Mechanical and Biochemical Stability of Hyaluronic Acid-Gelatin Hydrogels for Use in a Phono-Mimetic Vocal Fold Bioreactor

Latifi, Neda (Department of Mechanical Engineering, McGill University)
Heris, Hossein K. (Department of Mechanical Engineering, McGill University)
Boucher, Eric (Department of Anatomy and Cell Biology, McGill University)
Li, Nicole Y. K. (School of Communication Sciences and Disorders, McGill University)
Vali, Hojatollah (Department of Anatomy and Cell Biology, McGill University)
Mongeau, Luc (Department of Mechanical Engineering, McGill University)


Human phonation involves self-oscillations of the vocal folds within the larynx. An airflow-induced perfusion bioreactor was developed to mimic the biomechanical environment of the human vocal fold (VF) lamina propria. The bioreactor consisted of two synthetic VF replicas mounted within a custom-built case with continuous perfusion of cell culture medium (CCM) supply. The aim of the current study was to investigate the stability (biochemical (degradation) and mechanical (fatigue)) of a scaffold consisting hyaluronan (HA), gelatin (Ge), and polyethylene glycol (PEG) crosslinker for use in the perfusion VF bioreactor. The effects of the composition of CCM on the biochemical stability of the HA-Ge hydrogels were also studied.

Materials and Methods

Bioreactor Fabrication: The bioreactor is shown in Fig. 1. The VF replicas were fabricated using Ecoflex 10 silicon rubber. A cylindrical inner cavity was made to host the cell-scaffold mixture. The bioreactor case was fabricated using Ecoflex 30. Cell Culture in Flasks: Immortalized human VF fibroblasts (I-HVFFs) were grown in DMEM (1X) with 10% FBS, 1% Penicillin/Streptomycin, and 1% non-essential amino acids at 37ºC in 5% CO2 humidified atmosphere. Scaffold Fabrication: A cell-scaffold mixture consisting I-HVFFs solution, HA, Ge, and PEG crosslinker was prepared. Five different concentrations of Ge (0.25%, 0.1%, 0.05%, 0.025%, or 0%) and two different PEG crosslinkers (0.5% PEG di-acrylate (PEGDA) or 0.5% PEG tetra-acrylate (PEGTA)) were used. Biochemical Stability test: Scaffolds with different concentrations of the constituents were submerged in CCM with different FBS concentrations (2%, 5%, and 10%) in static condition at 37ºC. Another set of samples was submerged in distilled water to study hydrolysis. Gelatin zymography was used to assess the presence of matrix metalloproteases in our culture system. Rheometry: A TA instrument rheometer (AR 2000) was used to measure the storage and loss moduli of the hydrogels over the frequency range between 0.1 and 10 Hz. Mechanical Stability (Fatigue) Test: The scaffold was injected into the cavity of the bioreactor and phonated continuously for 72 hours. Subsequently, the gel was harvested and its mechanical integrity was evaluated. Phonation Stimulation: A cell-scaffold mixture consisting of optimized HA-Ge scaffold was injected into the VF replicas. Optimized CCM with 5% CO2 was perfused through the cell-scaffold mixture. Replicas were phonated using a variable speed centrifugal blower targeting an operating frequency of around 100 Hz for two hours a day over a period of 7 days. A similar bioreactor without mechanical excitation was used as negative control. The cell-scaffold mixture was harvested for cell viability and collagen type I immunohistochemical staining 7 days after injection.


The storage and loss moduli of HA-Ge-PEGDA and HA-Ge-PEGTA are shown in table 1. PEGTA yielded a stiffer hydrogel than PEGDA. Also, HA-Ge-PEGDA scaffolds broke down into several pieces following continuous phonation. But, HA-Ge-PEGTA resisted mechanical impact stresses. HA-Ge-PEGTA scaffolds submerged in water showed no change in shape or size after 7 days. The degradation rates of HA-Ge-PEGTA scaffolds submerged in CCM were different. The gels with 0.25% or 0.1% Ge were mostly degraded after 7 days. The gels with 0.05%, 0.025%, and 0% Ge were stable for at least 14 days. The concentration of FBS was found to affect scaffold degradation. Gelatin zymography experiments were consistent with those observations. The concentration of FBS was decreased to 5%. At this concentration, CCM supports cell proliferation and viability without adverse degradation of the HA-Ge-PEGTA scaffolds. The optimized hydrogel was injected into the VF bioreactor. The viability of the cells was over 90% after 7 days culture in the bioreactor, and collagen type I was observed in the cell periphery (See Fig. 2).

Discussion and Conclusion

Polymeric materials may fail under dynamic loading. Vocal folds undergo impact, shear, tensile and compression stresses during phonation. Scaffolds consisting of PEGTA were mechanically stable under such loading. PEGTA has four-arm for linking HA and Ge molecules, which yields in an increase in the strength of the HA-Ge scaffolds and their capacity for absorbing energy during cyclic loading. Although the biochemical stability of the HA-Ge hydrogels is affected by the crosslinker type and concentration, the presence of Ge in the structure of the scaffold makes it more prone to enzymatic degradation. Indeed presence of MMP2, both in the CCM and produced by the cells would have cleaved Ge molecules and contributed to the loss of the scaffold matrix. Decreasing Ge concentration improved the biochemical stability of the scaffold. However, concentration of Ge below 0.1% decreases the cell adhesion property of the scaffold. At this concentration, the HA-Ge-PEGTA matrix was stable over two-week period. Also, decreasing the concentration of FBS to 5% reduced the amount of detectable MMP2 which further improved the overall stability of the HA-Ge-PEGTA scaffold. This optimized scaffold and CCM supported cell viability and extracellular matrix protein synthesis in the VF bioreactor. Future studies will extend the culture time in order to encourage production of a self-generated extra-cellular matrix by the I-HVFFs and to study the effect of mechanotransduction on those parameters.

Fig. 1. The airflow-induced perfusion bioreactor. Synthetic vocal fold replicas were mounted into a custom-built case (A and B). The bioreactor was phonated using a centrifugal air blower (C).

Table 1. The storage and loss moduli of HA-Ge hydrogels containing 0.5% HA, 0.1% Ge, and 0.5% PEGDA or PEGTA.

Fig. 2. Collagen type I synthesis inside the HA-Ge-PEGTA hydrogel, (A) Zero time point sample, (B) negative control, and (C) phonated sample seven days after injection. Collagen and cell nucleus are shown in green and blue, respectively. Scale bar: 15 microns.


This work was supported by grant NIDCD R01-DC005788 (Mongeau, PI) from the National Institutes of Health. Neda Latifi and Hossein K. Heris are Co-first authors.


Latifi, N., H. K. Heris, S. Kazemirad and L. Mongeau (Accepted). "Development of a self-oscillating mechanical model to investigate the biological response of human vocal fold fibroblasts to phono-mimetic stimulation." In Proceedings of the ASME 2014 International Mechanical Engineering Congress & Exposition, IMECE2014. Li, N., H. Heris and H. Mongeau (2013). "Current understanding and future directions for vocal Fold mechanobiology." Journal of Cytology &Molecular Biology 1(1): 1-9.

Copyright ©1990 - 2019
Web Development by Inc.

Close Drag