Restoration of β-Globin Expression with Optimally Designed Lentiviral Vector for β-Thalassemia Treatment in Chinese Patients

Abstract

β-Thalassemia is one of the most prevalent genetic diseases worldwide. The current treatment for β-thalassemia is allogeneic hematopoietic stem cell transplantation, which is limited due to lack of matched donors. Gene therapy has been developed as an alternative therapeutic option for transfusion-dependent β-thalassemia (TDT). However, successful gene therapy for β-thalassemia patients in China has not been reported. Here, we present the results of preclinical studies of an optimally designed lentiviral vector (LV) named LentiHBB^T87Q in hematopoietic stem and progenitor cells (HSPCs) derived from Chinese TDT patients. LentiHBB^T87Q was selected from a series of LVs with optimized backbone and de novo cloning strategy. It contains an exogenous T87Q β-globin gene (HBB^T87Q) driven by a specific reconstituted locus control region, and efficiently expresses HBB mRNA and HBB protein in erythroblasts derived from cord blood HSPCs. To facilitate clinical transformation, we manufactured clinical-grade LentiHBB^T87Q (cLentiHBB^T87Q) and optimized its transduction procedure. Importantly, transduction of cLentiHBB^T87Q restored expression of HBB monomer and adult hemoglobin tetramer to relatively normal level in erythroblasts from bone marrow HSPCs of Chinese TDT patients that carry the most common mutation types and cover various genotypes, including β0/β0. Furthermore, viral integration sites (VISs) of cLentiHBB^T87Q were similar to other LVs safely used in previous clinical trials, and gene-ontology (term) analysis of VIS targeted genes suggests that no tumor-associated pathways were enriched in treated samples. Taken together, we have engineered the cLentiHBB^T87Q that can restore β-globin expression in the HSPCs-derived erythroblasts of Chinese TDT patients with minimal risk of tumorigenesis, providing a favorable starting point for future clinical application.

Introduction

β-thalassemia is the most prevalent hereditary blood disorder found in Mediterranean, Middle Eastern, Indian, South China, and Southeast Asian.^1,2 The patients in developing countries have shortened life expectancy and suffer from severe morbidity because of limited supportive therapies, imposing an enormous financial burden to the society and the patients' families.³ So far, the clinically available curative therapy for β-thalassemia is allogeneic hematopoietic stem cell transplantation (HSCT). However, for most patients, it is challenging to find the human leukocyte antigen (HLA)-matched donors.⁴ Recently, although the donor availability has been largely improved with the development of haploidentical HSCT (haplo-HSCT), patients receiving haplo-HSCT, particularly for those >7 years old, still experience life-threatening complications due to graft versus host disease (GVHD).^5,6 Meanwhile, haplo-HSCT donors are usually patients' parents with one mutated β-globin (HBB) allele showing a benign anemia phenotype, which may affect the patients' final outcome and life quality after transplantation.

Gene therapy for β-thalassemia either activates γ-globin through nuclease-mediated genome editing⁷ or introduces functional β-globin through lentiviral transduction to compensate loss of function of HBB. ^8,9 The strategy to activate γ-globin has entered clinical trial only in these 2 years, and requires more data and longer time to confirm its safety and efficacy. In contrast, reintroduction of β-globin with lentiviral vector (LV) has been successfully applied in clinical trials for over two decades. Therefore, reintroduction of β-globin is the focus of this article and referred to as gene therapy henceforward for convenience. Since gene therapy uses patients' own hematopoietic stem and progenitor cells (HSPCs), donor availability and GVHD risk are largely avoided, providing an attractive therapeutic option.^10
–12 Moreover, gene therapy with an efficient transduction and HBB expression could theoretically restore HBB to a normal level, and may result in better clinical outcome when compared with γ-globin activation or haplo-HSCT.

Previous studies have shown successful correction of β-thalassemia in animal models using different LVs.¹³ The first LV entering clinical trial was known as LentiGlobin, which uses a reconstituted locus control region (rLCR) to express HBB with a T87Q mutation (HBB^T87Q) that functions as wild-type allele but meanwhile has antipathologic characteristic in sickle cell disease (SCD).^14,15 In phase I/II clinical trials, 12 of 13 patients with β0/β⁺ genotype and 3 of 9 patients with β0/β0 genotype became blood transfusion independent (TI) after gene therapy. The other six of β0/β0 patients showed a significant reduction (average ∼73%) in annual transfusion volume. Consistent with the proposed mechanism, the overall clinical outcome was positively correlated with the inserted vector copy number (VCN) and exogenously expressed adult hemoglobin (HbA^T87Q) characterized by a T87Q amino acid substitution.¹⁶ Therefore, a higher VCN expressing sufficient HbA^T87Q is hypothesized to cure most β-thalassemia patients, including those with β0/β0 genotype. Indeed, in the latest phase III trial (NorthStar-III), with improved LV quality the mean VCN was increased to ∼2.5 and the mean HbA^T87Q was >9.5 mg/dL after gene therapy, enabling 75% of β0/β0 patients to become TI for >3 months.⁸ Up to date, LentiGlobin has treated >40 transfusion-dependent β-thalassemia (TDT) patients, and the earliest treated patients have been TI for over 36 months. Based on these promising data, gene therapy drug (Zynteglo) has been approved for treatment of non β0/β0 TDT patients over 12 years old in Europe.¹⁷ However, the market price of this drug is 1.8 million dollars, which is too expensive for most TDT patients in developing countries.

In another clinical trial,⁹ GLOBE LV drives exogenous HBB expression with a shorter LCR and, after gene therapy, six of seven treated patients decreased their blood transfusion volume dramatically, and three of them became TI for over 1 year. Interestingly, this study found that gene therapy was more effective in pediatric patients but did not increase the risk when compared with adult patients, suggesting that it is ethical and reasonable to recruit pediatric and juvenile patients in future clinical trials.⁹ Compared with γ-globin activation, which only went into clinical trials in these 2 years and requires more data and longer time to confirm its safety and efficacy, reintroduction of β-globin with LV has been successfully implied in clinical trials for over two decades.

The aforementioned independent clinical trials have strongly suggested that gene therapy is promising for β-thalassemia treatment. Based on these clinical studies, the key to a successful gene therapy is to safely induce sufficient HBB expression, which is regulated by rLCR. The endogenous HBB LCR region is ∼80 kb in which the identified critical regions contain DNase I hypersensitive sites (HSs), including HS1, HS2, HS3, and HS4, ranging from several hundreds of base pairs to >2 kb.¹⁸

Different rLCRs have been constructed with various HS combinations to drive HBB expression.^13,19

–22 Recently, Morgan et al. have developed a new system called LVMPRA (lentiviral vector-based, massively parallel reporter assay) to identify the boundaries of previously unknown intrinsic enhancer sequences of the human β-globin LCR and assembled novel therapeutic βAS3-globin LVs.^23,24 Existing clinically applied LVs including LentiGlobin and GLOBE are effective, but cannot help all TDT patients to get rid of transfusion, suggesting that there is still room for improvement.

Meanwhile, previous studies and clinical trials were mostly conducted in Caucasian and southeast Asian patients with genetic mutations of CD39, CD41/42, IVS-I-110, and HbE.^9,16 The most popular mutation types in south China including CD17, CD71/72, and IVS-II-654²⁵ were not tested yet, and different genetic backgrounds among populations may be a complex factor for gene therapy. Therefore, we aim to develop an affordable gene therapy for Chinese TDT patients. Here, by optimizing rLCRs in the third-generation self-inactivating LV, we obtained and manufactured clinical-grade LentiHBB^T87Q with robust HBB expression. With improved transduction procedures, we confirmed its efficacy in Chinese TDT patient samples and evaluated its safety profile regarding viral integration sites (VISs), laying a foundation for further advancing this gene therapy to clinical studies in China.

Materials and Methods

Ethics approval and consent to participate

Permission for this study was obtained from the Bioethics and Biological Safety Review Committee of BGI-Shenzhen (No. 16089-T2), the Medical Ethics Review Committee of Shenzhen Second People's Hospital (No. 20180515006), Shenzhen Children's Hospital (No. 201900203), and First Affiliated Hospital of Guangxi Medical University (No. 2019001). We confirm that all experiments were performed in accordance with relevant guidelines and regulations.

LV design, manufacture, and titration measurement

The self-inactivating (SIN) vector of LentiHBB contains a CMV promoter and enhancer, and deleted the U3 region of HIV in the 5′ LTR. The expression cassette consists of β-globin locus control region (LCR) and HBB minigene. LCR is composed of regions containing DNase I HSs, named HS2, HS3, and HS4. HBB minigene comes with a short deletion of intron 2 and an antisickling amino acid substitution (T87Q). The LCR and HBB sequence information will be provided upon request. All of the DNA fragments were commercially synthesized (Beijing Liuhe, Co. Ltd., Guangzhou, China) and cloned into the LentiHBB vector using the Gibson assembly (NEB, E2611L).

LVs of laboratorial grade were produced by cotransfection of HEK293T cells with the transfer plasmid LentiHBB, the packaging plasmid psPAX2 and the envelope plasmid pMD.2G. The pseudotyped virions were collected from the clear supernatant and concentrated by ultracentrifugation at 80,000 g for 2 h as previously described.²⁶ The clinical-grade LentiHBB^T87Q vector was manufactured according to good manufacturing practices (GMP) by OBiO Technology (Shanghai) Corp. Ltd.

The infectious titer of lentivirus was calculated by transducing HEK293T cells cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, glutamine, and polybrene (8 μg/mL). HEK293T cells were transduced with serial diluted lentivirus. The supernatant was replaced with fresh medium 24 h after transduction, and genomic DNA (gDNA) was extracted at the third day. VCN was measured by Real-time quantitative PCR (qPCR) as described below. Titer was calculated multiplying the VCN/cell with the number of transduced cells and then normalized with dilution times, and was presented as transforming units per milliliter (TU/mL).

HSPCs culture and lentiviral transduction

Cord blood (CB) CD34⁺ cells were purified with Human Cord Blood CD34 Positive Selection Kit II (17896; StemCell Technologies) and cultured in SFEM II medium (09605; StemCell Technologies) supplemented with CD34⁺ Expansion Supplement (02691; StemCell Technologies). Bone marrow (BM) and mobilized peripheral blood (mPB) CD34⁺ cells were purified with ClinMACS CD34 MicroBead Kit (130-046-702; Miltenyi) and cultured in SCGM (20802-0500; CellGenix) supplemented with 100 ng/mL recombinant human cytokines thrombopoietin (102-06; PrimeGene), 100 ng/mL fms-related tyrosine kinase 3 ligand (103-05; PrimeGene), 100 ng/mL stem cell factor (102-01; PrimeGene), and 25 ng/mL IL3 (101-03; PrimeGene). CD34⁺ cells for transduction were cultured at a concentration of 1 × 10⁶ to 2 × 10⁶ cells/mL at 37°C in a humidified incubator of 5% CO₂.

For lentiviral transduction, CD34⁺ cells were prestimulated in culture medium for 24 h, and then transduced with lentivirus for 24 h. The enhancers of protamine (P4020; Sigma), dmPGE2 (39746-25-3; Cayman), UM171 (72914; StemCell Technologies), and Poloxamer407 (9003-11-06; BASF) were added at indicated concentration immediately before the addition of lentivirus.

Methylcellulose colony forming unit assay

After lentiviral transduction, ∼2,000 HSPCs were plated into Methocult Classic H4434 (04434; StemCell Technologies) following the manual. After 14 days of culture, the number of colony forming units (CFUs)-erythroid, burst forming unit-erythroid (BFU-E), and CFU-granulocyte/erythroid/macrophage/megakaryocyte, CFU granulocyte/monocyte colonies were determined and scored based on colonies morphology, and individual BFU-E or pooled colonies were collected for VCN measurement and HBB expression analysis.

Erythroid differentiation assay

CD34⁺ cells were washed with Dulbecco's phosphate buffered saline (DPBS) after lentiviral transduction or control treatment, and transferred to SFEMII medium (09605; StemCell Technologies) supplemented with erythroid Expansion Supplement (02692; StemCell Technologies). Erythroid differentiation medium was changed every 2 days, and differentiated erythroblasts were collected for qPCR or high-performance liquid chromatography (HPLC) analysis from days 7 to 14.

gDNA extraction and VCN measurement

gDNA of LV transduced cells was extracted using the TIANamp Genomic DNA Kit (Tiangen, Beijin, China) following the manufacturer's manual. gDNA of BFU colonies was extracted by suspending the cell in 1 × PCR buffer (R050A; Takara) containing 0.2 mg/mL proteinase K (RT403; Tiangen), and then incubating them at 55°C for 20 min, 95°C for 5min. The extracted gDNA can be stored at 4°C for short-term usage or stored at −20°C for long-term storage.²⁷

VCN per diploid genome was determined by duplex TaqMan qPCR, using Premix Ex Taq kit (RR390A; Takara) and the StepOnePlus Real-Time PCR System (ABI), following the manufacturers' manuals. A plasmid (pMD18T-GAG-APOB) containing one copy of each GAG and APOB fragment was used to generate a standard curve, and copies of GAG and APOB were determined using this standard curve. The calculation of VCN (per diploid genome) was provided as follows¹⁶: VCN = [copies of GAG] ÷ [copies of APOB] × 2.

qPCR Primers and probe sequences used for APOB and GAG were as follows:

APOB qPCR F: 5-TGAAGGTGGAGGACATTCCTCTA-3;

APOB qPCR R: 5-CTGGAATTGCGATTTCTGGTAA-3;

APOB probe: 5-VIC-CGAGAATCACCCTGCCAGACTTCCGT-3-TAMRA;

GAG qPCR F: 5-GGTTGTAGCTGTCCCAGTATTTGTC-3;

GAG qPCR R: 5-GGAGCTAGAACGATTCGCAGTTA-3;

GAG probe: 5-FAM-ACAGCCTTCTGATGTTTCTAACAGGCCAGG-3-TAMRA.

Total RNA extraction and quantitative RT-PCR

Total RNA was extracted from HSPCs, erythroblasts, or BFU-E using TRIzol reagent (Life Technology, CA) following the manual instruction. RNA concentration and purity were determined by NanoDrop 2000 Spectrophotometer (Thermo Scientific). DNA contamination was eliminated with gDNA Eraser (RR047A; Takara), and reverse transcription of mRNA was conducted using the PrimeScript™ RT reagent Kit (RR047Q; Takara) with oligo(dT) primers.

Relative quantification PCRs were performed using TB Green Premix Ex Taq kit (RR820B; Takara) and the StepOnePlus Real-Time PCR System (ABI) following the manufacturer's manuals. The expression of HBB^WT and HBB^T87Q was calculated according to the 2^−ΔΔCt method using GAPDH as the internal reference control as previously described.²² Primers used for SYBR green qPCR were as follows: to measure GAPDH mRNA: GAPDH qPCR F: 5-ACCCACTCCTCCACCTTTGA-3; GAPDH qPCR R: 5-CTGTTGCTGTAGCCAAATTCGT-3; to measure total HBB mRNA: HBB qPCR F: 5-GAAGTCTGCCGTTACTGCCC-3; HBB qPCR R: 5-AGCCTTCACCTTAGGGTTGC-3; to measure relative expression percentage of HBB and HBBT^87Q : HBB^WT F: 5-TCAAGGGCACCTTTGCCACA-3; HBB^T87Q F: 5-TCAAGGGCACCTTTGCCCAG-3 and HBB^T87Q&WT R: 5-AATTCTTTGCCAAAGTGATGGG-3.

HPLC analysis of globin and hemoglobin

Hemolysates were made by lysing 10⁵ erythroblasts or BFU-E cells in 50 μL HPLC-grade water for 10 min. After centrifugation at 13,200 rpm for 10 min, 20 μL of the supernatant of hemolysates was used for HPLC analysis. The type of globin chains and hemoglobin tetramers was characterized using Aeris™ 3.6 μm WIDEPORE C4, 200 Å, 150 × 4.6 mm LC Column (00F-4486-E0; Phenomenex) at 220 nm, and 100 × 4.6 mm cation exchange column (104CT0315; Phenomenex) at 418 nm, respectively, on a HPLC machine (Agilent 1260 Infinity II). Globin samples were eluted with a gradient mixture of solution A (water/acetonitrile/trifluoroacetic acid, 700:300:0.7) and solution B (water/acetonitrile/trifluoroacetic acid, 450:550:0.5). Hemoglobin was eluted with a gradient mixture of solution A (Bis-Tris 40 mM, NaN₃ 2 mM; pH adjusted at 6.5 with acetic acid) and solution B (Bis-Tris 40 mM, NaN₃ 2 mM, NaCl 200 mM; pH adjusted at 6.8 with acetic acid) without degas program. Analysis and peak integration were performed using Open LAB CDS Chem station software (Agilent). Lysates of CB RBCs and adult peripheral blood RBCs from healthy donors were used as controls.

Identification of VIS

The genomic sequences flanking VIS were identified by linear amplification-mediated PCR (LAM-PCR) method and next-generation sequencing (NGS).^28,29 In brief, gDNA from cLentiHBB^T87Q transduced cells was fragmented by sonication using a Covaris E220 Ultrasonicator. The fragmented DNA was then subjected to end repair and 3′ adenylation, ligated to a custom adaptor include a 10 bp random unique molecular identifier and a specific sequencing adaptor for BGI-Seq 500 platform. The viral LTR-host gDNA junctions were amplified by a seminested PCR using LTR-specific primers and the adaptor-specific primers. The second-round adaptor-specific primer contains a unique 10-nucleotide barcode, which indexes the amplification products. The PCR products were purified with AmpureXP beads, pooled in equimolar ratios, subjected to DNA nanoball generation, and then sequenced using the BGI-seq 500 platform.

Bioinformatics analysis of VIS data

Raw reads were filtered by SOAPnuke³⁰ with the parameters “-l 15 -q 0.2-n 0.05.” Reads were aligned to the lentivirus sequences to remove vector backbone sequences, and then mapped to plasmid sequences to remove potential plasmid contamination. Finally, the remaining reads were aligned to the human reference genome (GRCh38.p12) using HISAT2.³¹ Samtools³² was used to sort and index the alignment files for visualization using the Integrative Genomics Viewer tool.³³ The sorted binary sequence analysis file (BAM files) was also used to generate WIG files using BEDTools.³⁴ To this end, coverage files were normalized using the total signal for each sample. We generated a FASTA file of insertion sites from genome sequences (GRCh38.p12) using the getfasta function from BEDTools³⁴ and used ggseqlogo³⁵ R packages (version 0.1) to create 15 bp sequence logos (5 upstream and 10 downstream) from the FASTA files. Insertion sites and sequence information was used as input files for functional annotation using ANNOVAR.³⁶ We selected the insertion sites located in gene exons. Gene-ontology (GO) enrichment analysis of the gene sets was then performed on each sample using the clusterProfiler R package.^37,38

Results

LV backbone selection and optimization

We selected the third-generation LV backbone with proven higher safety as it will specifically be packaged into functional lentivirus guided by its three assistant vectors.³⁹ To improve the lentiviral titers, we tested replacing its endogenous RSV promoter with CMV promoter.^39,40 In agreement with previous studies, we found that LV containing CMV promoter always resulted in a higher titer (Supplementary Fig. S1A). In accordance, CMV version of LentiHBBs induced a higher expression level of β-globin (HBB) when transduced into K562 cells, as it produces a higher VCN (Supplementary Fig. S1B). Thus, the LV backbone containing the CMV promoter was used in the following experiments (Fig. 1A).

Figure 1.

Construction and comparison of LVs containing different rLCRs in CB HSPCs. (A) Diagram of our LV backbone, which contains a CMV promoter and enhancer instead of the endogenous HIV promoter at the 5′ LTR, a deleted U3 region and a reversely inserted rLCR-HBB; (B) diagram of different rLCR-HBBs constructed into LentiHBB101 to LentiHBB107; (C) percentages of HBB^T87Q mRNA and VCN in BFU-E colonies from CB HSPCs transduced with different LVs; (D) statistics of HBB^T87Q versus VCN in BFU-E colonies shown in (C); (E) percentages of HBB^T87Q mRNA in differentiated erythroblasts from CB HSPCs transduced with LentiHBB103, LentiHBB106, and LentiHBB107; (F) statistics of HBB^T87Q versus VCN in differentiated erythroblasts shown in (E). CB1 and CB2 are two independent samples. BFU-E, burst forming unit-erythroid; CB, cord blood; CMV, cytomegalovirus; HSPCs, hematopoietic stem and progenitor cells; LTR, long terminal repeat; LVs, lentiviral vectors; rLCR, reconstituted locus control regions; VCN, vector copy number. Color images are available online.

Construction of LVs containing different rLCRs

The locus control region (LCR), critical for HBB regulation in vivo, contains four critical regions harboring DNase I HS. Prior studies using different rLCRs with different combinations of HSs confirmed that most of them can drive HBB expression. Therefore, we designed different rLCRs followed by HBB (rLCR-HBBs), cloned them into our CMV version of LV. To avoid blockade of lentiviral RNA transcription by HBB polyA, rLCR-HBBs were reversely inserted into the lentiviral backbone (Fig. 1A). LentiHBB101 and LentiHBB102 contained the same 2.9 kb rLCR,⁴¹ and a 562 bp deletion in the second intron of HBB was introduced in LentiHBB101 but not in LentiHBB102. LentiHBB103 to LentiHBB107 contained different rLCRs followed by the same HBB region that contains a 372 bp deletion in the second intron (Fig. 1B).^13,19

–22 Neither 562 bp deletion nor 372 bp deletion affected the diseasing-causing mutations located in the second intron.⁴² To avoid introducing other irrelevant sequences, which was the cases in LentiHBB101 and LentiHBB102, new rLCR-HBBs in LentiHBB103 to LentiHBB107 were synthesized de novo and then seamlessly cloned into LVs using Gibson Assembly.⁴³ Furthermore, the threonine at the 87th amino acid position was mutated to glutamine (T87Q) in HBB second exon. This mutated isoform (HBB^T87Q) showed a stronger antisickling property and was used to treat SCD.^14,15 More importantly, HBB^T87Q mRNA levels can be quantitatively distinguished from the endogenous HBB mRNA using quantification PCR with sequence-specific primers.

Expression of LVs in erythroblast cells derived from CB HSPCs

To measure exogenous HBB^T87Q expression levels, LentiHBBs were introduced to CB HSPCs. Transduced HSPCs were plated on methylcellulose to form single colonies, which were picked up for measurement of HBB expression and VCN. Percentages of each type of colonies were also calculated, and results indicated that transduction of different LentiHBBs did not affect the polyclonal ability of CB HSPCs (Supplementary Fig. S2A). LentiHBB101 and LentiHBB102 produced a lentiviral titer <1 × 10⁸ (Supplementary Fig. S1A) and induced a negligible expression of exogenous HBB^T87Q in BFU-E derived from the CB HSPCs (Fig. 1C, data not shown). In contrast, LentiHBB103 to LentiHBB107 induced considerable HBB^T87Q expression in BFU-E and showed a high percentage of positive BFU-E transduction up to 92.3.

The titers of LentiHBB103 to LentiHBB107 are shown in Supplementary Fig. S2B. We noticed that a higher titer was associated with smaller size of vector. To compare their expression efficiency, we measured the exogenous HBB^T87Q mRNA percentage (Fig. 1C), and then normalized the HBB^T87Q percentage to the inserted VCN in colonies of each group (Fig. 1D). We found that LentiHBB103, LentiHBB106, and LentiHBB107 were the top 3 LVs presenting efficient production of HBB^T87Q mRNA, with LentiHBB103 showing the least variation (Fig. 1D). We also found that LentiHBB103 had a higher correlation factor between HBB^T87Q percentage and its VCN (Supplementary Fig. S2C), suggesting that LentiHBB103 documents consistent HBB expression.

To differentiate HSPCs into erythroblasts, we cultured the transduced CB HSPCs in erythroid medium for 2 weeks, a procedure validated to induce erythroid differentiation, and then measured the expression levels of HBB^T87Q mRNA in these pooled and differentiated cells. We found that the exogenous HBB^T87Q mRNA expression level was sample dependent and positively correlated with the multiplicity of infection (MOI) (Fig. 1E). Compared with LentiHBB106 and LentiHBB107, LentiHBB103 induced higher expression levels of HBB^T87Q mRNA after VCN normalization in two independent CB samples (Fig. 1F). Thereafter, we chose LentiHBB103 to perform the following studies, and renamed it as LentiHBB^T87Q. In summary, we generated a series of LVs containing different rLCRs, tested their HBB^T87Q expression in erythroblasts derived from CB HSPC samples, and identified the LentiHBB^T87Q with the highest efficacy and consistency.

LentiHBB^T87Q express HBB protein in a dosage-dependent manner

We further investigated whether the expressed HBB mRNA can be translated into HBB protein and form the hemoglobin tetramer in the differentiated erythroblasts. We transduced CB HSPCs with LentiHBB^T87Q with MOI25, MOI50, and MOI100 and induced their erythroid differentiation. Compared with the untransduced control, the transduced samples showed an increased level of total HBB mRNA. VCN and HBB^T87Q mRNA percentages were positively correlated with MOI values (Fig. 2A). We also collected cell lysates for HPLC analysis of globin monomer and hemoglobin tetramer. In untransduced control group, CB HSPC-derived erythroblasts expressed high levels of α-globin and γ-globin, but negligible β-globin (Fig. 2B). In agreement, the ratio of adult hemoglobin (HbA) to fetal hemoglobin (HbF) was as low as 0.16 in control group (Fig. 2B–D). In contrast, the erythroblasts differentiated from the LentiHBB^T87Q transduced CB HSPCs expressed a considerable amount of β-globin. When the MOI and percentages of HBB^T87Q mRNA were higher, the ratio of β-globin to α-globin was increased, and the corresponding ratio of γ-globin to α-globin was decreased. In agreement, HbA to HbF ratio increased with the increase of MOI gradient and VCN (Fig. 2B–D).

Figure 2.

LentiHBB^T87Q increases HBB^T87Q mRNA and HBB protein expression in erythroblasts derived from CB HSPCs in a dosage-dependent manner. (A) The expression level of total HBB mRNA and the percentage of HBB^T87Q in erythroblasts differentiated from CB HSPCs transduced with LentiHBB^T87Q; total HBB mRNA in transduced samples was normalized to control; (B) the globin monomer and hemoglobin tetramer of erythroblasts were determined by HPLC; green arrows indicate the β-globin monomer and blue arrows indicate HbA tetramer; (C) linear regression analysis among HBB^T87Q percentage, HbA/HbF, and VCN; (D) the statistic results of HPLC in (B). HbA, adult hemoglobin; HbF, fetal hemoglobin; HPLC, high-performance liquid chromatography. Color images are available online.

The statistic and regression analysis showed that HBB^T87Q mRNA percentage and HbA/HbF ratio are positively correlated with VCN (Fig. 2C, D). The high R ² value (>0.9) indicates that lentiviral transduction-mediated HBB^T87Q expression increases HbA/HbF ratio. In conclusion, our studies show that LentiHBB^T87Q efficiently and dose-dependently generates HBB protein in erythroblasts derived from CB HSPCs.

Optimization of lentiviral transduction in mPB HSPCs and validation of the optimized parameters with clinical-grade LentiHBB^T87Q

For successful clinical application of gene therapy, it is critical to increase LV transduction efficiency while maintaining the cell stemness. Thus, it is important to optimize the ex vivo cell manipulation procedures, including the culture system, cell activation time, MOI value, and the transduction enhancers. As mPB HSPCs are typically used for β-thalassemia gene therapy, we thus collected mPB HSPCs from a healthy carrier of the 17M mutation to optimize the procedures for lentiviral transduction.

First, we tested different combinations of cytokines, and determined their influence on transduction efficiency of LV and differentiation potential of mPB HSPCs. Among the tested three combinations, the one with the lowest content of cytokines showed the highest VCN and maintained the highest CD34⁺ percentage after LV transduction (Fig. 3A). Second, we assessed distinct transduction enhancers, including protamine, dmPGE2, UM171, and Poloxamer407,^44
–46 and tested their individual or combinational effects on LV transduction in mPB HSPCs. We found that dmPGE2 and Poloxamer407 combination (PGE–Pol) was the most efficient, which increased the VCN by six to seven times when compared with enhancer-free control (Fig. 3B). Meanwhile, PGE–Pol combination also showed the highest efficacy at different cell concentrations (Fig. 3C) ranging from 1.0 × 10⁶ to 2.0 × 10⁶, suggesting that cell concentration had minimal effect on the transduction efficiency.

Figure 3.

Optimization of lentiviral transduction in mPB HSPCs and validation of the optimized parameters with clinical-grade LentiHBB^T87Q (cLentiHBB^T87Q) at different scales. (A) Results of CD34⁺ percentage and VCN after lentiviral transduction, when mPB HSPCs were cultured in serum-free medium with different cytokine combinations; Combination A: 100 ng/mL TPO, 100 ng/mL FltL, 100 ng/mL SCF, and 25 ng/mL IL3; Combination B: 10 ng/mL TPO, 300 ng/mL FltL, 300 ng/mL SCF, and 75 ng/mL IL3; Combination C: 100 ng/mL TPO, 300 ng/mL FltL, 300 ng/mL SCF, and 75 ng/mL IL3; (B) results of VCN at 3 and 7 days after lentiviral transduction when mPB HSPCs were treated with different enhancer combinations during transduction. pro: protamine; UM: UM171; PGE: dmPGE2; pol: poloxamer407; (C) results of enhancer combinations on transduction efficiency when mPB HSPCs were cultured at different cell concentrations during transduction process; (D–F) validation of transduction efficiency with clinical-grade LentiHBB^T87Q (cLentiHBB^T87Q) at 1.0 × 10⁵ cell scale; (D) VCN and HBB^T87Q mRNA percentage were positively correlated with MOI in differentiated erythroblasts at 1.0 × 10⁵ cell scale; (E) average VCN was positively correlated with MOI; (F) linear regression of HBB^T87Q percentage with VCN was obtained in BFU-E colonies that were derived from mPB HSPCs transduced at 1.0 × 10⁵ cell scale; (G–M) validation of transduction efficiency with cLentiHBB^T87Q at 8.5 × 10⁶ cell scale; (G) VCN and HBB^T87Q mRNA percentage were positively correlated with MOI in differentiated erythroblasts at 8.5 × 10⁶ cell scale; (H, I) HPLC and corresponsive statistic data of hemoglobin analysis in differentiated erythroblasts as shown in (G); PB2 is same as PB sample, but run again in the end; (J, K) average VCN was positively correlated with MOI and linear regression of HBB^T87Q percentage with VCN was observed in BFU-E colonies that were derived from mPB HSPCs transduced at 8.5 × 10⁶ cell scale; (L, M) results of CD34⁺ percentages and percentages of different colonies in CFU assay after transduction of cLentiHBB^T87Q at 8.5 × 10⁶ cell scale. FltL, fms-related tyrosine kinase 3 ligand; MOI, multiplicity of infection; mPB, mobilized peripheral blood; ND, not determined; SCF, stem cell factor; TPO, thrombopoietin. Color images are available online.

Using the PGE–Pol enhancers, we transduced mPB HSPCs with a gradient of MOI, and measured the VCN and HBB^T87Q mRNA percentage. The results indicated a positive correlation of VCN and HBB^T87Q mRNA percentage with MOI (Fig. 3D), and a positive correlation of HBB^T87Q mRNA percentage with VCN in BFU-E (Fig. 3E, F), suggesting that the usage of enhancers increased both VCN and the expression of HBB^T87Q mRNA in mPB HSPC samples. We also examined the cell activation time (from 16 to 48 h) and LV transduction time (from 24 to 48 h), and found no significant differences (data not shown).

Last, we outsourced LV manufacture and obtained clinical-grade LentiHBB^T87Q, named cLentiHBB^T87Q. To validate the optimized procedures in clinical settings, we transduced mPB HSPCs (∼1.0 × 10⁷) that was purified with ClinMACS using cLentiHBB^T87Q. The VCN up to 1.31 and HBB^T87Q mRNA percentage up to 56.39 were positively correlated with MOI after erythroid differentiation (Fig. 3G). In this experiment, the control sample was obtained from a donor with a 17M mutation, which is a nonsense mutation and may be responsible for the shift in HbA position. Nevertheless, after cLentiHBB^T87Q transduction, we observed a clear further shift of HbA position that was dependent on the increase of MOI, though HPLC analysis cannot distinguish HbA^T87Q from WT HbA. In contrast, HbF position was not affected by the transduction. Furthermore, HbA/HbF ratio was increased and positively correlated with MOI (Fig. 3H, I), suggesting that increased expression of HbA was achieved by transduction of cLentiHBB^T87Q.

In addition, increased VCN upon increased MOI and positive correlation of HBB^T87Q mRNA percentage with VCN were observed in BFU-E colonies from cLentiHBB^T87Q transduced groups (Fig. 3J, K). Viability and differentiation potential of mPB HSPCs, as determined by flow cytometry and CFU assays, were not altered by cLentiHBB^T87Q transduction (Fig. 3L, M). Taken together, we have optimized the critical parameters of cell manipulation for ex vivo gene therapy and validated them with clinical-grade LentiHBB^T87Q at a relatively large scale in mPB HSPCs, providing an entrance point for future therapeutic applications.

Restoration of globin and hemoglobin with cLentiHBB^T87Q transduction in the erythroblasts differentiated from the BM HSPCs of Chinese TDT patients

To test the effect of cLentiHBB^T87Q on compensating HBB expression, we collected BM samples from Chinese TDT patients (Fig. 4A, Supplementary Table S1). Genotypes of these patients are β⁺/β0 and β0/β0 with the most common mutation alleles in South China, including CD41–42, CD71–72, 17M, and 43M, mutations affecting mRNA translation, IVS-II-654, mutation affecting mRNA processing, and −29(A>G), mutation affecting mRNA transcription.

Figure 4.

cLentiHBB^T87Q restores HBB mRNA and HBB protein expression in erythroblasts derived from Chinese TDT patients' BM HSPCs. (A) Summary of the patients' information; (B) relative HBB mRNA expression level, VCN and HBB^T87 mRNA percentage after lentiviral transduction in erythroblasts derived from transduced BM HSPCs of Chinese TDT patients; genotype of each patient is listed on top of the panel; (C) HPLC results of globin monomer and hemoglobin tetramer of erythroblasts from BM HSPCs. The restored β^T87Q and HbA^T87Q are indicated by green and blue arrows, respectively. Corresponsive statistic results of each HPLC panel are shown as columns on the right; (D) VCN and HBB^T87Q mRNA percentage were positively correlated with MOI in TDT patients' sample; (E) HPLC results of globin and hemoglobin in erythroblasts derived from patients' HSPC transduced with different MOI gradients. BM, bone marrow; TDT, transfusion-dependent β-thalassemia. Color images are available online.

BM HSPCs of TDT patients were purified and transduced with cLentiHBB^T87Q at MOI = 20, following the procedures optimized in our aforementioned studies. All transduced samples have a VCN >1.3 and showed increased total HBB mRNA levels, in which exogenous HBB^T87Q mRNA amounted to >50% and up to 96.5% (Fig. 4B). HPLC results indicated that expression of β-globin was profoundly increased, and the ratios of β- to α-globin were restored from <0.1 to ∼1.0 in the β0/β0 samples transduced with cLentiHBB^T87Q. Meanwhile, in some patient samples, the increased ratio of γ- to α-globin was reduced (Fig. 4C). In addition, erythroblasts from the transduced BM HSPCs of TDT patients mainly expressed HbA^T87Q, and largely eliminated the expression of HbF and other abnormal hemoglobin polymers that dominated in the untransduced controls (Fig. 4C).

We also transduced a sample with cLentiBGI103 at lower MOIs of 5 and 10. The lower MOIs also resulted in efficient expression of HBB^T87Q mRNA and HBB protein (Fig. 4D), suggesting the high efficacy of cLentiHBB^T87Q and the robustness of our procedures. Taken together, our results suggest that cLentiHBB^T87Q can transduce HSPCs from Chinese TDT patients ex vivo, and a VCN more than ∼1.2 and a percentage of HBB^T87Q mRNA higher than ∼50 were sufficient to produce significant improvements.

Characterization of VIS

To determine the safety spectrum of cLentiHBB^T87Q, we analyzed the VIS in cLentiHBB^T87Q transduced mPB HSPCs on day 7 after transduction (MOIs of 5, 10, and 20; as shown in Fig. 3G). The gDNA sequences flanking the VIS were retrieved by LAM-PCR method (Fig. 5A), sequenced using BGI NGS platform and mapped on the human genome (GRCh38.p12). Overall, we identified a large amount of unique VIS from transduced mPB HSPCs. The VIS numbers ranged from 20,885 to 38,395, and were positively associated with MOI. Less than 5% overlapping sites were detected in different samples, suggesting that the insertion of cLentiHBB^T87Q was largely random (Fig. 5B, C). The top 10 most prevalent VISs account for <15% of total VISs, and contain 2 popular sites, which were also detected as the top 2 sites in a previous clinical study (Fig. 5D).⁴⁷ VIS distribution profiles at the chromosome level and centered on transcription starting site (TSS) showed that consistent preference of cLentiHBB^T87Q for integration in gene-dense regions but far from TSS (>5 kb), comparable with the patterns observed in gene-therapy-treated patient samples of previous clinical studies (Fig. 5E, F).^9,47,48 Consensus sequences of preferred insertion sites of cLentiHBB^T87Q were consistent between groups (Fig. 5G). Finally, GO enrichment analysis of the genes targeted by VIS for biological processes, cellular components, and molecular functions demonstrated no enrichment of tumor-related pathways (Fig. 5H). Collectively, our data indicate that the VIS of cLentiHBB^T87Q is similar to those of lentiviruses safely used in previous clinical studies, and based on our bioinformatics analysis, it did not increase the risk of tumorigenesis.

Figure 5.

Characterization of VIS of cLentiHBB^T87Q transduced mPB HSPCs. (A) Diagram of VIS identification process with LAM-PCR and NGS; (B) VIS numbers and overlaps among groups; (C) the VIS number is positively correlated with VCN; (D) top 10 integration sites in each group of cLentiHBB^T87Q transduced mPB HSPCs; (E) chromosome distribution profiles of VIS from samples transduced with cLentiHBB^T87Q and from patient samples transduced with other LVs in previous clinical trials; (F) distribution of VIS centered on TSS; (G) the consensus sequences of preferred insertion sites of cLentiHBB^T87Q and another LV in a previous clinical trial; data cited from published papers are labeled in green in panel (E–G); (H) gene-ontology analysis of VIS-targeted genes in a representative sample (mPB HSPC-20) for biological processes, cellular components, and molecular functions. LAM-PCR, linear amplification-mediated PCR; NGS, next-generation sequencing; TSS, transcription starting site; VIS, viral integration sites. Color images are available online.

Discussion

Gene therapy, as an attractive alternative option in addition to allogeneic HSCT, has been developed for β-thalassemia treatment. Two independent clinical studies indicated that the efficacy of gene therapy depends on exogenous HBB expression, which is driven by rLCR.^9,16 We designed a series of LCRs based on previous studies,^19,49,50 and constructed them with de novo synthesis and seamless Gibson assembly. Our results showed that rLCRs containing different DNase I HSs drove HBB expression with different efficiency, and the rLCR with the longest HSs (LentiHBB104) did not necessarily deliver the best result (Fig. 1D), suggesting that the core elements in HSs are sufficient to drive a high expression of HBB and other irrelevant sequences might play a negative role.^22,51 Interestingly, the selected LVs containing the shorter rLCR named LentiHBB^T87Q can be manufactured conveniently under GMP condition and achieve a high titer up to 8.0 × 10⁸ TU/mL. The clinical-grade LentiHBB^T87Q (cLentiHBB^T87Q) transduced HSPCs efficiently at a MOI as low as 5 (Fig. 4D). Our data suggest that an optimally designed cLCR not only increases the HBB expression efficacy but also facilitates the LV manufacture.

It is important to increase the transduction efficiency of HSPCs ex vivo for a successful gene therapy. In a phase I/II clinical trial (HGB-205), cells with VCN >0.6 produced HbA^T87Q level >8 g/dL, and after infusion back to patients, dramatically alleviated thalassemia phenotypes,¹⁶ suggesting that a VCN >0.6 should be effective. In another phase I/II trial based on the LV GLOBE, three of four evaluable pediatric participants discontinued transfusions after gene therapy, and the median VCN of transduced cells was 0.9. In our studies, we optimized the combinations of cytokines and transduction enhancers for cell culture and transduction. VCN in HSPCs transduced with cLentiHBB^T87Q increased with the increase of MOI and reached up to ∼2.38 (Fig. 3D) at a small scale. However, mPB HSPCs transduced with cLentiHBB^T87Q at a larger scale only had a VCN of 1.31 when MOI was 20 (Fig. 3G). The lower VCN may be due to biological variation. Indeed, we noticed that VCNs were variable in the BM HSPCs from different TDT patients, although a MOI was used at 20 in all samples. Nevertheless, VCN was always >1.3 in all the transduced BM HSPCs (Fig. 4B).

More importantly, transduction of cLentiHBB^T87Q with our current procedures was sufficient to restore HBB and HbA expression ex vivo in all tested samples, including those with β0/β0 genotype (Fig. 4), suggesting that the efficacy of cLentiHBB^T87Q is comparable with the LVs used in previous clinical trials. We noticed that HbF and other aberrant hemoglobins, which are likely the excess α-globin precipitates, were increased in the erythroblasts differentiated from BM HSPCs of TDT patients. Interestingly, those hemoglobin polymers were largely eliminated after cLentiHBB^T87Q transduction, suggesting that exogenous β-globin^T87Q can competitively and stably bind to α-globin. Furthermore, the glutamine residue (T87Q) in β-globin^T87Q disrupts the hydrophobic pocket responsible for binding mutated valine in SCD and then inhibits HbS fiber polymerization.⁵² Therefore, like other LVs containing HBB^T87Q, cLentiHBB^T87Q is expected to treat SCD with improved potency.⁵³

In previous LV-based clinical trials, after gene therapy, almost all non-β0/β0 and part of β0/β0 TDT patients were able to maintain the independence of blood transfusion, and the remaining β0/β0 TDT patients also showed a long-lasting reduced requirement for red blood infusion, indicating that the gene-modified autologous HSPCs can be successfully implanted into the body without significant side effects. Our data showed that the clonal potential and CD34⁺ percentage of HSPCs were not affected by our LV transduction (Fig. 3L, M). Therefore, we hypothesized that the in vivo duration of our gene therapy may be long lasting and comparable with the previous ones using the same strategy. Further, in vivo xenotransplantation experiments will be required to demonstrate the long-term efficacy and safety of our treatment.

One potential risk for LV-mediated gene therapies is tumorigenesis associated with viral integration, but the self-inactivation third generation of lentivirus has been used in >100 clinical trials and shows a relatively safe profile.⁵⁴ Thousands of unique VISs of cLentiHBB^T87Q in transduced mPB HSPCs were identified (Fig. 5A–C). Consistent with prior studies,^9,16 VISs are enriched in gene-dense regions (Fig. 5E). Furthermore, VIS distribution and consensus sequence of the insertion locus show patterns comparable with those observed in previous clinical studies (Fig. 5E, F), suggesting that cLentiHBB^T87Q may have similar characteristics to those clinically proven “safe” LVs. Our bioinformatics analysis of the genes influenced by VIS reveals no upregulation of cancer-related pathways, confirming that VIS of cLentiHBB^T87Q is unlikely to introduce malignant cellular transformation (Fig. 5H).

In summary, we have engineered a highly effective LV named LentiHBB^T87Q for β-thalassemia treatment. We successfully manufactured clinical-grade LentiHBB^T87Q (cLentiHBB^T87Q), and show that the cLentiHBB^T87Q is robustly expressed in the mPB HSPCs of healthy donors, and restores HBB function in the BM HSPCs of Chinese TDT patients. Our systemic profiling of integration events in cLentiHBB^T87Q transduced HSPCs excludes the likelihood of increased tumorigenesis. Taken together, our data suggest that cLentiHBB^T87Q is a promising candidate for the future therapeutic application in Chinese TDT patients.

Data Availability

The data that support the findings of this study have been deposited into CNGB Sequence Archive (CNSA: https://db.cngb.org/cnsa/) of CNGBdb with accession number CNP0001136.

Footnotes

Authors' Contributions

W.O.Y., Y.G., Y.R.L., S.L., and C.L. conceived and designed the experiments. W.O.Y., G.D., and J.L. performed the experiments. Z.Z. and H.S. conducted the bioinformatics analysis. W.Z., G.Y., R.L., Y.L., Y.R.L., and S.L. provided the crucial clinical samples and collected the consent forms. Q.Z. and L.D. assisted in CD34⁺ cell purification. W.O.Y. and C.L. wrote the article. All authors read and approved the final article.

Acknowledgments

We thank members of BGI research for their helpful discussions; Dr. Longhou Fang for article proof reading and editing; and Dr. Andrew Alpert for technical support on HPLC analysis; we also thank Yu Qiu for kind help on developing the bioinformatics pipeline. We are grateful for the support provided by China National GeneBank.

Author Disclosure

No competing financial interests exist.

Funding Information

This work is supported by grants from National Natural Science Foundation of China (NSFC) (31970816), Guangdong Provincial Key Laboratory of Genome Read and Write (No. 2017B030301011), Science, Technology and Innovation Commission of Shenzhen Municipality (No. JCYJ20170817145218948), Shenzhen Key Medical Discipline Construction Fund (SZXK034), and Sanming Project of Medicine in Shenzhen (SZSM201512033).

Supplementary Material

Supplementary Table S1

Supplementary Figure S1

Supplementary Figure S2

References

Modell

, Darlison

. Global epidemiology of haemoglobin disorders and derived service indicators. Bull World Health Organ, 2008; 86:480–487.

Piel

, Patil

, Howes

, et al. Global epidemiology of sickle haemoglobin in neonates: a contemporary geostatistical model-based map and population estimates. Lancet, 2013; 381:142–151.

Williams

, Weatherall

. World distribution, population genetics, and health burden of the hemoglobinopathies. Cold Spring Harb Perspect Med, 2012; 2:a011692.

Pennings

, Schots

, Liebaers

. Ethical considerations on preimplantation genetic diagnosis for HLA typing to match a future child as a donor of haematopoietic stem cells to a sibling. Hum Reprod, 2002; 17:534–538.

Faraci

, Bekassy

, De Fazio

, et al. Non-endocrine late complications in children after allogeneic haematopoietic SCT. Bone Marrow Transplant, 2008; 41(Suppl 2):S49–S57.

Gaziev

, Galimberti

, Lucarelli

, et al. Bone marrow transplantation from alternative donors for thalassemia: HLA-phenotypically identical relative and HLA-nonidentical sibling or parent transplants. Bone Marrow Transplant, 2000; 25:815–821.

Karponi

, Zogas

. Gene therapy for beta-thalassemia: updated perspectives. Appl Clin Genet, 2019; 12:167–180.

Schneiderman

, Thompson

, Walters

, et al. Interim results from the phase 3 Hgb-207 (Northstar-2) and Hgb-212 (Northstar-3) studies of betibeglogene autotemcel gene therapy (LentiGlobin) for the treatment of transfusion-dependent β-thalassemia. Biol Blood Marrow Transplant, 2020; 26:S87–S88.

Marktel

, Scaramuzza

, Cicalese

, et al. Intrabone hematopoietic stem cell gene therapy for adult and pediatric patients affected by transfusion-dependent ss-thalassemia. Nat Medi, 2019; 25:234–241.

10.

Negre

, Eggimann

, Beuzard

, et al. Gene therapy of the β-hemoglobinopathies by lentiviral transfer of the β(A(T87Q))-globin gene. Hum Gene Ther, 2016; 27:148–165.

11.

Malik

Gene Therapy for hemoglobinopathies: tremendous successes and remaining caveats. Mol Ther, 2016; 24:668–670.

12.

Mansilla-Soto

, Riviere

, Boulad

, et al. Cell and gene therapy for the beta-thalassemias: advances and prospects. Hum Gene Ther, 2016; 27:295–304.

13.

Negre

, Bartholomae

, Beuzard

, et al. Preclinical evaluation of efficacy and safety of an improved lentiviral vector for the treatment of beta-thalassemia and sickle cell disease. Curr Gene Ther, 2015; 15:64–81.

14.

Takekoshi

, Oh

, Westerman

, et al. Retroviral transfer of a human beta-globin/delta-globin hybrid gene linked to beta locus control region hypersensitive site 2 aimed at the gene therapy of sickle cell disease. Proc Natl Acad Sci U S A, 1995; 92:3014–3018.

15.

Pawliuk

, Westerman

, Fabry

, et al. Correction of sickle cell disease in transgenic mouse models by gene therapy. Science, 2001; 294:2368–2371.

16.

Thompson

, Walters

, Kwiatkowski

, et al. Gene therapy in patients with transfusion-dependent beta-thalassemia. N Engl J Med, 2018; 378:1479–1493.

17.

Schuessler-Lenz

, Enzmann

, Vamvakas

. Regulators' advice can make a difference: European Medicines Agency approval of zynteglo for beta thalassemia. Clin Pharmacol Ther, 2020; 107:492–494.

18.

Forrester

, Takegawa

, Papayannopoulou

, et al. Evidence for a locus activation region: the formation of developmentally stable hypersensitive sites in globin-expressing hybrids. Nucleic Acids Res, 1987; 15:10159–10177.

19.

May

, Rivella

, Callegari

, et al. Therapeutic haemoglobin synthesis in beta-thalassaemic mice expressing lentivirus-encoded human beta-globin. Nature, 2000; 406:82–86.

20.

Roselli

, Mezzadra

, Frittoli

, et al. Correction of beta-thalassemia major by gene transfer in haematopoietic progenitors of pediatric patients. EMBO Mol Med, 2010; 2:315–328.

21.

Breda

, Casu

, Gardenghi

, et al. Therapeutic hemoglobin levels after gene transfer in beta-thalassemia mice and in hematopoietic cells of beta-thalassemia and sickle cells disease patients. PLoS One, 2012; 7:e32345.

22.

Weber

, Poletti

, Magrin

, et al. An optimized lentiviral vector efficiently corrects the human sickle cell disease phenotype. Mol Ther Methods Clin Dev, 2018; 10:268–280.

23.

Morgan

, Unti

, Aleshe

, et al. Improved titer and gene transfer by lentiviral vectors using novel, small beta-globin locus control region elements. Mol Ther, 2020; 28:328–340.

24.

Morgan

, Ma

, Unti

, et al. Creating new beta-globin-expressing lentiviral vectors by high-resolution mapping of locus control region enhancer sequences. Mol Ther Methods Clin Dev, 2020; 17:999–1013.

25.

Weatherall

DJ.

Phenotype-genotype relationships in monogenic disease: lessons from the thalassaemias. Nat Rev Genet, 2001; 2:245–255.

26.

Salmon

, Trono

Production and titration of lentiviral vectors. Curr Protoc Hum Genet 2007;Chapter 12:Unit 12.10.

27.

Boulad

, Wang

, Qu

, et al. Safe mobilization of CD34+ cells in adults with beta-thalassemia and validation of effective globin gene transfer for clinical investigation. Blood, 2014; 123:1483–1486.

28.

Rae

, Collins

, Hocum

, et al. Modified genomic sequencing PCR using the MiSeq platform to identify retroviral integration sites. Hum Gene Ther Methods, 2015; 26:221–227.

29.

Schmidt

, Schwarzwaelder

, Bartholomae

, et al. High-resolution insertion-site analysis by linear amplification-mediated PCR (LAM-PCR). Nat Methods, 2007; 4:1051–1057.

30.

Chen

, Chen

, Shi

, et al. SOAPnuke: a MapReduce acceleration-supported software for integrated quality control and preprocessing of high-throughput sequencing data. Gigascience, 2017; 7:gix120.

31.

Kim

, Langmead

, Salzberg

. HISAT: a fast spliced aligner with low memory requirements. Nat Methods, 2015; 12:357.

32.

, Handsaker

, Wysoker

, et al. The sequence alignment/map format and SAMtools. Bioinformatics, 2009; 25:2078–2079.

33.

Robinson

, Thorvaldsdóttir

, Winckler

, et al. Integrative genomics viewer. Nat Biotechnol, 2011; 29:24.

34.

Quinlan

, Hall

. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics, 2010; 26:841–842.

35.

Wagih

ggseqlogo: a versatile R package for drawing sequence logos. Bioinformatics, 2017; 33:3645–3647.

36.

Wang

, Li

, Hakonarson

. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res, 2010; 38:e164.

37.

, Wang

L-G

, Han

, et al. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS, 2012; 16:284–287.

38.

Ashburner

, Ball

, Blake

, et al. Gene ontology: tool for the unification of biology. Nat Genet, 2000; 25:25.

39.

Dull

, Zufferey

, Kelly

, et al. A third-generation lentivirus vector with a conditional packaging system. J Virol, 1998; 72:8463–8471.

40.

Miyoshi

, Blomer

, Takahashi

, et al. Development of a self-inactivating lentivirus vector. J Virol, 1998; 72:8150–8157.

41.

Ellis

, Pasceri

, Tan-Un

, et al. Evaluation of beta-globin gene therapy constructs in single copy transgenic mice. Nucleic Acids Res, 1997; 25:1296–1302.

42.

Thein

SL.

The molecular basis of beta-thalassemia. Cold Spring Harb Perspect Med, 2013; 3:a011700.

43.

Gibson

, Young

, Chuang

, et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat Methods, 2009; 6:343–345.

44.

Heffner

, Bonner

, Christiansen

, et al. Prostaglandin E2 increases lentiviral vector transduction efficiency of adult human hematopoietic stem and progenitor cells. Mol Therapy, 2018; 26:320–328.

45.

Uchida

, Nassehi

, Drysdale

, et al. High-efficiency lentiviral transduction of human CD34(+) cells in high-density culture with poloxamer and prostaglandin E2. Mol Ther Methods Clin Dev, 2019; 13:187–196.

46.

Ngom

, Imren

, Maetzig

, et al. UM171 enhances lentiviral gene transfer and recovery of primitive human hematopoietic cells. Mol Ther Methods Clin Dev, 2018; 10:156–164.

47.

Biffi

, Montini

, Lorioli

, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science, 2013; 341:1233158.

48.

Aiuti

, Biasco

, Scaramuzza

, et al. Lentiviral hematopoietic stem cell gene therapy in patients with Wiskott-Aldrich syndrome. Science, 2013; 341:1233151.

49.

Bank

, Dorazio

, Leboulch

. A phase I/II clinical trial of beta-globin gene therapy for beta-thalassemia. Ann N Y Acad Sci, 2005; 1054:308–316.

50.

Miccio

, Cesari

, Lotti

, et al. In vivo selection of genetically modified erythroblastic progenitors leads to long-term correction of beta-thalassemia. Proc Natl Acad Sci U S A, 2008; 105:10547–10552.

51.

Leboulch

, Huang

, Humphries

, et al. Mutagenesis of retroviral vectors transducing human beta-globin gene and beta-globin locus control region derivatives results in stable transmission of an active transcriptional structure. EMBO J, 1994; 13:3065–3076.

52.

Nagel

, Bookchin

, Johnson

, et al. Structural bases of the inhibitory effects of hemoglobin F and hemoglobin A2 on the polymerization of hemoglobin S. Proc Natl Acad Sci U S A, 1979; 76:670–672.

53.

Ribeil

, Hacein-Bey-Abina

, Payen

, et al. Gene therapy in a patient with sickle cell disease. N Engl J Med, 2017; 376:848–855.

54.

Milone

, O'Doherty

. Clinical use of lentiviral vectors. Leukemia, 2018; 32:1529–1541.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.11 MB

0.22 MB

Restoration of β-Globin Expression with Optimally Designed Lentiviral Vector for β-Thalassemia Treatment in Chinese Patients

Abstract

Introduction

Materials and Methods

Ethics approval and consent to participate

LV design, manufacture, and titration measurement

HSPCs culture and lentiviral transduction

Methylcellulose colony forming unit assay

Erythroid differentiation assay

gDNA extraction and VCN measurement

Total RNA extraction and quantitative RT-PCR

HPLC analysis of globin and hemoglobin

Identification of VIS

Bioinformatics analysis of VIS data

Results

LV backbone selection and optimization

Construction of LVs containing different rLCRs

Expression of LVs in erythroblast cells derived from CB HSPCs

LentiHBBT87Q express HBB protein in a dosage-dependent manner

Optimization of lentiviral transduction in mPB HSPCs and validation of the optimized parameters with clinical-grade LentiHBBT87Q

Restoration of globin and hemoglobin with cLentiHBBT87Q transduction in the erythroblasts differentiated from the BM HSPCs of Chinese TDT patients

Characterization of VIS

Discussion

Data Availability

Footnotes

Authors' Contributions

Acknowledgments

Author Disclosure

Funding Information

Supplementary Material

References

Supplementary Material

LentiHBB^T87Q express HBB protein in a dosage-dependent manner

Optimization of lentiviral transduction in mPB HSPCs and validation of the optimized parameters with clinical-grade LentiHBB^T87Q

Restoration of globin and hemoglobin with cLentiHBB^T87Q transduction in the erythroblasts differentiated from the BM HSPCs of Chinese TDT patients