Search

The stability of the complexes between DNA and nonviral vectors is a crucial parameter for efficient gene delivery into target cells. The stability must be high enough to prevent any dissociation during interaction with the plasma membrane but low enough to allow the dissociation that is required for efficient internalization into the nucleus. In this report, we investigated the stability of complexes of DNA with two cysteine surfactants (guanidinocysteine N-decylamide, C10−C^G+, and ornithinyl-cysteinyl-tetradecylamide, C14−CO), able to convert themselves, via oxidative dimerization, into cationic cystine lipids. To this end, we determined the critical aggregation concentration (cac) and the binding constants of the surfactants for DNA by using the fluorescence quenching of the DNA bis-intercalating agent, YOYO-1, that results from the dye clustering induced by the collapse of DNA. The cac's of C10−C^G+ and C14−CO monomeric forms are 2.5 and 1 μM, respectively, and are slightly less than the 5 μM value for CTAB, taken as a model of nondimerizable surfactant. Dimerization of C10−C^G+ and C14−CO reduces the cac to 400 and 1 nM, respectively. The strong stabilization induced by oxidation of C14−CO is further confirmed by the increase in the rigidity of the micellelike domains in the complexes, as deduced from the rotational correlation time of the hydrophobic probe 1,6-diphenylhexatriene. In keeping with the stability data, no dissociation of the (C14−CO)2/DNA complexes occurs in the presence of neutral vesicles (that mimic the external leaflet of the plasma membrane), while a significant dissociation was observed with (C10−C^G+)2/DNA complexes and an even larger one with CTAB/DNA complexes. Similarly, (C14−CO)2/DNA complexes do not dissociate in the presence of anionic vesicles (that mimic the cytoplasmic leaflet of the plasma membrane), while a complete dissociation and DNA release occurs with both (C10−C^G+)2/DNA and CTAB/DNA complexes. Both the initial interaction with the plasma membrane and the release of DNA in the cytoplasm are strongly dependent on the stability of the complexes obtained with this new class of nonviral vectors.

Chemical modification of 2 +/- 0.5 arginine residues of porcine pancreatic elastase by 1,2-cyclohexanedione leads to an 85 +/- 5% loss of activity with the specific substrate N-succinyltrialanine p-nitroanilide. Modification of additional arginines does not completely abolish the enzyme activity. The modification reaction is very fast (second order rate constant = 0.24 M-1 S-1) and involves only arginine residues. Acetyltetraalanine or trifluoroacetyltetraalanine decreases the rate of cyclohexanedione-induced inactivation of the enzyme but does not significantly change the number of modified arginine residues. Other dicarbonyl reagents, butanedione or phenylglyoxal, also react with elastase but at much lower rates. Cyclohexanedione-modified elastase is partially active against a series of synthetic substrates of varying chain length. The partial inhibition results from a 2- to 5-fold increase in Km while kappa cat is increased for most substrates. For N-succinyltrialanine p-nitroanilide both the acylation and deacylation rate constants are decreased. The Ki values of a series of acetylated and trifluoroacetylated inhibitors increase 2- to 5-fold. Modified elastase is still able to react with fibrous elastin and with plasma alpha 1-proteinase inhibitor but at significantly lower rates. Modification of one arginine residue alters the properties of the calcium-binding site of elastase as demonstrated by terbium luminescence experiments. The affinity of enzyme for terbium is decreased by a factor of 10 and the circularly polarized luminescence spectrum of the terbium-elastase complex is considerably flattened. Modification of further arginine residues does not increase the extent of these alterations. Circular dichroism shows that the overall conformation of elastase is not altered following arginine modification. We speculate that the two residues modified by cyclohexanedione are Arg 65, located at about 8 A from the metal ion-binding site, and Arg 217A, located at the S'3 subsite of elastase.

The microviscosity of micelles of the dimeric (gemini) surfactants alkanediyl-α,ω-bis(dimethyldodecylammonium bromide), referred to as 12-s-12 (s = carbon number of the alkanediyl spacer), has been obtained from measurements of the fluorescence anisotropy of micelle-solubilized diphenylhexatriene (DPH) and of the intensity ratio IM/IE of micelle-solubilized dipyrenylpropane (P3P, IM and IE = intensities of the monomer and excimer emissions). The micropolarity of these surfactant micelles was characterized by the value of the fluorescence intensity ratio I1/I3 of the first and third vibronic peaks in the emission spectra of micelle-solubilized pyrene. The microviscosity and micropolarity of the trimeric surfactant 12-3-12-3-12 and of the tetrameric surfactant (12-3-12-4-12-3-12) micelles have also been determined in order to gain information on the effect of the degree of oligomerization of the surfactant. The microviscosity and micropolarity of the asymmetric two-chain surfactants, the dimethyldodecylalkylammonium bromides, referred to as 12-s/2 (s/2 = carbon number of the variable alkyl chain varying from 1 to 10), which can be considered as the monomers of the 12-s-12 surfactants, have been determined for the sake of comparison. The I1/I3 vs s plots for the 12-s-12 and 12-s/2 series were found to be coincident, except for s = 2, revealing that the polarity sensed by pyrene in its micellar solubilization site is not affected by the “dimerization” of the surfactant. The micropolarity goes through a maximum for the two series of surfactants as a result of the change of location of the spacer or variable alkyl chain as s is increased. The microviscosity in the 12-s-12 series decreased rapidly as s increased above 3, then more slowly, and leveled out, but was nearly independent of s in the 12-s/2 series, in the range s/2 = 1−8. For the two series of surfactants the microviscosity decreased upon increasing temperature but increased when the micelle shape of the surfactant oligomer micelles changed from spherical to elongated. Also, the microviscosity increased significantly and nearly linearly with the number of dodecyl chains in going from DTAB to the dimer, trimer, and tetramer. These results are used to explain the low values of the quenching rate constants characterizing micellar solutions of surfactant oligomers and hydrophobic microdomains in solutions of water-soluble amphiphilic polymers (polysoaps), with respect to micelles of the corresponding monomeric surfactants.