Hematopoietic Stem Cells As Targets For Gene Therapy

A. Requirements for Effective Gene Therapy

The ability of the HSC to completely repopulate the entire hematopoietic system following transplantation makes the he-matopoietic system a particularly attractive target for a permanent correction of inherited or acquired defects (8-10,(114). Integration of new genetic material into the genome of HSC would ensure a continuous supply of modified hematopoietic cells in the transplant recipient. In addition, the repopulation of the thymus with the progeny of transduced HSC has been shown to induce immune tolerence to the transgene product, which would allow the production of a new protein in the recipient without triggering an immune response, a frequent problem encountered in gene therapy (115). In contrast, gene transfer into progenitor cells, which do not have the ability to repopulate or self-renew, would be transient and would require periodic gene transfer procedures to maintain a supply of corrected cells. Adeno-associated viruses have been shown to integrate into the genome of cells cultured in vitro, and transduction of HSC with recombinant AAV vectors has been attempted (116-118). At the present time, the frequency of recombinant AAV integration into the HSC genome has been too low to predict that AAV-mediated gene transfer to HSC would be an efficient gene transfer strategy. Similarly, adeno-virus vectors, which do not integrate into the host cell genome, would not be effective for HSC gene therapy (119). Oncore-trovirus and lentivirus vectors integrate into the host cell genome, and have been developed specifically for gene transfer into HSC (120-122). The development of oncoretroviral and lentiviral vectors is discussed in other chapters; this chapter will focus on the use of oncoretrovirus and lentivirus vectors to transduce HSC.

Successful oncoretrovirus-or lentivirus-mediated gene transfer into hematopoietic stem cells requires several critical events (123-125). First, the virus particle must enter the target cell. The interactions between the target cell and the virus are mediated by the envelope protein of the virus and specific receptor molecules on the surface of the target cells (126,127), and the choice of virus envelope can have dramatic effects on the efficiency of gene transfer. After the viral genome has entered the target cell, the reverse transcriptase (pol) protein packaged with the virus converts the RNA genome into a double-stranded DNA molecule (124,125). This step is regulated by the availability of deoxynucleotide triphosphate molecules (dNTP) in the cytoplasm. Cells in the G0 phase of the cell cycle (like HSC) typically have low levels of dNTPs, while cells in the G1 phase of the cell cycle have increased levels of cytoplasmic dNTPs in preparation for DNA synthesis (128,129). The double-stranded DNA enters the nucleus where the pol protein-mediated integration into the host cell DNA (Fig. 3) (124,125). After integration, the transferred gene must be expressed at the appropriate levels in the appropriate cell type to correct the genetic defect (121). For some hematopoietic diseases, such as ADA deficiency, the expression of the ADA enzyme in hematopoietic cells of all lineages might have a beneficial effect (10). However T-lymphocytes must express the ADA gene at relatively high levels for gene therapy to be effective (130,131). For other diseases such as IL-2 receptor common chain (7c) deficiency (X-SCID), expression in lymphoid cells in essential (132), while expression of 7c in myeloid cells would have no beneficial effect on the disease (133).

B. Oncoretrovirus-mediated Gene Transfer

The Moloney Murine Leukemia Viruses (MMuLVs) were originally studied as the causative agents of leukemia in mice (16,17). The MMuLV life cycle is very well understood (124,125), and despite the fact that replication competent MMuLVs can cause leukemia in mice and monkeys, MMuLVs were the first retroviruses to be adapted for gene therapy (122). Murine retroviruses are classified based on differences in their envelope proteins. The envelope protein of ecotropic retroviruses binds to a murine basic cationic amino acid transport protein, mCAT (134,135). The 3 amino acid sequence of this protein that serves as the virus binding site is not conserved in CAT from other mammals (136-138), restricting ecotropic retrovirus transduction to mouse cells. The envelope protein from amphotropic retroviruses binds to a phosphate transporter protein, Pit-2 (139-141). The binding site for the amphotropic envelope is conserved among the different mammalian Pit-2 molecules, giving amphotropic viruses a broad host range that includes human cells. Currently, most human gene therapy trials have involved amphotropic retroviruses (123).

Other retrovirus envelopes have been adapted for packaging recombinant retrovirus vectors. The envelope protein of the Gibbon Ape Leukemia Virus (GALV) binds to a distinct phosphate channel protein known as Pit-1 (142,143), which is approximately 60% identical to Pit-2 at the amino acid level (139,140). The Pit-1 amino acid sequence that binds the GALV envelope protein is conserved in most mammalian Pit-1 proteins with the exception of mouse Pit-1. Recombinant virus particles with a GALV envelope (144) have been used to improve gene transfer into mature lymphocytes (145,146). The 10A1 virus is a laboratory recombinant whose envelope protein binds to mouse and human Pit-1 and Pit-2 (147).

The Vesicular Stomatitis Virus type G (VSV-G) envelope recognizes a cell membrane phospholipid as a receptor, allowing VSV-G retrovirus particles to transduce almost all cell types regardless of species (148-150). In contrast to most other retroviruses, virus particles containing the VSV-G envelope can be agglutinated and concentrated to titers 2 orders of magnitude greater than can be obtained with other retroviruses (151-153). The envelope of the endogenous feline retrovirus RD114 recognizes a neutral amino acid transporter protein, and the binding site is conserved among larger mammals, but not mice (154). Recombinant retrovirus vectors with the RD114 envelope have been used to transduce high levels of CD34+ cells that engraft into NOD-SCID mice and fetal sheep (155,156).

A variety of recombinant MMuLV gene transfer vectors has been described in previous chapters, each of which has advantages that may be exploited for gene transfer applications [for review see (122)]. The LTRs from different murine retroviruses express downstream transgenes differently in different cell types. Each LTR has its own advantages in terms of the level of expression and the cells in which the highest levels have been achieved. In general, MMuLV LTRs are most active in lymphoid cells, while the Harvey Murine Sarcoma Virus (157), Myeloproliferative Sarcoma Virus (MPSV), and Mouse Stem Cell Virus (MSCV) LTRs promote relatively high levels of transgene mRNA in both myeloid and lymphoid cells (158-160).

It is well documented that transgene expression in hemato-poietic cells can be silenced over time (161). A detailed analysis of active and silent proviruses derived from the same hema-topoietic stem cell identified site-specific methylation within

Figure 3 The relationship between the cell cycle and oncoretrovirus-and lentivirus-mediated gene transfer. RNA virus-mediated gene transfer is critically dependent on the abundance of virus receptors and the availability of deoxyribonucleotides for reverse transcription. The lentivirus life cycle differs from the oncoretrovirus life cycle in that lentivirus DNA genomes can be transported into the nucleus while oncoretrovirus integration requires the breakdown of the nuclear envelope during M phase.

Figure 3 The relationship between the cell cycle and oncoretrovirus-and lentivirus-mediated gene transfer. RNA virus-mediated gene transfer is critically dependent on the abundance of virus receptors and the availability of deoxyribonucleotides for reverse transcription. The lentivirus life cycle differs from the oncoretrovirus life cycle in that lentivirus DNA genomes can be transported into the nucleus while oncoretrovirus integration requires the breakdown of the nuclear envelope during M phase.

the MMuLV LTR is associated with the silencing of gene expression (161). Mutation of the methylation sites combined with deletion of a negative regulatory region in the virus has dramatically improved the duration of gene expression in mouse hematopoietic cells (162,163). Gene expression from the MSCV LTR promoter has also been shown to be resistant to silencing (164).

C. Lentivirus-mediated Gene Transfer

After the RNA virus genome is converted to DNA, it must enter the nucleus for integration into the host cell DNA (123,124). This requirement limits oncoretrovirus-mediated gene transfer into HSC because oncoretrovirus DNA can only gain access to the genome when the nuclear envelope breaks down during mitosis. Since HSCs are usually quiescent or have a prolonged cell cycle (22,165-167), oncoretrovirus DNA must remain intact in the cytoplasm for a prolonged period of time. Lentiviruses, of which the Human Immunodeficiency Viruses HIV-1 and HIV-2 are the most familiar examples, have been shown to be able to integrate into the genome of nondividing cells (168). Similar results have been described with feline-derived lentivirus vectors (169,170). As described in previous chapters, the nuclear transport is mediated by two virion proteins, matrix and Vpr, which promote the transport of the HIV preintegration complex to the nucleus using the host cell nuclear transport machinery (171,172). The majority of lentivirus vectors are packaged with the VSV-G envelope (173), but recent studies have shown that lentivirus can also be packaged with amphotropic and RD114 envelopes (154,174). Compared to oncoretrovirus vectors, the lentivirus vectors integrated into irradiated or cell cycle arrested fibroblast cell lines at 5-to 20-fold greater efficiency (173,175). These studies have confirmed high rates of transgene integration into nondividing cells, including primitive human hematopoietic progenitor cells. However, direct comparisons in mouse models has shown that the efficiency of gene transfer into HSC is similar to that of oncoretrovirus vectors (176).

To date HIV vectors have been used only for gene transfer in animal models. The best application of the lentivirus technology has been the development of stable vectors containing the human p-globin gene and regulatory elements (177,178). For many years the inability of oncoretrovirus vectors to transfer an intact p-globin expression cassette had frustrated attempts to develop gene-based therapies for Sickle Cell Disease and the thalassemias (179). Several lines of evidence suggested that splicing of the p-globin retrovirus genomic RNA was responsible for the failure of an intact vector to integrate into the host cell genome. Lentivirus vectors are packaged in the presence of the rev protein, which suppresses splicing of the lentiviral RNA, and containing the p-globin cassette were introduced intact into mouse HSC, with expression of therapeutic levels of p-globin (177,178). It is clear that there will be many safety concerns to address in the future in the design of HIV packaging and gene-transfer vectors. It appears that many of the modifications made to improve the safety of onc-oretrovirus vectors could be used to improve the safety of lentivirus vectors.

Was this article helpful?

0 0

Post a comment