The chicken hypersensitivity site 4 core insulator blocks promoter interference in lentiviral vectors.

Lentiviral vectors, including double internal promoters, can be used to express two transgenes in a single vector construct; however, transcriptional activities from double internal promoters are often inhibited by promoter interference. To determine whether the chicken hypersensitivity site 4 insulator (cHS4) could block promoter interference, lentiviral vectors including an MSCV-U3 promoter (Mp) and an EF1α promoter (Ep) were generated, and transgene expression was evaluated among transduced cells. In the Ep-Mp configuration, transcriptional activity from Mp was much lower, while Mp-Ep had similar transcription levels from both promoters. The cHS4 core insulator increased expression levels from Mp in HeLa cells, hematopoietic cell lines, and mouse peripheral blood cells following hematopoietic stem cell transplantation transduced with the Mp-Ep configured vector. This blocking function was mainly mediated by barrier activity regions in the insulator but not by CCCTC-binding factor (CTCF) binding sites. Cytosine-phosphate-guanine (CpG) methylation did not contribute to this barrier activity. In summary, combining the cHS4 insulator in double promoter vectors can improve transgene expression levels in various cell lines and mouse hematopoietic repopulating cells. These findings are useful for developing hematopoietic stem cell gene therapy.

Overexpression of GDNF in the uninjured DRG exerts analgesic effects on neuropathic pain following segmental spinal nerve ligation in mice.

UNLABELLED: Glial cell line-derived neurotrophic factor (GDNF), a survival-promoting factor for a subset of nociceptive small-diameter neurons, has been shown to exert analgesic effects on neuropathic pain. However, its detailed mechanisms of action are still unknown. In the present study, we investigated the site-specific analgesic effects of GDNF in the neuropathic pain state using lentiviral vector-mediated GDNF overexpression in mice with left fifth lumbar (L5) spinal nerve ligation (SNL) as a neuropathic pain model. A lentiviral vector expressing both GDNF and enhanced green fluorescent protein (EGFP) was constructed and injected into the left dorsal spinal cord, uninjured fourth lumbar (L4) dorsal root ganglion (DRG), injured L5 DRG, or plantar skin of mice. In SNL mice, injection of the GDNF-EGFP-expressing lentivirus into the dorsal spinal cord or uninjured L4 DRG partially but significantly reduced the mechanical allodynia in association with an increase in GDNF protein expression in each virus injection site, whereas injection into the injured L5 DRG or plantar skin had no effects. These results suggest that GDNF exerts its analgesic effects in the neuropathic pain state by acting on the central terminals of uninjured DRG neurons and/or on the spinal cells targeted by the uninjured DRG neurons. PERSPECTIVE: This article shows that GDNF exerts its analgesic effects on neuropathic pain by acting on the central terminals of uninjured DRG neurons and/or on the spinal cells targeted by these neurons. Therefore, research focusing on these GDNF-dependent neurons in the uninjured DRG would provide a new strategy for treating neuropathic pain.

Leukemogenesis of b2a2-type p210 BCR/ABL in a bone marrow transplantation mouse model using a lentiviral vector.

The BCR/ABL fusion oncogene found in Philadelphia-positive leukemia exists in three principle forms: p190, p210 and p230. P210 BCR/ABL is commonly found in patients with chronic myelogenous leukemia (CML) and is further categorized into b3a2 or b2a2 subtypes on the basis of the BCR breakpoint. Although these 2 subtypes may be clinically heterogeneous, only the b3a2 BCR/ABL gene has been extensively studied at the molecular and cellular levels. In the present study, we compared the in vivo leukemogenic activity of the b3a2 and b2a2 BCR/ABL genes by using lentiviral transduction/transplantation mouse models. Lineage-depleted bone marrow cells of BALB/c mice were transduced with a lentiviral vector including either b2a2 or b3a2 BCR/ABL cDNA and then transplanted into lethally irradiated mice. In this model, p210 BCR/ABL subtype developed only B220(+), CD3e(-), Gr1(-), and Mac1(-) B-cell acute lymphoblastic leukemia but not myeloid leukemia. There were no differences in the incidence of leukemogenesis, the white blood cell count, the percentage of blast cells, or the survival rates between the b2a2 and b3a2 groups. We have demonstrated that b2a2-type BCR/ABL has leukemogenic activity similar to that of b3a2-type BCR/ABL.

Multistep pathogenesis of leukemia via the MLL-AF4 chimeric gene/Flt3 gene tyrosine kinase domain (TKD) mutation-related enhancement of S100A6 expression.

OBJECTIVE: Concerning MLL-AF4 leukemogenesis, previous mouse models suggest that the tumorigenesis capacity of MLL-AF4 alone is insufficient for causing leukemia. Based on the finding that an Fms-like tyrosine kinase 3 (Flt3) gene mutation in the tyrosine kinase domain (TKD) was observed in approximately 15% of mixed lineage leukemia (MLL), we investigated synergistic leukemogenesis effects of the two genes in vitro. MATERIALS AND METHODS: In a mouse interleukin-3 (IL-3)-dependent cell line, 32Dc, expression of MLL-AF4 and mutant Flt3 was induced using a lentiviral vector. We analyzed apoptosis induction in the absence of IL-3 and the granulocyte colony-stimulating factor-related induction of differentiation, gene expression profiling, and the mechanism involved in the synergistic effects of MLL-AF4 and Flt3-TKD. RESULTS: Neither Flt3-expressing 32Dc (32Dc(Flt3-TKD)) nor MLL-AF4-expressing 32Dc (32Dc(MLL-AF4)) acquired IL-3-independent proliferative capacity in semisolid/liquid media. However, Flt3-TKD+MLL-AF4-expressing 32Dc (32Dc(Flt3-TKD+MLL-AF4)) acquired a non-IL-3-dependent proliferative capacity by inhibiting apoptosis in the two media. The 32Dc(Flt3-TKD) and 32Dc(MLL-AF4) cells differentiated into granulocytes in the presence of granulocyte colony-stimulating factor. However, in the 32Dc(Flt3-TKD+MLL-AF4) cells, there was no differentiation. Subsequently, we performed gene expression profiling. The enhancement of Hox genes expression was not identified. However, expression of S100A6 was synergistically enhanced in the presence of both MLL-AF4 and Flt3-TKD genes. Moreover, anti-S100A6 small interfering RNA downregulated leukemic proliferation. CONCLUSION: We conclude that their synergistic enhancement of S100A6 expression plays an important role in MLL-AF4-associated leukemogenesis.

Optimized lentiviral vector design improves titer and transgene expression of vectors containing the chicken beta-globin locus HS4 insulator element.

Hematopoietic cell gene therapy using retroviral vectors has achieved success in clinical trials. However, safety issues regarding vector insertional mutagenesis have emerged. In two different trials, vector insertion resulted in the transcriptional activation of proto-oncogenes. One strategy for potentially diminishing vector insertional mutagenesis is through the use of self-inactivating lentiviral vectors containing the 1.2-kb insulator element derived from the chicken beta-globin locus. However, use of this element can dramatically decrease both vector titer and transgene expression, thereby compromising its practical use. Here, we studied lentiviral vectors containing either the full-length 1.2-kb insulator or the smaller 0.25-kb core element in both orientations in the partially deleted long-terminal repeat. We show that use of the 0.25-kb core insulator rescued vector titer by alleviating a postentry block to reverse transcription associated with the 1.2-kb element. In addition, in an orientation-dependent manner, the 0.25-kb core element significantly increased transgene expression from an internal promoter due to improved transcriptional termination. This element also demonstrated barrier activity, reducing variability of expression due to position effects. As it is known that the 0.25-kb core insulator has enhancer-blocking activity, this particular insulated lentiviral vector design may be useful for clinical application.

Long-term inhibition of glycosphingolipid accumulation in Fabry model mice by a single systemic injection of AAV1 vector in the neonatal period.

Fabry disease is caused by the deficiency of lysosomal alpha-galactosidase A (alpha-gal A) and usually develops clinical manifestations during childhood/adolescence. Adult Fabry model mice have been successfully treated by various viral vectors. Here, in order to examine the feasibility of preventive gene therapy, we compared AAV vector-mediated gene transfer into neonatal and adult model mice. AAV serotype 1 vector (AAV1) carrying human alpha-gal A cDNA driven by the CAG promoter was intravenously injected into adult (12 weeks old) and neonatal (2 days old) Fabry model mice, and were sacrificed for detailed examination 25 weeks after vector injection. AAV1 vector preferentially transduced the liver in male adult and sustained high concentration of alpha-gal A was detected in the liver, heart and plasma. In contrast, AAV1-mediated gene expression was suppressed in similarly treated female adult mice. When the vector was systemically injected into neonates, moderate increase in plasma alpha-gal A and cardiac-specific expression of alpha-gal A were observed independently of mouse sex. The high levels of alpha-gal A activity in the heart appear to be due to the strong activity of the CAG promoter in the heart. Globotriaosylceramide (Gb3) accumulation was efficiently inhibited in the liver and heart by a single injection into both adult and neonatal animals. The biodistribution of the AAV1 vector and levels of alpha-gal A expression are markedly different between adult and neonatal mice. Neonatal injection is effective to inhibit Gb3 accumulation and therefore, might help prevent failure of major organs during adulthood.

Globin lentiviral vector insertions can perturb the expression of endogenous genes in beta-thalassemic hematopoietic cells.

Although hematopoietic cell gene therapy using retroviral vectors has recently achieved success in clinical trials, safety issues regarding vector insertional mutagenesis have emerged. Vector insertion, resulting in transcriptional activation of proto-oncogenes, played a role in the development of lymphoid leukemia in an X-linked severe combined immunodeficiency trial, and caused myeloid clonal dominance in a trial for chronic granulomatous disease. These events have raised the question of whether gene therapy for other disorders such as beta-thalassemia and sickle cell disease may hold a similar risk. In this study, we prospectively evaluated whether gamma-globin lentiviral vectors containing enhancer elements from the beta-globin locus control region could alter the expression of genes near the vector insertion. We studied this question in primary, clonal murine beta-thalassemic erythroid cells, where globin regulatory elements are highly active. We found an overall incidence of perturbed expression in 28% of the transduced clones, with 11% of all genes contained within a 600-kilobase region surrounding the vector-insertion site demonstrating altered expression. This rate was higher than that observed for a lentiviral vector containing a viral long-terminal repeat (LTR). This is the first direct evidence that lentiviral vectors can cause insertional dysregulation of cellular genes at a frequent rate.

Efficient gene transfer via retrograde transport in rodent and primate brains using a human immunodeficiency virus type 1-based vector pseudotyped with rabies virus glycoprotein.

The primate lentiviral vector system based on human immunodeficiency virus type 1 (HIV-1) has been used for a wide range of gene therapy trials in animal models. Axonal transport in the retrograde direction, which is observed with some viral vectors, confers a considerable advantage to gene transfer into neuronal cell bodies that are localized in regions remote from the injection site of the vectors. However, retrograde axonal transport of the HIV-1-based lentiviral vector pseudotyped with vesicular stomatitis virus glycoprotein is reported to be inefficient. In the present study, we developed an efficient gene transfer system through retrograde transport in the brain with the HIV-1-based vector pseudotyped with rabies virus glycoprotein (RV-G). Injection of the RV-G-pseudotyped HIV-1 vector into the dorsal striatum of mice yielded an increase in gene transfer into neuronal populations in the cerebral cortex, thalamus, and ventral midbrain, each of which innervates the striatum. In addition, injection of the RV-G-pseudotyped vector into the monkey striatum (putamen) resulted in highly efficient transfer into neurons in the ventral midbrain (nigrostriatal dopamine neurons). Our results indicate that pseudotyping of the HIV-1 vector with RV-G enhances the efficiency of gene transfer through retrograde axonal transport in both mouse and monkey brains. This primate lentiviral vector system will provide a powerful approach to gene therapy for neurological and neurodegenerative diseases by means of enhanced retrograde transport.

Systemic cancer gene therapy using adeno-associated virus type 1 vector expressing MDA-7/IL24.

Melanoma differentiation-associated gene-7/interleukin-24 (mda-7/IL24), selectively induces apoptosis in cancer cells without harming normal cells. It also exerts immunomodulatory and antiangiogenic effects, as well as potent antitumor bystander effects, making it an ideal candidate for a new anticancer gene therapy. Here, we examined the feasibility of adeno-associated virus type 1 (AAV1) vector-mediated systemic gene therapy using mda-7/IL24. In vitro studies showed that medium conditioned by AAV1-mda7-transducedC2C12 cells induces tumor cell-specific apoptosis and inhibits angiogenesis in a human umbilical vein endothelial cell tube formation assay. To assess the in vivo effects of AAV1-mediated systemic delivery of MDA-7/IL24, we generated a subcutaneous tumor model by injecting Ehrlich ascites tumor cells into the dorsum of DDY mice. A single intravenous injection of AAV1-mda7 (2.0 x 10(11) viral genomes) significantly inhibited tumor growth. In addition, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL), and immunohistochemical analyses showed significant induction of tumor-cell-specific apoptosis and reduction of microvessel formation within the tumors, and there was a significant increase in survival among the AAV1-mda7-treated mice. These results clearly demonstrate that continuous systemic delivery of MDA-7/IL24 can serve as an effective treatment for cancer. Thus, AAV1 vector-mediated systemic delivery of MDA-7/IL24 represents a potentially important new approach to anticancer therapy.

Suppression of clonal dominance in cultured human lymphoid cells by addition of the cHS4 insulator to a lentiviral vector.

Lentiviral vectors efficiently transduce quiescent stem cells and are being evaluated for gene therapy of blood dis-orders. The risk of genotoxicity as a result of insertional mutagenesis is an important safety consideration. The hy-persensitive site 4 insulator from the chicken beta-globin locus (cHS4) possesses chromatin bar-rier and enhancer-blocking functions. A control lentiviral vector encoding green fluorescent protein was compared with a vector in which the cHS4 insulator element flanked the green fluorescent protein expression cassette in single cell isolates of transduced human T cells (Jurkat) after 9 days in culture. The insulator had minimal effect on mean fluorescent intensity and only modestly reduced the variability of green fluorescent protein expression among indi-vidual single cell isolates. Most unique integration sites were within genes, but the insulator-containing vector had a moderate predilection to integrate near the transcriptional start site compared with the control vector. Clonal domi-nance developed in cultures of cells containing the integrated vector genomes, as reflected by the recovery of mul-tiple single cell isolates containing the same integration site. We infer that certain integrations conferred a prolifera-tive or survival advantage by affecting gene expression through insertional mutagenesis, leading to this clonal dominance. This effect was diminished by including the insulator element in the vector genome.

Differential characteristics of HIV-based versus SIV-based lentiviral vector systems: Gene delivery to neurons and axonal transport of expressed gene.

The differential characteristics of lentiviral vectors based on human and simian immunodeficiency viruses (HIV and SIV) were investigated in rats and monkeys. Each vector was injected into the striatum, and the expression patterns of the marker gene green fluorescent protein (GFP) were analyzed in the basal ganglia. With respect to the capability of gene delivery to neural cells, the HIV-based vector exhibited a higher tropism to neurons than to astroglias in the striatum, and vice versa for the SIV-based vector. The preferential direction of axonal transport of striatally expressed GFP was also examined in the present study. The HIV-based vector allowed for both anterograde transport via the striatopallidal and striatonigral pathways and retrograde transport via the nigrostriatal pathway. The GFP labeling of axon terminals through anterograde transport was apparent regardless of the animal species, while that of neuronal cell bodies through retrograde transport was much more prominent in monkeys than in rats. As for the SIV-based vector, on the other hand, evidence for anterograde transport was obtained much more markedly in monkeys than in rats, and only weak or no retrograde transport occurred in either monkeys or rats. Our results indicate that HIV-based, but not SIV-based, lentiviral vectors possess the high tropism to neurons and permit retrograde transport of an expressed gene, especially in primates. The latter property might carry a potential benefit in gene therapy for Parkinson's disease, as stereotaxic injections of the vectors could be performed into the striatum, spatially larger than the substantia nigra, with greater certainty.

Mobilization and mechanism of transcription of integrated self-inactivating lentiviral vectors.

Permanent genetic modification of replicating primitive hematopoietic cells by an integrated vector has many potential therapeutic applications. Both oncoretroviral and lentiviral vectors have a predilection for integration into transcriptionally active genes, creating the potential for promoter activation or gene disruption. The use of self-inactivating (SIN) vectors in which a deletion of the enhancer and promoter sequences from the 3' long terminal repeat (LTR) is copied over into the 5' LTR during vector integration is designed to improve safety by reducing the risk of mobilization of the vector genome and the influence of the LTR on nearby cellular promoters. Our results indicate that SIN vectors are mobilized in cells expressing lentiviral proteins, with the frequency of mobilization influenced by features of the vector design. The mechanism of transcription of integrated vector genomes was evaluated using a promoter trap design with a vector encoding tat but lacking an upstream promoter in a cell line in which drug resistance depended on tat expression. In six clones studied, all transcripts originated from cryptic promoters either upstream or within the vector genome. We estimate that approximately 1 in 3,000 integrated vector genomes is transcribed, leading to the inference that activation of cryptic promoters must depend on local features of chromatin structure and the constellation of nearby regulatory elements as well as the nature of the regulatory elements within the vector.

The role of bone marrow-derived cells in bone fracture repair in a green fluorescent protein chimeric mouse model.

We investigated the role of bone marrow cells in bone fracture repair using green fluorescent protein (GFP) chimeric model mice. First, the chimeric model mice were created: bone marrow cells from GFP-transgenic C57BL/6 mice were injected into the tail veins of recipient wild-type C57BL/6 mice that had been irradiated with a lethal dose of 10Gy from a cesium source. Next, bone fracture models were created from these mice: closed transverse fractures of the left femur were produced using a specially designed device. One, three, and five weeks later, fracture lesions were extirpated for histological and immunohistochemical analyses. In the specimens collected 3 and 5 weeks after operation, we confirmed calluses showing intramembranous ossification peripheral to the fracture site. The calluses consisted of GFP- and osteocalcin-positive cells at the same site, although the femur consisted of only osteocalcin-positive cells. We suggest that bone marrow cells migrated outside of the bone marrow and differentiated into osteoblasts to make up the calluses.

Distinct genomic integration of MLV and SIV vectors in primate hematopoietic stem and progenitor cells.

Murine leukemia virus (MLV)-derived vectors are widely used for hematopoietic stem cell (HSC) gene transfer, but lentiviral vectors such as the simian immunodeficiency virus (SIV) may allow higher efficiency transfer and better expression. Recent studies in cell lines have challenged the notion that retroviruses and retroviral vectors integrate randomly into their host genome. Medical applications using these vectors are aimed at HSCs, and thus large-scale comprehensive analysis of MLV and SIV integration in long-term repopulating HSCs is crucial to help develop improved integrating vectors. We studied integration sites in HSCs of rhesus monkeys that had been transplanted 6 mo to 6 y prior with MLV- or SIV-transduced CD34(+)cells. Unique MLV (491) and SIV (501) insertions were compared to a set of in silico-generated random integration sites. While MLV integrants were located predominantly around transcription start sites, SIV integrants strongly favored transcription units and gene-dense regions of the genome. These integration patterns suggest different mechanisms for integration as well as distinct safety implications for MLV versus SIV vectors.

Extended beta-globin locus control region elements promote consistent therapeutic expression of a gamma-globin lentiviral vector in murine beta-thalassemia.

Since increased fetal hemoglobin diminishes the severity of beta-thalassemia and sickle cell anemia, a strategy using autologous, stem cell-targeted gene transfer of a gamma-globin gene may be therapeutically useful. We previously found that a gamma-globin lentiviral vector utilizing the beta-globin promoter and elements from the beta-globin locus control region (LCR) totaling 1.7 kb could correct murine beta-thalassemia. However, therapeutic consistency was compromised by chromosomal position effects on vector expression. In contrast, we show here that the majority of animals that received transplants of beta-thalassemic stem cells transduced with a new vector containing 3.2 kb of LCR sequences expressed high levels of fetal hemoglobin (17%-33%), with an average vector copy number of 1.3. This led to a mean 26 g/L (2.6 g/dL) increase in hemoglobin concentration and enhanced amelioration of other hematologic parameters. Analysis of clonal erythroid cells of secondary spleen colonies from mice that underwent transplantation demonstrated an increased resistance of the larger LCR vector to stable and variegating position effects. This trend was also observed for vector insertion sites located inside genes, where vector expression was often compromised, in contrast to intergenic sites, where higher levels of expression were observed. These data emphasize the importance of overcoming detrimental position effects for consistent therapeutic globin vector expression.

Efficient gene transfer into rhesus repopulating hematopoietic stem cells using a simian immunodeficiency virus-based lentiviral vector system.

High-titer, HIV-1-based lentiviral vector particles were found to transduce cytokine-mobilized rhesus macaque CD34(+) cells and clonogenic progenitors very poorly (< 1%), reflecting the postentry restriction in rhesus cells to HIV infection. To overcome this barrier, we developed a simian immunodeficiency virus (SIV)-based vector system. A single exposure to a low concentration of amphotropic pseudotyped SIV vector particles encoding the green fluorescent protein (GFP) resulted in gene transfer into 68% +/- 1% of rhesus bulk CD34(+) cells and 75% +/- 1% of clonogenic progenitors. Polymerase chain reaction (PCR) analysis of DNA from individual hematopoietic colonies confirmed these relative transduction efficiencies. To evaluate SIV vector-mediated stem cell gene transfer in vivo, 3 rhesus macaques underwent transplantation with transduced, autologous cytokine-mobilized peripheral blood CD34(+) cells following myeloablative conditioning. Hematopoietic reconstitution was rapid, and an average of 18% +/- 8% and 15% +/- 7% GFP-positive granulocytes and monocytes, respectively, were observed 4 to 6 months after transplantation, consistent with the average vector copy number of 0.19 +/- 0.05 in peripheral blood leukocytes as determined by real-time PCR. Vector insertion site analysis demonstrated polyclonal reconstitution with vector-containing cells. SIV vectors appear promising for evaluating gene therapy approaches in nonhuman primate models.

Development of gene therapy for hemoglobin disorders.

The hemoglobin disorders, severe beta-thalassemia and sickle cell anemia, are prevalent monogenetic disorders which cause severe morbidity and mortality worldwide. Gene therapy approaches to these disorders envision stem cell targeted gene transfer, autologous transplantation of gene-corrected stem cells, and functional, phenotypically corrective globin gene expression in developing erythroid cells. Lentiviral vector systems potentially appear to afford adequately efficient gene transfer into stem cells and are capable, with appropriate genetic engineering, of transferring a globin gene with the regulatory elements required to achieve high-level, erythroid-specific expression. Herein are results obtained in use of lentiviral vectors to insert a gamma-globin gene into murine stem cells with phenotypic correction of the thalassemia phenotype. Further, we have developed a drug-selection system for genetically modified stem cells based on a mutant form of methylguanine, methyltransferase, which allows selective amplification of genetically modified stem cells with phenotypic correction even in the absence of myeloablation prior to stem cell transplantation. These advances provide essential preclinical data which build toward the development of effective gene therapy for the severe hemoglobin disorders.

Successful treatment of murine beta-thalassemia using in vivo selection of genetically modified, drug-resistant hematopoietic stem cells.

Successful gene therapy of beta-thalassemia will require replacement of the abnormal erythroid compartment with erythropoiesis derived from genetically corrected, autologous hematopoietic stem cells (HSCs). However, currently attainable gene transfer efficiencies into human HSCs are unlikely to yield sufficient numbers of corrected cells for a clinical benefit. Here, using a murine model of beta-thalassemia, we demonstrate for the first time that selective enrichment in vivo of transplanted, drug-resistant HSCs can be used therapeutically and may therefore be a useful approach to overcome limiting gene transfer. We used an oncoretroviral vector to transfer a methylguanine methyltransferase (MGMT) drug-resistance gene into normal bone marrow cells. These cells were transplanted into beta-thalassemic mice given nonmyeloablative pretransplantation conditioning with temozolomide (TMZ) and O6-benzylguanine (BG). A majority of mice receiving 2 additional courses of TMZ/BG demonstrated in vivo selection of the drug-resistant cells and amelioration of anemia, compared with untreated control animals. These results were extended using a novel gamma-globin/MGMT dual gene lentiviral vector. Following drug treatment, normal mice that received transduced cells had an average 67-fold increase in gamma-globin expressing red cells. These studies demonstrate that MGMT-based in vivo selection may be useful to increase genetically corrected cells to therapeutic levels in patients with beta-thalassemia.

The degree of phenotypic correction of murine beta -thalassemia intermedia following lentiviral-mediated transfer of a human gamma-globin gene is influenced by chromosomal position effects and vector copy number.

Increased fetal hemoglobin (HbF) levels diminish the clinical severity of beta-thalassemia and sickle cell anemia. A treatment strategy using autologous stem cell-targeted gene transfer of a gamma-globin gene may therefore have therapeutic potential. We evaluated oncoretroviral- and lentiviral-based gamma-globin vectors for expression in transduced erythroid cell lines. Compared with gamma-globin, oncoretroviral vectors containing either a beta-spectrin or beta-globin promoter and the alpha-globin HS40 element, a gamma-globin lentiviral vector utilizing the beta-globin promoter and elements from the beta-globin locus control region demonstrated a higher probability of expression. This lentiviral vector design was evaluated in lethally irradiated mice that received transplants of transduced bone marrow cells. Long-term, stable erythroid expression of human gamma-globin was observed with levels of vector-encoded gamma-globin mRNA ranging from 9% to 19% of total murine alpha-globin mRNA. The therapeutic efficacy of the vector was subsequently evaluated in a murine model of beta-thalassemia intermedia. The majority of mice that underwent transplantation expressed significant levels of chimeric m(alpha)(2)h(gamma)(2) molecules (termed HbF), the amount of which correlated with the degree of phenotypic improvement. A group of animals with a mean HbF level of 21% displayed a 2.5 g/dL (25 g/L) improvement in Hb concentration and normalization of erythrocyte morphology relative to control animals. gamma-Globin expression and phenotypic improvement was variably lower in other animals due to differences in vector copy number and chromosomal position effects. These data establish the potential of using a gamma-globin lentiviral vector for gene therapy of beta-thalassemia.