Sustained competitiveness of p18^-/- HSCs after multiple rounds of cBMT. BMNCs from p18^-/- or p18^-/-p21^-/- knockout mice were mixed with an equal number (2 × 10⁶/each) of competitor cells (BMNCs from 2-month-old wild-type mice) and injected into lethally irradiated (10 Gy) recipient mice. Ten months after primary cBMT, 2 × 10⁶ BMNCs from the recipients were rechallenged with an equal amount of the competitor cells and secondarily transplanted into lethally irradiated recipients. This procedure was sequentially repeated twice (donor cells were pooled from 3-5 animals, and there were 10 recipients/each group for each cBMT). The actual interval between each cBMT is shown (A). Some mice were kept longer than the interval. I°, II°, III°, and IV° indicate first, second, third, and fourth cBMTs, respectively. 1:1 indicates that equal amounts of test cells and competitor cells were cotransplanted into the recipients. M and F indicate male and female, respectively. Numbers under the arrows indicate the months after each cBMT, and numbers in the parentheses are the cellular age of original transplanted HSCs. Peripheral blood was collected from the recipients after long-term engraftment during each interval between sequential cBMTs, and PCR-based semiquantitative analysis as described in our previous study16 was performed to determine the contribution of p18^-/- or p18^-/-p21^-/- to the overall hematopoietic reconstitution. The spleen- or blood-nucleated cells from wild-type and mutant mice were mixed at different ratios to obtain a standardization curve for the semiquantitative analysis. The average levels of engraftment in blood from p18^-/- or p18^-/-p21^-/- origin 12 months after secondary and 6 months after tertiary cBMTs are shown (B). Error bars indicate SD. □ and ▪ indicate p18^-/- and p18^-/-p21^-/- transplantation groups, respectively. Each group includes 8 animals, and the difference between p18^-/- and p18^-/-p21^-/- is significant (P < .05) based on the Student t test. The representative results after tertiary cBMT are shown (C) along with the standardization generated simultaneously (bottom panel). Numbers above the gel images indicate individual animals, and the normalized percentages of test cells of the total in blood are shown below the images. As a control to confirm the accelerated exhaustion of hematopoietic repopulation in the absence of p21 as reported previously,10 the identical cBMT procedure (A) was used to test the repopulating ability of BMNCs from p21^-/- mice at the age of 12 months (D). Eight to 12 months after the first cBMT, there was virtually no detectable p21^-/- band in blood after 30 cycles of the PCR, and there were 2 blood samples at a barely detectable level (1%-3%) of the engraftment after 35 cycles of the PCR as representatively shown here. This experiment was intended to confirm a previous claim of HSC exhaustion due to p21 absence.10 It was not performed at the same time as other 2 groups but the competitor cells were from the same type of donor mice as described in “Materials and methods.” Therefore, the results between the p21^-/- group and other groups, though imperfect, are still comparable since all these different test cells were directly assessed with the same type of competitor cells in the irradiated hosts. The singular effect of age would not substantially contribute to the exhaustion based on earlier studies by others showing no disadvantage of aged BMNCs in the competitive repopulation model.4,19,20

In vitro growth of LKS cells. CD34^-LKS cells were isolated from the recipients of p21^-/- primary cBMT (n = 3) or quaternary p18^-/-p21^-/- cBMT (n = 4). CD34^-LKS cells (100) were sorted into 96-well plates and cultured in IMDM supplemented with 100 ng/mL SCF, 50 ng/mL Flt-3, and 25 ng/mL TPO. Cell numbers were counted at 3 time points during the culture. The growth curves were plotted based on the means ± SD of triplicate cultures. An aliquot of the differentiated cells were lysed for the semiquantitative PCR analysis as described in Figure 2. The inserts in the graph are the PCR results showing the altered genotypic representation (knockout vs wild-type). Based on the single-cell analysis (Table 1), the initial ratios of p21^-/- and p18^-/-p21^-/- in the starting CD34^-LKS cells were 63.3% and 78.9%, respectively. No differentiation block was observed in both groups based on the cell morphology under a microscope.

Continued presence of p18^-/-HSC progeny without apparent exhaustion during sBMT. BMNCs were collected from p18^-/- or p18^+/+ male mice at 2 months of age and transplanted into lethally irradiated (10 Gy) female recipients (2 × 10⁶ cells/recipient). After long-term engraftment was established in the primary recipients, the same dose of BMNCs were pooled from the primary recipients and retransplanted into lethally irradiated female recipients. The same procedure was repeated for additional 3 times (A). 0°, I°, II°, III°, and IV° indicate nonrecipient, first, second, third, and fourth sBMTs, respectively. M and F indicate male and female, respectively. Numbers under the arrows indicate the months after each sBMT, and numbers in the parentheses are the cellular age of original transplanted HSCs. In vitro hematopoietic activity was assayed with the CAFC assay and CFC assay as previously described.10 The frequencies of day-35 CAFCs (left) and CFCs (right) normalized to the nonrecipient wild-type control after different sBMTs are shown (B). ^*P < .05 (Student t test, n = 3). Error bars indicate SD. □ and ▪ indicate p18^+/+ and p18^-/- groups, respectively. To further assess the exhaustion of hematopoietic regeneration, peripheral blood was collected from the recipients 12 months after IV° sBMT and the PCR-based semiquantitative analysis for mouse Y chromosome–specific sequence (Sry) normalized to the housekeeping gene myogenin (Myo) was applied to calculate the contribution of the original male donor cells (C). The spleen- or blood-nucleated cells from male and female mice were mixed at different ratios to acquire the standardization for the semiquantitative analysis. According to the standardization generated simultaneously (bottom panel), the normalized percentages of donor cells in the blood are indicated under the PCR images (top, middle panels). Numbers above the gel images indicate individual p18^-/- or p18^+/+ transplant recipients, and engraftment level (%) in each recipient is indicated under the gel image.

Multilineage differentiation that favors myelopoiesis after an extended long-term engraftment. (A) The recipients were killed after tertiary cBMT and their BMNCs were collected and stained with lineage markers. Myeloid (Mac-1⁺), T (CD3⁺)–, and B (B220⁺)–cell lineages were sorted for semiquantitative PCR analysis as described in Figure 2C. The actual results are shown. Numbers above the gel images indicate individual animals, and the calculated percentages of donor cells are indicated below the PCR images. (B) Contribution of p18^-/- or p18^-/-p21^-/- donor cells to different lineages in tertiary recipients (6 months after transplantation). Error bars indicate SD. □ and ▪ indicate p18^-/- and p18^-/-p21^-/- transplantation groups, respectively. ^*P < .05 (n = 3, Student t test).