Target Tissues And Therapeutic Models

Gene therapy techniques, as currently applied to the treatment of disease, tend to develop as follows. A vector is made, tried out in tissue culture and small animal models to see how it performs in terms of tissue tropism, the results assessed and a disease process identified that might be susceptible to modification. This is a perfectly rational way to proceed given the complexity of the technology and disease pathogenesis, but it has the disadvantage that the clinical element arrives late and the choice of clinical target has not always been well made. One of the exciting aspects about lentiviral vectors has been the early clear identification of tissues and disease areas where the technology seems to have great potential and there are large clinical needs. Time will tell if the development of these vectors will, in fact, lead to clinically useful therapies, but there is no doubt about the promise.

One of the first areas identified as a target was the brain. Injection of small amounts of high titer material leads to localized long-lived expression of the encoded genes (12,46,106-111c) (Fig. 4.). This has a seemingly natural fit with the supposed need for chronic therapies for chronic neu-rodegenerative diseases. The VSV-G pseudotyped material from HIV, FIV, and EIAV vectors have been shown to efficiently transduce neurons in various areas of the brain and express the transgene(s) for as long as anyone has looked so far. Such experiments have been performed in mice, rats, and monkeys. Thus, animal model data for the treatment of lyso-somal storage diseases (112,113), Huntington's disease (114), Parkinson's disease (79,115,116), and other diseases have been generated. One prominent example has been the recovery from a model of Parkinson's disease in primates (79) after administration of an HIV vector encoding glial cell derived nerve growth factor. It is hard to see how a protein can be delivered chronically to large numbers of people, as may be needed, without some kind of gene therapy. The flip side of this is that, of course, it is necessary have some control over expression of a growth factor in the brain, and this technology (e.g., 117-120) will also need to be clinically developed. Nevertheless, the results are encouraging enough that several groups are starting to undertake this.

A related area that shows promise is the delivery of genes to the eye for treatment of various types of retinopathies, including macular degeneration and diabetic retinopathy (121-125). This interest springs, largely, from the observation that lentiviral vectors seem to have particular tropism for pig-mented retinal epithelial cells in several species. Most proposals center on preventing the disorganized angiogenesis that follows disease onset and try to prevent it as it exacerbates clinical symptoms. There is currently a clinical trial underway using adenoviral vectors encoding the angiogenesis inhibitor pigment derived epithelium factor (126), but presumably len-

Figure 4 Efficient transduction of neural tissue in vivo with a lentiviral vector. The photomicrograph shows GFP expression in the striatum of a squirrel monkey at 18 days following infusion of 7.5 uL of an EIAV-based VSV-G pseudotyped vector [pSMART(1)G]) encoding GFP (2.4 X 1010 TU/mL) into the right caudate nucleus. Lentiviral vector was infused at a rate of 0.5 uL/min. Extensive expression of GFP can be seen throughout both cell bodies and processes. Cells morphologically similar to both neurons and glia expressed high levels of GFP. See the color insert for a color version of this figure.

Figure 4 Efficient transduction of neural tissue in vivo with a lentiviral vector. The photomicrograph shows GFP expression in the striatum of a squirrel monkey at 18 days following infusion of 7.5 uL of an EIAV-based VSV-G pseudotyped vector [pSMART(1)G]) encoding GFP (2.4 X 1010 TU/mL) into the right caudate nucleus. Lentiviral vector was infused at a rate of 0.5 uL/min. Extensive expression of GFP can be seen throughout both cell bodies and processes. Cells morphologically similar to both neurons and glia expressed high levels of GFP. See the color insert for a color version of this figure.

cells (61). The indications that are most discussed here are treatments that require systemic protein over long periods of time such as hemophilia A and B.

There is little investigation of transducing bone marrow in vivo but a lot of effort has been devoted to using the vectors to deliver genes to hematopoietic stem cells (HSCs). This is because the transplantation technology is in clinical use and the lentiviral vectors offer the opportunity to transduce the cells in vitro before reimplantation, without the need to make these cells pass through the cell cycle, as has been required for MLV-based vectors. This has been performed in mouse models and in human HSC in immune-deficient mouse models (133,134). Targets here include genetic diseases such as thalassemia (135,136), sickle-cell anemia (137), and chronic granulomatous disease (138). In addition, anti-HIV therapy has been investigated to provide blood cells derived from HSC that are protected from HIV or simply transducing T cells directly (139). Among the anti-HIV strategies are rescuable anti-HIV genes (140), antisense (141), and iRNA (142) to constitutively protect cells (143).

Because of the characteristics of the vector, there has only been a small amount of activity in cancer (e.g., 144,145), a target that has attracted a lot of attention from other vector systems. In addition, there have been attempts to use the vectors for lung application, such as cystic fibrosis (146), where some understanding of the issues around tranducing the apical side of the lung epithelial cells has been developed. Other targets include collagen deficiency (147) and kidney disease (148).

tiviral vectors may be a better fit. Once again, however, clinically it would be desirable to be able to switch the gene off. If a VSV-G lentiviral vector is injected intravenously in a mouse, it localizes mainly to the liver, spleen, and bone marrow (74,127,128). The levels decline somewhat over time in the liver and spleen but may stay at the initial level in the marrow. These levels are not completely consistent but are in the 0.1% to 30% range of total cell sampled. This observation points to some other promising target tissues, namely, liver and bone marrow. The spleen observation raised the possibility of inducing immune responses and has not attracted as much attention, except in vaccine-oriented applications.

The ability of lentiviral vectors to transduce liver tissue in mice and rats has been the subject of some debate, and there are definitely some issues that are not fully understood. Some investigators see the need to make the liver cells at least be part of a regenerating or proliferating tissue (34,129,130), whereas others experience no problem in generating therapeutic levels of systemic protein after liver targeted transduction (35,74,131). It seems likely that such differences are due to subtle vector differences or precise details of the delivery method, which have not yet been completely identified. It does seem that explanted primary hepatocytes that are not replicating are efficiently transduced (132), and pseudotyping with Sendai virus envelope has been used to try to target liver

Was this article helpful?

0 0

Post a comment