Gene therapy for hemophilia A: production of therapeutic levels of human factor VIII in vivo in mice

Coexpression of factor VIII heavy and light chain adeno-associated viral vectors produces biologically active protein

We are interested in using recombinant adeno-associated viral vectors in the treatment of hemophilia A. Because of the size constraints of recombinant adeno-associated viral vectors, we delivered the heavy and light chains of the human factor 8 (hFVIII) cDNA independently by using two separate vectors. Recombinant AAV vectors were constructed that utilized the human elongation factor 1alpha promoter, a human growth factor polyadenylation signal, and the cDNA sequences encoding either the heavy or light chain of hFVIII. Portal vein injections of each vector alone, a combination of both vectors, or a hFIX control vector were performed in C57BL/6 mice. An ELISA specific for the light chain of hFVIII demonstrated very high levels (2-10 microgram/ml) of protein expression in animals injected with the light chain vector alone or with both vectors. We utilized a chromogenic assay in combination with an antibody specific to hFVIII to determine the amount of biologically active hFVIII in mouse plasma. In animals injected with both the heavy and light chain vectors, greater than physiological levels (200-400 ng/ml) of biologically active hFVIII were produced. This suggests that coexpression of the heavy and light chains of hFVIII may be a feasible approach for treatment of hemophilia A.

Long-term expression of human coagulation factor VIII and correction of hemophilia A after in vivo retroviral gene transfer in factor VIII-deficient mice

Hemophilia A is caused by a deficiency in coagulation factor VIII (FVIII) and predisposes to spontaneous bleeding that can be life-threatening or lead to chronic disabilities. It is well suited for gene therapy because a moderate increase in plasma FVIII concentration has therapeutic effects. Improved retroviral vectors expressing high levels of human FVIII were pseudotyped with the vesicular stomatitis virus G glycoprotein, were concentrated to high-titers (10(9)-10(10) colony-forming units/ml), and were injected intravenously into newborn, FVIII-deficient mice. High-levels (>/=200 milliunits/ml) of functional human FVIII production could be detected in 6 of the 13 animals, 4 of which expressed physiologic or higher levels (500-12,500 milliunits/ml). Five of the six expressers produced FVIII and survived an otherwise lethal tail-clipping, demonstrating phenotypic correction of the bleeding disorder. FVIII expression was sustained for >14 months. Gene transfer occurred into liver, spleen, and lungs with predominant FVIII mRNA expression in the liver. Six of the seven animals with transient or no detectable human FVIII developed FVIII inhibitors (7-350 Bethesda units/ml). These findings indicate that a genetic disease can be corrected by in vivo gene therapy using retroviral vectors.

Biologic aspects of expression of stably integrated transgenes in cells of the skin in vitro and in vivo

The observation that transgenes can be stably integrated into the genome of fibroblasts using recombinant retroviruses enhanced interest in using these cells as a vector for gene therapy. This enthusiasm has lessened during the past 8 years, not because skin has lost the features that make it attractive for gene therapy, but rather because stable transgene expression in vivo has not been achieved. All investigators who have used genetically modified fibroblasts to study in vivo aspects of gene therapy have shown a decrease in transgene expression with time. This contrasts with transgene expression in similarly transduced fibroblasts in vitro, where expression is not lost or is lost very slowly. We have initiated an approach to bring further understanding to the biology of transgene expression by fibroblasts carrying stably integrated transgenes in an in vivo setting. Experiments described permit the following conclusions. Expression by and survival of genetically modified fibroblasts a) requires a persistent matrix scaffold in in vivo settings; b) is prolonged if the matrix is allowed to mature in vitro; c) is enhanced if the matrix is partially sequestered behind a coating of normal fibroblasts; and d) can be substantively prolonged in vivo by immortalizing the cells. These observations support the notion that prolonged expression of transgenes by fibroblasts can be achieved in vivo and that gene therapy utilizing fibroblasts and other cells of the skin has clinical utility.

Efficient adenoviral vector transduction and expression of functional human factor VIII in cultured primary human hepatocytes

Hemophilia A is a severe bleeding disorder caused by a deficiency in blood coagulation factor VIII (FVIII). Adenoviral vectors containing a potent human FVIII expression cassette encoding a truncated FVIII cDNA were developed that mediated sustained FVIII expression in normal and haemophiliac mice and complete phenotypic correction of the bleeding disorder in haemophiliac mice and dogs (Connelly and Kaleko, Haemophilia, 1998; 4: 380-8). Here, we evaluated two E1/E2a/E3-deleted adenoviral vectors encoding human FVIII, one containing the full-length cDNA and the second containing a truncated cDNA lacking the B-domain. Viral vectors encoding the human full-length FVIII cDNA have not been described previously. Hepatocyte transduction was efficient and dose dependent, ranging from 50% to 100%. High levels of functional FVIII were secreted from transduced cells at amounts up to 6000 mU-1 10(6)cells-1 60 h. B-domain deleted FVIII was expressed at levels at least 8-fold higher than the full-length FVIII protein, whereas FVIII RNA levels were similar with both vectors. These data provide the first demonstration of FVIII adenoviral vector function in primary human cells and verify the potential clinical utility of adenoviral vectors for the treatment of haemophilia A.

Elucidation of muscle-binding peptides by phage display screening

Muscle makes up the largest tissue volume of the body, yet its size makes muscle-specific therapy difficult. This becomes particularly relevant when approaches to gene therapy for inherited myopathies are evaluated. Thus, a mechanism to target constructs or pharmaceuticals to muscle following intravenous injection would be advantageous. By screening a random phage display library we have identified a heptapeptide sequence, ASSLNIA, with enhanced in vivo skeletal and cardiac muscle binding. Phage carrying this peptide showed a 9- to 20-fold (depending on control tissue) increase in muscle selectivity compared with phage with no insert. When the injected individual phage clone was localized by immunohistochemistry, it was found within focal areas of the membrane of myofibers. Thus, the peptide identified represents a ligand that is capable of accessing skeletal and cardiac muscle from the lumen of blood vessels and could therefore readily be exploited for targeted delivery to muscle cells.

Control of erythropoietin secretion by doxycycline or mifepristone in mice bearing polymer-encapsulated engineered cells

Cell encapsulation offers a safe and manufacturable method for the systemic delivery of therapeutic proteins from genetically engineered cells. However, control of dose delivery remains a major issue with regard to clinical application. We generated populations of immortalized murine NIH 3T3 fibroblasts that secrete mouse erythropoietin (Epo) in response to stimulation by doxycycline or mifepristone. Engineered cells were introduced into AN69 hollow fibers, which were implanted in the peritoneal cavity or recipient mice. Animals receiving doxycycline or mifepristone showed stable polyhemia and increased serum Epo concentrations over a 6-month observation period, whereas animals not receiving the inducer drug had normal hematocrits. Epo secretion could be switched on and off, depending on the presence of doxycycline in the drinking water. In contrast, polyhemia was hardly reversible after subcutaneous injections of mifepristone. These data show that a permanent and regulated systemic delivery of a therapeutic protein can be obtained by the in vivo implantation of engineered allogeneic cells immunoprotected in membrane polymers.

Bone marrow stromal cell-mediated gene therapy for hemophilia A: in vitro expression of human factor VIII with high biological activity requires the inclusion of the proteolytic site at amino acid 1648

To evaluate the potential of the ex vivo bone marrow stromal cell (BMSC) system as a gene therapy for hemophilia A, we studied the in vitro expression of human factor VIII (hFVIII) in canine BMSCs following transfection with plasmid vectors and transduction with retroviral vectors. Vectors were composed of B domain-deleted forms of hFVIII that either retain or delete the proteolytic site at amino acid 1648. On transfection of BMSCs, vectors supported expression and secretion of similar levels of up to 386 mU/10(6) cells/24 hr, even though only 3-9% of the cells expressed hFVIII while 42-48% of transfected cells harbored plasmid vector. Much higher percentages (approximately 70%) of cells expressing hFVIII were achieved when BMSCs were transduced by retroviral vectors, resulting in expression and secretion as high as 1000-4000 mU/10(6) cells/24 hr. Western analysis demonstrated that the B domain-deleted forms possessing the proteolytic site were secreted predominantly as heavy and light chain heterodimers that resemble native forms found in plasma. In contrast, the hFVIII lacking the proteolytic site was expressed mostly as unprocessed, single heavy-light chains. Both hFVIII forms were correctly cleaved and activated by thrombin. The proteolyzed hFVIII form possessed > or = 93% normal biological activity while the unproteolyzed form possessed consistently less than 55% normal biological activity and was therefore considered less suitable for therapeutic application. These results demonstrate that the BMSC system has potential utility in gene therapy for hemophilia A and stress the importance of selecting the appropriate hFVIII structure for prospective clinical use.

Genetic vaccination-induced immune responses to the human immunodeficiency virus protein Rev: emergence of the interleukin 2-producing helper T lymphocyte

Rev M10 is a trans-dominant negative inhibitor of HIV replication. Hence, stable transduction of CD4+ T cells with Rev M10 represents a novel gene therapy aimed at inhibiting HIV replication within these cells, thereby slowing the progression of AIDS. However, the immune system may recognize Rev M10 as foreign and target transduced cells for elimination. In the current study, mice were genetically immunized with a plasmid encoding Rev M10, to (1) identify immune parameters that may be induced by Rev M10 gene transfer, (2) determine the impact of repeated introduction of the Rev M10-encoding plasmid on the immune response to the transgene product, and (3) determine if cotransfection with a plasmid encoding TGFbeta1 would suppress the response. Kinetic studies revealed that Rev-specific IL-2-producing helper T lymphocytes (HTLs) appeared following the second genetic immunization, peaked after the third, and persisted at peak levels for at least 6 weeks. Rev-specific HTLs were CD4+, and the development of these cells was ablated by cotransfection with TGFbeta1. Other cytokines were not readily detectable when immune splenocytes were restimulated with Rev in vitro, and Rev-specific IgG antibodies were not present in the sera of these mice. To our knowledge, this represents the first report that genetic immunization with Rev M10 induces an immune response that is dominated by IL-2-producing HTLs. Further, this study demonstrates the potential utility of introducing immunosuppressive genes as a means to control the immune response to foreign transgene products.

The PIG-A Mutation and Absence of Glycosylphosphatidylinositol-Linked Proteins Do Not Confer Resistance to Apoptosis in Paroxysmal Nocturnal Hemoglobinuria

Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal stem cell disorder characterized by complement-mediated hemolysis and deficient hematopoiesis. The development of PNH involves an acquired mutation in the X-linked PIG-A gene, which leads to incomplete bioassembly of glycosylphosphatidylinositol (GPI) anchors and absent or reduced surface expression of GPI-linked proteins. The origin and mechanisms by which the PNH clone becomes dominant are not well understood, but recently resistance to apoptosis has been postulated. To test the hypothesis that the PIG-A mutation and absence of GPI-linked surface proteins directly confer resistance to apoptosis, we isolated peripheral granulocytes from 26 patients with PNH and 20 normal controls and measured apoptosis induced by serum starvation. Granulocytes from patients with PNH were relatively resistant to apoptosis (38.8% ± 14.1%) as compared with granulocytes from controls (55.0% ± 12.0%, P < .001). However, this resistance to apoptosis was not related to the dominance of the PNH clone because patients with a low percentage of GPI-deficient granulocytes had a similar rate of apoptosis as those with a high percentage of GPI-deficient granulocytes. Similarly, the resistance to granulocyte apoptosis was not influenced by the degree of neutropenia or a prior history of aplastic anemia. To investigate formally the importance of GPI-linked surface proteins in apoptosis, we introduced the PIG-A cDNA sequence into the JY5 GPI-negative B-lymphoblastoid cell line using two different methods: (1) stable transfection of a plasmid containing PIG-A, and (2) stable transduction of a retroviral vector containing PIG-A. We then measured rates of apoptosis induced either by Fas antibody, serum starvation, or γ-irradiation. With each stimulus, apoptosis of JY5 with stable surface expression of GPI-linked proteins was not statistically different from the parent JY5 cell line or the JY25 (GPI-positive) cell line. Our data confirm that granulocytes from patients with PNH have a relative resistance to apoptosis as compared with normal granulocytes. However, this resistance does not vary with the level of expression of GPI-linked proteins, and stable introduction of PIG-A cDNA with correction of GPI-linked surface expression does not change the rate of apoptosis. Taken together, our data do not support the hypothesis that the PIG-A mutation and absence of GPI-linked surface proteins directly confer resistance to apoptosis in PNH. We conclude that the resistance to apoptosis in PNH is not related to the PIG-A mutation, indicating that other factors must be important in the origin of this phenomenon and the clonal dominance observed in PNH.

The PIG-A Mutation and Absence of Glycosylphosphatidylinositol-Linked Proteins Do Not Confer Resistance to Apoptosis in Paroxysmal Nocturnal Hemoglobinuria

Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal stem cell disorder characterized by complement-mediated hemolysis and deficient hematopoiesis. The development of PNH involves an acquired mutation in the X-linked PIG-A gene, which leads to incomplete bioassembly of glycosylphosphatidylinositol (GPI) anchors and absent or reduced surface expression of GPI-linked proteins. The origin and mechanisms by which the PNH clone becomes dominant are not well understood, but recently resistance to apoptosis has been postulated. To test the hypothesis that the PIG-A mutation and absence of GPI-linked surface proteins directly confer resistance to apoptosis, we isolated peripheral granulocytes from 26 patients with PNH and 20 normal controls and measured apoptosis induced by serum starvation. Granulocytes from patients with PNH were relatively resistant to apoptosis (38.8% ± 14.1%) as compared with granulocytes from controls (55.0% ± 12.0%, P < .001). However, this resistance to apoptosis was not related to the dominance of the PNH clone because patients with a low percentage of GPI-deficient granulocytes had a similar rate of apoptosis as those with a high percentage of GPI-deficient granulocytes. Similarly, the resistance to granulocyte apoptosis was not influenced by the degree of neutropenia or a prior history of aplastic anemia. To investigate formally the importance of GPI-linked surface proteins in apoptosis, we introduced the PIG-A cDNA sequence into the JY5 GPI-negative B-lymphoblastoid cell line using two different methods: (1) stable transfection of a plasmid containing PIG-A, and (2) stable transduction of a retroviral vector containing PIG-A. We then measured rates of apoptosis induced either by Fas antibody, serum starvation, or γ-irradiation. With each stimulus, apoptosis of JY5 with stable surface expression of GPI-linked proteins was not statistically different from the parent JY5 cell line or the JY25 (GPI-positive) cell line. Our data confirm that granulocytes from patients with PNH have a relative resistance to apoptosis as compared with normal granulocytes. However, this resistance does not vary with the level of expression of GPI-linked proteins, and stable introduction of PIG-A cDNA with correction of GPI-linked surface expression does not change the rate of apoptosis. Taken together, our data do not support the hypothesis that the PIG-A mutation and absence of GPI-linked surface proteins directly confer resistance to apoptosis in PNH. We conclude that the resistance to apoptosis in PNH is not related to the PIG-A mutation, indicating that other factors must be important in the origin of this phenomenon and the clonal dominance observed in PNH.

Haemophilia A gene therapy

Gene therapy for haemophilia A would represent a significant improvement over the current treatment by providing prophylactic expression of FVIII and correction of the coagulation defect. Furthermore, a gene therapy protocol allowing simple, infrequent vector administration may extend haemophilia treatment to remote locations world-wide that currently lack access to FVIII replacement therapy. Within the last half decade, significant progress has been made on the development of gene therapy for the treatment of haemophilia A. Recent achievements include high level clotting factor expression in mice, dogs, and monkeys as well as phenotypic correction in haemophiliac mice and dogs. With the efforts that are currently directed toward the improvement of gene transfer vectors and the development of technologies to enable sustained clotting factor expression, gene therapy for haemophilia A will ultimately become a reality.

Sustained Phenotypic Correction of Murine Hemophilia A by In Vivo Gene Therapy

Hemophilia A is caused by a deficiency of blood coagulation factor VIII (FVIII) and has been widely discussed as a candidate for gene therapy. While the natural canine model of hemophilia A has been valuable for the development of FVIII pharmaceutical products, the use of hemophiliac dogs for gene therapy studies has several limitations such as expense and the long canine generation time. The recent creation of two strains of FVIII-deficient mice provides the first small animal model of hemophilia A. Treatment of hemophiliac mice of both genotypes with potent, human FVIII-encoding adenoviral vectors resulted in expression of biologically active human FVIII at levels, which declined, but remained above the human therapeutic range for over 9 months. The duration of expression and FVIII plasma levels achieved were similar in both hemophiliac mouse strains. Treated mice readily survived tail clipping with minimal blood loss, thus showing phenotypic correction of murine hemophilia A by in vivo gene therapy.

Sustained Phenotypic Correction of Murine Hemophilia A by In Vivo Gene Therapy

Hemophilia A is caused by a deficiency of blood coagulation factor VIII (FVIII) and has been widely discussed as a candidate for gene therapy. While the natural canine model of hemophilia A has been valuable for the development of FVIII pharmaceutical products, the use of hemophiliac dogs for gene therapy studies has several limitations such as expense and the long canine generation time. The recent creation of two strains of FVIII-deficient mice provides the first small animal model of hemophilia A. Treatment of hemophiliac mice of both genotypes with potent, human FVIII-encoding adenoviral vectors resulted in expression of biologically active human FVIII at levels, which declined, but remained above the human therapeutic range for over 9 months. The duration of expression and FVIII plasma levels achieved were similar in both hemophiliac mouse strains. Treated mice readily survived tail clipping with minimal blood loss, thus showing phenotypic correction of murine hemophilia A by in vivo gene therapy.

Sustained expression of high levels of human factor IX from human cells implanted within an immunoisolation device into athymic rodents

Immunoisolation of allogeneic cells within a membrane-bound device is a unique approach for gene therapy. We employed an immunoisolation device that protects allograft, but not xenograft, cells from destruction, to implant a human fibroblast line (MSU 1.2) in athymic rodents. Cells, transduced with the MFG-human factor IX retroviral vector, and expressing 0.9 microg/10(6) cells/day in vitro, were implanted in rats (four 40-microl devices, each containing 2 x 10(7) cells, two subcutaneously, two in epididymal fat) and in mice (two 20-microl devices, each containing 2 x 10(6) cells, subcutaneously). Plasma factor IX levels increased for 50 days, reaching maxima of 203 ng/ml (rat) and 597 ng/ml (mouse), and both continued at greater than 100 ng/ml for more than 140 days. A clone derived from the transduced cells, making 5 microg of factor IX/10(6) cells/day, was implanted within a device (one 20-microl device containing 2.5 x 10(6) cells), or without a device (1 x 10(7) cells implanted freely), either subcutaneously or in epididymal fat. The freely implanted cells expressed transiently, reaching more than 100 ng/ml in each site by day 4, but dropped to zero by day 20 (subcutaneous) or day 90 (epididymal fat). In devices, levels gradually increased to 100 ng/ml (subcutaneous) or 300 ng/ml (epididymal fat), remaining high for more than 100 days. These results show long-term, high-level expression of a human protein: (1) when cells are implanted within a cell transplantation device, but not when the cells are freely implanted, and (2) from a transgene driven by a viral promoter. An alloprotective device will enable the use of cloned cell lines that can be subjected to stringent quality control assessment that is impossible to achieve with autologous approaches.

Bone marrow stromal cells as targets for gene therapy of hemophilia A

Attempts to develop an ex vivo gene therapy strategy for hemophilia A, using either primary T cells or bone marrow (BM) stem/progenitor cells have been unsuccessful, due to the inability of these cell types to express coagulation factor VIII (FVIII). As an alternative, we evaluated the potential of BM-derived stromal cells which can be readily obtained and expanded in vitro. Human and murine BM stromal cells were transduced with an intron-based Moloney murine leukemia virus (MoMLV) retroviral vector expressing a B-domain-deleted human factor VIII cDNA (designated as MFG-FVIIIdeltaB). Transduction efficiencies were increased 10- to 15-fold by phosphate depletion and centrifugation, which obviated the need for selective enrichment of the transduced BM stromal cells. This resulted in high FVIII expression levels in transduced human (180 +/- 4 ng FVIII/10[6] cells per 24 hr) and mouse (900 +/- 130 ng FVIII/10[6] cells per 24 hr) BM stromal cells. Pseudotyping of the MFG-FVIIIdeltaB retroviral vectors with the gibbon ape leukemia virus envelope (GALV-env) resulted in significantly higher transduction efficiencies (100 +/- 20%) and FVIII expression levels (390 +/- 10 ng FVIII/10[6] cells per 24 hr) in transduced human BM stromal cells than with standard amphotropic vectors. This difference in transduction efficiency correlated with the higher titer of the GALV-env pseudotyped viral vectors and with the higher GALV receptor (GLVR-1) versus amphotropic receptor (GLVR-2) mRNA expression levels in human BM stromal cells. These findings demonstrate the potential of BM stromal cells for gene therapy in general and hemophilia A in particular.

Gene therapy for the hemophilias

There are many lines of evidence that suggest the eventual success of gene therapy as a treatment strategy for hemophilia. Because current treatment protocols using plasma-derived or recombinant proteins are far from ideal, the safe and efficient substitution of the defective gene by a normal copy of the gene, or at least its addition, would be of great benefit to the patient and may even be a potential cure. However, the construction of efficient gene therapy vehicles has proven quite difficult in the past and, so far, there is no system that promises to have all the desired features without any serious disadvantages. In general, either the levels of transgene expression are too low (because of the low titers achieved during the generation of the virus) or shortlived (e.g., because of the specific shut-off of the transferred promoter) as is often seen with retroviruses, or in the case of adenoviral vectors, expression is limited because of a strong immune response of the host. Clearly, much work remains to be done to optimize these promising though still imperfect vector systems. In the case of adenovirus, the development of less immunogenic vectors or in vivo modulation of the host immune system may hold promise for improvements. Reports by Yang et al. (1995) and Kay et al. (1995) are promising steps in the direction of immunomodulation. Both attenuate the immune reaction to the adenoviral vector by simultaneous application of either an interleukin or an immunoglobulin, respectively. When IL-2 was administered, the amounts of IgA were reduced and successful administration of a second dose of virus was possible. When CTLA4-Ig, an immunoglobulin that blocks the second signal during antigen presentation, was administered, a markedly prolonged expression of the transgene resulted. In vivo trials with AAV vectors have been carried out for some diseases (Flotte et al., 1993; Kaplitt et al., 1994) but not for hemophilia. Advances in high-titer AAV vector preparation will make this approach more feasible. The pace continues to quicken in the development of nonviral modes of gene delivery (Perales et al., 1994). Although these results are encouraging for the future of gene therapy as a treatment for genetic diseases, much work remains to be done to make this potential alternative a reality for treatment of hemophilia.

Promoter attenuation in gene therapy: interferon-gamma and tumor necrosis factor-alpha inhibit transgene expression

One of the major limitations to current gene therapy is the low-level and transient vector gene expression due to poorly defined mechanisms, possibly including promoter attenuation or extinction. Because the application of gene therapy vectors in vivo induces cytokine production through specific or nonspecific immune responses, we hypothesized that cytokine-mediated signals may alter vector gene expression. Our data indicate that the cytokines interferon-gamma (IFN-gamma) and tumor necrosis factor-alpha (TNF-alpha) inhibit transgene expression from certain widely used viral promoters/enhancers (cytomegalovirus, Rous sarcoma virus, simian virus 40, Moloney murine leukemia virus long terminal repeat) delivered by adenoviral, retroviral or plasmid vectors in vitro. A constitutive cellular promoter (beta-actin) is less sensitive to these cytokine effects. Inhibition is at the mRNA level and cytokines do not cause vector DNA degradation, inhibit total cellular protein synthesis, or kill infected/transfected cells. Administration of neutralizing anti-IFN-gamma monoclonal antibody results in enhanced transgene expression in vivo. Thus, standard gene therapy vectors in current use may be improved by altering cytokine-responsive regulatory elements. Determination of the mechanisms involved in cytokine-regulated vector gene expression may improve the understanding of the cellular disposition of vectors for gene transfer and gene therapy.

Intraperitoneal injection of genetically modified, human mesothelial cells for systemic gene therapy

An ideal cell type for ex vivo gene therapy should be easy to biopsy, propagate, and genetically engineer in culture, should be transplantable using simple procedures, and should express therapeutic proteins at useful levels. The mesothelial cell appears to satisfy these criteria. Several thousand proliferative mesothelial cells were present in typical specimens of nonpathologic human peritoneal fluid obtained by needle aspiration. These divided rapidly in a specialized medium to yield pure cultures of approximately 10(7) cells within 2 weeks. The replicative lifespan of mesothelial cells cultured from adults was approximately 42-52 population doublings, permitting expansion and cryopreservation of a lifetime supply of autologous cells from one fluid sample. Cells transduced with a human growth hormone (hGH) adenoviral vector secreted 100-300 microg of hGH/10(6) cells per day for at least 6 weeks in culture when maintained at quiescence. Intraperitoneal injection of transduced cells into athymic mice resulted in rapid systemic delivery of hGH, with peak plasma levels of 0.1-1 microg/ml declining over 3 weeks to <1 ng/ml. Mice receiving a second injection of engineered cells displayed the same plasma hGH levels and duration as naive mice. Cells labeled with a beta-galactosidase vector were identifiable by in situ enzymatic staining as clusters attached to peritoneal surfaces at multiple sites for at least 19 days after injection. Cells serially passaged through about three-quarters of their lifespan before transduction and injection were as effective at hGH delivery as earlier-passage cells. These results indicate the clinical potential for ex vivo gene therapy using mesothelial cells.

An ex vivo keratinocyte model for gene therapy of hemophilia B

We are investigating whether skin-targeted gene therapy may be used to treat hemophilia B by transplanting keratinocytes transduced by factor IX-expressing retroviral vectors. No pre-clinical animal model for keratinocyte-mediated gene therapy has shown long-term efficacy in vivo. It remains unclear whether this short-term expression is due to promoter shut-off or a reduced survival of grafted genetically modified cells. The purpose of this study was to determine the fate of primary human keratinocytes superficially grafted to nude mice in a silicone transplantation chamber. In addition, vectors containing keratinocyte-specific enhancers from the human papilloma virus-16 (HPV-16) and human keratin 5 and 14 genes were used upstream of the cytomegaloviral (CMV) immediate-early promoter/enhancer to control factor IX cDNA expression to avoid promoter shut-off. Factor IX was secreted by cultured keratinocytes after transduction by each of these chimeric promoter/enhancer vectors, although the levels varied according to the particular construct used. Keratinocytes transduced by the vector containing the HPV-16 enhancer were grafted into nude mice, and human factor IX was detected in plasma at 0.02-9 ng per ml for 4-5 wk for the duration of graft survival. The HPV-16 enhancer may be a useful addition to expression vectors for keratinocyte gene therapy. The transplantation chamber can be adapted to grafting retrovirally transduced keratinocytes for gene transfer studies.

Adenovirus-mediated gene transfer of viral interleukin-10 inhibits the immune response to both alloantigen and adenoviral antigen

Although adenoviral vectors are attractive for gene transfer, their effectiveness is limited by host antiviral immune responses. In this study, we determined if host antiallograft and antiviral immunity could be diminished with an adenoviral vector encoding the immunosuppressive cytokine viral interleukin-10 (vIL-10). AdSV40vIL-10, a vIL-10-expressing adenoviral vector with an SV40 promoter, induced significant prolongation of murine cardiac allograft survival to 32.2 +/- 1.7 days compared to 14.2 +/- 1.0 days for controls (p < 0.01). This effect was specific for vIL-10 encoding vector and could be inhibited by anti-vIL-10 monoclonal antibody (mAb). In vivo administration of adenovirus facilitated the generation of adenovirus-specific cytotoxic T lymphocytes (CTL), whereas treatment with AdSV40vIL-10 prevented CTL priming and generation of virus-specific immunity. AdSV40vIL-10 also induced extended expression of a beta-galactosidase reporter from a co-injected LacZ-encoding adenoviral vector. These results demonstrate that adenovirus-mediated gene transfer and expression of vIL-10 prolong allograft survival and inhibit the immune response to adenoviral antigens, thereby improving the persistence of the vector and extending transgene expression. The efficacy of adenoviral vectors can be improved by incorporating immunosuppressive genes into the vector.

Use of cloned genetically modified human fibroblasts to assess long-term survival in vivo

Because human fibroblasts are easily brought to tissue culture conditions and can be stably transduced with retroviral vectors encoding transgenes ex vivo, genetically modified fibroblasts are frequently considered in strategies to correct disease with gene therapy. This enthusiasm has been dampened by studies showing that transgene expression by genetically modified fibroblasts diminishes with time in vivo, but not in vitro, for reasons that are unclear. We elected to study this problem using cloned human fibroblasts that had been cloned by limiting dilution and stably transduced with a retroviral vector encoding lacZ ex vivo. These were seeded onto a nonbiodegradable nylon matrix that was transplanted to nude mice. Transgene expression was followed prospectively by histologic exam. Data show that human fibroblasts can withstand the pressure of cloning by limiting dilution. In addition, they can be passaged from 10 to > 20 times, and > 1 x 10(20) of genetically modified fibroblasts can be generated as progeny of one cell. Loss of transgene expression by the cloned genetically modified fibroblasts in vivo occurs in an orderly and progressive fashion, but is not complete by 4 months. Neither the loss nor the persistence of expression appear to be random. These observations are most compatible with the thesis that a major cause of the loss of transgene expression in vivo is secondary to apoptosis of the genetically modified fibroblast. Loss of expression of transgenes in senescent genetically modified fibroblasts occurs more rapidly than in their presenescent counterparts in the age-neutral, in vivo setting of the nude mouse.

Systemic delivery of human growth hormone or human factor IX in dogs by reintroduced genetically modified autologous bone marrow stromal cells

Canine bone marrow stromal cells were expanded to numbers in excess of 10(9) cells from the initial 10-20 ml of marrow aspirates and transfected to express high levels of human growth hormone (hGH) in vitro. Ex vivo-modified marrow stromal cells were used in a gene therapy model system for the systemic delivery of transgene products in dogs. Adherent bone marrow stromal cell cultures, established and expanded from iliac crest marrow aspirates from each of 8 dogs, were transfected with a hGH gene plasmid expression vector and shown to express from 0.54-3.84 micrograms/10(6) cells per 24 hr hGH in vitro. The transfected plasmid vector does not possess a eukaryotic origin of replication nor does it possess sequences required for efficient integration into the host cell genome. As such, expression was expected to be transient. Transfected cells were autologously reintroduced into each dog by either infusion into a foreleg vein or directly into iliac crest marrow. In two cases, the stromal cells were cryopreserved following transfection, and subsequently thawed and infused. In one case, the expanded stromal cells were first cryopreserved, and then thawed, recultured, transfected, and infused. Reintroduced cell numbers ranged from 2.2 x 10(7) to 2.6 x 10(9), with total hGH expression capacities ranging from 62 to 1,400 micrograms/24 hr. Plasma of each of the dogs contained detectable hGH for a mean of 3.1 days (SD +/- 0.8 day) ranging from 2 to 5 days following reinfusion of cells. Peak plasma levels ranged from 0.10 to 1.76 ng/ml. Similar hGH expression values, based upon total expression capacity of the cells infused and dog body weight, were obtained for all dogs. Vector-modified stromal cells were detectable, by polymerase chain reaction (PCR) analysis, in the peripheral circulation following reinfusion in all 4 dogs analyzed. In 3 of the dogs, modified stromal cells were detected for 8.5-15 weeks. In addition, modified stromal cells were detected in iliac crest marrow of 2 dogs for 9 and 13 weeks, respectively, following reinfusion. In another experiment, cultured bone marrow stromal cells were transfected with a human factor IX (hFIX) plasmid vector. Modified cells (5.57 x 10(8)), with a total hFIX expression capacity of 281 micrograms/24 hr, were reinfused, resulting in detectable hFIX in plasma continuously for 9 days with a peak level of 8 ng/ml on day 1. These results demonstrate that the ex vivo bone marrow stromal cell system is a potentially powerful method by which to deliver secreted transgene product to the systemic circulation of large animals.

Cutaneous gene therapy

Gene therapy efforts in a variety of tissues have foundered on fundamental technologic barriers, such as difficulties in achieving high-efficiency gene transfer to diseased tissues and in sustaining delivered transgene production. The skin offers an attractive tissue for development of approaches to therapeutic gene delivery by virtue of its accessibility for regulation by topical agents, the ease of gene transfer into cutaneous tissues, and the ready ability to monitor the impact of somatic gene transfer. With the ability of the skin to deliver therapeutic polypeptides to the systemic circulation and the recent molecular characterization of monogenic skin diseases, efforts to target genes to the skin are expected to accelerate. The current status of gene therapy efforts is reviewed, with a special focus on the skin.

Systemic hematological effects of granulocyte colony-stimulating factor produced by irradiated gene-transfected fibroblasts

Although long-term expression of therapeutic molecules is necessary for the treatment of permanent deficiencies, short-term expression of therapeutic molecules inducing local or systemic effects is preferable in clinical situations where temporary substitution is the goal. One such clinical setting is the administration of hematopoietic growth factors in cancer chemotherapy-induced myelosuppression. Several plasmid vectors containing the human granulocyte colony-stimulating factor (G-CSF) gene under transcriptional control of different regulatory elements were constructed. In vitro production of G-CSF by nonvirally transfected murine fibroblast clones initially increased after lethal irradiation and was detectable for at least 12 days. We also demonstrate that a single injection of irradiated G-CSF-secreting fibroblasts leads to accelerated hematopoietic recovery and mobilization of committed peripheral blood progenitor cells equivalent to that achieved by twice daily s.c. administration of high doses of recombinant human G-CSF. Using dicistronic vectors, high levels of G-CSF secretion were also obtained in human fibroblasts.

Construction of human factor IX expression vectors in retroviral vector frames optimized for muscle cells

Development of a highly refined human factor IX (hFIX) expression vector system is critical for establishing a durable hemophilia B gene therapy. Here we report construction of a series of retroviral vectors and identification of an optimal basic structure and components for expressing hFIX in skeletal muscle cells. These vectors, which are derived from Moloney murine leukemia virus (MoMLV) with its enhancer sequence in the 3' long terminal repeat (LTR) deleted, contained internal hFIX expression units inserted in forward configuration without or with a viral vector intron sequence (pdL or pdLIn vector frame, respectively) or in inverted configuration without a viral vector intron sequence (pdLi frame). Internal expression units contained a hFIX cDNA or hFIX minigene (hIXm1 or hIXm2) derived from the hFIX cDNA by insertion of a shortened first intron sequence of the hFIX gene. Regardless of the promoter and vector frame used, both hIXm1 and hIXm2 gave 10- to 14-fold higher hFIX expression compared to those with hFIX cDNA. Internal hFIX transcriptional control units of these vectors were composed of various promoters linked with or without the muscle creatine kinase enhancer (Me) sequence. Promoters tested included those of alpha-actin (alpha A775), beta-actin (beta A280), cytochrome oxidase (CO1250 and CO650), myogenin (Mg1031 and Mg353), and Rous sarcoma virus (RSV). beta A200, which was derived from beta A280 by eliminating potential polyadenylation sites, was also tested. As extensively examined with the myogenin promoter, presence of one or multiple copies of Me in the vectors elevated the expression activity in myotubes by 4.5- to 19-fold over those without Me, but not significantly in myoblasts. Similar enhancements in expression activity with Me were also observed with other promoters, except those of RSV and CO. The latter two showed only modest enhancements in the presence of Me. As assayed with myotubes in culture, the general order of hFIX expression activity of various promoters with four copies of Me in the three different vector frames was beta A280 approximately beta A200 > Mg353 > Mg1031 approximately RSV approximately CO650 approximately alpha A775 > CO1250. One exception was that CO650 showed significantly less activity in pdLi-type vectors than in the pdLIn vectors. Based on the systematic analyses of various structural components, a group of pdLi vectors consisting of beta A200, two to four copies of Me, and hIXm2 was identified to have the optimal basic vector structure to be used in retrovirus for hFIX expression in differentiated skeletal muscle cells. The present studies provide the critical first step for establishing a highly refined hemophilia B gene therapy based on skeletal muscle-targeted hFIX gene transfer.

Local production of the p40 subunit of interleukin 12 suppresses T-helper 1-mediated immune responses and prevents allogeneic myoblast rejection

The p40 subunit of interleukin 12 (IL-12p40) has been known to act as an IL-12 antagonist in vitro. We here describe the immunosuppressive effect of IL-12p40 in vivo. A murine myoblast cell line, C2C12, was transduced with retro-virus vectors carrying the lacZ gene as a marker and the IL-12p40 gene. IL-12p40 secreted from the transfectant inhibited the IL-12-induced interferon gamma (IFN-gamma) production by splenocytes in vitro. Survival of C2C12 transplanted into allogeneic recipients was substantially prolonged when transduced with IL-12p40. Cytokine (IL-2 and IFN-gamma) production and cytotoxic T lymphocyte induction against allogeneic C2C12 were impaired in the recipients transplanted with the IL-12p40 transfectant. Delayed-type hypersensitivity response against C2C12 was also diminished in the IL-12p40 recipients. Furthermore, serum antibodies against beta-galactosidase of the T-helper 1-dependent isotypes (IgG2 and IgG3) were decreased in the IL-12p40 recipients. These results indicate that locally produced IL-12p40 exerts a potent immunosuppressive effect on T-helper 1-mediated immune responses that lead to allograft rejection. Therefore, IL-12p40 gene transduction would be useful for preventing the rejection of allografts and genetically modified own cells that are transduced with potentially antigenic molecules in gene therapy.

High-level tissue-specific expression of functional human factor VIII in mice

Hemophilia A results from subnormal levels of blood coagulation factor VIII (FVIII) and is an attractive target for gene therapy. However, progress has been impeded by features of FVIII biology such as low mRNA accumulation and the instability of the protein. We have shown previously that a FVIII adenoviral vector, Av1ALH81, allowed high-level expression of human FVIII in mice sustained for several weeks. Here, we have generated a second FVIII adenoviral vector, Av1ALAPH81, in which an intron was introduced into the FVIII expression cassette. Administration of Av1ALAPH81 to mice resulted in significantly increased FVIII plasma levels, 1,046 +/- 163 ng/ml compared to 307 +/- 93 ng/ml of FVIII detected in mice that received Av1ALH81. Normal FVIII levels in humans are 100-200 ng/ml and therapeutic levels are as low as 10 ng/ml. Therapeutic levels are defined as the amount of FVIII necessary to convert severe hemophilia to a moderate or mild hemophiliac condition. The increased potency of the second FVIII adenoviral vector allowed the administration of significantly lower, less toxic vector doses, while retaining the potential for high FVIII expression. Furthermore, we demonstrate that adenoviral-mediated expression of human FVIII can be limited to the liver by inclusion of a liver-specific promoter, thereby achieving the first step in regulated expression of human FVIII in vivo.

Development and analysis of retroviral vectors expressing human factor VIII as a potential gene therapy for hemophilia A

To develop a potential gene therapy strategy for the treatment of hemophilia A, we constructed several retroviral vectors expressing a B-domain-deleted factor VIII (FVIII) cDNA. We confirmed previous reports that when the FVIII cDNA is inserted into a retroviral vector, the vector mRNA is decreased resulting in significantly (100- to 1,000-fold) lower vector titers. In an attempt to overcome this inhibition we pursued two independent strategies. First, site-directed mutagenesis was employed to change the structure of a putative 1.2-kb FVIII RNA inhibitory sequence (INS). Second, the FVIII gene was transcribed from a retroviral vector containing a 5' intron. Results demonstrated that the intron increased FVIII expression up to 20-fold and viral titer up to 40-fold but conservative mutagenesis of the putative FVIII INS region failed to yield a significant increase in FVIII expression or titer. Using the improved FVIII splicing vector, we transduced a variety of cell types and were able to demonstrate relatively high FVIII expression (10-60 ng of FVIII/10(6) cells/24 hr). These results underscore the usefulness of these transduced cell types for potential in vivo delivery of FVIII.

Long-term delivery of a lysosomal enzyme by genetically modified fibroblasts in dogs

We have evaluated the feasibility and efficacy of intraperitoneal implants (neo-organs) for protein delivery in large animals. Skin biopsies were taken from four healthy dogs. Primary fibroblast cultures were transduced with a retroviral vector coding for the human beta-glucuronidase. One to six lattices each containing 10(9) skin fibroblasts were implanted into the omentum of the donor animal. Laparotomies performed at regular intervals showed vascularized neo-organs without local inflammation. Human beta-glucuronidase levels equivalent to 0.8 to 3.1% of the endogenous canine activity were detected for up to 340 days on liver biopsy samples. These results indicate that neo-organs can be considered for the long-term delivery of therapeutic proteins or enzymes in humans.