3.1. Ex Vivo Gene Therapy Using Retroviral and Lentiviral Vectors 3.1.1. Retroviral Gene Transfer to Hematopoietic Cells: General Considerations
The use of hematopoietic stem cells (HSC) in bone marrow transplantation approaches have convincingly demonstrated the potential of this approach to repopu-late the different hematologic lineages in mice and in humans, and has found application not only in the treatment of hereditary diseases but also in myelo-reconstitution after high-dose chemotherapy and in controlling graft vs host disease. Because it was shown that HSC can be persistently transduced ex vivo with retroviral vectors (22), these vectors have been extensively used in gene replacement or augmentation and anticancer strategies involving the hematopoietic system.
Numerous technical refinements developed over more than a decade have greatly increased the efficiency of murine retroviral vectors for transduction of HSC, as reflected by the recent clinical utility of this approach (23,24). In a study performed by Alain Fischer and colleagues in Paris, an optimized protocol for ex vivo gene transfer into hematopoietic progenitor cells was employed to achieve successful retroviral vector-mediated expression of the interleukin receptor common y (yc) chain, a component of several cytokine receptors that is defective in severe combined immunodeficiency-Xl (SCID-X1) disease (25). Building on a decade of experience, that has led to incremental yet cumulative improvements in the efficiency of retrovirus-mediated gene transfer to hematopoietic progenitors, Fischer and colleagues applied the appropriate cytokine-mediated stimulation including, stem cell factor (SCF), Flt-3 ligand, megakaryocyte growth and differentiation factor (M-GDF), and interleukin-3 (IL-3), to induce CD34+ proliferation without loss of lymphoid or myeloid potential, resulting in increased trans-duction efficiency (26). Transduction of yc-deficient bone marrow cells was performed using a retrovirus containing the yc gene based on the simple MFG vector backbone produced from YCRIP packaging cells and optimized procedures including immobilization of the vector and target cells on fibronectin-coated tissue culture plates. Subsequent transplantation into a SCIDX1 mouse model resulted in normal levels of immunoglobulins, normal T- and B-cell interaction, and the presence of lymphocytes 47 wk post-treatment (25).
This success has further been extended to include the full correction of SCIDX1 in humans. The CD34+ cells taken from bone marrow of two patients, aged 8 and 11 mo, were transduced with an efficiency of between 20 and 40% (23). Two weeks after replacement of the transduced cells, the yc transgene was detected in the blood and T-cell levels had increased (23). At 10 mo post-treatment, both patients exhibited T and NK cells expressing the yc transgene and functioning at levels similar to normal controls of identical age (23). Since then, an additional nine patients have been treated in this manner, with successful reconstitution with corrected cells in all but one case. This was also the first demonstration of a selective growth advantage for genetically corrected cells reintroduced into humans, a hitherto hypothetical idea, that previously had not been possible to definitively demonstrate in the adenosine deaminase (ADA) gene therapy trials because of the continued administration of polyetylene glycol (PEG)-ADA. Long-term follow-up will be necessary to determine how much transduction was achieved in the earliest multipotent, self-renewing hematopoietic stem cell population. Nevertheless, this success illustrated the usefulness of MMLV vectors applied using optimized ex vivo transduction procedures in the setting of a well-thought-out clinical application in which even relatively low levels of corrected cells can achieve therapeutic efficacy through conferral of a selective advantage. Unfortunately, this trial also demonstrated for the first time in humans the potential for retroviral insertional events to contribute to the development of oncogenesis; an aspect, which will be discussed below and in other chapters separately.
More recently, however, it has become evident that lentiviral vectors are more efficient at transducing quiescent HSC than murine retroviral vectors. Thus, some of the specific applications of hematopoietic gene transfer with implications for cancer gene therapy using retroviral vectors are presented below, with emphasis on lentiviral vectors.
3.1.2. Efficiency of Retroviral/Lentiviral Vector Gene Delivery Into Long-Term Repopulating Hematopoietic Progenitors
As noted above, HSC are predominantly quiescent cells in the G0/G1 phase of the cell cycle. It has been demonstrated convincingly by several groups that lentiviral vectors, which encode proteins that permit active import of the viral genome into the nucleus of nondividing cells, are a more efficient system for gene transfer to HSCs than other retroviral vectors. Early observations based on the initially available generations of lentiviral vectors used to transduce purified human HSC, showed higher rates of ex vivo gene delivery by lentiviral vectors compared with retroviral vectors whether in the presence or absence of growth factors (27-29). Subsequently, animal models of hematopoietic reconstitution showed that CD34+ cells transduced with lentiviral vectors were capable of stable, long-term reconstitution of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice. These studies have shown long-term expression of marker genes from 15 to 22 wk in lymphoid, myeloid, and erythroid progeny and also engrafted human cells that retained the CD34+ phenotype (30,31). Importantly, analysis of human progenitor cells isolated from bone marrow of NOD/SCID recipient animals showed that the overall percentage of gene marking in colony-forming cells by microscopy was equivalent to the percentage of provirus sequences by polymerase chain reaction (PCR) analysis, indicating minimal transgene silencing in vivo (32). Later, repopulating assays in NOD/SCID mice with lentivirus-transduced human HSC cells included dose-response analyses to determine the minimal MOI (and hence lower numbers of virus integrants/cell) required to produce consistent gene marking in serial transplantation assays, demonstrating that even at low MOI (3) and in the absence of cytokines, lentiviral vectors were able to consistently produce marking in self-renewing, multi-potent and long-term repopulating hematopoi-etic cells (33-35). It is interesting to note that lentiviral vectors and MMLV vectors seem to transduce mouse HSC (lineage-negative cells obtained from bone marrow) with similar efficiency when performed in the presence of growth factors (IL-3, IL-6, and stem cell factor), demonstrating that, although not totally essential for transduction, entry into the cell cycle favors optimal lentiviral transduction (36,37). The comparison of lentiviral and oncoretroviral vectors has further been extended to nonhuman primate models. In baboons, efficient lentiviral gene transfer of HSC was dependent on the presence of cytokines during transduction (38). Surprisingly, as the result of a posten-try restriction to HIV infection in rhesus macaques, although low levels of marking in rhesus cells could be detected, it was generally poor (39-41). Further studies assessing lentiviral vector-mediated gene transfer into HSC and transplantation into primates are currently being performed by several groups to allow long-term evaluation of safety, maintenance of gene expression, and potential immune responses against transgene products in large animal models. Notably, in one recent study, it was noted that mobilized CD34+ cells transduced with lentiviral vectors expressing enhanced green fluorescent protein (EGFP) transplanted into myeloablated rhesus macaques resulted in the induction of specific immunological tolerance toward the foreign transgene (42).
3.1.3. Approaches Exploring Hematopoietic Reconstitution for Cancer Gene Therapy
Retrovirus-mediated gene transfer and overexpression of cytostatic drug-resistance genes has been envisaged as a myeloprotective strategy that would permit chemothera-peutic dose escalation beyond normally tolerated levels following bone marrow transplantation (BMT) and reconstitution with transduced hematopoietic progenitors. When used to express the human multiple drug resistance (MDR-1) gene, retroviral vectors significantly improved protection to cytostatic drugs in vitro in transduced hematopoietic cell lines and in HSC transplanted into mice (43,44).
When MDR-1 transduced HSC were injected into nonirradiated mice, high levels of long-term engraftment and conferral of chemoprotection were observed (45,46). Interestingly, it appears that MDR-1 overexpression may somehow alter the ability of HSC to respond to cytokines in culture, as significant expansion of the repopulating cell fraction of HSC after MDR-1 transduction at high-copy-number by retroviral gene transfer has been reported during ex vivo culture in the presence of IL-3, IL-6, and SCF, but without any drug selection; in contrast, such expansion was not observed after retroviral gene transfer of the dihyfrofolate reductase (DHFR) gene, also a drug selection marker (47). However, mice transplanted with these expanded stem cells developed a myeloproliferative disorder characterized by high peripheral white blood cell counts and splenomegaly (47). These preclinical results demonstrate that enforced stem cell self-renewal divisions can have adverse consequences. Nonetheless, using transduced HSC with significantly lower copies of MDR integrations per cell (about 1-2 vs >10), studies by other groups did not show any signs of myeloproliferative disorder in transplanted mice (46). Therefore, the rationale for conducting clinical trials with optimized retroviral vectors containing drug resistance genes such as MDR-1 to prevent chemotherapy-induced myelosuppression is still under consideration (46), although this elegant approach faces intense scrutiny prior to clinical implementation.
In this context, it should also be noted that this general strategy for in vivo drug selection and amplification of transduced hematopoietic progenitors is now being further explored with other drug resistance genes, such as the P140K variant methylgua-nine methyl transferase gene (MGMT). This drug resistance gene confers resistance to O6-benzylguanine (BG) and temozolomide (TMZ), as well as 1,3-bis (2-chloroethyl)-1-nitrosourea (BCNU), allowing efficient in vivo selection of transduced hematopoietic cells even without extensive prior myeloablation (48,49). In some studies, it has been reported that this strategy can achieve such efficient amplification and 80 to 90% of circulating cells are found to be transduced (50,51), to confer protection against chemotherapy-induced myelosuppression and therapeutically relevant levels of donor chimerism even in an allogeneic transplant setting (51). Dose-limiting hematopoietic toxicity of conventional chemotherapy remains a major problem, as observed in a recent phase II trial of BG/BCNU in patients with nitrosourea-resistant glioma (52), hence the pursuit of innovative gene transfer strategies to confer myeloprotection are certainly warranted.
Another interesting concept to improve the efficacy of anticancer treatments with gene therapy is to manipulate the immune response after hematologic transplants. Following allogeneic BMT or HSC transplantation in leukemia/lymphoma treatment, donor lymphocytes are known to mediate a graft-vs-leukemia effect (GVL). However, a major problem with this approach is the potential development of graft-vs-host diseases (GVHD). One promising solution to prevent GVHD is to genetically modify donor T-cells with a suicide mechanism that can be activated by administration of a prodrug if this life-threatening complication should occur. This strategy has been confirmed preclinically by several groups using retroviral vector-mediated gene transfer of the herpes simplex virus thymidine kinase (HSV-tír) suicide gene, which encodes an enzyme that phosphorylates and thereby activates the antiviral prodrug ganciclovir (GCV) (53-55). The clinical utility of this system was recently confirmed in a phase 1 study with eight leukemia/lymphoma patients who relapsed and were subsequently treated with donor lymphocytes transduced with the HSV-rt suicide gene. The transduced lymphocytes survived for up to 12 mo, resulting in antitumor activity in five patients. Three patients developed GVHD, which could be effectively controlled by GCV-induced elimination of the transduced cells (56).
Another potential scenario for the utilization of retroviral gene transfer is to "repair" genetic damage caused by translocations, widely seen in hematologic malignancies. Chronic myelogenous leukemia (CML) and Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL) are malignant diseases caused by gene rearrangement resulting in the formation of the abnormal fusion protein BCR/ABL. Despite the high remission rate initially obtained by the advent of new drugs such as Gleevec, a small molecule inhibitor of BCR/ABL kinase activity, it has become increasingly clear that the emergence of drug resistant clones eventually results in relapse; hence, more definitive treatment strategies are still being sought. The presence of the BCR/ABL oncoprotein is a necessary event for malignant transformation seen in CML and Ph+ ALL. Thus, genetic modification of HSC in order to eliminate expression of BCR/ABL might render transduced CML and Ph+ ALL stem and progenitor cells functionally normal. This approach has been successfully shown to work by the use of retroviral vectors expressing antisense RNA (57), ribozymes (58), and RNAi (59,60). More recently, third-generation lenti-viral vectors expressing ribozymes directed specifically against the fusion joint in the BCR/ABL transcript were used to transduce primary Ph+ ALL and CD34+ cells, resulting in growth inhibition and apoptosis specifically in the leukemic blasts (61). Thus, allied with the high efficiency of the lentiviral vector system to transduce HSC, this technology has the potential to develop into a realistic treatment modality for patients with CML or Ph+ ALL.
3.1.4. Transduction of Dendritic Cells for Anticancer Vaccines
Dendritic cells (DCs) provide the most potent pathway for initiating T- and B-cell immune responses (62). Myeloid DC precursors derived from peripheral blood, bone marrow or cord blood can be differentiated in vitro and used for immunization with peptides, protein, cDNA, RNA, or cell extracts (63,64). CD14+ monocytes are a naturally abundant cell population in the peripheral blood, which is an easily accessible source for production of DCs. Plastic-adherent peripheral blood monocytes can differentiate into "immature DCs" if a mixture of cytokines is added to the culture (65,66). After differentiation, DCs do not proliferate, and therefore attempts to transduce them with MMLV vectors were not successful.
Thus, adenoviral vectors, which are capable of transducing nonreplicating cells, have been traditionally used to transduce DCs. To reach efficient transduction, however, adenoviral vectors have to be used at high multiplicity of infection (MOI = 100-1000) (67-69) which can produce cytopathic and cytotoxic effects. Furthermore, the commonly used adenoviral vectors are themselves highly immunogenic in humans, which may hamper immune responses to weaker "self" tumor antigens (70), or trigger the rejection of transduced cells coexpressing adenoviral antigenic determinants (71). In addition, it was shown that transduction of mouse DCs with null adenoviral vectors at high MOI (>100) induces some degree of activation by itself (72), with unpredictable effects on the instruction of immune responses by these DC in vivo.
In contrast to these potential unwanted side effects of using the adenoviral vector system, lentiviral vectors offer an approach by which simple, efficient, persistent, non-toxic, and nonimmunogenic gene delivery into monocytes and DCs may be obtained. HIV-1 is naturally effective in infecting dendritic cells and monocytes, and a number of groups, including our own, have demonstrated that lentiviral vector transduction is a suitable methodology for efficient and persistent gene delivery into ex vivo differentiated DCs (13,73-75), and into monocytes obtained from peripheral blood mononuclear cells (PBMC) (76). Transduction of DCs with lentiviral vectors expressing the green fluorescent protein (GFP) did not alter their viability, immunophenotype or the ability to differentiate into mature DCs capable of stimulating autologous T-cell responses (75). In a demonstration of their immunostimulatory functionality, lentivirus-transduced DCs expressing an antigenic HLA-A2.1 restricted Flu peptide were able to effectively activate autologous Flu-specific CTL responses (74). We have shown that improved and safer third-generation self-inactivating lentiviral vectors very efficiently delivered GFP and CD40L genes into DCs with an average transduction efficiency of 70% (13). After transduction, DC maturation and activation was stimulated only by the vector containing the CD40L immunocytokine transgene, but not by a control vector expressing only the GFP marker transgene, indicating that
Cytokine treatment (GM-CSF, IL-4, FLT3-L)
T helper stimulation or soluble factors (CD40L, LPS, TNF-a )
Cytokine treatment (GM-CSF, IL-4, FLT3-L)
T helper stimulation or soluble factors (CD40L, LPS, TNF-a )
Immature Dendritic Cell
Fig. 6. Lentiviral vector-mediated genetic modification of dendritic cell precursors or immature dendritic cells. Conventional methods for dendritic cell differentiation and maturation require treatment with recombinant cytokines and stimulation with T-helper cells or other soluble factors (upper arrows). Instead, lentivirus-mediated gene transfer (lower arrows) can be used to achieve endogenous production of immunomodulators, thereby triggering autonomous differentiation.
lentiviral transduction per se is unlikely to cause DCs to differentiate, mature, activate, or otherwise engage in unpredictable immune stimulation. Transduction of DCs with the RRL-CD40L vector correlated with a mature DC phenotype as shown by morphology, upregulation of CD83 and other immunological relevant markers and production of IL-12 (13) (see Fig. 6). Autologous responses against an HLA-A2-restricted tumor associated antigenic peptide (gp100) and against an influenza peptide (Flu-M1) were significantly enhanced at non-saturating effector/target ratios when CD40L transduced DCs were used as antigen-presenting cells for in vitro stimulation of CD8+ cytotoxic T-lymphocytes (13).
Recently, we have evaluated a one-hit lentiviral transduction approach for genetic modification of monocytes in order to promote autocrine and paracrine production of factors required for their differentiation into immature DCs (76). High-titer third-generation self-inactivating lentiviral vectors expressing granulocyte-macrophage colony stimulating factor (GM-CSF) and interleukin-4 (IL-4) efficiently achieved simultaneous and persistent codelivery of the transgenes into purified human CD14+ monocytes (see Fig. 6). Coexpression of GM-CSF and IL-4 in monocytes was sufficient to induce their differentiation into lentivirus-modified DCs ("DC/LVs"), as evidenced by their morphology, immunophenotype, and immune-function. Mixed lymphocyte reactions showed that the T-cell stimulating activity of DC/LVs was superior to that of DCs grown by conventional methods. DC/LVs displayed efficient antigen-specific, major histocom-patability complex (MHC) Class-I restricted stimulation of autologous CD8+ T-cells, as shown by interferon (IFN)-y production and CTL assays. Importantly, DC/LVs exhibited a longer lifespan in culture and could be maintained metabolically active and viable in culture for 2 to 3 wk in the absence of exogenously added growth factors, compared with DCs cultured by conventional methods (76).
3.1.5. Autologous Leukemia/Lymphoma Cell Vaccines
Relapse remains one of the most important clinical problems in leukemia and lymphoma and immune therapeutic strategies designed to eradicate residual disease hold promise and are an attractive option. The demonstration of immune responses against leukemia and lymphoma associated antigens supports the concept that normal immune mechanisms can effectively target leukemia/lymphoma cells. It is thus possible that major improvements in long-term survival for leukemia/lymphoma patients could be potentially achieved if a host immune response to several leukemia antigens could be enhanced to eradicate minimal residual disease after use of induction and consolidation chemotherapy. Therefore, leukemia/lymphoma cell vaccines would be expected to result in the presentation of multiple antigens without requiring knowledge of the precise identity of each antigen. However, inefficient antigen presenting cell (APC) function, and the poor reactivity of autologous anti-leukemia/lymphoma T-cell mediated immunity, are associated with the inability of leukemia/lymphoma cells to provide sufficient co-stimulation to autologous T-cells. Indeed, we and others have found that requisite immune costimulators such as CD80 (B7.1) are frequently lacking leukemic cells. This provides the rationale to modify leukemia/lymphoma cells into efficient APCs by genetic manipulation, which has been tested and confirmed preclini-cally in a variety of models (for review see ).
As human leukemia/lymphoma cell vaccines have moved towards clinical trials, different types of vectors to genetically engineer human leukemia cells have been tested, including those based on MMLV (78), herpesviruses (79), adenoviruses (80), and plas-mids (78), but none of those proved to be efficient or consistent enough for clinical application to the development of autologous cell vaccines. However, primary leukemia/lymphoma cells are good candidates for lentiviral vector transduction, as they show poor proliferation in vitro (81). Thus, we and others have been able to show that HIV-derived lentiviral vectors pseudotyped with the VSV-G envelope were capable of efficiently transducing human leukemia cells and hematopoietic progenitor cells (78,82-84).
In our first endeavor to deliver CD80 and GM-CSF genes into human ALL cells, we used a second-generation lentiviral vector packaging system (see Fig. 7). Functional experiments were performed to evaluate the response of the patients' autologous T-cells against their lentivirus transduced leukemia cells (82). The stimulatory activity of non-transduced and transduced ALL cells was compared in primary and secondary auto-logous T-cell stimulation assays. These showed that ALL/CD80 cells, but not ALL/Mock or ALL/GFP, stimulated significant T-cell proliferation, which could be abrogated in the presence of an anti-CD80 blocking antibody or the fusion protein CTLA4-Ig, which blocks the engagement of CD28 by CD80 (82). These results demonstrated that the transduction of CD80 into ALL cells was capable of converting the leukemia cells into competent APC. Subsequently, we evaluated a third generation SIN lentiviral vector coexpressing GM-CSF and CD80 for transduction of primary acute myeloid leukemia (AML) cells (83). Allogeneic and autologous T-cell stimulation experiments demonstrated that transduction with RRL-CD80, RRL-GM-CSF and RRL-GM-CSF/CD80 significantly increased allogeneic T-cell proliferation, in contrast to a smaller increase in the autologous T-cell proliferation (83). We have more recently evaluated the insertion of the central polypurine tract and the central termination sequence (cPPT/CTS) into a SIN lentiviral vector encoding for GM-CSF and CD80 (77). Expression levels of GM-CSF and CD80 were consistently and significantly
Was this article helpful?
Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.