Novel hydrogel system eliminates subculturing and improves retention of nonsenescent mesenchymal stem cell populations

Mesenchymal stem/stromal cells (MSCs) are a heterogeneous population of multipotent progenitor cells that are found in a variety of tissue sources within the human body, including bone marrow, dental pulp, umbilical cord and adipose tissue [1]. MSCs have earned a spotlight in the fields of tissue engineering and regenerative medicine for their intrinsically diverse regenerative and secretory properties [2]. The multipotent nature of MSCs originally garnered immense interest in the field of biomedical engineering due to the capacity for MSCs to differentiate into osteogenic, chondrogenic, adipogenic, nervous, skeletal and cardiac lineages to potentially form a neotissue or whole-organ replacement for diseased and damaged tissue within the body [1–3]. These types of therapies have often focused on the incorporation of MSCs with scaffolds and hydrogels of various complexities in order to control the differentiation of MSCs toward specific lineages.

Other MSC-based therapies that are currently being investigated include endovascular or direct injection of MSCs into locations of damaged tissue [2]. These therapies require further optimization due to cell death and radial diffusion of MSCs from the sites of injection, although ongoing research into delivery vehicles for MSC therapies such as scaffolds and hydrogel encapsulation have proven to be promising next steps in improving their efficiency and efficacy [4–9]. Similarly, another MSC-based therapy that has garnered interest involves the utilization of MSC-derived secretory products, called the secretome, as a form of acellular MSC therapy that circumvents some of the common limitations of cell-based therapies. MSCs’ secreted byproducts include a heterogeneous variety of biomodulatory factors such as proteins, antioxidants, nucleic acids, exosomes and microvesicles, and have been shown to promote tissue regeneration and wound healing in a variety of applications [10–13]. Ultimately, for both cellular and acellular MSC-derived therapeutics, cells must be removed from a donor source and subsequently expanded in vitro to achieve sufficient cell numbers (often requiring anywhere from 10⁷ to 10⁹ MSCs at a minimum) to obtain a viable clinical product, which is associated with loss of a stem-like phenotype and often takes several weeks to achieve [14,15].

Traditionally, MSC in vitro expansion systems include the use of rigid, 2D plastic culture vessels to expand the cells as a monolayer and require multiple protease-dependent subculturing (passaging) events to achieve large enough cell quantities. Evidence suggests that traditional 2D culture systems are not ideal for stem cell expansion and can result in a loss of MSC multipotency, reduce viability, induce senescence and decrease secretion of regenerative factors [16–20]. Studies have demonstrated that stem cell differentiation and function are, in part, dependent on substrate mechanical properties [21,22], with more rigid substrates promoting osteogenic lineages and decreased viability, and softer substrates improving viability and the retention of ‘stem-like’ properties [21,23]. Thus, the decreased robustness of MSC populations seen in traditional 2D culture is likely due to a combination of the unphysiologically rigid substrate mechanics of 2D culture plastic, in addition to the overcrowding of cells in 2D monolayers and the need for continuous subculturing of cells. Ultimately, traditional 2D culture modalities likely result in less viable and more senescent cell populations over time, leading to impurities and/or an inconsistent secreted product that subsequently increases the variability between experimental assays and limits the potential downstream clinical benefits of MSC-derived therapies.

Early studies with 3D culture systems, including spheroid and organoid culture, have demonstrated their capacity to circumvent some of the limitations of 2D culture and improve the stem-like phenotype of MSC populations [24–26]. However, spheroid and organoid cultures are often limited in their size due to diffusional constraints and are prone to central regions of necrosis and cell death [27]. Additionally, spheroid/organoid culture can be labor-intensive and can require large and expensive bioreactor systems, such as stir-tank bioreactors, for long-term expansion applications that culminate in the need for artificial dissociation of the cellular clusters with proteases [24,28]. This process is especially true for industrial purposes with the goal of generating cellular or cellular-derived therapies. Moreover, a recent study investigating long-term culture of Wharton’s jelly MSCs demonstrated an increased population of senescent cells after 3D spheroid culture [29]. Thus, the development of more efficient tissue-mimetic 3D culture systems that improve the long-term expansion of robust and stem-like cell populations are currently still under investigation. One promising approach is the utilization of 3D hydrogel cultures, due in part to the tailorability of the substrate composition and mechanics [30,31]. To date, many hydrogel systems have aimed to control differentiation and/or act as a delivery vehicle of stem cells, rather than to expand MSCs and maintain their stem-like properties [32–35]. Additionally, most current hydrogel formulations lack the porous microarchitecture that aids in cellular migration and nutrient diffusion.

In this study we set out to develop a proof of concept for a novel 3D hydrogel system that improves expansion outcomes of MSC populations, such as adipose-derived MSCs (ASCs). The customizable 3D hydrogel system is bioprinted and was formulated to be a bioinert substrate that closely mimicked native adipose tissue mechanics (Supplementary Figure 1) and ultimately acts like a tailored bioreactor for MSC expansion and collection of biologics. The 3D hydrogel system was constructed with a unique ‘puzzle-piece’ macrostructure design that enables easy addition of supplementary hydrogels. Additionally, the unique porous microarchitectural design permits mass transport and promotes cellular migration and proliferation, eliminating the need to subculture cells via cellular migration between hydrogels within the microchannels. The initial utilization of a bioinert substrate allowed for the investigation and observation of the potential role of the mechanical and dimensional properties of the 3D system on ASC senescence and stem-like phenotypic properties without introducing a bioactive substrate. We hypothesized that the softer, 3D hydrogel substrate would provide a more natural mechanical environment for the ASCs and result in the retention of nonsenescent stem-like ASC populations, relative to traditional 2D culture methodologies. Moreover, the unique architectural design would allow for continuous expansion of ASCs via cellular migration between the attached hydrogels, thus eliminating exposure of cells to negative 2D subculturing procedures and subsequent sequelae.

Materials & methods

Cell culture

Human ASCs (Lot #18TL212639, 23-year-old female, Black), human keratinocytes (KCs; Lot #18TL318559, 62-year-old male, Caucasian) obtained from Lonza (Basel, Switzerland) were utilized for this study. MSC-GM Mesenchymal Stem Cell Growth Medium BulletKit™ was obtained from Lonza (#PT-3001) and used for ASCs. MesenCult™-ACF Plus Medium Kit (Stem Cell Technologies, BC, Canada; Cat. #05445) was used as the serum-free medium. DermaLife K Keratinocyte Medium Complete Kit was obtained from Lifeline Cell Technologies (MD, USA; #LL-0007) and used for KCs.

3D-printed hydrogel cell culture system

The bioinert 3D hydrogel system is approximately 1 × 1 × 1 cm and is a polyethylene glycol (PEG)-based system called a Tissue-Block (T-Block; Ronawk, KS, USA) which contains a unique microarchitectural design and was fabricated to resemble the mechanics of native adipose soft tissue (Supplementary Figure 1) and demonstrated no significant changes in mechanical properties over 3 months. The 3D hydrogels were placed into a glass six-well culture plate for culturing. Fibronectin is a commonly selected coating substrate for ASCs due to their natural secretion of fibronectin. Because the cells do not efficiently adhere/attach to the PEG-based polymer, both the 2D culture plastic/glass and 3D hydrogel were coated with fibronectin at a concentration of 5 μg/cm² to enhance the initial cell attachment. The concentration of fibronectin was standardized to surface area due to the inherent surface-area-to-volume differences between 2D and 3D systems. The approximate surface area of the 3D hydrogel was calculated from the 3D model used for bioprinting.

Expansion of ASCs & KCs

ASCs and KCs were seeded within a T-150 flask and cultured until ∼80% confluence before subculturing (passaging). Subculturing of cells was performed by removing culture media, washing thrice with Hanks’ balanced salt solution (HBSS, MA, USA; calcium-free, magnesium-free) and incubating with 0.05% Trypsin/EDTA (Lonza; Cat. #CC-3232) at 37°C for 5 min. Trypsin was neutralized with serum and the cells were centrifuged at 500 × g for 5 min, then pelleted and resuspended for reseeding on new 2D tissue culture plastic vessels. ASCs at passage 1 (P1) were reseeded onto 2D culture plastic or onto/within the bioprinted 3D hydrogel system via dropwise addition of a concentrated cell solution to the surface of the hydrogels. This process was repeated with the residual cell solution five times to ensure efficient cell seeding. This repetitive seeding process allowed for the cells to distribute throughout the microporous structure within the hydrogel system. Given that the increased surface area and attachment of additional hydrogels eliminated the need for subculturing for this study, a passage-equivalence time point was utilized to allow for analogous comparison with 2D culture. Thus, a passaging event typically occurred every 4–5 days in 2D culture for ASCs but not in 3D culture. After 4–5 days of 2D culture for P2 ASCs, the cells were then subcultured and considered to be P3 in 2D, and the 3D ASCs were then considered P3 passage-equivalent. Culture expansion was determined based on the known 2D and 3D surface areas, initial cell seeding density and average population doubling time of 2.25 days (experimentally determined in 2D; data not shown) for the ASCs in order to standardize cell numbers. ASCs were seeded at a standardized concentration of 5000 cells/cm² for assays. Media supplementation was standardized to 150 μl/cm² for ASC expansion to account for dilutional differences in surface-area-to-volume ratio between 2D and 3D culture.

ASC phenotype characterization

The MSC stem-like phenotype was evaluated for the ASCs at P1/2/6/10 via immunolabel characterization of three key MSC surface markers (CD73/90/105). ASCs were either continuously subcultured in a T-150 flask or allowed to expand within the 3D hydrogel system. At each respective passage time point, cells were seeded in 2D at a standardized density of ∼5000 cells/cm² onto a 96-well glass culture plate (Cellvis, CA, USA; Cat. #P96-1.5H-N) for 2 days, fixed, then assessed for ASC phenotype via immunolabeling for surface CD markers. For ASCs within the 3D culture system, cells were fixed and stained in situ. Positive staining for CD73/90/105 and negative staining for CD34/45 was used to denote a stem-like MSC phenotype for this study. After fixation with 4% paraformaldehyde, cells were washed thrice with HBSS and placed in blocking buffer (1% donkey serum in HBSS) for 1 h. After blocking, primary antibodies for CD73 (Abcam, MA, USA; Cat. #133582; 1:100), CD90 (Abcam; Cat. #181469; 1:100), CD105 (Abcam; Cat. #231774; 1:100), CD34 (Abcam; Cat. #81289; 1:200) or CD45 (Abcam; Cat. #40763; 1:200) were added and the cells incubated overnight at 4°C. The next day cells were washed thrice and secondary antibodies were applied for 1 h, followed by three additional washes. Cells were counterstained with Hoechst 33342 (Invitrogen, MA, USA; Cat. #H3570; 1:1000) and Alexa Fluor^® 647 Phalloidin (Invitrogen; Cat. #A22287; 1:1000). Immunofluorescence was assessed with a Revolve microscope using filters for 4′,6-diamidino-2-phenylindole (DAPI; EX-380/30, EM-450/50), fluorescein isothiocyanate (EX-470/40, EM-525/50); Texas Red (EX-560/40, EM-630/75) and Cy5 (EX-630/40, EM-700/75) (Echo, CA, USA) and 20× objective (Olympus, Tokyo, Japan; UPlanSApo, 0.75NA). ASC stem-like phenotypic quantification was carried out in quadruplicate (n = 4), with images taken from a total of ten random fields of view per biological replicate (2D = per well; 3D = per hydrogel), to achieve up to 40 total measured values for each sample. Total nuclei were counted, and total positive cells were evaluated. Phalloidin counterstain was used to aid in localization of positive staining. ASCs characterized at P1 were used as a baseline for comparison (Supplementary Figure 2).

ASC senescence characterization

Similar to the ASC phenotyping methodology above, assessment of ASC senescence was performed. ASCs were continuously subcultured in a T-150 culture flask until each respective assay time point, when they were seeded onto a 96-well glass culture plate. ASCs in the 96-well plate were allowed to acclimate in serum-based media for 2 days, fixed, then assessed for senescent activity via immunofluorescent labeling of β-galactosidase activity with the CellEvent™ Senescence Green Detection Kit (Invitrogen; Cat. #C10850), per manufacturer’s instructions. ASCs in the 3D system were assessed simultaneously at the P2/6/10 passage-equivalent time points. Similar to the method above, cells were counterstained with Hoechst and Phalloidin. Senescence characterization was carried out in quadruplicate (n = 4), with images taken from a total of ten random fields of view per biological replicate, to achieve up to 40 total measured values per sample. Total nuclei were counted, and total senescent positive and negative cells were determined.

Isolation of ASC-conditioned media

Media supplementation was standardized for ASC expansion to account for dilutional differences in surface-area-to-volume ratio between 2D and 3D cultures. Media were changed every 2 days. For ASC-conditioned medium (ASC-CM) collection, MSC-GM was removed and cells were washed with HBSS thrice, then cultured with serum-free MSC media for 48 h before collection (for both 2D and 3D cultures). Collected ASC-CM was then centrifuged at 1500 × g for 10 min to eliminate cell debris, Steriflip-filtered with a 0.22-μm filter and stored at -80°C until use.

KC activity after ASC-CM treatment

ASC-CM was collected at each respective time point per the protocol above. ASC-CM was then placed on KCs for up to 24 h to assess its capacity to modulate metabolic, proliferative or migratory activity to evaluate wound healing capabilities of the ASC-CM. For these studies, ASC-CM was added at a 1:1 ratio with KC growth medium. Experimental assays were performed per the manufacturer’s instructions. PrestoBlue fluorescence was obtained at 560/590 nm (n = 4) and used for evaluation of metabolic activity after 24 h of ASC-CM treatment. Hoechst was added to PrestoBlue samples and values were displayed as average relative fluorescence unit values of PrestoBlue/Hoechst signal in order to control for potential differences in cell numbers and obtain approximate metabolic activity per cell. PicoGreen fluorescence was obtained at 485/535 nm (n = 4) and used for evaluation of cell number as a surrogate measurement of KC proliferation after 24 h of ASC-CM culture. Total cell numbers per 96-well plate were calculated based on an average DNA content of 7.7 pg/cell. KC scratch assays were performed to evaluate changes in wound size as a surrogate measurement of KC migration (n = 3). Migration images were taken using an ImageXpress Micro XLS Imaging System (Molecular Devices, CA, USA). The entire wound was imaged for each wound triplicate and three different wound regions per wound triplicate were used to calculate wound area. The three wound area values were averaged per triplicate and per time point for each group and displayed as percentage wound area recovered (n = 3).

Statistical analysis

All data are reported as means with standard error of the mean. Characterization analyses of ASC populations with immunolabeling for senescence and CD markers were evaluated with a two-way analysis of variance. KC metabolic, proliferative and migratory activities were evaluated with a two-way analysis of variance. Data were tested for normality via Shapiro–Wilk and Kolmogorov–Smirnov tests and plotted with a QQ plot. GraphPad Prism 9.0.2 software (GraphPad, CA, USA) was used for the analyses, and a p-value < 0.05 was considered significant.

Results

Unique 3D hydrogel design eliminates subculturing

The ∼1-cm³ 3D-printed hydrogel system contains a unique ‘puzzle-piece’ macrostructure that allows the continuous addition of supplementary hydrogels (Figure 1A) and promotes the migration of cells from the primary seeded hydrogel into the newly attached hydrogels (Figure 1A & B). ASCs were cultured within/onto a single hydrogel system for 2 weeks and allowed to migrate throughout the hydrogel. Subsequently, an additional hydrogel was added for 5 days, and cells were allowed to migrate to the newly attached hydrogel. The cells were then stained and assessed for migration and proliferation between the two hydrogels (Figure 1B). ASCs were seen lining the porous channels beyond the superficial surface within the internal structure and can be seen forming networks within the hydrogel pores (Figure 1B & C). Cells were able to migrate both across the surface of attached hydrogels and within the microchannels, and can be seen with numerous cellular extensions and focal adhesions protruding from the cells to aid in migration (Figure 1C). The interaction of ASCs in 3D and the formation of 3D networks within the hydrogel microarchitecture are further highlighted in Supplementary Videos 1 & 2. Notably, ASCs can be seen lining the surface of the hydrogel and migrating down into one of the pore channels, ultimately interacting in three dimensions with other cell populations within the pore.

Retention of stem-like MSC surface markers in 3D hydrogel over time

Evaluation of stem-like ASC populations over time was performed via immunolabeling quantification of the CD markers CD73/90/105 (Figure 2A). The prevalence of stem-like CD markers declined over time in both 2D and 3D systems; however, 2D culture had a significantly enhanced rate of loss of these markers relative to the 3D system (Figure 2B). Similarly, there was an apparent decrease in intensity of positively stained cells over time in both 2D and 3D systems. Culture within the 3D system resulted in a significant retention at all passage time points for all three positive stem-like markers, except for the initial P2 comparison of CD105. The significant loss of a stem-like phenotype in 2D culture can be seen clearly when comparing ASCs in 3D at P6 and P10 versus ASCs in 2D at P2 (Figure 2B). Over the course of five or nine passage equivalents in 3D (i.e., 3 or 6 weeks in the 3D system for P6 or P10, respectively) and only one passage event in 2D (seeded at P1), the ASCs had similar expression patterns for multiple CD markers, with only a significant difference between 3D expression of CD105 at P10 relative to 2D CD105 at P2 (Figure 2B). Both systems maintained a low population (<5%) of positive-staining cells for CD34 and CD45 markers.

Delayed induction of senescence in 3D hydrogel over time

The prevalence of senescent ASC populations over time was evaluated via fluorescent labeling of β-galactosidase activity, a commonly utilized surrogate measurement of cellular senescence (Figure 3A). The baseline expression of β-galactosidase activity was ∼5% for both 2D and 3D ASCs at P2. Over the course of multiple passaging events in 2D, senescence was significantly increased in 2D, with 11.7 and 22.5% senescent cells at P6 and P10, respectively (Figure 3B). Conversely, in the 3D system there was no significant change in the prevalence of senescence over time in ASC populations for the time course of this study (6 weeks) (Figure 3B). Similarly, there were notable differences in cellular morphology in the 2D system versus the 3D system. More specifically, 2D ASCs appear to be more flattened, with a more heterogeneous morphological distribution and an apparent increase in cell size over time in culture, whereas the 3D ASCs maintained a more homogeneous morphology with no observable change in cell size or morphology.

Retention of ASC conditioned media wound healing capacity

Evaluation of the ASC-CM’s functional capacity to modulate a secondary cell population was utilized to demonstrate the dynamic interrelationship between ASC population phenotype and ability to promote wound healing activity in KCs (Figure 4D). KCs were treated with either 2D or 3D ASC-CM from each respective time point for up to 24 h. KCs treated with ASC-CM from 2D cultures showed a significant drop in recovered wound area when treated with P6 and P10 ASC-CM, relative to P2. There was also a slight decrease noted in KC migratory activity observed with 3D ASC-CM from P10 relative to P2 (Figure 4A & B). Overall, 3D ASC-CM maintained a significantly higher ability to enhance KC migration and close their respective wounds, relative to their 2D ASC-CM counterparts. The metabolic activity of KCs demonstrated a decreasing trend when treated with ASC-CM from 2D-expanded cells, with a significant decrease noted in P10 relative to P2 ASC-CM (Figure 4C). No significant differences were observed with ASC-CM from 3D cultures over time. Similarly, the proliferative activity of KCs treated with ASC-CM from 2D cultures demonstrated a decreasing trend, with a significant difference between P10 and P2 ASC-CM, but no significant change noted when KCs were treated with ASC-CM from 3D cultures (Figure 4D). Moreover, the proliferative activity of KCs treated with 3D ASC-CM from P6 and P10 was significantly higher than that of their 2D ASC-CM counterparts.

Discussion

The inherent regenerative properties of MSCs have garnered immense interest for advancing the field of regenerative medicine and tissue engineering. However, MSCs typically must first be removed from a donor tissue source and cultured outside the body within an artificial environment not native to human tissue. To date, commercially available in vitro expansion systems are almost exclusively 2D in nature. Rigid 2D systems are unphysiological for the cells and rapidly result in the loss of MSC multipotent stem-like features, with subsequent loss of viability and induction of senescence [16,18,36]. These changes lead to MSC populations with significantly reduced regenerative capabilities, which is compounded by a lack of standardized cell culture conditions, creating a significant bottleneck in the growth and development of regenerative therapeutics. Thus, there is a critical need to develop culture systems for MSC expansion that are 3D and more tissue-mimetic in their mechanical, architectural and substrate composition properties and which can ultimately circumvent many of the limitations of traditional 2D culture, such as the continuous need for subculturing.

Recent advancements in 3D systems have demonstrated progress toward producing MSC populations that are more stem-like. However, 3D systems such as spheroids, organoids, microspheres and many scaffold systems typically do not closely mimic the native mechanics of their cell/tissue source (e.g., adipose mechanics for ASCs) and often require large bioreactor systems and continuous subculturing to achieve large-scale cell numbers for clinical use. However, tissue-engineered hydrogel systems appear to be advantageous toward producing tailorable, tissue-mimetic systems for cell culture systems [30,31]. More specifically, the mechanotransductive response to the softer substrate of hydrogels is thought to aid in the retention of stem-like characteristics [23,37,38]. Unfortunately, most current hydrogel systems are manufactured to promote controlled differentiation of stem cells toward a specific tissue lineage and not to allow the cells to maintain a stem-like phenotype for long-term expansion [34,39–44]. Ultimately, an ideal MSC hydrogel expansion system would protect against senescence, while also improving the retention of a regenerative stem-like phenotype and permitting long-term expansion with minimal subculturing or user intervention.

Although MSC-based therapies have demonstrated promise, with over 1000 clinical trials to date listed with the US FDA, they have not advanced as quickly as previously thought. This is considered to be due, at least in part, to the detrimental impact senescent MSC populations may have on tissue regeneration [45,46]. Senescence is a progressive form of cell-cycle arrest, typically due to DNA and/or oxidative damage, which results in MSCs with impaired DNA-repair modalities that no longer proliferate and exhibit a loss of multipotency [47–49]. Moreover, senescent MSCs have been shown to secrete factors that negatively impact tissue regeneration and wound healing by impairing angiogenesis, increasing oxidative stress and exacerbating inflammation via the secretion of factors known as the senescence-associated secretory phenotype [50–52]. The composition of this phenotype can be heterogeneous and is dependent on the mechanism of senescence induction and environmental stimuli; therefore, this likely contributes to the heterogeneity in patient outcomes seen in clinical trials with both cell-based and acellular therapies [53–55]. Thus, developing an in vitro culture expansion system that limits/prevents the induction of senescence in healthy allogeneic or autologous MSC populations intended for patients would improve the efficacy and consistency of MSC-based clinical therapies.

In this proof-of-concept study, culture of ASCs within a traditional 2D culture system resulted in a significant increase in senescence, as previously established in literature. Conversely, the 3D hydrogel system resulted in no significant changes of senescence in ASC populations over the course of the 6-week study (i.e., 10 passage equivalents). Moreover, changes in cell morphology and size can be indicative of phenotypic changes; the apparent increase in cell size seen in the 2D ASCs may be associated with the induction of senescence and thus further supports the β-galactosidase imaging data and prior literature. Overall, these data support the previous hypothesis that 3D culture and substrate mechanics maintain a protective role against senescence. However, with our 3D hydrogel system the need for subculturing is eliminated, and secreted by-products are more readily accessible versus more traditional poured/molded hydrogels that lack a porous microarchitecture. Further studies will be needed to assess potential dynamic mechanisms at play (e.g., induction of telomerase, protection against oxidative damage or increased DNA repair mechanisms) to establish the protective role of 3D culture seen in this study. Moreover, evaluating what the inciting events are that drive the progression toward in vitro senescence, and whether they can be circumvented with appropriate culture systems, requires further investigation. This study offers insight into future design considerations for improving 3D cell culture modalities for MSC expansion.

Similarly, the stem-like phenotype of MSCs is critical to their regenerative potential and can rapidly change depending on the culture environment of the cells. MSC populations that differentiate and lose their stem-like characteristics result in variability of cellular phenotype and alterations in secretome composition, ultimately decreasing the consistency of regenerative MSC therapeutics, both cellular and acellular. Thus, MSC populations such as ASCs are often used within only a few passaging events in an attempt to circumvent the loss of regenerative potential. However, as we see in this study, even within one additional passaging event in 2D culture, ASCs significantly alter their expression of stem-like markers. Moreover, ASCs expanded in 3D culture for six or ten passaging equivalents (i.e., P6 or P10) over 6 weeks maintained similar expression levels of several markers relative to the baseline P2 ASCs, and a higher expression relative to their respective 2D counterparts, further highlighting the detrimental effects of 2D culture systems on MSC populations and the potential protective effects of 3D culture. The ability to improve the retention of stem-like properties within MSC populations for longer periods of time is desirable for a multitude of applications, including cell therapies, regenerative tissue engineering, immunotherapy and production of secreted biologics.

As previously mentioned, recent MSC therapies have expanded into investigating biologics as a potential regenerative therapy. MSC populations are highly adaptive in nature, and are known to sense their surrounding environmental stimuli and secrete factors accordingly. Thus, secreted bioactive compounds act to coordinate and bridge a variety of tissue reparative processes in an autocrine, paracrine or endocrine manner. In this study, the conditioned media from ASCs cultured in 2D versus the 3D system were collected over time and utilized as a therapeutic for a secondary cell population, KCs, as a means to functionally assess the phenotype of the ASC secretome toward promoting wound healing activity. Working under the hypothesis that the ASC phenotype deteriorates over time in 2D cultures but is sustained in the 3D system, we expected to see a gradual decline in regenerative capabilities of the ASC-CM from 2D cultures but minimal changes in ASC-CM from 3D cultures. This hypothesis was supported by the metabolic, proliferative and migratory data of KCs treated with ASC-CM. ASC-CM from 2D cultures demonstrated a steady decline in ability to augment KC activity and was consistently outperformed by its 3D counterpart over the course of the study. To our knowledge, this is the first direct comparison of ASC phenotypic changes over time in 2D and 3D cultures, and the first examination of how those changes potentially correlate to the regenerative wound healing capacity of the ASC secretome on KC functionality.

The limitations of this study include the formulation of the hydrogel system being intentionally bioinert in order to eliminate any contribution of a bioactive substrate, and thus it was not degradable. As a result, adequate removal of cells from this formulation was not feasible. Therefore, we utilized immunofluorescent labeling as an alternative to flow cytometry or RNA analysis to demonstrate senescence and phenotype of MSC populations. However, the ability to perform in situ visualization of an adherent population such as MSCs without the need to resuspend them is an advantage of immunolabeling over cytometry. Moreover, the relative comparison between 2D and 3D cultures, paired with the functional data of the ASC-CM, helps provide a supportive and holistic perspective that reinforces the immunolabeling data, although a more comprehensive analysis of the secretome is needed to assess both qualitative and quantitative changes. Future hydrogel formulations will aim to be biodegradable, such as insertion of enzyme-sensitive sequences or utilization of native matrix-derived compounds, while also incorporating the advantages and design of this proof-of-concept system, including the mechanics, modularity, architecture and elimination of subculturing. Lastly, follow-up studies to directly compare the performance of this 3D hydrogel system versus other 3D systems will provide a better understanding of the relative benefits of this new system.

Conclusion

In this study we introduce the proof-of-concept of a novel 3D hydrogel system to demonstrate the benefits of culturing MSCs in a system that more closely resembles their native tissue mechanics. Our 3D system contains a unique architectural design that does not impede effective mass and fluid transport while also allowing the movement of cells within and between attached hydrogels, in effect providing a continuous 3D culture system that eliminates the need to subculture cells. The continuity is achieved by the addition of supplemental hydrogels to previously seeded hydrogels, much like attaching together two puzzle pieces. The porous microarchitecture creates a ‘tunneling’ system for the cells to interact in the x-, y- and z- planes and to migrate within and between hydrogels, in addition to surface migration at attachment points. This is the first demonstration of such a system, to our knowledge. The results of this study provide a framework for future studies; specifically, after establishing the beneficial effects of this bioinert 3D hydrogel system on ASC populations, future studies will aim to further tailor the 3D system’s tissue-mimetic properties and control the regenerative activity of MSCs. Together, these data demonstrate the potential of a novel 3D culture system to enhance and retain the overall regenerative capacity of MSC populations over time, with the added benefit of further improving the regenerative and wound healing capabilities of MSC-derived biologics for a variety of tissue reparative applications.