Molecular implications of DKC1 inhibition in human hematopoietic stem and progenitor cells
Previous studies revealed that the knock-out of Dkc1 is embryonic lethal (14) and that only hypomorphic mutations have been found in X-DC patients (15, 16). In this study we aimed at generating X-DC-like hematopoietic stem and progenitor cells (HSPCs) based on the down-regulation of DKC1 with short hairpin RNA (shRNA) lentiviral vectors (LVs). shRNA-LVs carried a puromycin resistance gene to facilitate the selection of transduced HSPCs (see Materials and Methods).
The efficacy of seven different shRNA-LVs (Suppl. Table 1) to down-regulate the expression of DKC1 was screened in healthy donor CD34+ cells (Suppl. Figure 1A). In subsequent experiments, we showed that three of these shRNA-LVs, iDKC1, iDKC4 and iDKC7, significantly decreased DKC1 mRNA levels to 34-47% compared to levels determined in cells transduced with the scrambled shRNA LV (Figure 1A). Vector copy numbers (VCN) determined in these cells showed the presence of 1-8 copies per cell in all groups (Suppl. Figure 1B), revealing that inhibitory effects upon DKC1 were related to the interfering proviruses.
To investigate the molecular implications resulting from the inhibition of DKC1, we first evaluated the expression of TERC in CD34+ cells transduced with scrambled and DKC1-shRNA LVs. As shown in Figure 1B, TERC mRNA levels in cells transduced with iDKC1-, iDKC4- or iDKC7-LVs were respectively decreased to 27.7%±10.8%, 49.1%±18.6% and 19.8%±11.5%, compared to levels determined in the control group. In subsequent analyses, changes in the telomerase functionality of DKC1-interfered CD34+ cells were quantified. To this end, we measured telomerase activity of DKC1-interfered and control CD34+ cells by the TRAP assay. These results showed marked decreases in the telomerase activity of CD34+ cells that had been transduced with iDKC1-, iDKC4- or iDKC7-LVs, which showed values of 42.8%±19%, 61.5%±3.6% and 43.7%±15.6%, respectively, of values determined in the control group (Figures 1C and Figure 1D).
In the following experiments we investigated the implication of DKC1 interference in the DNA damage determined in CD34+ cells. Analyses of γH2AX foci in the nucleus of cells transduced with iDKC1-, iDKC4- or iDKC7-LVs revealed that only 19% of cells transduced with the scrambled shRNA LV showed more than 10 γH2AX foci per cell. However, an important increase in the proportion of CD34+ cells with γH2AX foci was observed in cells transduced with either the iDKC1- (76%), iDKC4- (42%) or the iDKC7- (61%) LVs (Figure 2A). In next studies, we determined the expression of phosphorylated p53 and p21 (CDKN1A) in CD34+ cells transduced with the different constructs. As shown in Figure 2B, phosphorylated p53 expression was higher in CD34+ cells transduced with the DKC1-shRNA LVs. When the expression of p21 was tested, iDKC1- and iDKC4-LVs enhanced its levels (2.7±0.7 and 2.4±0.26 fold, respectively) compared to the control group, though this was not observed in iDKC7-transduced cells (Figure 2C). Levels of caspase 3 and Annexin V+ cells were also increased in CD34+ cells transduced with either type of DKC1-shRNA LVs, although levels did not reach statistical significance (Figures 2B and 2D and Suppl. Figure 2). Taken together these results suggest the induction of DNA damage, cell senescence and apoptosis of DKC1-interfered HSPCs (Figure 2).
The interfered expression of DKC1 impairs the in vitro growth and ablates the in vivo repopulating ability of human HSPC
To determine whether the knock-down of DKC1 affects the functionality of human HSPCs, DKC1-interfered CD34+ cells were in vitro cultured for 10 days (see Materials and Methods) to evaluate implications in cell growth. In these studies, the portion of CD34+ cells at the end of the culture period was similar among the different experimental groups (Suppl. Figure 3). While transduced cells with the scrambled shRNA-LV showed a marked cell expansion during this period (117±87.31 fold compared to initial cell numbers), levels of expansion observed in iDKC1- and iDKC4-transduced CD34+ cells were only 13±6.99 and 15.3±2.42 fold compared to input cell numbers (Figure 3A). These values represent a significant decrease to 20±8% and 10±4%, respectively, of cell expansions corresponding to the control group (CD34+ cells transduced with the scrambled shRNA LV) (Figure 3B). As happened with p21 levels (Figure 2C), defects in cell proliferation were not observed with iDKC7-transduced cells (Figure 3B). In additional studies we evaluated changes in the telomere length in DKC1-interfered cells, although no differences were observed among the different experimental groups (Suppl. Figure 4). This suggests that much longer incubation periods would be required to observe a significant telomere shortening, although defects in the ability of DKC1-interfered cells to grow in culture limited the possibility of evaluating changes in the telomere length long-term after DKC1-interference.
When the clonogenic potential of DKC1-interfered cells was assessed, a significant reduction in the number of colonies generated by CD34+ cells transduced with any of the three anti-DKC1 LVs was observed (Figure 3C and Suppl. Fig 5). Again, reductions were more significant in cells transduced with the iDKC1- and iDKC4-LVs, which reduced the clonogenic potential to 20%±4% and 52%±5%, respectively, compared to the control group.
Based on the results obtained in cells transduced with iDKC1-LV, in a final set of experiments we assessed the repopulation potential of CD34+ cells transduced with this LV, and with a control LV (scrambled shRNA-LV). To this aim, 8x105 transduced cells, which contained an average number of 30,000 transduced CD34+ cells (Suppl. Figure 3), were transplanted into NSG mice. As shown in Figure 3D, CD34+ cells transduced with the control LV showed an evident in vivo repopulating ability (see orange dots in Figure 3D). In these animals the presence of human hematopoietic progenitors (CD34+), as well as of myeloid (CD33+) and lymphoid cells (CD19+) were observed (Suppl. Figure 6), confirming the multi-lineage repopulation ability of human HSPCs transduced with the scrambled shRNA LVs. In sharp contrast with these observations, 7 out of the 8 recipients that were transplanted with iDKC1-transduced CD34+ cells failed to repopulate recipient NSG mice (see light blue dots in Figure 3D). Interestingly, when VCNs were tested in the BM of mice engrafted with cells of the control group, the presence of integrated LV copies was observed in all cases (0.3 to 1 VCNs/cell; Suppl. Figure 7). However, no copies of the iDKC1 provirus were detected in BM cells from the animal engrafted with cells transduced with the DKC1-shRNA LV (Suppl. Figure 7). This reveals that this specific recipient was repopulated with cells that have survived the puromycin selection, although did not integrate in their genome the iDKC1-interfering provirus. As expected, the presence of the DKC1-shRNA provirus was neither observed in the non-engrafted NSG recipients (Suppl. Figure 7), since no human hematopoietic cells were observed in these recipients (Figure 3D).
Based on the hematopoietic studies conducted in these experiments we conclude that the inhibited expression of DKC1 impairs the in vitro growth properties and the in vivo repopulating ability of human HSPCs.