Amino Acid Remodeling Precedes Malignant Transformation in Oral Epithelial Cells

Abstract

Background The accessibility of the oral cavity and the presence of clinically identifiable oral potentially malignant disorders provide an important opportunity for oral cancer prevention. However, current clinical and histopathological assessments cannot reliably predict which lesions will undergo malignant transformation. Here, we investigated amino acid alterations and related proteomic remodeling that characterize potentially malignant oral epithelial cells relative to healthy and malignant cells. Methods Healthy, potentially malignant, and malignant oral epithelial cell models were analyzed using Raman spectroscopy and liquid chromatography-mass spectrometry (LC-MS)-based proteomics. Raman spectral profiles were compared across the three states of epithelial transformation, focusing on aromatic amino acid-associated spectral regions. Proteomic changes were evaluated using differential abundance and pathway analyses, and the potentially malignant remodeling patterns were integrated with Raman findings. Results Raman spectroscopy identified distinct biochemical profiles across the three cellular states, with prominent alterations in the 1000-1165 cm-1 region. Phenylalanine-associated features near 1001-1004 cm-1 were increased in the potentially malignant state and subsequently decreased in malignant cells. In contrast, tyrosine-associated bands near 1164 cm-1 were reduced in potentially malignant cells and increased in malignant cells. Proteomic analysis identified extensive stage-associated remodeling of amino acid metabolism, transport, and protein synthesis. Early alterations included changes in GOT1, GOT2, GPT2, IVD, HIBCH, DBT, DLD, ALDH6A1, ALDH4A1, SHMT1, CTH, and other amino acid-associated proteins. Integration of the datasets identified convergence between aromatic amino acid-associated Raman features and early proteomic remodeling. Conclusions Potentially malignant oral epithelial cells exhibited distinct Raman and proteomic signatures that differentiated them from both healthy and malignant cells. The correspondence between aromatic amino acid-associated Raman features and changes in proteins involved in amino acid metabolism and transport supports the presence of a distinct biochemical state associated with potential malignancy. These findings identify candidate molecular features for further validation in patient-derived lesions and future studies of malignant transformation risk.