Altered niche occupancy of radioresistant MKs following sublethal radiation. (A) MK lineage injury kinetics following 4 Gy TBI. Circulating platelets (black) and marrow MKs (red) remain at steady-state levels for 3 days (one-way analysis of variance, P = .45), while upstream MKPs (blue) are rapidly lost. MKP (colony assay) and MK (imaging flow cytometry) numbers are normalized to per femur values and all compartments expressed as percent of unirradiated control. Mean absolute numbers for 0 Gy controls: 13 550 MKP/femur, 56 022 MK/femur, 484 × 10³ platelets/μL. (B) MFI/A of surface CXCR4 on primary MKs by imaging flow cytometry of flushed marrow samples prepared from unirradiated controls (black) or days 1 to 3 post-4 Gy TBI (green). MKs from irradiated mice display increased surface CXCR4 expression. (C) Representative images of femoral IHC for MECA32 (vasculature, green) with hindlimbs isolated from unirradiated mice (top panel) or day 2 post-4 Gy TBI (bottom panel). The marrow vasculature dilates after TBI. (D) Quantification of vascular area within MECA32+ marrow vessels by IHC. Vascular dilation occurs by day 1 after injury and remains constant from days 1 to 4 post-4 Gy TBI. (E) Quantification of Gp1Bβ+ MKs physically associated with MECA32+ vessels by femoral IHC prepared from unirradiated control mice (black) or days 1 to 4 post-4 Gy TBI (green). MK association with vasculature changes dynamically following radiation, increasing between days 0 and 1 and between days 2 and 3, but decreasing between days 3 and 4. (F) Quantification of Gp1Bβ+ MKs in the endosteal niche (within 100 μm of the endosteal surface within diaphysis) by IHC in unirradiated control mice (black) and days 1 to 4 post-4 Gy TBI (green). MKs increase in the endosteal niche between days 1 and 2, and decrease between days 2 and 3. Image processing and analysis is described in the “Immunohistochemistry” section. The vascular niche and endosteal niche measurements were not made in a mutually exclusive manner and there may be overlap in occupancy between these niches. Error bars represent standard error of the mean of ≥3 independent experiments (n = 3-12 total mice per group). Statistical analyses either comparing irradiated samples to unirradiated control (B and D) or indicated samples (E and F) were performed using a 2-tailed Student’s t test. Bar represents 100 μm (C). *P < .04; **P < .001.