Characterizing DPPM inhibition of butyrylcholinesterase: integrated enzymatic kinetics and Raman spectroscopy with chemometric analysis

Butyrylcholinesterase (BChE) may serve as a scavenger enzyme protecting against various toxic compounds, but we still don't fully understand how it interacts with fentanyl analogues. We investigated how Despropionyl <i>meta</i>-methyl fentanyl (DPPM) inhibits equine BChE by combining traditional enzyme kinetics with Raman spectroscopy and machine learning approaches. We measured enzyme activity across different substrate and inhibitor concentrations, then fit the data to four different inhibition models using nonlinear regression. Statistical comparison using Akaike Information Criterion clearly showed that mixed inhibition best explained our results (AIC = 269.62), with DPPM binding more strongly to the free enzyme (<i>K</i>_ic = 528.7 μM) than to the enzyme-substrate complex (<i>K</i>_iu = 1471.0 μM). Raman spectroscopy revealed structural changes when the enzyme was inhibited, and we used principal component analysis to separate the enzyme's spectral signature from background interference. Three machine learning algorithms - artificial neural networks, random forest, and support vector machines - could distinguish between active and inhibited enzyme with 92% accuracy using leave-one-out cross-validation. The spectral features that worked best for classification included changes at 1028 cm^-1 (phenylalanine), 1539 and 1557 cm^-1 (protein backbone), and lower frequencies (454-875 cm^-1) associated with larger-scale protein movements. These results provide direct molecular evidence for the mixed inhibition mechanism we found through kinetics. Our work shows how combining different analytical techniques can reveal details about enzyme inhibition. The findings also have practical implications for understanding toxicity of fentanyl and its different analogues and could help develop better detection methods for these dangerous compounds.