*Article* **Mechanical Characterization of Human Trabecular and Formed Granulate Bone Cylinders Processed by High Hydrostatic Pressure**

**Janine Waletzko-Hellwig 1,\*, Michael Saemann 2, Marko Schulze 3, Bernhard Frerich 1, Rainer Bader <sup>2</sup> and Michael Dau <sup>1</sup>**


**Abstract:** One main disadvantage of commercially available allogenic bone substitute materials is the altered mechanical behavior due to applied material processing, including sterilization methods like thermal processing or gamma irradiation. The use of high hydrostatic pressure (HHP) might be a gentle alternative to avoid mechanical alteration. Therefore, we compressed ground trabecular human bone to granules and, afterwards, treated them with 250 and 300 MPa for 20 and 30 min respectively. We characterized the formed bone granule cylinders (BGC) with respect to their biomechanical properties by evaluating stiffness and stress at 15% strain. Furthermore, the stiffness and yield strength of HHP-treated and native human trabecular bone cylinders (TBC) as control were evaluated. The mechanical properties of native vs. HHP-treated TBCs as well as HHP-treated vs. untreated BGCs did not differ, independent of the applied HHP magnitude and duration. Our study suggests HHP treatment as a suitable alternative to current processing techniques for allogenic bone substitutes since no negative effects on mechanical properties occurred.

**Keywords:** high hydrostatic pressure; mechanical characterization; uniaxial compression test; bone substitutes; allograft; bone regeneration

#### **1. Introduction**

The reconstruction of severe bone defects, which originate, e.g., from infections, pathologic fractures, tumors or trauma, still remains a clinical challenge [1]. Although there are a number of different possibilities for reconstructing bone, including xenografts like demineralized bone matrices, autologous bone is still considered to be the gold standard [2–4]. For reconstruction surgery, autologous bone can be obtained from various donor sites such as the iliac crest, and it is specified as osteogenic, osteoinductive, osteoconductive and biocompatible, with low immunological potential and adequate mechanical strength [5]. Most other bone substitutes cannot comply with all of these requirements. Nevertheless, harvesting autologous bone is naturally limited, the occurrence of donor-side morbidities is not unusual and it requires an advanced surgical procedure [2]. The most frequently chosen alternatives to autografts are allografts [5]. Allogenic bone grafts have osteoconductive properties and avoid donor-side morbidity in the recipient. Additionally, customized types of allografts like blocks, stripes or granules are possible [2]. However, postoperative infections due to residual microbiota, proteins, etc. following allograft transplantation of human origin are a risk [6]. To prevent this, different sterilization methods including

**Citation:** Waletzko-Hellwig, J.; Saemann, M.; Schulze, M.; Frerich, B.; Bader, R.; Dau, M. Mechanical Characterization of Human Trabecular and Formed Granulate Bone Cylinders Processed by High Hydrostatic Pressure. *Materials* **2021**, *14*, 1069. https://doi.org/ 10.3390/ma14051069

Academic Editor: Jung-Bo Huh

Received: 29 January 2021 Accepted: 19 February 2021 Published: 25 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

thermal processing, gamma radiation or physical and chemical decellularization have been established in recent years. Due to the removal of cellular components, any osteogenic properties are lost. However, this circumstance can be overcome with the revitalization of the graft using the recipient's own stem cells [6,7]. Unfortunately, the mechanical strength of the allografts usually suffers when using common decellularization and sterilization methods [6].

A reasonable alternative to current decellularization methods could be treatment with high hydrostatic pressure (HHP). HHP is commonly associated with processing of food and beverages [8]. This process inactivates microbes by membrane modifications, deactivation of key enzymes and inhibition of relevant metabolic processes like protein biosynthesis [9]. In comparison to conventional thermal food processing, HHP has the advantage that flavors and vitamins are unaffected by pressures up to 800 MPa [9,10]. In recent years, HHP has gained attention in pharmaceutical research. Rigaldie et al. have shown that HHP can be used to sterilize sensitive drugs like insulin with no effect on molecular integrity [11]. Furthermore, it was shown that HHP had a devitalizing effect on different mammalian cell lines, ex situ and in situ [12,13]. In the latter, it was already shown that this form of devitalization had no negative influence on the mechanical behavior of e.g., blood vessels [13].

The aim of our present study was to evaluate the mechanical properties of human trabecular bone cylinders (TBC) and bone granules pressed to cylinders (bone granule cylinders, BGC), both treated with HHP. While TBCs with an interconnected, trabecular structure can be used for larger bone defects, the use of BGCs as filling material for nonload-bearing bone defects is conceivable. A previous study at the cellular level showed that osteoblasts, as part of trabecular bone, follow either apoptotic or necrotic means of cell death, depending on the applied HHP magnitude. A pressure range of 100–150 MPa for 10 min did not have a negative influence on the metabolic activity and cell death could not be detected. Applied pressures of 250 MPa and more led to a significant reduction in metabolic activity compared to the control group. However, it was found that a pressure of 250–300 MPa tended to lead to apoptosis, while a pressure of 450–500 MPa had a necrotic effect on the osteoblasts [14]. The level and duration of HHP applied to tissues should be selected carefully, as necrosis can be a crucial factor in clinical transplantation due to the conceivably strong immunological response of the recipient [12]. Relying on the previous cell-based study, pressures of 250 and 300 MPa were used in the present experiments, as it was assumed that the biological effects would be similar. However, due to the changes in sample geometry compared to the cell pellets, the treatment periods for TBCs and BGCs were increased from 10 min to 20 and 30 min, respectively. The mechanical properties were analyzed by performing uniaxial compression tests and comparing stiffness and strength.

#### **2. Materials and Methods**

#### *2.1. Sample Preparation, HHP Treatment and Creation Granules-Based Bone Cylinders*

Trabecular bone specimens were taken post-mortem from human femur condyles and the femoral heads of body donors (Institute of Anatomy, Rostock University Medical Center; ethics approval A 2016-0083). Both were harvested within 72 h post-mortem in order to prevent the samples from being affected by decomposition processes. Afterward, all samples were rinsed once with sterile phosphate-buffered saline (PBS) (Sigma Aldrich, Munich, Germany), supplemented with 1% penicillin/streptomycin (Sigma Aldrich, Munich, Germany). Femur condyles and femoral heads were stored at −20 ◦C and covered with cling film until further preparation for HHP treatment and mechanical testing.

Before preparation of the TBCs for the compression tests, femoral condyles were slowly thawed at 4 ◦C. The defrosted condyles were partitioned into different sections (Figure 1). Within each predefined section, cylinders with a diameter of 6 mm were obtained from the proximal side using a trepan drill (Ustomed Instrumente, Tuttlingen, Germany). This was performed at room temperature under constant cooling with physiological saline solution (B. Braun, Melsungen, Germany) to prevent damage from heat. The plane ends

of the cylinders were rectified with the help of a scalpel to achieve parallel end faces perpendicular to the drilling axis and to shorten cylinders to a length of approximately 10 mm. Care was taken to ensure that specimens consisted of only trabecular bone, and that the ends of the cylinders were parallel to each other with a deviation of less than 5◦. Specimens that did not satisfy these criteria were discarded.

**Figure 1.** Sample preparation: (**a**) Knee condyles of human femurs were partitioned into the shown sections. Within shown sections, long cylinders were drilled along the femoral axis from proximal to distal using a trepan drill. (**b**) Long cylinders were sectioned into smaller ones with a length of 10 mm and a diameter of 6 mm using a scalpel, and care was taken that the ends were parallel to each other with a deviation of less than 5◦.

For the compression test, trabecular cylinders from the identical harvesting location underwent different HHP treatments (control: *n* = 20; group A: 250 MPa, 20 min, *n* = 18; group B: 250 MPa, 30 min, *n* = 19; group C: 300 MPa, 20 min, *n* = 16; group D: 300 MPa, 30 min, *n* = 14) and were tested afterwards and compared one by one. Therefore, if a cylinder from the lateral region of a left femur condyle was taken for HHP treatment, the corresponding cylinder from the lateral region of a right femur was used as a control. The different group sizes arose due to the rejection of samples that did not satisfy the above-mentioned criteria and the alternating approach described before.

To investigate the influence of HHP on the mechanical properties of the bone granule cylinders (BGCs), the femoral heads were sawed into bone blocks with a size ranging between 0.05 and 0.1 cm3. Afterward, the bone blocks were processed by a bone mill (Ustomed Instrumente, Tuttlingen, Germany) to granules with a size between 1 and 2 mm.

For HHP treatment, the granules were transferred into 2 mL cryogenic tubes filled with sterile PBS. Different treatment protocols (control: *n* = 7; group A: 250 MPa, 20 min, *n* = 9; group B: 250 MPa, 30 min, *n* = 10; group C: 300 MPa, 20 min, *n* = 8; group D: 300 MPa, 30 min, *n* = 6) were applied at a constant temperature of 30 ◦C. The untreated specimens of the control group were stored for the same time in PBS at 30 ◦C.

Before performing the uniaxial compression test, the bone granules were pressed to cylinders with a diameter of 6 mm and a length of about 10 to 12 mm. To generate cylinders of similar density, between 0.75 and 1 g of granules per cylinder were put into a hollow cylinder (Figure 2) and compressed with a uniaxial testing machine (ZwickRoell, Ulm, Germany) using a predefined compression regime (Figure 3). A compression speed of 0.5 mm/s was applied, and the compression stopped after reaching an end load of 1000 N for 5 min. Afterwards, the cylinders formed from the granules were taken out of the hollow cylinders and stored at room temperature until performing the unconfined uniaxial compression test.

#### *2.2. Unconfined Uniaxial Compression Test*

The unconfined uniaxial compression tests were conducted at room temperature using a uniaxial testing machine (Z050, ZwickRoell, Ulm, Germany) and a 2.5 kN load cell (Zwick-Roell, Ulm, Germany). A preload of 0.1 N was applied at a test speed of 0.05 mm/s, which was chosen based on previous studies, and which represents a physiological range [15,16]. The test runs were terminated at an engineering strain of 80%. The test setup is shown in

Figure 4. TBCs and BGCs that lost their axial alignment during the uniaxial compression test were discarded.

**Figure 2.** (**a**) Two segments of the hollow cylindrical body with a diameter of 6 mm. (**b**) Both segments screwed together to form a hollow cylinder. Bone granules were placed into the hollow cylinder and pressed together with a testing machine.

**Figure 3.** Predefined compression regime to compact bone granules to cylinders.

**Figure 4.** *Cont*.

(**b**)

**Figure 4.** Test setup for the uniaxial unconfined compression test for the bone granule cylinders (BGCs) (**a**) and trabecular bone cylinders (TBCs) (**b**) with a representation of the respective test specimens.

#### *2.3. Evaluation of the Results and Statistics*

For the TBCs, the stiffness and the yield strength (first stress maximum after linear behavior) were compared. For BGCs, the stiffness and the stress at 15% strain were compared because the stress–strain curves of the bone granule specimens did not exhibit local maxima due to the lack of an interconnected trabecular structure.

For all human bone specimens, the generated engineering stress–strain curves were analyzed using a self-developed MATLAB script (v. R2018a, MathWorks, Natick, MA, USA). The linear-elastic region of the stress-strain curves was automatically identified and used for calculation of the stiffness via regression. For TBCs, the yield strength was identified as the first local dominant maximum after linear behavior. For BGCs, the engineering stress at 15% strain was calculated as a comparable alternative to the yield strength. Additionally, all curves of each group were averaged using Origin (v. 2018b, OriginLab, Northampton, MA, USA).

Statistical analyses were done by one-way ANOVA tests using GraphPad Prism Version 7 (GraphPad Software, San Diego, CA, USA), and results are presented as box-andwhisker plots. *p*-values ≤ 0.05 were seen as significant.

#### **3. Results**

#### *3.1. Effects of HHP Treatment on the Mechanical Properties of Trabecular Bone Cylinders (TBCs)*

To evaluate the effects of HHP treatment on the mechanical properties of TBCs, samples were treated with HHPs of 250 and 300 MPa for 20 and 30 min each. The parameters of stiffness and yield strength were used for mechanical characterization. Results from samples that were tilted and/or slipped during testing or showed macroscopic defects were excluded. All results are shown as box plots in Figure 5 and summarized in Table 1. Additionally, a summary of the averaged stress-strain curves for all tested groups is shown in Figure 6.

Analyses showed no significant differences between the untreated and HHP-treated specimens (TBCs and BGCs), neither for stiffness nor yield strength. Comparing the different HHP magnitudes and durations applied to the specimens, no significant differences were determined within the treated groups. Considering the curves averaged within each group in Figure 6, it is shown that the courses of the stress–strain curves are similar. No effects of the HHP treatments can be observed in the stress–strain curves.

**Figure 5.** Analysis of stiffness (**a**) and yield strength (**b**) of trabecular bone cylinders (TBC) treated with and without high hydrostatic pressure (HHP). Mechanical properties were tested using a uniaxial compression test. Data are shown as box plots with median and interquartile ranges from 25 to 75%. Statistical analyses were performed using a one-way ANOVA. Sample size: control group (*n* = 20); 250 MPa, 20 min (*n* = 18); 250 MPa, 30 min (*n* = 19); 300 MPa, 20 min (*n* = 16); 300 MPa, 30 min (*n* = 14).

**Table 1.** Overview of the results, including sample size *n*, the mean and the standard deviation for trabecular bone cylinders after the uniaxial compression test for stiffness and yield strength.


**Figure 6.** Averaged stress-strain curves of the compressed TBCs.

#### *3.2. Compression of Granules to Cylindrical Samples*

Pressed granulate bone cylinders were prepared using the described setup and technique and resulted in a length between 8 and 12 mm. An example of the compressed granules can be found in Figure 7.

**Figure 7.** Bone granules with an average size of 1 to 2 mm (**a**). These were compressed to cylindrical samples using a hollow cylinder (**b**).

#### *3.3. Effect of HHP Treatment on the Mechanical Properties of Granules Bone Cylinders*

To assess the influence of HHP on the mechanical properties of the BGCs, stiffness and stress at 15% strain were chosen as comparative parameters. The results are shown in Figure 8 and in Table 2. The averaged stress–strain curves for granulated bone cylinders are shown in Figure 9.

**Figure 8.** Analysis of stiffness (**a**) and stress at 15% strain (**b**) of pressed bone granules treated with and without HHP. Mechanical properties were tested using a uniaxial compression test. Data are shown as box plots with median and interquartile ranges from 25 to 75%. Statistical analyses were performed using a one-way ANOVA. Sample sizes: control group (*n* = 7); 250 MPa, 20 min (*n* = 10); 250 MPa, 30 min (*n* = 10); 300 MPa, 20 min (*n* = 8); 300 MPa, 30 min (*n* = 6).

**Table 2.** Overview of the results, including the sample size *n*, the mean and standard deviation for BGCs after the uniaxial compression test for the stiffness and stress at 15% strain.


**Figure 9.** Averaged engineering stress-strain curves of pressed BGCs.

For BGCs, neither the stiffness nor the strain at 15% stress showed any significant differences between the groups. Figure 9 shows, as for the TBCs, the averaged stressstrain curves for granulate bone cylinders. Here, too, a similar course for all groups can be determined.

Comparing the stiffness of TBCs and BGCs, the latter comprises only a fraction of the native cylinders due to its missing intertrabecular structure.

#### **4. Discussion**

The reconstruction of bone defects is still challenging. In particular, cases that are correlated with severe bone loss or cases of patients with disorders in healing processes are clinically demanding [5]. Autologous bone is still considered to be the gold standard for bone defect reconstruction despite the known donor site morbidity and the limited amount of harvestable autologous bone [17]. A major drawback of the alternative, allogenic bone, which is less limited in quantity, is the alteration of mechanical properties due to current devitalization and sterilization methods, including thermal processing and gamma irradiation [18–20].

HHP as a gentle devitalization and sterilization method has been used in the food industry for several years, but it has also gained attention in the medical and pharmaceutical sectors [9]. Depending on the applied pressure, various studies have shown that mammalian cells can be devitalized while preserving an intact tissue matrix, which has already been shown for blood vessels and uterine tissues [12,13,21]. HHP has the potential to serve as a novel way to process allogenic bone substitute materials.

Within this study, it was shown that HHP had no effect on the macroscopic mechanical properties of human trabecular bone, as had already been shown for other tissues [13,21,22]. Specifically, HHP-treated TBCs showed no significant differences in either stiffness or yield strength. It was noticeable that all groups showed high variance in their mechanical properties, which is typical for biological samples, as gender and physical conditions of tissue donors can influence the results. Nevertheless, when looking at the averaged stress– strain curves of the individual groups with very similar curve progressions, it was shown that the eventual effects of HHP were small compared to the naturally occurring effects. The range for the compressive strength of trabecular bone is specified as 2 to 48 MPa according to the literature [7]. The measured strength of TBCs in this study was at the lower end of this range, at around 4 MPa for all groups. For this reason, the effect of HHP on trabecular bone specimens from other regions that have typically denser bone than the femoral condyles should be analyzed in further studies.

For pressed BGCs, no significant differences between the groups were found with stiffness and stress at 15% strain, i.e., no changes in the mechanical properties when comparing untreated and HHP-treated groups could be observed, but they were clearly below the values of the TBCs. The stress-strain curves of the BGCs also differ substantially from that of the TBCs. These curves exhibit a continuous, monotonic progression without local maxima due to the lack of an interconnected trabecular structure. Furthermore, this results in the significantly lower strength of BGCs when compared to native trabecular specimens. In addition, with these BGCs as well as the TBCs, it is noticeable that the values for both stiffness and stress at 15% strain vary widely. Many providers of bone substitute materials advertise both granules and bone blocks, which have been processed thermally or with gamma radiation [23,24]. According to various studies, these sterilization methods greatly reduce the mechanical properties of the allografts [19]. Commonly used irradiation doses between 20 and 30 kGy do not reduce the stiffness of bone, but they significantly reduce the ultimate stress and, to a smaller extent, the bending strength [18,25]. Thermal sterilization up to 60 ◦C has no effect on the mechanical properties, but higher temperatures (up to 100 ◦C) reduce the mechanical strength significantly [19].

There are several indications as to why HHP does not seem to affect the mechanical properties of treated biological tissues, which could explain the results observed in this study. An important role in the toughness of bone is played by collagen type I, which makes up the main part of extracellular matrix proteins that could be found in trabecular bone [26]. Pivotal to the mechanical properties of bone tissue is the formation of calcium-apatite crystals in the collagen fibrils interface [27,28]. This can also be seen in the correlation between calcification and bone stiffness [29]. Diehl et al. evaluated the effect of HHP on the biological properties of extracellular matrix (ECM) proteins and showed that HHP treatment had no influence on collagen type I and other common ECM proteins, such as fibronectin and vitronectin, in regards to their biological behavior in when compared to untreated ECM proteins [30].

These findings are supported by various studies at the molecular level [31,32]. Proteins are a complex organization of subunits with primary, secondary, tertiary and quaternary structures. The successive hierarchy describes increasing complexity of organization, which is affected by HHP treatment in different ways [31,32]. The primary structure (polypeptide chain) consists of covalent bonds, which are not affected by HHP. The secondary structure of proteins, created by the formation of hydrogen bonds between the polypeptides, is irreversibly degraded at pressures higher than 700 MPa. Tertiary structures, made up of hydrophobic interactions and ionic bonds, are broken up at HHPs higher than 200 MPa, and quaternary structures, with non-covalent bonds including Van der Waal's forces, are dissolved at HHPs between 100 and 150 MPa [31,32]. In contrast to the complete protein destruction that occurs at very high or low temperatures or during gamma irradiation, these structural changes caused by HHP are reversible at pressures between 100 and 300 MPa. This means proteins can rearrange after HHP treatment [18]. Based on this observation, some literature also describes a new association of previously incorrectly folded protein structures after HHP treatment [33–35].

In the present study, HHP between 250 and 300 MPa was applied, which induced reversible changes in protein structure. This could be the reason for the maintained mechanical properties of the specimens. However, our study is limited by several factors. As mentioned above, only TBCs from femoral condyles were studied. The same applies to BGCs, which were solely extracted from femoral heads. Furthermore, only one mechanical test was performed, and the specimen size did not reflect the influence of HHP on an entire femur. Additionally, only trabecular bone was analyzed in the study at hand, and the effect of HHP on cortical bone tissue should be analyzed in further studies. Another limitation is due to the use of biological samples; different tissue donors vary in physiological characteristics such as age, physical activity or pre-existing diseases. The structure and morphology of bone varies as well, which is reflected in the mechanical properties of whole bones and also bone tissue. A further limitation is contamination with bacteria and germs. The sterilization effect of HHD on these organisms was not the subject of this study.

The results presented fit well with the investigations described above. Further mechanical tests, such as a three-point-bending test, could be performed in addition to the uniaxial compression test already shown here. Samples composed of both cortical and trabecular bone tissue or whole bones could also be analyzed with regard to the effects of HHP on

their mechanical properties. This could give a good overall view of the influence of HHP on bone at the macroscopic level. Simultaneously, the effects of HHP at the microscopic and molecular levels should not be neglected. Here, a structural analysis of proteins after HHP treatment and an analysis of the inorganic bone components via electron microscopy is conceivable. If HHP is discussed as an alternative to the previous methods of processing allografts, the inherent mechanical properties are of great importance. With regard to clinical applications, the devitalizing efficiency and immunological safety should also be studied in the future. In further studies, it will still be necessary to assess the effects of HHP from a microbiological and virological point of view. Here, the study of bacterial and virological load before and after HHP treatment could be conceivable following previous works [36,37].

In the case of the BGCs, the compression method could also be optimized depending on the targeted clinical application. Loosely packed granules may well have an advantage for the ingrowth of cells, and more densely packed granules could possibly be used as a load-bearing structure. In addition, compression parameters or shapes (e.g., blocks) deviating from those shown here should be investigated.

As shown in the study, the clinical use of HHP-treated TBCs and BGCs is conceivable. Although the formed granulated bone cylinders are more fragile than native trabecular bone, granulate cylinders might be used as shaping filler material for non-load-bearing bone defects, acting as osteoinductive and osteogenic scaffolds.

In conclusion, this study showed that HHP treatment has a pivotal advantage over conventional processing methods of bone substitute materials by maintaining the mechanical properties in combination with effective cell devitalization.

#### **5. Patents**

A patent application with the number DE 10 2020 131 181.8 was submitted to the German Patent and Trademark Office.

**Author Contributions:** Conceptualization, J.W.-H. and M.S. (Michael Saemann); methodology, M.S. (Michael Saemann), M.S. (Marko Schulze); software, M.S. (Michael Saemann); formal analysis, J.W.- H.; investigation, J.W.-H. and M.S. (Michael Saemann); resources, R.B.; data curation, J.W.-H. and M.S. (Miachel Saemann); writing—original draft preparation, J.W.-H.; writing—review and editing, M.S. (Michael Saemann), M.S. (Marko Schulze), M.D., R.B., B.F.; visualization, J.W.-H., M.S. (Michael Saemann), M.D.; supervision, M.D., R.B.; project administration, R.B., M.D.; funding acquisition, R.B., M.D., B.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This joint research project HOGEMA is supported by the European Social Fund (ESF), reference: ESF/14-BM-A55-0012/18, and the Ministry of Education, Science and Culture of Mecklenburg-Vorpommern, Germany.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Rostock University and approved by the Institutional Ethics Committee Rostock University, Germany. Prior to preparation of human femoral condyles and femoral heads, ethical approval (A 2016-0083) including IRB information was obtained.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to the nondisclosure agreement with the project sponsor.

**Acknowledgments:** We thank Mario Jackszis for his excellent technical support during preparation and mechanical testing of bone specimens. We acknowledge financial support by Deutsche Forschungsgemeinschaft and Universität Rostock/Universitätsmedizin Rostock within the funding programme Open Access Publishing.

**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.

#### **References**


## *Article* **The Influence of Hyaluronic Acid Biofunctionalization of a Bovine Bone Substitute on Osteoblast Activity In Vitro**

**Solomiya Kyyak 1, Andreas Pabst 2, Diana Heimes <sup>1</sup> and Peer W. Kämmerer 1,\***


**Abstract:** Bovine bone substitute materials (BSMs) are used for oral bone regeneration. The objective was to analyze the influence of BSM biofunctionalization via hyaluronic acid (HA) on human osteoblasts (HOBs). BSMs with ± HA were incubated with HOBs including HOBs alone as a negative control. On days 3, 7 and 10, cell viability, migration and proliferation were analyzed by fluorescence staining, scratch wound assay and MTT assay. On days 3, 7 and 10, an increased cell viability was demonstrated for BSM+ compared with BSM− and the control (each *p* ≤ 0.05). The cell migration was enhanced for BSM+ compared with BSM− and the control after day 3 and day 7 (each *p* ≤ 0.05). At day 10, an accelerated wound closure was found for the control compared with BSM+/− (each *p* < 0.05). The highest proliferation rate was observed for BSM+ on day 3 (*p* ≤ 0.05) followed by BSM− and the control (each *p* ≤ 0.05). At day 7, a non-significantly increased proliferation was shown for BSM+ while the control was higher than BSM− (each *p* < 0.05). The least proliferation activity was observed for BSM− (*p* < 0.05) at day 10. HA biofunctionalization of the BSMs caused an increased HOB activity and might represent a promising alternative to BSM− in oral bone regeneration.

**Keywords:** bone substitute; bovine; xenograft; oral regeneration; biofunctionalization; hyaluronic acid; osteoblasts

#### **1. Introduction**

Presently, the demand for soft tissue and hard tissue regeneration is frequently increasing where bone is one of the most transplanted tissues because of a multitude of congenital or acquired diseases [1]. Nevertheless, the field of bone transplantation and regeneration faces limitations regarding infections, immunological reactions, failed osteointegration and graft resorption [2]. To avoid graft harvesting and to support a better and faster regeneration, numerous materials are combined to find suitable alternatives to autogenous bone grafts [3]. Bone substitute materials (BSMs) of a xenogeneic, an allogeneic and an alloplastic origin are well established and widely used as suitable alternatives in numerous fields of medicine [4–7]. In the range of craniomaxillofacial regeneration, BSMs can cover a wide variety of clinical indications such as alveolar ridge preservation and augmentation, sinus floor elevation and the bony reconstruction of congenital or acquired maxillofacial malformations and defects [8–11].

Xenogeneic BSMs of bovine origin are long-term established and widely spread [12]. The hydroxyapatite-based substance [13] is known for its biocompatibility, sufficient osteoconduction and low up to no resorption [14,15] and its similarity to human bone due to its microstructure [16,17] and crystalline phase [18]. In contrast to autogenous grafts, BSMs do not contain organic components such as osteogenic cells or growth factors such as BMP-2 (bone morphogenic protein-2) and a VEGF (vascular endothelial growth factor) and they also may not contain collagen structures and fibers, enabling an osteoconductive and inductive regenerative potential in autogenous grafts. Thus, different BSM preparation

**Citation:** Kyyak, S.; Pabst, A.; Heimes, D.; Kämmerer, P.W. The Influence of Hyaluronic Acid Biofunctionalization of a Bovine Bone Substitute on Osteoblast Activity In Vitro. *Materials* **2021**, *14*, 2885. https://doi.org/10.3390/ma14112885

Academic Editor: Franz E. Weber

Received: 27 April 2021 Accepted: 24 May 2021 Published: 27 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

methods and processes could affect the regeneration and surface characteristics of xenogeneic BSMs [19–22]. Accordingly, BSM sintering under a temperature >1000 ◦C seems to remove all organic compounds, thereby excluding an immune reaction and disease transmission and increasing crystallinity and volume stability [13,21,23–25]. Furthermore, it has been observed that even after a high temperature treatment, xenogeneic BSMs preserve their surface characteristics and a good biological performance [20–22,26]. Additionally, the carbonate content of high temperature treated hydroxyapatite stimulates human osteoblast (HOB) attachment and proliferation [27]. Nevertheless, it appears that BSMs may not be able to perform with an equal regenerative potency compared with autogenous grafts caused by the acellular and inorganic matrix. To overcome this limitation, BSM biofunctionalization has become more and more popular and has been tested in different ways. Recent studies analyzed combinations of BSMs with growth factors (e.g., BMP-2, VEGF) and PRF (platelet-rich fibrin). The findings of these studies illustrated that such biofunctionalized BSMs have the potency to accelerate and increase bone formation and vascularization as characteristic hallmarks of fast and sufficient bone regeneration [28–33]. As BSM modification with growth factors is technically challenging and restricted by legal requirements in most countries, further substances might be of interest for BSM biofunctionalization.

Hyaluronic acid (HA) is one of the largest components of the extracellular matrix. It is a long polysaccharide composed of macromolecules of many repetitive units of glucuronic acid and N-acetyl-glucosamine, remaining the same within all species [34–36]. It is stated that HA may regulate cell proliferation, differentiation, adhesion and gene expression [37]. These characteristics have aroused interest in HA in cutaneous research, cartilage grafting [38,39] and even bone reconstruction [34,40,41]. Thus, Kawano et al. reported that HA enhanced BMP-2 osteogenic bioactivity [35]. It has been discussed that HA retards bone resorption and osteoclast genesis through its receptor, CD44 [42]. HA may demonstrate lubricity under peculiar circumstances [43] and has been studied to have a bacteriostatic effect [44]. Sasaki et al. suggested that high molecular HA serves as a retainer for osteoinductive growth factors, thus stimulating osteogenic cell differentiation [45]. In addition, HA may positively influence angiogenesis and (neo-) vascularization because of its possible effects on endothelial cells, thus in turn indirectly stimulating new bone formation [45,46].

Different variations of HA molecules and their possible influence on tissue formation have been discussed. Guo et al. suggested that the molecular weight of HA strongly influences pro- and/or anti-inflammatory reactions of various tissues as far as peculiar angiogenic processes [47,48]. For example, Pilloni et al. observed that HA of a high molecular weight is dose-independent and not able to present any significant effects on bone formation [49]. However, further studies showed opposite findings [50]. This led to a significant interest in HA as an additive to different polymers and BSMs in bone engineering and regeneration.

Thus, the objective of this study was to analyze the influence of a commercially available BSM with (+) and without (−) HA biofunctionalization on viability, migration ability and the proliferation rate of human osteogenic cells. The zero hypothesis claims that this HA biofunctionalization has no influence on osteoblast activity.

#### **2. Materials and Methods**

#### *2.1. Bovine Bone Substitutes*

A commercially available xenogeneic bone substitute material (BSM−) of bovine origin (cerabone®, granularity: 1–2 mm; botiss biomaterials GmbH, Zossen, Germany) and a commercially available BSM with an HA modification (BSM+; cerabone® Plus, granularity: 0.5–1 mm; botiss biomaterials GmbH) were used.

#### *2.2. Cell Culture*

Commercially available human osteoblasts (HOBs) were applied in the present study (HOB; PromoCell, Heidelberg, Germany). A HOB medium was supplemented with Dulbecco's modified Eagle's medium (DMEM; Gibco Invitrogen, Karlsruhe, Germany), fetal calf serum (FCS; Gibco Invitrogen), streptomycin (100 mg/mL; Gibco Invitrogen), dexamethasone (100 nmol/L; Serva Bioproducts, Heidelberg, Germany) and L-glutamine (Gibco Invitrogen). The HOBs were cultured according to standard protocols in an incubator at 37 ◦C, 95% humidity and 5% of CO2. Reaching a 70% confluence, the HOBs were passaged using 0.25% trypsin (Seromed Biochrom KG, Berlin, Germany) until passage five. The plates were filled with 100 mg BSM+/− together with 5 × <sup>10</sup><sup>4</sup> HOB per well, respectively (27 wells per group, two groups). The plates with HOBs alone served as a negative control group (overall 27 wells). A further incubation was performed under the same conditions as by cell passaging. The measures were conducted in three time points in triplicate for each group and for each time point (days 3, 7 and 10; overall 81 wells).

#### *2.3. Cell Viability*

To analyze the HOB cell viability, CellTracker staining (Life Technologies, Thermo Fisher Scientific, Darmstadt, Germany; catalog number: C34552) was performed on days 3, 7 and 10. Red dye was prepared and used according to the manufacturer's protocol. After the removal of the culture media, red dye was added into the wells. After 30 min, the red dye was removed and a serum-free medium was applied. The wells were further incubated for 30 min at 37 ◦C. After the removal of the serum-free medium, a fluorescence microscope (BZ-9000; Keyence, Osaka, Japan) for cell imaging was used where one image per well in ten-fold magnification was conducted. The cell quantification was managed by means of ImageJ software (ACTREC, Navi Mumbai, India) [51] by the following steps: the conversion of the images into grayscale, the correction of the background by image subtraction, automatic thresholding for cell structure extraction from the background and the final calculation of the percentage area fraction (%). The measures were carried out in triplicate for each group and for each time point by three time points (on days 3, 7 and 10; overall 9 wells per group).

#### *2.4. Cell Migration*

The cell attachment was measured by means of a scratch wound assay. A scratch wound was performed at the bottom of the wells with a sterile pipet tip (p200; Gilson, Middleton, USA) on days 3, 7 and 10 [52]. Immediately after the scratch, a fluorescence microscope (BZ-9000; Keyence, Osaka, Japan) for cell imaging was used. Twenty-four hours later, red dye staining was obtained for preparing images with the aforementioned microscope (one image for each well, 9 wells per group, ten-fold magnification). An area of migrated cells into the gap was quantified by the percentage area (%) using ImageJ software as described before [51]. The measures were carried out in triplicate for each group and for each time point (three time points).

#### *2.5. Cell Proliferation*

The proliferation activity was measured by a 3-(4,5-Dimethylthiazol-2-yl) -2,5-dipheny ltetrazolium bromide (MTT) assay on days 3, 7 and 10. An MTT solution (200 μL, 2 mg/mL) was applied to the cell culture medium in the wells followed by 4 hours of incubation at 37 ◦C. After the removal of the culture medium and washing up by phosphate buffered saline, a lysis buffer (Isopranol (49 mL) with 2N NCl (1 mL; 1 mL per well) was added. The measurement was performed without the BSM in separate wells using a fluorescence microplate reader with a wavelength of 570 nm (Versamax; Molecular Devices, San Jose, CA, USA). The measures were carried out in triplicate for each group on days 3, 7 and 10 (overall 9 wells for each group).

#### *2.6. Statistics*

The mean values were interpreted into a standard error of the mean (SEM) in the cases of parametric data and into median values for non-parametric data. The numbers were rounded (to two decimal places). The normal distribution was defined by a Shapiro-Wilk

test. In the case of a normal distribution, to compare two subgroups a two-sided Student's *t*-test for paired samples was applied. In the case of non-normal distributions, a Mann-Whitney test was used. For a comparison of all subgroups, a Kruskal-Wallis rank sum test was performed. *p*-values ≤ 0.05 were considered to be significant. Data were illustrated with bar charts including error bars.

#### **3. Results**

#### *3.1. Cell Viability*

On day 3, the highest cell viability was observed for BSM+ when compared with BSM− (*p* = 0.028, *t*-test) and the control (*p* = 0.24, *t*-test). The cell viability of the control group was significantly higher than BSM− (*p* < 0.001, *t*-test) On day 7, the highest cell viability was seen for BSM+ compared with BSM− (*p* < 0.001, *t*-test; *p* < 0.05, KWT) and the control (*p* = 0.014, *t*-test; *p* < 0.05, KWT) followed by the control when compared with BSM− (*p* = 0.006, *t*-test; *p* < 0.05, KWT). At day 10, the cell viability of BSM+ was significantly higher when compared with the controls (*p* = 0.004, *t*-test) and BSM− (*p* = 0.002, *t*-test) (Table 1, Figures 1 and 2). Although the cell viability values for BSM+ were the highest of all groups through the whole period, the greatest tendency to increase was observed in BSM− in which the cell viability raised almost five times compared with BSM+ and the control with approximately two times (Figure 3a).

**Table 1.** Cell Viability. Percentage area fraction (%) of fluorescence-stained HOBs at a ten-fold magnification for BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3, 7 and 10. The mean values are for parametric data and the median values are for non-parametric data.


**Figure 1.** Fluorescence imaging (red cell tracker) in groups with BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3 (**A**–**C**), 7 (**D**–**F**) and 10 (**G**–**I**).

**Figure 2.** Cell Viability. Percentage area fraction (%) of fluorescence-stained HOBs at a ten-fold magnification for BSM− (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3, 7 and 10. \* = *<sup>p</sup>* <sup>≤</sup> 0.05, *<sup>t</sup>*-test; \*\*\* = *p* ≤ 0.0001, *t*-test; ## = *p* ≤ 0.05, KWT.

**Figure 3.** *Cont.*

**Figure 3.** Tendency through the period of 3, 7 and 10 days within groups BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus) and the control (HOB alone). (**a**) Cell viability, (**b**) migration ability, (**c**) proliferation rate.

#### *3.2. Cell Migration*

On day 3, the highest cell migration rate was found for BSM+ followed by BSM− and the control (each *p* > 0.05, *t*-test). On day 7, the highest value was observed for BSM+ (*p* < 0.05, KWT). The controls showed a significantly increased proliferation rate when compared with BSM− (*p* = 0.007, *t*-test). On day 10, the best wound closure was observed for the control followed by BSM+ and BSM− (*p* > 0.05 each, *t*-test) (Table 2, Figure 4). The migration ability in BSM+ increased from day 3 to day 7 by five and a half times and then decreased almost two times until day 10, being on day 10 almost on the same level with BSM− and the control group (Figure 3b).

**Figure 4.** Migration ability: Percentage area fraction of the scratch gap (%) of fluorescence-stained HOBs at a ten-fold magnification for BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3, 7 and 10. \* = *p* ≤ 0.05, *t*-test; ## = *p* ≤ 0.05, KWT.

**Table 2.** Migration ability: Percentage area fraction of the scratch gap (%) of fluorescence-stained HOBs at a ten-fold magnification for BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3, 7 and 10. The mean values are for parametric data and the median values are for non-parametric data.


#### *3.3. Cell Proliferation*

On day 3, the highest cell proliferation was observed for BSM+ in comparison with BSM− (*p* = 0.011, *t*-test; *p* < 0.05, KWT) and the control (*p* < 0.001, *t*-test; *p* < 0.05, KWT) followed by BSM− and the control (*p* < 0.05 each, KWT). On day 7, an increased proliferation rate was shown for BSM+ in comparison with BSM− (*p* = 0.019, *t*-test; *p* < 0.05, KWT) and the control (*p* < 0.05, KWT) while the control demonstrated increased values compared with BSM− (*p* = 0.046, *t*-test; *p* < 0.05, KWT). On day 10, the least proliferative activity was measured for BSM− (*p* > 0.05, MWT). Here, the highest proliferation rate was demonstrated for BSM+ (*p* > 0.05, MWT) (Table 3, Figure 5). The groups generally showed a tendency to increase up to day 7 and decrease until day 10. The highest raise rate was observed in the control on day 7. However, BSM+ stayed far on the top throughout the whole period (Figure 3c).

**Table 3.** Cell Proliferation: MTT assay, absorbance at 570 nm for BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and the control (HOB alone) on days 3, 7 and 10. The mean values are for parametric data and the median values (\*) are for non-parametric data.


**Figure 5.** Cell Proliferation: MTT assay, absorbance at 570 nm for BSM<sup>−</sup> (cerabone®), BSM+ (cerabone® Plus including HA) and Control (HOB alone) on days 3, 7 and 10. \* = *p* ≤ 0.05, *t*-test; \*\* = *p* ≤ 0.001, *t*-test; ## = *p* ≤ 0.05, KWT.

#### **4. Discussion**

This in vitro study analyzed the effects of HA in combination with commercially available BSMs of bovine origin on human HOB cell viability, migration ability and proliferation rate. The overall findings demonstrated a significant benefit of HA biofunctionalization of BSMs on the above-mentioned HOB cell features responsible for bone regeneration. In brief, the modification of bovine BSM with HA significantly increased the biological activity of HOBs in comparison with the same BSM alone. The cell viability presented a smooth increase through the whole period where BSM+ stayed distinctly the highest of all groups. HA additivity activated the migration ability on days 3, 7 and 10. The cell proliferation in its turn was significantly affected on day 7 and presented only a slight difference among groups on day 3 and day 10.

In a previous study, we evaluated different commercially available BSMs of bovine origin in regard to their biological effect on human HOBs. Here, the high temperature (>1200 ◦C) sintered bovine BSM, which was included in the present study whether alone or in combination with an injectable PRF, seemed to have the best effects on HOB cell viability, metabolic activity and gene expression of alkaline phosphatase (ALP), osteonectin and BMP-2 when compared with other BSMs of bovine origin prepared at lower temperatures [29]. Hence, the aforementioned BSMs of bovine origin commercially modified with HA or pure were included in the present study. According to our results, the combination of HA manufactured by bacterial fermentation and bovine BSMs presents an increase in the biological activity of HOBs in comparison with the same BSM alone. Accordingly, the cell viability in all groups presented a smooth increase throughout the whole period where they stayed distinctly the highest in groups with HA modification. Moreover, HA biofunctionalization activated the proliferation rate of HOBs on days 3, 7 and 10. The cell proliferation in its turn was significantly affected on day 7 and presented only a slight difference among groups on days 3 and 10. Our findings, that HA positively affects HOB bioactivity, were in accordance with other in vitro and in vivo studies although, to the best of our knowledge, there are no in vitro studies dealing with information about the effects of HA in combination with BSMs of bovine origin on HOBs. Kawano et al. concluded that HA enhances the osteogenic activity of HOBs in vitro via the down-regulation of BMP-2 antagonists and the phosphorylation of extracellular signal-regulated kinase [35]. Thus, chemically cross-linked hyaluronan-based hydrogels with HA and BMP-2 demonstrated cancellous bone formation in ectopic sites after five weeks [53]. HA functionalization of a titanium surface seems to enhance HOB proliferation and alkaline phosphatase activity [54,55]. Furthermore, HA has been studied to modify the composition of the extracellular matrix, affecting its fibrillary and non-fibrillar components [56]. Sasaki et al. suggested that HA acts as a detent for growth factors even enhancing HOB activity [45]. Interestingly, HA and its side groups happen to reduce bacterial adhesion and prevent biofilm formation [57].

In spite of intensive research in this area, there are no evident studies proving a HA-specific mechanism of interactions and pathways considering osteogenesis [58]. It has been reported that HA affects wound healing by enhancing the CD44 surface marker consequently activating early inflammation and cell migration into granulation tissue [59]. Due to the similarity with the extracellular matrix, HA seems to be biocompatible inducing a low immune response. Furthermore, it accelerates cell adhesion, migration and proliferation and, as a result, to some extent new tissue formation [60]. However, HA presents a low mechanical strength and a high degradation rate, thus being limited and requiring appropriate modifications [61]. A combination of HA with gelatin and alginate into a three-dimensional composite scaffold showed to be high load bearing without fractural deformation [62]. Mathews et al. presented a scaffold with a chitosan-collagen-HA ratio of 1:1:0.1 in which lower HA concentrations and more uniform pores seemed to enhance HOB differentiation-promoting effects [63]. Furthermore, HA appears to be capable of encapsulating bioactive factors by cross-linking [64,65]. Nevertheless, the general process of HA cell bioactivation, due to its complexity, especially including osteogenesis, is still unclear [58]. Presently, the effect of HA as an enhancer of the biological properties of a

synthetic scaffold, an activator of osteogenesis and as a vector for osteoinductive substances is approved [66].

It is known that HA combined with BSM+ is of a bacterial origin non-cross linked high molecular weight hyaluronic acid (h-HA) with a molecular mass of 1.9–2.1 MDa. However, it belongs to the limit of our study that the amount of it added to the BSM was not given. It has been reported that the molecular weight of HA is greatly decisive regarding the effect on the biological activity of cells and pro-inflammatory characteristics [45,67]. However, there is no consensus in the literature regarding the ideal constitution and concentration of HA for better bone regeneration [68]. Thus, Boeckel et al. observed a decrease in HOB viability under presence of HA and referred this not to chemical composition but rather to the molecular weight of HA [68]. The same findings were found in other studies [49,69,70]. Hence, it was suggested by that h-HA positively alters the cellular parameters of HOBs and influences peculiar inflammatory mediators, acting as an adjustor of HOB biological capacities [71]. Furthermore, Agarwal et al. demonstrated that h-HA in comparison with a low molecular one (l-HA) presented a significantly increased osteogenic differentiation of HOBs based on an upregulation of ALP, collagen and EM mineralization as well as the effects of l-HA, in its turn, on HOB proliferation and adhesion [50].

Another limitation of our study was sample staining using the Cell Tracker 5 chloromethylfluorescein diacetate for cell viability, which also permeates dead cell membranes. However, stained live cells are >100-fold brighter than dead cells and could be easily distinguished from the dead population [72]. It also belongs to the limits of the study that bone substitutes of two different particle sizes were compared: 0.5–1 mm versus 1–2 mm. However, the difference was not significant and may not have affected the results [73–75]. The critical difference was studied to be between the particles of less than 0.4 mm and more than 1 mm [73]. However, another study contradicted this statement, concluding that particles of 0.1–0.3 mm and 0.5–0.7 mm were not significantly different in terms of their osteogenic potential [74]. Another study suggested that the granularity was not of a significant relevance but was rather dependent on the clinical defect size. It seems that the microstructure characteristics of the material rather than its granularity plays an important role [75].

The implementation of HA in combination with bone substitute materials may be very promising to overcome any limitations in the soft and hard tissue regeneration. HA modified BSMs have the advantage of being classified as commercially available medical devices ready to use. Further in vitro and in vivo studies of HA in combination with BSMs of different origins will carve out the significance of dosage and the molecular weight of HA in bone engineering as far as there are no specific mechanisms of interactions and pathways considering HA involved in osteogenesis [58]. Clinical trials will focus on visible benefits such as the bone regeneration capacity and long-term stability in vivo.

#### **5. Conclusions**

HA biofunctionalization of BSMs enhancing the viability, migration ability and proliferation rate of human osteogenic cells on days 3, 7 and 10 might be able to accelerate and improve bone regeneration and might represent a promising alternative to native BSMs.

**Author Contributions:** Conceptualization, S.K. and P.W.K.; methodology, S.K. and P.W.K.; software, S.K. and P.W.K.; validation, S.K. and P.W.K.; formal analysis, S.K., D.H. and P.W.K.; investigation, S.K. and P.W.K.; resources, P.W.K.; data curation, S.K., D.H. and P.W.K.; writing—original draft preparation, S.K., A.P. and P.W.K.; writing—review and editing, S.K., A.P. and P.W.K.; visualization, S.K. and P.W.K.; supervision, P.W.K.; project administration, P.W.K.; funding acquisition, P.W.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** Botiss biomaterials GmbH kindly provided the bone substitute material for the research.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data are available on request.

**Conflicts of Interest:** The authors A.P. and P.W.K. received speaker fees and research support from botiss biomaterials GmbH and Straumann AG for other studies. This had no influence on the current study. For this study, free samples of cerabone® and cerabone® Plus were received from botiss biomaterials GmbH.

#### **References**

