BARRIERS TO DNA BASED DRUGS

INTRODUCTION

Ex vivo gene therapy approaches are predominantly concerned with gene delivery barriers presented by the cell itself. These intracellular barriers can include the plasma membrane, the endosome, or the nuclear membrane. For sustained expression, maintenance of the transgene within the nucleus might also be considered a barrier. In vivo gene therapy and, to some extent, ex vivo therapies, must be concerned not only with intracellular barriers but also with extracellular barriers. Extracellular barriers, such as those posed by a specific tissue or the immune system, can greatly diminish the effectiveness of in vivo gene delivery.

I. Intracellular barriers.

The plasma membrane is the first obstacle to be overcome in delivering genes into a cell. Our previous discussions on gene transfer vectors explored a variety of approaches used to deliver genes into a cell. All of the gene transfer vectors obtain entry into the cell either by endocytosis or membrane fusion, but a detailed mechanism of entry for many gene transfer vectors is still not fully understood. Clearly, viral vectors and targeted vectors rely on binding to cell surface molecules prior to cellular internalization. Liposomes, naked or complexed plasmid DNA must also interact with the plasma membrane, but may do so in a non-receptor binding fashion prior to cellular internalization. A schematic diagram of the potential intracellular fate of a gene transfer vector is shown below.

Once across the plasma membrane, gene transfer vectors must escape from the endosome into the cytosome. Inability to escape from the endosome remains problematic for many non-viral gene transfer vectors.

The last intracellular hurdle to gene delivery vectors is the nuclear envelope. Nuclear entry of exogenous DNA appears to be regulated by the nuclear pore complex(NPC). The NPC accommodates both passive diffusion and active transport. Molecules smaller than 10 nm in diameter or proteins less than 15 kDa, passively diffuse through the NPC into and out of the nucleus. Larger macromolecules require active transport for nuclear entry. The exact mechanism by which exogenous DNA passes through the NPC has not yet been determined, although it may be very similar to the transport of proteins larger than 15 kDa into the nucleus. The diagram below describes some of the steps involved in protein transport into the nucleus (1). Nucleases present either within the cytosol or the nucleus are obviously a concern of any DNA-based drug. The effect of nucleases upon exogenous DNA can most easily be reduced by rapid integration of the exogenous DNA.

II. Extracellular barriers.

Direct administration of gene transfer vectors into a patient, i.e., in vivo gene therapy, must contend not only with intracellular barriers, but also with extracellular barriers. Extracellular barriers presented to a gene transfer vector predominantly fall into one of two categories, specific tissue barriers or immunological barriers.

Tissue barriers can present a substantial hurdle for gene transfer vectors. For example, any gene therapy aimed at treating CNS disorders must contend with the blood-brain barrier. Currently, gene transfer vectors are directly injected into the brain at or near the desired brain region. Ultimately, development of a targeted CNS gene therapy agent which could be administered intravenously would be preferred to a surgical procedure for vector administration. Such a vector would need to be able to cross the blood-brain barrier. Also, the nerve itself is somewhat resistant to gene transfer by its very nature. Direct injection of a gene transfer vector into a nerve is not only technically difficult, but usually harmful to the nerve. The myelin sheath surrounding a nerve impedes the ability for vectors to transfer genes into a nerve. These unique characteristics of the CNS may require a vector possessing a natural tropism for the CNS, such as herpes simplex virus (HSV) vectors, to be developed. Connective tissue barriers can also be a substantial hurdle to gene delivery. The multiple connective tissue layers found in muscle can diminish the spread and overall infectivity of vectors administered intramuscularly. Epithelial cell linings can also interfere with a vector's ability to infect deeper cell layers. Also, postmitotic cells will not be infected by retroviral vectors. These few examples of potential tissue barriers to gene transfer agents serve as examples of how a particular vector system can be influenced by the target tissue or tissues surrounding it.

Immunological barriers of gene transfer agents are best understood for viral vectors. Gene transfer vectors are susceptible to inactivation by serum components are well as cellular and humoral immune responses.

Serum components can inactivate a variety of gene transfer agents. For example, naked plasmid DNA can interact with a variety of positively-charged serum proteins which impede its ability to deliver genes in vivo. Serum proteases/nucleases also degrade both viral and non-viral vectors.

Complement inactivation is the predominant serum component responsible for vector inactivation. One example, is the binding of complement component C1 to MMLV surface proteins. Complement C1 binding initiates the classical complement pathway, in an antibody-independent fashion, resulting in vector inactivation. It is presumed that similar complement interactions can play a role in inactivating other gene delivery systems. If complement interactions are not observed the first time a vector is administered but antibodies against the vector are formed, future administration of the vector will have to contend with the binding of circulating antibodies and subsequent complement inactivation.

Cellular immune responses against recombinant adenoviruses has been observed following intramuscular injection. Lymphocyte infiltrates have been observed as early as 24 hours after adenoviral vector administration. CD8+ cells were also observed near the injection site. These cellular-mediated responses are thought to be triggered by the expression of the viral proteins from the regions encoded by the recombinant vector genome. Cellular-mediated responses result in the elimination of those cells which were expressing the transgene.

Antibodies against both retroviral and adenoviral vectors have been observed following in vivo administration. In some cases these antibodies existed prior to vector administration. There appear to be a multitude of antibodies circulating within primate serum that interact with retroviral coat proteins and envelope components. Currently, hurdles posed by the immune system can be reduced through the use of immunosuppressant drugs. Tolerance, a state of specific immunological unresponsiveness, against viral vector proteins has been proven to be an effective means to overcome these hurdles in experimental models but may prove difficult to reproduce in clinical settings.

The degree to which intra- and extracellular barriers interfere with gene delivery is dependent upon the vector system and the target tissue. The unique characteristics of each vector system and the barriers it would encounter represent the growing frontiers of new vector development. This has also led to the investigation of the pharmacokinetics and pharmacodynamic parameters of DNA- based drugs.