The following section outlines the 3 phases of a retroviral-based hematopoietic stem cell (HSC) and lymphocyte gene transfer protocol: 1) hematopoietic cell procurement, 2) exvivo gene transfer, and 3) reinfusion of the corrected cells. (Fig. 1)
A. Gene Transfer to Hematopoietic Stem Cells
Collection of large numbers of autologous hematopoietic stem cells is the first step in an HSC gene therapy protocol. The 3 sites from which HSC can be harvested are: 1) bone marrow, 2) cytokine or chemotherapy/cytokine-mobilized peripheral blood, and 3) umbilical cord blood. Although the precise phe-notype of a true HSC is unknown, large numbers of HSCs are contained in a population of cells expressing the CD34 antigen (12). CD34+ cells make up 0.5-5% of nucleated cells in the bone marrow and only a fraction of these are HSCs. HSCs can be safely aspirated from bone marrow of the poste
rior superior iliac crest in a minor operative procedure. The major side effects of this procedure include mild discomfort at the aspiration site and an occasional hematoma. For smaller individuals, symptomatic anemia may require a blood transfusion, which, if anticipated in advance, can be an autologous unit. The major disadvantage of large volume bone marrow aspiration is that it must be done under general anesthesia, which adds risk to the procedure. Because of these issues, repeated large-scale marrow harvests are not desirable.
In most individuals, administration of Granulocyte-Colony Stimulating Factor (G-CSF) at 10-16 ^g/kg per day subcutane-ously for 5 or 6 days results in a transient 20- to 100-fold increase in the frequency of CD34+ hematopoietic progenitor and stem cells in the circulation (a phenomenon termed stem cell mobilization). Other growth such as Flt-3-ligand (Flt3-L) (13-15), Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) (16-19), and Stem Cell Factor (SCF) (20,21), which are used alone or in combination, are also effective for stem cell mobilization. Large numbers of CD34+ cells (2-10 X 106 cells/kg patient weight) may be obtained from a single or repeated apheresis procedure following mobilization with growth factors. Furthermore, the mobilization process may be safely repeated at 4- to 8-week intervals. Unlike the procurement of bone marrow, no operative procedure is needed and the entire process can be done without a hospital admission.
Dunbar and coworkers (22) have demonstrated improved gene marking in nonhuman primates when peripheral blood HSC mobilized with G-CSF and SCF are compared to bone marrow HSC. The same group later showed superior in vivo gene marking of nonhuman primate HSC mobilized with G-CSF and SCF compared to HSC mobilized with G-CSF and Flt-3 ligand or G-CSF alone (23). High resolution cell cycle analysis of cytokine mobilized peripheral blood HSC has revealed that these cells are more likely to have entered cell cycle and express higher levels of the amphotropic receptor mRNA (24-26). While certainly easier to collect, human gene marking studies have yet to demonstrate unequivocally that cytokine-mobilized peripheral blood HSCs are better targets for retroviral transduction than bone marrow HSCs. However, the logistical advantages to obtaining peripheral blood HSCs dictate these as the preferred source for gene therapy and other transplant purposes except when the donor has a poor CD34 + mobilization response to cytokine adminstration.
Umbilical cord blood contains a higher concentration of primitive hematopoietic progenitors than bone marrow (27). Recent data suggests that the HSCs from umbilical cord blood may also express higher levels of the amphotropic retrovirus receptor (28), which may result in more efficient transduction with amphotropic retroviral vectors. On average 20 X 106 CD34+ cells can be collected from the placenta at the time of delivery, which is approximately 10-fold less than what can be collected from mobilized peripheral blood of an adult. Because of the low efficiency of HSC transduction with current techniques, clinical application for gene-corrected umbilical cord blood stem cells may only be practical in neonates.
2. Ex Vivo Gene Transfer
Optimization of ex-vivo retroviral transduction conditions for HSCs has proven to be a formidable task. It appears that 48
to 72 h of culture in growth factors is required for quiescent lineage negative CD34+ cells to enter the cell cycle and thus become receptive targets for retroviral tranduction (29). However, studies have shown that prolonged ex vivo culture of HSCs results in loss of long-term repopulating ability as a consequence of lineage commitment and of an acquired defect in the ability of cycling cells to engraft (30,31). This loss of repopulating potential with ex vivo culture may be gradual and to some degree reversible. Takatoku and coworkers manipulated the ex vivo culture conditions and growth factor combinations such that HSCs from nonhuman primates were first induced into cell cycle by using a combination of active cytokines, thereby facilitating retroviral transduction. Then, before the cells were transplanted back into the animal, they were returned to a quiescent state by incubating the cells in media containing only stem cell factor. This method of ''resting'' stem cells prior to reinfusion resulted in improved en-graftment of gene-marked cells (32).
Measurable gene transfer into HSCs in the clinical setting has been reported with ex vivo transduction periods ranging from 6 to 72 hours (33,34). While the report using a 6-hour regimen appeared to succeed in achieving measurable gene transfer without use of growth factors (33), most investigators have found that growth factors and an ex vivo culture period of 48-96 hours are required for optimum transduction. Growth factors are also essential to prevent apoptosis of HSCs during prolonged ex vivo culture.
Some investigators have used autologous bone marrow stromal cell layers in an attempt to enhance transduction with retrovirus vectors while preserving reconstitutive potential (35-38). Despite encouraging results in small animal models, there is no evidence that use of bone marrow stromal layers has resulted in better outcome in human clinical studies. Not only does the establishment of autologous stromal culture require a bone marrow aspiration, but this approach greatly increases the complexity and handling involved during clinical scale-up. Moreover, to achieve the highest level of transduc-tion efficiency of HCSs grown on stromal layers, it is still essential to add growth factors to the culture.
Much current investigation is focused on determination of growth factor combinations used without stromal layers that can achieve highest transduction while maintaining reconstitutive potential of the transduced HSC. Flt3-L (39) and thrombo-poietin (TPO) (39,40) have emerged as important agents to add to the ex vivo culture in relatively high concentrations (100-300 ng/ml) to achieve these dual goals. These growth factors work optimally in synergy with other growth factors. SCFs at more modest concentrations of 50-100 ng/ml appear to be important as well. Interleukin 3 (IL-3) and interleukin (IL-6) have also been widely used ex vivo in clinical trials to enhance cycling and may help to prevent apoptosis of HSC. However, use of concentrations of IL-3 higher than 30-50 ng/ml may be detrimental to preservation of cells with long-term engraftment potential. Other factors essential for maintenance and development of lymphoid progenitors from stem cells, such as IL-7, have not been used clinically, but may in the future prove to be a growth factor for the transduction of lymphocytes (41,42).
A number of techniques have been devised to encourage the interaction of hematopoietic progenitors with viral particles during the ex vivo transduction period. Most investigators have opted to transduce a cell population enriched for CD34 expression. CD34+ cell enrichment enables an improved stem cell/viral particle ratio, thereby optimizing the multiplicity of infection while using less of the valuable clinical grade ret-roviral supernatant. A variety of stem cell selection devices that use monoclonal antibodies specific for the CD34 antigen have been employed in experimental clinical protocols [Reviewed in (43,44)]. These devices are able to select large numbers of CD34 + cells from a bone marrow or mobilized peripheral blood apheresis graft at 50-80% efficiency yielding a product which consists of 60-90% CD34+ cells.
For reasons that are not well defined, centrifugation of target cells during incubation in a retrovirus vector supernatant increases transduction efficiency, a technique that has been termed ''spinocculation'' (45). The g-forces employed to achieve the effect are as low as 1200 x g for 20 min, making it unlikely that the effect is due to sedimentation of individual virus particles. Cocultivation of the target cells on a confluent layer of retrovirus producer cells has also been shown to enhance transduction. However, regulatory issues related to the safety of cocultivation of HSC with producer lines make this approach impractical for clinical application.
One of the more exciting techniques to be described is the finding that a specific proteolytic fragment of fibronectin facilitates stem cell-retroviral particle interaction when this peptide is used to coat the surface of the culture vessel (46). Fibronectin, a prominent component of the extracellular matrix of bone marrow stromal cells, has numerous hematopoi-etic cell-binding domains. Moritz and coworkers have demonstrated binding of both viral particles and hematopoietic target cells to a proteolytic fragment of fibronectin that contains the CS1 binding site (47). The CS1 binding site of fibronectin interacts with the VLA4 adhesion molecule found on hemato-poietic stem cells (48). Thus, when HSCs are incubated with retroviral particles in the presence of this specific fibronectin fragment, transduction efficiency is improved.
The availability of a clinical grade recombinant human C-terminal fibronectin fragment (CH-296) has facilitated its use in several clinical gene therapy trials targeting HSC. Even with retrovirus vectors of modest titer, acceptable transduction of CD34 + cells can be achieved. Of note is that with retrovirus vectors at titers >5 x 106 infective particles per mL, the use of CH-296 fibronectin fragment coated culture vessels can achieve transduction of 50-70% of CD34+ cells routinely at clinical scale. It is also possible that use of fibronectin fragment coating may help to preserve the long-term engraftment potential of cultured HSC (49).
Following transduction, the HSCs are infused into a peripheral vein of the patient. Within 24 h, the majority of the HSCs have homed to the bone marrow. Experience with stem cell transplantation for treatment of hematological malignancies has shown that the bone marrow can be completely reconstituted by transplanted HSCs (autologous or allogeneic) following myeloablative doses of chemotherapy and/or radiation. Because loss of long-term repopulating ability may occur during ex vivo transduction of HSC, it is unethical to rescue hemato-poiesis in a myeloablated patient with ex vivo manipulated HSCs only. However, preclinical and clinical studies suggest that some degree of cytoreductive therapy administered prior to infusion is required to establish clinically relevant levels of gene marking. Using clinically applicable tools, long-term marking at levels of 5-10% can be achieved in nonhuman primates following the administration of high-dose total body irradiation (30,50,51). Long-term multilineage gene marking has not been demonstrated in large animals or humans without bone marrow conditioning. Aiuti and colleages were the first to employ nonmyeloablative bone marrow conditioning in a human gene therapy study for children with adenosine deami-nase-deficient severe combined immunodeficiency (discussed in more detail below) (52).
B. Gene Transfer to Lymphocytes
With few exceptions, large numbers of lymphocytes circulate in the peripheral blood and are therefore easily harvested from gene therapy candidates using apheresis. Lymphocytes may also be harvested from special sites such as tumors. These cells are of particular interest since they may possess unique antitumor properties.
Compared to HSC, fewer hurdles exist in the quest to optimize retroviral gene transfer of lymphocytes. Since these cells are terminally differentiated, loss of phenotype during ex vivo manipulation is not a concern. T-lymphocytes, which are the most common target for lymphocyte-based gene therapy, are expanded in culture with agents such as IL-2 and/or monoclonal antibodies to CD3. While being cultured, many of the cells are stimulated into active phase of the cell cycle and become susceptible to permanent retroviral integration. It has been observed that with long-term culture of T lymphocytes, enrichment of CD8 T lymphocytes relative to CD4 T lymphocytes develops. This issue may be addressed by altering the ratio of cells added to the initial culture mixture (53).
Techniques that have been shown to improve lymphocyte transduction efficiency include: 1) the use of retroviral vectors pseudotyped with the Gibbon Ape Leukemia Virus (GALV) envelope protein, 2) upregulation of amphotropic or GALV retroviral receptor expression by growth in phosphate depleted media, and 3) transduction in a culture vessel coated with the CS1 fibronectin fragment (54,55). Incorporation of these techniques together in the same protocol can yield transduc-tion efficiency of 50-60%.
Reinfusion of the gene-corrected lymphocytes takes place as would any routine infusion of cell products. Cytoreductive conditioning of the recipient appears not to be necessary to achieve persistence of the transplanted lymphocytes, particularly where the therapeutic gene may provide a survival advantage (i.e., correction of ADA SCID).
The transition from a laboratory-based gene transfer assay to one that is ready for inclusion in a clinical protocol can be quite challenging. The most obvious differences relate to the number of hematopoietic progenitors that must be transduced at one time. Large volumes of retroviral supernatant must be produced in a facility licensed to provide clinical grade material. It is common to find a decrement in the viral titer of the clinical material compared to what is obtained in the laboratory. Besides the requisite sterility and endotoxin testing, the product must always be tested for the presence of replication-competent retrovirus. Regulatory and proprietary issues regarding use of reagents or devices often hinder the ability to replicate in the clinical setting what is done with ease in the laboratory. Performing retroviral transduction in a clinically approved facility may require modifications of a laboratory optimized assay (56). Use of closed-system, gas-permeable, culture containers compatible with the standard sterile transfer techniques used by blood banks is one method that has been adapted for this purpose (57).
Was this article helpful?