Table 3. Drawbacks and advantages of model systems for mitochondrial screens^*.

	Suitability for high-throughput screening format	Conservation with human proteins	Multi-organ physiology readout	Feasibility of genome-wide screen	Examples of drug discovery
Target-based approach
Isolated target	++++	++++	NA	NA	ERR ligands137–141,144 PPAR ligands128–130 SIRT1 activators89,93,95–100,214 GPBAR1 agonists215–221 AMPK agonists222–223
Phenotype- based approach
Yeast	+++	++	+	++++	MDIVI-1 (REF. 186) Chlorhexidine229
Immortalized cell lines	+++	+++	+	++	M1 hydrazone185 Meclizine232 Podophyllotoxins233
Primary cells	−/+	++++	+	−/+	BRD6897 (REF. 225)
Caenorhabditis elegans	++	++	++++	+++

AMPK, AMP-activated protein kinase; ERR, oestrogen-related receptor; GPBAR1, G protein-coupled bile acid receptor 1; NA, not applicable; PPAR, peroxisome proliferator-activated receptor; SIRT1, sirtuin 1.

^*Although targeted approaches use isolated targets and directly assess the change in target activity, phenotypic screening approaches can use different models with variable degrees of complexity. Models are ranked by degree of complexity from top to bottom in this table. Several criteria can help to determine the choice of the model. For example, the worm C. elegans is less suitable for high-throughput screening than an assay using isolated proteins or immortalized cell lines, but it models multi-organ physiology and is amenable to high-content screens. The last column provides examples of compounds and drugs that have been discovered using the different models and that are discussed in this article. The ‘−’ and ‘+’ signs are relative comparisons of the different models for each criteria, ranging from not suitable (−/+) to highly suitable (++++).