3D printed hydrogel scaffold promotes the formation of hormone-active engineered parathyroid tissue

The parathyroid glands are localized at the back of the thyroid glands in the cervical region and are responsible for regulation of the calcium level in the blood, through specialized cells that sense Ca2+ and secrete parathyroid hormone (PTH) in response to a decline in its serum level. PTH stimulates the skeleton, kidneys and intestines and controls the level of Ca2+ through specialized activities. Iatrogenic removal of the parathyroid gland, as well as damage to its vascular integrity during cauterization are some of the common complications of thyroid surgery. Therefore, regeneration and/or replacement of malfunctioning parathyroid tissue is required. Tissue engineering is an emerging and promising field for patients with organ failure with recent pioneering clinical applications. The success of tissue engineering strategy depends on the use of proper cells, bioactive factors that stimulate the activities of these cells and scaffolds that are produced to recapitulate the tissue structure and support the function of the engineered tissues. 3D printing is a developing strategy for the production of these scaffolds by providing a delicate control over their structure and properties. In this study, human primary parathyroid cells were successfully isolated and their viability and ability to secrete PTH upon stimulation with different levels of Ca2+ were shown in vitro. These cells were then seeded onto 3D printed alginate scaffolds and 3D bioprinted within alginate bioink, and cell viability as well as the ability to secrete PTH upon stimulation were also demonstrated. Therefore, functional hormone-active parathyroid tissue substitute was engineered in vitro through 3D printed hydrogels and autologous cells.


Introduction
Parathyroid glands are members of the endocrine system, and they are responsible for regulating the Ca 2+ level in the blood through secretion of parathyroid hormone (PTH) [1]. Specialized cells in the glands detect the level of serum Ca 2+ via Ca-sensing receptors and they either secrete or shut down the secretion of PTH to maintain the homeostasis. For example, when the serum Ca 2+ concentration is lower than 10 mg dl −1 , PTH secretion is activated and reaches to the highest rate when serum Ca 2+ concentration is ca. 7.5 mg dl −1 . When the Ca 2+ concentration is in the physiological range (10 mg dl −1 ) or higher, then PTH secretion is suppressed [2]. This cycle is vital for the physiological activities and also for the function of various organs including the kidneys and the skeleton [3,4]. For example, PTH promotes bone resorption and delivery of both Ca 2+ and P to the extracellular fluid while it affects the kidneys and increase renal reabsorption of Ca 2+ and reduces P reabsorption [2].
Parathyroid glands are localized at the back of the thyroid gland and vary in between 3 to 4 mm in size. Due to their very small size and indifference of color and/or texture with the thyroid, their iatrogenic removal or damage to their vascular integrity during cauterization in surgical procedures are common complications. These conditions cause loss of function of the parathyroid glands to produce PTH, which require lifelong hormone replacement therapies or parathyroid transplantation [5][6][7]. Major drawbacks of transplantation procedures include histo-incompatibility of donor organs as well as their scarcity.
Tissue engineering approach provided an alternative for transplantation procedures to form viable and functional tissues and organs in the laboratory conditions. This technology combines principles of health sciences and engineering to regenerate tissues/organs by using autologous cells and biomaterials [8,9]. Engineering functional parathyroid tissue has its own specific challenges as well as certain advantages. When compared to other tissues, engineered parathyroid is more prone to transplantation since the damage in the defect site often mechanical due to surgical complications, instead of other underlying more complicated mechanisms. However, a major challenge is the culture of parathyroid cells in a functional form in the in vitro conditions [10,11]. Probably due to this major challenge, there are a few studies in the literature on parathyroid tissue engineering. Considering these facts as well as the vital function of the organ, it is important to develop strategies for engineering of functional parathyroid tissue, which is attempted in this study.
It is well known that mimicking the natural microenvironment is an important strategy for functional tissue engineering and scaffold design and modification is a major strategy towards that goal [12]. To produce scaffolds with desired geometries and topologies, various biofabrication techniques were applied [13]. Gas foaming, freeze drying, solvent casting, electrospinning are the some of the traditional techniques in which pore shape, size and/or distribution cannot be easily controlled and predetermined shapes may not be formed [14]. On the other hand, complex microstructures and shapes, personalized scaffold design with controlled pore structure and topology can be produced rapidly and with high accuracy through 3D printing [15][16][17][18]. 3D bioprinting applies bioinks to create 3D structures with desired and controlled characteristics such as external shape, pore geometry and uniform cellular distribution in 3D [12,14]. 3D cell-laden structures are printed with a layer-by-layer approach to form viable tissue substitutes [12,19,20].
There are various approaches for 3D bioprinting according to different deposition methods employed; such as inkjet, extrusion-based, and laser-based bioprinting systems [18,[21][22][23][24]. Micro droplets are used to produce the desired structure in inkjet or droplet-based 3D bioprinting approaches [12]. This method is very similar to traditional 2D inkjet printing which uses a cartridge and is especially preferred in cases where high resolution 3D printing of complex structures is desired. However, biomaterials applicable with these systems are limited due to the need of fast gelation. Moreover, production of the large organs starting from droplets could become an unpractical process [24,25]. Extrusionbased bioprinting is another approach which relies on a computer-controlled fluid dispensing system, where the dispending system controls the extrusion and the robotic system controls the 3D organization for fabrication [12,26]. In extrusion-based systems, pneumatic pressure or mechanical pistons aid the extrusion of the cell-laden or cell-free biomaterials through a syringe tip. Printability and the resolution of the product is highly dependent on the rheological properties of the bioink used. The bioink should support the stability of the print following extrusion from the syringe tip and post-processing strategies including cooling, cross-linking etc are generally required [27]. When compared to droplet-based strategy, a more continuous filament formation is possible with extrusion-based biofabrication with relatively lower resolution, and high cell densities can be 3D printed within various different compositions of bioinks [12,24,25,28].
Based on these, the aim of this study was to use an extrusion-based 3D printing system to produce engineered parathyroid tissue by the use of autologous cells within a 3D printed biodegradable hydrogel matrix to mimic natural tissue characteristics. Development of a strategy to maintain functionality of human primary parathyroid cells which are responsive to stimulation with different doses of Ca 2+ in the 2D as well as in the 3D culture was achieved. These results will aid in the generation of engineered, hormone-active, fully-functional parathyroid tissue highly demanded for clinical applications.

Cell isolation and culture
Human parathyroid tissue was obtained in accordance with the Clinical Research Ethics Committee approval (No: 2019 60/02, Dışkapı Yıldırım Beyazıt Research and Training Hospital) from patients who have undergone thyroidectomy/parathyroidectomy. Patients who underwent unilateral thyroid surgery for benign reasons were included in the study. In case of intraoperative iatrogenic parathyroid devascularization of the patients, the tissue was autoimplanted between the sternocleidomastoid muscle fibers of the patient, and a maximum of 10% was separated for use in our experimental study. Tissue was implanted after the confirmation that the excised tissue was parathyroid by histopathological examination from the cryosections.
Of the nine patients who agreed to be included in the study, two were male and seven (77.7%) were female. One patient was excluded from the study due to diagnosis as papillary thyroid carcinoma in the postoperative pathology examination. The age of patients included in the study ranged from 37 to 56, with a mean of 44.12 ± 6.6. Postoperative pathological evaluation revealed seven nodular hyperplasia and one follicular adenoma. In the preoperative laboratory evaluation of the patients, mean PTH and calcium values were calculated as 20.21 ± 3.1 pg ml −1 and 9.33 ± 0.63 mg dl −1 , respectively. No clinical and/or laboratory hypocalcemia and hypoparathyroidism were encountered in the 18 month followup of the patients after discharge.
Set aside tissue was transferred to the laboratory in an ice-cold RPMI-1640 (Biological Industries) supplemented with 1% penicillin/streptomycin (P/S, Biological Industries). It was then rinsed with phosphate buffered saline (PBS) and removed from gland capsule, connective tissues, visible fat and blood vessels under sterile conditions. Tissue was minced with a sharp lancet and treated with 15 ml of digestion solution composed of 1.5 ml collagenase type II (Sigma Aldrich), 500 µl DNAse I (Serva) within PBS at 37 • C for 1 h with gentle agitation. The suspension was then transferred into fresh falcon tubes, rinsed with PBS and centrifuged at 3500 rpm for 10 min. Percoll (Sigma Aldrich) solution was added onto the resuspension (75% in PBS (v/v)) to fragment and remove the dead cells, debris and blood cells following centrifugation. Culture media composed of RPMI-1640 supplemented with 10% fetal bovine serum (FBS) (Biological Industries) and 1% P/S was used to rapidly resuspend the pellet. The suspension was filtered into a fresh falcon tube through a 40 µm sterile cell strainer (Greiner bio-one). Filtered solution was rinsed with PBS and centrifuged at 3500 rpm for 10 min. After centrifugation, the pellet was resuspended within fresh culture media and the cells were transferred into flasks (or wells) and incubated under standard culture conditions and used at passage 1 for further studies.

3D printing and bioprinting
Alginate (Sigma Aldrich) was selected as the base hydrogel ink material due to its high printability as well as appropriateness to mimic the natural microenvironment of the cells that have low anchorage dependency [29]. Various concentrations of alginate (4%, 6% and 8% w/v) were used to optimize printability and cell viability. Scaffolds were 3D printed on an Envisiontec 3D Bioplotter equipment through a 25G nozzle. CaCl 2 (0.25 M, 0.5 M, 1.5 M) was used to crosslink the 3D structures.
3D printed alginate scaffolds were sterilized with 70% EtOH treatment and rinsed vigorously with PBS prior to cell seeding. Parathyroid cells were suspended in culture medium and ca. 60 000 cells were seeded on each scaffold. Scaffolds were incubated for 45 min to ensure cell attachment prior to feeding with culture medium.
Parathyroid cell-laden alginate scaffolds were also bioprinted for immunocytochemistry analysis. Alginate powder was sterilized with UV light for 30 min prior to use. Parathyroid cells were mixed within 4% w/v alginate hydrogel to prepare the bioink under sterile conditions. Scaffolds were 3D bioprinted with an Axo A2 bioprinter system (Axololt Biosystems) at 37 • C under sterile conditions.

Mechanical testing
Mechanical properties of the scaffolds were determined by uniaxial tensile testing using a universal testing machine (Shimadzu AGS-X). Tensile testing was applied on samples with dimensions of length: 15 mm, width: 5 mm and thickness: 0.5 mm (n = 5) and samples were elongated under 5 kN load at a speed of 1 mm min −1 . The applied load vs. strain of the samples were recorded as a function of time. Young's Modulus (E) values of scaffolds were calculated.

Cell viability, morphology and immunocytochemistry
Viable cell number was assessed with Alamar blue cell proliferation assay (Thermo Fisher) at days 3, 7 and 14. Briefly, samples were rinsed with PBS and then Alamar Blue solution (10% v/v) was added on the wells and incubated at 37 • C for 90 min in dark. After 90 min, solutions were collected in 96 wells and examined with ELISA plate reader. All samples were studied in triplicate.
After being cultured for 3, 7, and 14 d, samples were stained with Alexa Fluor 488 Phalloidin /DAPI (Thermo Fisher) to evaluate cytoskeletal organization of the cells in both 2D and 3D environment. For 2D observations, cells were cultured on FBS coated cover glass. 3.5% paraformaldehyde was used for cell fixation. After fixation procedure, both 2D and 3D samples were rinsed with PBS and then treated with Triton-X to disintegrate the cell membrane for 5 min. Then, wells were rinsed twice with PBS. To block nonspecific binding of the dye, bovine serum albumin (BSA) (1% (w/v) in PBS) was used and was treated with the sample for 40 min. After removal of the samples, Phalloidin stain was added and incubated with the samples for 1 h with protection from light. To counterstain the nuclei, for 5 min at room temperature DAPI was used. After rinsing with PBS, samples were stored at +4 • C in PBS and was examined with a fluorescence microscope.
Characterization of cells in 3D environment was made by immunofluorescence staining for CaSR on day 3. After fixation procedure, 3D scaffold were incubated with primary antibody for CaSR (1:100) diluted in 0.1% BSA serum at 4 • C overnight. Afterwards, samples were incubated with fluorochrome conjugated secondary antibody (Alexa Fluor ® 488-conjugated goat anti mouse IgG (1:200)) for 4 • C overnight. Nuclei were counterstained with DAPI and visualized on a Zeiss Fluorescent microscope.

Ca 2+ response test and measurement of PTH levels in 2D and on 3D culture
Stimulation with Ca 2+ at 0.5 mM, 1.25 mM, and 3 mM concentrations were used to test the hormoneactivity of parathyroid cell culture on days 1, 3, 5, 7 and 14 for both 2D and 3D culture environment. To stimulate the cells, Ca 2+ solutions were incubated with the cells for 1 h. The media were then collected and stored at −20 • C until analysis. The released PTH to the media were analyzed via the corresponding PTH ELISA kit.

Histopathological analysis
Paraffin embedded tissue sections of 4 µm thickness were processed with tissue fixation and dehydration. The Mouse anti-rat monoclonal PTH antibody (1:100; Cell Marque USA) staining on the BenchMark XT platform with Ventana's detection systems. The expression of PTH was observed under a microscope (Nikon E600).

Statistical analysis
All quantitative results were expressed as means ± standard deviation (n > 3). Data was analyzed with statistically significant values defined as p < 0.05 based on one-way analysis of variance followed by Tukey's test for determination of the significance of difference between different groups (p ⩽ 0.05).

Cell isolation, characterization and culture on tissue culture plastic
Parathyroid cells were isolated via enzymatic digestion of the tissue. Plated cells were examined under light microscope and cuboid and spherical-like cellular morphology was observed ( figure 1(A)). Cells proliferated and increased in number throughout the 14 d of culture on tissue culture plastic (TCP) ( figure 1(B)). It was observed that while the morphological features resemble that of a typical cuboid parathyroid cell on 3 on the TCP, the dominant cellular morphology was fibroblastic (spindle-shaped actin filament organization) on days 7 and 14 on TCP ( figure 1(C)).

Scaffold preparation with 3D printing of alginate
Alginate scaffolds were prepared with 3D printing using different hydrogel formulations. Printability and structural stability of the scaffolds after 3D printing were assessed, and scaffolds prepared with 8% (w/v) alginate ink selected due to highest structural stability and print fidelity (figure 2, left panel) among 4%, 6% and 8% w/v alginate solutions prepared within dH 2 O. Alginate concentration was not further increased in order not to negatively interfere with the cell viability, since it is known that higher alginate concentrations may reduce cellular viability and sprouting [30] Concentration and duration of crosslinking of alginate with CaCl 2 solution was also studied by observation of the stability of the 3D printed structures ( figure 2, right panel). It was observed that while use of 0.25 M CaCl 2 did not reveal stable structures post-printing (data not shown), and use of CaCl 2 at a concentration of 1.5 M resulted in very stiff structures that might interfere with cell viability. Therefore, use of 0.5 M CaCl 2 for 10 min was selected as the optimal condition for 3D printing of alginate scaffolds.
Scaffolds were 3D printed with the abovementioned optimized parameters with polygonal internal architecture (figure 3(A)), since parathyroid chief cells are also known to possess polygonal morphology. Scaffolds were printed at 3 layers at 37 • C, 4.8 bar pressure and 3 mm s −1 print speed. It was observed that alginate scaffolds were 3D printed with high print fidelity as well as high structural stability, with preserved structural integrity upon drying and re-wetting ( figure 3(B)). Mechanical properties of 3D printed 8% (w/v) alginate scaffolds were examined by tensile testing at days 0 and 9 after printing, cultured within PBS with no cells seeded. Stress-strain graphs were obtained and stiffness of the 3D printed hydrogel structures were calculated from these graphs (figure 3(C)). It was observed that the stiffness of the samples increased almost nine-fold during culture within PBS.

Cell culture on 3D printed alginate scaffolds
Isolated primary human parathyroid cells were seeded onto the sterilized, 3D printed alginate scaffolds. It was observed by light microscopy that cells populate mostly within the pores of the scaffolds 3 d after cell seeding ( figure 4(A)). Determination of the viable cell number within the scaffolds with Alamar blue assay revealed that although the cells had a similar proliferation rate between days 3-7 on TCP and within the scaffolds, the proliferation rate ceased and decreased after day 7, until day 14 ( figure 4(B)). Cellular morphology on the scaffolds was examined via staining of actin filaments with FITC-labelled Phallodin, and counterstained with DAPI at days 3, 7 and 14 of culture. Compared to the cellular morphology on days 3, 7 and 14 on TCP and on the scaffolds (figures 1(C) and 4(C)), it was observed that cells maintained their globular morphology on the scaffolds, rather than obtaining a fibroblast-like spindleshape morphology as on the TCP. This indicated  Similarly, when the number of viable cells were compared on TCP and on the scaffolds, it was observed that the continuous increase in the cell number observed on TCP was not reached on the scaffolds. This trend was in correlation with the observation of more spindle-shaped, proliferating cell morphology observed on TCP. 3D printed alginate scaffolds, both due to the proper biomimetic stiffness provided for the parathyroid cells [31], as well as due to the lack of proper cell binding sites and unfavorable cell sprouting properties [29], was favorable for maintaining the globular cellular morphology native to the parathyroid cells, while decreasing the presence of fibroblast-like cells.  Surface calcium receptors (CaSR) are membranebound on functional parathyroid cells. Anti CaSR marker was stained to detect the presence of parathyroid cells on 3D bioprinted scaffold (figure 5).

Determination of cellular functionality within 3D printed scaffolds
Surface calcium receptors (CaSR) are membranebound on functional parathyroid cells and sense even  minor changes in the serum Ca 2+ concentration level. In this study, different Ca 2+ concentrations (0.5 mM, 1.25 mM and 3.0 mM) were used to stimulate the isolated parathyroid cells to secrete and/or inhibit PTH secretion and therefore evaluate their functionality in in vitro culture.
1.25 mM Ca 2+ was used as a physiological (neutral) condition, while decreased Ca 2+ level (0.5 mM) was expected to stimulate the PTH release; whereas increased Ca 2+ level (3.0 mM) was expected to inhibit the PTH secretion [10]. This effect was tested with cells cultured on TCP as well as on the 3D printed alginate scaffolds.
On the other hand, a meaningful difference was not seen between different Ca 2+ concentrations in figure 6(a). Yet, results from hydrogel scaffold, in figure 6(b), were more reliable and it can be seen lower dose of Ca2 + enhance PTH production and PTH secretion was inhibited by higher Ca2 + concentration. Most meaningful hormone induction seen on day 7. After day 7, hormone release decreased.
PTH secretion data showed that, cultured cells were actively healthy cells which could secrete their natural hormones. Moreover, T3 thyroid hormone secretion proved that there were not just healthy parathyroid cells but also healthy thyroid cells (figure 7). Since there was not another cell group that could secrete mentioned hormones, these results also count as biomarkers of parathyroid/thyroid cells.

Discussion
Parathyroid and thyroid cells were characterized via immunostaining and hormone secretion assays. This step is of utmost importance since in literature culturing parathyroid cells in long-term is mentioned as a difficult procedure because of their high differentiation level and low proliferation activities [10,11]. On the other hand, thyroid and parathyroid glands are close in proximity and distinguishing them in surgery room with naked eye is a struggle. Thus, iatrogenic removal of the thyroid tissue is a common complication. Therefore, parathyroid cell culture can be very easily contaminated by the thyroid cells. One of the distinguishing ways to differ these two cell types is the Ca 2+ dose dependent PTH release, since only parathyroid cells contain CaSR receptor. PTH release from the cells upon stimulation with Ca 2+ was quantified with PTH ELISA kit. Moreover, T3 thyroid hormone secretions were assessed as a proof of functional thyroid cells ( figure 7). To combine all of these results, it can be said functional parathyroid and thyroid cells were cultured in 2D and 3D environment in our study.
Producing functional endocrine organs with 3D bioprinting/printing is not a common procedure in the literature. Studies so far have concentrated on the 3D bioprinting of less vascularized, less innervated, and less functional tissues like cartilage and/or skin as contrast to highly vascularized and functional organs like endocrine glands [32]. In a recent study, researchers did not isolate and studied the functionality of the individual cells in vitro, rather they have minced the isolated parathyroid tissue, mixed within a bioink and 3D printed the tissue [33]. There are a few other studies in the literature on engineering parathyroid tissue. For example, Ritter et al cultured parathyroid cells in 3D collagen gel matrix [34]. In another study, bioprinting a thyroid gland with collagen hydrogel was performed [35]. Furthermore, Matrigel, decellularized extracellular matrix and alginate can be used as a biomaterials for bioartificial endocrine glands [32,[36][37][38][39][40]. In this study, alginate was used as biomaterial to construct the scaffold for parathyroid cells.
To overcome the poor mechanical properties of the hydrogels and find more printable biomaterial, concentration of alginate was increased and 8% (w/v) was used and as the crosslinker 0.5 M CaCl 2 crosslink agent was preferred. Considering the natural structural features of the parathyroid gland and polygonal shape of the chief cells, cylindrical architecture with honeycomb design was preferred for the scaffold. The first layer of the scaffold was printed with close pores and the rest of the two layers were printed with wider pores to eliminate the cell leakage and increase the cell adhesion on the scaffold ( figure 3).
To analyze the cell viability in a quantitative way and to assess the cell morphology in TCP and on 3D scaffolds Alamar Blue and Phalloidin/DAPI tests were done, respectively. It was observed that cell proliferation in TCP was much higher than on scaffold. In TCP, cell proliferation was increased continuously ( figure 4(B)). On the other hand, with fluorescent microscope examination of Phalloidin/D-API staining showed fibroblastic cytoskeleton in TCP (figure 1). However, on alginate scaffold after day 7, there was a decrease on cell proliferation ( figure 4). Also, globular cells were detected with Phalloidin/D-API which resembled the morphological appearance of parathyroid/thyroid cell morphology ( figure 4). Alginate is generally used in a combination with other hydrogels like gelatin to increase the viability, attachment, spreading and proliferation of fibroblast because pure alginate biomaterial is not known as a proper environment for cells which need high anchorage and sprouting and does not promote cell adhesion effectively [29]. Thus, it can be said, in TCP proliferation of the parathyroid cells were limited because of the overgrowth of the fibroblast and fibroblasts kept proliferate even after day 7, on day 14. On the other hand, fibroblastic cells cannot survive on 3D scaffold due to the property of the biomaterial, so their elimination creates a beneficial environment for proliferation of the parathyroid cells.
Additionally, functional parathyroid cells should detect external Ca 2+ changes with surface calcium receptors and PTH is secreted as a response of that. Therefore, cells were incubated with using three different Ca 2+ concentrations (0.5, 1.25 and 3.0 mM) to evaluate calcium responsiveness and to assess hormone active functional cells. To stimulate PTH secretion 0.5 mM was used, 1.25 mM was chosen as neutral stage and to suppress the PTH production 3.0 mM was selected [10]. Since PTH and T3 hormone secretions were achieved, PTH secretion was not stimulated/suppressed by Ca 2+ concentration. These findings indicate that PTH and T3 hormones were specific to parathyroid/thyroid cells. It can be said that PTH secretion was higher nearly twice on the 3D scaffold. Also, after day 7, decrease of the PTH shows correlation with Alamar blue and Phalloidin/DAPI results. This outcome also proves that culturing paratiroid cells on long-term is difficult and they can proliferate on 3D scaffolds more easily than fibroblastic cells.

Conclusion
Parathyroid gland can be damaged after thyroidectomy/parotidectomy or neck surgeries and these unwanted temporary or permanent effects can be lead hypoparathyroidism. Since hypoparathyroidism has no permanent treatment, in both research and therapeutic field tissue engineered parathyroid gland has so much meaning. Results showed that not only parathyroid cells, but also thyroid cells were cultured on 3D printed alginate hydrogel and these cells were specified with specific hormone assessments and IHC stains. Thus, 3D fabricated product can be used as a potential hypoparathyroidism/hypothyroidism treatment. Furthermore, PTH is preferred for fracture healing, prosthesis and implant fixing since it accelerates skeletal repairing creases implant fixation with effecting vascularization, bone formation and reproliferation. Therefore, for further investigations, parathyroid cells can be cultured with bone cells and behavior of bone cells can be investigated in a PTH secreted coculture and animal testing can be done on mice models.

Data availability statement
The data cannot be made publicly available upon publication because no suitable repository exists for hosting data in this field of study. The data that support the findings of this study are available upon reasonable request from the authors.