Where do dyes go inside living cells? Predicting uptake, intracellular localisation, and accumulation using QSAR models

Abstract

Uptake of dyes into living cells and organisms is of concern to several diverse groups of people. These include those not wishing dyes to enter cells (e.g. manufacturers and users of textile dyes, or laboratory workers using dyes as analytical reagents) and those requiring dye entry (e.g. biologists imaging cell contents, or clinicians using photoactive dyes as antitumour drugs). This diversity results in the need to consider an extremely wide range of dyes -and indeed of cells and organisms. An overview of methods for predicting uptake and intracellular localisation is provided, followed by a more detailed account of the concepts and procedures involved in decision-rule quantitative structure-activity relationship (QSAR) models. Some of these models permit the prediction of which dyes are likely to enter cells, and which dyes will be excluded. Other models predict where internalised dyes will localise within the live cells. Use of QSAR models to understand intracellular accumulation, redistribution, loss from the cell, and metabolic modification of dyes is also considered. In particular, the relationship of such predictions to toxicity is discussed. An extended case example is provided, describing the modelling of dye binding to nucleic acids in single-cell systems. A further case example then illustrates dye localisation in multicellular organisms. Finally, conclusions, critiques, and probable future directions concerning the QSAR modelling approach to dye uptake and localisation are given. A summary of key QSAR decision rules in the form of decision logic tabulations is provided.

Lipid analogs reveal features critical for hemolysis and diminish granadaene mediated Group B Streptococcus infection

“…The ability of the polyene moiety of pP7X, pP7, P9, and pP9 to span the plasma membrane partially, but not entirely, could explain the observed attenuated hemolytic activity relative to granadaene. Additionally, calculated partition coefficients for more active lipids are in agreement with a dynamic and favorable lipophilic interaction with lipidic cellular domains 38 .…”

Engineering Mitochondriotropic Carbon Dots for Targeting Cancer Cells

“…Mathematical models and computational approaches that describe and rationalize the transport and distribution at cellular and subcellular levels of small molecules predict that polar weak bases preferably accumulate in lysosomes [ 77 ]. Accumulation in mitochondria of cationic compounds will only be favored if the permeability of their ionic and neutral forms is roughly equal—as is the case for delocalized cations such as TPP—and also if they are only moderately lipophilic [ 77 , 78 ]. It is, therefore, clear that mitochondrial localization is affected not only by the presence of delocalized lipophilic cations but also by the overall basicity and lipophilicity of the compounds.…”

Reactive dyes for living cells: Applications, artefacts, and some comparisons with textile dyeing

“…9 Accounts of the wider uses of fluorescent probes are of course available, including one in this journal. 10…”

“…structure activity (QSAR) models. 10 Note that some probes mimic metabolites, and so their localisation is also influenced by transporters and selective pores in cell membranes, but such complications will not be discussed here.…”

“…AI and log P are, respectively, the electric charge, the amphiphilicity index (a measure of the size of the lipophilic domain) and the log of the octanol-water partition coefficient (a measure of overall lipophilicity). The quantitative structure activity (QSAR) models providing the predictions of accumulation sites are summarised in Horobin 10. How differences in physicochemical properties influence cellular localisation of MTT Tetrazolium and MTT Formazan, the reactant and product species of a reactive probe Outline of the procedure used to evaluate the likelihood of reactant/product localisation artefacts.Step 1 generates localisation predictions for reactant and product species (see green boxes), and step 2 assesses the significance of the predictions (see orange boxes).…”