Tissue-mimetic culture enhances mesenchymal stem cell secretome capacity to improve regenerative activity of keratinocytes and fibroblasts in vitro

Mesenchymal stem/stromal cells (MSCs) are a heterogenous population of multipotent and highly secretory cells currently being investigated in the field of wound healing for their ability to augment tissue responses. The adaptive response of MSC populations to the rigid substrate of current 2D culture systems has been considered to result in a deterioration of regenerative ‘ stem-like ’ properties. In this study, we characterise how the improved culture of adipose-derived mesenchymal stem cells (ASCs) within a tissue-mimetic 3D hydrogel system, that is mechanically similar to native adipose tissue, enhances their regenerative capabilities. Notably, the hydrogel system contains a porous microarchitecture that permits mass transport, enabling efficient collection of secreted cellular compounds. By utilising this 3D system, ASCs retained a significantly higher expression of ASC ‘ stem-like ’ markers while demonstrating a significant reduction in senescent populations, relative to 2D. Additionally, culture of ASCs within the 3D system resulted in enhanced secretory activity with significant increases in the secretion of proteinaceous factors, antioxidants and extracellular vesicles (EVs) within the conditioned media (CM) fraction. Lastly, treatment of wound healing cells, keratinocytes (KCs) and fibroblasts (FBs), with ASC-CM from the 2D and 3D systems resulted in augmented functional regenerative activity, with ASC-CM from the 3D system significantly increasing KC and FB metabolic, proliferative and migratory activity. This study demonstrates the potential beneficial role of MSC culture within a tissue-mimetic 3D hydrogel system that more closely mimics native tissue mechanics, and subsequently how the improved phenotype augments secretory activity and potential wound healing capabilities of the MSC secretome.


| INTRODUCTION
Mesenchymal stem/stromal cell (MSC) therapeutics have garnered immense interest in scientific research and discovery due to the intrinsic regenerative capabilities and wide-ranging applications of MSCs. 1 MSCs are a heterogenous population of progenitor cells easily isolated from a variety of different tissue locations, including bone-marrow, adipose, dental pulp and umbilical cord. 2,3 MSCs maintain multilineage differentiation potential and have shown the capacity to differentiate towards a number of different end-stage mature tissue types. [3][4][5] Thus, MSCs are at the forefront of tissue engineering and regenerative medical research in attempts to develop next generation bioengineered organs and tissue via combining multipotent MSC populations with bioengineered tissue scaffolds. [6][7][8][9][10][11] MSC-based therapies have also expanded to include cell-based therapies. [12][13][14] This is due to the inherent homing ability of MSCs to locations of tissue damage upon injection, in addition to the highly secretory nature of MSCs. [15][16][17][18] Numerous studies have investigated the use of MSC therapies in both animal and human trials. [19][20][21] Although the dynamics and efficacy of MSC-based therapy remains unresolved, the preliminary outcomes are potentially inspiring, so much so that over 1000 registered MSC-based clinical trials are currently listed with the FDA. [22][23][24] Interestingly, recent studies have demonstrated that the clinical benefit seen in many MSC-based therapies may potentially rely less on their multipotent nature and more on their adaptive secretory response after implantation. 18,25,26 The ability for MSCs to dynamically sense and adapt to specific environmental stimuli permits the subsequent modulation of local tissue environments in a state-dependent manner via secretion of biomodulatory factors. [27][28][29][30] Notably, MSCs have demonstrated the capacity to secrete cytokines, growth factors and extracellular vesicles (e.g., exosomes and microvesicles) that are immunomodulatory, trophic, mitogenic, pro-angiogenic and can augment matrix deposition to promote neotissue formation. [30][31][32][33][34] Therefore, MSCs contain a variety of intrinsic regenerative capabilities due to the compositional plasticity of their bioactive secretory product. Furthermore, studies have shown that wound healing and other tissue reparative processes can be enhanced after treatment solely with acellular MSC secretory products. For example, conditioned media (CM) from MSCs in 2D culture have demonstrated the capacity to directly enhance the migration, proliferation, and matrix production of fibroblasts and keratinocytes, which are key cell populations involved in the regulation of wound healing and tissue regeneration in a variety of settings. [35][36][37][38][39] However, the role of a tissuemimetic hydrogel culture system in augmenting the MSC secretory response is still under investigation. These observations have prompted a new age of MSC and MSC-derived acellular biologic therapies and adjuvants as a means to modulate the regenerative bioactivity of already utilised wound therapy modalities, in order to further promote tissue genesis and restores tissue function. 40 Wound healing is a broad classification for the highly complex series of synchronised signalling events that occur in the setting of tissue damage. 41,42 Two important cell populations involved with wound healing are Keratinocytes (KCs) and Fibroblasts (FBs). 43 KCs orchestrate the process of re-epithelialization, which is a fundamental step of proper wound healing that results in 'closure' of the wound. 44,45 Lack of wound closure can result in polymicrobial infections, desiccation and reinjury. Similarly, FBs are intimately involved in a number of processes during wound healing. 46 FBs must migrate into the wound tissue, aid in the deposition of temporary granulation tissue and subsequently remodel the tissue matrix. 46 The bioactivity of KCs and FBs is highly dependent on cellular crosstalk and communication within the local tissue environment. Thus, paracrine and autocrine signalling activity are critical to the success of native wound healing. 45,47 In the body, our tissue microenvironments provide a range of mechanotransductive cues that regulate cellular activity within that environmental niche, including tissue mechanics. 48 Moreover, studies have demonstrated that substrate mechanical properties play a key role in cellular differentiation, function, viability and overall phenotypic properties within in vitro culture systems. 49,50 Notably, tissues in our body typically range from $0.1 to 100 kPa in tissue stiffness, with the brain measuring in around $0.3 kPa, fat $3 kPa, muscle $10 kPa and precalcified bone $100 kPa. 51 Conversely, traditional 2D tissue culture plastic is a much stiffer substrate, often reaching around $1000 kPa or greater. 51 Evidence suggests that traditional 2D culture is not ideal for expansion of many cells, particularly 'stem-like' cells such as MSCs, and can result in a loss of MSC multipotency, induction of senescence and decreased bioactivity. [52][53][54][55][56] This is due to a combination of an unnaturally stiff 2D culture substrate, over-crowding of cells and the stress of continuous subculturing (i.e., passaging) to achieve adequate cell numbers for experiments and/or clinical therapies. Cellular senescence is a progressive, phenotypically diverse and multi-staged process of cellular aging and DNA damage, typically considered to result in irreversible cell cycle arrest. 57,58 The secretome of senescent cells, including MSCs, have been shown to negatively impact tissue regeneration and wound healing by inhibiting angiogenesis, exacerbating inflammation, increasing oxidative stress and inducing senescence in secondary cell populations. 58 [62][63][64] Due to this, a number of attempts have been made to generate 3D systems that are tissue-mimetic in nature, such as the fabrication of 3D hydrogels that recapitulate the native in vivo tissue mechanics cells would be exposed to.
In this study, we utilise a 'bioinert' 3D hydrogel system that mechanically mimics adipose tissue in order to culture Adiposederived Mesenchymal Stem Cells (ASCs). Subsequently, the relationship between the ASC population and subsequent modulation of the ASC secretory byproducts were evaluated to determine the ability of the secretome to augment specific wound healing functions. More specifically, the loss of 'stemness' and induction senescence of ASCs cultured in a traditional 2D system will be compared to ASCs within the 'bioinert' 3D hydrogel system and the secretory byproducts of each will be used to assess for modulation of migration and proliferation of keratinocytes and fibroblasts in vitro. Utilisation of a 'bioinert' 3D substrate provided the opportunity to directly observe the role of the mechanical, dimensional and architectural properties of a 3D system on ASC senescence and 'stem-like' phenotypic properties without artificially supplementing with a biological additive or 'bioactive' substrate. Additionally, the 3D hydrogel system contains a porous microarchitectural design that permits mass transport and easy collection of acellular byproducts secreted from cells, essentially serving as a 'bioreactor'. Therefore, we hypothesized that the mechanical properties of an adipose-like tissue-mimetic system would result in a more robust ASC population and subsequently alter ASC secretory activity to enhance wound healing functions in keratinocytes and fibroblasts. 2.2 | Three-dimensional (3D) printed hydrogel cell culture system The 'bioinert' 3D hydrogel system is $1-cm 3 and is a PEG-based 3D printed cell culture and expansion system called a Tissue-Block (T-Block; Ronawk; Kansas, USA) that contained a unique microarchitectural design with a continuous porous channelling system that had a pore diameter of 300-μm. The hydrogel is printed utilising a pre-defined and specific microstructure that results in the creation of voided/porous regions that create continuous microchannels and results in a porous fraction that account for $42% of the total volume. The unique microarchitectural design promotes cellular migration and proliferation, while also permitting mass transport and nutrient exchange, via the microchannels. The 3D hydrogels were placed into a glass 6-well culture plate and utilised for cell culture to culture ASCs in a more tissue-mimetic environment and as a 'bioreactor' to generate/collect secreted byproducts from the ASCs for analysis. Fibronectin is a commonly selected coating substrate for ASCs due to their natural secretion of fibronectin.
Since the cells do not naturally adhere/attach to the PEG-based (polyethylene glycol) hydrogel, both the 2D culture plastic/glass and 3D hydrogel system were coated with fibronectin 24-h before cell seeding at a standardised concentration of 5-μg/cm 2 to enhance the initial cell attachment for experimental assays carried out in this study. The concentration of fibronectin was standardised 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. After the T-Blocks were coated, ASCs were added dropwise and allowed to migrate and distribute throughout the porous microchannels. It is important to note that the cells are not embedded/encapsulated within the hydrogel itself but rather form 3D networks within the porous microchannels. Additionally, oxygen distribution within hydrogel was carried out by O2M Technologies (Chicago, IL).

| Expansion of ASCs, KCs and FBs
ASCs, KCs and FBs were seeded on 2D plastic and cultured until $80% confluency before subculturing (i.e., passaging). Subculturing of cells was performed by removing culture media, washing 3Â with Hank's balanced salt solution (HBSS; calcium-free, magnesium-free) and incubating with 0.05% Trypsin/EDTA (Lonza; Cat. #CC-3232) at 37 C for 5-min. Trypsin was neutralised with serum and cells were centrifuged at 500g for 5-min, pelleted and resuspended for reseeding and use in experimental assays or continued expansion. Subculturing events occurred every 4-5 days for ASCs, 5-7 days for KCs and 4-5 for FBs. After an initial characterisation of 'Passage 1 (P1)' ASCs, subcultured ASCs at 'P2' through 'P5' were utilised for the assays in this study. The increased surface area of a single 3D hydrogel system eliminated the need for subculturing for the time course of this study, therefore a passage-equivalence timepoint was utilised to allow for analogous comparison with 2D culture. For example, in this study, ASCs were seeded in 2D and 3D at 'P2', a passaging event occurred every 4-5 days in 2D for a total of three passages. After three passaging events in 2D ASCs became 'P5', thus the passage-equivalent in 3D was 'P5' after 14 days in 3D culture, even though the cells never underwent another passage after the initial 'P2' seeding. Culture expansion for 14 days 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) for the ASCs in order to standardise cell numbers. ASCs were seeded at a standardised concentration of $5000-cells/cm 2 for assays. KCs and FBs were seeded at a concentration of $7500-cells/cm 2 for 2D and only grown in 2D for the experiments in this study. (14 days), cells were assessed for 'stem-like' phenotype, as mentioned above. ASCs seeded in 3D system at 'P2' were assessed simultaneously at the 'P5' passage-equivalent timepoint for 3D.

| Analysis of ASC proliferation and migration within porous hydrogel system
The ability for ASCs to adhere to, and propagate within, the hydrogel system was evaluated via confocal microscopy after prolonged culture based on similarly discussed protocols. In short, ASCs were seeded onto/within the hydrogel system at a standard seeding concentration.
MSC-GM media was changed every two (2) days for two (2) weeks (14 days  GAPDH was used as an endogenous control for these samples as well. Reverse transcription and gene expression analysis was performed as described above and all RNA analyses were performed with biological triplicates (n = 3).

| Isolation of ASC conditioned media
Media supplementation was standardised to $250-μL/cm 2 for ASC expansion to account for dilutional differences in surface area-tovolume ratio between 2D and 3D culture. Media was changed every 2 days. When Conditioned Medium (CM) from ASC culture was desired, MSC-GM media was removed, cells were washed with HBSS and serum-free MSC media was added for 48-h before collection for both 2D and 3D. Collected ASC-CM was then centrifuged at 1500g for 10-min to eliminate cell debris, Steriflip filtered with a 0.22-μm filter and stored at À80 C for long-term storage until use.  Table S2. Without an internal reference control on the proteome array blots, 2D and 3D samples were performed and developed pairwise to standardise exposure time for each pair. For example, analysis of one (1) 2D blot and one (1) 3D blot array was performed in tandem and placed into chemiluminescent detector machine (ProteinSimple, Inc., San Jose, CA, USA;

| Proteomic microarray for secreted soluble proteins
FluorChem E) at same time and exposed simultaneously. The relative fold change of 3D-to-2D signal was calculated for each pair of 2D and 3D samples using ImageJ software and performed in biological triplicate (n = 3). The average fold change was then calculated.

| ASC-CM extracellular vesicle (EV) production
ASC-CM was collected as previously discussed and total EV production was assessed via relative protein content of the EV fraction. EVs were isolated via centrifugation at 3500g for 15-min through a 100-kDa filter tube (Cytiva; Cat. #28932363) followed by washing with PBS and re-centrifugation at 3500g for 5-min through the 100-kDa filter, for a total of 3 washes. The concentrate was then taken, and EV were precipitated overnight using a ExoQuick-TC kit

| EV in vitro tracking
EVs previously isolated from the ASC-CM were fluorescently labelled with the lipophilic membrane stain Dil (Invitrogen; Cat. #D282) at a concentration of 1-μM. The labelled EVs were washed with PBS and re-centrifuged at 3500g for 5-min through the 100-kDa filter, for a total of 3 washes and centrifugations to remove excess dye. DiIlabelled EVs were added to KC-GM or FB-GM and co-cultured overnight at 37 C with KCs and FBs, respectively. The following day, the cells were washed 2Â with PBS and then fixed with 4% paraformaldehyde, followed by 2Â additional PBS washes. KCs and FBs were counterstained with Hoechst 33342 (Invitrogen; Cat. #H3570; 1:1000) and Alexa Fluor 488 Phalloidin (Invitrogen; Cat. #A12379; 1:500) and observed under a fluorescence microscope.

| Statistical analysis
All data are reported as means with standard deviation. Characterisation analyses of ASC population with immunolabelling for senescence were evaluated with a two-way ANOVA to include control population. Immunolabelling for CD markers was also evaluated with a two-way ANOVA. Proteome array and KC scratch assay were also evaluated with a two-way ANOVA methodology. Analysis of antioxidant activity, EV protein content, and plate reader spectroscopy data for KC metabolic and proliferative activity were evaluated with a oneway ANOVA. Data was tested for normality via Shapiro-Wilk and Kolmogorov-Smirnov tests and plotted with a QQ plot. GraphPad Prism 9.0.2 software (La Jolla, CA) was used for the analyses and a p < 0.05 was considered significant.

| RESULTS
3.1 | Culture of ASCs within adipose-like porous hydrogel system The movement of fluid through the hydrogel porous microarchitecture was captured photographically and can be seen quickly (1-2 s) permeating from the surface of the hydrogel to the base, where the liquid was absorbed via a Kimwipe ( Figure 1A). Similarly, the ability of nutrients and gases to distribute throughout the 3D system was indirectly assessed via a conjugated oxygen isotope and magnetic resonance, which demonstrated relatively homogenous distribution throughout the hydrogel ( Figure S2). Additionally, ASCs were cultured within the hydrogel system for 14 days and allowed to migrate throughout the hydrogel within the porous microchannels.
ASCs were then stained and assessed with confocal microscopy for their ability to distribute, via migration and proliferation, throughout the hydrogel's pore system ( Figure 1B). ASCs were seen lining the porous channel beyond the superficial surface, deep into the hydrogel system and can be seen forming 3D cellular networks within the hydrogel's pore structures. Moreover, there is no indication of cell death or necrosis within the more centralised region of the hydrogel.
The mechanical properties of the hydrogel were characterised to assess for similarities with native adipose tissue. Notably, based on literature, the modulus (E) of adipose tissue is about 1-10 kPa when evaluated via a stress-strain curve immediately preceding plastic deformation, 65,66 with an average considered to be $3 kPa. 51 DMA was performed to determine the elastic compressive secant modulus (E), max strain and max stress of the hydrogel system. The modulus (E) was determined to be $3.7 kPa immediately prior to plastic deformation ( Figure S3), similar to what has been reported for native adipose tissue.

| 3D hydrogel expansion decreases ASC senescence
ASC senescence was assessed via a marker targeting β-galactosidase activity (Figure 2A,B), a commonly utilised surrogate measurement of cellular senescence and expression of p16 and p53 ( Figure 2C). Image quantification demonstrated a significant increase in senescent-positive cells at 'P5' within the 2D culture system ($11.5%) relative to 'P5' cells within the 3D culture system ($3.6%) and the initial 'P2' ASC population ($5.5%) ( Figure 2B). Therefore, there was an $3.2Â fold increase F I G U R E 2 Tissue-mimetic hydrogel culture decreases ASC senescence. ASCs seeded at 'P2' within the 3D hydrogel system or continuously subcultured for 2 weeks in traditional 2D culture until reaching 'P5'. The 'P5' ASCs were used for characterisation in 2D and 'P5' passage-equivalent were used for 3D. (A) At the conclusion of culture period, ASCs were fixed and stained for senescence/β-galactosidase (Green), Hoechst 33342 (Blue) and Phalloidin-AF647 (Not Shown). (B) ASCs cultured in 2D (Black Bar) or 3D (Teal Bar) were evaluated using a 20Â objective. ASCs seeded in 2D at 'P2' were used as an initial control population and denoted as the dashed line (Black). All image quantification data is displayed as a bar graph and is the result of averaging each group of technical replicates (different images within each biological replicate) to quantify senescence. Samples done is quadruplicate (n = 4). Scale bar = 100 μm. (C) Relative fold change in gene expression for 'P5' ASCs in both 2D and 3D was assessed for changes in senescence-associated markers, p16 (left) and p53 (right), relative to 'P2' baseline control cells. Samples done in triplicate (n = 3). All error bars are standard deviation. One-way ANOVA with Tukey's post-hoc was used for statistical analysis. Significance is denoted as *p < 0.05 or ****p < 0.0001 for 2D versus 3D comparison and # p < 0.05 or #### p < 0.0001 for comparison relative to 'P2' control.
in senescent ASCs within 2D culture relative to 3D culture. Gene expression for senescent ASC populations was assessed via p16 and p53 ( Figure 2C). The relative fold change in p16 expression for 'P5' ASCs in 2D ($2.7Â) was significantly increased relative to 3D ($1Â) and to the 'P2' ASC baseline population. Whereas expression of p53 in 2D ($2.3Â) demonstrated an increasing trend but was not significantly different relative to ASCs in 3D ($1.3Â) or 'P2' controls. ASCs demonstrated no significant changes in senescence activity for cells cultured within the 3D system relative to 'P2' baseline ASCs.

| 3D hydrogel expansion improves retainment of ASC phenotype
Evaluation of 'stem-like' ASC populations was assessed via immunolabel quantification of cell surface CD markers ( Figure 3A,B) and expression of key phenotypic markers ( Figure 3C and Figure S4). Both 2D and 3D resulted in a decrease in 'stem-like' positive markers (CD73/90/105) over time, relative to the initial population ( Figure 3B and Figure S1). ASCs exhibited 16%, 30% and 27% in 2D and 38%, 52% and 45% positive expression for CD73, CD90 and CD105, respectively. Thus, the relative decline in ASC surface markers was greater in 2D ( Figure 3B). Both 2D and 3D ASCs maintained a relatively low population (<5%) of ASCs staining positive for both CD34 and

| 3D hydrogel enhances ASC EV production
ASC production of EVs was quantified within the ASC-CM from 2D and 3D culture to determine relative differences in EV composition.
All three protein quantification modalities (BCA, Bradford, QuickDrop) demonstrated similar trends with a significant increase in EV protein content within the ASC-CM from 3D culture ( Figure 5A). The purified EV fraction was also assessed via NTA to further determine particle concentration and size distribution between 2D and 3D EVs. NTA determined a similar trend in EV production for 3D, with approximately 9.2 Â 10 9 particles/mL in 2D and 6.5 Â 10 10 particles/mL in 3D ( Figure 5B). The EV particle size distribution between 2D and 3D both had similar trends and followed a Gaussian distribution, with $98% of all particles measured falling within the exosomal range of 25-250 nm in diameter ( Figure 5B), suggesting that the EV population is likely mostly exosomal in nature. Subsequently, an in vitro tracking and uptake study was performed with the EV fraction, where the EVs were stained with a lipophilic membrane dye (DiI) and KCs and FBs were treated with the DiI-labelled EVs ( Figure 5C). The EVs can be F I G U R E 4 Altered secretory activity of ASCs within tissue-mimetic hydrogel. ASCs seeded at 'P2' within the 3D hydrogel system or subcultured one additional time in traditional 2D culture until reaching 'P3'. The 'P3' ASCs were used for characterisation in 2D and 'P3' passage-equivalent were used for 3D. ASC-CM was collected from the 'P3' and 'P3' passage-equivalent ASC cultures, for 2D and 3D, respectively. Relative chemiluminescence was determined for each proteome array membrane. Average fold change was calculated and displayed as 3D:2D ratio. Assay was performed in triplicate (n = 3) and averaged. Significant differences in 3D relative to 2D are denoted (Teal Bars). Proteins indicating no significant difference are denoted with black bars. Error bars are standard deviation. One-way ANOVA with Tukey's posthoc used for statistical analysis. Significance is denoted as *p < 0.05, **p < 0.001, ***p < 0.001 and ****p < 0.0001 for 2D versus 3D comparison. seen being taken up and endocytosed by both the KCs and the FBs within 18 h ( Figure 5C). Interestingly, KC morphology and cytoskeletal rearrangement was evident (denoted by changes in phalloidin staining), especially in KCs taking up a larger proportion of EVs ( Figure S6). F I G U R E 5 Enhanced production of EVs within tissue-mimetic hydrogel. (A) ASCs seeded at 'P2' within the 3D hydrogel system or continuously subcultured for 2 weeks in traditional 2D culture until reaching 'P5'. The 'P5' ASCs were used for characterisation in 2D and 'P5' passage-equivalent were used for 3D. ASC-CM was collected from the 'P5' and 'P5' passage-equivalent ASC cultures. The EV fraction of ASC-CM was isolated, purified and analysed for relative protein content via three different modalities, BCA (leftmost), Bradford (middle) and QuickDrop (rightmost). Concentration of EV protein fraction displayed as average 'μg/mL' within the initial cell culture volume before concentrating with 100-kDa filter. Concentration within control media was analysed and is displayed as dashed line (Black). (B) Isolated EV fractions were then analysed with NTA for determining concentration of particles/mL within media (leftmost) and to assess size distribution of the measured particles and the cumulative frequency of the different EV particle sizes (rightmost) to determine whether measure particles are truly within EV size range. Assays were performed in triplicate (n = 3) and averaged. Error bars are standard deviation. One-way ANOVA with Tukey's post-hoc used for statistical analysis. Significance denoted as ****p < 0.0001 for 2D versus 3D comparison and #### p < 0.0001 for 3D comparison relative to media control. (C) Representative images of fluorescent-labelled Keratinocytes (Top) and Fibroblasts (Bottom) that were treated with media supplemented with labelled-EVs from 3D ASC-CM for 18 h. Samples imaged with a 40Â objective. Scale bar = 30 μm.

| Secretome from 3D culture enhances KC and fibroblast activity
To determine the significance of utilising secretome from ASC populations in 3D rather than 2D, the functional capacity of ASC-CM was assessed for its ability to modulate the activity of secondary cell  Figure 6D,H). . Metabolic and Proliferative activity was performed with five replicates (n = 5). Migratory activity was assessed via scratch assay recovery. The voided space created by a pipette tip was evaluated for recovery of area via migration of KCs and FBs. Whole well images were acquired and the recovery area of three different locations per well were averaged. Migration samples were performed in triplicate (n = 3) for a total of nine images per treatment group. Average area closed/recovered are denoted for each time point for KCs and FBs after treatment with ASC-CM from 2D (Black Circles) or 3D (Teal Diamonds). Significance is denoted as *p < 0.05, **p < 0.01 and ****p < 0.0001, and # p < 0.05 or ## p < 0.01 for 3D comparison relative to media control. Error bars are standard deviation. One-way ANOVA was used for Metabolic and Proliferative assays. Twoway ANOVA was used for Migratory assay. Scale bar = 200 μm.

| DISCUSSION
The multipotent and highly secretive nature of MSCs provide a unique opportunity for advancing the field of regenerative medicine and wound care. However, limitations of current in vitro 2D expansion methodologies currently hinder the growth of MSC-based research and development. In the unnatural and harsh environment of traditional 2D culture systems, MSCs lose their multipotent 'stem-like' features, become non-viable and senescent and subsequently decrease their production of regenerative secretory products. [52][53][54] Thus, a shift towards 3D culture systems that more closely mimic native in vivo tissue environments are being investigated.
Recently, investigations into microcarriers, spheroids/organoids and 3D scaffolds or hydrogel culture systems have been investigated.
There have been a number of studies demonstrating the benefits of 3D culture of MSC-like populations relative to traditional 2D culture and how the culture system effects MSC phenotype. Moreover, the importance of mechanical regulation of MSC phenotype is known. However, '3D culture' and 'tissue-mimetic' culture are broadly used terms. Microcarrier systems do allow for enhanced expansion of stem cells, but they are more pseudo-3D and are not tissue-mimetic in nature. The production of native matrix components with cells in 3D spheroidal and organoid culture has demonstrated enhanced abilities for generating MSCs that are more viable and maintain 'stem-like' properties, though they do not produce a native tissue-mimetic environment and typically require a largescale bioreactor system for continuous culturing. 67,68 Moreover, mass transport and diffusional constraints limit the size and utility of 3D spheroids/organoids for many applications. 67,68 Recently, tissue engineered 3D hydrogel systems have provided advantages towards producing a tailorable tissue-mimetic system that improve the viability of 'stem-like' cells and allow for greater control over differentiation potential of MSCs in culture. 69,70 However, many 3D hydrogel systems can be limited in size due to diffusional constraints, such as that seen with gels that are poured/moulded/ casted which lack a porous architecture, or are derived of a bioactive substrate that ultimately induces differentiation of MSC populations, which results in lack of clarity of the role of the mechanics.
To date, there is yet to be any studies investigating how the culture of ASCs within a bioinert but mechanically analogous system effects ASC populations and their secretome in the context of wound healing. Notably, previous studies have yet to utilise a 3D hydrogel system to assess the relationship between senescence and 'stem-like' properties of ASC populations to changes in ASC secretive bioactivity.
In this study, we utilise a 'bioinert' 3D hydrogel system to improve in vitro culture conditions of ASCs. The porous microarchitecture of this hydrogel system is unique and allows for cellular migration and proliferation within the hydrogel. Notably, cells are seen within the porous channels attaching and creating an interconnected network ( Figure 1B). Moreover, the ability to effectively permit the transport of fluids and small molecules within a 3D system is critical to effectively allow nutrient and waste exchange. [70][71][72] Thus, the porous nature of this system permits mass transport ( Figure 1A) and the easy collection of cellular-derived biologics, such as EVs and proteins, within the secretome (Figures 4 and 5). Therefore, this system was selected because it provided an improved culture environment for ASCs over traditional 2D culture. 73 Additionally, this 3D system enabled more efficient isolation of cellular byproducts (i.e., biologics) relative to non-porous hydrogel systems, which ultimately lack the capacity for long-term culture and biologics collection that would be necessary for future clinical therapies. For the context of this study, a hydrogel that was mechanically similar to native adipose tissue was generated ($3 kPa; Figure S3). 51,66,[73][74][75] Additionally, the 'bioinert' In this study, culture of ASCs within a traditional 2D system resulted in a significant increase in senescent cells relative to the initial 'P2' ASC populations, whereas 3D culture did not (Figure 2A-C).
It is important to note the timeframe of this study of 14 days (or 3 passage events in 2D), as this is a common timeframe found in studies in literature, and is often the minimum required expansion time in vitro to achieve adequate MSC numbers for clinical therapies. The observed differences in senescence for 3D ASCs will need to be further investigated to determine the exact mechanism and dynamics of the ASC senescent populations. Ultimately, this data shows that the prevalence of unfavourable senescent ASC populations can be significantly reduced by culturing ASCs in a tissue-mimetic 3D system, without the need of added supplements or other biomodulatory factors.
To our knowledge, this is the first demonstration of a 'bioinert' 3D Denoting a possible protective role of this tissue-mimetic 3D system from MSC phenotypic changes, in part, potentially due to the softer mechanical properties and 3D architecture. Future experiments will aim to investigate the difference between this 'bioinert' system and an analogous system with additives to generate a more tissue-mimetic system and further augment ASC regenerative capabilities.
Overall, this study reiterates the detrimental effects of traditional 2D culture systems on the 'stem-like' properties of ASCs over the course of only three passaging events (14 days). As previously mentioned, many studies in literature use MSC populations typically in the passage 3-6 range for experiments, meaning there is likely variability in the phenotype of these populations, both between studies as well as within the same study. This study provides insight into future design consideration for an MSC culture expansion system and the potential importance of tissue mechanics ad MSC phenotype. Ultimately, biomedical research with MSCs often aims to exploit their 'stem-like' properties and characteristics, where loss of 'stemness' has been shown to decrease the overall regenerative properties of MSCs. 76,77 Thus, the ability to improve the retainment of MSCs with higher 'stem-like' populations is desirable for a multitude of applications, including cell therapies, regenerative tissue engineering and production of secreted biologics.
To evaluate the significance of culturing ASCs in a tissue-mimetic 3D system rather than traditional 2D on the secretome composition, ASC-CM was collected, analysed and utilised in conditioned media studies with KCs and FBs. Although characterisation of specific ASC secretome components have previously been assessed in different 'primed' 2D and 3D systems (e.g., increased secretion of VEGF in fibrin-based gels or increased HGF secretion in 3D spheroids), [78][79][80] direct comparison of secretome differences between 2D culture and a 'bioinert' 3D hydrogel system have yet to be investigated (i.e., direct investigation into role of 3D mechanics and architecture).
First, ASC-CM was evaluated for the relative composition of key secreted proteins that are important for wound healing and regeneration. Our data demonstrated that $15% of the proteins tested were significantly increased in 3D (Figure 4), whereas the remaining proteins demonstrated no significant changes. This suggests that the relative composition of soluble proteinaceous factors secreted from ASCs is altered within the 3D hydrogel system. Additionally, this data may suggest a possible increase in total secreted protein production, though more studies are needed to assess global protein production.
Notably, solubilised proteins have been shown to directly augment the bioactivity of cellular populations, including the wound healing activity of fibroblasts and keratinocytes. [81][82][83] Therefore, this 3D system demonstrates a potential means to produce solubilised proteins for regenerative therapies. Total protein concentration was attempted but due to phenol red within the media, QuickDrop spectrophotometry and BCA analysis were not possible. Future studies will look to optimise utilisation of phenol-free media and assess total proteome.
Similarly, the ASC-CM was further evaluated for antioxidant activity. Increased oxidative activity is one of the many compounding factors that can hinder the tissue regenerative response and promote the progression towards non-healing wounds and tissue. 84,85 Therefore, antioxidants are critical compounds in the wound healing and tissue regeneration processes, with antioxidant treatments previously demonstrating the capacity to improve the wound healing process. [84][85][86] Stem cells are known to secrete antioxidants, 87 though upon becoming senescent stem cells have demonstrated a shift in the secretion of oxidants relative to antioxidant compounds. In this study, ASC-CM from the 3D system exhibited a significantly higher level of total antioxidant activity relative to 2D ( Figure S5). As we have previously stated, senescent stem cells have previously been associated with changes in oxidative activity and have been shown to induce upregulation of oxidative damage in surrounding cell populations, thus there is potentially a relationship between the relative senescence and the relative antioxidative/oxidative balance in 2D and 3D culture systems. [58][59][60] Further studies investigating the dynamic relationship between senescence, stemness and antioxidant/oxidant activity are warranted to elucidate any potential connection.
Another form of cargo secreted by ASCs other than antioxidants and solubilised proteins are EVs. In general, EVs consist of a myriad of vesicle-like particles such as apoptotic bodies (500-5000 nm), microvesicles (MVs; 200-1000 nm) and exosomes (Exo; 25-250 nm). 88,89 EV have garnered a lot of interest of late, due to the diverse array of biomodulatory cargo that they can carry. Depending on the EV type and origin, EVs can proteins, small molecules and nucleic acids for MVs or exosomes; whereas apoptotic bodies tend to secrete chromatin and organelles. [88][89][90] Moreover, the relative composition of the contents within EVs have been shown to change in a state-dependent manner. 91 Thus, the highly adaptive nature of stem cells are known to secrete a variety of compounds within EVs; specifically, the exosomal fraction of ASCs have demonstrated the ability to enhance the rate of wound healing and tissue regeneration. 31,92 In this study, similar to the total protein fraction, the relative concentration of EVs was significantly increased in the ASC-CM from our 3D system ( Figure 5A). Suggesting that ASCs potentially favour secretion of EVs in 3D (and in vivo) and/or are more secretive globally overall in 3D, relative to 2D. Moreover, the size distribution data suggests that the EVs evaluated in this study tended to be smaller and more likely exosomal or MV in nature. The subsequent in vitro tracking assays validated the ability for KCs and FBs to endocytose ASC-derived EVs. Upon internalisation into the cytoplasm of cells, EV/Exosomes have been shown to modulate cellular signalling. 90,92 Moreover, as observed in this study, a notable change in cellular morphology is demonstrated to a greater extent is cellular populations that endocytosed more EVs ( Figure S6), denoting a potential key role of EVs in wound healing physiology that should be further explored. Therefore, one possible mechanism of action of ASC secretome modulation of wound healing activity is via the controlled secretion of EV/Exosomes. 92,93 Further studies investigating the composition of EV/Exosomes contents between 2D and 3D would provide a better understanding not only in total secretive activity, but the nature of that secretive activity and composition of biomodulatory cargo. To our knowledge, this is the first holistic demonstration of how the secretory production of soluble proteins, antioxidants and EV/exosomes from ASCs can be further enhanced within a tissue-mimetic hydrogel system, potentially due to the improved ASC phenotype within the system. Lastly, to assess the significance of the altered ASC secretome composition on wound healing capacity, ASC-CM co-culture assays were performed to assess for 'functional changes' in the ASC-CM ( Figure 6). For both KCs and FBs, proliferative activity was significantly increased when treated with ASC-CM from 3D culture, but only KCs exhibited an increase in metabolic activity after treatment with ASC-CM. Similar to previous literature, ASC-CM from traditional 2D culture did expedite the rate recovered scratch area of KCs and FBs, relative to control. However, 3D ASC-CM further enhanced the rate of recovery for KCs and FBs, relative to traditional 2D ASC-CM.
Therefore, the augmentation of ASC secretory composition within the  49,94,95 Future studies will aim to investigate the factors driving the benefits demonstrated by 3D culture in this study and aim to determine whether 3D culture modulates the quality and/or quantity of ASC secretory compounds.
Limitations of this study include the utilisation of one ASC donor source. Ideally, future studies will include multiple donor sources to make sure differences seen are not donor-specific. However, since the study directly compared 2D versus 3D culture and not the overall ASC-CM efficacy, the direct comparison between the two systems is reasonable because the relative donor-specific attributes would affect both the 2D and 3D systems equally. Of note, since the goal of this study is to demonstrate the possible utility of an acellular biologic therapy derived of allogeneic ASCs for a wound therapy, a younger and healthier source of ASCs, and an older source of keratinocytes and fibroblasts was desired. Another limitation is the lack of breadth of the proteome array. Although this proteome array provides more information than individual ELISAs, this array only tested for 55 secreted proteins. Therefore, it is not an accurate depiction of the entire proteome. However, in this study we utilise a number of supplementary assays of ASC-CM characterisation to corroborate the differences seen in the proteome array. Future studies will investigate the key proteins in ASC-CM that resulted in augmented KC and FB activity. One possible method will be to perform mass spectroscopy analysis to provide a holistic perspective of the ASC-CM proteome in 2D and 3D. Future studies will also investigate the composition of the EV/Exosomes and provide a clearer perspective of the dynamic changes occurring in ASC secretive activity. Ultimately, 3D in vitro and in vivo wound models will be the subject of future studies after demonstrating the validity of this tissue-mimetic system. However, performing 2D models did allow the authors to directly assess the relative effect of the ASC secretome on individual functions for each cell line separately. Lastly, performing a study to directly compare this system with other 3D culture systems widely used in biomedical research will provide a better understanding of the relative benefits of this system.
As previously discussed, bioactive compounds coordinate and bridge a variety of tissue reparative processes and hold immense regenerative capabilities. Studies have shown that ASCs are capable of modulating secondary cell populations via the secretion of bioactive compounds such as growth factors, cytokines, exosomes, and microvesicles in an autocrine, paracrine, and endocrine manner. Thus, the dynamic and diverse milieu of biomodulatory compounds secreted from ASCs offers a unique approach for tailored clinical therapeutics. 47 However, there is still much about the secretive bioactivity of ASC populations that is unknown. The role of culture expansion, viability, senescence, 'stem-like' phenotype, and other stochastic processes on the overall composition and functionality of the ASC secretome remains largely unexplored. With the growing interest for regenerative acellular biologics due to limitations of autologous and allogeneic cell-based therapies, this study aimed to introduce how a tissue-mimetic 3D hydrogel system improves the culture expansion of ASCs and the subsequent production of acellular biologics that can augment wound healing activity of secondary cell populations.

| CONCLUSION
In conclusion, this study was a 'proof-of-concept' study meant to serve as a template for future study designs and illustrate the benefits of culturing MSCs within a porous tissue-mimetic 3D system that mechanically relates to their native tissue (rather than 3D spheroids or microcarriers). The results of this study shed further insight into the heterogeneity of MSC populations and the effect of inadequate culture conditions on the regenerative capacity of the cells. Moreover, this study will lay the foundation for future studies looking to tailor the properties of this 3D system to generate a more tissue-mimetic system via incorporating biomodulatory compounds to enhance the control over phenotype and secretive bioactivity. For example, after establishing the beneficial effects of this 'bioinert' 3D hydrogel system on ASC populations, extracellular matrix (ECM) substrates (e.g., fibrin, collagen, gelatin, elastin), biologics (e.g., growth factors or cytokines) and environmental stimuli (e.g., hypoxia) can be introduced as stimuli into the 3D system to further modulate ASC activity. Thus, this system provides means to generate customizable ASC-derived biologics for future regenerative therapeutic modalities for a variety of applications, including integration into current wound healing modalities.