Molecular, hierarchical and acoustical cues promote mesenchymal stromal cell differentiation toward tympanic membrane regeneration

doi:10.21203/rs.3.rs-4214901/v1

Download PDF

Article

Molecular, hierarchical and acoustical cues promote mesenchymal stromal cell differentiation toward tympanic membrane regeneration

https://doi.org/10.21203/rs.3.rs-4214901/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Stem cell therapies using mesenchymal stromal cells (MSCs) have recently emerged as a promising approach for the treatment of tympanic membrane (TM) injuries. However, the role of essential biochemical, biophysical, and biomechanical signals in guiding the MSC differentiation for TM applications remains unexplored. Therefore, this study aims to address the existing knowledge gap by applying three distinct stimulation mechanisms – molecular, hierarchical, and acoustical – on biofabricated TM scaffolds. In this regard, relevant bioactive molecules were identified to trigger the desired expression of TM-specific genes on electrospun meshes. Subsequently, additive-manufactured filaments were deposited on the nanofibrous meshes to investigate the influence of 3D hierarchy. Finally, acoustical stimulation was applied using a custom-built bioreactor setup, partially mimicking the native tissue niche. The acousto-vibrational characterization of the stimulated samples revealed an amplified oscillatory behavior at specific frequencies, which was shown to positively impact the TM wound healing mechanism. In summary, this work demonstrates that a synergistic integration of suitable bioactive agents, 3D hierarchical structures, and acoustical vibrations promotes the formation of an aligned extracellular matrix relevant for TM regeneration.

Biological sciences/Biotechnology/Tissue engineering

Biological sciences/Biotechnology/Regenerative medicine

Biological sciences/Stem cells/Mesenchymal stem cells

Biological sciences/Biotechnology/Biomimetics

Eardrum

Tissue engineering

Biofabrication

Mesenchymal stromal cells

Acoustical stimulation

Bioreactor

Collagen type II

The human auditory system relies on an intricate interplay of mechanically vibrating tissues and sensorineural cells for perceiving incoming acoustic waves. In this regard, the tympanic membrane (TM), commonly known as the eardrum, is the first point of action within the air conduction pathway of human hearing.¹ It is a thin tissue located at the intersection of the outer and middle ear, that transmits the sound pressure in the auditory canal as mechanical vibrations of the adjoining ossicles. Multiple studies have attributed the distinctive collagen arrangement of TM, comprising radially and circumferentially aligned fibers, to its effective sound transmission.^2,3 Therefore, any injury to the TM results in a complete disruption of the conduction pathway, leading to impaired hearing capabilities. The chronic inflammation of the middle ear, chronic suppurative otitis media (CSOM), has been identified as the most widespread cause of TM perforations and the subsequent loss of hearing.⁴ Autologous grafting has conventionally been the “gold standard” in the treatment of perforated TM. However, this technique often yields a suboptimal hearing restoration.⁵ Tissue engineering has recently emerged as a promising alternative in this direction, where numerous regenerative therapies have been reported for a biomimetic reconstruction of the damaged eardrum.

The field of tissue engineering has gained considerable attention in the last few decades, especially after its promising integration with stem cell therapy.⁶ A wide range of tissues has been explored in this regard for their potential reconstruction using uncommitted progenitor cells such as totipotent zygote, haematopoietic stem cells, embryonic stem cells, and mesenchymal stromal cells (MSCs).⁷ Among them, MSCs have been extensively investigated due to their ability to differentiate into adipocytes, chondrocytes, osteoblasts, and fibroblasts.⁸ First isolated from the bone marrow, these non-haematopoietic stromal cells are now commonly derived from various adult connective tissues. Moreover, a growing relevance of MSC-based therapies for TM injuries has been reported in recent years.⁹ Multiple studies have proposed their application for regenerating the perforated eardrum.^2,3,10–16 However, with the exception of one, none of them have actually attempted their differentiation towards TM-like phenotype.¹⁴ In this unique case, a previously reported protocol for cartilage differentiation was adapted for TM applications, although without evaluating the influence of all the medium constituents.¹⁴ Therefore, an investigation on the medium composition is imperative to identify the necessary biochemical cues for differentiating the applied MSCs towards TM regeneration.

In general, the stem cell microenvironment is known to play a critical role in steering cell fate. Conventionally, bioactive agents, such as growth factors, morphogens, and hormones, have been the primary contributors to shaping the desired microenvironment for engineering functional tissue constructs. However, recently, alternative routes in the form of topographical, mechanical, and electrical stimulation have demonstrated an equally significant effect in regulating the associated signaling pathways.¹⁷ As a result, current efforts in this direction, incorporate relevant physical cues in combination with well-established biochemical stimuli to drive the cellular differentiation.¹⁸ The mechanosensitivity of stem cell niches toward specific geometrical and biomechanical features can induce their phenotypical change into the desired cell lineages. A variety of physical interactions ranging from shear stress, tensile strain, and hydrostatic pressure have been applied to induce the required mechanotransduction pathways.¹⁹ Despite this, not many stimulatory approaches have yet been investigated for actively modulating the regeneration of human eardrum using stem cell therapies. In this regard, the patented oscillatory technology of Danti et al. remains the only dynamic culture environment reported so far for the mechanical stimulation of eardrum prostheses.²⁰ In vitro assessment of scaffolds cultured in these cyclic strain bioreactors demonstrated a favorable influence of the vertical oscillations over MSC infiltration and differentiation towards TM-like fibroblasts.^10,14 However, the resultant macroscopic forces within these fully-immersed constructs fail to mimic the native microenvironment of the human eardrum.

This could be attributed to the fact that the human TM is a peripherally suspended tissue that is consistently subjected to nano-vibrational motion by the incoming sound waves. Thus, replicating these nano-acoustical vibrations in vitro might have a more pronounced effect on steering the cell fate towards TM regeneration, as compared to a macroscopic stimulation. Prior studies in this direction have, in fact, shown a favorable influence of acoustic waves in promoting the expression of extracellular proteins such as collagen type II by MSCs.²¹ However, the majority of the investigations to date have been performed with low-intensity ultrasound stimulation, that is, sound frequencies beyond the human audible range.^21–25 In addition, these sound waves have been predominantly applied onto a standard culture environment, where they traverse through the culture media before reaching the cells. Therefore, such mechanical stimulation models are not a good representation of the acousto-vibrations experienced by the native TM during its transmission mechanism. In this regard, a custom-designed bioreactor capable of mimicking the air conduction pathway of human hearing would be a more appropriate system to unravel the necessary biomechanical cues for TM regeneration.

Overall, a synergistic integration of biochemical, biophysical, and biomechanical signals seems essential for guiding the stem cell differentiation towards the desired lineage. Yet, as highlighted earlier, the role of these critical factors in the differentiation of MSCs for TM applications has still not been fully evaluated. Hence, the present study aims to address this existing knowledge gap in the implementation of MSC-driven tissue engineering therapies for eardrum injuries. Three distinct stimulation mechanisms, namely molecular, hierarchical, and acoustical, were chosen to assess the influence of the corresponding biochemical, biophysical, and biomechanical cues on biomimetic constructs (Fig. 1). To achieve this, the molecular signaling was investigated using a cocktail of bioactive agents in the culture media. The hierarchical mechanotransduction was triggered by an anatomically relevant scaffold topography replicated using a combination of electrospinning (ES) with fused deposition modelling (FDM). Finally, the acoustical stimulation was applied through a bioreactor setup that drew inspiration from the native tissue niche and facilitated a direct exposure of TM scaffolds to the incoming sound waves. Therefore, by exploring a sequential combination of these stimulatory approaches, this study intends to understand the role of cellular microenvironment in guiding the MSC differentiation towards the production of eardrum-relevant extracellular matrix (ECM).

2.1. Biofabrication of TM scaffolds

Two classes of scaffolds were biofabricated to investigate the role of 3D hierarchy in steering the MSC differentiation towards TM regeneration. Hierarchical TM scaffolds were biofabricated by applying a systematic integration of ES with FDM. Considering plain electrospun membranes represent the simplest version of PEOT/PBT scaffolds relevant for TM tissue engineering, they were chosen as the control to the hierarchical constructs. Qualitative assessment of the resultant scaffolds was performed to validate the homogenous distribution of their electrospun nanofibers and additive-manufactured filaments (Fig. S1). The electrospun nanofibers exhibited an average diameter of 715 ± 103 nm, while the thickness of the additive-manufactured filaments measured at 50.87 ± 5.45 µm.

2.2. Molecular stimulation

Bioactive molecules are frequently used to modulate the stem cell differentiation pathways. Therefore, molecular stimulation was investigated as the first activation approach to direct the MSC fate towards TM-like cells. Figure 2A presents a schematic illustration of the media compositions assessed. A sequential supplementation of the selected formulations was implemented to test four key bioactive factors: AsAP, TGF-β1, proline, and dexamethasone. The M3 and M4 compositions were envisioned as the differentiation media, whereas the proliferation medium, M1, was chosen as the negative control. In addition, an intermediate control in the form of M2 was included to bridge the gap between the proliferation and differentiation compositions.

The relative metabolic activity showed a gradual increment with time across all the media compositions (Fig. 2B). However, the rate of this increase was significantly lower for scaffolds in M1 and M2 in comparison to their differentiation counterparts. Overall, the M3 composition demonstrated the highest metabolic activity across all the time points. A similar trend was observed in the gene expression as well, where M3 emerged as the most optimal composition for the majority of genes (Fig. 2C). In this regard, the presence of AsAP and TGF-β1 alone was noted to have a stronger influence on the upregulation of COL2A1 and FGF10, as compared to the additional inclusion of proline and dexamethasone to the same mixture. Although none of these bioactive molecules were found to be effective in promoting the FGF7 expression. Finally, all the media compositions revealed a marked downregulation of COL1A1 over time, which is desirable for TM regeneration.

Immunofluorescence imaging of the cultured scaffolds revealed a substantial increase in collagen type II production upon the addition of AsAP and TGF-β1 in culture media (Fig. 2D). As a result, M3 and M4 exhibited a consistent deposition of extracellular collagen type II over the electrospun meshes, in contrast to M1 and M2.

2.3. Acoustic bioreactors: design and characterization

The acoustic bioreactors were designed as an add-on device for a conventional 6-well plate (Fig. 3). Within each well plate, the setup comprised two discrete elements: (1) well inserts to culture the samples at an air-liquid interface, and (2) an array of piezoelectric disc benders to generate the required sound waves. Figure 3A schematically illustrates the optimization of the well insert design to determine the ideal distance for acoustical waves to travel from the piezoelectric disc to the sample without experiencing notable damping. Three sample placement heights (h_SP) were investigated for this purpose using customized spacers (Fig. S2A). Throughout the frequency range of the applied sound signals, it was observed that a quantifiable electrical signal was detected only at certain frequencies. These frequencies corresponded to the resonant frequencies of the disc benders and resulted in a marked amplification of the output voltages (Fig. S2B; presented for h_SP = 5 mm). A similar trend was detected across all the tested conditions, where the maximum activity was noted in the range of 500–5000 Hz (Fig. 3B). Moreover, by applying a threshold voltage of 1.2 V (10% of input voltage) to the obtained dataset, a significant variation between the different h_SP was revealed. Figure 3C summarizes this comparative analysis with the heights of 3 mm and 5 mm outperforming their 0 mm and 7 mm counterparts. Overall, the source-sample gap was demonstrated as a critical parameter in the production of acoustic bioreactors, highlighting a substantial influence on the sound reception capability.

The optimized well inserts were manufactured by milling acrylic slabs (Fig. S3A-B). The top view of the produced inserts depicts the area designated for the sample placement, while the bottom view displays the compartment intended for the culture medium (Fig. S3C). Figure 3D illustrates the final layout of a 6-well plate, prepared for stimulation, with the TM scaffolds held at an air-liquid interface. In addition to these, control inserts were fabricated for culturing the TM scaffolds under the standard submerged environment (Fig. S3D). Ultimately, the electronic components of the bioreactor were assembled in a plug-and-play fashion to enable easy testing and replacement of individual transducers in case of damage (Fig. 3E).

Following the design and fabrication of the optimized bioreactors, a systematic characterization was carried out to evaluate the acousto-vibrational outcome of the stimulated scaffolds. A COMSOL-based computational model was applied as the first investigative approach to predict their nano-vibrational response across the human audible range (Fig. 4A-C). Theoretical simulations conducted within the bioreactor environment revealed that plain scaffolds demonstrate a lower resonant frequency with respect to the hierarchical scaffolds, for the same mode of vibration (Fig. 4B). This was accompanied by a reduced displacement as well across all the eigenfrequencies (Fig. 4C). Furthermore, upon comparing the air-liquid interface of the bioreactor with an air-air configuration, a notably higher displacement was obtained for the latter, thereby highlighting the damping influence of the liquid (Fig. S5C).

Subsequent to the in silico analysis, experimental setups were devised to characterize the acousto-vibrational response empirically. For this, a preliminary qualitative assessment was performed to visualize the oscillatory motion induced within the acoustically stimulated samples (Fig. S4). A graphical representation of the employed test setup is presented in Fig. S4A. The recorded videos revealed the stained dot to be consistently going in and out of focus, thereby confirming scaffold vibration along the orthogonal axis (Fig. S4B, Video S1, Supporting Information). This was further corroborated with the help of fluctuating intensity values measured across sequential frames of the same recording (Fig. S4B-C). However, due to the limited frame rate offered by the stereomicroscope, it was not feasible to accurately sample and reconstruct these rapid oscillations.

Thus, to facilitate a precise quantification of the sound-induced vibrations in TM scaffolds, a high-speed laser displacement sensor was employed as shown in Fig. 4D. A real-time displacement measurement of plain and hierarchical constructs was carried out while they were exposed to a frequency sweep of 1 Hz to 4000 Hz. This was conducted for both air-air and air-liquid interfaces (Fig. S5A). For the air-air characterizations, a significant amplification in the vibrational response was observed at specific frequencies, thereby validating the eigenfrequency analysis of the computational model (Fig. 4E). Two major peaks were identified in this respect, one around 1200 Hz and another at 3200 Hz, which prompted a comparative investigation of two distinct frequency ranges: (F1) 1 Hz to 2000 Hz, and (F2) 2000 Hz to 4000 Hz. Overall, F2 exhibited a higher maximum displacement with respect to F1. In addition, a greater vibrational motion was detected in the plain TM scaffolds (F1: 32.3 ± 2.3 µm and F2: 45.6 ± 0.7 µm) as compared to their stiffer hierarchical counterparts (F1: 25.5 ± 0.6 µm and F2: 33.0 ± 0.4 µm).

Finally, Fig. 4F illustrates the predicted damping of oscillations as the scaffold's underlying surface transitions from air to liquid. Theoretical simulations projected a 27-fold reduction in plain electrospun membranes and a 32-fold reduction in hierarchical ones (Fig. S5B). As a result, a maximum displacement of 1.7 µm was estimated across all samples, which was below the detection limit of the employed test bench for the air-liquid interface (Fig. S5C).

3.4. Acoustical stimulation: validation and optimization

An in vitro validation of the acoustic bioreactors was performed before assessing their ability to drive the MSC differentiation towards TM regeneration. These investigations, conducted utilizing the acousto-vibrational parameters from the previous section, served to establish the stimulation workflow and confirm its resultant cytocompatibility.

The cytocompatibility of the bioreactor was tested under both static and dynamic conditions, with respect to the standard culture environment (control). Fig. S6A shows the experimental setup implemented for culturing and acoustically stimulating the dynamic samples. No cytotoxic effects were detected from either the bioreactor or the resulting sound waves (Fig. S6B). Moreover, a significant improvement in the cell metabolic activity was observed for the bioreactor-cultured constructs as compared to control samples cultured under the conventional configuration (Fig. S6C). Following the in vitro validation of the bioreactors, all subsequent differentiation studies were performed after the formation of a cellular monolayer on the TM scaffolds (Fig. S6D).

The first necessary step for implementing the acoustical stimulation was to identify the optimal frequency range for its application. Thus, a preliminary screening was carried out based on the audio parameters evaluated during the vibrational characterization (Fig. 5A-C). Cultured MSCs were exposed to two distinct frequency ranges, F1 and F2, and their responses were compared against a static control (Fig. 5A). A significant improvement in the cell metabolic activity was obtained upon the application of sound waves, irrespective of the frequency (Fig. 5B). However, no statistically significant differences were observed between F1 and F2 at any time point. In addition, the gene expression results also demonstrated the dynamic samples to be outperforming their static counterparts (Fig. 5C). The acoustical stimulation of MSCs was noted to trigger a marked downregulation of COL1A1, accompanied by an upregulation of COL2A1 and FGF10. This was found to be considerably more pronounced for F2 as compared to F1. In contrast, no detectable trend was evident for the FGF7 expression across all conditions. Overall, the incoming sound waves were determined to have a favorable influence on the MSC behavior, especially in the frequency range of 2000–4000 Hz.

3.5. Integrated stimulation

The ultimate objective of this study was to evaluate distinct cellular cues aimed at guiding the MSC differentiation toward the production of TM-like ECM. Following the initial optimization of molecular and acoustical approaches on plain electrospun scaffolds, anatomically relevant 3D architecture was introduced to assess the influence of hierarchy-induced mechanotransduction. Consequently, a trilateral stimulation (molecular, hierarchical, and acoustical) was implemented for a duration of 4 weeks subsequent to one week of proliferation under standard conditions (Fig. 5D).

Quantitative investigation of the integrated stimulation highlighted a favorable influence of the applied acousto-vibrations on MSC response. In this regard, cells cultured in the dynamic environment were metabolically more active than the ones cultured without any sound stimulus (Fig. 5E). This was observed unanimously for both the scaffold types across all the differentiation time points, week 2 through week 5. In addition, the acoustically stimulated samples demonstrated a significantly lower expression of COL1A1 and higher expression of COL2A1 and FGF10 as compared to their static counterparts (Fig. 5F). Analogous to the cell metabolic activity, an identical trend was revealed for both plain and hierarchical constructs, indicating minimal influence of the presented scaffold geometry on mechanosensitivity of the cultured MSCs.

Furthermore, qualitative assessment of the integrated stimulation using immunofluorescence imaging validated the gene expression quantifications for collagen type II (Fig. 6). A pronounced cell attachment was obtained with both the scaffold types on day 1 (W0), followed by a homogeneous cell proliferation at day 7 (W1). In case of the hierarchical scaffolds, regions along the additive-manufactured filaments showed higher cell density and alignment compared to areas without them. Thus, the inclusion of 3D hierarchy allowed the differentiated MSCs to maintain their desired orientation throughout the 5-week culture period, facilitating an aligned deposition of collagen type II-rich ECM.

Finally, the protein quantification results indicated a consistent secretion of collagen type II across all the tested conditions, while a significant decrement in the collagen type I concentration was obtained with acoustically stimulated hierarchical scaffolds (Fig. S7).

According to the latest World Health Organization report, 2.5 billion people are projected to experience varying degrees of hearing loss by 2050. TM perforations are one of the most widespread injuries to the human ear, causing partial or complete hearing loss as a result of impaired sound conduction. Therefore, the current work aims to advance the ongoing efforts of using MSCs in combination with tissue-engineered scaffolds to generate suitable implants for TM regeneration.

Over the last few decades, human MSCs have emerged as one of the most frequently utilized cell sources for regenerative applications.²⁶ Numerous studies have highlighted their relevance for a wide range of tissues, spanning from bone,²⁷ cartilage,²⁸ fat,²⁹ to cardiovascular,³⁰ and neural.³¹ The versatility of MSC-driven therapeutics can be attributed to their capacity for multilineage differentiation, in conjunction with their inherent abilities to self-renew and regulate immune responses.³² These unique characteristics are known to be intricately guided by a myriad of signals and cues originating from their surrounding microenvironment. Hence, by manipulating a precise combination of biochemical, biophysical, and biomechanical stimuli, the employed MSCs can be steered towards the desired phenotype and functions. In this study, three distinct stimulation routes were identified and sequentially combined for driving the MSC differentiation on biomimetic eardrum scaffolds. Among them, biochemical activation was examined as the first stimulatory approach to replicate the TM wound healing, followed by the activation of mechanotransduction pathways.

The ECM of the human eardrum is predominantly composed of collagen type II.³³ However, during the healing phase of a perforated TM, the de novo ECM tissue typically produced is rich in collagen type I, which is gradually converted into type II.³⁴ Thus, a successful regeneration of the TM would involve a significant decrease in collagen type I expression accompanied by an increase in type II expression. In addition to the expression of collagen type I and type II, the genotypic profile of TM-derived cells revealed a strong association of FGF7 and FGF10 proteins with TM repair.³⁵ Several investigations in this direction have highlighted the role of FGF7 and FGF10 in TM wound healing.^36,37 Therefore, the efficiency of MSC differentiation in the context of TM applications was quantitatively evaluated herein with respect to the gene expression of COL1A1, COL2A1, FGF7, and FGF10. This was further corroborated by qualitative assessment of the extracellular collagen type II deposited on the cultured scaffolds along with their cell metabolic analyses. Studies have shown that MSCs have a higher metabolic activity during differentiation in contrast to when cultured in basic medium over the same period.²⁸

The differentiation medium for the biochemical activation of the cultured MSCs was formulated with respect to four key bioactive molecules: AsAP, TGF-β1, proline, and dexamethasone. AsAP, a stable derivative of vitamin C, was applied for its well-known ability to induce collagen type II secretion during the chondrogenic differentiation of MSCs.³⁸ In addition to AsAP, the family of TGF-β growth factors, in particular TGF-β1, has been established as a potent stimulator of extracellular collagen production.^39,40 Furthermore, the incorporation of proline, a non-essential amino acid in humans was investigated, considering its role in the collagen biosynthesis.^41,42 Finally, dexamethasone, a synthetic glucocorticoid widely employed for downregulating the COL1A1 expression in stromal cells was included as well.⁴³ The biochemical stimulation achieved with M3 corroborates the findings previously obtained by Moscato et al., where an identical media composition was applied for differentiating MSCs for TM applications.¹⁴ The authors reported a successful commitment of MSCs into fibroblast lineage-producing TM-like collagens, on star-branched poly(ε-caprolactone)-based electrospun scaffolds, using medium supplemented with AsAP and TGF-β1.

In the past, AsAP supplemented media has been tested to enhance the extracellular collagen production by MSCs cultured on PEOT/PBT-based TM scaffolds.² In addition to this biochemical stimulus, the same study demonstrated that the presence of additive-manufactured filaments facilitated the cellular alignment and subsequent collagen deposition.² The present work aimed to further investigate the influence of 3D hierarchy in promoting the production of extracellular collagen type II. It is known that cells activate mechanotransduction pathways by sensing mechanical signals from their surrounding microenvironment, triggering specific cellular functions in response.⁴⁴ These mechanical signals can either emerge as contact stresses resulting from topographical or geometrical cues, or as dynamic stresses induced by external stimuli such as electricity, magnetism, or light.⁴⁵ Recently, multiple reports in literature have demonstrated that the application of sound waves can produce a similar effect.^46–48 Therefore, considering the native environment of the human eardrum, the current study investigated acousto-vibrational stimulation as an additional approach to promote MSC-driven deposition of de novo ECM relevant for TM regeneration. Consequently, the cultured MSCs experienced two distinct classes of mechanical stresses: contact stress in the form of 3D hierarchy and dynamic stress caused by incoming sound waves.

The acoustical stimulation was applied through a custom-built bioreactor setup, drawing inspiration from the air conduction pathway of human hearing. The normal audible spectrum spans from 20 Hz to 20 000 Hz; however, it is within the midrange of 200 Hz to 4000 Hz where most of the human hearing resides.⁴⁹ This is the range of frequencies in which the auditory system is most sensitive, enabling a clear perception of everyday conversations, television broadcasts, and musical instruments across various genres.⁵⁰ Frequencies above 4000 Hz, also known as treble, primarily comprise high-pitched sounds such as electronic alarms and whistles, which are less commonly encountered. Therefore, considering the higher prevalence of low and midfrequency acoustic signals in an average human’s life, this study selected a frequency range of 1 Hz to 4000 Hz to understand its influence on MSC differentiation. The chosen range was further split into F1 and F2 to present a comparative analysis between the lower and upper mid-range frequencies.

In general, it is known that as the distance between the source and the sensor increases, there is a gradual decrease in the detected amplitude. This phenomenon was carefully analyzed to optimize the bioreactor design. In this regard, a h_SP of 5 mm was concluded to effectively transmit the generated acoustical signals, while allowing an adequate space for media storage. Following the bioreactor fabrication, a complete characterization of their resultant acousto-vibrational response was critical before initiating any cellular investigation. A combination of qualitative and quantitative techniques was applied to evaluate the vibrations induced within the acoustically stimulated TM scaffolds. Initially, a stereomicroscope-based imaging technique facilitated a visual validation of the vibrating scaffolds. However, considering the camera's limited capture rate, a laser-based quantification method was subsequently implemented to precisely measure their displacement. In accordance with the Nyquist theorem, which dictates that a waveform can be accurately reproduced if the sampling frequency is at least twice the original frequency, it was established that the stereomicroscope's 100 frames per second (FPS or Hz) could accurately sample audio signals only up to 50 Hz. In contrast, the use of laser displacement sensors extended this capacity to a frequency of 10 000 Hz.

The integrated stimulation of the eardrum scaffolds provided three simultaneously applied cues to promote the deposition of TM-relevant ECM. A strong influence was noted with respect to the molecular and acoustical stimulation. The relevance of 3D hierarchy was evaluated in conjunction with the supplemented media and biomimetic vibrations, although without a pronounced resulting matrix difference. Despite this, the 3D hierarchy guided the aligned deposition of collagen type II-rich ECM, which is an essential requirement for full eardrum regeneration.

To conclude, this work explored the role of cellular microenvironment in driving the MSC differentiation toward eardrum regeneration. Three critical cues, namely biochemical, biophysical, and biomechanical, were investigated within the context of biomimetic TM scaffolds, by applying a sequential combination of molecular, hierarchical, and acoustical stimulations, respectively. Traditionally, biochemical signaling has been implemented as the primary pathway for guiding the differentiation of stem cells into the desired lineage. However, to date, no comprehensive identification of the relevant bioactive molecules for TM regeneration has been reported. The current study evaluated this by analyzing four distinct media compositions and found AsAP and TGF-β1 to be significantly more important than proline and dexamethasone. In addition to their response to biochemical stimuli, MSCs are widely recognized for their high mechanosensitivity. Therefore, physical forces in the form of sound-induced vibrations and the scaffold topography were incorporated to mimic the native eardrum microenvironment. In this regard, an innovative bioreactor was designed, produced, and optimized for acoustically stimulating hierarchical TM constructs at an air-liquid interface. The acousto-vibrational characterization of the stimulated samples revealed an amplified oscillatory behavior at specific frequencies, corresponding to their resonant frequencies. Overall, a superior biological response was obtained in the presence of incoming sound waves, especially when applied in the range of 2000 Hz to 4000 Hz. This was confirmed with respect to the cell metabolic activity and expression of specific genes associated with wound healing in the TM. A simultaneous downregulation of collagen type I and upregulation of collagen type II was achieved with the dynamic culture in acoustic bioreactors. In contrast, the 3D hierarchy did not have a significant impact on modulating the MSC response; but it did contribute to aligning the cultured cells and the subsequent extracellular collagen deposition. Future efforts in this direction will focus on extended application of the stimuli, with the goal of creating fully matured eardrum implants comprising a radially and circumferentially arranged ECM rich in collagen type II protein.

4.1. Biofabrication of TM scaffolds

TM scaffolds were fabricated as previously described by Anand et al.^2,12 Briefly, two categories of scaffolds were manufactured using the validated grade of PEOT/PBT copolymer, that is, 300PEOT55PBT45 (kindly provided by PolyVation B.V., the Netherlands): plain and hierarchical.^2,3,11,12 The plain constructs were biofabricated with ES alone, whereas a successive combination of ES and FDM was applied for manufacturing their hierarchical counterparts.

Electrospinning: Electrospun meshes were fabricated on Fluidnatek LE-100 (Bioinicia S.L., Spain) using the optimized parameters reported for PEOT/PBT-based TM scaffolds.² In short, 17% (w/v) of the copolymer was dissolved in a 70:30 (v/v) solvent mixture of trichloromethane (anhydrous, Sigma-Aldrich, the Netherlands) and hexafluoro-2-propanol (analytical reagent grade, Biosolve B.V., the Netherlands). The resultant polymer solution was ejected through a 0.8 mm spinneret at a flow rate of 0.9 mL·h^− 1 on top of a stainless-steel collector (diameter = 200 mm, length = 300 mm) rotating at 200 revolutions per minute (rpm). A potential difference of 14 kV was applied between the spinneret (11 kV) and the collector (-3 kV) while maintaining a working distance of 10 cm. The ES process was conducted for a duration of 30 min at an ambient temperature of 23°C and 40% relative humidity.

Fused deposition modelling

A previously identified biomimetic geometry was chosen, envisioning the full reconstruction of the human eardrum.¹² Following the aforementioned ES step, polymeric patterns were deposited onto the nanofibrous meshes using FDM, resulting in the formation of hierarchical TM constructs. Analogous to the ES technique, the optimized parameters reported by Anand et al. were employed for depositing the PEOT/PBT-based microfilaments.² PEOT/PBT copolymer was molten at 190°C and extruded through a 70 µm ceramic-tip needle (ID70, DL003005AC, DL technologies, USA) by applying a combined effect of pneumatic pressure (750 kPa) and screw rotation (60 rpm).

4.2. Qualitative assessment of TM scaffolds

A qualitative evaluation of the biofabricated scaffolds was conducted using scanning electron microscopy (SEM; Jeol JSM-IT200, the Netherlands) for the electrospun nanofibers, and stereomicroscopy (Nikon SMZ25, the Netherlands) for the FDM microfilaments. The SEM micrographs were captured at an accelerating voltage of 10 kV and working distance of 10 mm, on gold sputter coated samples (45 seconds, Quorum Technologies SC7620, UK).

The subsequent image processing and analysis was performed on Fiji software (version 2.9.0, https://fiji.sc/) to determine the fiber diameters obtained during ES (n = 3 samples × 30 measurements each) and during FDM (n = 3 samples × 5 measurements each), respectively.

4.3. Cell seeding on TM scaffolds

The biofabricated TM scaffolds were disinfected in 70% ethanol (VWR International B.V., the Netherlands) for 4 h, followed by an overnight evaporation in a biosafety cabinet. The resultant disinfected samples were placed in 24-well plates (non-treated, 734–0949, VWR International B.V., the Netherlands) and held by O-rings (FKM 75 51414, 11.89 × 1.98 mm, ERIKS N.V., the Netherlands) to prevent them from floating. Thereafter, they were washed twice with Dulbecco's phosphate-buffered saline (DPBS, Sigma-Aldrich, the Netherlands) and incubated in tissue culture media overnight.

Minimum essential medium (α-MEM; GlutaMAX supplement, no nucleosides, 11534466, Fisher Scientific, the Netherlands) supplemented with 10% heat-inactivated fetal bovine serum (FBS; F7524, Sigma-Aldrich, the Netherlands) and 1% penicillin/streptomycin (PS; 100 U·mL^− 1, Thermo Fisher Scientific, the Netherlands) was used as the tissue culture media for the initial seeding and proliferation of human MSCs (Primary, PT-2501, Lonza, the Netherlands). A total of 1 × 10⁵ cells were seeded onto each scaffold and cultured for 7 days, with media changes every alternate day. After 7 days of proliferation, the stimulatory mechanisms were introduced to investigate the MSC differentiation.

4.4. Media compositions for molecular stimulation

Distinct tissue culture media was evaluated to understand the role of bioactive molecules in triggering the differentiation of MSC-cultured TM scaffolds. Four media compositions (M1 – M4) were sequentially formulated and studied independently. The proliferation medium, α-MEM supplemented with 10% FBS and 1% PS, was chosen as the negative control and henceforth termed M1. The next composition, M2, was a slight modification of M1, where the α-MEM was replaced by Dulbecco's modified Eagle's medium (DMEM; high glucose, GlutaMAX supplement, pyruvate, 12077549, Fisher Scientific, the Netherlands), while keeping the remaining components unchanged. M3 was a further extension of M2, supplemented with 57.8 µg·mL^− 1 of L-ascorbic acid 2-phosphate (AsAP; A8960, Sigma-Aldrich, the Netherlands) and 10 ng·mL^− 1 of transforming growth factor-β1 (TGF-β1; 100 − 21, PeproTech, the Netherlands). Finally, M4 was an enriched version of M3 with the additional inclusion of 40 µg·mL^− 1 of proline (P5607, Sigma-Aldrich, the Netherlands) and 0.1 µM of dexamethasone (D8893, Sigma-Aldrich, the Netherlands).

4.5. Design and optimization of acoustic bioreactors

The fabrication and implementation of the acoustic bioreactors was envisioned as an add-on device for commercial 6-well plates (non-treated, 22.23.136, VWR International B.V., the Netherlands). Each bioreactor comprised two distinct components: (1) a piezoelectric disc bender to generate the sound waves, and (2) a customized well insert to hold the sample and culture media. The piezoelectric sound source was obtained commercially, whereas the well insert was designed and produced in-house.

A circular cased transducer (Diaphragm External Piezo Buzzer, 724–3166, RS Components, the Netherlands) with a diameter of 35 mm and thickness of 530 µm was chosen to fit within the individual rings marked on the well plate lids, corresponding to the well below. The transducer was affixed to the inner surface of the lid using a double-sided tape (16084-20, Ted Pella, USA), with the ceramic disc exposed outward to facilitate the propagation of sound waves into the underlying well. The well insert was designed to allow cell culture under air-liquid interface conditions. The modeled structure was milled from a 15 mm thick acrylic sheet (1000060015, Perlaplast, the Netherlands) using a CNC milling machine (monoFab SRM-20, Roland DG, Japan). The overall shape and size of the produced insert was adjusted to fit an individual well of the well plate: cylindrical with diameter = 34.3 mm and height = 15 mm. Within the insert, the diameters for sample placement were defined by the standardized dimensions reported for the PEOT/PBT-based TM scaffolds: outer diameter (OD_SP) = 15.7 mm and inner diameter (ID_SP) = 12 mm.^2,12 In contrast, the height for sample placement (h_SP, measured from the top) was optimized to enhance the efficiency of sound detection at that distance.

Three distinct h_SP (3 mm, 5 mm, and 7 mm) were investigated by applying a custom-built test bench. Two piezoelectric transducers were arranged complementarily, with the first transducer converting electrical energy to sound energy as a buzzer and the second transducer acting as a sensor, converting sound energy to electrical energy. A gap of 3 mm, 5 mm, and 7 mm was created between the two transducers using acrylic spacers, fabricated with the same diameters as the OD_SP and ID_SP. The buzzer was set to discrete frequencies, starting from 100 Hz and continuing up to 20 000 Hz, with an interval of 100 Hz between each frequency. The electrical input (12 V, ramp function) was provided to the buzzer with a waveform generator (DG1022Z, RIGOL Technologies EU GmbH, Germany), whereas the electrical output of the sensor was measured on an oscilloscope (XDS3064AE, OWON Technology, P.R. China).

4.6. Production of acoustic bioreactors

Subsequent to the design optimization, a bulk fabrication of the acoustic bioreactors was carried out. Six piezoelectric transducers corresponding to each well of a 6-well plate were attached to its lid. The transducers were connected in parallel through a pair of non-fused terminal strips (703–3833, RS Components, the Netherlands) and a pair of splicing connectors (Wago 221–413, RS Components, the Netherlands). The components were assembled together with jumper wires in a ‘plug and play’ fashion without any use of soldering. In case of the well inserts, a batch of 18 pieces were mass produced from each round of milling (Fig. S3A). The milling was performed with a 2 mm carbide milling bit, except the tweezer hole, which was drilled with a 1 mm carbide milling bit.

4.7. Characterization of acoustic bioreactors

Theoretical and experimental characterizations of the acoustic bioreactors were performed with respect to the nano-vibrational response of the corresponding TM scaffolds. Both plain and hierarchical constructs were analyzed to identify the optimal frequency range for their acoustical stimulation.

Theoretical

Computational models (COMSOL Multiphysics 6.1, Comsol B.V., the Netherlands) of the bioreactor were designed to simulate the structural behavior of the stimulated samples. CAD files of the bioreactor setup and TM scaffolds were imported within the acoustics module of COMSOL. The pre-defined physics of acoustic-structure interaction was applied to investigate the structural mechanics of the plain and hierarchical constructs in response to incoming acoustic waves up to a frequency of 20 000 Hz. An eigenfrequency study was performed to identify their resultant resonant frequencies (RF) along with the maximum displacement noted at that frequency.

Experimental

The acoustic bioreactors were characterized experimentally in terms of the nano-vibrational response triggered within the stimulated scaffolds. A qualitative and quantitative evaluation of the vibrating membranes was carried out based on a wide range of acoustical frequencies. The qualitative assessment was conducted using a stereomicroscope (Nikon SMZ25, the Netherlands) at a magnification of 15×. Plain TM scaffolds were subjected to discrete frequency sweeps covering a range of 200 Hz each, starting from 1 Hz and going up to 5000 Hz. A sweep time of 2 s was chosen per cycle within a given frequency range. Real-time videos were acquired for a duration of 1 min to visualize the vibrational motion of a microscopic dot marked on the electrospun mesh (Video S1, Supporting Information). Individual frames were extracted from the video and processed on Fiji software. The images were analyzed by measuring the pixel intensity of the vibrating dot, with higher values indicating ‘in focus’ whereas the lower ones denoted ‘out of focus’.

Following the qualitative verification of the acousto-mechanical vibrations, a quantitative investigation was performed to measure the amplitude of these oscillatory movements. A high-speed laser displacement sensor (LK-G5000 Series, Keyence, the Netherlands) and a confocal displacement sensor (CL-3000 Series, Keyence, the Netherlands) were employed. The air-air measurements were conducted using the LK-H057K sensor head of the LK-G5000 Series. The laser was directed onto the center of the scaffold's surface, while it was stimulated from the opposite side. For air-liquid interface, the bioreactor setup was positioned on a glass petri dish, and measurements were taken through the glass bottom using the CL-P030 sensor head of the CL-3000 Series. The scaffolds were subjected to audio signals through frequency sweeps ranging from 1 Hz to 4000 Hz, and the characterization was performed at a sampling cycle of 100 µs. Subsequently, the maximum amplitude of scaffold vibration was quantified within two distinct ranges: (F1) 1 Hz to 2000 Hz, and (F2) 2000 Hz to 4000 Hz. All measurements were conducted for n = 5 samples.

4.8. In vitro validation of acoustic bioreactors

Preliminary biological assessments were carried out to validate the cytocompatibility of the bioreactor setup along with the accompanying acoustical stimulation. Plain TM scaffolds were seeded with 1 × 10⁵ MSCs each, and thereafter, transferred to the bioreactor after one day in culture. The scaffolds were subjected to sound waves for 4 h per day over a three-day period. The stimulation involved a frequency sweep ranging from 100 Hz to 10 000 Hz, with each cycle lasting 30 s. After three days of daily stimulation, the samples were kept for an additional day in culture without stimulation. Subsequently, quantitative and qualitative assessments of the TM scaffolds were performed in terms of their cell metabolic activity and immunofluorescence imaging, respectively.

4.9. Acoustical stimulation

Following the in vitro confirmation of the bioreactors’ cytocompatibility, the influence of acoustical stimulation was investigated on the MSC differentiation, specifically targeting the deposition of relevant ECM for eardrum regeneration. Analogous to the molecular stimulation, a monolayer of MSCs was allowed to form over the TM scaffolds prior to the differentiation protocol. After 7 days of culture in 24-well plates, the samples were moved to the acoustic bioreactors. Dynamic samples were cultured in a dedicated incubator to maintain consistent conditions during the periods of stimulation and non-stimulation.

A frequency selection study was conducted to tune the stimulation parameters for MSC differentiation. Plain TM constructs were used in this regard, in combination with the optimized media composition. Two sets of frequency sweeps were chosen, based on the acousto-vibrational response of the fabricated scaffolds: (F1) 1 Hz – 2000 Hz, and (F2) 2000 Hz – 4000 Hz. A 20 s cycle was applied for both the ranges, with a total duration of 10 000 s, thereby resulting in 500 cycles of acousto-mechanical vibrations per day. The cultured scaffolds were subjected to a daily stimulation over a period of two weeks, with one day in culture without stimulation after each week. The non-stimulation day was utilized for changing the culture media and quantifying the cell metabolic activity. At the end of the time period, samples were collected and processed to assess the role of F1 and F2 in steering the MSC fate towards the expression of TM-specific genes.

The selected set of frequency sweep was implemented during the final differentiation study, which investigated the effect of an integrated hierarchical, molecular, and acoustical stimulation. Following the monolayer formation, both plain and hierarchical TM scaffolds were cultured in the bioreactor for a period of 4 weeks, in accordance with the audio parameters described earlier. Ultimately, the stimulated constructs were characterized to verify the relevance of these approaches for MSC-driven tissue engineering of the human eardrum.

4.10. Characterization of stimulated TM scaffolds

Metabolic activity: The cell metabolic activity was investigated as the key indicator of MSC health and viability. This was performed using resazurin-based PrestoBlue™ reagent (Invitrogen™, Thermo Fisher Scientific, the Netherlands), diluted 1:10 times in M1 media. Samples were washed twice with DPBS and incubated with 500 µL of the diluted reagent at 37°C for 1 h. Thereafter, 100 µL of the reacted media was collected and fluorescence measurements were made on a multi-mode plate reader (CLARIOstar, BMG LABTECH, Germany) at an excitation wavelength of 535 nm and an emission wavelength of 615 nm. The values obtained from n = 15 measurements (5 biological replicates × 3 technical replicates) at each time point were normalized with respect to week 1.

Gene expression: Four genes were chosen to evaluate the relevance of MSCs for treating TM perforations: collagen type I (COL1A1), collagen type II (COL2A1), fibroblast growth factor 7 (FGF7) and fibroblast growth factor 10 (FGF10).^35,36 Cells seeded on scaffolds were washed twice with DPBS and subsequently lysed by TRIzol (Life technology, USA). A series of 3 freeze-thawing cycles was performed to aid lysing the cells and to release the biological content from the scaffolds. Total RNA was isolated based on phenol/chloroform method using GlycoBlue (Invitrogen™, Thermo Fisher Scientific, the Netherlands) as an RNA co-precipitant. RNA concentration and purity were determined spectrophotometrically using a BioDrop µLITE spectrophotometer (Biochrome, USA). Purity control standards were established for A260/A280 and A260/A230 ratios within the range of 1.8 to 2.2. First-strand cDNA was reverse-transcribed from total RNA by the use of iScript cDNA synthesis kit (Bio-Rad Laboratories, USA) according to the manufacturer's instructions. The expression of COL1A1, COL2A1, FGF7, and FGF10 genes was determined by means of real-time quantitative reverse transcription polymerase chain reaction (RT-PCR). Human amplification primers are listed in Table S1. SYBR Green master mix (Bio-Rad Laboratories, USA) was used, and real time PCR was carried out on a Bio-Rad CFX96 thermal cycler (Bio-Rad Laboratories, USA). Human beta-tubulin (TUBB2A) was selected as a housekeeping gene. Relative expression was calculated for n = 6 measurements (3 biological replicates × 2 technical replicates) using the 2^(−ΔΔCt) method relative to the control group.

Immunofluorescence imaging: The cell-seeded scaffolds after 1 day, 1 week, and 5 weeks in culture were washed twice with DPBS and fixed with 3.7% (v/v) formaldehyde solution (diluted in DPBS, ACS reagent, Sigma-Aldrich, the Netherlands) for 30 min at ambient temperature. To avoid background fluorescence signal during imaging, the fixed samples were incubated with 0.1% (w/v) of Sudan Black B dye (Sigma-Aldrich, the Netherlands) in 70% ethanol, for 45 min at ambient temperature. Thereafter, they were exposed to 0.1% Triton™ X-100 (diluted in DPBS, Sigma-Aldrich, the Netherlands) for 10 min, and subsequently treated with 1% bovine serum albumin (diluted in DPBS, VWR Chemicals, the Netherlands) for 1 h. The permeabilized and blocked cells were incubated with anti-collagen II primary antibody (1:200, ab34712, Abcam, UK) overnight at 4°C. Next day, the samples were washed three times with DPBS and stained with species-specific secondary antibody conjugated to fluorophore Alexa Fluor 647 (1:500, Thermo Fisher Scientific, A21245, the Netherlands) for 1 h, Alexa Fluor 488 phalloidin (1:100, Thermo Fisher Scientific, A12379, the Netherlands) for 45 min, and 4,6-diamideino-2-phenylindole dihydrochloride (DAPI; 1:1000, Sigma-Aldrich, D9542, the Netherlands) for 20 min. Finally, the stained scaffolds were imaged under an inverted microscope (Nikon Eclipse Ti-E, the Netherlands) to detect the deposition of extracellular collagen type II.

Protein secretion

The secretion of collagen type I and type II by the cultured MSCs was quantified using enzyme-linked immunosorbent assay (ELISA). After four weeks of integrated stimulation, culture media from all samples were collected and stored at -80°C for subsequent analysis. Protein quantification was performed on the collected supernatants using commercial DuoSet ELISA kits designed to detect human Pro-Collagen I alpha 1 and human Pro-Collagen II (R&D Systems, the Netherlands), following the manufacturer’s protocol. Absorbance was measured instantly at 450 nm and 540 nm using a multi-mode plate reader (CLARIOstar, BMG LABTECH, Germany). Each group was investigated with n = 6 samples for the measurements.

4.11. Statistical analysis

All the obtained values have been expressed as mean ± standard deviation. The samples were assigned randomly to different experimental groups, and the number of replicates (n) for each experiment has been specified in the figure captions. Statistical analysis was performed using GraphPad Prism 8 (GraphPad Software, USA), where the statistical significances were determined by applying a one-way or two-way analysis of variance (ANOVA) followed by a Tukey's honestly significant difference (HSD) post-hoc test (*p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.0001 and ns for p > 0.05).

Acknowledgments

This study was funded by the 4NanoEARDRM project, under the frame of EuroNanoMed III, an ERA-NET Cofund scheme of the Horizon 2020 Research and Innovation Framework Programme of the European Commission, and the Netherlands Organization for Scientific Research (NWO, grant agreement number OND1365231) and the Horizon 2020 cmRNAbone project (grant agreement number 874790). The authors thank Keyence Netherlands for their assistance with the vibrational characterization of the acoustic bioreactors.

Author contributions

Shivesh Anand: Conceptualization, Investigation, Validation, Formal analysis, Writing – Original draft preparation. Claudia Del Toro Runzer: Investigation, Validation, Formal analysis. Elizabeth Rosado Balmayor: Supervision, Funding acquisition, Writing – reviewing and editing. Martijn van Griensven: Supervision, Funding acquisition, Writing – reviewing and editing. Lorenzo Moroni: Conceptualization, Resources, Funding acquisition, Supervision, Writing – reviewing and editing. Carlos Mota: Conceptualization, Resources, Funding acquisition, Supervision, Writing – reviewing and editing.

Competing interests

The authors declare no competing interests.

Anand, S., Danti, S., Moroni, L. & Mota, C. Regenerative therapies for tympanic membrane. Progress in Materials Science, 100942 (2022).
Anand, S. et al. Mimicking the Human Tympanic Membrane: The Significance of Scaffold Geometry. Advanced Healthcare Materials 10, 2002082 (2021).
Mota, C. et al. Multiscale fabrication of biomimetic scaffolds for tympanic membrane tissue engineering. Biofabrication 7, 025005 (2015).
Schilder, A. G. et al. Otitis media. Nature reviews Disease primers 2, 1-18 (2016).
Blanshard, J. D., Robson, A. K., Smith, I. & Maw, A. R. A long term view of myringoplasty in children. J Laryngol Otol 104, 758-762 (1990).
Bianco, P. & Robey, P. G. Stem cells in tissue engineering. Nature 414, 118-121 (2001).
Kwon, S. G., Kwon, Y. W., Lee, T. W., Park, G. T. & Kim, J. H. Recent advances in stem cell therapeutics and tissue engineering strategies. Biomaterials Research 22, 1-8 (2018).
Pountos, I., Corscadden, D., Emery, P. & Giannoudis, P. V. Mesenchymal stem cell tissue engineering: techniques for isolation, expansion and application. Injury 38, S23-S33 (2007).
Maharajan, N., Cho, G. W. & Jang, C. H. Application of mesenchymal stem cell for tympanic membrane regeneration by tissue engineering approach. International journal of pediatric otorhinolaryngology 133, 109969 (2020).
Danti, S. et al. Tissue engineering of the tympanic membrane using electrospun PEOT/PBT copolymer scaffolds: A morphological in vitro study. Hearing, Balance and Communication 13, 133-147 (2015).
Danti, S. et al. Chitin Nanofibril Application in Tympanic Membrane Scaffolds to Modulate Inflammatory and Immune Response. Pharmaceutics 13, 1440 (2021).
Anand, S. et al. Chitin nanofibrils modulate mechanical response in tympanic membrane replacements. Carbohydrate Polymers, 120732 (2023).
Jang, C. H. et al. Mesenchymal stem cell-laden hybrid scaffold for regenerating subacute tympanic membrane perforation. Materials Science and Engineering: C 72, 456-463 (2017).
Moscato, S. et al. Tympanic Membrane Collagen Expression by Dynamically Cultured Human Mesenchymal Stromal Cell/Star-Branched Poly (ε-Caprolactone) Nonwoven Constructs. Applied Sciences 10, 3043 (2020).
Goncalves, S. et al. Histologic changes of mesenchymal stem cell repair of tympanic membrane perforation. Acta oto-laryngologica 137, 411-416 (2017).
Shahal, D., Goncalves, S. & Angeli, S. I. Mesenchymal Stem Cells for Treatment of Delayed-Healing Tympanic Membrane Perforations Using Hyaluronate-based Laminas as a Delivery System: An Animal Model With Histopathologic Study. Otology & neurotology 43, e497-e506 (2022).
Wang, S., Hashemi, S., Stratton, S. & Arinzeh, T. L. The effect of physical cues of biomaterial scaffolds on stem cell behavior. Advanced Healthcare Materials 10, 2001244 (2021).
Burdick, J. A. & Vunjak-Novakovic, G. Engineered microenvironments for controlled stem cell differentiation. Tissue Engineering Part A 15, 205-219 (2009).
Zhang, P. et al. The physical microenvironment of hematopoietic stem cells and its emerging roles in engineering applications. Stem Cell Research & Therapy 10, 1-13 (2019).
Danti, S., Berrettini, S., Marrazza, S. & Stefanini, C. (Google Patents, 2019).
Lee, H. J., Choi, B. H., Min, B. H., Son, Y. S. & Park, S. R. Low‐intensity ultrasound stimulation enhances chondrogenic differentiation in alginate culture of mesenchymal stem cells. Artificial organs 30, 707-715 (2006).
Chu, Y.-C., Lim, J., Hwang, W.-H., Lin, Y.-X. & Wang, J.-L. Piezoelectric stimulation by ultrasound facilitates chondrogenesis of mesenchymal stem cells. The Journal of the Acoustical Society of America 148, EL58-EL64 (2020).
Amini, A., Chien, S. & Bayat, M. Impact of ultrasound therapy on stem cell differentiation-a systematic review. Current stem cell research & therapy 15, 462-472 (2020).
Xia, P., Shi, Y., Wang, X. & Li, X. Advances in the application of low-intensity pulsed ultrasound to mesenchymal stem cells. Stem Cell Research & Therapy 13, 1-14 (2022).
Min, S. et al. Production enhancement of human adipose-derived mesenchymal stem cells by low-intensity ultrasound stimulation. Scientific Reports 12, 22041 (2022).
Fitzsimmons, R. E., Mazurek, M. S., Soos, A. & Simmons, C. A. Mesenchymal stromal/stem cells in regenerative medicine and tissue engineering. Stem cells international 2018 (2018).
Di Luca, A. et al. Tuning cell differentiation into a 3D scaffold presenting a pore shape gradient for osteochondral regeneration. Advanced healthcare materials 5, 1753-1763 (2016).
Decarli, M. C. et al. Bioprinting of stem cell spheroids followed by post‐printing chondrogenic differentiation for cartilage tissue engineering. Advanced Healthcare Materials, 2203021 (2023).
Zhang, Y., Marsboom, G., Toth, P. T. & Rehman, J. Mitochondrial respiration regulates adipogenic differentiation of human mesenchymal stem cells. PloS one 8, e77077 (2013).
Zhang, Y., Liang, X., Lian, Q. & Tse, H.-F. Perspective and challenges of mesenchymal stem cells for cardiovascular regeneration. Expert review of cardiovascular therapy 11, 505-517 (2013).
Hernández, R. et al. Differentiation of human mesenchymal stem cells towards neuronal lineage: clinical trials in nervous system disorders. Biomolecules & therapeutics 28, 34 (2020).
Tan, L., Liu, X., Dou, H. & Hou, Y. Characteristics and regulation of mesenchymal stem cell plasticity by the microenvironment—specific factors involved in the regulation of MSC plasticity. Genes & Diseases 9, 296-309 (2022).
Knutsson, J., Bagger-Sjöbäck, D. & von Unge, M. Collagen type distribution in the healthy human tympanic membrane. Otology & Neurotology 30, 1225-1229 (2009).
Stenfeldt, K., Johansson, C. & Hellström, S. The collagen structure of the tympanic membrane: collagen types I, II, and III in the healthy tympanic membrane, during healing of a perforation, and during infection. Archives of Otolaryngology–Head & Neck Surgery 132, 293-298 (2006).
Redmond, S. L., Levin, B., Heel, K. A., Atlas, M. D. & Marano, R. J. Phenotypic and genotypic profile of human tympanic membrane derived cultured cells. J Mol Histol 42, 15-25 (2011).
Santa Maria, P. L., Redmond, S. L., Atlas, M. D. & Ghassemifar, R. Keratinocyte growth factor 1, fibroblast growth factor 2 and 10 in the healing tympanic membrane following perforation in rats. Journal of molecular histology 42, 47-58 (2011).
Hassmann-Poznańska, E. et al. Analysis of gene expression profiles in tympanic membrane following perforation using PCR Array in rats—preliminary investigation. International Journal of Pediatric Otorhinolaryngology 77, 1753-1759 (2013).
Barlian, A. & Yanti, N. L. W. E. Glycosaminoglycans Content and Type II Collagen Localization in Chondrogenic Differentiation of Adipose-Derived Mesenchymal Stem Cells Induced by L-Ascorbic Acid 2-Phosphate. Journal of Mathematical & Fundamental Sciences 52 (2020).
Blaney Davidson, E. N., Vitters, E. L., van den Berg, W. B. & van der Kraan, P. M. TGF β-induced cartilage repair is maintained but fibrosis is blocked in the presence of Smad7. Arthritis research & therapy 8, 1-8 (2006).
Kim, K. K., Sheppard, D. & Chapman, H. A. TGF-β1 signaling and tissue fibrosis. Cold Spring Harbor perspectives in biology 10, a022293 (2018).
Karna, E., Szoka, L., Huynh, T. Y. L. & Palka, J. A. Proline-dependent regulation of collagen metabolism. Cellular and Molecular Life Sciences 77, 1911-1918 (2020).
Kay, E. J., Koulouras, G. & Zanivan, S. Regulation of extracellular matrix production in activated fibroblasts: roles of amino acid metabolism in collagen synthesis. Frontiers in oncology 11, 719922 (2021).
Beresford, J., Bennett, J., Devlin, C., Leboy, P. & Owen, M. Evidence for an inverse relationship between the differentiation of adipocytic and osteogenic cells in rat marrow stromal cell cultures. Journal of cell science 102, 341-351 (1992).
Sun, Y., Chen, C. S. & Fu, J. Forcing stem cells to behave: a biophysical perspective of the cellular microenvironment. Annual review of biophysics 41, 519-542 (2012).
Wan, X., Liu, Z. & Li, L. Manipulation of stem cells fates: the master and multifaceted roles of biophysical cues of biomaterials. Advanced Functional Materials 31, 2010626 (2021).
Sun, C. et al. Acoustically accelerated neural differentiation of human embryonic stem cells. Acta Biomaterialia 151, 333-345 (2022).
Jonnalagadda, U. S. et al. Acoustically modulated biomechanical stimulation for human cartilage tissue engineering. Lab on a Chip 18, 473-485 (2018).
Cho, H. et al. Sound affects the neuronal maturation of neuroblastoma cells and the repair of damaged tissues. Electronic Journal of Biotechnology 57, 1-11 (2022).
Schedin, O. Target Spectrums For Mastering: A comparison of spectral stylistic conventions between rock and vocal-based electronic music. Audio Technology thesis, Luleå University of Technology, (2021).
Wereski, M. The threshold of hearing. The STEAM Journal 2, 20 (2015).

There is NO Competing Interest.

SupplementaryInformationAnandetal.docx
Supplementary Information File

Download PDF

Version 1

posted

You are reading this latest preprint version

Molecular, hierarchical and acoustical cues promote mesenchymal stromal cell differentiation toward tympanic membrane regeneration

Status:

Version 1

Abstract

Figures

1. Introduction

2. Results

2.1. Biofabrication of TM scaffolds

2.2. Molecular stimulation

2.3. Acoustic bioreactors: design and characterization

3.4. Acoustical stimulation: validation and optimization

3.5. Integrated stimulation

3. Discussion

4. Methods

4.1. Biofabrication of TM scaffolds

4.2. Qualitative assessment of TM scaffolds

4.3. Cell seeding on TM scaffolds

4.4. Media compositions for molecular stimulation

4.5. Design and optimization of acoustic bioreactors

4.6. Production of acoustic bioreactors

4.7. Characterization of acoustic bioreactors

4.8. In vitro validation of acoustic bioreactors

4.9. Acoustical stimulation

4.10. Characterization of stimulated TM scaffolds

4.11. Statistical analysis

Declarations

Acknowledgments

Author contributions

Competing interests

References

Additional Declarations

Supplementary Files

Status:

Version 1