Identification of <i>M</i>. <i>tuberculosis</i> mutants defective for intracellular growth using a high-content imaging screen.

<p>(A) Protocol for the high-content imaging assay for the transposon mutant library growth in macrophages. Wild-type <i>Mtb</i> at MOIs varying two-fold from 2:1 to 0.25:1 was included in each plate to allow both accurate normalization between plates and validation of a cutoff for hits based on control population behavior. Because the transposon mutant library does not express a fluorescent protein, <i>Mtb</i> was externally stained with auramine-rhodamine after fixation to permit visualization. (B) Correlation of PC1 score of technical replicates in the full screen. (C) Density function of PC1 values for mutants and controls. While hits were selected based on comparison with OD-matched mutants, the distribution of hit mutants falls almost entirely below the wild-type control distribution as well, confirming that hit mutants also behave differently than wild-type.