Abstract

While genetic delivery of recombinant anti-sickling β-globin genes have been shown to correct murine sickle cell anemia (SCA), correction of SCA by delivery of a natural hemoglobin, fetal hemoglobin (HbF), the proportion of genetically modified hematopoietic stem cells (HSC), or amount of HbF necessary to correct the disease is unknown. We designed a lentivirus vector carrying γ-globin exons with β-globin regulatory elements and non-coding sequences, GbG. First, GbG or mock transduced Berkeley sickle HSC were transplanted using a myeloablative (lethal irradiation) transplant model, to acheive full donor chimerism. GbG mice showed high HbF expression (HbF 41 ± 5% measured by HPLC) that was sustained in primary (6 mo) and secondary (7.5 mo) transplant recipients, and resulted in effective correction of hematological and functional RBC parameters, and reduction of inflammation that results from sickle cell disease. We found significantly reduced irreversibly sickled cells (2.3 ± 0.7% in GbG versus 10.2 ± 0.3% in mock mice; p<0.001), minimal sickling of RBC when exposed to graded hypoxia using tonometry, improved RBC deformability (performed by ektacytometry), and a four-fold increase in RBC half-life (by in vivo biotin labeling) in the GbG group of mice. There was correction of anemia, and reduction in hemolysis (measured by LDH levels), reticulocytes, and leukocytosis (Table 1). This was accompanied by a dramatic improvement in chronic organ damage that is seen in untransplanted Berkeley/mock group of mice: there was a significant reduction in spleen weights and normalization of splenic follicular architecture, correction in bone marrow myeloid:erythroid ratios, and a notable absence of kidney infarction and atrophy, and liver infarction and extramedullary hematopoiesis that was observed in mock mice. Untransplanted Berkeley and mock mice showed shortened survival consistent with a severe SCA phenotype. Genetic correction with GbG improved survival to 100% compared to a 20% survival in the mock transplanted. Notably, in our proof-of principle studies, comparable functional sickle RBC correction was also seen in the Townes knock-in sickle mice (Wu et al, Blood 2006) transduced with GbG. Myeloablative conditioning in this setting allowed non-competitive repopulation of donor genetically modified HSC, resulting in high HbF and correction of disease. However, myeloablation in SCA is associated with peri-transplant mortality and long-term effects, and may not be necessary for achieving correction of phenotype. To address this, we used a unique reduced-intensity conditioning transplant model. We transplanted GbG-modified Berkeley HSC into sub-lethally irradiated Berkeley mice. In this model, when HbF was <10%, there was a small and variable improvement in hematological and functional sickle RBC parameters. However, when HbF was γ10%, there was consistent long-term correction in RBC sickling, deformability, RBC survival, and improvement in hematological parameters for 10–11 months (Table 1). Impressively, when HbF was γ10%, there was a significant reduction in splenomegaly, absence of liver and kidney pathology, and a dramatically improved overall survival of the mice, comparable to that seen in the myeloablative model. Comparison of the proportion of F-cells (HbF containing RBC) and HbF/F-cell to the assays showing correction of SCA revealed that >30% HbF/F-cell and >60% F-cells consistently corrected SCA. The mean HSC transduction (assessed by secondary HbF+ CFU-S at 6 months post transplant) was 50% and 30% in the myeloablative and reduced intensity transplant models, respectively, with 1–3 GbG copies/ cell. Furthermore, three GbG mice showed correction of SCA with 20% HSC transduction, a clinically achievable goal. Taken together, this study is the first demonstration of correction of SCA with gene therapy using γ-globin, and defines critical determinants for effective gene therapy of this disease.

Mouse Model Hb (g/dl) RBC 106/ul) Reticulocyte (%) WBC (K/ul) 
*p<0.05; ** p<0.001 
Mock Myeloablative 7.6±0.7 5.8±0.4 40.0±3.0 29.7±1.4 
GbMyeloablative 10±0.8* 9.4±0.8** 15.8±3.2** 10.6±3.1** 
GbG, HbF ≥ 10% Reduced intensity 9.3±0.6* 8.1±0.5** 21.2±1.9** 13.4±1.1** 
Mouse Model Hb (g/dl) RBC 106/ul) Reticulocyte (%) WBC (K/ul) 
*p<0.05; ** p<0.001 
Mock Myeloablative 7.6±0.7 5.8±0.4 40.0±3.0 29.7±1.4 
GbMyeloablative 10±0.8* 9.4±0.8** 15.8±3.2** 10.6±3.1** 
GbG, HbF ≥ 10% Reduced intensity 9.3±0.6* 8.1±0.5** 21.2±1.9** 13.4±1.1** 

Disclosures: No relevant conflicts of interest to declare.

[AP and TH contributed equally to this work].

Author notes

Corresponding author