Loco-regional cancer drug therapy: present approaches and rapidly reversible hydrophobization (RRH) of therapeutic agents as the future direction.

Insufficient drug uptake by solid tumors remains the major problem for systemic chemotherapy. Many studies have demonstrated anticancer drug effects to be dose-dependent, although dose-escalation studies have resulted in limited survival benefit with increased systemic toxicities. One solution to this has been the idea of loco-regional drug treatments, which offer dramatically higher drug concentrations in tumor tissues while minimizing systemic toxicity. Although loco-regional delivery has been most prominent in cancers of the liver, soft tissues and serosal peritoneal malignancies, survival benefits are very far from desirable. This review discusses the evolution of loco-regional treatments, the present approaches and offers rapidly reversible hydrophobization of drugs as the new future direction.

Rapidly reversible hydrophobization: an approach to high first-pass drug extraction.

We have investigated a rapidly reversible hydrophobization of therapeutic agents for improving first-pass uptake in locoregional drug therapy. This approach involves the attachment of a hydrophobic moiety to the drug by highly labile chemical linkages that rapidly hydrolyze upon injection. Hydrophobization drastically enhances cell-membrane association of the prodrug and, consequently, drug uptake, while the rapid lability protects nontargeted tissues from exposure to the highly active agent. Using the membrane-impermeable DNA intercalator propidium iodide, and melphalan, we report results from in vitro cellular internalization and toxicity studies. Additionally, we report in vivo results after a single liver arterial bolus injection, demonstrating both tumor targeting and increased survival in a mouse tumor model.

Mechanism of plasmid delivery by hydrodynamic tail vein injection. I. Hepatocyte uptake of various molecules.

BACKGROUND: The hydrodynamic tail vein (HTV) injection of naked plasmid DNA is a simple yet effective in vivo gene delivery method into hepatocytes. It is increasingly being used as a research tool to elucidate mechanisms of gene expression and the role of genes and their cognate proteins in the pathogenesis of disease in animal models. A greater understanding of its mechanism will aid these efforts and has relevance to macromolecular and nucleic acid delivery in general. METHODS: In an attempt to explore how naked DNA enters hepatocytes the fate of a variety of molecules and particles was followed over a 24-h time frame using fluorescence microscopy. The uptake of some of these compounds was correlated with marker gene expression from a co-injected plasmid DNA. In addition, the uptake of the injected compounds was correlated with the histologic appearance of hepatocytes. RESULTS: Out of the large number of nucleic acids, peptides, proteins, inert polymers and small molecules that we tested, most were efficiently delivered into hepatocytes independently of their size and charge. Even T7 phage and highly charged DNA/protein complexes of 60-100 nm in size were able to enter the cytoplasm. In animals co-injected with an enhanced yellow fluorescent protein (EYFP) expression vector and fluorescently labeled immunoglobulin (IgG), hepatocytes flooded with large amounts of IgG appeared permanently damaged and did not express EYFP-Nuc. Hepatocytes expressing EYFP had only slight IgG uptake. In contrast, when an EYFP expression vector was co-injected with a fluorescently labeled 200-bp linear DNA fragment, both were mostly (in 91% of the observed cells) co-localized to the same hepatocytes 24 h later. CONCLUSIONS: The appearance of permanently damaged cells with increased uptake of some molecules such as endogenous IgG raised the possibility that a molecule could be present in a hepatocyte but its transport would not be indicative of the transport process that can lead to foreign gene expression. The HTV procedure enables the uptake of a variety of molecules (as previous studies also found), but the uptake process for some of these molecules may be associated with a more disruptive process to the hepatocytes that is not compatible with successful gene delivery.

Mechanism of plasmid delivery by hydrodynamic tail vein injection. II. Morphological studies.

BACKGROUND: The efficient delivery of plasmid DNA (pDNA) to hepatocytes by a hydrodynamic tail vein (HTV) procedure has greatly popularized the use of naked nucleic acids. The hydrodynamic process renders onto the tissue increased physical forces in terms of increased pressures and shear forces that could lead to transient or permanent membrane damage. It can also trigger a series of cellular events to seal or reorganize the stretched membrane. Our goal was to study the uptake mechanism by following the morphological changes in the liver and correlate these with the fate of the injected plasmid DNA. METHODS: We utilized both light microscopic (LM) and electron microscopic (EM) techniques to determine the effect of the HTV procedure on hepatocytes and non-parenchymal cells at various times after injection. The LM studies used paraffin-embedded livers with hematoxylin and eosin (H&E) staining. The immune-EM studies used antibodies labeled with sub-nanometer gold particles followed by silver enhancement to identify the location of injected pDNA at the subcellular level. The level of overall damage to liver cells was estimated based on alanine aminotransferase (ALT) release and clearance. RESULTS: Both the LM and EM results showed the appearance of large vesicles in hepatocytes as early as 5 min post-injection. The number of vesicles decreased by 20-60 min. Plasmid DNA molecules often appeared to be associated with or inside such vesicles. DNA could also be detected in the space of Disse, in the cytoplasm and in nuclei. Non-parenchymal cells also contained DNA, but HTV-induced vesicles could not be observed in them. CONCLUSIONS: Our studies suggest an alternative or additional pathway for naked DNA into hepatocytes besides direct entry via membrane pores. It may be difficult to prove which of these pathways lead to gene expression, but the membrane pore hypothesis alone appears insufficient to explain why expression happens preferentially in hepatocytes. Further study is needed to delineate the importance of each of these putative pathways and their interrelationship in enabling oligonucleotide (siRNA) activity and pDNA expression.

Efficient in vitro and in vivo expression of covalently modified plasmid DNA.

The tracking of plasmid DNA (pDNA) movement within cells requires the attachment of labels to the DNA in a manner such that: (a) the pDNA remains intact during the labeling process and (b) the labels remain stably attached to the DNA. Keeping these two criteria in mind, we have recently developed a series of alkylating reagents that facilitate the one-step, covalent attachment of compounds directly onto nucleic acids in a nondestructive manner. Using these DNA-alkylating reagents, we have attached a wide range of both fluorescent and nonfluorescent reporter molecules onto pDNAs. We now show that even with the covalent attachment of various marker compounds, the pDNA remains expression competent. The ability to create labeled, expression-competent DNA allows for the simultaneous tracking of both pDNA location and reporter gene expression within living or fixed cells.

The role of a microscopic colloidally stabilized phase in solubilizing oligoamine-condensed DNA complexes.

DNA complexes of spermine and spermidine become resolubilized at very high concentrations of the oligoamine. It has been postulated that high oligoamine concentrations shift the DNA from the globule back to the coil phase. The present study indicates that DNA resolubilization at high concentrations of spermine and spermidine is explained by formation of small particles of condensed DNA that cannot be precipitated by centrifugation. The fact that DNA stays condensed during resolubilization was confirmed using a relatively new condensation assay and three independent microscopic techniques. A considerable portion of DNA was found to be in particles with diameter <100 nm. Formation of such small particles is likely to be caused by colloidal forces. The ability to form small, condensed DNA particles in solutions that contain high concentrations of oligocation should aid in the design of synthetic DNA vectors for gene transfer and gene therapy and in the handling of DNA for diagnostic studies.

Entrapment and condensation of DNA in neutral reverse micelles.

DNA condensation and compaction is induced by a variety of condensing agents such as polycations. The present study analyzed the structure of plasmid DNA (DNA) in the small inner space of reverse micelles formed from nonionic surfactants (isotropic phase). Spectroscopic studies indicated that DNA was dissolved in an organic solvent in the presence of a neutral detergent. Fluorescent quenching of ethidium bromide and of rhodamine covalently attached to DNA suggested that the DNA within neutral, reverse micelles was condensed. Circular dichroism indicated that the DNA structure was C form (member of B family) and not the dehydrated A form. Concordantly, NMR experiments indicated that the reverse micelles contained a pool of free water, even at a ratio of water to surfactant (Wo) of 3.75. Electron microscopic analysis also indicated that the DNA was in a ring-like structure, probably toroids. Atomic force microscopic images also revealed small, compact particles after the condensed DNA structures were preserved using an innovative cross-linking strategy. In the lamellar phase, the DNA was configured in long strands that were 20 nm in diameter. Interestingly, such DNA structures, reminiscent of "nanowires," have apparently not been previously observed.

Self-quenched covalent fluorescent dye-nucleic acid conjugates as polymeric substrates for enzymatic nuclease assays.

A fluorescent method is described for assessing nuclease activity. The technique is based on the preparation of quenched fluorophore-nucleic acid covalent conjugates and their subsequent dequenching due to degradation by nucleases. The resulting fluorescence increase can be measured by a spectrofluorometer and exhibits subpicogram per milliliter sensitivity level for RNase A and low picogram per milliliter level for DNase I. The method is adaptable for quantitative nuclease inhibitor testing.