Gene therapy may be defined as the administration of exogenous DNA, in the form of intact gene or genes, for therapeutic purposes. Gene therapies usually have two major components, the DNA-containing molecule itself (the 'construct'), and an administrating adjuvant (the 'vector'). Vectors are usually necessary because genes are large, hydro-philic molecules that do not readily cross lipid membranes, although in some cases constructs are injected directly, without a vector (termed 'naked DNA').
Vectors may be viral or non-viral. Viral vectors include:
• Potentially pathogenic DNA viruses. These include adenoviruses and pox or vaccinia viruses. Both virus types can replicate in mammalian cytoplasm, whether or not the host cell is in mitosis or quiescent, and usually elicit a host immune response.
• Herpes simplex virus I (HSV1). This also contains double-stranded DNA, but replicates in the nucleus of cells that are successfully infected, which need not themselves be dividing.
• Non-pathogenic adeno-associated viruses. These parvoviruses carry single-stranded DNA and are able to integrate into a broad range of non-dividing cells.
• Retroviruses. These RNA-containing viruses exist in an envelope derived from host cell membrane, and thus do not usually elicit vigorous immune responses. Retroviruses also tend to replicate only in dividing cells.
It is perhaps surprising that naked DNA can cause gene expression. Current examples where this concept has been proven include genes injected into skeletal and smooth muscle. Gold-coated DNA containing gold particles may also be inserted into cells by a 'gene gun', where electrostatic or gas pressure-powered displacement from a plastic matrix occurs.
Liposomal envelopes can also transport substances across cell membranes which would otherwise be repelled by the hydrophilicity of the gene construct. Liposomes may be constructed that are either anionic or cationic, and can also be coated with antibodies that will target specific antigen-presenting cells.
DNA protein conjugates are again without a vector. These act as ligands for specific cell surface receptors. Some cell types will then convey the gene construct during the ordinary processes of endocy-tosis.
Two-stage delivery systems are also under development, often manipulating somatic tissue ex vivo. A good example would be following the biopsy of some bone marrow from a patient. The gene therapy can be introduced into the biopsy material, using either a viral or a non-viral vector in vitro, until expression has definitely been established. Thereafter, the biopsy may be used as an autolo-gous bone marrow transplantation, with the intention of its proliferation and generation of the desired protein product.
Early human gene transfer experiments in lymphocyte marking studies began in 1989. These early studies showed that gene transfer was feasible and could be well-tolerated, although there was no demonstrable therapeutic benefit. The first human gene therapy clinical trial was in 1990, in a patient with adenosine deaminase (ADA) deficiency; initial responsiveness proved not to be uniform when the series of cases was extended, possibly due to the fact that the disease phenotype could be elicited by a variety of genotypes.
There are some a priori characteristics for diseases that are likely to be attractive targets for gene therapy. There is a fundamental contrast between the gene therapy of inherited and acquired disorders. For inherited disorders, the absolute or relative deficiency of a particular protein needed for health may be correctable, e.g. the enzyme needed to reverse Gaucher's disease.
However, theoretically, there may be congenital disorders involving relative deficiencies of a particular protein, where therapeutic-induced overexpression is as likely to be as harmful as underexpression. Controlling gene expression is likely to be more difficult than merely inducing it. The thallassemias are a priori a good example of this problem. Overproduction of the missing haemoglobin chain is unlikely to be helpful to the patient. Similarly, when the principal desired target for gene therapy is a specific target organ, then overexpression of genes in other tissues may create tolerability problems.
The pharmacokinetics of gene therapy, and its relationship to dynamic effects, are very different form the orthodox pharmaceutical situation (Table 17.1). Ledley and Ledley (1994) have proposed a corollary of traditional PK-PD modeling, predicated upon the specific events in the cellular response to gene uptake and activation. These authors have developed a six-compartment model, which appears to have general applicability, to evaluate the apparent kinetic properties of a therapeutic gene product. This leads to the possibility of designing dosing regimens and relating them to measurements of expression and efficacy responses.
Acquired disorders are likely to have alternative therapeutic approaches. An increase in the production of some cytokine that is a normal response to a tumour might be a fairly direct strategy. On the other hand, some other biochemical process that indirectly compensates for a disease or injury could be another. A good example might be the differential sensitization of cells in a tumor to a particular cytotoxic drug, thus obtaining enhanced thera peutic response, permitting the use of lower doses of cytotoxic, and minimizing dose-limiting systemic adverse effects.
There are two areas of specific tolerability concerns associated with gene therapies, related to the expressed gene product and the vector. Both are immunological in nature, and may lead to therapeutic ineffectiveness.
If the gene therapy causes the production of a protein that was previously absent in the body, then an immune response is likely. This is simply analogous to patients who used to become resistant to xenobiotic insulins, and has also been demonstrated in the case of human factor VIII in some hemophiliacs. Escalating doses may be needed to maintain efficacy, or efficacy may be eventually lost. Furthermore, viral vectors are liable to replicate and also to elicit immune responses, just as for any vaccination, creating many of the same problems. Resistance to way gene therapy can result from immunization against either the construct or the vector.
One approach has been to develop strains of many of the viruses listed above as 'replication-defective' or 'replication-incompetent'. These viruses are cultured initially in conditions that provide some nutrient or element of the replicating machinery exogenously. Spontaneous mutations then create strains of virus that cannot replicate unless the crucial element is provided, and it is assumed that after human administration this will be the case. There is always the concern that, after injection, the virus will find some way to overcome its incompetency, e.g. by recruitment of the host cell machinery for this purpose. Similarly, non-viral vectors may offer an advantage by presenting the patient with less foreign antigen.
Was this article helpful?