Identification of dormant label-retaining and nondormant HSPCs in situ by quantitative BM imaging. (A-C) Representation of the experimental approach for endogenous labeling of dormant LR HSPCs. Transgenic mice expressing tetracycline transactivator in HSPCs from the scl gene locus (A; SCL-tTA) and the histone 2B-GFP fusion protein (H2B-GFP) under the control of the tetracycline-responsive regulatory element (B; TRE; H2B-GFP) were crossed to confine GFP expression to the HSPCs (C, left; SCL-tTA; H2B-GFP). In the presence of DOX, H2B-GFP transcription was inhibited, resulting in dilution of GFP signal by proliferating cells. (D-G) Identification of BM LR and non-LR HSPCs in situ by quantitative imaging. (D) Frequency distribution of GFP levels (MFI) of cKit⁺GFP⁺ (see “Methods” for identification criteria) cells from imaging of SCL-tTA (110 cells), H2B-GFP control (67 cells), and DOX-treated SCL-tTA;H2B-GFP mice (96 cells). (E) Flow cytometry of SCL-tTA, H2B-GFP, and double transgenic SCL-tTA;H2B-GFP mice treated for 150 days with DOX revealed the high purity of HSC marker expression in cKit⁺GFP^high LR cells (n = 3). (F) Dot plot showing the absolute GFP MFI (MFI^GFP) of cKit⁺GFP⁺ single cells from control and treated mice shown in panel D (SCL-tTA: MGI^GFP = 4.8; H2B-GFP: MGI^GFP = 19.2; DOX-treated SCL-tTA;H2B-GFP: MGI^GFP = 35.99). Note the leakiness of GFP expression in the cKit⁺GFP⁺ BM cells of the H2B-GFP mouse. LR HSPC identification therefore required precise GFP quantification. LR HSPCs were identified as cKit⁺ cells with higher MFI^GFP than background levels in the single H2B-GFP mouse line (H2B-GFP: maximum MGI^GFP = 35.3). (G) High-resolution images of BM LR and non-LR HSPCs, illustrating the need for computational quantification. (H) Frequency distribution of LR and non-LR HSPCs after thresholding based on H2B-GFP background fluorescence from aged-matched animals.