess the intrinsic pro-adipogenic capacity of the composite biomaterials and prevented any possible masking by the commonly used long-term application of adipogenic differentiation reagents. The scaffolds with incorporated DAT consistently showed the highest adipogenic outcome and hence provided an adipo-inductive microenvironment for hMSCs, which holds great promise for applications in soft tissue regeneration.
Future studies should combine all three addressed projects in a more in vivo-related manner, comprising a co-cultivation setup of neutrophils, macrophages, and MSCs. The MEW-scaffold, particularly due to its ability to combine surface functionality and adjustable morphology, has been proven to be a successful approach for wound healing and paves the way for subsequent tissue regeneration.
Adipose tissue defects and related pathologies still represent major challenges in reconstructive surgery. Based on to the paradigm ‘replace with alike’, adipose tissue is considered the ideal substitute material for damaged soft tissue [1-3]. Yet the transfer of autologous fat, particularly larger volumes, is confined by deficient and unpredictable long term results, as well as considerable operative morbidity at the donor and recipient site [4-6], calling for innovative treatment options to improve patient care.
With the aim to achieve complete regeneration of soft tissue defects, adipose tissue engineering holds great promise to provide functional, biologically active adipose tissue equivalents. Here, especially long-term maintenance of volume and shape, as well as sufficient vascularization of engineered adipose tissue represent critical and unresolved challenges [7-9]. For adipose tissue engineering approaches to be successful, it is thus essential to generate constructs that retain their initial volume in vivo, as well as to ensure their rapid vascularization to support cell survival and differentiation for full tissue regeneration [9,10]. Therefore, it was the ultimate goal of this thesis to develop volume-stable 3D adipose tissue constructs and to identify applicable strategies for sufficient vascularization of engineered constructs. The feasibility of the investigated approaches was verified by translation from in vitro to in vivo as a critical step for the advancement of potential regenerative therapies.
For the development of volume-stable constructs, the combination of two biomaterials with complementary properties was successfully implemented. In contrast to previous approaches in the field using mainly non-degradable solid structures for mechanical protection of developing adipose tissue [11-13], the combination of a cell-instructive hydrogel component with a biodegradable porous support structure of adequate texture was shown advantageous for the generation of volume-stable adipose tissue. Specifically, stable fibrin hydrogels previously developed in our group [14] served as cell carrier and supported the adipogenic development of adipose-derived stem cells (ASCs) as reflected by lipid accumulation and leptin secretion. Stable fibrin gels were thereby shown to be equally supportive of adipogenesis compared to commercial TissuCol hydrogels in vitro. Using ASCs as a safe source of autologous cells [15,16] added substantial practicability to the approach. To enhance the mechanical strength of the engineered constructs, porous biodegradable poly(ε caprolactone)-based polyurethane (PU) scaffolds were introduced as support structures and shown to exhibit adequately sized pores to host adipocytes as well as interconnectivity to allow coherent tissue formation and vascularization. Low wettability and impaired cell attachment indicated that PU scaffolds alone were insufficient in retaining cells within the pores, yet cytocompatibility and differentiation of ASCs were adequately demonstrated, rendering the PU scaffolds suitable as support structures for the generation of stable fibrin/PU composite constructs (Chapter 3).
Volume-stable adipose tissue constructs were generated by seeding the pre-established stable fibrin/PU composites with ASCs. Investigation of size and weight in vitro revealed that composite constructs featured enhanced stability relative to stable fibrin gels alone. Comparing stable fibrin gels and TissuCol as hydrogel components, it was found that TissuCol gels were less resilient to degradation and contraction. Composite constructs were fully characterized, showing good cell viability of ASCs and strong adipogenic development as indicated by functional analysis via histological Oil Red O staining of lipid vacuoles, qRT-PCR analysis of prominent adipogenic markers (PPARγ, C/EBPα, GLUT4, aP2) and quantification of leptin secretion. In a pilot study in vivo, investigating the suitability of the constructs for transplantation, stable fibrin/PU composites provided with a vascular pedicle gave rise to areas of well-vascularized adipose tissue, contrasted by insufficient capillary formation and adipogenesis in constructs implanted without pedicle. The biomaterial combination of stable fibrin gels and porous biodegradable PU scaffolds was thereby shown highly suitable for the generation of volume-stable adipose tissue constructs in vivo, and in addition, the effectiveness of immediate vascularization upon implantation to support adipose tissue formation was demonstrated (Chapter 4).
Further pursuing the objective to investigate adequate vascularization strategies for engineered adipose tissue, hypoxic preconditioning was conducted as a possible approach for in vitro prevascularization. In 2D culture experiments, analysis on the cellular level illustrated that the adipogenic potential of ASCs was reduced under hypoxic conditions when applied in the differentiation phase, irrespective of the oxygen tension encountered by the cells during expansion. Hypoxic treatment of ASCs in 3D constructs prepared from stable fibrin gels similarly resulted in reduced adipogenesis, whereas endothelial CD31 expression as well as enhanced leptin and vascular endothelial growth factor (VEGF) secretion indicated that hypoxic treatment indeed resulted in a pro-angiogenic response of ASCs. Especially the observed profound regulation of leptin production by hypoxia and the dual role of leptin as adipokine and angiogenic modulator were considered an interesting connection advocating further study. Having confirmed the hypothesis that hypoxia may generate a pro-angiogenic milieu inside ASC-seeded constructs, faster vessel ingrowth and improved vascularization as well as an enhanced tolerance of hypoxia-treated ASCs towards ischemic conditions upon implanatation may be expected, but remain to be verified in rodent models in vivo (Chapter 5).
Having previously been utilized for bone and cartilage engineering [17-19], as well as for revascularization and wound healing applications [20-22], stromal-vascular fraction (SVF) cells were investigated as a novel cell source for adipose tissue engineering. Providing cells with adipogenic differentiation as well as vascularization potential, the SVF was applied with the specific aim to promote adipogenesis and vascularization in engineered constructs in vivo. With only basic in vitro investigations by Lin et al. addressing the SVF for adipose repair to date [23], the present work thoroughly investigated SVF cells for adipose tissue construct generation in vitro, and in particular, pioneered the application of these cells for adipose tissue engineering in vivo.
Initial in vitro experiments compared SVF- and ASC-seeded stable fibrin constructs in different medium compositions employing preadipocyte (PGM-2) and endothelial cell culture medium (EGM-2). It was found that a 1:1 mixture of PGM-2 and EGM-2, as previously established for co-culture models of adipogenesis [24], efficiently maintained cells with adipogenic and endothelial potential in SVF-seeded constructs in short and long-term culture setups. Observations on the cellular level were supported by analysis of mRNA expression of characteristic adipogenic and endothelial markers. In preparation of the evaluation of SVF-seeded constructs under in vivo conditions, a whole mount staining (WMS) method, facilitating the 3D visualization of adipocytes and blood vessels, was successfully established and optimized using native adipose tissue as template (Chapter 6).
In a subcutaneous nude mouse model, SVF cells were, for the first time in vivo, elucidated for their potential to support the functional assembly of vascularized adipose tissue. Investigating the effect of adipogenic precultivation of SVF-seeded stable fibrin constructs in vitro prior to implantation on the in vivo outcome, hormonal induction was shown beneficial in terms of adipocyte development, whereas a strong vascularization potential was observed when no adipogenic inducers were added. Via histological analysis, it was proven that the developed structures were of human origin and derived from the implanted cells. Applying SVF cells without precultivation in vitro but comparing two different fibrin carriers, namely stable fibrin and TissuCol gels, revealed that TissuCol profoundly supported adipose formation by SVF cells in vivo. This was contrasted by only minor SVF cell development and a strong reduction of cell numbers in stable fibrin gels implanted without precultivation. Histomorphometric analysis of adipocytes and capillary structures was conducted to verify the qualitative results, concluding that particularly SVF cells in TissuCol were highly suited for adipose regeneration in vivo. Employing the established WMS technique, the close interaction of mature adipocytes and blood vessels in TissuCol constructs was impressively shown and via species-specific human vimentin staining, the expected strong involvement of implanted SVF cells in the formation of coherent adipose tissue was confirmed (Chapter 7).
With the development of biodegradable volume-stable adipose tissue constructs, the application of ASCs and SVF cells as two promising cell sources for functional adipose regeneration, as well as the thorough evaluation of strategies for construct vascularization in vitro and in vivo, this thesis provides valuable solutions to current challenges in adipose tissue engineering. The presented findings further open up new perspectives for innovative treatments to cure soft tissue defects and serve as a basis for directed approaches towards the generation of clinically applicable soft tissue substitutes.
Tumorzellen, Stromazellen, Extrazellulärmatrix (EZM) und lösliche Faktoren in der Tumormikroumgebung beeinflussen und verstärken sich gegenseitig in der Ausbildung eines malignen Phänotyps. Sowohl die fibrotische EZM als auch eine kleine Subpopulation von pluripotenten Tumorstammzellen sind bekanntermaßen für die Steigerung der Tumoraggressivität verantwortlich. Inwiefern diese beiden unabhängigen Faktoren im Kontext von Brustkrebs miteinander in Beziehung stehen, ist jedoch bis heute unklar.
Um untersuchen zu können, welchen Beitrag Tumorzellen, Stromazellen, EZM und lösliche Faktoren einzeln und im Zusammenspiel zur Malignität eines Tumors leisten, ist die Entwicklung geeigneter in-vitro-Modelle unabdingbar. Daher war es das Ziel dieser Arbeit, ein 3D-Mikrotumormodell zu generieren, in dem eine Analyse dieser genannten Faktoren stattfinden könnte. An diesem Modell wurden darüber hinaus erste Untersuchungen von im Tumorkontext bekannten EZM-Proteinen durchgeführt. Um die dreidimensionale Anordnung von Tumorzellen und ihrer Gewebeumgebung adäquat wiedergeben zu können, beinhalteten die 3D-Tumorsphäroide sowohl Brustkrebszellen (MDA-MB-231) als auch Stromazellen (hASCs).
Die EZM als wichtiger Bestandteil der (Tumor-) Mikroumgebung sollte übersichtshalber durch Hämatoxylin-Eosin-Färbung und detaillierter durch immunhistochemische Analyse nach zwei verschiedenen Kulturzeitpunkten charakterisiert werden, um EZM-Veränderungen im zeitlichen Verlauf darzustellen. Im Fokus der Analyse standen die beiden wichtigsten profibrotischen EZM-Proteine Fibronektin und Kollagen I, die maßgeblich an der Pathogenese von Brustkrebs beteiligt sind. Zudem wurde das Vorkommen des Myofibroblastenmarkers α-SMA untersucht.
An den Sphäroiden einer Kontrollgruppe, die lediglich hASCs beinhaltete, sollte vergleichend eine Analyse der genannten EZM-Proteine sowie α-SMA durchgeführt werden. Um schließlich den Einfluss der von Tumorzellen sezernierten löslichen Faktoren in der Tumormikroumgebung herauszustellen, wurden Sphäroide aus hASCs in tumorkonditioniertem Medium gezüchtet und darin ebenfalls Matrixproteine und α-SMA untersucht.
Abschließend erfolgte eine Korrelation der EZM-Analyse mit dem Vorhandensein von Tumorstammzellen in den 3D-Tumorsphäroiden. Dafür wurden die Tumorstammzellen mithilfe eines GFP-basierten Reporters für den Stammzellmarker NANOG (NANOG-GFP-Reporterzelllinie) in mikroskopischen Aufnahmen der 3D-Tumorsphäroide nachgewiesen und im Kontext mit der EZM lokalisiert.
Bei der Implantatversorgung von Patienten mit Osteoporose besteht weiterhin eine hohe Komplikationsrate vor allem durch aseptische Prothesenlockerungen. Eine vielversprechende Möglichkeit diese zu minimieren stellt eine Funktionalisierung der Implantate mit Strontium dar.
Ziel der vorliegenden Arbeit war es dabei die Wirkung lokal verfügbaren Strontiums auf osteoklastäre und osteoblastäre Zellen zu untersuchen.
Mittels elektrochemischer Abscheidung erfolgte die Beschichtung von Titanproben mit strontiumdotiertem Struvit, wobei sieben verschiedene Dotierkonzentrationen zwischen 6 µg und 487 µg Strontium pro Probe hergestellt wurden. Die Untersuchungen an osteoklastären RAW 264.7 Zellen erfolgten mittels Bestimmung von Zellzahl und -aktivität, verschiedener mikroskopischer Methoden sowie auf genetischer Ebene. Osteoblastäre MG63-Zellen wurden orientierend anhand von Zellzahl und Zellaktivität untersucht.
Zellbiologisch konnte ein hemmender Einfluss von Strontium auf Differenzierung sowie Proliferation und Aktivität osteoklastärer Zellen gezeigt werden. Die Dotierkonzentration mit den günstigsten Eigenschaften war unter vorliegenden Versuchsbedingungen 487 µg Strontium pro Probe, da sich hierbei zudem eine erhaltene ostoblastäre Proliferation und Aktivität zeigte.
Elektrochemisch gestützte Abscheidung kupfer- und zinkdotierter Magnesiumphosphatschichten auf Titan
(2020)
Zur Entwicklung von Implantaten, welche eine komplikationsärmere Einheilung aufweisen, wurde eine dünne, homogene Beschichtung von Titanprobenkörpern mit Struvit mithilfe elektrochemischer Abscheidung generiert. Hierbei wurden dem Basiselektrolyt in den Versuchsreihen unterschiedliche Konzentrationen an Kupfer-(II)-nitrat-3-hydrat- und/oder Zinknitrat-6-hydratlösung hinzugefügt. Die experimentelle Freisetzung erfolgte in drei unterschiedlichen physiologischen Nährmedien: simulated body fluid (SBF), fetal calf serum (FCS) und Dulbecco’s Modified Eagle Medium (DMEM). Es konnte gezeigt werden, dass eine antibakteriell wirkende Menge an Kupfer- und Zinkionen freigesetzt wurde. Zusammenfassend stellt die elektrochemische Abscheidung von mit Kupfer- und Zink-dotierten Struvit auf Titanoberflächen einen vielversprechenden Ansatz in der Implantologie hinsichtlich der Einheilzeit im Knochen sowie der Risikominimierung des Verlustes dar.
As a major component of the articular cartilage extracellular matrix, hyaluronic acid is a widely used biomaterial in regenerative medicine and tissue engineering. According to its well-known interaction with multiple chondrocyte surface receptors which positively affects many cellular pathways, some approaches by combining mesenchymal stem cells and hyaluronic acid-based hydrogels are already driven in the field of cartilage regeneration and fat tissue. Nevertheless, a still remaining major problem is the development of the ideal matrix for this purpose. To generate a hydrogel for the use as a matrix, hyaluronic acid must be chemically modified, either derivatized or crosslinked and the resulting hydrogel is mostly shaped by the mold it is casted in whereas the stem cells are embedded during or after the gelation procedure which does not allow for the generation of zonal hierarchies, cell density or material gradients. This thesis focuses on the synthesis of different hyaluronic acid derivatives and poly(ethylene glycol) crosslinkers and the development of different hydrogel and bioink compositions that allow for adjustment of the printability, integration of growth factors, but also for the material and biological hydrogel, respectively bioink properties.
Mineral biocements are brittle materials, which usually results in catastrophic failure during mechanical loading. Here, previous works demonstrated the feasibility of reducing brittleness by a dual-setting approach, in which a silica sol was simultaneously gelled during the setting of a brushite forming cement. The current thesis aimed at further improving this concept by both using a novel silicate based cement matrix for an enhanced bonding between cement and silica matrix as well as multifunctional silica precursors to increase the network density of the gel. Due to its well-known biocompatibility and osteogenic regeneration capacity, baghdadite was chosen as mineral component of such composites. This required in a first approach the conversion of baghdadite ceramics into self-setting cement formulations. This was investigated initially by using baghdadite as reactive filler in a brushite forming cement (Chapter 4). Here, the ß-TCP component in a equimolar mixture of ß-TCP and acidic monocalcium phosphate anhydrous was subsequently replaced by baghdadite at various concentrations (0, 5, 10, 20, 30, 50, and 100 wt%) to study the influence on physicochemical cement properties such as mechanical performance, radiopacity, phase composition and microstructure. X-ray diffraction profiles demonstrated the dissolution of baghdadite during the cement reaction without affecting the crystal structure of the precipitated brushite phase. In addition, EDX analysis showed that calcium is homogeneously distributed in the cement matrix, while zirconium and silicon form cluster-like aggregates ranging in size from a few micrometers to more than 50 µm. X-ray images and µ-CT analyses indicate improved X-ray visibility with increased incorporation of baghdadite in brushite cement, with an aluminum equivalent thickness nearly doubling at a baghdadite content of 50 wt%. At the same time, the compressive strength of brushite cement increased from 12.9 ± 3.1 MPa to 21.1 ± 4.1 MPa at a baghdadite content of 10 wt%. Cell culture medium conditioned with powdered brushite cement approached physiological pH values when increasing amounts of baghdadite were added to the cement (pH = 6.47 for pure brushite, pH = 7.02 for brushite with 20 wt% baghdadite substitution). Baghdadite substitution also affected the ion content in the culture medium and thus the proliferation activity of primary human osteoblasts in vitro. The results demonstrated for the first time the suitability of baghdadite as a reactive cement additive for improving the radiopacity, mechanical performance, and cytocompatibility of brushite cements.
A second approach (Chapter 5) aimed to produce single component baghdadite cements by an increase of baghdadite solubility to initiate a self-setting cement reaction. For this, the material was mechanically activated by longer grinding times of up to 24h leading to both a decrease in particle and crystallite size as well as a partial amorphization of baghdadite. Baghdadite cements were formed by adding water at a powder to liquid ratio of 2.0 g/ml. Maximum compressive strengths were determined to be ~2 MPa after 3 days of setting for a 24-hour ground material. Inductively coupled plasma mass spectrometry (ICP-MS) measurements showed an incongruent dissolution profile of the set cements, with preferential dissolution of calcium and only minor release of zirconium ions. Cement formation occurs under alkaline conditions, with the unground raw powder resulting in a pH of 11.9 during setting, while prolonged grinding increases the pH to about 12.3.
Finally, mechanically activated baghdadite cements were combined with inorganic silica networks (Chapter 6) to create dual-setting cements with a further improvement of mechanical performance. While a modification of the cement pastes with a TEOS derived sol was already thought to improve strength, it was hypothesized that using multi-arm silica precursors can further enhance their mechanical performance due to a higher network density. In addition, this should also reduce pore size of both gels and cement and hence will be able to adjust the release kinetics of incorporated drugs. For this, multi-armed silica precursors were synthesized by the reaction of various multivalent alcohols (ethylene glycol, glycerine, pentaerythrit) with an isocyanate modified silica precursor. After hydrolysis under acidic conditions, the sols were mixed with baghdadite cement powders in order to allow a simultaneous gel formation and cement setting. Since the silica monomers have a high degree of linkage sites, this resulted in a branched network that interpenetrated with the growing cement crystals. In addition to minor changes in the crystalline phase composition as determined by X-ray diffraction, the novel composites exhibited improved mechanical properties with up to 20 times higher compressive strength and further benefit from an about 50% lower overall porosity than the reference pure baghdadite cement. In addition, the initial burst release of the model drug vancomycin was completely inhibited by the added silica matrix. This observation was verified by testing for the antimicrobial activity with Staphylococcus aureus by measuring the inhibition zones of selected samples after 24 h and 48 h, whereas the antimicrobial effectiveness of a constant vancomycin release could be demonstrated.
The current thesis clearly demonstrated the high potential of baghdadite as a cement formulation for medical application. The initially poor mechanical properties of such cements can be overcome by special processing techniques or by combination with silica networks. The achieved mechanical performance is > 10 MPa and hence suitable for bone replacement under non-load bearing conditions. The high intrinsic radiopacity as well as the alkaline pH during setting may open the way ahead to further dental applications, e.g. as root canal sealers or filler in dental composites. Here, the high pH is thought to lead to antimicrobial properties of such materials similar to commonly applied calcium hydroxide or calcium silicates, however combined with an intrinsic radiopacity for X-ray imaging. This would simplify such formulations to single component materials which are less susceptible to demixing processes during transport, storage or processing.
The human body has very good self-healing capabilities for numerous different injuries to a variety of different tissues. This includes the main human mechanical framework, the skeleton. The skeleton is limited in its healing without additional aid by medicine mostly by the defect size. When the defect reaches a size above 2.5 cm the regeneration of the defect ends up faulty. Here is where implants, defect fillers and other support approaches developed in medicine can help the body to heal the big defect still successfully.
Usually sturdy implants (auto-/allo-/xenogenic) are implanted in the defect to bridge the distance, but for auto- and allogenic implants a suitable donor site must be found and for all sources the implant needs to be shaped into the defect specific site to ensure a perfect fit, the best support and good healing. This shaping is very time consuming and prone to error, already in the planning phase. The use of a material that is moldable and sets in the desired shape shortly after applying negates these disadvantages. Cementitious materials offer exactly this property by being in a pasty stage after the powder and liquid components have been mixed and the subsequently hardening to a solid implant. These properties also enable the extrusion, and therefore may also enable the injection, of the cement via a syringe in a minimal invasive approach.
To enable a good injection of the cement modifications are necessary. This work aimed to modify commonly used calcium phosphate-based cement systems based on α-TCP (apatitic) and β-TCP (brushitic). These have been modified with sodium phytate and phytic acid, respectively. Additionally, the α-TCP system has been modified with sodium pyrophosphate, in a second study, to create a storable aqueous paste that can be activated once needed with a highly concentrated sodium orthophosphate solution.
The powder phase of the α-TCP cement system consisted of nine parts α-TCP and one part CDHA. These were prepared to have different particle sizes and therefore enable a better powder flowability through the bimodal size distribution. α-TCP had a main particle size of 20 μm and CDHA of 2.6 μm. The modification with sodium phytate led to an adsorption of phytate ions on the surface of the α-TCP particles, where they started to form complexes with the Ca2+ ions in the solution. This adsorption had two effects. The first was to make the calcium ions unavailable, preventing supersaturation and ultimately the precipitation of CDHA what would lead to the cement hardening. The second was the increase of the absolute value of the surface charge, zeta potential, of the powder in the cement paste. Here a decrease from +3 mV to -40 mV could be measured. A strong value for the zeta potential leads to a higher repulsion of similarly charged particles and therefore prevents powder agglomeration and clogging on the nozzle during injection. These two modifications (bimodal particles size distribution and phytic acid) lead to a significant increase in the paste injectability. The unmodified paste was injectable for 30 % only, where all modified pastes were practically fully injectable ~90 % (the residual paste remained in the nozzle, while the syringe plunger already reached the end of the syringe).
A very similar observation could be made for the β-TCP system. This system was modified with phytic acid. The zeta potential was decreased even stronger from -10 ± 1.5 mV to -71.5 ± 12 mV. The adsorption of the phytate ions and subsequent formation of chelate complexes with the newly dissolved Ca2+ ions also showed a retarding effect in the cements setting reaction. Where the unmodified cement was not measurable in the rheometer, as the reaction was faster than the measurement setup (~1.5 min), the modified cements showed a transition through the gel point between 3-6 min. This means the pastes stayed between 2 and 4 times longer viscous than without the modification. Like with the first cement system also here the effects of the phytate addition showed its beneficial influence in the injectability measurement. The unmodified cement was not injectable at all, due to the same issue already encountered at the rheology measurements, but all modified pastes were fully injectable for at least 5 min (lowest phytate concentration) and at least 10 min (all other concentrations) after the mixing of powder and liquid.
The main goal of the last modification with sodium pyrophosphate was to create a paste that was stable in aqueous environment without setting until the activation takes place, but it should still show good injectability as this was the desired way of application after activation. Like before also the zeta potential changed after the addition of pyrophosphate. It could be lowered from -22 ± 2mV down to -61 to -68 ± 4mV (depending on the pyrophosphate concentration). The pastes were stored in airtight containers at room temperature and checked for their phase composition over 14 days. The unmodified paste showed a beginning phase conversion to hydroxyapatite between 7 and 14 days. All other pastes were still stable and unreacted. The pastes were activated with a high concentrated (30 wt%) sodium orthophosphate solution. After the activation the pastes were checked for their injectability and showed an increase from -57 ± 11% for the unmodified paste to -89 ± 3% (practically fully injectable as described earlier) for the best modified paste (PP005).
It can be concluded that the goal of enabling full injection of conventional calcium phosphate bone cement systems was reached. Additional work produced a storage stable paste that still ensures full injectability. Subsequent work already used the storable paste and modified it with hyaluronic acid to create an ink for 3D extrusion printing. The first two cement systems have also already been investigated in cell culture for their influence on osteoblasts and osteoclasts. The next steps would have to go more into the direction of translation. Figuring out what properties still need to be checked and where the modification needs adjustment to enable a clinical use of the presented systems.