A linear, amphoteric poly(amidoamine) nicknamed AGMA1, based on 4-aminobutylguanidine, or agmatine, was successfully prepared by Michael-type polyaddition of monoprotonated agmatine and 2,2-bis(acrylamido)acetic acid (BAC). Copolymers between AGMA1 and the biocompatible poly(amidoamine) ISA23 (deriving from the polyaddition of 2-methylpiperazine with BAC) were also prepared. Acid-base titrations gave for AGMA1 three acid dissociation constants, with pKa values of 2.25, 7.45, and >or=12.1, corresponding to a strong acid, a medium-weak base, and a strong base, respectively. The charge distribution profiles show that this polymer is prevailingly cationic at all physiological pH values, the positive net average charge per unit varying from about 0.5 at pH 7.4 to about 1.0 at pH 5, with an isoelectric point at pH approximately 10. Zeta-potential measurements confirmed this. Despite that, AGMA1 is nontoxic and nonhemolytic in vitro within all pH ranges tested (4-7.5). This is in contrast with the previously observed behavior of amphoteric PAAs, for instance ISA23, that are weakly hemolytic at pH 7.4 but highly hemolytic at pH 5/5.5. The lack of hemolytic activity of AGMA1 even at acidic pH values seems typical of the agmatine-BAC sequences and may be ascribed to their RGD-like structure. In fact, AGMA1-ISA23 copolymers behave in a way increasingly similar to that of ISA23; that is, they become hemolytic at low pH values as their ISA23 content increases.

AGMA1, a prevailingly cationic amphoteric polyamidoamine obtained by polyaddition of (4-aminobutyl)guanidine (agmatine) to 2,2-bis(acrylamido)acetic acid, was studied as a potential DNA carrier and transfection promoter. Fluorescein-labeled AGMA1 was prepared by conjugation with fluorescein isothiocyanate and its cell uptake, blood permanence, and body distribution studied. In spite of its cationic character, AGMA1 is neither toxic nor hemolytic in the pH range 4.0-7.4, circulates for a long time in the blood without preferentially localizing in the liver, easily enters HT-29 cells, gives stable complexes with DNA, and is endowed with good transfection efficiency, suggesting the ability to transport in the cytoplasm a DNA payload without any measurable membranolytic activity. If compared with other transfection promoters, including polyamidoamines of different structures, AGMA1 is apparently endowed with a unique combination of desirable requirements for a nonviral DNA polymer carrier and warrants potential as a transfection agent in vivo.

Bioresponsive poly(amidoamine)s (PAA)s are currently under development as endosomolytic polymers for intracellular delivery of proteins and genes. Here for the first time, small-angle neutron scattering (SANS) is used to systematically investigate the pH-dependent conformational change of an endosomolytic polymer, the PAA ISA 23. The radius of gyration of the ISA23 was determined as a function of pH and counterion, the aim being to correlate changes in polymer conformation with membrane activity assessed using a rat red blood cell haemolysis assay. With decreasing pH, the ISA23 radius of gyration increased to a maximum (R(g) approximately 80 A) around pH = 3, before subsequently decreasing once more. At high pH and therefore high ionic strengths, the polymer is negatively charged and adopts a rather compact structure (R(g) approximately 20 A), presumably with the dissociated carboxylic groups on the exterior of the polymer coil. At low pH, the coil again collapses (R(g) < 20 A), presumably due to the effects of the high ionic strength. It is concluded that the nature of the salt form has no direct bearing on the size of the polymer coil, but it does indirectly determine the prevailing pH and, hence, polymer conformation. Pulsed-gradient spin-echo NMR measurements were in good agreement with the SANS estimates of the radius of gyration, although ISA23 polydispersity does complicate the data interpretation/comparison. These results support the proposed mode of action of PAAs, namely a coil expansion on passing from a neutral pH (extracellular) to an acidic pH (endosomal and lysosomal) environments. The results do, however, suggest that the charge on the polymer shows a closer correlation with the haemolysis activity rather than the polymer conformation.

The pH-responsive poly(amidoamine)s (PAAs) have been previously described. Whereas ISA23 enhances transfection in vitro and ISA1 promotes the cytosolic delivery of the non-permeant toxins this process shows poor efficiency. The aim of this study was to prepare and evaluate PAA conjugates containing the membrane disrupting peptide melittin (MLT). It was hypothesised that PAA conjugation would reduce the haemolytic activity of MLT at pH 7.4, however, upon delivery to tumours by the EPR effect, the polymer would uncoil in an acidic environment exposing MLT and allowing it to interact with membranes. PAA-MLT conjugates were prepared using MLT as a comonomer together with bis-acryloylpiperazine, 2-methylpiperazine and bis-hydroxyethylethylenediamine (ISA1-like), or bis-acrylamidoacetic acid and 2-methylpiperazine (ISA23-like). The melittin content of the conjugates was 6-19% (w/w). Although ISA1-MLT improved gelonin delivery compared to the parent polymer ISA1 (␣ 13-fold increase) and showed pH-dependent haemolytic activity at a polymer concentration of 0.05 mg/ml, this conjugate also displayed high haemolytic activity at pH 7.4. In contrast, ISA23-MLT like the parent compound ISA23 did not deliver gelonin. However, this conjugate could have potential as a novel polymeric anticancer conjugate due to its lack of haemolytic activity at pH 7.4 and retention of cytotoxicity.

A comprehensive structural characterisation of cross-linked insoluble poly(amidoamine) (PAA) networks was performed by high-resolution magic angle spinning (HRMAS) NMR spectroscopy. Model samples with 20%, 40% and 80% cross-linking degrees were prepared and the best conditions to obtain high-resolution spectra in the gel phase determined. Whereas the samples with 20% and 40% cross-linking degrees could be exhaustively resolved and described, the sample with 80% cross-linking degree could not be characterised by this technique owing to insufficient mobility of the polymer segments. Even with this limitation, the method developed in this study can be reasonably considered as a general one, which enables exhaustive characterisation of cross-linked PAA networks of biomedical interest.

No abstract

No abstract