Presentation Information
[2ACCE-06-TA]Peptide-functionalized surfaces for the regulation of myofibroblast differentiation
○Takumi Taga1, Shunsuke Ito1, Kodai Watanabe2, Tatsuo Takagi1, Ayato Sugiyama1, Akiyo Fujimoto1, Kenjiro Tanaka1, Hiroyuki Imanaka2, Ryuji Kato1,3 (1. Graduate School of Pharmaceutical Sciences, Nagoya University (Japan), 2. Graduate School of Environmental, Life, Natural Science and Technology, Okayama University (Japan), 3. Institute of Nano-Life-Systems, Institutes of Innovation for Future Society, Nagoya University (Japan))
Keywords:
peptide functionalization,peptide,myofibroblast,machine learning,fibrosis
[Purpose]
In tissue regeneration and wound healing, fibroblasts play a key role in restoring normal tissue architecture through the production and degradation of extracellular matrix (ECM). In contrast, myofibroblasts, which differentiate from fibroblasts, contribute to fibrosis through excessive ECM deposition. Recent studies indicate that myofibroblast differentiation is strongly influenced by interactions with the surrounding microenvironment, and understanding these interactions may contribute to the development of antifibrotic drugs and the design of medical materials that prevent tissue adhesion.
In vivo, cells receive various signals from the ECM microenvironment, which regulate cell adhesion, proliferation, and differentiation. The ECM possesses not only biological but also physicochemical properties, such as stiffness, charge, and hydrophobicity, which critically influence cell behavior. Our group has focused on short peptides and developed peptide-functionalized surfaces that mimic both the biological and physicochemical properties of the ECM. In this study, we induced myofibroblast differentiation on a variety of peptide-functionalized surfaces and used machine learning to identify the surface-property rules underlying this differentiation process.
[Methods]
We screened a library of 70 tripeptides for their effects on fibroblast-to-myofibroblast differentiation. The sequences were selected through an in silico clustering analysis of the physicochemical properties of 8,000 possible tripeptides. The peptides were then immobilized onto polystyrene surfaces using an L-3,4-dihydroxyphenylalanine (L-DOPA)-based immobilization method. Myofibroblast differentiation was induced on the 70 peptide-functionalized surfaces, and the expression of the differentiation marker α-smooth muscle actin (αSMA) was quantified. A machine learning model was then constructed using the physicochemical properties of the peptides as input variables and αSMA expression as the output variable. For precise control of peptide surface density, we employed the peptide-presenting protein CutA1.
[Results]
Several peptide sequences that suppressed myofibroblast differentiation were identified. A machine learning-based regression model was developed with a coefficient of determination (R²) of 0.726. The model suggested that the overall polarity of the tripeptide and the hydrophobicity of the N-terminal residue are key factors influencing myofibroblast differentiation. Furthermore, the peptide-presenting protein CutA1 enabled fine control of myofibroblast differentiation.
[Conclusion]
Peptide-functionalized surfaces and machine learning enabled the identification of surface-property rules that regulate myofibroblast differentiation, providing a basis for the efficient design of antifibrotic biomaterials.
In tissue regeneration and wound healing, fibroblasts play a key role in restoring normal tissue architecture through the production and degradation of extracellular matrix (ECM). In contrast, myofibroblasts, which differentiate from fibroblasts, contribute to fibrosis through excessive ECM deposition. Recent studies indicate that myofibroblast differentiation is strongly influenced by interactions with the surrounding microenvironment, and understanding these interactions may contribute to the development of antifibrotic drugs and the design of medical materials that prevent tissue adhesion.
In vivo, cells receive various signals from the ECM microenvironment, which regulate cell adhesion, proliferation, and differentiation. The ECM possesses not only biological but also physicochemical properties, such as stiffness, charge, and hydrophobicity, which critically influence cell behavior. Our group has focused on short peptides and developed peptide-functionalized surfaces that mimic both the biological and physicochemical properties of the ECM. In this study, we induced myofibroblast differentiation on a variety of peptide-functionalized surfaces and used machine learning to identify the surface-property rules underlying this differentiation process.
[Methods]
We screened a library of 70 tripeptides for their effects on fibroblast-to-myofibroblast differentiation. The sequences were selected through an in silico clustering analysis of the physicochemical properties of 8,000 possible tripeptides. The peptides were then immobilized onto polystyrene surfaces using an L-3,4-dihydroxyphenylalanine (L-DOPA)-based immobilization method. Myofibroblast differentiation was induced on the 70 peptide-functionalized surfaces, and the expression of the differentiation marker α-smooth muscle actin (αSMA) was quantified. A machine learning model was then constructed using the physicochemical properties of the peptides as input variables and αSMA expression as the output variable. For precise control of peptide surface density, we employed the peptide-presenting protein CutA1.
[Results]
Several peptide sequences that suppressed myofibroblast differentiation were identified. A machine learning-based regression model was developed with a coefficient of determination (R²) of 0.726. The model suggested that the overall polarity of the tripeptide and the hydrophobicity of the N-terminal residue are key factors influencing myofibroblast differentiation. Furthermore, the peptide-presenting protein CutA1 enabled fine control of myofibroblast differentiation.
[Conclusion]
Peptide-functionalized surfaces and machine learning enabled the identification of surface-property rules that regulate myofibroblast differentiation, providing a basis for the efficient design of antifibrotic biomaterials.
Comment
To browse or post comments, you must log in.Log in