| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
promoter: in vivo model for platelet-targeting gene therapy
,
,1
* Research Division of Cell and Molecular Medicine, Center for Molecular Medicine, Jichi Medical School, Tochigi, Japan;
Hematology Division of Department of Medicine, Jichi Medical School, Tochigi, Japan;
DNAVEC Corp., Ibaraki Japan; and
Research Division of Genetic Therapeutics, Center for Molecular Medicine, Jichi Medical School, Tochigi, Japan
1Correspondence: Research Division of Cell and Molecular Medicine, Center for Molecular Medicine, Jichi Medical School, Minamikawachi, Tochigi 329-0498, Japan. E-mail: yoisaka{at}jichi.ac.jp
ABSTRACT
Platelets release several mediators that modify vascular integrity and hemostasis. In the present study, we developed a technique for efficient transgene expression in platelets in vivo and examined whether this targeted-gene-product delivery system using a platelet release reaction could be exploited for clinical applications. Analysis of luciferase reporter gene constructs driven by platelet-specific promoters (the GPIIb, GPIb
, and GPVI) revealed that the GPIb promoter was the most potent in the megakaryoblastic cell line UT-7/TPO and human CD34+-derived megakaryocytes. Transduction of UT-7/TPO; CD34+-derived megakaryocytes; and c-Kit+, ScaI+, and Lineage– (KSL) murine hematopoietic stem cells with a simian immunodeficiency virus (SIV)-based lentiviral vector carrying eGFP resulted in efficient, dose-dependent expression of eGFP, and the GPIb promoter seemed to bestow megakaryocytic-specific expression. Transplantation of KSL cells transduced with SIV vector containing eGFP into mice showed that there was preferable expression of eGFP in platelets driven by the GPIb promoter [7–11% for the cytomeglovirus (CMV) promoter, 16–27% for the GPIb promoter]. Furthermore, transplantation of ex vivo-transduced KSL cells by SIV vector carrying human factorVIII (hFVIII) driven by the GPIb promoter induced the production of detectable transcripts of the hFVIII gene and the hFVIII antigen in bone marrow and spleen for at least 90 days and partially corrected the hemophilia A phenotype. Platelet-targeting gene therapy using SIV vectors appears to be promising for gene therapy approaches toward not only inherited platelet diseases but also other hemorrhagic disorders such as hemophilia A.—Ohmori, T., Mimuro, J., Takano, K., Madoiwa, S., Kashiwakura, Y., Ishiwata, A., Niimura, M., Mitomo, K., Tabata, T., Hasegawa, M., Ozawa, K., Sakata, Y. Efficient expression of a transgene in platelets using simian immunodeficiency virus-based vector harboring glycoprotein Ib promoter: in vivo model for platelet-targeting gene therapy.
Key Words: hemophilia A • lentiviral vector • stem cell transplantation
BLOOD PLATELETS, the principal cells responsible for primary hemostasis, play major roles in thrombosis, atherosclerosis, tumor metastasis, and inflammation. At the site of vascular injury, activated platelets aggregate and release several mediators that modify vascular integrity and hemostasis (1)
. Platelet-derived bioactive products are released by exocytosis of the three types of granules (dense-core, alpha, and lysosome) within platelets; this process is mediated by soluble NSF attachment protein receptor proteins and VAMP-3 (2)
. Taking advantage of the platelet-release reaction as a delivery system for a specific factor would be a reasonable approach for therapy for individuals deficient in the factor because it provides a way to enhance the local concentration of target substances at the site of vascular injury, while minimizing the influence of plasma proteins that may inhibit their activities.
In transgenic settings, platelet-targeting gene transfer has been reported to enable the storage of the targeted substance within platelets. Platelet expression of urokinase-type plasminogen activator (u-PA) using a megakaryocyte-specific platelet factor 4 promoter enabled u-PA to be stored in platelets and then released within developing thrombi when the platelets became activated (3)
. Furthermore, platelet-specific expression of coagulation factor VIII (FVIII) could be achieved in a transgenic setting with the resultant FVIII predominantly or exclusively stored in platelet granules rather than being released into the plasma (4)
. When transgenic mice were crossed onto a FVIII null background, whole blood clotting time was partially corrected (4)
. These findings have facilitated the development of methods for gene therapy that use platelets to deliver therapeutic agents to the site of vascular injury.
The present study aimed at megakaryocyte- and platelet-directed gene transfer using a platelet-specific promoter that could be directly applicable to gene therapy for not only inherited platelet disorders but also coagulation factor deficiencies such as hemophilia. Since megakaryocytes have a finite life span, hematopoietic stem cells are preferable targets for genetic transfer to establish in vivo long-term expression of the target protein in platelets. When retroviral vector containing the Glycoprotein (GP) IIIa gene driven by the GPIIb promoter was transduced into CD34+ cells from a Glanzmann thrombasthenia patient with defects in the GPIIIa gene, GPIIb/IIIa were detected after in vitro megakaryocyte differentiation (5)
. Furthermore, retrovirus transduction of FVIII driven by the virus promoter into human CD34+ hematopoietic stem cells reportedly enabled FVIII-transduced megakaryocytes to store human FVIII with von Willebrand factor (VWF; ref 6
). Although the concept of platelet-targeting gene therapy is very attractive and some reports have clinical implications as described above (3
4
5
6)
, there have been no evaluations of whether gene transduction of hematopoietic stem cells effectively results in sufficient genetic information being given in platelets so that they synthesize enough transgene products to correct the phenotype of the hemorrhagic disorder.
In this study, we used GPIb promoter to achieve platelet-specific gene expression, based on analyses of the promoter activities of three platelet-specific genes, GPIIb, GPIb, and GPVI, in megakaryocytes and then we assessed the gene transfer efficiency to platelets in vivo by transducing hematopoietic stem cells with simian immunodeficiency lentivirus vector (SIV). We also examined whether FVIII ectopically expressed in platelets corrected the hemorrhagic phenotype of FVIII-deficient mice by using the platelet release reaction.
MATERIALS AND METHODS
Mice
Hemophilic A mice with targeted destruction of exon 16 of the FVIII gene were kindly provided by H. H. Kazazian Jr. (University of Pennsylvania, Philadelphia, PA; ref. 7
). C57BL/6 (B6-Ly5.2) mice were purchased from Japan SLC (Shizuoka, Japan). C57BL/6 mice congenic for the Ly5 locus (B6-Ly5.1) were purchased from Sankyo-Lab Service (Tsukuba, Japan). All animal procedures were approved by the institutional Animal Care and Concern Committee at Jichi Medical School, and animal care was performed in accordance with the guidelines of the committee (8)
.
Cytokines and antibodies
Recombinant human thrombopoietin (TPO) and recombinant human stem cell factor (SCF) were gifts from Kirin Brewery (Gunma, Japan). The following materials were obtained from the indicated suppliers: recombinant human interleukin (IL)-6, recombinant human soluble IL-6 receptor (sIL-6R), recombinant human basic fibroblast growth factor (b-FGF), and recombinant human Flt3-Ligand (Flt3-L; PeproTech EC, London, UK); recombinant human IL-3 (IL-3; TECHNE, Minneapolis, MN); antimouse c-Kit monoclonal antibody (MoAb; clone 2B8), antimouse Sca-1 MoAb (clone D7), antimouse Ly5 (CD45) MoAb (clone 30-F11), and antimouse Ly5.1 (CD45.1) MoAb (clone A20; BD Pharmingen, CA); antimouse GPIb MoAb (clone Xia.G5; Emfret Analytics, Wurzberg, Germany); and anti-human GPIIb/IIIa MoAb (clone 5B12), anti-human GPIb MoAb (clone AN51), and anti-human CD34 MoAb (clone BIRMA-K3; DakoCytomation, Glostrup, Denmark).
Cell culture
The human megakaryoblastic cell line UT-7/TPO was kindly provided by Dr. Norio Komatsu (Yamanashi University, Yamanashi, Japan; ref. 9
). The cells were cultured in Iscove’s modified Dulbecco’s medium (IMDM) supplemented with 10% FBS and 10 ng/ml of TPO. Human umbilical vein endothelial cells (HUVEC) were obtained and maintained as described previously (10)
. U937, a myelomonocytic cell line, and K562, an erythroleukemic cell line, were obtained from the American Type Culture Collection and were cultured in RPMI 1640 supplemented with 10% FBS. 3T3 fibroblasts were cultured with Dulbecco’s modified Eagle’s medium (DMEM)/F-12 supplemented with 10% FBS. Human aortic smooth muscle cells (SMC) were purchased from the Applied Cell Biology Research Institute (Kirkland, WA) and maintained in DMEM/F-12 with 10% FBS and 10 ng/ml of b-FGF.
Megakaryocyte differentiation
Human umbilical cord blood was obtained during normal full-term deliveries. The institutional review board of Jichi Medical School approved the study protocols, and informed consent was obtained from all donors. Human CD34+ cells derived from cord blood were isolated using the AutoMACS magnetic cell sorting system (Miltenyi Biotec., Auburn, CA) according to the manufacturer’s instructions. The purity of isolated CD34+ cells was >90% (data not shown). CD34+ cells were expanded and differentiated into megakaryocytes with IMDM containing 1% fatty acid-free BSA, 200 µg/ml of human iron-saturated transferrin, and 10 µg/ml of human recombinant insulin, supplemented with 50 ng/ml of TPO and 10 ng/ml of IL-3 (11)
. After 14 days of culture, 75–90% of cells were positive for GPIIb/IIIa.
Construction of luciferase reporter plasmid, transient transfection, and luciferase assay
The DNA fragments for the promoter, which reportedly each had maximum promoter activity [GPIIb promoter: –554 to+33 (12)
, GPIb promoter:- 254 to+330 (13)
, GPVI promoter: –320 to + 28 (14)
], were amplified by polymerase chain reaction (PCR) using human genomic DNA as template. Since the region surrounding the translation start site of GPIb gene at position –5 from the initiator ATG codon is important for transgene expression (15)
, we selected the DNA fragment just before the first ATG sequence, as reported previously (13)
. The oligonucleotide primer pairs used for the cloning of the promoter sequence were as follows: 5'-CCATTCCAGAAGGTGTGAAG-3' (sense) and 5'-GTTCCTCAGCCCTGTCCTG-3' (antisense) for GPVI promoter (GenBank #AF521646); 5'-CTAAAGCTTGGCTCAAGACG-3' (sense) and 5'-CTTCCTTCTTCCACAACCTC-3' (antisense) for GPIIb promoter (GenBank #M33319); and 5'-GTTCTGGGATTACAGGCATGAG-3' (sense) and 5'-GAGGACCTGTGGGCAAGGGAG-3' (antisense) for GPIb promoter (GenBank #M22403).
After being subcloned into pCR-Blunt-TOPO (Invitrogen, Carlsbad, CA), the fragment was subsequently cloned into pGL3-basic, a promoterless luciferase plasmid (Promega, Madison, CA). Five hundred thousand cells (UT-7/TPO, CD34+-derived megakaryocytes, K562, and U937) plated in IMDM (without FBS or growth factors) or subconfluent cells (HUVEC and SMC) in each well of six-well plates were transfected with 4 µg of plasmid DNA using DMRIE-C Reagent (Invitrogen). After 4 h, an equal volume of IMDM supplemented with 2x growth factors was added to each well. Cells were incubated for 48 h at 37°C, and luciferase activities were assayed according to the manufacturer’s instructions (Luciferase Assay System, Promega).
Construction and production of SIV vectors
Replication-defective self-inactivating (SIN) SIV vector was created by deletion of the U3 region of 3' long terminal repeat (LTR), which contains two NF-
B sites, three Sp-1 sites, and a TATA box, as described previously (ref. 16
; Gene Bank association number: #X07805). The transduction efficiency of the SIN vector did not differ from the intact U3 sequence-containing vector (16)
.
The full-length human FVIII (hFVIII) cDNA was a generous gift from Dr. J.A. van Mourik (Blood Coagulation, Sanquin, Amsterdam, Netherlands), and the human B domain-deleted (BDD) FVIII (hBDD-FVIII) cDNA was generated by PCR-based mutagenesis as described previously (17)
. eGFP or hBDD-FVIII driven by the CMV promoter (SIV-CMV-eGFP/hFVIII) or GPIb promoter (SIV-GPIb-eGFP/hFVIII) was inserted between the LTR-containing elements of an SIV-derived vector (see Results).
The gene transfer plasmid was transfected together with three packaging plasmids (encoding gag-pol, rev, and VSV-G env) into 293T cells using Lipofectamine Plus reagent (Invitrogen). After 12 h, the culture medium was replaced to start harvesting virus particles; harvesting was undertaken at 48 h and virus particles were concentrated by ultracentrifugation. Transduction units of SIV vectors carrying eGFP were measured by infection of 293 cells followed by measurement of eGFP expression by FACS analysis. The average infectivity of the SIV-CMV-eGFP vector was in the range of 2–5 x 108 TU/ml. To compare viral infectivity between different promoters or targeted genes, viral particle titer was simultaneously measured by real-time quantitative reverse transcriptase (RT)-PCR. Viral RNA was isolated using QIAamp viral RNA mini kit (QIAGEN, Valencia, CA), and the isolated RNA was reverse-transcribed using SuperScript II (Invitrogen). Quantification of vector particles was performed by measuring copies of vector-specific post transcriptional regulatory element derived from the wood-chuck hepatitis virus (WPRE) sequences by real-time quantitative PCR using the QuantiTect Probe PCR system (QIAGEN). The WPRE sequence was amplified with WPRE forward primer 5'-GCTTTCATTTTCTCCTCCTT-3' and WPRE reverse primer 5'-GGCCACAACTCCTCATAA-3'. The FAM-labeled probe sequence was 5'-ATCCTGGTTGCTGTCTC-3'. The PCR started with an initial incubation step of 15 min at 95°C. Thermal cycling consisted of 45 cycles of 94°C for 15 s, 56°C for 30 s, and 76°C for 30 s. During the annealing phase of PCR, reporter fluorescent dye from a specific probe was detected with a ABI PRISM 7700 Sequence Detector System (PE Applied Biosystems, Foster City, CA). The standard curve of the viral titer was estimated by serial dilution of the gene transfer vector plasmid. The average virus particle titer of the SIV-based vector was in the range of 1–2 x 1010 TU/ml. The infectivity of SIV vectors [multiplicity of infection (MOI)] was estimated from the ratio of particle titer to SIV-CMV-eGFP.
For the transduction of UT-7/TPO and CD34+-derived megakaryocytes with SIV vectors, 1 x 105 cells were resuspended in 100 µl of culture medium containing 8 µg/ml of polybrene. Cells were transduced with SIV vectors at various MOIs indicated in the test for 24 h. Cells were then resuspended in 300 µl PBS containing 0.5% BSA and 2 mM EDTA, and eGFP-positive cells were subjected to FACS analysis. When CD34+ cells were transduced with SIV containing eGFP followed by differentiation of megakaryocytes in vitro, we could not observe sustained eGFP expression during the differentiation. This was probably due to the short life span of the eGFP-expressing differentiating cells in vitro. Hence, CD34+-derived megakaryocytes, at an indicated day after the start of differentiation, were transduced with SIV to examine the influence of megakaryopoiesis on promoter activity.
Hematopoietic stem cell isolation and viral transduction
Bone marrow cells obtained from mouse femur and tibia were depleted for cells expressing lineage cell markers B220, CD5, CD11b, Gr-1, and Ter-119 by magnetic sorting using a Lineage Cell Depletion kit (Miltenyi Biotec.) and then sorted for Sca-1+ and c-kit+ cells (KSL) by FACS (FACSAria Cell Sorter, Becton-Dickinson).
When freshly isolated KSL cells were directly infected with SIV vectors, PI-positive cells (dead cells) increased after the transduction and the efficacy of the transduction decreased (data not shown). Hence, we examined the culture conditions of KSL cells to improve SIV transduction and cell viability. When isolated KSL cells were cultured with IL-3, IL-6, and SCF, SIV transduction of the eGFP gene into KSL cells resulted in lower transduction efficiencies (25–35%: MOI of 30) and a significant increase in PI-positive cells (20–30%; data not shown). On the other hand, high proportions of eGFP positive cells were obtained by culturing the KSL cells with SCF, IL-6, sIL-6R, Flt-3L, and TPO for 3–7 days (see Results), and PI-positive cells after eGFP transduction were significantly decreased (8–12%). Hence, KSL cells were precultured with IMDM 1% fatty acid-free BSA, 200 µg/ml of transferrin, and 10 µg/ml of insulin, supplemented with 100 ng/ml of SCF, 10 ng/ml of TPO, 100 ng /ml of IL-6, 100 ng/ml of Flt-3L, and 400 ng/ml of sIL-6R before viral transduction in accordance with a method for human CD34+ expansion (18)
. Infection of cells with SIV was carried out in a plate coated with 50 µg/ml of RetroNectin (TakaraBio, Tokyo, Japan). Cultured KSL cells (1x105 or 1x106 cells) were resuspended in 100 µl or 1 ml of IMDM with 10% FBS. The cells were transduced with SIV vectors at various MOIs as indicated in the text for 12 h in the presence of the same cytokine combination and incubated at 37°C. Since the use of polybrene (up to 4 µg/ml) per se did not significantly improve the transduction efficiencies of mouse KSL and human CD34+ cells (data not shown), our procedure for stem cell transduction was performed without polybrene. Cells were then resuspended in 100 µl of PBS containing 1% BSA for stem cell transplantation or incubated for the indicated number of days for FACS analysis.
Stem cell transplantation
Bone marrow cells were obtained from B6-Ly5.1 to allow us to distinguish between donor and recipient cells. Recipient mice (8–12 wk old B6-Ly5.2) were irradiated with a single lethal dose of 9.5 Gy (60Co, Gamma Cell; Norton International, Ontario, Canada). One hundred thousand cultured KSL cells from B6-Ly5.1 without or with SIV transduction were injected together with 5 x 105 freshly isolated Ly5.2 unfractionated bone marrow cells. Since the transplantation procedure without unfractionated bone marrow cells increased mortality to 60–80%, we simultaneously transplanted KSL cells with nontransduced bone marrow cells as competitor cells. To assess the reconstitution of bone marrow, peripheral blood was drawn from the retro-orbital sinus with heparin-coated micropipettes and analyzed for the percentages of Ly5.1 (donor-derived) lymphoid and myeloid cells by flow cytometry. In our transplantation procedure, engraftment by Ly5.1 cells was 40–55% in lymphoid and myeloid cells 4 months after transplantation. For secondary transplantations, 2 x 106 unfractionated marrow cells collected from primary recipient mice 4 months after primary transplantation were used to reconstitute lethally irradiated recipients (9.5 Gy). The animals with relatively high eGFP levels after transplantation were chosen as donors for the secondary bone marrow transplantation.
Detection of the transcripts of hFVIII transgene
Detection of hFVIII transgene transcripts was performed by RT-PCR (19)
. RNA was isolated from mouse organs using an RNA isolation kit (RNeasy Protect Kit; QIAGEN). RNA samples were subjected to RT-PCR using a pair of primers for hFVIII (19)
and an RT-PCR kit (SuperScript One-Step RT-PCR System, Invitrogen). A primer pair for mouse GAPDH mRNA (R&D Systems, Minneapolis, MN) was used in the control RT-PCR experiments.
Quantification of SIV vector-derived mRNA expression was examined by real-time quantitative RT, using the QuantiTect Probe RT-PCR system (QIAGEN). Mouse GAPDH (mGAPDH) control reagents (QIAGEN) were used to estimate the amount of RNA analyzed. A standard curve was established by analyzing duplicate aliquots of serial dilutions of SIV gene transfer vector. Fifty nanograms of purified RNA from tissues as indicated in the text were used as a template. For each sample, the amounts of WPRE sequences compared with the standard control curve were estimated, and the quantities of the WPRE sequences were determined by dividing the copy number of the WPRE sequences by those of the mGAPDH sequences.
Integration of vector sequence into genomic DNA
Genomic DNA was extracted from bone marrow cells of mice 4–5 months after transplantation using a DNA isolation kit (DNeasy Tissue Kit; QIAGEN). WPRE sequence derived from the SIV vector was amplified using real-time quantitative PCR with a QuantiTect Probe PCR system as described above. For each sample, the amounts of WPRE sequences compared with the standard control curve were estimated, and the integration of the vector sequence was calculated from dividing the copy number by the cell count corresponding to the DNA quantity.
Immunohistochemistry
Bone marrow cells attached to glass slides using a Cytospin3 (Shandon, ThermoShandon) were fixed with 3% paraformaldehyde in PBS and then permeabilized with 0.2% Triton X-100. After being blocked with 1% BSA, samples were incubated with polyclonal anti-hFVIII antibody (Ab) conjugated with biotin at 4°C for 2 h, washed with PBS, and then incubated with streptavidin conjugated with Alexa 594 (Molecular Probes, Eugene, OR) and FITC-labeled ant-imouse GPIb MoAb. Immunofluorescent staining was observed and photographed using a fluorescent microscope with an attached camera.
For detection of hFVIII molecules in mouse tissues by immunohistochemistry, the spleen was fixed with 4% paraformaldehyde in PBS for 2 h at 4°C, incubated with PBS containing sucrose (10–30%), and then frozen in the presence of OCT compound in dry ice/ethanol. Sections were prepared from frozen tissues at –25°C and attached to polylysine-coated glass slides. For detection of hFVIII, tissue sections were blocked with 1% rabbit serum in PBS containing Triton-X 100 (0.1%) and incubated with sheep polyclonal anti-human FVIII Ab (Cedarlane Laboratories, Homby, Ontario, Canada) at 4°C for 16 h. After being washed with PBS, sections were incubated with biotin-conjugated rabbit anti-sheep IgG Ab followed by the avidin-biotin complex (ABC) reagents (Vectastain ABC Elite kit; Vector, Burlingame, CA) and a 3,3'-diaminobenzidine kit (Vector).
Measurement of hFVIII antigen, neutralizing antibodies, and phenotypic correction
hFVIII antigens were measured by an anti hFVIII-specific ELISA kit (Affinity Biological, Hamilton, Ontario, Canada) and compared with those in pooled human normal plasma. Two hundred and seventy microliters of whole blood were drawn from the superior vena cava of anesthetized mice using a syringe containing 30 µl of sodium citrate and 1 µM PGI2 to inhibit platelet activation. When indicated, 1 µM phorbol myristyl acetate (PMA) and 50 µg/ml of collagen were added to the blood to activate platelets. After centrifugation, the platelet-poor plasma was frozen at –80°C until assayed for FVIII antigen. Analyses of neutralizing antibodies against hFVIII developed in mice were performed by the Bethesda method as described using FVIII deficient plasma and normal pooled plasma (8)
. Phenotypic correction was tested in some of the transplanted mice by anesthetizing them with diethyl ether and clipping 1.5 cm of their tails. The mice were then observed for survival after 24 h.
RESULTS
Comparison of luciferase reporter expression driven by platelet-specific promoters
To achieve efficient expression of genes targeted in platelets, we first compared the promoter activities of three platelet-specific genes, GPIIb, GPIb, and GPVI, in megakaryocytic cells. Figure 1
A shows a schematic diagram of the platelet-specific promoters used in this study with their unique regulatory elements. The luciferase reporter gene was used to compare promoter activities among different promoters. We used the nucleotide region in promoters that had the highest activity in previous studies (12
13
14)
. The GPIb promoter directed the most powerful expression of luciferase in UT-7/TPO cells, a megakaryoblastic cell line (Fig. 1B
), and its relative efficiency was even more marked in CD34+-derived megakaryocytes (Fig. 1C
). Since GPIb and GPVI were reportedly expressed in endothelial cells (20
,21)
, we examined whether these platelet-specific promoter activities were stimulated in endothelial cells. Under conditions in which the PAI-1 promoter (22)
efficiently directed luciferase expression, the GPIIb, GPIb, and GPVI promoters did not direct luciferase expression in HUVECs (Fig. 1)
. We further examined promoter specificity by using other cells including K562, an erythroleukemic cell line; U937, a myelomonocytic cell line; human aortic artery SMC and 3T3 fibroblast cell. In these cells, the platelet-specific promoters drove less reporter gene than SV40/Enhancer and CAG promoter (Fig. 1)
. Hence, the specificities of the platelet glycoprotein promoters did not differ, and the activity of GPIb promoter was the strongest in megakaryocytes.
|
Efficient transformation of megakaryocytic cells by SIV-based vectors
Next, we constructed SIV-based lentiviral vectors containing the eGFP gene under the control of either the CMV promoter (SIV-CMV-eGFP), GPIb promoter (SIV-GPIb-eGFP), GPIIb (SIV-GPIIb-eGFP), or GPVI (SIV-GPVI-eGFP). The transgene located downstream of the promoter was inserted between the LTR-containing elements of a SIV-derived vector (Fig.2A
). To increase gene expression in the transduced cells, a post-transcriptional regulatory element derived from woodchuck hepatitis virus (WPRE) was inserted downstream of the gene expressed (Fig. 2A
). To investigate eGFP gene transduction of megakaryocytes, UT-7/TPO or CD34+-derived megakaryocytes were cultured for 24 h in the presence of various concentrations of the indicated SIV vector. SIV-GPIIb-eGFP, SIV-GPIb-eGFP, SIV-GPVI-eGFP, and SIV-CMV-eGFP efficiently transduced the eGFP gene into UT-7/TPO and CD34+-derived mega-karyocytes (Fig. 2B
, C
). There were no significant differences in eGFP expressions among cells transfected with SIV-GPIIb-eGFP and SIV-GPIb-eGFP (Fig. 2B
and C).
|
We next investigated whether ex vivo megakaryocyte differentiation affects gene expression. eGFP expression in CD34+ hematopoietic progenitor cells transduced with SIV-GPIb-eGFP or SIV-GPVI-eGFP was <10%, whereas 
20% of the cells transduced with SIV-GPIIb-eGFP were positive for eGFP (day 0 in Fig. 2D
), indicating that the GPIIb promoter worked at an earlier stage of megakaryopoiesis. Furthermore, the GPIb compared with the GPIIb promoter seemed to work during a later phase of megakaryocyte maturation, and the percentages of eGFP expression did not change after differentiation (day 14 in Fig. 2D
). eGFP expression driven by the CMV promoter was not affected by megakaryocyte differentiation (Fig. 2D
). We selected the GPIb promoter as the platelet-specific promoter for in vivo experiments because the promoter activity of GPIb was the strongest in differentiated megakaryocytes (Fig. 1)
and the promoter drove in the later phase of megakaryopoiesis (Fig. 2)
.
Establishment of efficient transduction of KSL cells with SIV vector
We next optimized the transduction protocol for KSL cells by using an SIV vector containing the eGFP gene driven by the CMV promoter. The transduction efficiency of eGFP in cultured KSL cells reached 60–80% (Fig. 3
A). The plateau value for transduction was observed with a MOI of 10–30. One day (24 h) after incubation with the viral vector was sufficient to achieve the efficient expression of the transduced gene (Fig. 3B
). eGFP expression then gradually declined (Fig. 3B
); the decrease might have been due to the reduction in cell viability, because PI-positive cells (dead cells) increased with time (data not shown). Although lentiviral vectors can express transgenes for long periods even in the absence of integration in CD34+ cells (23)
, KSL cells could not maintain eGFP expression for long periods in vitro. Accordingly, we cultured KSL at an MOI of 30 for 24 h and transplanted the cells into recipient mice in the following experiments.
|
Preferential eGFP expression in platelets using SIV vectors harboring GPIb promoter in vivo
To compare the strength and the specificity of the CMV and GPIb promoters and to assess eGFP transduction by SIV vectors in vivo, KSL cells transduced with SIV-CMV-eGFP or SIV-GPIb-eGFP were transplanted to recipient mice (Ly5.2). One hundred thousand cultured KSL cells (Ly5.1) transduced with SIV-CMV-eGFP or SIV-GPIb-eGFP (MOI of 30) were transplanted together with 5 x 105 competitor cells (Ly5.2). When KSL cells transduced with SIV-CMV-eGFP were transplanted, eGFP expression was observed in 35–45% of CD45+ cells and 7–11% of platelets in peripheral blood (Fig. 4
A and B). Interestingly, transduction of the SIV vector harboring the GPIb promoter resulted in efficient gene marking to platelets (16–27%); however, only marginal eGFP expression was observed in CD45+ and red blood cells (Fig. 4A, B
). We next analyzed bone marrow cells from transplanted mice using specific markers to identify macrophages, granulocytes, B lymphocytes, T lymphocytes, and erythroblasts. Whereas eGFP was expressed in these lineages of cells of mice that received KSL cells transduced with SIV-CMV-eGFP, the GPIb promoter drove just marginal eGFP expression in these cell lineages, confirming the specificity of its activity in megakaryocytes and platelets in vivo (Fig. 4C
). We next performed a second bone marrow transplantation using marrow cells obtained from mice that had been transplanted 4 months earlier. As shown in Fig. 4D
, eGFP expression driven by the CMV and GPIb promoters in hematopoietic cells was sustained after the second stem cell transplantation, indicating that the promoter maintains transgene expression during differentiation of hematopoietic stem cells.
|
Expression of hFVIII and phenotypic correction in hemophilia A mice transplanted with KSL cells transduced with SIV-GPIb-hFVIII
To determine whether platelet-directed gene therapy enables sustained expression of FVIII, we constructed two SIV-based vectors containing hBDD-FVIII cDNA under the control of either the CMV (SIV-CMV-hFVIII) or GPIb promoter (SIV-GPIb-hFVIII). We transplanted mice with 1 x 105 transduced KSL cells after lethal 
-irradiation. We first analyzed the presence of the hFVIII gene transcripts in organs of the transplanted recipients at 3 months after transplantation. Analyses for hFVIII transcripts driven by the CMV promoter revealed that the hFVIII gene was expressed mainly in bone marrow and to a lesser extent in the spleen (Fig. 5
A, middle panel). Interestingly, hFVIII mRNA was predominantly found in both the spleen and bone marrow in the recipients of KSL cells transduced with SIV-GPIb-hFVIII (Fig. 5A
, lower panel). To quantify the mRNA expression concentration in each organ, vector-specific WPRE expression was measured by real-time quantitative RT-PCR. As expected, based on the results from RT-PCR, bone marrow and spleen are the major sites for mice transplanted with KSL cells transduced with SIV vectors (Fig. 5B
). Furthermore, we examined vector integration into the genome of bone marrow hematopoietic cells after transplantation and detected that 0.19–2.3 vector copies/genome were integrated in cells of transplanted mice (CMV promoter: 1.07±0.95; n=4; GPIb promoter: 0.98±0.62; n=4). In accordance with the data on hFVIII transcripts, hFVIII molecules were immunohistochemically detected in bone marrow and the spleen in both types of transduced mice (Fig. 6
). Interestingly, cells expressing GPIb concurrently expressed hFVIII in bone marrow obtained from mice transduced with SIV-GPIb-hFVIII (Fig. 6A
).
|
|
Finally, we evaluated whether platelet-specific gene transduction using SIV-GPIb-hFVIII resulted in phenotypic correction of FVIII-deficient hemophilia A mice. The plasma FVIII antigen concentration with or without platelet activation was measured in transplanted FVIII-deficient mice at 30 and 60 days after transplantation. We detected FVIII activity in the transplanted mice in which 1–2% correction was noted in the plasma of mice transplanted with KSL cells transduced with SIV-GPIb-hFVIII (Fig. 7
A). When platelets were stimulated with collagen and PMA, the plasma FVIII concentration increased to 2–3.5% (Fig. 7A
). The mortality rate after tail clipping was significantly improved in transduced mice (Fig. 7B
). Furthermore, ectopically expressed hFVIII levels had not attenuated, and the appearance of inhibitor against hFVIII was not detected in mice transplanted with KSL cells transduced with SIV-GPIb-hFVIII at day 60 after the transplantation (data not shown). We simultaneously performed transplantation experiments using SIV-CMV-hFVIII. Plasma levels of hemophilia A mice transplanted with KSL transduced with SIV-CMV-hFVIII were 3–6% after the transplantation, and phenotypic correction was also observed, as reported previously (19
, 24
, 25)
.
|
DISCUSSION
In this study, we examined the gene transduction of platelets and megakaryocytes by using an SIV lentiviral vector harboring a platelet-specific promoter in vivo. Since the strategy of using platelets as potential targets for producers of transgene products has already been proposed in transgenic mice (3
, 4)
, it was shown that it is possible to apply this strategy to correct hemorrhagic disorders including hemophilia by efficient platelet-directed gene transduction in vivo. However, detailed comparisons of platelet-specific promoters and the efficiency of the transduction of transgenes in vivo have not previously been reported. In our system, the transduction of hematopoietic stem cells with an SIV lentiviral vector resulted in the expression of the transgene in 20% of platelets and also resulted in a phenotypic correction of hemophilia A mice, suggesting that platelet-targeting gene therapy has the potential for further clinical applications. This is a first study to achieve a phenotypic correction of a coagulation abnormality such as hemophilia A by using platelet-directed gene transduction.
Megakaryocytes have a finite life span of 10–21 days (26)
; therefore, hematopoietic stem cells are a more practical target than megakaryocytes for genetic transfer to establish long-term expression of a target protein in platelets. Because lentiviruses are capable of infecting certain types of quiescent cells, there has been significant interest in the application of lentivirus-derived vectors to the transduction of hematopoietic cells; indeed, it has been shown that lentiviral vectors can efficiently transduce hematopoietic stem cells (27)
. We used the SIV lentiviral system for efficient platelet-targeting gene transduction because of its probable safety. The SIV lentiviral system was derived from SIVagmTYO1 and is nonpathogenic to its natural host and to experimentally infected Asian macaques (16)
. Replication-competent virus particles were not detected in vector-infected cells, and the risk of development of replication-competent lentivirus particles in HIV carrier patients may be significantly lower than that for the HIV-based vectors (19)
. Accordingly, SIV vectors have a safety advantage for clinical applications of gene therapy.
Most reported studies have used the GPIIb promoter for megakaryocyte- and platelet-specific gene transduction. We used the GPIb promoter as a platelet-specific promoter in this study because the promoter activity of GPIb was more potent than that of GPIIb in UT-7/TPO and CD34+-derived megakaryocytes. Another reason we selected this platelet-specific promoter was that the GPIb promoter works at a late stage of megakaryopoiesis. Although the GPIIb gene is expressed in platelets and megakaryocytes, it is an early gene for megakaryopoiesis (28)
. In conditional knockout mice in which the thymidine kinase gene was driven by the GPIIb promoter, the administration of gancyclovir led to a dramatic reduction in the platelet count (29)
. In bone marrow, erythroid and myeloid progenitors were also affected, which indicated the presence of GPIIb in progenitor cells (29)
. Indeed, 18% of human CD34+ hematopoietic stem cells already expressed GPIIb, and so the appearance of GPIb was markedly delayed as compared with that of GPIIb, indicating that GPIb is a later marker of megakaryocytic maturation. Platelet-targeting gene therapy using the GPIb promoter was therefore expected to allow more specific and restricted expression of gene products in platelets than that using the GPIIb promoter.
Another important finding here was that the eGFP gene driven by the CMV promoter showed significantly decreased expression in platelets, despite the high transduction efficiencies of CD45+ cells in vivo. Generally, the reduction of transgene expression caused by a shortened protein half-life is even more pronounced in terminally differentiated blood cells (30)
. The decreased expression might have been mediated by the down-regulation of the transgene during differentiation; the stability of the encoded protein is at least as relevant for the expression of a transgene as the choice of the promoter or cis-elements influencing RNA processing in differentiated cells (30)
. In this context, the use of the GPIb promoter, which drives expression in late megakaryocyte differentiation, might be important for gene transduction of terminally differentiated anucleate platelets.
Our strategy of platelet-directed gene transduction has potential for not only inherited platelet disorders (such as Glanzmann’s thrombasthenia and Bernard-Sourlier syndrome) but also other hemorrhagic disorders. Hemophilia A is an X chromosome-linked bleeding disorder caused by defects in the FVIII gene and affecting 1:5000 males (31)
. Hemophilia is considered suitable for gene therapy because it is caused by a single gene abnormality and therapeutic coagulation factor levels may well vary over in a broad range (5–100%; ref 31
). Although sustained therapeutic expression of FVIII has been achieved in preclinical studies using a wide range of gene transfer technologies targeted at different tissues (32)
, emergence of neutralizing Ab often limits their clinical applications (33)
. The targeting of hematopoietic stem cells is not an exception. Although lentiviral FVIII gene transduction of hematopoietic stem cells is able to produce therapeutic levels of FVIII (19
, 24
, 25
, 34)
, the emergence of neutralizing antibodies to FVIII has resulted in decreased levels of FVIII activity (34)
. Platelet-directed gene therapy for hemophilia A has a possible advantage for therapeutic applications, because the use of the platelet-specific system may limit the development of inhibitors by preventing the expression of FVIII in antigen presenting cells. Furthermore, 10–30% of populations with hemophilia A develop inhibitors to infusion products, which leads to the disruption of coagulation and severe bleeding (31)
. Under these conditions, platelet-directed gene therapy of hemophilia A is very attractive because platelets could specifically store the protein in the bloodstream and then specifically release it at sites of thrombus formation, thereby minimizing the influence of any circulating inhibitors. For further clinical application, the long-term observations are required to substantiate long-term in vivo gene expression because our observation periods were limited in this study.
During the course of this study, the therapeutic expression of GPIIb/IIIa in GPIIIa-deficient mice using HIV-lentivirus vector containing GPIIIa cDNA under the control of the GPIIb promoter was reported (35)
. That study used a heterogeneous population of bone marrow cells as a source for stem cell transplantation and gene transduction. We demonstrated efficient transduction of KSL murine hematopoietic cells by a SIV vector harboring the GPIb promoter and phenotypic correction of hemophilia A mice. Primitive KSL cells are a nearly homogeneous population, and a single KSL cell frequently can provide long-term multilineage engraftment of lethally irradiated mice (36)
. Targeting of primitive hematopoietic stem cells is thought to be a safer approach, because the number of transduced cells needed for reconstitution is much lower than that needed when using a heterogeneous bone marrow population. The development of leukemia in two children with severe combined immunodeficiency disease who were transplanted with retroviral vector-transduced bone marrow cells caused renewed concern about the risks associated with the integration of proviral sequences into chromosomal DNA (37)
. One way to possibly reduce the risks of insertional mutagenesis would be to use transduction protocols that minimize the total number of genetically modified cells (38)
. From this aspect, our procedure using KSL cells transduced with SIV lentiviral system is a practical approach for platelet-specific gene modification in clinical applications.
ACKNOWLEDGMENTS
The authors thank Dr. H. H. Kazazian Jr. (University of Pennsylvania, Philadelphia, PA) for FVIII-deficient mice (hemophilia A mice), Dr. A. Kume (Jichi Medical School) for K562, and Dr. N. Komatsu (Yamanashi University, Yamanashi, Japan) for UT-7/TPO. This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education and Science; Health and Labor Science Research Grants for Research from Ministry of Health, Labor and Welfare; and Grants for "High-Tech Center Research" Projects for Private Universities: matching fund subsidy from MEXT (Ministry of Education, Culture, Sports, Science, and Technology), 2002–2006.
Received for publication September 25, 2005. Accepted for publication March 9, 2006.
REFERENCES
gene. J. Biol. Chem. 270,24532-24539 gene is a major determinant of the plasma membrane levels of the platelet GP Ib-IX-V complex. Blood 94,186-191 in human endothelium. J. Biol. Chem. 265,19833-19838IIbß3 integrin and GPIb-V-IX complex identify distinct stages in the maturation of CD34+ cord blood cells to megakaryocytes. Blood 96,4169-4177IIb gene is transcriptionally active in primitive hematopoietic progenitor cells. Blood 90,2995-3004IIbß3, in a murine model for Glanzmann thrombasthenia. Blood 106,2671-2679This article has been cited by other articles:
|
|
 |
|
  H. Nishikii, K. Eto, N. Tamura, K. Hattori, B. Heissig, T. Kanaji, A. Sawaguchi, S. Goto, J. Ware, and H. Nakauchi Metalloproteinase regulation improves in vitro generation of efficacious platelets from mouse embryonic stem cells J. Exp. Med., August 4, 2008; 205(8): 1917 - 1927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Ide, B. Gangadharan, K.-Y. Chiang, C. B. Doering, and H. T. Spencer Hematopoietic stem-cell gene therapy of hemophilia A incorporating a porcine factor VIII transgene and nonmyeloablative conditioning regimens Blood, October 15, 2007; 110(8): 2855 - 2863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ohmori, Y. Kashiwakura, A. Ishiwata, S. Madoiwa, J. Mimuro, and Y. Sakata Silencing of a Targeted Protein in In Vivo Platelets Using a Lentiviral Vector Delivering Short Hairpin RNA Sequence Arterioscler. Thromb. Vasc. Biol., October 1, 2007; 27(10): 2266 - 2272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kimura, T. Ohmori, R. Ohkawa, S. Madoiwa, J. Mimuro, T. Murakami, E. Kobayashi, Y. Hoshino, Y. Yatomi, and Y. Sakata Essential Roles of Sphingosine 1-Phosphate/S1P1 Receptor Axis in the Migration of Neural Stem Cells Toward a Site of Spinal Cord Injury Stem Cells, January 1, 2007; 25(1): 115 - 124. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||
2 test.