Non-Invasive classification of macrophage polarisation by 2P-FLIM and machine learning

Abstract

In this study, fluorescence lifetime imaging of NAD(P)H-based cellular autofluorescence is applied as a non-invasive modality to classify two contrasting states of human macrophages by proxy of their governing metabolic state. Macrophages were obtained from human blood-circulating monocytes, polarised using established treatments, and metabolically challenged using small molecules to validate their responding metabolic actions in extracellular acidification and oxygen consumption. Fluorescence lifetime imaging microscopy (FLIM) quantified variations in NAD(P)H-derived fluorescent lifetimes in large field-of-view images of individual polarised macrophages also challenged, in real-time with small molecule perturbations of metabolism during imaging. We uncover FLIM parameters that are pronounced under the action of carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) which strongly stratifies the phenotype of polarised human macrophages. This stratification and parameters emanating from a FLIM approach, served as the basis for machine learning models. Applying a random forest model, identified three strongly governing FLIM parameters, achieving a ROC AUC value of 0.944 when classifying human macrophages.