Abstract
Compared to cultured 2D cell monolayers, 3D multicellular spheroids are more realistic tumor models. Nonetheless, spheroids remain under-utilized in preclinical research, in part, because there is a lack of fluorescence sensors that can noninvasively interrogate all the individual cells within a spheroid. This present study describes a deep-red fluorogenic molecular probe for microscopic imaging of cells that contain a high level of nitroreductase enzyme activity as a biomarker of cell hypoxia. A first-generation version of the probe produced "turn-on" fluorescence in a 2D cell monolayer under hypoxic conditions; however, it was not useful in a 3D multicellular tumor spheroid because it only accumulated in the peripheral cells. To guide the probe structural optimization process, an intuitive theoretical membrane partition model was conceived to predict how a dosed probe will distribute within a 3D spheroid. The model identifies three limiting molecular diffusion pathways that are determined by a probe's membrane partition properties. A lipophilic probe with high membrane affinity rapidly becomes trapped in the membranes of the peripheral cells. In contrast, a very hydrophilic probe molecule with negligible membrane affinity diffuses rapidly through the spheroid intercellular space and rarely enters the cells. However, a probe molecule with intermediate membrane affinity undergoes sequential diffusion in and out of cells and distributes to all the cells within a spheroid. Using the model as a predictive tool, a second-generation fluorescent probe was prepared with a smaller and more hydrophilic molecular structure, and optical sectioning using structured illumination or light sheet microscopy revealed roughly even probe diffusion throughout a tumor spheroid. The membrane permeation model is likely to be broadly applicable for the structural optimization of various classes of molecules and nanoparticles to enable even distribution within a tumor spheroid.