Purpose: The blood brain barrier (BBB) represents a significant hurdle in getting therapeutic compounds into brain to treat different neurological diseases. Compared to lipophilic small molecule drugs, the situation is more dire for macromolecules such as oligonucleotides (ODN), necessitating targeted delivery to the brain. Amongst many other approaches attempted to date, polyethyleneimine (PEI) has been the most popular non-viral vector utilized for delivering genes/oligonucleotides. However, it has the disadvantages of being cytotoxic to cells, tends to agglomerate and gets easily taken up by reticuloendothelial system without appropriate shielding. Previously our lab conjugated Biotin-poly(ethylene glycol)-succinimidyl ester (BIO-PEG-NHS) with low molecular weight PEI, which was then again linked to streptavidin conjugated transferrin receptor antibody 8D3 (8D3-BPP). 8D3-SA-BPP was successfully used in targeted delivery of NFKB decoys to treat experimental stroke model of mouse model (MCAO). However due to the PEG dilemma associated with shielding effects, reported hypersensitivities and being non-biodegradable, the search for alternative shielding agents is ongoing. Hydroxyethyl Starch (HES), a biodegradable polymer, which has long been safely used as a plasma volume expander in humans, has been recently introduced as a shielding agent in conjugation with PEI to deliver genes in vitro and shows promise to be used with PEI for targeted delivery to the brain. However, since HES and PEI are available in various molecular weights and chemical structures, some optimization studies are necessary to elucidate the best combination of HES and PEI in terms of molecular weight and DNA condensation properties.
Methods: HES-PEI conjugates having combinations of different molecular weights (10 or 20 kDa HES, 4 or 20 kDa Linear PEI) were synthesized by Schiff’s base formation followed by reductive amidation. A molar excess of HES was used to ensure 1:1 conjugation with the linear PEI. The conjugates were then subjected to cation exchange chromatography and dialysis for purification, followed by lyophilization. The confirmation of synthesis was done through NMR and size exclusion chromatography combined with Multi Angle light scattering. The HES moiety in the conjugates were then biotinylated with Biotin-PEG4-Hydrazide, purified and lyophilized (Figure 2). The degree of biotinylation of the conjugates were then assessed by the HABA assay . The optimum conjugates for further study were then selected based on cytotoxicity assay, particle size analysis, zeta potential measurements and oligonucleotide condensation studies with Ethidium Bromide. Particle size analysis, zeta potential measurements and condensation studies were done with NFΚB decoy ODN having the following sequence 5’ -CCTTGAAGGGATTTCCCTCC-3’ and 3’ -GGAACTTCCCTAAAGGGAGG-5’. The studies were done in different n/p ratios (3:1, 6:1 and 9:1) in absence or presence of serum and at different HES-PEI:PEI mixtures (50:50 and 10:90).
Results: The HABA assay showed that the conjugates containing HES 10 kDa and HES 20 kDa were biotinylated approximately 2.55% and 9% by weight respectively. The conjugates shielded the PEI at varying degrees when compared to naked PEI, as seen with particle size analysis and zeta potential in absence and presence of serum. Compared to naked PEI, the zeta potential of HES-PEI conjugates were significantly lower, ranging from -10 to +10 mV indicating shielding effects against agglomeration. However, the HES20kDa-PEI20kDa and HES10kDa-PEI20kDa conjugates yielded particles sizes ranging from approximately 200-400 nm in different conditions and n/p ratios, indicating they may not be suitable for in-vivo uptake studies. HES10kDa-PEI4kDa and HES20kDa-PEI4kDa conjugates were found to have much more favorable particles sizes which stayed well below 200 nm in different n/p ratios. However, the HES10kDa-PEI4kDa particle sizes were larger when mixed with different ratios of PEI while HES20kDa-PEI4kDa retained the particle sizes below 200 nm even after being mixed with upto 50:50 HES-PEI:PEI. HES conjugates had significantly lower cytotoxicity compared to PEI alone as seen up to 24 hours. The DNA condensation studies yielded the ODNs were condensed optimally at n/p ratio of 6 in line with existing literature and our previous lab findings with 8D3-BPP.
Conclusion: The positive results from the HABA assay, which utilizes streptavidin indicates that there will be appropriate binding of 8D3-SA to the conjugate. Even though there is a lot of biotin on the surface of the conjugates, since the molecular weight of IgG is 150 kDa and the molecular weight of the conjugate is approximately 24 kDa (HES20-PEI4) we do not anticipate more than one antibody binding to each conjugate. Our experiments show HES20kDa-PEI4kDa to be the most optimal conjugate for further uptake studies in brain endothelial cells as well as different animal models in the future.
Franz Hack– PhD Candidate, University of Jena
Behnam Noorani– Research Assistant, Texas Tech University Health Sciences Center
Dagmar Fischer– Professor of Pharmaceutical Technology and Biopharmacy, University of Jena
Bickel Ulrich– Professor, Texas Tech University Health Sciences Center