Tissue printing for engineering transplantable human parathyroid patch to improve parathyroid engraftment, integration, and hormone secretion in vivo

The study was a cooperative research work performed by staff at two institutions, Seoul National University Bundang Hospital (SNUBH) and the Pohang University of Science and Technology (POSTECH). Human parathyroid samples were provided by SNUBH and the 3D tissue printing processes for the fabrication of PTs-printed patches, composed of printable adECM bioink and porous PCL support, were performed at POSTECH. The study protocol was approved by the Institutional Review Board of SNUBH (B-1904/535-301).

Porcine adECM was prepared as described in another article with slight modifications [21, 22]. In brief, fresh adipose tissue collected from a slaughterhouse was centrifuged to remove oil and blood, washed with 1× phosphate-buffered saline (PBS) solution and decellularized with 0.5% sodium dodecyl sulfate (SDS) solution for 48 h. The decellularized tissue was treated with isopropanol for 48 h to remove lipids, washed several times with 1× PBS solution, sterilized by treatment with 0.1% peracetic acid solution in 4% ethanol for 4 h, and washed several times with 1× PBS and distilled water. Lyophilized adECM was frozen in liquid N₂ and crushed into a powder using a mortar and pestle. To formulate 2% (w/v) adECM bioink, each 100 mg aliquot of adECM powder was added to a solution of 10 mg of pepsin (P7125, Sigma-Aldrich, St. Louis, MO, USA) in 0.5 M acetic acid and incubated for 3 d. After complete solubilization, the pH of the adECM solution was adjusted by dropwise addition of cold 10 M NaOH solution on ice to avoid thermal gelation. The pH-adjusted adECM bioink was stored at 4 °C until being used.

Rheological examinations were performed using an Advanced Rheometric Expansion System (TA Instrument, USA) with a 25 mm diameter parallel plate geometry. To evaluate the viscosity, steady shear sweep analysis of adECM pre-gel was performed at a constant temperature of 15 °C. Dynamic frequency sweep analysis was conducted to measure the frequency-dependent storage and loss moduli of dECM gels in the range of 1–1000 rad s⁻¹ at 2% strain after incubation for 30 min at 37 °C. A temperature sweep experiment was performed to study the gelation kinetics of dECM pre-gels. The storage and loss moduli were measured to evaluate the structural stability after the thermal gelation process. All measurements were conducted in triplicate.

The study goals and protocols were explained in advance to four patients with primary and secondary hyperparathyroidism between May and October 2019 at Seoul National University Bundang Hospital (SNUBH). The participating patients were provided written informed consent. Three and one-half parathyroid glands were removed from the patients with secondary hyperparathyroidism during surgery.

Patients' parathyroid glands from SNUBH were stored in complete Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin (P/S). The intact parathyroid glands were cut into fragments smaller than 0.5 mm in diameter and were passed through 500 µm nylon mesh filter bags to prevent nozzle clogging prior to adECM encapsulation. The filtered PTs (30 mg ml⁻¹) were uniformly mixed into 2% (w/v) adECM bioink at a volume ratio of 1:10 while avoiding unwanted air bubbles. Specifically, 60 mg of PTs in 200 μl of culture medium was mixed into 2 ml of 2% adECM bioink.

In this study, the Food and Drug Administration (FDA)-approved PCL was used as a supporting framework to prepare transplantable PTs-adECM, owing to its acceptable bioprintability, biodegradation behavior, and biocompatibility [23]. A porous PCL mesh (10 mm diameter, 200 µm porosity, and 1 mm thickness) was printed using an in-house 3D bioprinting system [23]. This mesh acted as a mechanical support to enhance and stabilize the integration of printed PTs-encapsulated adECM into the host subcutaneous tissue. Following this process, the PTs encapsulated in adECM bioink were sequentially printed onto the supportive PCL mesh to form well-distributed minced PTs. These patches were validated by comparison with PTs alone. The workflow of this study is shown (figure 1).

Zoom In Zoom Out Reset image size

Groups of PTs-adECM and PTs alone were cultured for 7 d in complete 154CF medium (Life Sciences; Cascade, NY, USA), containing human keratinocyte growth supplements and 0.07 mM calcium chloride. Cell viability over 7 d was monitored by staining using Live/Dead Cell Assay Kits (Invitrogen; Carlsbad, CA, USA), where living and dead cells colored green and red, respectively. The stained images of the respective groups were quantified using Image J software (version 1.48).

Tensile testing was performed on PTs alone and PTs-printed PCL patch using a universal mechanical testing machine (3340 Series Single Column Systems, Instron, USA). The tensile strengths of two groups were measured at a steady strain rate of 2 mm min⁻¹. Young's modulus was then calculated from the initial slope of the linear region (5%–15% strain) in the stress–strain curve (data not shown).

PTH secreted by cells cultured in vitro was qualitatively assessed through immunofluorescence staining examination. Briefly, the samples were incubated overnight at 4 °C anti-human PTH primary antibody (cat. no. ab154792, Abcam; Cambridge, MA, USA, 1:500 dilution) and washed three times in 1× PBS for 5 min each. The samples were then incubated for 1 h at room temperature with Alexa Fluor 488-conjugated anti-rabbit secondary antibody (cat. no. ab150077, Abcam; Cambridge, MA, USA, 1:200 dilution) and washed three times in 1× PBS for 5 min each. Nuclei were subsequently stained by incubation with 4', 6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), and the stained samples were imaged using a confocal microscope (Carl Zeiss LSM 880, Jena, Germany). To quantitatively compare the secretion of PTH, the soluble protein extract from the tissues (10 mg piece per 1 ml extraction buffer at the beginning) was analyzed on days 1, 4, and 7 using enzyme linked immunosorbent assays (ELISA) Kits (MyBiosource; San Diego, CA, USA) according to the manufacturer's instructions.

All animal protocols were approved by the POSTECH ethics Committee (POSTECH-2018-0031). PT alone and PTs-printed patches were subcutaneously implanted into 24 healthy male Sprague-Dawley rats (n = 3 per group for each time point). The rats were sacrificed after week 0, 1, 2, and 4; as a control group, week 0 means the time point immediately after the implantation. Initial parathyroid engraftment, long-term stabilization, and hormone secretion were further evaluated.

Samples were fixed with 4% solution of paraformaldehyde (PFA) for hematoxylin and eosin (H&E) and immunohistochemical (IHC) staining. Representative images of stained PTH in each group were captured using an automatic digital slide scanner (Panoramic Viewer; 3DHistech, Budapest, Hungary). IHC optical density scores were quantified as described in detail in literature [24].

All data were expressed as mean ± standard deviation (SD). Differences between groups were determined using a Student's t-test. Significance between groups was established for p < 0.01. All the measurements were repeated at least three times and representative results were shown.

The prepared adECM pre-gel at 15 °C showed shear thinning behavior in the established shear rate range (figure 2(A)). Gelation kinetics of the dECM pre-gels was successfully demonstrated at varying temperatures from 4 °C to 37 °C. The result revealed that the dECM pre-gel started gelation immediately at 15 °C and formed a crosslinked gel at 37 °C within 30 min. In addition to the shear thinning behavior and gelation kinetics, the storage and loss moduli after the gelation were investigated. The storage modulus was larger than the loss modulus, meaning that the prepared adECM bioink could maintain its stable structure after the thermal gelation. With the prepared printable bioink, we finally embedded PTs in adECM bioink (PTs-adECM). The PTs alone were also prepared in vitro. To determine an appropriate time for subcutaneous implantation after extraction from patients, cell viabilities of the two groups were assessed on days 1, 4, and 7. After 4 d, few dead cells (red) were observed, with most of the cells in the two groups surviving (green) and at a comparable level (figure 2(B)). By day 7, however, the dead cells in the two groups had increased with no significant difference between the groups. The quantitative results of cell viability over 7 d supported this observation (figure 2(C)). These findings indicate that, to maintain parathyroid function, PTs cultured in vitro should be transplanted into subcutaneous tissue within 7 d.

Zoom In Zoom Out Reset image size

Qualitative comparison of the levels of PTH secreted by PTs-adECM and PTs alone showed that PTH was expressed by PTs in both groups (figure 2(D)). Quantitative analysis of cumulative PTH secreted into the culture medium over 7 d showed that both PTs-adECM and PTs alone released similar amounts of normal PTH into the medium, indicating that PTs remained functional after incorporation into adECM bioink (figure 2(E)). Because PTs-adECM is also bioprintable, it may allow the incorporation of a printable PCL support that can provide better mechanical stabilization in vivo.

To prepare functional and implantable tissue patches containing PTs, the minced PTs-adECM was printed onto the pre-designed supportive PCL membrane in the functional and implantable form (figure 3(A)). As a control group, a group of PTs alone was prepared. The well-positioned PTs in 2% adECM, designed pores, and line width were observed in the prepared PTs-printed patch and compared with the printed PCL only (figure 3(B)). The observations demonstrated that the tissue printing process was successfully implemented. Due to the influence on the PCL support, Young's modulus of PTs-printed patch was approximately 16.3 folds larger than that of PTs-adECM only (without PCL support) (figure 3(C)). Obviously, this mechanical enhancement indicates that the engineered patch could provide better stabilization and integration to the native subcutaneous tissue. The PTs-printed patch and PTs alone were cultured and stabilized in vitro for 1 d. Finally, they were implanted subcutaneously in the rats for further comparison and observations (figure 3(D)).

Zoom In Zoom Out Reset image size

After one week, both the printed patches and PTs alone generally remained in their original position (figure 4(A)). At week 4, however, the group of PTs alone had migrated considerably from their original site of injection and the numbers of PTs had markedly decreased. By contrast, the engineered patches remained in their original position at week 4 and were well-integrated into the host tissue, indicating that these preparations, consisting of PTs-printed patches, adECM, and supportive PCL, were rapidly immobilized within the host tissue.

Zoom In Zoom Out Reset image size

In addition, the initial in vivo integration of PTs-printed patches and PTs alone into subcutaneous tissue was evaluated by H&E staining at week 1. The PTs-printed patches were located at the inner region of the host tissue with the PTs present throughout the implanted patches (figure 4(B)). Implantation of PTs alone, however, resulted in the presence of aggregates found on the surface of the host tissue, despite the minced tissues being initially well distributed.

We hypothesized that the well-integrated PTs-printed patches would eventually show normal parathyroid functions, such as the secretion of PTH. Histologic evaluation showed that PTH was not expressed prior to subcutaneous implantation of PTs-printed patch or PTs alone (figure 5(A)). On week 1, PTH was locally expressed in rats implanted with PTs-printed patch, whereas little PTH was expressed following implantation of PTs alone. These findings suggest that the engineered patch could better interact with the host tissue and provide relatively prompt stabilization under in vivo conditions.

Zoom In Zoom Out Reset image size

Evaluation of the degree of PTH secretion on weeks 2 and 4 after implantation of PTs-adECM PCL revealed that PTH was expressed over a wider area after two weeks than one week, indicating that the stabilized PTs had greater functional activity after two weeks than one week. Similarly, PTH was still expressed throughout the implanted region at week 4. However, PTH expression in the control group implanted with PTs alone was extremely limited on week 2 with no significant increase at week 4. More importantly, it was difficult to locate the position of the implanted PTs.

To corroborate these results, the IHC optical density scores for PTH secretion were compared in the two groups (figure 5(B)). Although these scores did not differ significantly in the two groups before subcutaneous implantation, these scores differed significantly four weeks after implantation. These observations indicate that these engineered patches were well stabilized in vivo, resulting in the secretion of PTH.