Harnessing Thermogel Actuation for Driving Directional Stromal Cell Communication and Migration into Columnar Arrays

doi:10.21203/rs.3.rs-3827648/v1

Download PDF

Article

Harnessing Thermogel Actuation for Driving Directional Stromal Cell Communication and Migration into Columnar Arrays

https://doi.org/10.21203/rs.3.rs-3827648/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

During developmental processes, cells frequently condense along a preferred axis, creating columnar arrangements—a pivotal step in shaping elongating tissue structures and facilitating gradual building of tissue complexity. Despite advances in biofabrication technology that has allowed researchers to recreate these axial arrangements in vitro in 3D culture, maintaining these patterns for periods of cultivation beyond 7 days has proven challenging, given cells' tendency to exhibit random migratory patterns. In this study, we introduce EXPECT (EXtrusion Patterned Embedded ConstruCTs), a thermosensitive hydrogel based on poly(N-isopropylacrylamide) designed with specific rheological properties enabling the creation of embedded, macroscopic, cell-laden channels within the hydrogel using 3D printing. EXPECT, coupled with mild temperature changes at regular intervals, suppressed the random migratory tendencies of mesenchymal stromal cells (MSCs), guiding the cells to laterally intercalate and aggregate longitudinally. This resulted in the formation of continuous stacked arrangements of MSCs sustained over 36 days of culture. Additionally, EXPECT led to the elongation of initially spaced MSC spheroids toward each other, culminating in their fusion into narrowed, columnar assemblies. Our study presents a versatile and readily applicable approach for orchestrating and maintaining cell communication and movements along a preferred axis outside the developmental niche. By addressing a key limitation in current in vitro 3D culture systems and inducing cell movements reminiscent of both convergent extension and directed chemotaxis, we present a novel tool for studying various facets of development, disease, and repair.

Physical sciences/Engineering/Biomedical engineering

Biological sciences/Stem cells/Adult stem cells/Mesenchymal stem cells

Biological sciences/Biological techniques/Biological models/Musculoskeletal models

In the development of aligned tissue structures, such as bone and tendon, a pivotal step involves the condensation of cells along a longitudinal axis resulting in the formation of narrowing and elongating columnar patterns [1–6]. Maintaining spatially aligned movements play a fundamental role in achieving successful morphogenesis, especially in the development of hierarchically complex tissues.

Replicating this carefully orchestrated alignment using in vitro 3D culture holds significant importance for researchers seeking to understand the mechanisms involved in the formation of elongating tissues or working to advance regenerative strategies for such tissues. Presently, acoustic vibration [7–10] or magnetic field [11–13] are the primary techniques for generating columnar cell assemblies within precursor macromer solutions. Following the application of these stimulatory methods, the organized assemblies are "fixed" through the gelation of the 3D hydrogel network. Nevertheless, it is essential to acknowledge that spatial disorganization becomes apparent over time, usually by the seventh day of culture [7, 14], due to the migratory behavior of the cells. This is problematic as the complexity inherent in tissue-scale structures demands sufficient time, along with various other microenvironment factors [15], for development in culture. Moreover, within the domain of spheroid cultures, studies on elongation morphogenesis [16], tissue repair [17], or cell cluster migration [18, 19] have documented the occurrence of cell collectives moving synchronously in columnar patterns in vitro. However, instances of coordinated reshaping are limited to short time frames, usually spanning hours to days, after which isotropic spheroid disaggregation or disorder tends to manifest. The aforementioned examples drawn from the literature over the last 15 years imply that the stochastic migratory behavior of cells in vitro has imposed limitations across various fields of study.

While the significance of chemotactic gradients in guiding cell migration along a preferred axis in vivo is well-established [20–22], the prolonged observation of this phenomenon using in vitro 3D cultures is infrequently documented. This scarcity is attributed to a common limitation in current in vitro culture techniques—the lack of mechanisms to hinder chemoattractants from uniformly diffusing throughout the culture system [23–25]. As a result, this diffusion gradually leads cells to migrate in various directions instead of adhering to a coordinated pattern. The random diffusion of chemoattractants can even override biophysical cues such as grooves or fluid flow [26, 27]. Given the significant relevance of collective cell alignment and axial migration in various areas of tissue growth and morphogenesis, there is a pressing need for an effective in vitro approach to sustain cellular self-organization along a preferential axis.

Motivated by this challenge in fundamental research and translation of regenerative medicine, we introduce a thermosensitive hydrogel named EXPECT (EXtrusion Patterned Embedded ConstruCT), which is based on a copolymer of poly(N-isopropylacrylamide) grafted with chondroitin sulfate (pNIPAAm-CS). EXPECT was designed to exhibit shear thinning, yield stress, and self-recovery properties, enabling us to create embedded cell-laden cylindrical channels within the hydrogel using 3D printing. Additionally, EXPECT demonstrates a lower critical solution temperature (LCST), approximately at 32°C [28, 29]. By periodically cooling the 3D printed constructs below the LCST for short intervals throughout the culture period, we induced the condensation of single-cell mesenchymal stromal cells (MSCs) toward each other. This resulted in narrowing their arrangement and promoting the formation of stacked arrangements that remained stable over a 36-day culture period. Furthermore, the temperature-driven actuation of EXPECT led to the elongation of initially spaced MSC spheroids toward each other, culminating in their fusion into narrowed, columnar assemblies.

The LCST of pNIPAAm-based 2D surfaces is well known to be linked to the reversible adhesion of various cell types and proteins to polymer surfaces [30–36]. Our findings indicate that by constructing 3D channels with surfaces comprised of pNIPAAm-derived polymers, we can leverage its temperature sensitivity to spark and dynamically sustain cell directional migratory decision-making, convergent extension,[6] and fusion along a preferential axis– aspects currently lacking in existing in vitro methodologies. This marks a breakthrough development in biomaterials and biofabrication, presenting a versatile new approach that can lead to advances in bioassembly, developmental biology, and disease modeling.

Development of the thermosensitive EXtrusion Patterned Embedded construCT (EXPECT)

We approached the design of the EXPECT hydrogel with two goals in mind. First, we decided to place the cells locally together inside an oriented embedded channel within a 3D hydrogel network, to facilitate cellular communication. Embedded 3D printing offered one viable strategy for achieving this. Previously, groups have reported the writing of sacrificial inks within granular gel media exhibiting yield stress and self-healing properties [37–39]. After 3D printing, the sacrificial ink is removed by perfusion and the resulting embedded hollow channels can be used for a variety of applications, such as organoid fusion or providing enhanced nutrient diffusion [40, 41]. Carbopol® 940 (CP) is one of the most extensively studied granular gel media [42–47] known for its good printing performance and cellular compatibility. The microparticles of CP consist of polyacrylic acid with a high molecular weight, featuring methyl and methylene substituent groups arranged in a syndiotactic manner along the polymer chain [48]. Entangled and close-packed polymer chains of CP form sponge-like microgels, in which chains extend from the core and can interact with other particles. The extended chains strengthen inter-particle interactions and forming a fully entangled microparticle network [49]. In our study, we selected CP as one of the components of EXPECT, which would allow us to align cells within an embedded channel of a ring geometry. Because enclosed rings are a geometry that offers the minimum perimeter for a given enclosed area, we projected that the pattern would enable us to minimize the spatial distance between cells. At the same time, the ring could be extruded in a macroscopic size (10 mm diameter), so we anticipated that migratory movements along the ring would constitute straight ahead, directional movement on the cellular scale.

Our second goal for the design of EXPECT was to provide a means of sustaining cell migratory activities in a forward and backward motion, promoting unidirectional organizational behavior. Although embedded bioprinting has marked a notable progression in biomanufacturing, granular media exhibit a high degree of isotropy, lacking any strong asymmetric cues capable of guiding embedded cells in a specific direction throughout extended in vitro culture periods [35, 50, 51]. To address the limitations of using a hydrogel like CP alone, we introduced an additional component, a copolymer of polyN-isopropylacrylamide-graft-chondroitin sulfate (pNIPAAm-CS), into the hydrogel. This copolymer, developed and characterized previously [28, 29], demonstrates LCST behavior in aqueous solution at approximately 32 ºC. The LCST behavior of pNIPAAm-based systems is well-documented to involve a shift in molecular conformation from a solvated, rod-like state to a hydrophobic, globular state [36, 52, 53]. Based on previous investigations, it is known that culturing cells on 2D substrates containing pNIPAAm above its LCST promotes cell spreading and attachment, while the solvated molecular state below the LCST triggers the immediate release of attached cells and proteins [30–33]. Capitalizing on this knowledge, we explored the possibility of printing cells in an embedded ring channel that incorporates a pNIPAAm-based thermosensitive surface. We hypothesized that cooling below the LCST could effectively reduce the random cell migration out of these channels by inducing periodic detachment from the matrix. Figure 1 illustrates a schematic representation of the approach employed in this study.

Establishing the composition of EXPECT

After determining the main components of the embedding medium, the next step was to design a system that achieved a balance of desired characteristics from each phase. To begin, pNIPAAm-CS copolymers were synthesized in house by free-radical polymerization [28, 29, 54]. After synthesis and purification, we blended aqueous solutions of 1%, 3%, and 5% (w/v) pNIPAAm-CS with 0.8% (w/v) CP. In order to compare the results, we used 0.8% (w/v) CP alone, which is in a well-established composition range for embedded bioprinting [45].

We conducted amplitude sweep tests at 25°C (Supplementary Fig. 1a) and found that the formulations exhibited solid-like behavior within the linear viscoelastic region, below 1% strain. Additionally, the formulations could fluidize when subjected to a specific yield stress unique to each formulation, with increases in pNIPAAm-CS content leading to a reduced flow stress (Supplementary Fig. 1b). This suggested that the presence of pNIPAAm-CS chains reduced cohesive CP-CP inter-particle interactions. Further supporting this concept, print fidelity characterization of the patternable hydrogel was carried out by means of printing a free-floating horizontal grid structure with a low viscosity of 6% (w/v) gelatin bioink supplemented with 0.1% (w/v) Coomassie Blue dye. Printing in a high concentration of 5% pNIPAAm-CS + 0.8% CP resulted in deformation of the printed gid compared to the other embedding formulations (Supplementary Fig. 1c).

Despite possible losses in print fidelity at higher concentrations of pNIPAAm-CS, incorporating this temperature-responsive component into the CP microparticles rendered the hydrogel sensitive to temperature changes between 25°C to 37°C°C, exemplifying an increase in the storage modulus (G') between 66 and 730% (Supplementary Fig. 1d). The magnitude of this increase was proportional to the concentration of pNIPAAm-CS in the hydrogel formulation (Fig. 1e). To strike a balance between printability and thermal response, we decided to proceed with intermediate concentrations of pNIPAAm-CS in the thermosensitive embedding medium (3% pNIPAAm-CS + 0.8% CP). Additionally, we discovered that by blending a small concentration of gelatin (1% (w/v)) with this composition, we could recover key rheological properties at 25°C compared to 0.8% CP alone, summarized in the following sections.

Rheological properties of EXPECT

The shear recovery behavior of EXPECT (3% pNIPAAm-CS + 0.8% CP + 1% gelatin) was comparable to that of 0.8% CP alone. In an oscillatory step strain test (Fig. 2a), we measured the storage and loss moduli (G' and G'', respectively) while alternating the strain between 1% and 250%. Over three cycles, EXPECT maintained self-healing properties, with a mean recovery of 94.6 ± 6.5%, compared to the CP control exhibiting a mean recovery of 96.9 ± 4.0% (p = 0.4). These results indicate that EXPECT can effectively reform its network after nozzle translation in an extrusion-based printing process.

In the amplitude sweep test with oscillatory strain at 25°C (Fig. 2b), we determined that the flow stress of EXPECT, calculated at the point where the curves crossed over (G' = G''), was 366 ± 43 Pa, slightly higher than the CP control (277 ± 4 Pa) (Fig. 2c). This suggests that EXPECT can provide equivalent or improved spatial stability to a bioink extruded within its structure. Print fidelity studies with 6% (w/v) gelatin bioionk containing 0.1% (w/v) Coomassie Blue dye (Fig. 2d) further confirmed this, where multiple measures of printing accuracy, including printability (PR) value, pattern angle, and pattern width, were similar to the same bioink extruded within CP.

When the temperature increased from room temperature (25°C) to physiological temperature (37°C), EXPECT exhibited temperature-responsiveness (Fig. 2e), with G' increasing by 195 ± 65% between these temperatures. In comparison, the control CP showed a slight increase in G' of 5 ± 1% (Fig. 2f, p = 0.007). Macroscopic images of EXPECT at 25°C and 37°C (Fig. 2g) indicate that the system is a self-supporting gel at both temperatures, with the phase transition of pNIPAAm copolymer demonstrated visually by the shift to a white opaque appearance. Overall, our results indicate that the material's temperature responsiveness was successfully implemented without compromising 3D printing accuracy.

Small Angle X-ray Scattering (SAXS) analysis of EXPECT

We performed a structural analysis utilizing SAXS to explore the EXPECT system, aiming to understand and interpret its thermoresponsive characteristics. To achieve this, we collected scattering data at different concentrations and temperatures, encompassing both sub-LCST (25°C) and super-LCST (37°C) conditions, in the q range, which is sensitive to structures ranging from 27 to 140 nm in size (Fig. 3, Supplementary Fig. 2, and Supplementary Fig. 3). From our SAXS measurements, we extracted the Guinier slope (denoted as -n = d_m, representing the fractal dimension, as shown in Fig. 3a). This parameter offers insights into the arrangement of polymer chains and surface characteristics within the EXPECT system or its individual components.

As a starting point, we ran the SAXS analysis on pNIPAAm homopolymer at a concentration of 1.5% (w/v). We observed a significant change in d_m as a function of temperature, finding 1.5% pNIPAAm transitioning from 1.17 at 25.0°C to 4.09 at 37°C, as shown in Fig. 3b. This shift indicated a transformation from a swollen, disentangled polymeric network to collapsed globules with a nearly smooth surface (as illustrated in Fig. 3a i and ii) [55]. This transition pattern remained consistent for all three tested concentrations of pNIPAAm homopolymers (0.3%, 1.5%, and 3% (w/v), as shown in Supplementary Fig. 2a and 2b), with minor variations in the Guinier slope. Previous SAXS analyses of pNIPAAm microgels have reported similar -n slopes (or d_m values) with values of approximately 1.5 below the LCST and around 4 above the LCST [56, 57]. This suggests that SAXS analysis is a suitable method for investigating temperature transitions in this system.

Further SAXS studies were performed on aqueous mixtures of pNIPAAm homopolymers with CP and gelatin. At 25°C, a d_m value of 1.04 was extracted for this mixture (Supplementary Fig. 3a, blue), indicating rod-like polymeric chains with weak physical entanglement (Fig. 3a iii). As the temperature increased to 37°C, the d_m of the mixture shifted to a value of 3.14 (Supplementary Fig. 3a, red), signifying a transition to a collapsed, entangled globular structure with a rough surface (Fig. 3a iv). In the absence of pNIPAAm, a mixture of CP and gelatin yielded d_m values of 1.11 and 1.08 at 25°C and 37°C, respectively (Supplementary Fig. 3b), indicating an extended coil structure at both temperatures (Fig. 3a v and vi). This confirms that the observed temperature responsiveness is attributable to pNIPAAm.

We then proceeded to evaluate the SAXS measurements for pNIPAAm-CS to understand how the addition of chondroitin sulfate affects this transition. At 25°C, the d_m values were 2.39 and 2.17 for concentrations of 1.5% and 3% (w/v), respectively (Fig. 3c and Supplementary Fig. 2c). These findings suggest increase in internal entanglement or chain branching (Fig. 3a vii [55]) due to the chondroitin sulfate grafting [54] and conform to the expected structure of the synthesized copolymer in the swollen state at 25°C [58]. Despite this structural change, a transition from d_m = 2.39 to 2.66 was still observed for pNIPAAm-CS at 25°C and 37°C (Fig. 3c), indicating that the copolymer retained its temperature sensitivity even after the grafting with chondroitin sulfate (Fig. 3a viii). Furthermore, this transition was consistent for both concentrations (1.5% and 3% (w/v), as shown in Supplementary Fig. 2c and 2d), indicating that the transition in pNIPAAm-CS is not concentration-dependent in the concentration range investigated.

The full EXPECT composition (3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin) also showed a structural transition with temperature (d_m from 2.16 → 2.62, Fig. 3d-e) (Fig. 3a ix and x). Without gelatin, 3% (w/v) pNIPAAm-CS mixed in aqueous solution with 0.8% (w/v) CP underwent transition (d_m from 1.55 → 2.85, Fig. 3d-e). Interestingly, pNIPAAm-CS in aqueous solution with 1% (w/v) gelatin did not show any transition (d_m from 2.51 → 2.46). However, this does not necessarily indicate a complete absence of transition but rather suggests that a less pronounced molecular re-arrangement occurs, considering that both components are thermoresponsive, adding complexity to the structural events since two different molecules are involved.

The transformation of pNIPAAm-based polymers, shifting from a collapsed globule to a swollen, disentangled polymeric network upon cooling through the LCST, has been extensively documented in connection with a transition from adhesive to non-adhesive behavior towards cells and proteins [30–33]. It is suggested that this process occurs in two stages: initially, cell detachment is triggered by the hydration of polymer chains, followed by a cell shape change that relies on F-actin dynamics, facilitating further detachment [59]. This detachment phenomenon has been observed on 2D surfaces for various cell types, including endothelial cells [60, 61], hepatocytes [60], keratinocytes [62], cardiomyocytes [63, 64], tenogenic and osteogenic progenitor cells [34] and has also been demonstrated for extracellular matrix (ECM) proteins [65, 66]. We anticipate that these thermosensitive properties, as validated through our combined rheological and SAXS data, play a crucial role in orchestrating the aligned cellular movements of cells embedded within EXPECT, described in the following sections.

Pilot cell studies with EXPECT

We conducted two pilot studies using an L929 murine fibroblast line. In the first study, we suspended 1 × 10⁶ cells/mL in a sacrificial 6% (w/v) gelatin bioink and extruded them forming 10 mm ring-patterns embedded in the finalized composition of EXPECT (3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin). The cultures were maintained under standard conditions of 37°C, 5% CO₂, and 90% relative humidity for 14 days to assess the biocompatibility of the embedding medium.

Using fluorescent confocal imaging, we observed TRITC-conjugated phalloidin stained cytoskeleton (red) and DAPI-stained nuclei (blue, Supplementary Fig. 4a). Between days 1 and 7, we noticed a tendency for cell spreading and alignment of the F-actin cytoskeleton. This coincided with an increase in metabolic activity (Supplementary Fig. 4b) and DAPI count (Supplementary Fig. 4c, right axis). However, between days 7 and 14, although the DAPI count did not increase, we observed a notable increase in the disorganization of the patterns, with increased spacing between the cells. To quantify the changes in the pattern width, we converted the DAPI/phalloidin images to grayscale, manually rotated the images to align the major axis of the printed structure to the y-direction and summed the pixel intensity along the column (i.e., the y-axis of the images). From the intensity data, we measured the width of the columnar pattern (i.e., width of the peak) at each time point. We found a statistically significant increase from 212 ± 78 µm at day 7 to 499 ± 70 µm at day 14 (Supplementary Fig. 4c, left axis, p < 0.0001). Also observed in previous reported studies, we postulate that this increase in pattern width over the culture period reflects stochastic movements of the cells in vitro and which gradually worsens the initially placed patterns [7, 14, 16].

To address the issue of pattern maintenance over the culture period, our subsequent study focused on investigating whether the temperature-sensitive behavior of EXPECT could offer a solution. In the next pilot study, we extruded 4 × 10⁶ cell/mL fibroblasts in the same ring pattern. After fabrication, a temperature-actuated study group was designated to undergo a cooling cycle to 25°C for 15 minutes at days 1, 3, 5 and then once every 5 days thereafter, while otherwise maintaining normal culture conditions for a 21-day culture period. As a comparison, a static group was kept at normal culture conditions throughout the study.

Co-images of DAPI and phalloidin from day 21 (Fig. 4a) qualitatively showed that the cells in the temperature-actuated group tended to aggregate more. Quantitative analysis of the imaging data revealed that cell area was slightly lower under temperature-actuated compared to static conditions (Fig. 4b), possibly an indicator of increased cell spreading under static conditions. This behavior aligned with our hypothesis that EXPECT is cell-adhesive at 37°C. Additionally, we segmented the nuclei present in a single image through StarDist plugin for ImageJ [67], extracted the center of mass of each nucleus, and used these coordinates to calculate the nearest neighbor to each particle (Fig. 4c). The results showed that the distance between the nearest neighboring particles was smaller, and the distribution was tighter for the temperature-actuated group compared to the static group. Based on these data, we anticipated that temperature actuation could be useful in minimizing the accumulated migratory distances between cells. We decided to continue our investigations by applying this method with primary bone-marrow derived MSCs to explore if it reduced the migratory distance in this cell type.

Temperature-actuated aligned intercalation of MSC single cells

We next prepared a single cell suspension of 4 × 10⁶ cell/mL MSCs in 6% (w/v) gelatin, 3D printed the same free-floating ring pattern into EXPECT, and maintained the constructs under static or temperature-actuated conditions for 36 days. Temperature actuation was introduced on days 1, 3, 5, and subsequently every 5 days. Co-images, combining DAPI and phalloidin, were captured on days 0, 7, and 36 post-printing. The experimental timeline is illustrated in Fig. 5a.

From fluorescence imaging on day 1, immediately after printing but prior to temperature actuation, the cells were distributed across a pattern width of 390 ± 114 µm. Under static conditions, by day 36, we observed some reductions in cellular organization along the channel. The pattern width significantly increased to 537 ± 76 µm (p = 0.0235) and appeared less regular (Fig. 5b). By adding temperature-actuation to EXPECT, we observed a striking difference in cell behavior. By day 7, the cells tended to self-organize into longitudinal patterns along the embedded channels, and this configuration was maintained up to day 36. To analyze this further, we conducted image analysis to measure the width of the columnar patterns at each time point. As can be observed from the day 7 data depicted in Fig. 5c, the narrower distribution of the curves for temperature-actuated conditions indicated a higher degree of anisotropy compared to static conditions. Additionally, we normalized the maximum width of the columnar pattern (Fig. 5d) at days 7 and 36 for each sample group to day 0, immediately after printing. Interestingly, under temperature-actuated conditions, pattern width was reduced by approximately 50% by day 7, and this reduction was sustained at day 36. In contrast, when examined under static conditions, we noted a 115 ± 20% increase in pattern width by day 7 and a further 137 ± 20% increase by day 36 compared to day 0 (p < 0.001). These findings were consistent in two additional independent experiments, one conducted with an extra donor (see Supplementary Fig. 5).

In summary, we successfully induced and sustained lateral intercalation of cells within the EXPECT system, as evidenced by the reduced width of their columnar patterns. Building upon previous findings in pNIPAAm-based systems [30–33], we posit that that the observed temperature-induced effects are associated with the potential periodic release of cells and their endogenously expressed proteins, including chemoattractants, from the surface of the pNIPAAm-based channel, combining to decrease lateral migration away from the channels. Changes in the viscoelastic properties or fluid dynamics resulting from temperature-induced alterations in the polymer network also may have played a role in guiding and sustaining cells in the longitudinal pattern.

In the realm of tissue engineering, the significance of cell proximity and 3D patterns for tissue functionality is well acknowledged. While various methods, such as non-adhesive micro-patterned surfaces [68, 69], acoustic waves [9, 10], or pressure [70, 71] have been developed to guide initial cell assembly, limited attention has been given to the maintenance of these spatial arrangements. Singular applications or static maintenance of these cues are insufficient to drive cell movements to condense along a predetermined axis of alignment over extended in vitro culture periods beyond 7 days [7, 14, 16, 18, 72]. In contrast, our approach introduces a contactless method, leveraging materials design and minor temperature changes, that can be applied periodically during culture to drive and maintain cellular assembly in desired patterns.

This innovation contributes to the field of tissue engineering by addressing the nuanced challenge of maintaining functional patterning and offering a novel perspective for gaining mechanistic insights into the impact of spatial factors on functional bioassembly across prolonged culture periods. Notably, our approach has also led to the formation of cell distributions with lateral widths smaller than those produced by the 3D printer nozzle. Consequently, we propose this method as an enhancement to improve the resolution of extrusion-based 3D printing, one of the most widely utilized and adaptable techniques in biomanufacturing.

Temperature-actuated convergent extension and bioassembly of MSC spherical aggregates

We further investigated the effect of EXPECT temperature actuation on MSCs assembled into spherical aggregates of approximately 200,000 cells using AggreWell™ plates. Using the same cell density (4 × 10⁶ cell/mL) and ring size, we compared the changes in aggregate roundness, pattern width, and length under static versus temperature-actuated conditions. On the first day after bioprinting, the individual aggregates appeared slightly oblong, probably due to the mechanical pressure during the bioprinting. Over the 36-day culture period, we found no statistically significant changes in aggregate roundness or pattern width under static conditions (p > 0.48). However, we did observe some dissociation among the aggregates during the culture period, as they appeared looser over time (Fig. 6a). There was no measurable evidence of the aggregates migrating towards each other or fusing, as the average longitudinal length of any continuous pattern remained approximately 200 µm throughout the culture duration (p > 0.999). These outcomes are in line with our expectations and correspond with the current findings in controlled assembly [73–76], where aggregates are typically placed adjacent to each other to achieve fusion, whereas, in our study, we spaced them apart at the start of the experiment.

In contrast, under temperature-actuated conditions, aggregate roundness decreased from 0.77 ± 0.14 µm to 0.46 ± 0.10 µm between days 1 and 7 (p < 0.0001). Moreover, pattern width tended to decrease within this same time period from 181 ± 33 µm to 156 ± 46 µm (p = 0.64). These changes in shape descriptors denote the narrowing of the cellular patterns, consistent with what we observed by day 7 in the previous study with single cells. Additionally, the aggregates took on a bipedal morphology by day 7 (Fig. 6a, yellow arrows). Remarkably, with temperature actuation, the shape changes were measured again on day 36, with mean roundness further decreasing to 0.23 ± 0.14 (p < 0.0001) and pattern width tending to decrease to 119 ± 22 µm (p = 0.18). Additionally, we measured statistically significant increases in the longitudinal length of the continuous cellular pattern over the 36-day culture period from 206 ± 44 µm at day 0 to 480 ± 158 µm at day 36 (p < 0.0001). This change is attributable to the elongation of the individual aggregates, which enabled instances of fusion of the aggregates at day 36.

The temperature-actuation of EXPECT exerted a comparable influence on both cell aggregates and individual cells, facilitating organized self-assembly along a specific axis. Under temperature actuation, spheroidal groups of MSCs underwent reshaping, adopting a bipedal morphology akin to what has been documented on 2D gel surfaces—a characteristic associated with directionally migrating spheroids [18]. Importantly, the persistent cell motility observed in the referenced study was maintained for 1–6 hours. The presence of a bipedal morphology in cell aggregates with temperature actuation until day 36, in contrast to the isotropic disorganization observed under static conditions, suggests that temperature actuation sustained the directional migratory behavior of MSCs.

Within the domain of biofabrication, there has been a notable emphasis on employing bottom-up [77, 78] and middle-out [79] assembly strategies that utilize spheroidal aggregates as foundational units for constructing larger musculoskeletal or organ structures. In these approaches, spheroidal cell aggregates, often MSCs [9, 51, 80, 81] are strategically positioned within hydrogels, either through embedded 3D printing or acoustic vibration, with the goal of promoting fusion through spatial proximity. However, achieving precise geometric fusion poses inherent challenges for several reasons. Firstly, it requires a high level of positional accuracy for the aggregates, placing substantial technical demands on the biofabrication process. Additionally, spherical tissue building blocks inherently exhibit isotropic and amorphous qualities and are susceptible to random disassociation over time [16, 18, 72, 82]. As a result, the methodology developed in this work has the potential to enhance anisotropic directional microorganization in bottom-up and middle-out strategies intended to scale up to larger tissues.

The temperature-sensitive characteristics of pNIPAAm have been explored in conjunction with various cell types and diverse copolymer chemistries [30, 34, 35, 83]. As a result, the applications of this method can potentially extend beyond MSCs and fibroblasts, exerting a broader influence in multiple areas of tissue engineering. The ability to achieve directional cell orchestration and maintain spatial axial assembly holds significant promise for enhancing signal transmission efficiency and facilitating the complex maturation of tissues with axial organization, including those involved in tendon, bone, neural, and cardiovascular repair. Going beyond its applications in tissue engineering, our approach involves initiating and dynamically sustaining cellular movements reminiscent of convergent extension [6] and directed cell migration [84]. Consequently, the methodology in this study offers a valuable model framework for exploring developmental mechanisms, particularly those linked to axial elongation. Moreover, it holds promise for investigating diseases or repair mechanisms where the directed chemotaxis of cells is crucial, such as in the case of cancer metastasis.

This work introduces freeform printing and temperature actuation to guide cell spatial communication between cells and cell collectives. The EXPECT hydrogel, composed of a blend of pNIPAAm-CS, gelatin, and CP enabled us to create embedded channels of cell-laden sacrificial bioink within the hydrogel. A short temperature actuation consisting of 15-minute cooling intervals to 25°C periodically throughout the culture period brought evident modulation of fibroblasts and MSCs. Temperature actuation directed cellular intercalation, longitudinal extension, and end-to-end fusion into columnar-like cell patterns that remained stable over a 36-day culture period. Our research represents a significant advancement in biomaterials and biofabrication, offering a versatile method to both coordinate and sustain collective cell communication and movements along a preferred axis beyond the developmental niche. By overcoming a longstanding limitation in in vitro 3D culture, this work holds implications for advancing bioassembly techniques for aligned tissues. Furthermore, it lays the groundwork for modeling developmental mechanisms and investigating diseases and repair processes where directed chemotaxis is a critical factor.

Poly(N-isopropylacrylamide)-g-chondroitin sulfate Synthesis

N-isopropylacrylamide (NIPAAm) monomer (Acros Organics, Geel, Belgium) was dissolved in n-Hexane at 60°C and recrystallized at 4°C, in order to remove stabilizers. Chondroitin 4-sulfate sodium salt (CS, from bovine trachea, Sigma, St. Louis, USA) was functionalized with methacrylic anhydride (MA) according to the protocol established by Bryant et al. [85], using a molar ratio of 25:1 (MA:CS). The functionalization degree of CS was assessed by ¹H-NMR analysis according to Wang et al. [86] and found to be 35.4%. Free radical polymerization of pNIPAAm-CS was performed as described previously [28, 29, 87] with a molar ratio of NIPAAm:CS of 2500:1.

Preparation of EXPECT Hydrogels

Aqueous solutions of 1%, 3%, and 5% (w/v) pNIPAAm-CS + 0.8% (w/v) Carbopol® 940 (CP, Acros Organics Geel, Belgium), as well as 3% (w/v) pNIPAAm alone, and 3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin (from porcine skin, 175 g Bloom, type A, Sigma, St. Louis, USA) were prepared. To produce the hydrogel for cell culture, the required reagents were UV-sterilized for 12 hours, and an aqueous solution of 3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin was prepared. For all formulations, the pH was adjusted to 7.4 with 50% sodium hydroxide (NaOH) and the hydrogel was stored at 4°C until testing.

Rheological Characterization

Rheological testing was conducted with an Anton Paar MCR-302 rheometer (Anton Paar GmbH, Austria). All measurements were taken with a gap size of 0.5 mm and a parallel plate with a diameter of 25 mm. Unless stated otherwise, the rheological properties were assessed at a temperature of 25°C, with a frequency of 1 Hz and with n = 3 samples per group. To prevent sample drying during the measurements, silicone oil was applied around the plate to seal the hydrogel. A 0.8% (w/v) CP hydrogel served as a control group for all tests.

Amplitude Sweep Test

An amplitude sweep test with oscillatory strain was used to determine storage modulus (G') and loss modulus (G'') within the range of 0.01% and 1000% strain. The flow stress was calculated at the point of the curves cross-over (G' = G'') using the integrated analysis tool of the RheoCompass^™ software (version 1.26.22, Anton Paar GmbH, Austria).

Temperature Ramp Test

An oscillatory temperature ramp test with 1% constant strain (within linear viscoelastic region (LVR) as established by primary amplitude test) was used to evaluate viscoelastic properties as a function of temperature. The test was divided into three intervals, the first of which was a 60 s measurement at 25°C. After this, the temperature was increased to 37°C in the second interval for 600 s. In the third interval, the temperature was reduced to 25°C and the measurements were taken for 300 s. To calculate the absolute increase of G' caused by the increase of temperature from 25°C to 37°C, the following equation was used:

$$\varDelta {G}^{{\prime }}= {G}_{t=y s}^{{\prime }}- {G}_{t=x s}^{{\prime }}$$

where ${G}_{t=y s}^{{\prime }}$ defines the 4th data point of the first interval and ${G}_{t=x s}^{{\prime }}$the 30th data point of the second interval.

Step Strain Test (Recovery)

In an oscillatory step strain test, G' and G'' were measured while the strain was alternated between a low strain, set to 1%, and a high strain, set to 250%, over the course of seven intervals. The step strain test was started with 1% strain in the 1st interval, followed by 250% strain in the 2nd interval, and so on. Each interval was 120 s, except the 7th, which lasted 160 s at 1% strain.

Print Fidelity Characterization

Freeform 3D printing was performed with a 3D Discovery™ Printer and BioCAD software (version 1.1–12) from RegenHu Ltd. (Villaz-St-Pierre, Switzerland) using a PTFE-lined needle with 0.3 mm inner diameter (Nordson EFD, Westlake, USA). An acellular, 6% (w/v) gelatin bioink was extruded into patternable hydrogels composed of 1%, 3% or 5% (w/v) pNIPAAm-CS + 0.8% (w/v) CP (supplementary Fig. 1) or 3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin (Fig. 2), in a horizontal grid geometry to evaluate print fidelity. The bioink was deposited at a height of 2 mm above the bottom inside the patternable hydrogel at a velocity of 0.5 mm/s, and with a pressure of 0.5–1.5 bar. For visualization of the deposited pattern, the bioink was supplemented with 0.1% (w/v) Coomassie Blue dye. A 1% (w/v) CP solution and a glass slide served as controls. Print fidelity was quantitatively evaluated from digital, and optical microscopy photos by taking n > 30 measurements of strand diameter per group, and n = 6 measurements per group of both angle, and surface area, with ImageJ (version 1.53c, National Institutes of Health (NIH), MD, USA).

Small angle X-ray scattering (SAXS) Analysis

SAXS was performed at B21 beamline, Diamond Light Source, UK [88, 89] with sample loading into MPS sticks enclosed with Kapton tape [89, 90]. Briefly, samples were collected and measured at 25 or 37°C, for 21 frames with 1 s exposure each.

Culture of L929 Murine Fibroblasts

L929 murine fibroblasts (85011425, mouse C3H/An connective tissue, Sigma, St. Louis, USA) were expanded in Low Glucose Dulbecco's Modified Eagle Medium (Gibco, Carlsbad, USA), supplemented with 3.7 g/L sodium bicarbonate (Sigma-Aldrich, St. Louis, USA), 1% (v/v) Penicillin/Streptomycin (Pen/Strep, Gibco, Carlsbad, USA) and 10% (v/v) fetal bovine serum (FBS, Corning Incorporated, New York, USA). Cells were cultured at 37°C with 5% CO₂ and 90% relative humidity. The expansion medium was changed three times per week and cells were passaged at 80% confluency and used at passage 15. To assess cell viability of the fibroblasts, a CellTiter-Blue® metabolic activity assay (CTB, Promega Corporation, Madison, USA) was performed at days 0, 1, 7 and 14 with n = 4 hydrogels according to the manufacturer protocol. The read-out was done with a Tecan Infinite M Plex plate reader (Tecan Group Ltd., Männedorf, Switzerland) at 590 nm fluorescence.

Expansion of human mesenchymal stromal cells

Both studies presented in the main body of the paper were performed using cells isolated from the same donor (male, born in 1958), at University Hospital of Freiburg in Germany with the full approval of the patient. Furthermore, for supplementary Fig. 5, an additional donor was evaluated (female, born in 1961), from the same university hospital with the patient’s consent.

For the human MSC growth and expansion prior printing, Minimum Essential Medium Eagle - alpha modification (10.08g α-MEM powder, Gibco, Thermo Fischer, Switzerland and 2.2 g sodium bicarbonate (56014-1KG, Sigma-Aldrich, Switzerland) for 1L of basal medium), supplemented with 10% corning serum (FBS, Corning Incorporated, New York, USA), 1% (v/v) penicillin-streptomycin (Pen/Strep, Gibco, Carlsbad, USA), and 5 ng/ml basic Fibroblast Growth Factor (bFGF, Fitzgerald Industries International, Acton, MA, USA) was used. Cells were cultured at 37°C with 5% CO₂ and 90% relative humidity. The expansion medium was changed three times per week, cells were passaged at 80% confluency and used at passage 4 for bioprinting.

For MSC aggregate generation, AggreWell™ 400 Microwell Culture Plates (STEMCELLS technologies, Canada) were utilized. The wells were pretreated with AggreWell™ Rinsing solution for the prevention of cell adhesion to the well’s surface and the promotion of efficient spheroid formation. The cells were detached from the flasks using Trypsin-EDTA (Gibco, Thermo Fisher). The viable cells were determined by a hematocytometer and Trypan Blue exclusion, while the seeding of the cells took place at a concentration of 2 million cells per well following the manufacturer’s protocol for the aggregate generation. The cells were kept in the incubator overnight (24 hours) at 37°C with 5% CO₂ to form the spheroids prior to 3D printing.

3D printing

For freeform 3D printing, L929 murine fibroblasts (passage 15) were suspended in 6% (w/v) gelatin dissolved in the fibroblast medium described above, at a density of 1.0 × 10⁶ cells/mL (supplementary Fig. 4) or 4.0 × 10⁶ cells/mL (Fig. 4). For the MSC studies, cells (single cells or spheroidal aggregate format) were suspended in 6% (w/v) gelatin in the MSC growth medium described above at a total density of 4.0 x 10⁶ cells/mL.

The cell-laden bioinks were extruded through a PTFE-lined needle with 0.3 mm inner diameter and 12.7 mm length, inside 1 mL of the patternable hydrogel composed of 3% (w/v) pNIPAAm-CS + 0.8% (w/v) CP + 1% (w/v) gelatin. The printing pattern consisted of three circles with a radius of 5 mm, stacked in Z-direction with 1 mm spacing, and was deposited at a starting height of 2 mm inside the hydrogel. Printing parameters were as follows: 25°C ambient temperature, 16% relative humidity, 0.2–0.8 bar printing pressure, 25°C bioink temperature, and 0.5 mm/s deposition velocity. The cells were cultured in expansion medium for 14, 21 or 36 days at 37°C with 5% CO₂ and 90% relative humidity. The culture medium was changed three times per week.

For the purpose of the chondrogenic induction, chondrogenic medium was used for the MSCs post printing. The chondrogenic medium was composed of High glucose Dulbecco’s Modified Eagle Medium (Hg DMEM, Lot: 2551077,13.38 g H.g DMEM with 3.7 g of sodium bicarbonate (56014-1KG, Sigma-Aldrich, Switzerland) and 0.11 g of sodium pyruvate (P5280-25G, Sigma-Aldrich, Switzerland) for 1L basal medium), supplemented with 1% (v/v) Non-essential amino acid cell culture supplement (NEAA, Lot: 2554642, Gibco, Thermo Fischer, UK), 1% (v/v) insulin, human transferrin and selenous acid (ITS-plus,Lot: 3046004, Corning, Discovery Labware, Inc., Bedford, USA), 1% (v/v) L-Ascorbic Acid 2-phosphate sesquimagnesium salt hydrate (AA2P, A8960-5G, Sigma-Aldrich, Switzerland), 0.1% (v/v) dexamethasone (D2915-100MG, Lot: 029K1187, Sigma-Aldrich, Switzerland) and 0.1% (v/v) Transforming growth factor beta (TGFβ, Fitzgerald Industries International, USA). After the printing, the hydrogels were placed in the incubator (at 37°C with 5% CO₂ and 90% relative humidity) for about an hour for solidification and then the chondrogenic medium was added. The culture of the MSCs and spheroids within the hydrogels took place under the same conditions. The medium was changed 3 times per week.

Temperature actuation of printed EXPECT samples

Immediately post-printing, hydrogel from all groups was removed for a baseline imaging (referred to as day 0). Subsequently, on days 1, 3, and 5, and every 5 days thereafter, temperature actuation was implemented. For fibroblast studies, temperature actuation continued until day 20, with sample harvesting for imaging on day 21. In the case of MSC studies, the final temperature actuation cycle was applied until day 35, after which the hydrogels were equilibrated for 24 hours at 37°C before being harvested for imaging on day 36. The temperature actuation involved exposing culture plates to 25°C (ambient conditions outside the incubator) for 15 minutes. In contrast, static (control) samples were kept outside the incubator under the same conditions, except on a 37°C surface, to prevent the phase transition of the hydrogels.

Cytoskeletal Staining

For the evaluation of cell morphology, the samples were fixed with 4% buffered paraformaldehyde (Formafix AG, Hittnau, Switzerland) for 12 hours at 37°C and washed three times with phosphate buffered saline (PBS). Whole hydrogel samples were incubated in TRITC-conjugated Phalloidin (ActinRed^™ 555 ReadyProbes^™ Reagent, Invitrogen, Thermo Fisher Scientific, Waltham, USA) and DAPI (Sigma, St. Louis, USA). All staining steps were performed at 37°C unless stated otherwise. Samples were permeabilized with 0.25% Triton® X-100 (Sigma, St. Louis, USA) for 90 minutes. Next, the samples were incubated for 3 hours in ActinRed^™ 555 (concentration of two drops per mL, as suggested by the manufacturer) together with DAPI at 5 µg/mL in PBS. Subsequently, the samples were washed with 0.05% TWEEN® 20 (Sigma, St. Louis, USA) in PBS and stored in MilliQ water until image acquisition.

Image Acquisition and Analysis

Images of the fluorescently labelled cells were acquired with an Airyscan LSM800 Confocal Microscope (Carl Zeiss Microscopy GmbH, Germany) and the associated Zen software (Zen 2.3, blue edition, Carl Zeiss Microscopy GmbH, Germany) using the following channels: ESID-T1 (transmitted light), AF568 (561 nm) and DAPI (405nm). The average cell area was estimated with a methodology similar to the one reported by Tognato et al [91]. Briefly: images were converted to 8-bit grayscale, and an automatic threshold to channel comprising the phalloidin signal was applied, the total area of the thresholded pictures was obtained and converted to µm. The number of cells on each image was evaluated by means of cells' nuclei segmented and counted through StarDist and particle analyser plugin in Fiji. Lastly, the total area obtained from each image was divided by the number of nuclei in each field of view obtaining an average cell area. Nuclei locations were also employed for the particle nearest neighbour estimation, through an automated Python script. The channels width was estimated as follows: we converted the DAPI/Phalloidin images to 8-bit grayscale, and we rotated each image such that the printed structure is aligned to the vertical direction of the image (i.e., y-axis of the image), and summed the pixels intensity along the y-axis of the images. From the intensity-distance plot, we obtained the width of the printed structure (i.e., width of the peak) at each time point. For determination of the length of the patterning, the same procedure was followed, with the exception that the images were rotated to align with the x-axis. All the images were analysed using imageJ Fiji (1.53q NIH, US) based on java 1.8.0_322 (64-bit).

Statistical Analysis

GraphPad Prism 9 (version 9.3.1, GraphPad Software, LLC, San Diego, USA) was used for the graphical visualization and statistical analysis of the data. The quantitative results were tested for normality and lognormality. Subsequently, normally distributed results were evaluated for their statistical significance with one- or two-way analysis of variance (ANOVA), respectively. Multiple comparisons were analyzed post-hoc with a Tukey Honest Significance Difference Test. Quantitative results are presented as bar plots with mean value and standard deviation, or as XY graphs. P-values p < 0.05 were considered significant, significance levels are as follows: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001.

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. DL and AP acknowledge financial support from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 857287 (BBCE – Baltic Biomaterials Centre of Excellence). JRW acknowledges grant support from a private foundation that wishes to remain anonymous.

Kronenberg HM (2003) Developmental regulation of the growth plate. Nature 423(6937):332–336
Abad V et al (2002) The role of the resting zone in growth plate chondrogenesis. Endocrinology 143(5):1851–1857
Kalson NS et al (2015) A structure-based extracellular matrix expansion mechanism of fibrous tissue growth. Elife, 4
Richardson SH et al (2007) Tendon development requires regulation of cell condensation and cell shape via cadherin-11-mediated cell-cell junctions. Mol Cell Biol 27(17):6218–6228
Garzon-Alvarado DA, Garcia-Aznar JM, Doblare M (2009) A reaction-diffusion model for long bones growth. Biomech Model Mechanobiol 8(5):381–395
Wallingford JB, Fraser SE, Harland RM (2002) Convergent extension: the molecular control of polarized cell movement during embryonic development. Dev Cell 2(6):695–706
Sriphutkiat Y, Kasetsirikul S, Ketpun D, Zhou Y (2019) Cell alignment and accumulation using acoustic nozzle for bioprinting. Sci Rep 9(1):17774
Tognato R, Armiento AR, Bonfrate V, Levato R, Malda J, Alini M, Eglin D, Giancane G (2019) Serra, T., stimuli-responsive nanocomposite for 3D anisotropic cell‐guidance and magnetic soft robotics. Adv Funct Mater 29:1804647
Petta D et al (2020) Sound-induced morphogenesis of multicellular systems for rapid orchestration of vascular networks. Biofabrication, 13(1)
Tognato R et al (2023) Sound-based assembly of three-dimensional cellularized and acellularized constructs. Mater Today Bio 22:100775
Li Q et al (2020) Programmed magnetic manipulation of vesicles into spatially coded prototissue architectures arrays. Nat Commun 11(1):232
Shi W, Huang J, Fang R, Liu M (2020) Imparting Functionality to the Hydrogel by Magnetic-Field-Induced Nano-assembly and Macro-response. ACS Appl Mater Interfaces 12(5):5177–5194
Wychowaniec JK, Brougham DF (2022) Emerging Magnetic Fabrication Technologies Provide Controllable Hierarchically-Structured Biomaterials and Stimulus Response for Biomedical Applications. Adv Sci (Weinh) 9(34):e2202278
Armstrong JPK et al (2018) Engineering Anisotropic Muscle Tissue using Acoustic Cell Patterning. Adv Mater 30(43):e1802649
Treacy NJ et al (2023) Growth and differentiation of human induced pluripotent stem cell (hiPSC)-derived kidney organoids using fully synthetic peptide hydrogels. Bioact Mater 21:142–156
Marikawa Y, Tamashiro DA, Fujita TC, Alarcon VB (2009) Aggregated P19 mouse embryonal carcinoma cells as a simple in vitro model to study the molecular regulations of mesoderm formation and axial elongation morphogenesis. Genesis 47(2):93–106
Gogele C et al (2021) Cyclically stretched ACL fibroblasts emigrating from spheroids adapt their cytoskeleton and ligament-related expression profile. Cell Tissue Res 384(3):675–690
Beaune G et al (2018) Spontaneous migration of cellular aggregates from giant keratocytes to running spheroids. Proc Natl Acad Sci U S A 115(51):12926–12931
Jebeli M et al (2023) Multicellular aligned bands disrupt global collective cell behavior. Acta Biomater 163:117–130
Poukkula M, Cliffe A, Changede R, Rorth P (2011) Cell behaviors regulated by guidance cues in collective migration of border cells. J Cell Biol 192(3):513–524
Tweedy L, Knecht DA, Mackay GM, Insall RH (2016) Self-Generated Chemoattractant Gradients: Attractant Depletion Extends the Range and Robustness of Chemotaxis. PLoS Biol 14(3):e1002404
Shellard A, Mayor R (2019) Supracellular migration - beyond collective cell migration. J Cell Sci, 132(8)
Huang S (2016) Where to Go: Breaking the Symmetry in Cell Motility. PLoS Biol 14(5):e1002463
Wong JY, Velasco A, Rajagopalan P, Pham Q (2003) Directed movement of vascular smooth muscle cells on gradient-compliant hydrogels. Langmuir 19(5):1908–1913
Perez MA, Prendergast PJ (2007) Random-walk models of cell dispersal included in mechanobiological simulations of tissue differentiation. J Biomech 40(10):2244–2253
Clark P et al (1987) Topographical control of cell behaviour. I. Simple step cues. Development 99(3):439–448
Belotti Y, McGloin D, Weijer CJ (2020) Chemotaxis overrides barotaxis during directional decision-making in Dictyostelium discoideum. bioRxiv, : p. 2020-01
Christiani T et al (2021) Using embedded alginate microparticles to tune the properties of in situ forming poly(N-isopropylacrylamide)-graft-chondroitin sulfate bioadhesive hydrogels for replacement and repair of the nucleus pulposus of the intervertebral disc. JOR Spine 4(3):e1161
Wiltsey C et al (2015) Thermogelling bioadhesive scaffolds for intervertebral disk tissue engineering: preliminary in vitro comparison of aldehyde-based versus alginate microparticle-mediated adhesion. Acta Biomater 16:71–80
Chen YS et al (2013) Poly(N-isopropylacrylamide) hydrogels with interpenetrating multiwalled carbon nanotubes for cell sheet engineering. Biomaterials 34(30):7328–7334
Choi A et al (2020) Rapid harvesting of stem cell sheets by thermoresponsive bulk poly(N-isopropylacrylamide) (PNIPAAm) nanotopography. Biomater Sci 8(19):5260–5270
Madathil BK, Kumar PR, Kumary TV (2014) N-isopropylacrylamide-co-glycidylmethacrylate as a thermoresponsive substrate for corneal endothelial cell sheet engineering. Biomed Res Int, 2014: p. 450672
Lin JB et al (2012) Thermo-responsive poly(N-isopropylacrylamide) grafted onto microtextured poly(dimethylsiloxane) for aligned cell sheet engineering. Colloids Surf B Biointerfaces 99:108–115
Berntsen L, Forghani A, Hayes DJ (2022) Mesenchymal Stem Cell Sheets for Engineering of the Tendon-Bone Interface. Tissue Eng Part A 28(7–8):341–352
Kim H et al (2019) Recent Advances in Engineered Stem Cell-Derived Cell Sheets for Tissue Regeneration. Polym (Basel), 11(2)
Litowczenko J et al (2021) Synthesis, characterization and in vitro cytotoxicity studies of poly-N-isopropyl acrylamide gel nanoparticles and films. Mater Sci Eng C Mater Biol Appl 118:111507
Shiwarski DJ, Hudson AR, Tashman JW, Feinberg AW (2021) Emergence of FRESH 3D printing as a platform for advanced tissue biofabrication. APL Bioeng 5(1):010904
Mirdamadi E et al (2020) FRESH 3D Bioprinting a Full-Size Model of the Human Heart. ACS Biomater Sci Eng 6(11):6453–6459
Dasgupta Q, Black LD 3 (2019) A FRESH SLATE for 3D bioprinting. Science 365(6452):446–447
Skylar-Scott MA et al (2019) Biomanufacturing of organ-specific tissues with high cellular density and embedded vascular channels. Sci Adv 5(9):eaaw2459
Wu W, DeConinck A, Lewis JA (2011) Omnidirectional printing of 3D microvascular networks. Adv Mater 23(24):H178–H183
Ozbolat V, Dey M, Ayan B, Ozbolat IT (2019) Extrusion-based printing of sacrificial Carbopol ink for fabrication of microfluidic devices. Biofabrication 11(3):034101
Hua W et al (2021) Fluid Bath-Assisted 3D Printing for Biomedical Applications: From Pre- to Postprinting Stages. ACS Biomater Sci Eng 7(10):4736–4756
Hinton TJ et al (2016) 3D Printing PDMS Elastomer in a Hydrophilic Support Bath via Freeform Reversible Embedding. ACS Biomater Sci Eng 2(10):1781–1786
Bhattacharjee T et al (2016) Liquid-like Solids Support Cells in 3D. ACS Biomater Sci Eng 2(10):1787–1795
Jin Y, Compaan A, Bhattacharjee T, Huang Y (2016) Granular gel support-enabled extrusion of three-dimensional alginate and cellular structures. Biofabrication 8(2):025016
Ning L et al (2020) Embedded 3D Bioprinting of Gelatin Methacryloyl-Based Constructs with Highly Tunable Structural Fidelity. ACS Appl Mater Interfaces 12(40):44563–44577
Barry BWaM (1979) The rheological properties of carbopol gels I. Continuous shear and creep properties of carbopol gels. Int J Pharm 2(1):1–25
Kim JY, Song JY, Lee EJ, Park SK (2003) Rheological properties and microstructures of Carbopol gel network system. Colloid Polym Sci 281(7):614–623
Ayan B et al (2020) Aspiration-assisted freeform bioprinting of prefabricated tissue spheroids in a yield-stress gel. Commun Phys, 3
Daly AC, Davidson MD, Burdick JA (2021) 3D bioprinting of high cell-density heterogeneous tissue models through spheroid fusion within self-healing hydrogels. Nat Commun 12(1):753
Kokufuta MK, Sato S, Kokufuta E (2012) LCST behavior of copolymers of N-isopropylacrylamide and N-isopropylmethacrylamide in water. Colloid Polym Sci 290:1671–1681
Winnik FM (1990) Fluorescence studies of aqueous solutions of poly (N-isopropylacrylamide) below and above their LCST. Macromolecules 23(1):233–242
Christiani TR et al Synthesis of Thermogelling Poly(N-isopropylacrylamide)-graft-chondroitin Sulfate Composites with Alginate Microparticles for Tissue Engineering. J Vis Exp, 2016(116).
Glinka CJ (2000) Methods of X-Ray and Neutron Scattering in Polymer Science. Oxford University Press, New York
Deen GR, Alsted T, Richtering W, Pedersen JS (2011) Synthesis and characterization of nanogels of poly(N-isopropylacrylamide) by a combination of light and small-angle X-ray scattering. Phys Chem Chem Phys 13(8):3108–3114
Tirumala VR, Ilavsky J, Ilavsky M (2006) Effect of chemical structure on the volume-phase transition in neutral and weakly charged poly(N-alkyl(meth)acrylamide) hydrogels studied by ultrasmall-angle x-ray scattering. J Chem Phys 124(23):234911
Velychkivska N, Sedláček O, Shatan AB, Spasovová M, Filippov SK, Chahal MK, Janisova L, Brus J, Hanyková L, Hill JP, Winnik FM (2022) Phase Separation and pH-Dependent Behavior of Four-Arm Star-Shaped Porphyrin-PNIPAM4 Conjugates. Macromolecules 55(6):2109–2122
Cooperstein MA, Canavan HE (2010) Biological cell detachment from poly(N-isopropyl acrylamide) and its applications. Langmuir 26(11):7695–7707
Okano T et al (1995) Mechanism of cell detachment from temperature-modulated, hydrophilic-hydrophobic polymer surfaces. Biomaterials 16(4):297–303
Reed JA, Lucero AE, Cooperstein MA, Canavan HE (2008) The effects of cell culture parameters on cell release kinetics from thermoresponsive surfaces. J Appl Biomater Biomech 6(2):81–88
Yamato M et al (2001) Thermo-responsive culture dishes allow the intact harvest of multilayered keratinocyte sheets without dispase by reducing temperature. Tissue Eng 7(4):473–480
Fujimoto KL et al (2009) Synthesis, characterization and therapeutic efficacy of a biodegradable, thermoresponsive hydrogel designed for application in chronic infarcted myocardium. Biomaterials 30(26):4357–4368
Wang T et al (2009) Novel thermosensitive hydrogel injection inhibits post-infarct ventricle remodelling. Eur J Heart Fail 11(1):14–19
Kushida A et al (1999) Decrease in culture temperature releases monolayer endothelial cell sheets together with deposited fibronectin matrix from temperature-responsive culture surfaces. J Biomed Mater Res 45(4):355–362
Kushida A, Yamato M, Kikuchi A, Okano T (2001) Two-dimensional manipulation of differentiated Madin-Darby canine kidney (MDCK) cell sheets: the noninvasive harvest from temperature-responsive culture dishes and transfer to other surfaces. J Biomed Mater Res 54(1):37–46
Schmidt U, Weigert M, Broaddus C, Myers G (2018) Cell Detection with Star-Convex Polygons, in Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).
Napolitano AP, Chai P, Dean DM, Morgan JR (2007) Dynamics of the self-assembly of complex cellular aggregates on micromolded nonadhesive hydrogels. Tissue Eng 13(8):2087–2094
Livoti CM, Morgan JR (2010) Self-assembly and tissue fusion of toroid-shaped minimal building units. Tissue Eng Part A 16(6):2051–2061
Timmins NE, Nielsen LK (2007) Generation of multicellular tumor spheroids by the hanging-drop method. Methods Mol Med 140:141–151
Alhasan L et al (2016) Rapid Enhancement of Cellular Spheroid Assembly by Acoustically Driven Microcentrifugation. ACS Biomater Sci Eng 2(6):1013–1022
Gogele C et al (2021) Correction to: Cyclically stretched ACL fibroblasts emigrating from spheroids adapt their cytoskeleton and ligament-related expression profile. Cell Tissue Res 384(3):789
Li YE et al (2020) Toward a neurospheroid niche model: optimizing embedded 3D bioprinting for fabrication of neurospheroid brain-like co-culture constructs. Biofabrication, 13(1)
Burdis R et al (2022) Spatial patterning of phenotypically distinct microtissues to engineer osteochondral grafts for biological joint resurfacing. Biomaterials 289:121750
Roth JG et al (2023) Spatially controlled construction of assembloids using bioprinting. Nat Commun 14(1):4346
Caprio ND, Burdick JA (2023) Engineered biomaterials to guide spheroid formation, function, and fabrication into 3D tissue constructs. Acta Biomater 165:4–18
Guimaraes CF et al (2022) Engineered living bioassemblies for biomedical and functional material applications. Curr Opin Biotechnol 77:102756
Burdis R, Kelly DJ (2021) Biofabrication and bioprinting using cellular aggregates, microtissues and organoids for the engineering of musculoskeletal tissues. Acta Biomater 126:1–14
Blatchley MR, Anseth KS (2023) Middle-out methods for spatiotemporal tissue engineering of organoids. Nat Rev Bioeng 1(5):329–345
Nilsson Hall G et al (2021) Cartilaginous spheroid-assembly design considerations for endochondral ossification: towards robotic-driven biomanufacturing. Biofabrication, 13(4)
Kim MH, Banerjee D, Celik N, Ozbolat IT (2022) Aspiration-assisted freeform bioprinting of mesenchymal stem cell spheroids within alginate microgels Biofabrication, 14(2)
Beckingham LJ et al (2019) Three-dimensional cell culture can be regulated by vibration: low-frequency vibration increases the size of olfactory ensheathing cell spheroids. J Biol Eng 13:41
da Silva RM et al (2008) Poly(N-isopropylacrylamide) surface-grafted chitosan membranes as a new substrate for cell sheet engineering and manipulation. Biotechnol Bioeng 101(6):1321–1331
Reig G, Pulgar E, Concha ML (2014) Cell migration: from tissue culture to embryos. Development 141(10):1999–2013
Bryant SJ et al (2004) Synthesis and Characterization of Photopolymerized Multifunctional Hydrogels: Water-Soluble Poly(Vinyl Alcohol) and Chondroitin Sulfate Macromers for Chondrocyte Encapsulation. Macromolecules 37:6726–6733
Wang L, Shen S, Lu S (2003) Synthesis and characterization of chondroitin sulfate–methacrylate hydrogels. Carbohydr Polym 52:389–396
Wiltsey C et al (2013) Characterization of injectable hydrogels based on poly(N-isopropylacrylamide)-g-chondroitin sulfate with adhesive properties for nucleus pulposus tissue engineering. J Mater Sci Mater Med 24(4):837–847
Cowieson NP, Edwards-Gayle CJC, Inoue K, Khunti NS, Doutch J, Williams E (2020) Beamline B21: High-throughput small-angle X-ray scattering at diamond light source. J Synchrotron Radiat 27(5):1438–1446
Sceglovs A, Wychowaniec JK, Skadins I, Reinis A, Edwards-Gayle CJ, D’Este M, Salma-Ancane K (2023) Effect of steam sterilisation on physico-chemical properties of antibacterial covalently cross-linked ε-polylysine/hyaluronic acid hydrogels Carbohydrate Polymer Technologies and Applications, 6: p. 100363
Hamley CKN, Inoue IW, Cowieson K, N., and, Rambo R (2021) Design of a multipurpose sample cell holder for the Diamond Light Source high-throughput SAXS beamline B21. J Synchrotron Radiat 28:318–321
Tognato R, Bonfrate V, Giancane G, Serra T (2022) Fabrication of anisotropic collagen-based substrates for potential use in tissue engineering. Smart Mater Struct 31(7):074001

There is NO Competing Interest.

SupplementaryFiguresrev1.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Harnessing Thermogel Actuation for Driving Directional Stromal Cell Communication and Migration into Columnar Arrays

Status:

Version 1

Abstract

Figures

Introduction

Results and Discussion

Establishing the composition of EXPECT

Rheological properties of EXPECT

Small Angle X-ray Scattering (SAXS) analysis of EXPECT

Pilot cell studies with EXPECT

Temperature-actuated aligned intercalation of MSC single cells

Temperature-actuated convergent extension and bioassembly of MSC spherical aggregates

Conclusions

Materials and Methods

Poly(N-isopropylacrylamide)-g-chondroitin sulfate Synthesis

Preparation of EXPECT Hydrogels

Rheological Characterization

Print Fidelity Characterization

Small angle X-ray scattering (SAXS) Analysis

Culture of L929 Murine Fibroblasts

Expansion of human mesenchymal stromal cells

3D printing

Temperature actuation of printed EXPECT samples

Cytoskeletal Staining

Image Acquisition and Analysis

Statistical Analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1