Permanent and panerythroid correction of murine β thalassemia by multiple lentiviral integration in hematopoietic stem cells

Abstract

Achieving long-term pancellular expression of a transferred gene at therapeutic level in a given hematopoietic lineage remains an important goal of gene therapy. Advances have recently been made in the genetic correction of the hemoglobinopathies by means of lentiviral vectors and large locus control region (LCR) derivatives. However, panerythroid β globin gene expression has not yet been achieved in β thalassemic mice because of incomplete transduction of the hematopoietic stem cell compartment and position effect variegation of proviruses integrated at a single copy per genome. Here, we report the permanent, panerythroid correction of severe β thalassemia in mice, resulting from a homozygous deletion of the β major globin gene, by transplantation of syngeneic bone marrow transduced with an HIV-1-derived [β globin gene/LCR] lentiviral vector also containing the Rev responsive element and the central polypurine tract/DNA flap. The viral titers produced were high enough to achieve transduction of virtually all of the hematopoietic stem cells in the graft with an average of three integrated proviral copies per genome in all transplanted mice; the transduction was sustained for >7 months in both primary and secondary transplants, at which time ≈95% of the red blood cells in all mice contained human β globin contributing to 32 ± 4% of all β-like globin chains. Hematological parameters approached complete phenotypic correction, as assessed by hemoglobin levels and reticulocyte and red blood cell counts. All circulating red blood cells became and remained normocytic and normochromic, and their density was normalized. Free α globin chains were completely cleared from red blood cell membranes, splenomegaly abated, and iron deposit was almost eliminated in liver sections. These findings indicate that virtually complete transduction of the hematopoietic stem cell compartment can be achieved by high-titer lentiviral vectors and that position effect variegation can be mitigated by multiple events of proviral integration to yield balanced, panerythroid expression. These results provide a solid foundation for the initiation of human clinical trials in β thalassemia patients.

With an estimated 80,000,000 carriers worldwide (1), β thalassemia is a major genetic cause of morbidity and mortality. The hallmark of the disease is reduced or absent β globin synthesis (2). The imbalance, resulting in excess α chains, causes abnormal maturation of erythroblasts and their premature destruction in the marrow (2). Amelioration of the disease can be achieved by repeated transfusions in combination with the continuous administration of chelating agents to prevent iron overload (2). Yet, many patients die prematurely of their disease or of secondary iron overload (2). The only currently available curative treatment is replacement of the marrow with an allogeneic transplant, but this procedure is only available to patients with a matched donor and it exposes the patients to a significant risk of mortality and morbidity that includes graft-versus-host disease (2).

Cure of β thalassemia by the transplantation of genetically modified autologous hematopoietic stem cells (HSCs) has been recognized for more than two decades as an important new treatment strategy. However, formidable hurdles have hindered the progress of this approach. Transgenic mouse experiments have revealed the need to transfer a complete β globin gene together with previously discovered elements of the locus control region (LCR) (3, 4) to achieve high and erythroid-specific expression (5, 6). Although retroviral vectors remain the main modality of achieving efficient chromosomal integration of transferred genes, oncoretroviruses containing [β globin gene/LCR] DNA segments have proved extremely prone to rearrangements and low viral titers (7, 8). Through systematic analysis of patterns of proviral transmission of these vectors, we previously formulated and documented the hypothesis that internal splicing and poor nucleo-cytoplasmic export of the genomic viral RNA before packaging were the primary determinants of the observed untoward phenomena (9, 10). The permutation of LCR elements also resulted in vector stability, perhaps by modifying the pattern of splice-site distribution along the vector sequence (11). Although mutagenesis of unwanted splice sites and inclusion of various RNA export elements provided useful remedies allowing us to document long-term transfer and expression in mouse transplant experiments (12, 13), lentiviruses have become the most effective vectors available, primarily because they provide the uniquely powerful Rev/Rev-responsive element machinery, the most efficient system known for export of long and unspliced RNA (14).

Lentiviral vectors were recently successfully exploited to achieve efficient and stable transfer of large [β globin gene/LCR] segments resulting in therapeutic levels of β globin gene expression in both β thalassemia (15, 16) and sickle cell disease mouse models (17). In contrast to the results we obtained with an antisickling lentiviral vector in which virtually all HSCs from the graft seemed transduced with several integrated proviruses per cell, May et al. (15) reported incomplete transduction of the HSC compartment with an average of 0.75 integrated provirus per cell and persisting heterocellular expression of the transferred β globin gene. Although similar in their overall architecture, the two lentiviral vectors differ in the specific segments of promoter and LCR chosen, and the antisickling vector also incorporates the central polypurine tract/DNA flap of HIV-1, which has been reported to facilitate gene transfer to quiescent cells (18). Here, we set out (i) to determine whether a lentiviral vector similar to the antisickling vector, but for the presence of a wild-type human β globin gene, would completely transduce the donor HSC compartment in long-term transplant experiments in β thalassemic mice with a severe phenotype, and (ii) to study the in vivo expression profile of the transferred human β globin gene. Panerythroid permanent expression of the human β globin gene was observed in all transplanted β thalassemic mice with virtually complete correction of all observable disease manifestations.

Materials and Methods

Lentiviral Vector Design and Production.

The lentiviral globin gene vector used is schematically shown in Fig. 1. This vector is based on the design previously described in detail and proven successful for the correction of sickle cell disease (17), except that it now incorporates the wild-type human β globin gene (Fig. 1). Recombinant virus pseudotyped with vesicular stomatitis virus glycoprotein-G was produced and subsequently concentrated 1,000-fold by two rounds of ultracentrifugation as described (17). A lentiviral vector carrying the gene that encodes enhanced GFP driven by the elongation factor 1-α promoter was also generated and used as a control in some experiments (details available on request). The absence of replication competent virus was verified by mobilization assay. Viral titers were determined functionally by quantitative Southern blot analysis of transduced NIH 3T3 cells with proviral copy number controls.

Fig 1.

(A) Diagram of the human β globin (β^A) lentiviral vector. HIV LTR, HIV type-1 long terminal repeat; ψ+, packaging signal; cPPT, central polypurine tract/DNA flap; RRE, Rev-responsive element; E, exon; IVS, intervening sequence; βP, β globin promoter (from SnaBI to Cap site); HS, hypersensitive site; ppt, polypurine tract. The 3′ β globin enhancer (up to downstream AvrII site), the 372-bp IVS2 deletion (indicated by the triangle) and DNase I hypersensitive sites, HS2 (SmaI to XbaI), HS3 (SacI to PvuII) and HS4 (StuI to SpeI) of the LCR are shown. (B) Proportion of peripheral blood RBCs expressing human β globin assessed by FACS after staining the cells with an antibody specific for human β globin. RBCs were from THAL mice transplanted 7 months previously with bone marrow exposed to the lenti-β globin vector. The viral titer used in experiment 1 (E1), was ≈2 × 10⁸ infectious units/ml and in experiments 2 and 3 (E2, E3) was ≈1.5 × 10⁹ infectious units/ml. (C) Expression of human β globin in RBCs of reconstituted mice. (Upper Left) FACS analysis of RBCs from a representative recipient of lenti-GFP-virus-transduced THAL bone marrow cells. RBCs from an unmanipulated mouse were used as a negative control. (Upper Right and Lower Left) FACS analysis of RBCs from a representative recipient of lenti-β globin-virus-transduced THAL bone marrow cells at 2 and 7 months after transplantation. (Lower Right) FACS analysis of RBCs from one of the secondary recipients transplanted with bone marrow cells from a primary donor at 6 months after transplantation.

Mice.

β-Thalassemia mice homozygous for a deletion of the murine β-major gene (C57BL/6 Hbb^th-1/Hbb^th-1) (19), hereafter referred to as THAL mice, and control C57BL/6(B6) mice were bred from parental stocks obtained from The Jackson Laboratory. The identity of homozygous THAL mice was confirmed by isoelectric focusing analysis of RBC lysates to detect characteristic single, slow-migrating Hb tetramers consisting of two murine α and two murine β minor globin chains.

Bone Marrow Transduction and Transplantation.

Bone marrow cells (24 × 10⁶) from male THAL mice injected intravenously 4 days previously with 5-fluorouracil (100 mg/kg) were stimulated overnight in Iscove's medium supplemented with 1% BSA, 10 μg/ml bovine pancreatic insulin, and 200 μg/ml human transferrin (BIT; StemCell Technologies, Vancouver), 10⁻⁴ M 2-mercaptoethanol, 2 mM glutamine, 10 ng/ml human interleukin-11 (IL-11, Genetics Institute, Cambridge, MA), 100 ng/ml human flt3-ligand (Immunex, Seattle, WA) and 300 ng/ml murine steel factor (expressed in COS cells and purified at the Terry Fox Laboratories, Vancouver). The next day, harvested cells were pelleted and resuspended in 0.9 ml of the aforementioned medium containing the same growth factor combination with concentrated, vesicular stomatitis virus glycoprotein-G-pseudotyped GFP- or β globin-lentivirus at a final virus concentration of 1.5 × 10⁹ infectious units/ml (functional titer measured by Southern blot analysis of transduced NIH 3T3 cells). Infection was performed for 5 h on fibronectin (5 μg/cm², Sigma)-coated Petri dishes in the presence of 5 μg/ml protamine sulfate. After infection, 2 × 10⁶ cells were transplanted, without selection, by i.v. injection into each female THAL recipient given 900 cGy (110 cGy/min 37 Cs γ-rays) of total body irradiation.

RNA and DNA Analyses.

Human β globin RNA in peripheral reticulocytes was quantified by RNase protection assay as described (9). Southern blot analysis was performed by using standard methods (20). A 1.6-kb BamH1 fragment of the human β globin gene, [³²P]dCTP-labeled by random priming, was used as a probe.

Globin Protein Analysis.

The proportion of RBCs expressing human β globin protein was assessed by fluorescence-activated cell sorter (FACS) analysis of RBCs that had been fixed and stained with a biotinylated anti-human antibody (Perkin–Elmer) and Streptavidin-PE (12). RBC lysates from freshly collected blood were analyzed by isoelectric focusing by using the Resolve Hb test kit (Perkin–Elmer) as described (21). The globin composition was determined by HPLC with a denaturing solvent that separates the globin chains and a Vydac large-pore (3,000 A) C₄ column with a modified acetonitril/H₂O/trifluoroacetic acid gradient as described (21). The amount of unpaired α globin chains associated with RBC membranes was determined by urea Triton-polyacrylamide gel electrophoresis analysis (22). In brief, after extensive washing of membrane ghosts, loading of equivalent amounts of protein, and Coomassie blue staining of the gel, proteins were analyzed by densitometry at 570 nm. The proportion of membrane-associated α globin chain was expressed as a percentage of total membrane proteins.

Hematologic Parameters and RBC Density Gradient.

Blood from the tail vein was used to analyze RBC indices and reticulocyte counts by using the Sysmex SE 9500 system (Sysmex Corp. of America, Long Grove, IL). Blood smears were stained with methylene blue for manual reticulocyte counts to validate the Sysmex reticulocyte counts in the majority of cases and these numbers correlated well. Blood smears were also stained with Wright-Giemsa by using an automatic stainer. Smears were reviewed blinded by two independent hematologists. RBC densities were examined on Percoll–Larex gradients (Larex International, St Paul), as described (23). The Student's t test was used to determine whether hematological parameters differed between treatment groups.

Histopathology.

Livers and spleens were fixed in 10% neutral buffered formalin. Tissues were paraffin-embedded. Five-micrometer sections were stained with hematoxylin-eosin and Perls iron stain and subsequently examined by light microscopy.

Results

Persistent Panerythroid Expression of Lentivirus-Encoded Human β Globin in β Thalassemic Mice.

The [β globin gene/LCR] lentiviral vector was optimized for viral titers and β globin gene expression by choosing specific segments of the β globin gene, its promoter, and the β-LCR on the basis of results of transgenic mouse experiments with single integrated copies and recombination-mediated cassette exchange (24) at the same sites of chromosomal integration in erythroid cell lines (data not shown). In addition, the vector contains the HIV-1 central polypurine tract/DNA flap for increased transduction efficiency (Fig. 1). After transient production in 293T cells on cotransfection with a plasmid encoding the pseudotyping vesicular stomatitis virus glycoprotein-G envelope and subsequent concentration by ultracentrifugation as described in Material and Methods, the [β globin gene/LCR] lentiviral vector reached functional titers of 1.5 × 10⁹ infectious units/ml, as assessed by Southern blot analysis of transduced NIH 3T3 cells with proviral copy number controls. The titers achieved were only 5-fold lower than those obtained with a similar lentiviral vector containing only the GFP gene driven by the elongation factor 1-α promoter.

To test the therapeutic efficacy of this vector, we used a mouse model for severe β thalassemia, referred to as THAL mice, which bear a homozygous deletion of the mouse β major gene and manifest a hypochromic, microcytic anemia with considerable anisocytosis, poikilocytosis, reticulocytosis, the presence of inclusion bodies in a high proportion of their circulating RBCs, and abnormally dehydrated erythrocytes (19). Fig. 1B shows the results of FACS analyses, with an antibody specific for human β globin chain, of peripheral blood RBCs of lethally irradiated THAL mice transplanted 7 months previously with syngeneic THAL bone marrow cells transduced with either the [β globin/LCR] or the control GFP lentiviral vector as described in Material and Methods. In all mouse recipients of cells transduced with the high-titer [β globin/LCR] vector preparations, ≈95% of RBCs were positive for human β globin protein (Fig. 1B). The high-level reconstitution obtained with the high viral titer preparations was associated with a mean proviral copy number of ≈3 per transduced cell, as assessed by Southern blot analysis of bone marrow, thymus, and spleens with proviral copy number controls (data not shown). Time-course FACS analyses showed that reconstitution of the mice with human β globin⁺ RBCs was rapid, with pancellular expression observed as early as 2 months after transplant, and stable thereafter, even in THAL mice secondarily transplanted with marrow cells harvested from the primary THAL recipients (Fig. 1C).

Similar transplantation experiments with a lower titer viral preparation (2 × 10⁸ infectious units/ml) resulted in a lesser proportion of RBCs expressing human β globin, indicating incomplete transduction of donor HSCs and position effect variegation of cells with a single integrated copy (Fig. 1B, E1; manuscript in preparation).

Correction of α Globin Protein Imbalance with Therapeutic Levels of Human β Protein in Corrected THAL Mice.

Human β globin protein constituted on average 32.4 ± 4% (27–39%) of all β globin chains in RBC lysates of THAL mice in which anemia was corrected by transplantation (Fig. 2 A and B), which was further supported by the results of isoelectric focusing analyses of blood samples from the transplanted THAL mice. An additional Hb species was documented in all corrected mice, consistent with the presence of Hb tetramer containing two murine α and two human β globin chains (Fig. 2C). Quantification of human β globin RNA in peripheral blood cells by RNA protection assay showed up to 130% levels relative to total mouse α globin RNA (data not shown). As previously observed, human β globin protein levels represent only a fraction of those of the encoding RNA, presumably because of intrinsic differences in association constants between mouse and human globin chains to form Hb tetramers and/or differential competition for globin translation between human and mouse mRNA species (25).

Fig 2.

Analysis of human β^A globin protein and membrane-associated α globin protein in RBC of THAL recipients of lenti-β globin-transduced THAL bone marrow cells. (A and B) Quantification of human β^A by HPLC. (A) Representative HPLC profile of peripheral blood from a recipient 4 months after transplant. (B) Amount of human β^A in individual corrected THAL recipients as measured by HPLC and expressed as percent of total β globin. (C) Isoelectric focusing of RBC lysates from transplanted primary mice showing the expected species of Hb with two murine α and two human β chains. Lanes 1–4, primary recipients; lanes 5 and 6, secondary recipients of lenti-β globin-transduced THAL bone marrow cells. (D) Urea-Triton polyacrylamide gradient gel of RBC membranes to detect membrane-associated α globin. Lanes 1–4, recipients of lenti-GFP transduced and lanes 5–8, recipients of lenti-β globin-transduced THAL bone marrow cells.

The transplanted THAL mice had complete clearance of the excess of membrane-bound α globin chains, which represent ≈1% of RBC membrane-associated proteins in normal mice (1.4 ± 0.7% in corrected THAL mice versus 15.3 ± 1% in GFP transplanted THAL controls) (Fig. 2D).

Correction of Hematologic Parameters and Abnormal RBC Morphology in Thalassemic Recipients.

THAL mice showing pancellular erythroid expression of human β globin also showed a marked improvement in all RBC indices (Fig. 3). Compared with pretransplant values, significant elevations (P < 0.001) occurred in RBC number (from 7 × 10⁶ to 9.7 × 10⁶ ± 0.9 per mm³), hematocrit (from 28 to 40 ± 2.3%), and Hb concentration (from 8 to 12.4 ± 0.7 g/dl) with RBC levels rising to within the normal range (no significant difference compared with normal, unmanipulated B6 mice) and just under normal levels for the hematocrit and Hb concentration. This correction of anemia was further reflected in a dramatic reduction in reticulocyte numbers from levels of >20% to 3.4 ± 0.8%, again reaching the normal range. As expected, no improvement occurred in any parameters when THAL mice were transplanted with bone marrow transduced with a GFP control vector (Fig. 3).

Fig 3.

Improvement of hematological parameters in THAL recipients of lenti-β globin-transduced THAL bone marrow cells. Results shown are the mean ± SD for untransplanted, control THAL mice (n = 5), normal unmanipulated B6 mice (n = 5), and THAL mice transplanted 6 months previously with lenti-β globin-transduced cells (n = 8) or control lenti-GFP-transduced cells (n = 6). Changes in all hematologic parameters seen in THAL recipients of lenti-β globin-transduced cells were highly significant in comparison with nontransplanted, or control (GFP)-transplanted THAL mice (*, P < 0.001). Values for the RBC number and reticulocyte count in these corrected mice reached levels within the normal range of control B6 mice (P = 0.6 and 0.1, respectively).

Findings from morphologic examination of blood smears were consistent with these results. Correction of the anemia characteristic of THAL mice was associated with a marked reduction in RBC anisocytosis, poikilocytosis, and polychromasia (Fig. 4 A–D). Although a few abnormal cells were observed, >75% of the RBCs were normochromic and normocytic in both primary and secondary recipients (Fig. 4 B and C). Control mice transplanted with lenti-GFP virus-transduced bone marrow cells remained severely anemic with marked reticulocytosis and maintained their abnormal red cell morphology (Fig. 4A). To assess further the correction of the anemia, RBC density analyses were performed. Thalassemic RBCs have a decreased mean hemoglobin concentration and correspondingly lower density as compared with normal RBCs (23). In THAL recipients of [β globin gene/LCR] lentivirus-transduced marrow, the density of RBCs narrowed dramatically toward the normal range (Fig. 4K).

Fig 4.

Phenotypic changes in RBCs and decrease in hemosiderin accumulation in the spleen and liver of transplanted THAL mice. Blood smear of a THAL mouse transplanted with the control GFP (A) or [β globin gene/LCR] lentivirus (B)-transduced bone marrow cells 7 months before analysis. (C) Blood smear obtained 6 months after the transplantation of THAL mice with marrow from a primary mouse reconstituted for 2 months with [β globin gene/LCR] lentivirus-transduced cells. (D) Blood smear prepared from a normal, unmanipulated adult B6 mouse. (E, G, I, and F, H, J) Perls staining of spleen and liver. (E and F) The spleen and liver of a representative, unmanipulated THAL mouse; (G and H) spleen and liver of a [β globin gene/LCR] corrected THAL mouse killed 6 months after transplantation; and (I and J) spleen and liver of an unmanipulated control B6 mouse. Arrows shown in H indicate cells in which mild iron accumulation was observed. (K) Percoll-Larex continuous density gradient analysis of RBCs. Lanes 1–4, primary recipients; lanes 5 and 6, secondary recipients of [β globin gene/LCR] lentivirus-transduced THAL bone marrow cells. (L) Spleens from unmanipulated THAL and B6 mice, and THAL recipients of GFP or [β globin gene/LCR] lentivirus-transduced THAL bone marrow.

A minimal correction took place in the hematological parameters as well as the degree of anemia of mice in which only ≈30% of RBCs were human β globin⁺ after transduction with a lower-titer virus stock. In these mice, Hb concentration increased by 1 g, and only 30–40% of the RBCs became normocytic and normochromic (data not shown).

Reversal of Disease Pathology in Spleen and Liver of Thalassemic Recipients.

Recipients of [β globin gene/LCR] lentivirus-transduced bone marrow cells showed further evidence of a marked improvement in ineffective erythropoiesis. Spleen weight for B6, Thal/GFP, and Thal/Lenti-β^A-treated mice was 110, 610, and 110 mg, respectively (Fig. 4L). An increase in the number of mature RBCs was observed in the red pulp of the spleens of Thal/Lenti-β^A mice as compared with unmanipulated THAL mice (data not shown). Moreover, the mature/immature nucleated red cell ratio was reduced from 20:80 in unmanipulated Thal mice to 40:60 in Thal/Lenti-β^A mice. These values corresponded well with the proportion of circulating normochromic normocytic red cells. The red pulp in the Lenti-β^A-treated mice was only moderately expanded, compared with the conspicuous expansion of the red pulp in control THAL mice.

Untreated THAL mice showed significant extramedullary erythropoiesis in the liver, whereas in the corrected THAL mice, extramedullary erythropoiesis was mild to moderate (data not shown). No erythropoiesis was observed in the livers of normal B6 mice. Perls iron staining (Fig. 4 E–J) showed decreased iron accumulation in the recipients of [β globin gene/LCR] lentivirus transduced marrow, thus providing further evidence of reduced destruction of RBCs and improved erythropoiesis. The Perls iron staining was markedly reduced in the spleen and was negligible in the liver of the transplanted mice (Fig. 4 G and H). In contrast, THAL mice had pronounced accumulation of hemosiderin in both the spleen and the liver (Fig. 4 E and F). Normal B6 mice had mild hemosiderin only in the spleen, and the liver was negative (Fig. 4 I and J).

Discussion

By using a lentiviral-based vector to deliver an expression cassette for human β globin, we have achieved reproducible, persistent, and long-term correction of ineffective erythropoiesis and nearly complete cure of the thalassemic phenotype in a mouse model of β-thalassemia. Significantly, this correction was possible in the absence of preselection for transduced cells before transplantation and was associated with essentially complete reconstitution by genetically modified stem cells that resulted in stable, high-level, pancellular expression of the human β globin gene in erythroid cells. These results constitute a significant advance over results previously obtained by using retroviral-based gene delivery systems and probably reflect several important features of the lentivector design leading to enhanced efficiency of gene transfer and ultimately to increased expression of the transduced β globin gene. The lentiviral vector incorporated several elements to enable production of stable, high-titer virus including a central polypurine tract/DNA flap element and rev-responsive element of HIV. After concentration, viral preparations with titers exceeding 10⁹/ml were achieved with a vector carrying an unmodified human β globin gene and extensive regions of the locus control region including elements of HS2, 3, and 4 without any evidence of viral instability through abnormal splicing. The subsequent infection of murine bone marrow cells at high MOI resulted in essentially 100% gene transfer to repopulating cells with multiple proviral integrations per transduced cell. The high gene-transfer efficiency, multiple-copy integrations, and incorporation of extensive globin gene regulatory sequences together probably account for the superior expression as evidenced by both pancellular erythroid expression of the human β globin gene at levels sufficient to correct the α globin protein imbalance and overcome the ineffective erythropoiesis seen in this thalassemia model. The long-term correction of anemia was also associated with dramatic reversal of iron accumulation at the sites of extramedullary hematopoiesis, such as spleen and liver.

Our findings add to the growing evidence that lentiviral vectors offer significant advantages as a platform for developing gene therapy of hemoglobinopathies. May et al. (15) have also recently demonstrated a significant and persistent amelioration of thalassemic phenotype in another mouse model of β-thalassemia (Hbb^th3/+), although some persistence of heterocellular expression of the human β globin gene transferred was noted in that study. Our results suggest further modifications to vector design and/or infection conditions may prove useful to achieve adequate copy number, infection efficiency, and expression level to ensure therapeutic efficacy. Indeed, by using a lower viral titer stock, we observed lower integration copy number, averaging 1 per stem cell, human β globin expression in only 30% of repopulated erythroid cells, and no improvement in any hematopoietic parameters. This finding is in contrast to previous estimates from transgenic mouse models that chimerism at levels of ≈20% of RBC with “normal” β globin expression is sufficient for cure (26, 27). In an effort to decrease further position effect variegation even at single proviral copy, we are now incorporating chromatin insulators at various locations of the [β globin gene/LCR] lentiviral vector (28).

Overall, these results establish, in a murine model, the feasibility of reproducibly achieving essentially complete correction of thalassemia by using a lentiviral-based strategy of gene transfer to transplantable hematopoietic stem cells. These findings provide further support for moving this platform to the treatment of human hemoglobin diseases.

Note Added in Proof.

In addition to having the effect of narrowing the range of position effect variagation, the presence of more than one integrated proviral copies may also decrease fluctuations in gene expression from cell to cell that arise from stochastic microscopic events inherent to the transcriptional process (intrinsic noise) (29).

Abbreviations

• LCR, locus control region

• FACS, fluorescence-activated cell sorter

• HSC, hematopoietic stem cell