Mechanical Flexibility of DNA: A Quintessential Tool for DNA Nanotechnology.

The mechanical properties of DNA have enabled it to be a structural and sensory element in many nanotechnology applications. While specific base-pairing interactions and secondary structure formation have been the most widely utilized mechanism in designing DNA nanodevices and biosensors, the intrinsic mechanical rigidity and flexibility are often overlooked. In this article, we will discuss the biochemical and biophysical origin of double-stranded DNA rigidity and how environmental and intrinsic factors such as salt, temperature, sequence, and small molecules influence it. We will then take a critical look at three areas of applications of DNA bending rigidity. First, we will discuss how DNA's bending rigidity has been utilized to create molecular springs that regulate the activities of biomolecules and cellular processes. Second, we will discuss how the nanomechanical response induced by DNA rigidity has been used to create conformational changes as sensors for molecular force, pH, metal ions, small molecules, and protein interactions. Lastly, we will discuss how DNA's rigidity enabled its application in creating DNA-based nanostructures from DNA origami to nanomachines.

Reselection Yielding a Smaller and More Active Silver-Specific DNAzyme.

Ag10c is a recently reported RNA-cleaving DNAzyme obtained from in vitro selection. Its cleavage activity selectively requires Ag+ ions, and thus it has been used as a sensor for Ag+ detection. However, the previous selection yielded very limited information regarding its sequence requirement, since only ∼0.1% of the population in the final library were related to Ag10c and most other sequences were inactive. In this work, we performed a reselection by randomizing the 19 important nucleotides in Ag10c in such a way that a purine has an equal chance of being A or G, whereas a pyrimidine has an equal chance of being T or C. The round 3 library of the reselection was carefully analyzed and a statistic understanding of the relative importance of each nucleotide was obtained. At the same time, a more active mutant was identified, containing two mutated nucleotides. Further analysis indicated new base pairs leading to an enzyme with smaller catalytic loops but with ∼200% activity of the original Ag10c, and also excellent selectivity for Ag+. Therefore, a more active mutant of Ag10c was obtained and further truncations were successfully performed, which might be better candidates for developing new biosensors for silver. A deeper biochemical understanding was also obtained using this reselection method.

Folding of the silver aptamer in a DNAzyme probed by 2-aminopurine fluorescence.

The RNA-cleaving Ag10c DNAzyme was recently isolated via in vitro selection and it can bind two Ag+ ions for activity. The Ag10c contains a well-defined Ag+ binding aptamer as indicated by DMS footprinting. Since aptamer binding is often accompanied with conformational changes, we herein used 2-aminopurine (2AP) to probe its folding in the presence of Ag+. The Ag10c was respectively labeled with 2AP at three different positions, both in the substrate strand and in the enzyme strand, one at a time. Ag+-induced folding was observed at the substrate cleavage junction and the A9 position of the enzyme strand, consistent with aptamer binding. The measured Kd at the A9 position was 18 µM Ag+ with a Hill coefficient of 2.17, similar to those obtained from the previous cleavage activity based assays. However, labeling a 2AP at the A2 position inhibited the activity and folding. Compared to other metal ions, Ag+ has a unique sigmoidal folding profile indicative of multiple silver binding cooperatively. This suggests that Ag+ can induce a local folding in the enzyme loop and this folding is important for activity. This study provides important biophysical insights into this new DNAzyme, suggesting the possibility of designing folding-based biosensors for Ag+.

Metal Sensing by DNA.

Metal ions are essential to many chemical, biological, and environmental processes. In the past two decades, many DNA-based metal sensors have emerged. While the main biological role of DNA is to store genetic information, its chemical structure is ideal for metal binding via both the phosphate backbone and nucleobases. DNA is highly stable, cost-effective, easy to modify, and amenable to combinatorial selection. Two main classes of functional DNA were developed for metal sensing: aptamers and DNAzymes. While a few metal binding aptamers are known, it is generally quite difficult to isolate such aptamers. On the other hand, DNAzymes are powerful tools for metal sensing since they are selected based on catalytic activity, thus bypassing the need for metal immobilization. In the last five years, a new surge of development has been made on isolating new metal-sensing DNA sequences. To date, many important metals can be selectively detected by DNA often down to the low parts-per-billion level. Herein, each metal ion and the known DNA sequences for its sensing are reviewed. We focus on the fundamental aspect of metal binding, emphasizing the distinct chemical property of each metal. Instead of reviewing each published sensor, a high-level summary of signaling methods is made as a separate section. In principle, each signaling strategy can be applied to many DNA sequences for designing sensors. Finally, a few specific applications are highlighted, and future research opportunities are discussed.

Two Completely Different Mechanisms for Highly Specific Na+ Recognition by DNAzymes.

Our view of the interaction between Na+ and nucleic acids was changed by a few recently discovered Na+ -specific RNA-cleaving DNAzymes. In addition to nonspecific electrostatic interactions, highly specific recognition is also possible. Herein, two such DNAzymes, named EtNa and Ce13d, are compared to elucidate their mechanisms of Na+ binding. Mutation studies indicate that they have different sequence requirements. Phosphorothioate (PS) substitution at the scissile phosphate drops the activity of EtNa 140-fold, and it cannot be rescued by thiophilic Cd2+ or Mn2+ , whereas the activity of PS-modified Ce13d can be rescued. Na+ -dependent activity assays indicate that two Na+ ions bind cooperatively in EtNa, and each Na+ likely interacts with a nonbridging oxygen atom in the scissile phosphate, whereas Ce13d binds only one Na+ ion in a well-defined Na+ aptamer, and this Na+ ion does not directly interact with the scissile phosphate. Both DNAzymes display a normal pH-rate profile, with a single deprotonation reaction required for catalysis. For EtNa, Na+ fails to protect the conserved nucleotides from dimethyl sulfate attack, and no specific Na+ binding is detected by 2-aminopurine fluorescence, both of which are different from those observed for Ce13d. This work suggests that EtNa binds Na+ mainly through its scissile phosphate without significant involvement of the nucleotides in the enzyme strand, whereas Ce13d has a well-defined aptamer for Na+ binding. Therefore, DNA has at least two distinct ways to achieve highly selective Na+ binding.

A Silver-Specific DNAzyme with a New Silver Aptamer and Salt-Promoted Activity.

Most RNA-cleaving DNAzymes require a metal ion to interact with the scissile phosphate for activity. Therefore, few unmodified DNAzymes work with thiophilic metals because of their low affinity for phosphate. Recently, an Ag+-specific Ag10c DNAzyme was reported via in vitro selection. Herein, Ag10c is characterized to rationalize the role of the strongly thiophilic Ag+. Systematic mutation studies indicate that Ag10c is a highly conserved DNAzyme and its Ag+ binding is unrelated to C-Ag+-C interaction. Its activity is enhanced by increasing Na+ concentrations in buffer. At the same metal concentration, activity decreases in the following order: Li+ > Na+ > K+. Ag10c binds one Na+ ion and two Ag+ ions for catalysis. The pH-rate profile has a slope of ∼1, indicating a single deprotonation step. Phosphorothioate substitution at the scissile phosphate suggests that Na+ interacts with the pro-Rp oxygen of the phosphate, and dimethyl sulfate footprinting indicates that the DNAzyme loop is a silver aptamer binding two Ag+ ions. Therefore, Ag+ exerts its function allosterically, while the scissile phosphate interacts with Na+, Li+, Na+, or Mg2+. This work suggests the possibility of isolating thiophilic metal aptamers based on DNAzyme selection, and it also demonstrates a new Ag+ aptamer.

An Exceptionally Selective DNA Cooperatively Binding Two Ca2+ Ions.

Ca2+ is a highly important metal ion in biology and in the environment, and thus there is extensive work in developing sensors for Ca2+ detection. Although many Ca2+ -binding proteins are known, few nucleic acids can selectively bind Ca2+ . DNA-based biosensors are attractive for their high stability and excellent programmability. We report a RNA-cleaving DNAzyme, EtNa, cooperatively binding two Ca2+ ions but to only one Mg2+ . Four DNAzymes with known Ca2+ -dependent activity were compared, and the EtNa had the best selectivity for Ca2+ . The EtNa is 90 times more active in Ca2+ than in Mg2+ . Phosphorothioate (PS) modification showed that both non-bridging oxygen atoms at the scissile phosphate contribute equally to Ca2+ binding. The pH-rate profile suggests two concurrent deprotonation reactions. EtNa was further engineered for Ca2+ sensing, and found to have a detection limit of 17 µm Ca2+ and excellent selectivity. The detection of Ca2+ in tap water was performed, and the result was comparable with that by ICP-MS. This study offers new fundamental insights into Ca2+ binding by nucleic acids and improved metal selectivity by having multiple cooperative metal binding sites.

Label-Free Ag⁺ Detection by Enhancing DNA Sensitized Tb(3+) Luminescence.

In this work, the effect of Ag⁺ on DNA sensitized Tb(3+) luminescence was studied initially using the Ag⁺-specific RNA-cleaving DNAzyme, Ag10c. While we expected to observe luminescence quenching by Ag⁺, a significant enhancement was produced. Based on this observation, simple DNA oligonucleotide homopolymers were used with systematically varied sequence and length. We discovered that both poly-G and poly-T DNA have a significant emission enhancement by Ag⁺, while the absolute intensity is stronger with the poly-G DNA, indicating that a G-quadruplex DNA is not required for this enhancement. Using the optimized length of the G7 DNA (an oligo constituted with seven guanines), Ag⁺ was measured with a detection limit of 57.6 nM. The signaling kinetics, G7 DNA conformation, and the binding affinity of Tb(3+) to the DNA in the presence or absence of Ag⁺ are also studied to reveal the mechanism of emission enhancement. This observation is useful not only for label-free detection of Ag⁺, but also interesting for the rational design of new biosensors using Tb(3+) luminescence.

Deciphering the role of Atg5 in nucleotide dependent interaction of Rab33B with the dimeric complex, Atg5-Atg16L1.

Autophagy is a lysosomal degradation pathway that degrades cytosolic constituents, including whole organelles and intracellular pathogens. Previous studies on various autophagy related genes revealed the importance of the Atg12-Atg5-Atg16 complex in autophagy. Atg16L1 is an effector of Golgi-resident Rab33B and the molecular mechanism of the interaction of Rab33B with either Atg16L1 or in complex with Atg5 is still elusive. In the current study, using the pull down and calorimetric approaches, we have dissected the molecular insights into the interaction of Rab33B with different regions of mouse Atg16L1 as well as with the dimeric complex, Atg5-mAtg16L1. Our in vitro observation suggests that Atg5 is pre-requisite for the augmented nucleotide dependent interaction of Rab33B with the dimeric complex, Atg5-Atg16L1. Moreover, the results reported here suggest that Arg-24 of Atg16L1 is crucial for its interaction with Atg5 which will have further implication in the binding of the dimeric complex to Rab33B.

A Silver DNAzyme.

Silver is a very common heavy metal, and its detection is of significant analytical importance. DNAzymes are DNA-based catalysts; they typically recruit divalent and trivalent metal ions for catalysis. Herein, we report a silver-specific RNA-cleaving DNAzyme named Ag10c obtained after six rounds of in vitro selection. Ag10c displays a catalytic rate of 0.41 min(-1) with 10 µM Ag(+) at pH 7.5 with 200 mM NaNO3, while its activity is completely inhibited with the same concentration of NaCl. Ag10c is highly specific for Ag(+) among all the tested metals. A catalytic beacon biosensor is designed by labeling a fluorophore and a quencher on the DNAzyme. Fluorescence enhancement is observed in the presence of Ag(+) with a detection limit of 24.9 nM Ag(+). The sensor shows a similar analytical performance in Lake Huron water. This is the first monovalent transition metal dependent RNA-cleaving DNAzyme. Apart from its biosensor application, this study strengthens the idea of exploring beyond the traditional understanding of multivalent ion dependent DNAzyme catalysis.

A New Na(+)-Dependent RNA-Cleaving DNAzyme with over 1000-fold Rate Acceleration by Ethanol.

Enzymes working in organic solvents are important for analytical chemistry, catalysis, and mechanistic studies. Although a few protein enzymes are highly active in organic solvents, little is known regarding nucleic acid-based enzymes. Herein, we report the first RNA-cleaving DNAzyme, named EtNa, that works optimally in concentrated organic solvents containing only monovalent Na(+). The EtNa DNAzyme has a rate of 2.0 h(-1) in 54% ethanol (with 120 mM NaCl and no divalent metal ions), and a Kd of 21 mm Na(+). It retains activity even in 72% ethanol as well as in DMSO. With 4 mm Na(+), the rate in 54% ethanol is >1000-fold higher than that in water. We also demonstrated the use of EtNa to measuring the ethanol content in alcoholic drinks. In total, this DNAzyme has three unique features: divalent metal independent activity, Na(+) selectivity among monovalent metals, and acceleration by organic solvents.

Searching for a DNAzyme Version of the Leadzyme.

The leadzyme refers to a small ribozyme that cleaves a RNA substrate in the presence of Pb(2+). In an optimized form, the enzyme strand contains only two unpaired nucleotides. Most RNA-cleaving DNAzymes are much longer. Two classical Pb(2+)-dependent DNAzymes, 8-17 and GR5, both contain around 15 nucleotides in the enzyme loop. This is also the size of most RNA-cleaving DNAzymes that use other metal ions for their activity. Such large enzyme loops make spectroscopic characterization difficult and so far no high-resolution structural information is available for active DNAzymes. The goal of this work is to search for DNAzymes with smaller enzyme loops. A simple replacement of the ribonucleotides in the leadzyme by deoxyribonucleotides failed to produce an active enzyme. A Pb(2+)-dependent in vitro selection combined with deep sequencing was then performed. After sequence alignment and DNA folding, a new DNAzyme named PbE22 was identified, which contains only 5 nucleotides in the enzyme catalytic loop. The biochemical characteristics of PbE22 were compared with those of the leadzyme and the two classical Pb(2+)-dependent DNAzymes. The rate of PbE22 rises with increase in Pb(2+) concentration, being 1.7 h(-1) in the presence of 100 µM Pb(2+) and reaching 3.5 h(-1) at 500 µM Pb(2+). The log of PbE22 rate rises linearly in a pH-dependent fashion (20 µM Pb(2+)) with a slope of 0.74. In addition, many other abundant sequences in the final library were studied. These sequences are quite varied in length and nucleotide composition, but some contain a few conserved nucleotides consistent with the GR5 structure. Interestingly, some sequences are active with Pb(2+) but none of them were active with even 50 mM Mg(2+), which is reminiscent of the difference between the GR5 and 8-17 DNAzymes.

Tandem DNAzymes for mRNA cleavage: choice of enzyme, metal ions and the antisense effect.

The concept of DNAzyme-based gene silencing via mRNA cleavage was proposed over twenty years ago. A number of studies regarding intracellular gene silencing have been reported as well. However, questions have been raised regarding the lack of enzyme activity in physiological buffer conditions and it is being doubted that in the previously reported studies, gene silencing might be simply due to an antisense effect. In this work, two classical DNAzymes for RNA cleavage are studied using both chimeric substrates and extracted mRNA. We concluded that the activity of the 8-17 DNAzyme is much higher than that of 10-23 in the same conditions. To illustrate and compare the effect of specific cleavage versus antisense effect in the best possible way, we used tandem DNAzymes. Specific mRNA cleavage occurred with Zn(2+), while with Mg(2+), even the inactive control DNAzymes showed a similar response, suggesting that the antisense effect might be the dominating phenomenon causing gene silencing. This study has thus clarified the choice of DNAzyme sequence, the effect of metal ions and a potential source of antisense effect.