Infectious retrovirus is inactivated by serum but not by cerebrospinal fluid or fluid from tumor bed in patients with malignant glioma

Imaging in gene therapy of patients with glioma

Over 10 years ago, the first successful gene therapy paradigms for experimental brain tumors models have been conducted, and they were thought to revolutionize the treatment of patients with gliomas. Application of gene therapy has been quickly forced into clinical trials, the first patients being enrolled in 1994, with overall results being disappointing. However, single patients seemed to benefit from gene therapy showing long-term treatment response, and most of these patients bearing small glioblastomas. Whereas the gene therapy itself has been performed with high sophistication, limited attention has been paid on technologies, which (i) allow an identification of viable target tissue in heterogenous glioma tissue and which (ii) enable an assessment of successful vector administration and vector-mediated gene expression in vivo. However, these measures are a prerequisite for the development of successful gene therapy in the clinical application. As biological treatment strategies such as gene and cell-based therapies hold promise to selectively correct disease pathogenesis, successful clinical implementation of these treatment strategies rely on the establishment of molecular imaging technology allowing the non-invasive assessment of endogenous and exogenous gene expression in vivo. Imaging endogenous gene expression will allow the characterization and identification of target tissue for gene therapy. Imaging exogenously introduced cells and genes will allow the determination of the 'tissue dose' of transduced cell function and vector-mediated gene expression, which in turn can be correlated to the induced therapeutic effect. Only these combined strategies of non-invasive imaging of gene expression in vivo will enable the establishment of safe and efficient vector administration and gene therapy protocols for clinical application. Here, we review some aspects of imaging in gene therapy trials for glioblastoma, and we present a 'proof-of-principle' 2nd-generation gene therapy protocol integrating molecular imaging technology for the establishment of efficient gene therapy in clinical application.

Clinical-scale selection of anti-CD3/CD28-activated T cells after transduction with a retroviral vector expressing herpes simplex virus thymidine kinase and truncated nerve growth factor receptor

Activation of T cells is necessary for efficient retroviral-mediated gene transfer. In addition, if the population of infused cells is to be limited to transduced cells, a means of positive selection is required. We describe a clinical scale procedure for activation of donor T cells with anti-CD3/CD28 beads followed by transduction with a retroviral construct expressing the herpes simplex virus thymidine kinase (HSV-tk) and human nerve growth factor receptor (NGFR). Optimization of transduction parameters was performed, testing the timing of transduction, centrifugation, and the use of serum. In large-scale experiments, 3-5 x 10(8) peripheral blood mononuclear cells (PBMC) were activated with anti-CD3/CD28 beads and expanded to day 13. Transduction was accomplished using MFG-TKiNG supernatant produced from the PG13 packaging line 48 hr after T-cell activation. The mean transduction frequency was 37.5% based on NGFR expression, and the mean expansion observed was 42.6-fold (mean final cell number 1.85 x 10(10)). A comparison of the ability of the Baxter Isolex 300i and the Miltenyi CliniMACS to perform purification of NGFR+ cells suggests that greater purity can be achieved with the CliniMACS device (67.4% vs. 97.7%), while the yield of transduced cells appears higher with the Isolex 300i (41.3% vs. 23.5%). We conclude that a strategy based on activation of human T cells with anti-CD3/CD28 beads can result in sufficient transduction, expansion, and purification based on NGFR expression for clinical trials.

Similar efficiency of DNA-liposome complexes and retrovirus-producing cells for HSV-tk suicide gene therapy of peritoneal carcinomatosis

Several experimental approaches have been tested for suicide gene delivery into tumor cells, including viral and non-viral vectors. In this study, we compared the efficiency of Herpes Simplex Virus type 1 thymidine kinase gene (HSV-tk) delivery by retrovirus-producing cells and DNA/liposome complexes for the treatment of peritoneal carcinomatosis induced in syngeneic rats by DHD/K12 colorectal adenocarcinoma cells. After in vitro determination of the best transduction conditions, rats were treated with multiple intraperitoneal injections of plasmid DNA containing one or two copies of CMV-driven HSV-tk gene (pCMV-TK and p(CMV-TK)2, respectively) associated with LipofectAMINE, each injection being followed by a Ganciclovir (GCV) course. Animals treated by DNA/liposome complexes and GCV or with retrovirus-producing cells and GCV showed a similar increase of survival as compared to the control group. After DNA/ liposome injections, expression of the tk transgene was detected in tumor nodes (epiploon) and also in liver, lung, spleen, bowels and brain. The expression was not homogeneous throughout the different organs and most likely reflected the transfection of only a limited number of cells.

Neonatal gene therapy. transfer and expression of exogenous genes in neonatal sheep following direct injection of retroviral vectors into the bone marrow space

We investigated whether gene transfer into hematopoietic cells could be achieved by direct injection of retroviral vector supernatant into the bone marrow space of newborn sheep. Six sheep (5 weeks old) were injected bilaterally with either 1 mL of G1nBgSvNa8.1 vector supernatant (titer: 1 x 10(7)) in each hip (n = 5) or with 3 mL of the same vector preparation/hip (n = 1). In addition, one 3-month-old sheep was injected unilaterally with 1 mL of the same vector preparation. Blood and marrow of these animals were analyzed for the transgene before injection and at intervals thereafter. At 1 week postinjection, an average of 11.6% of the lymphocytes and 25.5% of the granulocytes/monocytes in the marrow, and an average of 0.9% of the lymphocytes and 1.8% of the granulocytes/monocytes in the blood contained and expressed the LacZ gene. The presence/expression of the transgene has persisted for at least 13 months within the blood and bone marrow of these animals. These findings demonstrate that the direct injection of small volumes of high-titer retroviral supernatant into the bone marrow of newborn sheep results in transduction of hematopoietic cells that persists for at least 13 months postinjection.

In utero transfer and expression of exogenous genes in sheep

OBJECTIVE

We have previously reported that directly injecting low-titer retroviral vector supernatant into pre-immune sheep fetuses resulted in the transfer and long-term expression of the bacterial NeoR gene within the hematopoietic system of these animals for over 5 years. In the present studies, we investigated whether using a higher titer vector would enable more efficient transduction and expression of the transgenes within the hematopoetic cells in sheep injected in utero.

MATERIALS AND METHODS

Sixteen pre-immune sheep fetuses were injected intraperitoneally with the G1nBgSvNa8.1 helper-free retroviral vector supernatant encoding the bacterial NeoR and LacZ genes (titer: 1x10(7) cfu/mL).

RESULTS

Over the 2-year time course of these studies, the presence and expression of the NeoR and LacZ genes were demonstrated in 12 of the 14 animals evaluated by several immunological and biochemical methods. Seven of the 12 sheep examined by flow cytometric analysis contained > or =6% transduced peripheral blood lymphocytes. Vector distribution was widespread without any detectable pathology. Importantly, PCR analyses and breeding experiments demonstrated that the germ line was not altered.

CONCLUSIONS

These studies confirmed that direct injection of an engineered retrovirus is a feasible means of safely delivering foreign genes into a developing fetus and thus achieving long-term expression of the transgenes within the recipient's hematopoietic cells. Furthermore, expression of the NeoR gene from these studies was higher than that reported in our previous study in which a lower titer vector was used.

Protection of MLV vector particles from human complement

Murine cell-derived MLV vector particles usually are highly sensitive to human complement-mediated lysis. Expression of the human complement inhibitor CD59 on murine packaging cells resulted in partial protection of these cells from lysis caused by human complement proteins. Furthermore, CD59 was incorporated into MLV vector particles released by these packaging cells, leading to an improved resistance of the virions against human complement-mediated inactivation.

Complement-protected amphotropic retroviruses from murine packaging cells

The application of retroviruses generated from murine cells for in vivo gene therapy is restricted primarily because of the rapid inactivation of these viruses by the human complement system. To circumvent this disadvantageous property of murine retroviruses we have generated infectious amphotropic retroviruses that exhibit strong protection against human complement attack. The membrane of these viruses contains a fusion protein, DAFF2A, that is composed of the catalytic domain of the human complement regulatory protein (CRP) decay-accelerating factor (DAF) and the envelope protein of the amphotropic murine leukemia virus (MuLV) 4070A (EnvA). The fusion of two other CRPs, MCP and CD59, to the same amphotropic Env moiety did not lead to equivalent results. The fusion protein DAFF2A was stably expressed in mouse NIH 3T3-based helper cells and independently identified with either alpha-DAF MAb or alpha-Env PAb on the cell membrane. Western blot analysis confirmed the expected molecular weight of the fusion protein. Viral titers obtained from NIH 3T3 helper cell pools were 5 x 10(5) CFU for wild-type amphotropic EnvA virus and 1 x 10(5) CFU for DAFF2A virus, respectively. By blocking the catalytic domain of DAF by pretreatment with alpha-DAF MAb DAFF2A, recombinant virions could be converted to wild-type with respect to sensitivity against human serum. Since the method for producing virions that are protected against human serum should be applicable to any cell type it offers a novel tool for human in vivo gene therapy.

Stable gene expression with VSV-G pseudotyped-retrovirus vector in the rat liver

BACKGROUND

Pseudotyped-retrovirus-mediated gene transfer to the regenerating rat liver was investigated in vivo and the findings were compared with those for retrovirus-mediated gene transfer.

MATERIALS AND METHODS

Four weeks prior to gene transfer, the spleen was transpositioned to the left subcutaneous position to develop a port-splenic shunt. Twenty-four hours after a partial hepatectomy (68%) was performed, the liver was perfused in situ and kept in contact with either a pseudotyped-retrovirus vector encoding LacZ (7 x 10(7) cfu/ml, Group 1) or a retrovirus vector encoding LacZ (1 x 10(4) cfu/ml, Group 2) for 30 min. The animals were sacrificed at various points after gene transfer, and X-gal staining, reversed polymerase chain reaction (RT-PCR), and ONPG assay were performed to detect the transferred LacZ cDNA.

RESULTS

In X-gal staining, the transferred LacZ cDNA started to show a strong beta-galactosidase activity in 30 to 50% of the hepatocytes at 3 days after gene transfer. Positive staining continued to be recognized until 28 days with a slight decrease in its intensity thereafter. On the other hand, Group 2 animals showed weak staining, which was observed in about 10 to 15% of the hepatocytes from 3 days after gene transfer and then decreased thereafter. In RT-PCR, positive mRNA of LacZ was detected constitutively until 28 days after gene transfer in Group 1, whereas two-thirds of the samples showed a negative band in Groups 2 at 3 days after gene transfer.

CONCLUSION

In conclusion, the pseudotyped-retrovirus vector was useful in establishing a stable and strong expression of the in vivo gene transfer, while targeting the regenerating liver.

Complement and anti-alpha-galactosyl natural antibody-mediated inactivation of murine retrovirus occurs in adult serum but not in umbilical cord serum

Many retroviral vectors for hematopoietic cell and other clinical gene therapy are derived from murine packaging cell lines. The exposure of these retroviruses and packaging cell lines to adult human serum (AS) inactivates them by complement and anti-alpha-galactosyl natural antibody-mediated mechanisms. We show that virus stability and infection efficiency of CRIP/BAG, a murine packaging cell line derived amphotropic retrovirus vector is reduced > 95% following a 30-min incubation in AS. This inactivation is prevented by replacing AS with umbilical cord serum (CS), wherein full retroviral transduction efficiency is maintained after 30 min of incubation. The loss of retrovirus transduction efficiency in AS was smaller upon blockage of anti-alpha-galactosyl antibodies with galactose alpha 1-3-galactose. Serum levels of CH 100, as well as C1q complement which inactivates retroviruses by an antibody-independent mechanism were similar in AS and CS. The high stability of CRIP/BAG retrovirus vector in CS is likely due to its lower levels of maternally derived anti-alpha-galactosyl antibodies. These results have implications for in vivo gene transfer in adults and also in newborns since neonates do not produce natural antibodies during the initial months of life.

In utero gene therapy: transfer and long-term expression of the bacterial neo(r) gene in sheep after direct injection of retroviral vectors into preimmune fetuses

We investigated whether directly injecting retroviral vectors into preimmune fetuses could result in the transfer and long-term expression of exogenous genes. Twenty-nine preimmune sheep fetuses were injected with helper-free retroviral vector preparations. Twenty-two fetuses survived to term, 4 of which were sacrificed at birth. Of the remaining 18 animals, 3 were controls and 15 had received vector preparations. Twelve of these 15 animals demonstrated transduction of hematopoietic cells when blood and marrow were analyzed by neo(r)-specific PCR. Eight experimental sheep have been followed for 5 years, during which time we have consistently observed proviral DNA and G418-resistant hematopoetic progenitors. The G418-resistant colonies were positive when analyzed by neo(r)-specific PCR. neo(r) gene expression was also demonstrated using several immunological and biochemical methods. The transduction of hematopoietic stem cells was confirmed when lambs transplanted with bone marrow from in utero-transduced sheep exhibited neo(r) activity in marrow and blood. Vector distribution was widespread in primary animals without pathology. PCR analysis indicates that the germ line was not altered. These studies demonstrate that direct injection of an engineered retrovirus is a feasible means of safely delivering a foreign gene to a developing fetus and achieving long-term expression without modifying the germ line of the recipient.

FUT-175, a synthetic inhibitor of the complement pathway, protects against the inactivation of infectious retroviruses by human serum

Serum-induced inactivation of retroviruses is the most critical limitation for in vivo gene transfer therapy. To solve this problem, we searched for reagents that protect retroviruses from inactivation. The effects of the protease inhibitors FOY-007 and FOY-305 and of an inhibitor of the complement pathway FUT-175, all of which have been used clinically, were investigated. All of these agents protected against the inactivation of retroviruses by human serum, with 1 microM FUT-175 providing the most effective protection. Thus, the co-administration of FUT-175 with retroviruses may make retrovirus-mediated in vivo gene transfer feasible for the treatment of patients. FUT-175 dose-dependently inhibited the classical pathway of complement in a hemolysis protection assay of sensitized sheep erythrocytes with guinea pig serum or by cell-lysis assay of mouse fibroblasts with human serum. However, increasing the FUT-175 concentration by 10-fold (10 microM) did not produce further protection against retroviral inactivation in most human sera. There was also no correlation between the serum-induced inactivation of retroviruses and either the amount of anti-alpha-galactosyl (anti-alpha-Gal) antibody or the complement activity in human serum. These results suggest that retroviruses are not inactivated by utilizing the same pathway leading to cell lysis by the classical complement system.