p53-mediated hematopoietic stem and progenitor cell competition

Abstract

Cell competition was originally described in Drosophila as a process for selection of the fittest cells. It is likely to play an important role in tissue homeostasis in all metazoans, but little is known about its role and regulation in mammals. By using genetic mosaic mouse models and bone marrow chimeras, we describe here a form of cell competition that selects for the least damaged cells. This competition is controlled by p53 but is distinct from the classical p53-mediated DNA damage response: it persists for months, is specific to the hematopoietic stem and progenitor cells, and depends on the relative rather than absolute level of p53 in competing cells. The competition appears to be mediated by a non-cell-autonomous induction of growth arrest and senescence-related gene expression in outcompeted cells with higher p53 activity. p53-mediated cell competition of this type could potentially contribute to the clonal expansion of incipient cancer cells.

Figure 1. Differential levels of DNA damage trigger HSPC competition

A, model of stress-induced cell competition. Cells with intermediate level of stress (grey) have growth potential in non-competitive conditions, but are underrepresented in the tissue in presence of undamaged cells (white). Cells with high level of damage (black) may undergo senescence, apoptosis, or necrosis, regardless of the presence of competitors. B, design of the experiment shown in panels C, D and E. Total BM was isolated from CD45.2 WT or p53+/− mice 4 days after exposure to 1 Gy of IR, and either used as is or mixed with BM from untreated mice carrying the CD45.1 congenic marker roughly at a ratio of (9:1). 15×10⁶ cells were injected into lethally irradiated recipients, and the rest was stained with antibodies against CD45.1 and HSPC markers and analyzed by FACS to determine the initial contribution of HSPCs from the donors. Total HSPC numbers (C) and percent chimerism (D) were determined in the BM of the recipient mice 16 weeks later by staining for CD45.1 and HSPC markers. E, chimerism in the PB was measured every 2 weeks after transfer. Error bars represent s.e.m., n ≥ 4; * p ≤ 0.05. F, design of the experiment shown in panel G and Figure S1C-G. Untreated mdm2+/− or WT CD45.2 donor BM was used as is or mixed with BM from untreated WT CD45.1 mice, and BM transfer and analysis performed as in (B) 8 weeks after transfer. Error bars represent s.e.m., n ≥ 4; *, p≤0.05; **, p≤0.01; ***, p≤0.001. See also Figure S1.

Figure 2. Design and characterization of the mosaic mouse model to study cell competition

A, schematic of the inducible mp53 knock-in allele integrated into Rosa26 locus. B, Thymocyte apoptosis was assayed in mice of the indicated genotypes 20 hrs after exposure to 5 Gy by propidium iodide (PI) exclusion assay. The dye can only penetrate damaged cell membranes; percentage of PI+ (dead) cells is shown. The data is representative of two independent experiments, n =2 in each. C, mp53 cell population size does not change in steady-state conditions. Left panel, mosaicism in the PBL of the R26-mp53 CreER (solid lines) and R26-eYFP CreER (dotted lines) at 15 weeks (left Y-axis) versus HSPCs in the BM at 80 weeks (right Y-axis). Lines represent individual animals. Right panel, comparison of mosaicism levels in the PBLs at 5 weeks of age (prebleed) and in the HSPC gate at 8 to 20 weeks of age. Here and throughout the paper, fold change in GFP^high HSPCs is calculated as follows: the percent of GFP^high HSPCs (analysed at the end of the experiment) is divided by percent of GFP^high PBLs in the prebleed (analysed at 5 weeks of age) of each mouse. See also Figure S2.

Figure 3. Characterization of the p53-mediated cell competition

A, B, Long-term effects of p53-mediated competition. Mosaicism was induced in R26-mp53; CreER mice by tamoxifen injection. 5 days later, percentage GFP^high cells was determined in the PB, and mice were treated with 2.5 Gy. Percentage of GFP^high cells in PB subsets and in HSPCs in the BM was assayed 200 days later. C, D, E, Cell type specificity of p53-mediated competition. For (D), total BM from R26-mp53 Mx1Cre (please see Materials and Methods for details) mosaic mice was transferred into lethally irradiated recipients (C, upper schematic); for (E), total splenocytes from R26-mp53 Mx1Cre mosaic donors were transfered into Rag2 −/− recipients (C, lower schematic). After one month, half of the chimeras in each group was subjected to 2.5 Gy. Percentage of GFP^high cells was measured one month after the 2.5 Gy treatment in HSPC compartment of the BM chimeras (D) or in splenocyte subsets of the lymphocyte-only chimeras (E). F, effect of the IR dose on mp53 cell expansion. Percentage of GFP^high cells was determined one month after exposure of R26-mp53 Balancer mosaic mice to the indicated doses, and normalized to the prebleed. G, kinetics of the mp53 cell expansion. R26-mp53; CreER mosaic mice were analyzed at indicated time points after exposure to 2.5 Gy. H,I, IR-induced HSPC competition can occur after the acute DDR phase. R26-mp53; CreER mice were injected with tamoxifen either 2 days before (a) or 7 days after (b) 2.5 Gy exposure. Unirradiated control mice (c) were injected with tamoxifen at the same day as group (b). Shown is ratio of GFP^high HSPCs (gated on lin-c-Kit+ Sca-1+ CD48-CD150+ CD34− cells) at 8 weeks after IR to the percent of GFP^high cells in the PB at day 9 after IR. All data is representative of at least 2 independent experiments, each with n≥3. Error bars represent s.e.m. *, p≤0.05; **, p≤0.01; ***, p≤0.001. Experiments in F, G were done with both CreER and Balancer strains. See also Figure S3.

Figure 4. Stress-competition is not cell-autonomous

A, B, mp53 cell expansion is frequency-dependent. Percentage of GFP^high cells was determined in R26-mp53 Balancer mice one month after 2.5 Gy exposure. Right Y-axis: percent of GFP^high HSPCs (green squares). Left Y-axis: fold increase in GFP^high HSPCs (blue triangles) and in GFP^high lineage-positive cells of the BM (red circles). Each number on the X axis corresponds to percent GFP^high cells in the prebleed of an individual mouse. C, D, relative p53 status determines the outcome of competition. p53 +/+, p53 +/− or p53 −/− BM was mixed with mosaic R26- mp53;Mx1Cre BM. The mixtures were transfered into lethally irradiated WT recipients. 1 month after the transfer, percent of chimerism was determined in the PB (prebleed), and half of the mice were treated with 2.5 Gy. Percentage of GFP^high HSPCs was assayed 1 month after the IR treatment, and normalized to the prebleed. E, the experiment was done as in (C) except for p53 +/− CD45.1 donor BM was used instead of R26-mp53. F,G, endogenous p53 is required for the effects of mp53 on cell competition. Percentage of GFP^high HSPCs were determined in R26-mp53 CreER mosaic mice crossed onto p53 +/+, p53 +/−, or p53 −/− background, one month after exposure to 2.5 Gy, and normalized to prebleed. All data is representative of at least 2 independent experiments, each with n≥3. Error bars represent s.e.m. *, p≤0.05; **, p≤0.01, ***, p≤0.001.

Figure 5. Contribution of proliferation and apoptosis to p53-mediated competition

A, B, Protection from IR-induced apoptosis by Bcl2 does not inhibit p53-mediated competition. R26-mp53 CreER mice were crossed to H^k-Bcl-2 transgene (Bcl2) and compared to non-transgenic R26-mp53; CreER littermates (-). Mice were treated and analyzed as in Figure 4F. C, HSPC proliferation is insufficient to induce p53-mediated competition. Percentage of GFP^high HSPCs were assayed in R26-mp53; CreER mosaic mice at indicated times after treatment with 2.5 Gy IR or 3 mg 5-fluorouracil (5-FU), and normalized to prebleed. D, E, HSPCs were sorted from WT, R26-mp53Δ or R26-mp53 mosaic mice 3 weeks after 2.5 Gy exposure. mRNA was amplified and analysed by microarray (not shown) and qPCR (E). All data is representative of at least 2 independent experiments, each with n≥3. Error bars represent s.e.m. *, p≤0.05; **, p≤0.01; ***, p≤0.001. See also Figure S4.

Figure 6. Senescence is implicated in IR-induced competition

A, B, expression of senescence marker genes was measured by qPCR as in Figure 5D. C, D, E, WT or cdkn2a−/− BM was mixed with either p53 +/− BM (D) or R26-mp53 mosaic BM (E) and transfered into lethally irradiated WT recipients. Mice were treated and analyzed as in Figure 4C, D, E. All data is representative of 2 independent experiments, each with n≥3. Error bars represent s.e.m. *, p≤0.05; **, p≤0.01; ***, p≤0.001. See also Figure S5.

Figure 7. Two categories of p53 target genes and their roles in cell behavior

A, Adhesion and signal transduction-related genes differentially expressed in HSPCs under competitive versus non-competitive conditions were assayed as described in Figure 5D, with the addition of WT and mp53 HSPC sorted from un-irradiated mosaic R26-mp53 mice. All data is representative of 2 independent experiments, each with n≥3. Error bars represent s.e.m. *, p≤0.05; **, p≤0.01; ***, p≤0.001. B, p53 regulates expression of two classes of genes: Group A genes are targets of the canonical p53-DDR, which include DNA repair, cell cycle arrest and cell intrinsic apoptosis. This group of genes affects cell fates such as proliferation versus cell cycle arrest, survival versus apoptosis, based on absolute levels of p53 activity and regardless of the presence of competitor cells. Group B genes are involved in cell communication with the environment. Arrows indicate up- or down-regulation by p53; genes that were expressed in HSPCs in p53- and competition-dependent manner (in this study) are italicized. C, p53 can have two types of activities: when the level of damage or stress reaches a certain threshold, the cell can undergo p53 dependent apoptosis or cell cycle arrest regardless of the status of other cells in the population. This effect is mediated by p53 target genes from the group A. If the level of damage or stress is below the threshold, the cells with lower level of p53 activity would outcompete cells with higher level of p53 activity. These competitive interactions are likely mediated by p53 target genes from group B. See also Figure S6.