Abstract
Normal erythroid maturation is highly regulated to maximize the production of HbA heterotetramer (α2β2) and minimize the accumulation of potentially toxic free α- or β-globin subunits. In β-thalassemia, β-globin gene (HBB) mutations cause buildup of free α-globin, which forms intracellular precipitates that impair erythroid cell maturation and viability. Endogenous mechanisms exist to detoxify free α-globin, as evidenced by clinical observations that most individuals with HBB haploinsufficiency (β-thalassemia trait) are asymptomatic. Previously, we showed that free α-globin is eliminated in erythroid precursors by the ubiquitin-proteasome (UPS) system and by a lysosomal-dependent process presumed to be autophagy (Khandros et al., Blood 2012;119:5265). Our current study investigates the latter.
Atg5 is required for basal autophagy and autophagosome formation via conjugation with Atg12 in most cell types. We investigated whether Atg5 played a selective role in eliminating free α-globin during unbalanced globin chain synthesis. β-thalassemic mice (HbbTh3/+) carrying a conditional Atg5 allele (Atg5fl) in which exon 3 is flanked by loxP sites were interbred with mice expressing erythroid-specific Cre recombinase (EpoR-Cre). Loss of Atg5 in β-thalassemic erythroid precursors inhibited autophagy, as evidenced by the accumulation of the markers p62/SQSTM1 and LC3. However, there was minimal effect on the β-thalassemic phenotype as measured by RBC count, Hb, hematocrit, reticulocyte count, and spleen weight. Moreover, loss of Atg5 did not alter the level of insoluble α-globin in β-thalassemic RBC fractions, as assessed by triton-acetic acid-urea (TAU) gel electrophoresis to resolve the α- and β-globin chains. Thus, Atg5, a key component of canonical autophagy, is not required for the elimination of free α-globin in β-thalassemia.
The Ulk1 (unc-51−like kinase 1) protein kinase participates in canonical and alternative autophagy pathways and mediates mitophagy during erythropoiesis. To determine the role of Ulk1 in a-globin degradation, we introduced a null allele into HbbTh3/+ mice. Ulk1−/− HbbTh3/+ mice were born at normal Mendelian ratio, but exhibited a high rate of perinatal death. Only 20% of double-mutant mice survived for 30 days (n=5), compared to 70% of Ulk1+/+ HbbTh3/+ mice (n=13) (P <0.05). To examine the consequences of Ulk1 loss in β-thalassemic erythroblasts, we transplanted whole bone marrow cells from double mutants and controls (CD45.2) into lethally irradiated wild-type hosts (CD45.1; >87% engraftment after 30 days). Loss of Ulk1 exacerbated the β-thalassemic phenotype, as evidence by reduced Hb (10.85 ± 0.18 vs 10.21 ± 0.18 g/dL, n=14 vs 8; P <0.01), reduced RBC number (7.37 ± 0.13 vs 6.35 ± 0.09 × 106/mL; P <0.0001), increased reticulocyte count (17.99 ± 2.76 vs 39.41 ± 2.14%; P <0.0001), and increased spleen weight (0.24 ± 0.01 vs 0.34 ± 0.03 g; P <0.01). Moreover, insoluble α-globin was increased by approximately 2-fold in Ulk1−/− HbbTh3/+ reticulocytes, ascompared to Ulk1+/+ HbbTh3/+ cells (Figure). To assess the turnover of newly synthesized free α-globin, we pulse-labeled reticulocytes with 35S amino acids and chased with or without a proteasome inhibitor (MG132) or a lysosome inhibitor (chloroquine or bafilomycin A1). Analysis of control (Ulk1+/+ HbbTh3/+) reticulocytesshowed that newly synthesized insoluble free α-globin was cleared by separate proteasomal and lysosomal pathways. In marked contrast, the lysosomal-dependent α-globin degradation pathway was completely eliminated in Ulk1−/− HbbTh3/+ reticulocytes. Thus, we conclude that Ulk1 mediates the degradation of free α-globin in β-thalassemia.
Our findings illustrate a new mechanism through which protein quality-control pathways can modulate the severity of β-thalassemia by eliminating unstable free α-globin, and they open up possibilities for new therapeutic approaches, as Ulk1 activity can be modulated pharmacologically.
No relevant conflicts of interest to declare.
