3.1.3. Cementum Formation

Cementum regeneration is closely related to the treatment of the periodontal complex comprised of alveolar bone, periodontal ligament, gingiva, and cementum (Figure 1). Besides the structural support a sca ffold material provides to the a ffected tissue, sca ffolds used for regeneration of the periodontal complex are often used as a delivery vehicle for various bioactive compounds such as proteins, growth factors, or gene vectors to favor the regenerative process and induce the recruitment and homing of endogenous stem cells from surrounding tissues. The development of multicompartment sca ffolds aims to meet the diverse challenges of the di fferent tissues to be regenerated in periodontal defects in a single sca ffold [141]. Additionally, besides synthetic sca ffolds, cell-based sca ffolds such as cell sheets are part of current research. In this approach, cell types that are relevant for the periodontal regeneration are cultivated in vitro extensively, until strong cell–cell interactions are established and an extracellular matrix has formed, thus allowing transplantation of the cell sheet as a sca ffold-like material [142].

Recently, Fakheran and peers evaluated the regenerative potential of Retro MTA, a calcium silicate cement, in combination with tricalcium phosphate in vivo and showed that newly formed bone and cementum was significantly higher than in the untreated control group. Moreover, the poor biodegradation rate of MTA is improved due to the combination with biodegradable TCP [143]. In a preclinical study to treat periodontal defects in dogs, Wei et al. used an inorganic calcium phosphate-based sca ffold material loaded with BMP2 [144]. The CaP-based biomaterial alone leads to a significantly increased regeneration of mineralized tissue as well as to an improved attachment of the teeth to the surrounding tissue compared to untreated control and a deproteinized bovine bone mineral that serves as commercial control. When loaded with BMP2, these positive results could even be improved two- and three-fold regarding height and area of the remineralized tissues, respectively. Noteworthy, the encapsulated BMP2 had a greater impact on osteogenesis than on cementogenesis [144]. Following the multicompartment-sca ffold approach, Wang and collaborators applied a bilayered material containing growth factors. The hybrid material containing an FGF2-loaded propylene-glycol alginate gel coating the root surface for ligament regeneration and a BMP2-loaded (PLGA)/calcium phosphate cement for periodontal regeneration was tested in vivo with non-human primates. Following a promising study in rodents, the authors reported significantly enhanced regeneration of cementum and periodontal ligament and a high vascularization of the newly formed periodontal ligament (PDL), thereby confirming the positive results of the previous study [145,146].

Vaquette el al. developed bilayered sca ffold materials based on polycaprolactone and combined them with cell sheets: while a fibrous three-dimensional compartment with macropores should favor alveolar bone regeneration, a flexible porous membrane aims at delivering the cell sheet and regenerates the periodontal ligament [147]. In their study, the authors evaluated the in vivo regenerative potential of the hybrid materials with di fferent cell types forming the cell sheet, namely gingival cells, periodontal ligament cells (PDLCs), and bone marrow-derived mesenchymal stem cells (BM-MSCs). Results from histomorphometry and micro-computed tomography (μ-CT) show that sca ffolds containing BM-MSCs and PDLCs had greater regenerative potential due to superior new bone and cementum formation compared to the sca ffolds containing gingival cell sheets. However, the regenerative potential of sca ffolds containing BM-MSCs and PDLCs did not di ffer significantly compared to the performance of the non-cellularized control sca ffold. Thus, the biphasic sca ffold alone is also a promising candidate

for further studies [147]. Table 2 summarizes recently published studies emphasizing regenerative approaches of enamel, dentin, and cementum.

**Table 2.** Compilation of recently published studies emphasizing regenerative approaches of enamel, dentin, and cementum.


#### *3.2. Drug Release Systems Useful in Tissue Engineering—To be Adapted to Tooth Engineering*

As discussed in the previous section, whole tooth regeneration is one of the most challenging fields in regenerative medicine—also regarding drug release aspects. In stem cell-based approaches, a cocktail of different drugs would be required to tightly tailor the differentiation of the corresponding cells involved in amelogenesis, dentinogenesis, and cementogenesis, respectively. This means that, besides appropriate scaffolds, compounds have to be developed for drug encapsulation and controlled release of those substances involved during tooth formation (such as growth factors and receptor ligands, as listed in Figure 2). Thus far, drug release approaches in tooth regeneration are mainly restricted to the delivery of antibiotics to avoid inflammation [66].

In analogy to other tissues and organs engineered using stem cell-based approaches, the drug delivery systems (DDS) are mainly classified into the following release mechanisms: di ffusion through water-filled pores; di ffusion through the polymer; osmotic pumping; and erosion [148]. In the past two decades, novel release materials have been designed and prepared that could be classified into the following three groups: (a) polymer-based systems; (b) ceramics-based systems; and (c) hybrid systems (e.g., organic/inorganic and polymer/ceramic) [62,149]. Many of them are prepared as nanomaterials (e.g., spheres, capsules, and rods) [64].

To develop a DDS that allows kinetically controlled release of drugs supporting the required stem cell di fferentiation processes, a variety of material characteristics would have to be considered. Parameters that influence the release behavior of polymer-based release materials include the following: molecular weight (number and weight average, respectively, Mn/M w) and corresponding polydispersity index (PI), number and nature of end-groups, and the polymer morphology mainly determined by the monomer 3D structure (amorphous and crystalline/semi-crystalline with the degree of crystallinity). All of them are able to influence the size and shape, as well as density and porosity of the entire DDS that includes the encapsulated drugs. In addition, the active substance (drug) itself influences the release kinetics via interaction with the encapsulation material. Thus, the drug hydrophilicity/hydrophobicity (resulting from chemical composition, functional groups, hydrogen bonds, etc.) is one of the most limiting aspects, as well as its ability to act as surfactants or plasticizer which would interfere with the release mechanism. Huang et al. comprehensively reviewed the release mechanisms discovered within the last five years, including drugs for tooth regeneration [65]. Most recent developments include tunable conductive polymers to be used for controlled delivery [150]. As stated in Section 3.1, in tooth regeneration, drugs (such as growth factors and FGF-2) are usually simply added to the sca ffold material—not ye<sup>t</sup> encapsulated and released from tailored delivery materials [14,19,66,67,151–157]. Recently, Moon et al. reported a study using nitric oxide release to support the pulp–dentin regeneration [158]. However, in this case, release kinetics cannot be controlled or adjusted to the di fferentiation processes of the corresponding cells. Very few studies reported the application of specific drug encapsulation materials, mainly using hydrogels [63–65,159–161]. Hydrogels can easily be prepared using natural and artificial polymers (sometimes a combination of both classes). One of the most prominent groups of hydrogels is based on polysaccharides [149,162,163]. Furthermore, other polymers such as polyvinyl alcohol (PVA), polylactic acid and polyglycolic acid (PGA), polyacrylic acid (PAA), and polyethylene glycol (PEG) are intensively studied regarding their ability to form hydrogels used for controlled delivery [160,164]. Hydrogels o ffer various advantages; most importantly, they are tunable in their chemical structure resulting in controlled degradability. In a comprehensive review, Li et al. discussed various multiscale release kinetic mechanisms of hydrogels and classified them according to the structural interactions. Thus, the kinetics are significantly determined by the hydrogel mesh size, network degradation, swelling, and mechanical deformation. In addition, kinetics depend on various interactions of the hydrogel components such as conjugation, electrostatic interaction, and hydrophobic association [164].

For hard tissue such as bone, our group could recently show that it is possible to guide osteogenesis via purinergic receptor ligand release. Osteogenesis of mesenchymal stem cells is influenced by various purinergic receptors (P1, P2X, and P2Y) [122,124,165–168]. Thus, a release of specific agonists and/or antagonists enables tailoring of the corresponding receptor up- or downregulation. Furthermore, besides osteogenesis, purinergic receptors are also involved in angiogenesis—a process also required during tooth regeneration [68,169,170].

In a recently published paper, we reported the synthesis and testing of novel hybrid release materials based on hydroxyapatite and agarose used to improve the release kinetics of drugs applied for guided osteogenesis [171]. Scanning electron microscopy (SEM) revealed details regarding the influence of the drying treatment: lyophilized (LYO) versus supercritically-dried (SCD) gels were tested and compared. As shown in Figure 3, SEM confirmed a homogeneous distribution of the elements involved in the hybrid (carbon, calcium, and phosphorus). In addition to SEM, energy-dispersive X-ray spectroscopy (EDX) results are given in [171]).

**Figure 3.** SEM images of agarose lyophilized (LYO) (**<sup>a</sup>**–**<sup>c</sup>**) and supercritically-dried (SCD) (**d**–**f**) and agarose/hydroxyapatite (33/76 w%) composite LYO (**g**–**i**) and SCD (**k**–**<sup>m</sup>**) at three different magnifications. The scale bars are 10 μm (left), 1 μm (middle), and 0.2 μm (right), respectively. Reproduced from Witzler et al., 2019 [171]. Open Access Copyright Permission (Creative Commons CC BY license).

Hitherto, hybrid systems are mainly studied as release materials for hard tissue regeneration [67]. Here, sustained delivery is required for guided stem cell differentiation, a burst release is favorable to achieve anti-inflammatory and antibacterial effects. Since both processes are also relevant in tooth formation, hybrid materials would be promising candidates to be investigated as release materials to improve cascades, as shown in Figure 2. In previous studies, the HA/agarose hybrids were loaded with model drug compounds for guided differentiation of MSCs. Different release kinetic models were evaluated for adenosine 5--triphosphate (ATP) and suramin (Figure 4) [171]. Although both drugs are highly water-soluble, the release could be slowed to four days, which is significantly longer than comparable systems reported in the literature [172].

Future efforts should be directed toward the development of tailored drug loading and/or encapsulation materials to be used for the controlled release of bioactive substances during tooth formation [157,173]. As shown in Figure 2 and Table 1, there are various signaling molecules and corresponding activators and suppressor molecules involved in the formation of enamel, dentin, and cementum. For a number of these substances, loading and controlled-release from non-cytotoxic materials already exist, as shown in Table 3. Release materials mainly consist of natural or artificial polymers, but also hybrids composed of organic and inorganic components. The focuses of the studies are release kinetics and corresponding mechanisms. However, some drugs are being successfully applied in vivo.

**Figure 4.** Release data of (**a**) adenosin triphosphate (ATP) and (**b**) suramin from agarose/hydroxyapatite (AG100HA0) (black), AG50HA50 (orange), and AG33HA67 (blue) scaffolds. Data fit: Weibull equation. Reproduced from Witzler et al., 2019 [171]. Open Access Copyright Permission (Creative Commons CC BY license).




**Table 3.** *Cont.*

In detail, a sequential and on-demand release of multiple drugs (signaling molecules, activators, and suppressors) would be required to control and guide the signaling cascades of amelogenesis, odontogenesis, and cementogenesis [164]. Moreover, on-demand release systems usually require specific stimuli as reported for example for conductive polymer-based delivery devices [150]. Finally, theoretical modeling could provide a more fundamental understanding of release kinetics [189].

## **4. Whole Tooth Regeneration**

The regeneration of a whole tooth as an organ replacement therapy is considered to be the ultimate goal of regenerative dentistry. For patients, this therapeutic option could represent a dream for the replacement of decayed or lost teeth to overcome prosthodontic or implantology treatment using artificial replacements. Whole-tooth generation could be performed as a hybrid strategy where, e.g., biologically created tissue compartments such as the periodontal ligament or a tooth crown would be combined with a metallic or ceramic implant or where a biological regenerated tooth root ("bio-root") would be combined with a prosthetic crown (see, e.g., [190–192]). In the following years, efforts in creating a whole tooth from only cells and tissues ("bio-tooth") will be very likely in the focus. However, despite all efforts and achieved results in basic and translational research, this approach is still challenging [48,58,69,193,194].

#### *4.1. Reactivation the Odontogenic Potency*

On the background of teeth evolution, a genetic approach to generate whole teeth may be an option in the far future. Teethed fishes, reptiles, or amphibians are polyphyodonty, which means that several tooth generations can be formed and erupted. This highly regenerative capacity was reduced during evolution. In mammals, many species including human are only diphyodont with the capacity to form a second dentition or even monophyodont such as the mouse [71,72,195]. Revitalizing the odontogenic potency for the lost tooth regeneration capacity may be an interesting approach to induce tooth formation in vivo in the adult. One prerequisite for tooth replacement is the existence of a successional dental lamina (SDL) carrying the capacity for inducing odontogenesis. Even in monophyodont animals, rudimentary SDL has been identified. In addition, in the human species, rudimentary laminae are preserved, which might be responsible for a third dentition but this, however, has been observed very seldomly. On a molecular level, tooth replacement is regulated by signaling pathways [71]. For example, in alligators or snakes, stem cells in the SDL express Sox2, which is initiated by the Wnt/β-catenin pathway an interacts with BMP signaling [195]. Dysregulation of Wnt-signaling is discussed to be important for the de-activation of rudimentary SDL as it occurs in the mouse. Therefore, the revitalization by stabilizing Wnt signaling by application of appropriate factors or genes could be a strategy for the induction of re-growing teeth in the future [195,196].

## *4.2. Tissue Recombination Approaches*

The basic principle of this "classical" approach is to mimic the natural development and formation of a tooth and to recapitulate the signaling cascades regulating tissue interactions during odontogenesis. For over a hundred years, progress has been made in understanding tooth development in di fferent species including human, identifying tissue interactions and factors involved on the morphological, cellular and molecular levels [18,58,65,71,193,195,197]. Classical tissue recombination experiments undertaken in developmental biology research have shown that mouse embryonic tooth germs can be dissociated and later re-aggregated. After temporary ectopic grafting of these cell aggregates, e.g., into the anterior eye chamber, subcutaneously, or under the renal capsule, tooth-like organs with mineralized tissues (dentin and enamel) could be grown (e.g., [198]). This method has been improved in the last years by using collagen drops for the organoid culture of 5–7 days or seeding the re-aggregated germ cells on biodegradable polymers [199–201]. The final goal of these experiments was to implant the constructs into the jaws of postnatal animals to generate a whole "bio-tooth". In line with this cultured rat tooth, bud cells seeded onto biodegradable sca ffolds for 12 weeks formed tooth-like crowns consisting of pulp, dentin, enamel, and periodontal ligament after implantation into rat jaws [200].

A breakthrough came with experiments of the group of Ikeda, who could demonstrate that the implantation of re-aggregated autogenous germ cells into the extraction socket of pigs succeeded in the formation, development, and eruption of teeth, which could be brought into occlusion and fulfilled all functions of normal teeth [50]. Over half (56%) of the implanted constructs had erupted. Later, it was also possible to create a unit of a regrown tooth with surrounding alveolar bone [202]. Whole-tooth restoration using autologous bioengineered tooth germ transplantation was also successful in canines [51]. An allogeneic approach was undertaken by Wu and colleagues, who transplanted re-associated tooth germs into the jawbones of minipigs [203]. A xenogeneic approach was published by Wang and co-workers in 2018 [52]. Cells from unerupted deciduous molar germs of pigs were recombined and transplanted first in mouse renal capsules and finally in jawbones. However, problems are caused by the limited sources of tooth germ cells and risks of immune rejection when using allogeneic or xenogeneic cells. In humans, there are many hindrances, e.g., that tooth germs may not be easily accessible, but also ethical and legal constraints must be considered. An alternative could be the use of adult stem cells (see Section 4.2) or of iPSCs [53,54].

Di fferent types of adult dental stem cells, e.g., from the pulp, or di fferentiated orofacial cells, e.g., from the gingiva, can be used as sources to create iPSCs with a similar epigenetic pattern. These cells show the ability to di fferentiate into epithelial or mesenchymal tooth germ cells [58,92]. Cai and co-workers generated iPS cells from cells out of human urine, which were di fferentiated to epithelial sheets and recombined with embryonic mouse dental mesenchyme [91]. Tooth-like structures were generated in which the epithelial cells di fferentiated into enamel-secreting ameloblasts. The formation of enamel, the hardest tissue of the body (see Section 2.2.1), is an important step in generating whole teeth, but also would be of importance for repair or regeneration of enamel loss in conservative dentistry. Thus, it is of major interest to find tissue sources able to generate dental epithelial cells which can be di fferentiated into enamel-secreting ameloblasts. Aside from iPSCs, examples for this are epithelial cells from the skin or gingiva as well as epithelial rests of Malassez, which can be found in the PDL, co-culture of these cells with di fferent types of dental mesenchymal cells can lead to ameloblast di fferentiation or even formation of enamel-like structures [58,87,89].

#### *4.3. Adult Stem Cell Approaches*

The optimal method to create whole teeth would be the use of autogenous dental cells from patients demanding tooth regeneration. For whole tooth bioengineering, di fferent strategies in the application of these cells have been developed. One idea was to combine adult stem cells with cells of the progenitor cells of embryonic tooth. Adult stem cells should have an odontogenic competence and should function as a "tooth inducer" when combined with mesenchymal cells or they should express a

dental mesenchymal competence when combined with dental epithelium. Already in 2002, Young et al. cultured cells obtained from unerupted porcine tooth buds [199]. The aggregates were grown on biodegradable sca ffolds in vitro or transplanted. This led to the formation of a primitive tooth crown with pulp, dentin, and enamel formation. Later, similar bioengineered tooth-like structures could be obtained by using rat and human cells [204,205]. In 2004, Ohazama and colleagues used non-dental adult MSCs in combination with inductive embryonic dental epithelium first transplanted under the renal capsule and transplanted them in adult jaws. Tooth formation including root occurred and the teeth erupted. In addition, bone was induced [206]. Adipose-derived MSCs alone were able to generate tooth bud-like structures in vitro [90]. Human gingival epithelial cells were used by Volponi Angelova and associates and combined with embryonic mouse tooth mesenchyme, which yielded an entire tooth outside of an embryo [207].

However, for all these experiments, relatively large amounts of adult cell populations were necessary that should be able to retain any odontogenic potential and, in addition, a large number of embryonic cells was needed as well. In a case of embryonic mouse tooth mesenchyme, a minimum cell number of 4 × 10<sup>4</sup> to 4 × 10<sup>5</sup> was su fficient according to the experiments of Hu et al. (2006) [208]. Therefore, to do so, cells from multiple embryos must be harvested. Another problem is the loss of the inductive capacity already after 24–48 h in culture, which makes the in vitro expansion of these cells using standard methods impossible [209]. This phenomenon can be explained by the fact that mesenchymal stem cells lose their dense packaging formed by cellular condensation and thus their linked cell contacts, which is a prerequisite for an inductive capacity in vivo. Ongoing research focuses therefore also on how an odontogenic potential can be maintained in vitro [194]. 3D micro-culture systems such as the hanging drop method in liquid media may allow the preservation of such signals. However, many cells are necessary for these methods [210]. Gene expression studies must be undertaken to identify signaling factors, which are lost in 2D cell cultures. In a study using postnatal dental pulp stem cells, Yang and collaborators could obtain "a rescue" in cultured cells due to the combination with uncultured mesenchymal tooth germ cells [57]. This rescue or community effect is responsible for the reactivation of inductive signals. Forthcoming, iPS cells (see Section 4.2) may be an appropriate cell substitute to overcome these biological problems.

In the future, research will presumably focus on using adult stem cells from dental and non-dental sources to test recombination or co-culturing for their e ffects on tooth development. Zhang and coworkers optimized such a method by recombinant 3D-tissue engineering of intact dental tissues and cell suspensions from postnatal porcine teeth and human third molars [211]. After osteogenic culturing and subcutaneous transplantation in athymic nude rat hosts, tooth-like constructs forming all dental hard substances could be harvested. Recently, tooth buds could be generated by co-culturing postnatal dental stem cells with human HUVEC cells encapsulated in gelatine hydrogel [56]. Only postnatal dental stem cells were used by Yang et al. (2016), who di fferentiated odontoblasts and osteoblasts from pig dental pulp stem cells and seeded them with gingival epithelium on a bioactive sca ffold. Implantation into extraction sockets of 13.5-month-old pigs revealed the development of teeth in seven of eight animals. The regenerated molar teeth expressed dentin-matrix protein-1 and osteopontin [212].

#### *4.4. Problems in Whole Tooth Regeneration*

Despite the progress in some basic strategies for tooth regeneration, we still face a lot of problems [18,48]. An important condition for a proper functional occlusion in a dentition where teeth should be replaced by regeneration is the correct anatomical size and shape of the crown. Especially the relief of the occlusal surface with its specific pattern of fissures and cusps is relevant for a functional occlusion. The proper size and shape of a crown are determined by epithelial morphogenesis forming spatially regulated cellular condensations as signaling centers, called knots [71,197]. These knots (initiation knot, primary enamel knot, and secondary enamel knot) regulate crown development and cusp number, morphology, and pattern by expressing di fferent factors such as FGF, BMP, Wnt, or Shh, as already mentioned. The number of tooth cusps in the mouse depends on the activity of Shh, EDA, and Activin A pathways [71,197]. The tooth size is independently regulated from the cusp number and is not only dependent on epithelial, but also mesenchymal influences. Therefore, it was suggested that the tooth size could be controlled by prolonging the activity of tooth epithelial stem cells and increasing the number of mesenchymal stem cells in recombination experiments [197]. The di fferent tooth types such as molars or incisors have specific morphological features not only of the occlusal surface but also of the crown and root morphology. This will also be an important aspect for future tooth engineering [193]. The quality and the biomechanical loading of dental hard tissues are also important for occlusion and mastication. In already developed models of tooth regeneration, only a low level of enamel mineralization could be observed.

Tooth health is also dependent on proper vascularization and innervation. While vascularization occurs in di fferent models already published [213], the question is whether this would be also su fficient for the long-lasting survival of regenerated teeth. E fforts have been made to induce neurogenesis and formation of nerve fibers, e.g., by using exogenous agents such as semaphorin 3 receptor inhibitors, by application of immunomodulation using cyclosporin A, or implication of bone marrow stromal cells [48]. Recently, Strub et al. recombined embryonic dental epithelium with a mixture of dental mesenchymal cells and bone marrow-derived cells and cultured and implanted these cells subcutaneously. The tooth-like tissues obtained were innervated with axons entering the newly formed pulp [214].

Other problems include the formation of a proper periodontium or infections occurring during or after transplantation. If whole tooth constructs can be implanted, the role of the tissue environment will play an important role in the success: How is the quality of the jawbone? How will the implantation be a ffected by age or systemic diseases of the patients? How resistant will the newly created tooth be against probable infections? Finally, the costs of creating a "bio-tooth" are also unpredictable ye<sup>t</sup> [48,156].

## **5. Conclusions and Future Perspectives**

Progress in regenerating whole teeth will need scientific research on di fferent levels such as identification of appropriate cell sources with tooth inductive signals. For this further research on the feasibility of iPS cells for this approach is important. Furthermore, the identification of master genes in gene regulatory networks responsible for tooth induction and tooth formation is necessary for successful manipulation of, e.g., adult cells to form bioengineered dental tissues, and to control tooth crown, size and tooth identity.

Applying the acquired knowledge about signaling pathways shaping dental tissue genesis might stimulate novel cell culture techniques establishment and functionalized sca ffolds development. Functionalized biomaterials will presumably play a central role in hard dental tissue regeneration such as dentin and cementum and probably the main role in enamel regeneration since this tissue is acellular and cannot be reproduced in vitro relying solely on a cell-based approach. Although several potentially appropriate biomaterials have already been investigated and tested, only very few examples were used in clinical studies until now. Future e fforts in stem cell-based approaches will very likely be directed toward biomaterials that allow sequential and on-demand drug release of multiple drugs in order to tailor timely the di fferent cascade processes during amelogenesis, dentinogenesis, and cementogenesis, respectively.

On the translational level, methods to improve 3D organogenesis, 3D printing applications, or the appropriate application of stimulatory molecules and drugs should be tested intensively. Solutions must be found for the proper mineralization of dental hard tissue formed by the regeneration process to ensure the natural properties of teeth in occlusion and mastication. Finally, there are considerable financial investment problems that should be taken into account. Then, but only then, whole biological tooth regeneration may even be a blueprint for the regeneration of other complex organs [70].

**Author Contributions:** Conceptualization and writing—original draft preparation, J.B., D.B., M.S., E.T., W.G.; writing—review and editing, E.T., M.S.; supervision and project administration, E.T., M.S.; funding acquisition, E.T., M.S., W.G. All authors have read and agree to the proofread version of the manuscript.

**Funding:** This article was supported by the Bundesministerium für Bildung und Forschung (BMBF) FHprofUnt to E.T., FKZ: 13FH012PB2; EFRE co-financed NRW Ziel 2: "Regionale Wettbewerbsfähigkeit und Beschäftigung", 2007–2013, Ministerium für Innovation, Wissenschaft und Forschung (MIWF) NRW FH-Extra to E.T., FKZ: z1112fh012; DAAD PPP Vigoni to E.T., FKZ: 314-vigoni-dr and FKZ: 54669218; BMBF-AIF to E.T., FKZ: 1720X06; BMBF IngenieurNachwuchs to E.T., FKZ: 13FH019IX5; MIWF NRW FH Zeit für Forschung to E.T., FKZ 005-1703-0017. BMBF IngenieurNachwuchs to S.W.; FKZ: 13FH569IX6; and National Council for Scientific and Technological Development of Brazil (*Conselho Nacional de Desenvolvimento Científico e Tecnológico*—*CNPq*, *Brasil*) under "*Cotas do Programa de Pós-Graduação*" No. 142137/2019-7 to J.B.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
