Substitution of blood coagulation factor X-binding to Ad5 by position-specific PEGylation: Preventing vector clearance and preserving infectivity.

The biodistribution of adenovirus type 5 (Ad5) vector particles is heavily influenced by interaction of the particles with plasma proteins, including coagulation factor X (FX), which binds specifically to the major Ad5 capsid protein hexon. FX mediates hepatocyte transduction by intravenously-injected Ad5 vectors and shields vector particles from neutralization by natural antibodies and complement. In mice, mutant Ad5 vectors that are ablated for FX-binding become detargeted from hepatocytes, which is desirable for certain applications, but unfortunately such FX-nonbinding vectors also become sensitive to neutralization by mouse plasma proteins. To improve the properties of Ad5 vectors for systemic delivery, we developed a strategy to replace the natural FX shield by a site-specific chemical polyethylene glycol shield. Coupling of polyethylene glycol to a specific site in hexon hypervariable region 1 yielded vector particles that were protected from neutralization by natural antibodies and complement although they were unable to bind FX. These vector particles evaded macrophages in vitro and showed significantly improved pharmacokinetics and hepatocyte transduction in vivo. Thus, site-specific shielding of Ad5 vectors with polyethylene glycol rendered vectors FX-independent and greatly improved their properties for systemic gene therapy.

Unexpected pulmonary uptake of adenovirus vectors in animals with chronic liver disease.

When adenovirus vectors are injected intravenously, most of the virions are quickly taken up by the reticuloendothelial system, primarily by the liver macrophages known as Kupffer cells. However, little is known about the behavior of adenovirus vectors when there is pre-existing liver disease. To study this, we examined the biodistribution of intravenously injected vector in a rat model of cirrhosis induced by bile duct ligation. Using quantitative PCR and fluorescently tagged adenovirus vectors, we observed a significant reduction in vector uptake by the cirrhotic liver and increased accumulation in the lungs. Immunocytochemistry and electron microscopy demonstrated that this was due to changes in the reticuloendothelial system, with the vector being taken up by large numbers of pulmonary intravascular macrophages in the lungs of cirrhotic rats. Interestingly, expression of vector-encoded luciferase was significantly reduced in the livers of cirrhotic rats, but was not increased in the lungs. These data demonstrate that the biodistribution of adenovirus vectors in rats is altered by cirrhosis, which suggests the possibility that these vectors might behave unexpectedly in patients with pre-existing liver conditions, particularly since pulmonary reticuloendothelial changes are known to occur in human disease.

Humoral immune responses to adenovirus vectors in the brain.

We have investigated the humoral immune response to E1-deleted adenovirus vectors encoding the lacZ gene introduced into the brains of mice. Injection of these non-replicating vectors into the brain induced systemic antibody production to adenovirus vectors in dose dependent manner. Apparent antibody production to beta-galactosidase, the product of the lacZ gene, was detected later than anti-adenovirus antibody. Neutralizing antibody was not detected. This study demonstrates that E1-deleted adenovirus vectors injected into the brain trigger humoral immune responses to the adenovirus and its gene products, but they are not sufficient to block the infection of cells by adenovirus upon repeat injection.

Large-plaque mutants of Sindbis virus show reduced binding to heparan sulfate, heightened viremia, and slower clearance from the circulation.

Laboratory strains of Sindbis virus must bind to the negatively charged glycosaminoglycan heparan sulfate in order to efficiently infect cultured cells. During infection of mice, however, we have frequently observed the development of large-plaque viral mutants with a reduced ability to bind to heparan sulfate. Sequencing of these mutants revealed changes of positively charged amino acids in putative heparin-binding domains of the E2 glycoprotein. Recombinant viruses were constructed with these changes as single amino acid substitutions in a strain Toto 1101 background. All exhibited decreased binding to heparan sulfate and had larger plaques than Toto 1101. When injected subcutaneously into neonatal mice, large-plaque viruses produced higher-titer viremia and often caused higher mortality. Because circulating heparin-binding proteins are known to be rapidly sequestered by tissue heparan sulfate, we measured the kinetics of viral clearance following intravenous injection. Much of the parental small-plaque Toto 1101 strain of Sindbis virus was cleared from the circulation by the liver within minutes, in contrast to recombinant large-plaque viruses, which had longer circulating half-lives. These findings indicate that a decreased ability to bind to heparan sulfate allows more efficient viral production in vivo, which may in turn lead to increased mortality. Because Sindbis virus is only one of a growing number of viruses from many families which have been shown to bind to heparan sulfate, these results may be generally applicable to the pathogenesis of such viruses.

Characterization of a Chinese hamster ovary cell line developed by retroviral insertional mutagenesis that is resistant to Sindbis virus infection.

The alphavirus Sindbis virus (SV) has a wide host range and infects many types of cultured cells in vitro. The outcome of infection is dependent on the strain of virus used for infection and the properties of the cells infected. To identify cellular determinants of susceptibility to SV infection we mutagenized Chinese hamster ovary (CHO) cells by retroviral insertion with a vector containing the neomycin resistance gene that allowed selection for integration into transcriptionally active genes. Cells were then selected for survival after infection with SV. The most resistant cell line (CHO-18.4m) exhibited delayed virus replication and virus-induced cell death, had a single retroviral insertion, and was defective in SV binding to the cell surface. Further analysis revealed that CHO-18.4m cells were deficient in the expression of the sulfated glycosaminoglycans heparan sulfate and chondroitin sulfate. This further confirms the importance of heparan sulfate as an attachment molecule for SV in vitro and demonstrates the usefulness of this technique for identifying cellular genes that are important for virus replication.

Local gene therapy with CTLA4-immunoglobulin fusion protein in experimental allergic encephalomyelitis.

It has been reported previously that the induction phase of experimental allergic encephalomyelitis (EAE) is highly sensitive to systemic blockade of stimulation via MHC class II molecules and co-stimulation via the CD28:CD80/CD86 pathways. In contrast, the effector phases of EAE were relatively unaffected by similar treatments using MHC class II antigen (Ag)-specific mAb and cytotoxic T lymphocyte antigen (CTLA)4-Ig fusion proteins in some studies. This has been attributed to different sensitivities of effector cell function or the poor penetrance of inhibitory proteins into the central nervous system (CNS). To examine this question further, MHC class II Ag-specific mAb and CTLA4-Ig were delivered directly into the CNS following EAE induction, and both were found to inhibit disease. While it was found that systemic administration of mouse CTLA4-Ig could also inhibit the progression of effector immune responses when administered shortly before or during clinical disease, these were significantly more active when delivered directly into the CNS, which probably involved an action on both CD28 ligands, CD80 and CD86. Although mouse CTLA4-human Ig was therapeutically less efficient than mouse CTLA4-mouse Ig protein, probably due to the enhanced immunogenicity and lower functional activity, gene delivery of CTLA4-human Ig into the CNS using a non-replicating adenoviral vector was more effective than a single injection of CTLA4-human Ig protein. Gene delivery significantly ameliorated the development of EAE, without necessarily inhibiting unrelated peripheral immune responsiveness. Local gene delivery of CTLA4-Ig may thus be an important target for immunotherapy of human autoimmune conditions such as multiple sclerosis.

Co-injection of adenovirus expressing CTLA4-Ig prolongs adenovirally mediated lacZ reporter gene expression in the mouse retina.

There is growing interest in gene delivery to the eye in order to develop gene therapy for the many ocular disorders which may be amenable to this approach. To date, recombinant adenoviruses (AV) have been the main vector used for gene delivery to anterior and posterior segments in animal models. As with delivery to other organs, immune responses to vector and transgene limit the duration of expression in the eye. Using an E1-deleted adenoviral vector carrying a lacZ reporter gene, we have previously demonstrated that a T cell-mediated immune response reduces the level of intra-ocular transgene expression over time and limits it to around 3 weeks in mice. This report describes a strategy for prolonging gene expression by blocking the B7-CD28 interactions between antigen presenting cells (APC) and T cells in order to prevent the costimulatory signals required for T cell survival and proliferation. This was achieved by the co-injection of AV encoding a secreted immunomodulatory molecule (CTLA4-Ig) which consists of the extra-cellular domain of mouse CTLA4 fused to the Fc region of human IgG. Subretinal co-injection of AV encoding beta galactosidase with AV encoding CTLA4-Ig results in prolonged expression in retinal cells compared with subretinal injection of only adenovirus encoding beta galactosidase.

Ex vivo adenovirus-mediated gene transfer and immunomodulatory protein production in human cornea.

One attractive strategy to prevent or control allograft rejection is to genetically modify the donor tissue before transplantation. In this study, we have examined the feasibility of gene transfer to human corneal endothelium, using a number of recombinant adenovirus constructs. Ex vivo infection of human corneas with adenoviral vectors containing lacZ, under transcriptional control of either cytomegalovirus (CMV) or Rous sarcoma virus (RSV) promoters, provided high-level gene expression, which was largely restricted to endothelium. Expression of the reporter gene persisted at relatively high levels for up to 7 days, followed by a decline to indetectable levels by 28 days. RT-PCR analysis of lacZ transcription showed a similar picture with a short period (3-7 days) of RNA transcription after infection. In contrast, adenoviral DNA persisted for at least 56 days. Subsequently, we examined the expression of a potential therapeutic gene, CTLA-4 Ig fusion protein. Following infection of human corneas with adenoviral vectors encoding CTLA-4 Ig protein, high levels of the fusion protein were detected in corneal culture supernatants for up to 28 days. This protein was functionally active, as determined by binding to B7.1 (CD80)-expressing transfectants. This study suggests that genetic alteration of donor cornea before transplantation is a feasible approach for preventing or controlling allograft rejection. Similar gene-based strategies might also be feasible to prevent rejection of other transplanted tissues or organs.

Immune responses to adenoviral vectors during gene transfer in the brain.

We have investigated the immune response to E1-deleted adenovirus vectors encoding the lacZ gene introduced into the brains of adult mice. Injection of these nonreplicating vectors caused a marked inflammatory response in the brain as assessed by immunocytochemistry and flow cytometry of leukocytes. Infiltrating leukocytes were detectable within 2 days of injection and reached a maximum by 9 days. Thereafter, the number of infiltrating cells decreased, but a small number persisted in the brain until day 60. Between 2 and 4 days after injection, the percentage of CD8+ cells detectable increased whereas the percentage of CD4+ cells present in the infiltrating population did not significantly increase until day 6, peaking on day 15. Activated CD25+ T cells were detectable between days 6 and 15. beta-Galactosidase (beta-Gal), the product of the lacZ gene encoded by the vector, was also detected, both at the injection site in the striatum and also in the substantia nigra. Expression peaked between 4 and 6 days but a small number of beta-Gal+ cells was still seen at 60 days after injection. This study demonstrates that a quantitative analysis of the immune responses caused by a nonreplicating adenovirus vector is possible in the brain. E1-deleted adenoviral vectors trigger a strong inflammatory response in the brain, but this immune response is not sufficient to eliminate completely expression of genes encoded by the adenoviral construct.

Persistent transgene expression in the hypothalamus following stereotaxic delivery of a recombinant adenovirus: suppression of the immune response with cyclosporin.

Replication deficient, recombinant adenoviruses (Ads) have been used successfully to transfect several forebrain and brainstem nuclei, but have yet to be demonstrated as useful vectors for transgene delivery in the structurally diverse and highly vascularised nuclei of the hypothalamus. In the present study we have assessed the ability of an Ad expressing the lac-Z gene to transfect cells of the paraventricular nucleus (PVN) of the hypothalamus in vivo. We show that: (1) we can achieve stable expression of the lacZ gene in cells of the magnocellular PVN for at least 2 months; (2) there were no obvious differences in the level of AVP mRNA in the PVNs injected with Ad compared with those injected with vehicle suggesting that Ad treatment is not disrupting normal cellular function in the injection region; (3) the introduction of Ads results in a limited immune response; (4) systemic treatment with cyclosporin dramatically reduces its magnitude. We conclude that Ad vectors represent useful tools for neuroendocrinological and gene therapeutic studies of the hypothalamus.

Immune responses to adenovirus vectors in the nervous system.

Non-replicating adenovirus vectors are being developed as vehicles for gene transfer into cells of the nervous system. An important requirement for successful gene transfer is the absence of deleterious cytotoxic or inflammatory side effects of the delivery system. Despite offering relatively stable reporter gene expression, currently available adenovirus vectors also elicit immune responses in the brain, both at the site of vector delivery and at synaptically linked distant sites. However, although an anti-viral T-lymphocyte response eliminates the vector and damages local tissue in many peripheral organs, the immune response to adenovirus in the brain is less effective and enables the vector to persist. Nevertheless, in this persistent state the adenovirus vector remains a potential target for a destructive immune response that can also cause local demyelination. The development of strategies to minimize this damaging immune response, through either vector modification or immunomodulation, will be crucial for the future success of genetic therapies in the brain.

Role of T cells in inflammation caused by adenovirus vectors in the brain.

In many organs, E1-deleted human adenovirus vectors trigger antiviral T cell responses which limit the duration of vector-encoded gene expression. When injected into the brain, however, long-term expression is possible in spite of the ensuing inflammatory response. To examine the role of T cells in the immune response in the brain, monoclonal antibodies were used to systemically deplete CD4+ and/or CD8+ T cell subsets from mice at the time of vector injection. The early phase of the inflammatory response, characterized by high MHC I expression and recruitment of mononuclear cells, was unaffected by T cell depletion. Six days after injection, however, inflammation was markedly reduced by CD8-depletion and eliminated by CD4-depletion. Vector expression of the marker protein beta-galactosidase did not differ between depleted and undepleted mice. In contrast, when mice had been previously exposed to adenovirus vector in the periphery, beta-galactosidase expression in the brain was transient, showing that T cells can effectively target vector-transduced cells in this organ. We conclude that adenovirus vectors are able to achieve long-term expression in the brain because such a route of injection triggers an ineffective T cell response.

Immunological instability of persistent adenovirus vectors in the brain: peripheral exposure to vector leads to renewed inflammation, reduced gene expression, and demyelination.

Nonreplicating adenovirus vectors are being developed as vehicles for the delivery of therapeutic genes in vivo. Whereas in many organs an antiviral T cell response eliminates the vector and damages local tissue, when adenovirus vectors are injected into the brain the subsequent immune attack can be ineffective, allowing the vector to persist. In the present study, E1-deleted human adenovirus vectors were injected into the caudate nucleus of rats. Two months later, expression of protein from the vector was still evident and little inflammation was seen. A subcutaneous injection of adenovirus vector at this time, however, led within 2 weeks to severe mononuclear inflammation and microglial activation in the caudate. This caused local demyelination and a decrease in detectable protein expression from the vector. Interestingly, intense microglial activation and numerous lymphocytes and monocytes were also seen in brain areas containing neurons capable of retrogradely transporting the adenovirus vector from the caudate. Control experiments established that this inflammation in distant brain areas was not a nonspecific consequence of degeneration. These experiments demonstrate that although adenovirus vectors can persist in the brain without causing chronic inflammation, they remain the potential target of a damaging cell-mediated immune response brought about by a subsequent peripheral exposure to vector. The finding of lymphocytes in brain areas that project to the caudate further shows that viral antigens that are retrogradely transported by neurons can also be the target of a T cell attack.

Adenovirus gene transfer causes inflammation in the brain.

We report that injecting an E1-deleted, non-replicating, human adenovirus type 5 vector into the brain leads to an inflammatory response. Much of this inflammation is induced directly by the virion particles themselves rather than through the expression of new proteins from the vector. The severity of inflammation was found to depend on the strain of inbred rat used: PVG rats have less inflammation than AO rats in response to a vector injection. Twelve hours after injection of adenovirus vectors into the striatum of AO rats, leukocytes were seen marginating to the walls of nearby blood vessels. By two days there was a large increase in major histocompatibility complex class I expression and a heavy infiltration of leukocytes, mainly macrophages and T cells. Retrograde transport of adenovirus to neurons of the substantia nigra was associated with a delayed and less intense inflammation at this distant site. Although AO and PVG rats showed comparable responses in the striatum up to six days, at later times PVG rats had less intense inflammation. In spite of the inflammatory response, vector-driven expression of the marker protein beta-galactosidase and an adenovirus early protein was seen for at least two months following the injection, although expression declined with time. The observation that adenovirus gene transfer leads to an inflammatory response in the brain must be taken into account when planning and interpreting experiments with these vectors. Furthermore, we conclude that using an appropriate strain of rat can diminish some aspects of the inflammation.

Immunological consequences of HSV-1-mediated gene transfer into the CNS.

Defective HSV-1 viral vectors were prepared using amplicon methods. The amplicon contained the cytomegalovirus immediate-early promoter and the lacZ gene as a reporter in addition to the HSV elements required for replication and packaging in vitro. Viral vectors were stereotaxically injected into the rat dentate gyrus and the resulting expression and immune response were investigated. Beta-galactosidase activity was detected in several thousand neurons from as early as 24 hours to as late as 10 days after injection. A significant immune response to the vector inoculation developed, which was characterised by diffuse MHC class I up-regulation from 48 hours and the infiltration of MHC class II+ cells and activated T lymphocytes and macrophages from day 4. These features persisted for at least 31 days. Of particular interest was a small group of neurons in the posterior hypothalamus which were found bilaterally to express beta-galactosidase. The immune response at this distant uninjected site was delayed in onset but its features were similar to that found at the primary site of inoculation.