(247j) Patterning and Controlled Condensation of Cells Embedded within Thermosensitive Support Baths for Anisotropic Tissue Engineering

INTRODUCTION: Anisotropic tissues are ubiquitous in the musculoskeletal system. When it comes to regenerating these tissues, remarkable challenges exist in reproducing the cell-scale complexities associated with anisotropic architectures. Traditional tissue engineering strategies tend to focus on scaffolds with spatially homogenous cell densities. However, this contrasts with in vivo tissue development, where cells communicate efficiently within patterned regions of high cell density leading to self-assembly of complex tissue. Freeform 3D bioprinting into hydrogel support baths, such as Carbopol (CP), enables the freeform extrusion of high-density cellular bioinks into patterns mimicking native tissue organization. However, support baths are designed to be sacrificial and are limited in long-term (>7 days) cell culture due to rapid dissolution. We present a support bath composed of poly(N-isopropylacrylamide)-graft-chondroitin sulfate (pNIPAAM-CS) [1] blended with gelatin and Carbopol 940®. We exploit the cell compatible, temperature-triggered gelation of pNIPAAM at (32°C) to enable bioprinting of cells at 25°C into embedded anisotropic channels. Additionally, since this support bath provides a thermosensitive surface for on-off cell adhesion, it was hypothesized that this could be exploited for stimulating the formation of closely associated and oriented cellular patterns. Here we use small angle X- ray scattering (SAXS) to characterize the nanostructure of the temperature sensitive support bath and link the findings to the in vitro behavior of bioprinted cells.

METHODS: SAXS was performed at B21 beamline, Diamond Light Source, UK with sample loading and beamline parameters set as in the reference [2]. Dilute support bath samples (0.3 w/v% pNIPAAM homopolymer, 0.08 w/v % CP and 0.1 w/v % gelatin) were collected and measured at 25ºC or 37 °C, for 21 frames with 1 s exposure each. From the SAXS measurements, we extracted the Guinier slope defining fractal dimension (n) to provide information on the dimensional structure of the support bath particles. Cell studies were conducted in parallel where L929 murine fibroblasts (P14) or human bone marrow derived mesenchymal stromal cells (MSCs, P4) which were suspended at a density of 4x10⁶ cells/mL in 6 % w/v porcine gelatin dissolved in cell culture medium. Bioinks were microextruded through a 300 µm (inner diameter) needle into 1 mL of support bath at 25°C composed of 3 w/v % pNIPAAM-CS, 0.8 w/v % CP and 1 w/v % gelatin and cultured for either 21 days (fibroblasts) or 5 weeks (MSCs). Throughout the culture period, constructs were either maintained under static temperature at 37°C or dynamic conditions by cooling to 25°C for 10 min every 5 days. Embedded cells were visualized at the end of the study with TRITC-phalloidin and DAPI counterstain and a semi-quantitative analysis was carried out with ImageJ (NIH, Bethesda, MD, USA) on n ≥ 5 images per time point to characterize the effects of static versus dynamic culture on cell patterns. For the fibroblasts, the distance between each pair of cells was plotted as a heat map. For the MSCs, intensity was plotted as a function of x-position across the image.

RESULTS: The Guinier slope of n=-1.04 was extracted at 25 °C (Figure 1A, blue) suggesting that the bath consisted of rod-like polymeric chains with weak physical entanglement [3] (Figure 1B, scheme) . Thus, at 25 °C, the blended network is fully hydrated and also behaves like a shear thinning and self-healing fluid, enabling omnidirectional bioink extrusion (rheological data not shown). As temperature was increased to 37 °C, n shifted to a value of -3.14 (Figure 1A, red), indicating that the network transitioned to a collapsed entangled globule with a rough surface [3] (Figure 1C, scheme). For the non-thermoresponsive bath, without the pNIPAAM, values of n=-1.11 and n=-1.08 were obtained at 25 and 37°C, respectively (Figure 1D) indicating an extended coil structure. This indicated that the temperature responsiveness of the bath is attributable to the pNIPAAM component. Confocal microscope images of DAPI and Phalloidin-TRITC labelled fibroblasts at 21 days of static culture revealed a tendency for filopodia formation (yellow arrows, Figure 2A). In addition, heat maps indicated larger distances between nearest neighboring fibroblasts for static versus dynamic culture conditions (Figure 2A,B, respectively). Similarly, mesenchymal stromal cells (MSCs) cultured under static conditions exhibited elongation (yellow arrows, Figure 2C), possible migration beyond their seeded channel placement (red arrows, Figure 2C), and overall pattern width of approx. 400 µm. Dynamic culture promoted closer MSC approximation within the embedded channels and allowed maintenance of cell pattern width of approx. 200 µm after 5 weeks of culture (Figure 2D).

DISCUSSION: The collapsed globular state of the bath at 37 °C exposes the hydrophobic surface of pNIPAAM, thus allowing the adsorption of proteins and cell attachment, whereas the hydrated state at 25ºC induces immediate cell release [4]. Our approach advances freeform bioprinting beyond the state-of-the-art, as we have a non-sacrificial embedding medium that can not only mechanically stabilize an embedded, cell-laden channel generated by a fugitive bioink, but also allow for spatiotemporal control over cellular binding sites. In the current study, temperature-triggered on-off cell binding was implemented to promote cellular condensation within aligned patterns, which could have implications for cell-cell communication and anisotropic self-assembly across multiple areas of musculoskeletal repair.

REFERENCES: [1] T. Christiani+, JOR Spine 2021, 4, 1161. [2] C. J. C. Edwards-Gayle+, Journal of Synchrotron Radiation 2021, 28, 318. [3] R.J. Roe+, Blackwell Publishing Ltd: Brookfield Center, 2000; 56, 79. [4] V. Grabstain +, Biotechnology Progress 2003, 19, 1728-1733.

ACKNOWLEDGEMENTS: Funding for this study was provided by the AO Foundation and AO Spine. The authors are grateful to Diamond Light Source (Didcot, UK) for awarding beam time (SM29767) to this project. J.K.W. acknowledges support by the European Union’s Horizon 2020 (H2020-MSCA-IF-2019) research and innovation programme under the Marie Skłodowska-Curie grant agreement 893099 — ImmunoBioInks. D.L acknowledges support from European Union’s Horizon 2020 Research and Innovation Program under grant agreement No 857287 (BBCE – Baltic Biomaterials Centre of Excellence).