Inability of DNAzymes to cleave RNA in vivo is due to limited Mg $$^{2+}$$ 2 + concentration in cells

Abstract

Sequence specific cleavage of RNA can be achieved by hammerhead ribozymes as well as DNAzymes. They comprise a catalytic core sequence flanked by regions that form double strands with complementary RNA. While different types of ribozymes have been discovered in natural organisms, DNAzymes derive from in vitro selection. Both have been used for therapeutic down-regulation of harmful proteins by reducing drastically the corresponding mRNA concentration. A priori DNAzymes appear advantageous because of the higher haemolytic stability and better cost effectiveness when compared to RNA. In the present work the 10-23 DNAzyme was applied to knockdown expression of the prion protein (PrP), the sole causative agent of transmissible spongiform encephalopathies. We selected accessible target sequences on the PrP mRNA based on a sequential folding algorithm. Very high effectivity of DNAzymes was found for cleavage of RNA in vitro, but activity in neuroblastoma cells was very low. However, siRNA directed to the identical target sequences reduced expression of PrP in the same cell type. According to our analysis, three Mg[Formula: see text] bind cooperatively to the DNAzyme to exert full activity. However, free ATP binds the Mg[Formula: see text] ions more strongly and already stoichiometric amounts of Mg[Formula: see text] and ATP inhibited the activity of DNAzymes drastically. In contrast, natural ribozymes form three-dimensional structures close to the cleavage site that stabilize the active conformation at much lower Mg[Formula: see text] concentrations. For DNAzymes, however, a similar stabilization is not known and therefore DNAzymes need higher free Mg[Formula: see text] concentrations than that available inside the cell.

Molecular Features and Metal Ions That Influence 10-23 DNAzyme Activity

“…A growing number of 10-23 DNAzyme variants is currently in preclinical model studies and a small selection in clinical trials focusing on the treatment of basal cell carcinoma and Th2-driven asthma [15,20] (also see e.g., Reference [31] for more detailed review of clinically relevant aspects). While the DNAzyme approach offers a very attractive therapeutic strategy, contradicting observations are found in in vivo experiments, including studies that report on limited catalytic activity under physiological conditions [11,22]. The current situation implies that there is an urgent need to better understand the fundamental processes of DNA-mediated catalysis to enable the design of DNAzymes with improved in vivo activity and unravel the full therapeutic potential of DNAzymes.…”

“…The thermodynamic stability of the DNAzyme:RNA complex depends on reactants' structures, which have to be dissolved at least partially prior to association, on the lengths of the formed helices, the basepair stacking in the helices, and the ionic strength. A well-designed DNAzyme sequence should form neither intramolecular nor intermolecular basepairs with itself; the RNA or at least the target region should also have a low degree of structure [11,46,47]. The stability of the complex increases with increasing length of the helical arms and strongly stacking basepairs [48,49], but a too high stability of the helices inhibits dissociation of the DNAzyme:product complex (3) and thus catalytic turnover [11,50].…”

“…A well-designed DNAzyme sequence should form neither intramolecular nor intermolecular basepairs with itself; the RNA or at least the target region should also have a low degree of structure [11,46,47]. The stability of the complex increases with increasing length of the helical arms and strongly stacking basepairs [48,49], but a too high stability of the helices inhibits dissociation of the DNAzyme:product complex (3) and thus catalytic turnover [11,50]. Increasing ionic strength overcomes the electrostatic repulsion of the polyelectrolytic nucleic acid single strands (1), leads to higher stability and denaturation temperature of the complex, but may be varied only in vitro while ionic conditions are given in a selected in vivo system.…”

“…Most studies described below have not been carried out under reaction conditions at saturating substrate concentrations as required by the Michaelis-Menten model. There is, however, sufficient experimental evidence in the literature (for examples see References [1,11,33,[52][53][54][55]) to assume that the Michaelis-Menten model is suitable to describe 10-23 DNAzyme catalysis. Under single-turnover conditions, that is, DNAzyme is in excess over RNA or the reaction is started by addition of M 2+ to the preformed Dz:RNA complex, as done in many of the analyses described below, a simple first-order reaction Dz:RNA…”

“…However, a major obstacle for its in vivo application is the high dependency of the DNAzyme on divalent metal ions. Indeed, recent studies suggest that the 10-23 DNAzyme is catalytically inactive under the conditions inside cells and that visible knockdown effects could be attributed to antisense effects [11,22]. Despite the efforts of several groups to improve catalytic performance by designing DNAzymes that function at lower metal ion concentration, with different metal ion specificity, or even in the complete absence of metal ion cofactors, the activity of the DNAzyme under conditions resembling the cell remains low [23][24][25][26].…”