5.1 Aims and Study design
The present pilot study explored the tissue behavior and cell survival capability within a new in vitro skeletal muscle tissue-based biomaterial organoid bioreactor.
Eighteen 3D printed β-tricalcium phosphate/hydroxyapatite devices were either wrapped in a sheet of rat muscle tissue (n = 9) or first implanted in a heterotopic muscle pouch (n = 9) that was then excised and cultured in vitro for up to 30 days. Normal muscle tissue without implants, uncultured, served as the endogenous control to which all samples were compared to. Specimens were harvested at 5 days, 15 days and 30 days (n = 3 per time point), respectively and underwent qRT-PCR and histological analyses. Supernatants of tissue cultures were assayed for angiogenic/vasculogenic protein production, fresh medium was the control.
5.2 3D printed β-tricalcium phosphate/hydroxyapatite devices (β-TCP/HA) devices
Eighteen devices were provided by BioMed Center Innovation gGmbH (Bayreuth, Germany). According to the BioMed Center, the 3D-printed β-TCP/HA bioceramic devices (Figure 1A) had been manufactured using a mixture of tri-calcium phosphate and hydroxyapatite powders (Merck, Kenilworth, NJ, USA)) at a ratio of 40%:60%, respectively. The mixture had previously been spray-nozzle granulated from a water-based slurry with addition of organic dispersing and binding agents using a custom spray-dryer (Trema, Kemnath) and cut off at 100 µm using a classing sieve (Retsch, Haan, Germany). The lower fraction of the granulate was coated with organic adhesion-improving agents by means of fluidized bed coating; the final printing powder had size distribution values of d10 = 34.87 μm, d50 = 61.86 μm and d90 = 93.33 μm. After mixing the powder with a combination of organic additives (trade secret), the scaffolds were then printed out in a Z310 3D-Printer (3D Systems, Rock Hill, USA) using the standard colorless ink provided with the printer. After de-powdering, the scaffolds were sintered at 1250°C, producing a solid, organic-free, porous bioceramic device with macroscopic pore channels (670.52 +/- 97.60 µm) resulting from printing design and smaller internal pores (80.95 +/- 23.38 µm) as described above. The devices were then allowed to cool, after which they were cleaned using deionized water, packed and sterilized by vacuum pulse autoclaving.
5.3 Skeletal muscle-based biomaterial culturing models
Commercially available, four adult male Rattus norvegicus Fischer 344/DuCrl rats (Charles River Laboratories, Sulzbach, Germany), were utilized in the pilot study, and equally split between the two tissue models. Animals were euthanized with an overdose of isoflurane (Abbot, Chicago, USA). This was done in accordance to the rules and regulations of the Animal Protection Laboratory Animal Regulations (2013), European Directive 2010/63/EU and approved by the Animal ethics research committee (AESC) of the Ludwig Maximillian’s University of Munich (LMU), Bavaria, Germany Tierschutzgesetz §1/§4/§17 (https://www.gesetze-im-internet.de/tierschg/TierSchG.pdf) with respect to animal usage for pure tissue or organ harvest only.
Two skeletal muscle tissue biomaterial-based models were designed and tested:
5.3.1 Tissue wrapping model
For the tissue wrapping model, n=9 β-TCP/HA devices, were first immersed in normal growth medium composed of Dulbecco’s modified Eagle medium–high glucose (DMEM-hg) (Biochrom GmbH, Berlin, Germany), 40 IU/mL penicillin (Biochrom GmbH) and 40 IU/mL streptomycin (Biochrom GmbH).
Two F-344 adult male rats (Charles River) were euthanized under sterile conditions, the abdominal skeletal muscle tissue harvested, placed in normal DMEM-hg after which 3D printed β-TCP/HA devices were wrapped in the sheets of muscle tissue (Figure 1B-F). Nine β-TCP/HA devices were then wrapped with a skeletal muscle sheet, and divided into 3 culturing periods set at 5, 15 and 30 days. Each culturing period contained 3 tissue bags. Muscle tissue without β-TCP/HA devices was cultured in parallel to tissue bags and acted as controls. Medium was changed every 2 days. Fresh muscle tissue was used in the normalization of qRT-PCR.
5.3.2 Tissue pouch model
Nine β-TCP/HA devices were prepared by placing them in normal growth medium as explained in the section of the tissue wrapping model. Rats were then euthanized under sterile conditions, β-TCP/HA devices were immediately implanted in intramuscular pouches created by sharp and blunt dissection (Figure 1G-J). Once all β-TCP/HA devices had been implanted, muscle tissue pouches with biomaterials were excised using 8 mm biopsy punches (PFM medical, Cologne, Germany). Nine muscle pouches with β-TCP/HA were created, and divided into 3 culturing periods set at 5, 15 and 30 days. Each culturing period contained 3 tissue pouches. Muscle tissue without β-TCP/HA devices were cultured in parallel to tissue pouches and acted as controls. Medium was changed every 2 days. Fresh muscle tissue was used in the normalization of qRT-PCR.
After the allotted culturing period, specimens with β-TCP/HA devices were harvested and cut midways, with one-half flash frozen in liquid nitrogen for qRT-PCR assays and the other half fixed in 4% paraformaldehyde (Microcos GmbH, Garching, Germany) to be processed for histological and histomorphometric analysis.
5.4 Bacterial contamination assay
The 30-day organoid pouch model devices, during histological analysis, were observed containing a filamentous fibrous-like material. To exclude the likelihood that the fibrous like material was not of bacterial origin and in fact fibrin fibrils, the medium collect for quantitative protein analysis from these samples was subjected to a bacterial contamination test. Under sterile conditions collected culture medium was plated out on a standard Luria Broth Agar (LA) plates (1g Tryptone, 1.5g Technical agar, 0.5g Yeast extract, 0.5g NaCl (all (Sigma-Aldrich)) in 100ml dH2O), with a normal LA plate with fresh DMEM-hg (Biochrom GmbH) medium set as control. After 72 hours of incubation at 37 °C with 5% CO2, plates were assessed for bacterial colony formation by one blinded analyst (Yan Chevalier).
5.5 QRT-PCR
QRT-PCR was performed to determine the relative gene expression quantity of tissue growth related genes especially angiogenesis and endothelial tissue formation genes, VEGF-A, COL4A1 and TGF-β1 including known osteogenesis signaling and structural markers, specifically RUNX-2 and BMP-2.
Specimen fragments for qRT-PCR were ground to powder in the presence of liquid Nitrogen. Total RNA was then isolated using a modified RNA Trizol extraction procedure (Chomczynski & Mackey, 1995). Briefly, 1 ml Trizol (Invitrogen, San Diego, CA, USA) was added to the powderised tissue, where through the addition of chloroform (Sigma-Aldrich) the aqueous RNA containing phase was transferred to Isopropanol (Sigma-Aldrich). RNA was then pelleted out in an overnight centrifugation step at 4 °C, which were then washed with 75% ethanol dried and resuspended in 32 μl RNase free water. The concentration of the RNA was determined using a NanoDropTMLite (Thermo Scientific, Waltham, USA) and quality assessed with a Bioanalyzer 2100 (Agilent Technologies, CA, USA). RNA integrity numbers lower than 8 were not accepted. RNA was then reverse transcribed into complementary DNA (cDNA) using the QuantiTect Reverse Transcription cDNA Synthesis Kit (Qiagen, Hilden, Germany).
QRT-PCR was then performed, in duplicate with FastStart Essential DNA Green Master (Roche, Basel, Switzerland) in a final reaction volume of 10 µl, using a LightCycler® 96 thermocycler (Roche). Each reaction contained 10 ng cDNA; 2x FastStart Essential DNA Green Master and 10 µM of each primer (Table 1). Primers were designed using Integrated DNA Technologies PrimerQuest Tool (https://eu.idtdna.com/Primerquest/Home/Index). Use of GeNorm (http://medgen.ugent.be/~jvdesomp/genorm/) established that ribosomal protein large P0 (RPLP0), succinate dehydrogenase complex subunit A (SDHA), RNA polymerase II subunit E (POLR2E) and TATA binding protein (TBP) were the most appropriate internal reference genes to use in this experiment. All amplified PCR products underwent Sanger sequencing (GATC Biotech, Cologne, Germany) and were then analyzed utilizing nucleotide analysis (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch) to confirm that the correct sequence had been amplified. QRT-PCR thermocycling parameters included a pre-incubation of 3 min at 95°C, followed by a three-step amplification program of 40 cycles consisting of a denaturation, annealing and extension step set at 95°C for 10 s, 60 °C for 15s and 72°C for 30s, respectively. Relative gene expression was normalized against four reference genes. Gene expression from the harvested tissue/device models was normalized to the four reference genes and fresh abdominal skeletal muscle tissue using the Qbase+ software (http://www.biogazelle.com). Gene expression results were represented as mean calibrated normalized relative quantities (CNRQs) ± standard error, which reflect the log10 2-ΔΔCt.
5.6 Histological evaluation
Specimens were fixed in 4% paraformaldehyde (Microcos GmbH) for 24h after which they were processed for paraffin wax embedding. Prior to cutting 10μm sections the surface of each paraffin block was decalcified [52]. In order to validate our gene expression patterns with respect to tissue survivability within the two tissue models, histological sections were stained using either the hematoxylin (Morphisto GmbH, Frankfurt, Germany) and eosin (H&E) staining [53] (Morphisto GmbH) or the Movat pentachrome staining [54] (Morphisto GmbH). Stained sections were subsequently analyzed under PreciPoint M8 microscope (PreciPoint, Freising, Germany).
5.7 Quantitative angio-/vasculogenic protein assays
The amount of VEGF-A produced by the two bioreactors and controls were determined using Magnetic Luminex® Assays (R&D systems, Minneapolis, USA). Supernatants of tissue cultures were harvested at 5 days, 15 days and 30 days for either the wrapping model specimens or the pouch model specimens and controls. VEGF-A contents in supernatants were measured according to the manufacturer’s instructions. Results were generated using xPONENT® 4.2 for MAGPIX® Software (R&D systems, Minneapolis, USA).
5.8 Statistical analysis
Data were analyzed using GraphPad Prism v8.0.1 (GraphPad Software, San Diego, USA). The results were represented as mean ± standard error (SEM). Measurements were performed in triplicate or duplicate when no valid data could be obtained from one sample per group. The Mann-Whitney test was used to detect statistical differences with α = 0.05. Statistical significance was indicated by ns for no significance, * for p<0.05, ** for p<0.01 and *** for p<0.001.