Toward gene therapy for β-thalassemia: new models, new approaches

Gene therapy for hemoglobinopathies has suffered from problems of vector instability, low viral titers, and variable expression for over a decade. In a pioneering study, Sadelain's group had shown that a lentiviral vector was able to stably transmit the human β-globin gene and its regulatory elements, resulting in therapeutic correction of β-thalassemia in Hbb^th3/+ mice (May et al, Nature. 2000;406:82-86). But the Hbb^th3/+ mice carry deletion of the β^major and β^minor genes only on one allele (Ciavatta et al, Proc Natl Acad Sci U S A. 1995;92:9259-9263; Yang et al, Proc Natl Acad Sci U S A. 1995;92:11608-11612) and resemble the human thalassemia intermedia phenotype. The degree of correction accomplished by the TNS-9 vector (about 3 g/dL increase in hemoglobin level per proviral copy) would be subtherapeutic in humans with Cooley anemia (β-thalassemia major). Homozygous deletion of both the β^minor and β^major globins is embryonic lethal in mice because, unlike in humans, the switch to adult globin production occurs in utero.

In this issue, Rivella and colleagues (page 2932) have developed a model of mouse thalassemia major by transplanting fetal liver stem cells from thalassemia homozygous fetuses into lethally irradiated healthy adults. They report a recapitulation of thalassemia major phenotype observed in humans, starting as early as 6 weeks following transplantation. Genetic correction of the thalassemia major bone marrow with the TNS-9 vector followed by a transplantation rescues the otherwise lethal anemia. But increases in hemoglobin level are no higher than those previously reported by this group using the same vector (May et al), predictably converting the thalassemia major phenotype to that of severe thalassemia intermedia. Nevertheless, this model of murine β-thalassemia major is representative of human disease and, eventually, may become the ultimate model to test therapeutic strategies.

Using a different approach, Persons and colleagues (Blood. 2003;101:2175-2183) have used the human γ-globin gene in a lentiviral vector, driven by a minimal β-globin promoter and the β-globin locus control region (LCR) elements to correct murine thalassemia intermedia. They show therapeutic increases in fetal hemoglobin production with 2 or more copies of provirus per cell in Hbb^th3/+ mice. The levels of expression are less robust than were levels of β-globin or a mutant β-globin expression previously observed by the Sadelain (May et al) and Leboulch (Pawliuk et al, Science. 2001;294:2368-2371) laboratories, respectively, probably due to a smaller β-globin gene promoter or LCR fragments. Nevertheless, their vector produces the highest levels of fetal hemoglobin protein reported in primary cells from an integrating viral vector. Additionally, γ-globin vectors have the advantage of a therapeutic potential both in thalassemia and sickle cell disease. Persons et al have taken gene therapy for β-thalassemia a step further: they have used self-inactivating lentiviral vectors, where the viral long-terminal repeat is deleted upon integration into cells, inactivating viral transcription and improving their biosafety.

While both these studies represent important strides toward gene therapy for thalassemia, they also highlight the obstacles yet to be conquered. Both studies show presence of chromatin position effects and underscore the need for better vectors, which would yield higher and predictable increases in hemoglobin to be therapeutic in human β-thalassemia major.