AK2 deficiency compromises the mitochondrial energy metabolism required for differentiation of human neutrophil and lymphoid lineages

Abstract

Reticular dysgenesis is a human severe combined immunodeficiency that is primarily characterized by profound neutropenia and lymphopenia. The condition is caused by mutations in the adenylate kinase 2 (AK2) gene, resulting in the loss of mitochondrial AK2 protein expression. AK2 regulates the homeostasis of mitochondrial adenine nucleotides (ADP, ATP and AMP) by catalyzing the transfer of high-energy phosphate. Our present results demonstrate that AK2-knocked-down progenitor cells have poor proliferative and survival capacities and are blocked in their differentiation toward lymphoid and granulocyte lineages. We also observed that AK2 deficiency impaired mitochondrial function in general and oxidative phosphorylation in particular - showing that AK2 is critical in the control of energy metabolism. Loss of AK2 disrupts this regulation and leads to a profound block in lymphoid and myeloid cell differentiation.

Figure 1

Impaired cell survival/proliferation and blockade of T-cell differentiation in AK2-deficient cells. (a) BM mononuclear cells from an RD patient (P3) were analyzed by flow cytometry, in order to evaluate the presence of the multilymphoid progenitor population (i.e., the CD10+CD24− population that accounts for 12% of the CD34+Lin− subset). (b) Two BM samples from RD patients (P4 and P6) and one sample from a healthy donor were sorted in order to purify CD34+ hematopoietic progenitors. To monitor T-cell differentiation, cells were seeded on OP9-hDelta1 cells. After 35 days of culture, we analyzed by flow cytometry for the presence of a CD4+CD8+ T-cell population. (c–g) CB CD34+ progenitors were transduced with shAK2 or shCont, sorted for GFP+ cells and plated on OP9-hDelta1 cells until D35. (c) The change over time in the percentage of GFP+ cells in shCont-transduced cells (black squares) or shAK2-transduced cells (black circles) (P=0.049, P=0.033, P=0.014 and P=0.012 at D14, D21, D28 and D35, respectively, n=4). (d) Mitochondrial membrane potential was analyzed in shCont-transduced cells (white bars) and shAK2-transduced cells (black bars) using the Mitoprobe DiIC(5) reagent 3 and then 7 days after the initiation of T-cell differentiation. The percentage of depolarization (corresponding to the proportion of viable 7AAD-negative cells that were negative for DiIC(5)) is shown (P=0.011 on D3 for shCont versus shAK2, n=6). (e) Cell proliferation was analyzed following incorporation of EdU 3 and then 7 days after the initiation of T-cell differentiation (P=0.0006 and P=0.008 on D3 and D7, respectively, n=6). (f) Flow cytometry analysis of GFP+ cells 28 days after the initiation of T-cell differentiation in the shCont and shAK2 conditions. (g) The proportion of DP CD4+CD8+ cells 28 days after initiation of T-cell differentiation in the shCont (white bars) and shAK2 conditions (black bars) (P=0.048); *P<0.05, **P<0.01, ***P<0.001

Figure 2

Impairment of survival and blockade of NK cell differentiation in AK2-deficient cells. CB CD34+ progenitors were transduced with shAK2 or shCont, sorted for GFP+ cells and seeded in NK cell differentiation medium for 21 days. (a) The change over time in the number of GFP+ cells during NK cell differentiation after transduction with shCont (black squares) or shAK2 (black circles). One of the three representative experiments is shown. (b) Mitochondrial depolarization was analyzed using the Mitoprobe DiIC(5) reagent 3 and then 7 days after initiation of NK cell differentiation in the shCont condition (white bars) and the shAK2 condition (black bars) (P=0.028 at D3, n=3). (c) Cell proliferation (incorporation of EdU) was analyzed 3 and 7 days after initiation of NK cell differentiation in the shCont condition (white bars) and the shAK2 condition (black bars) (n=3, P=not significant (NS)). (d) Flow cytometry analysis of GFP+ cells 21 days after initiation of NK cell differentiation in the shCont condition (dotted line) and the shAK2 condition (solid line). (e) The CD56+ cell number was evaluated 21 days after initiation of NK cell differentiation in the shCont (black squares) or shAK2 conditions (black circles) (P=0.014, n=3); *P<0.05

Figure 3

Impairment of granulocyte differentiation (but not monocyte differentiation) in AK2-deficient cells. CB CD34+ progenitors were transduced with shAK2 or shCont, sorted for GFP+ cells and seeded for 12 days in a methylcellulose culture (a) or for 14 days in a granulocyte culture with G-CSF or a monocyte culture with M-CSF (b–e). (a) The number of CFUs in a methylcellulose assay in the shCont condition (white bar) and shAK2 condition (black bar) (P=0.0062, n=6). The CFUs include both CFU-G and CFU-M colonies. (b) The total number of GFP+ cells in a G-CSF culture after transduction with shCont (white bar) or shAK2 (black bar) (left panel, P=0.026, n=3) and the proportion of EdU+ cells in shCont (black squares) or shAK2 (black circles) (right panel, P=0.0013, n=3). (c) The total number of CD15+CD11b+ granulocytes in a G-CSF culture after transduction with shCont (white bar) or shAK2 (black bar) (P=0.012, n=3). (d) The total number of GFP+ cells in an M-CSF culture after transduction with shCont (white bar) or shAK2 (black bar) (left panel) and the percentage of EdU+ cells in each condition (right panel, P=NS, n=3). (e) The total number of CD14+CD11b+ monocytes after an M-CSF culture after transduction with shCont (white bar) or shAK2 (black bar) (P=NS, n=3); *P<0.05, **P<0.01

Figure 4

Impaired survival and blockade of neutrophil cell differentiation in the AK2-deficient HL60 cell line. The promyelocytic cell line HL60 was transduced with shAK2 or shCont, selected with puromycin and seeded in culture with 10 μM ATRA for 4 days, in order to induce neutrophil differentiation. (a) The change over time in total cell count during neutrophil differentiation in shCont (black squares) or shAK2 (black circles) (P=0.013 on D3 and P=0.002 on D4, n=4). (b) The CD15+CD11b+ neutrophil count in differentiated HL60 cells transduced with shCont (white bar) or shAK2 (black bar) (P=0.033 on D2, P=0.008 on D3 and P=0.0003 on D4; n=4). (c) Mitochondrial depolarization was analyzed using Mitoprobe DiIC(5) reagent in HL60 cells transduced with shCont (white bar) and shAK2 (black bar) (n=4, P=NS); *P<0.05, **P<0.01, ***P<0.001

Figure 5

AK2 deficiency compromises energy metabolism during neutrophil differentiation. HL60 cells were transduced with shAK2 or shCont, selected with puromycin and seeded in culture with 10 μM ATRA for 4 days, in order to induce neutrophil differentiation. (a) Lactate and pyruvate levels in the cell culture supernatant of cells transduced with shCont (black squares) or shAK2 (black circles). There were significant differences in lactate (P= 0.0013 and 0.014 at D3 and D4, respectively) and pyruvate levels (P= 0.013 and 0.020 at D3 and D4, n=4) when comparing the two conditions. (b) Glucose consumption by cells transduced with shCont (black squares) or shAK2 (black circles) was calculated by measuring the difference in glucose concentration in the cell medium between the initial concentration and each time point. There were significant differences on D2, D3 and D4 (P=0.024, 0.0006 and 0.035, respectively, n=4). (c) Mitochondria function, as evidenced by the activities of CS, LDH and COX complex IV on D4 of culture in HL60 cells transduced with shCont (black squares) and shAK2 (black circles) (P=NS for CS and LDH, P=0.009 for COX and P=0.04 for COX/LDH, n=6); *P<0.05, **P<0.01, ***P<0.001

Figure 6

AK2 is required for mitochondrial respiration. HL60 cells were transduced with shAK2 or shCont, selected with puromycin and seeded in culture with 10 μM ATRA, to induce neutrophil differentiation. After 4 days of culture, oxygen consumption was compared in shCont (black square) and shAK2 (black circle) conditions. (a) Routine respiration (R, P=0.0007), (b) oligomycin-insensitive proton leak respiration (O, P=NS), (c) phosphorylating respiration (driving ATP synthesis) (R-O, P=0.0071) and (d) maximum electron transfer capacity (assessed using the uncoupler FCCP (P=0.0076)); *P<0.05, **P<0.01, ***P<0.001

Figure 7

Gene expression profiling of AK2-deficient cells, showing alterations in mitochondrial metabolism. (a) A heat map corresponding to the 770 genes deregulated in the shAK2 #1 and shAK2 #2 conditions, relative to the shCont condition. The hierarchical clustering of the four CB (CB1 to 4) replicates is shown. Significant differences are based on a 1.2-fold difference and a P-value<0.05. (b) A Venn diagram comparing upregulated (U) and downregulated (D) genes in the shAK2 #1 (blue) and shAK2 #2 (red) conditions, relative to the shCont condition. (c) A schematic model of AK2's function with respect to oxidative and glycolytic metabolism. Metabolites and enzymes measured in our assays in Figure 5 are framed. The respiratory chain protein genes found to be downregulated in shAK2 condition are shown in green