Despite myriad challenges, clinicians see room for progress.
Image credit: Shutterstock/David Tadevosian.
New Research In
Physical Sciences
Featured Portals
Articles by Topic
Biological Sciences
Featured Portals
Articles by Topic
-
Communicated by I. M. Gelfand, Rutgers, The State University of New Jersey, Piscataway, NJ, October 15, 2008 (received for review September 8, 2008)
Abstract
Biopolymers exhibit rough energy landscapes, thereby allowing biological processes to access a broad range of kinetic and thermodynamic states. In contrast to proteins, the energy landscapes of nucleic acids have been the subject of relatively few experimental investigations. In this study, we use calorimetric and spectroscopic observables to detect, resolve, and selectively enrich energetically discrete ensembles of microstates within metastable DNA structures. Our results are consistent with metastable, “native” DNA states being composed of an ensemble of discrete and kinetically stable microstates of differential stabilities, rather than exclusively being a single, discrete thermodynamic species. This conceptual construct is important for understanding the linkage between biopolymer conformational/configurational space and biological function, such as in protein folding, allosteric control of enzyme activity, RNA and DNA folding and function, DNA structure and biological regulation, etc. For the specific DNA sequences and structures studied here, the demonstration of discrete, kinetically stable microstates potentially has biological consequences for understanding the development and onset of DNA expansion and triplet repeat diseases.
The concept of rough biopolymer energy landscapes is important for understanding biopolymer properties and function. Complex landscapes allow partial population of discrete, kinetically stable (metastable) species (1–12). These substates, in turn, may serve as biologically significant ligands, beyond the traditional, thermodynamically most stable, time-averaged, “native” macrostrate. Although it is well recognized that macroscopic states are composed of microscopic substates, it has been customary for experimentalists to think of a biologically relevant native state as a singular macrostate with a unique, well defined structure characterized by a homogeneous/smooth energy profile. Many biophysical measurements probe properties that are averaged over the ensemble of microstates that collectively correspond to a macroscopic native state. However, experimental evidence for the presence of, and dynamic interchange between, multiple states that constitute a global macrostate is provided by direct time-resolved measurements, and by structural dispersions in crystallographic and NMR studies (13–22). Such results underscore the potential importance as ligands or receptors in biological mechanisms of kinetically stable, trapped microstates embedded within the rough energy landscapes of purportedly singular macrostates. For example, the detection of transiently populated structural states in the native unbound protein ensemble that resemble those of the proteins in their bound state provides evidence for the so-called “conformational selection model” of protein binding, thereby suggesting significant biological functions for microstates in proteins (23–26).
By contrast with proteins, relatively few studies have probed the energy landscapes of nucleic acids, particularly via calorimetric techniques (27–35). As a result, the potential energetic/biological role(s), if any, of kinetically stable (metastable) microstates that make up the time-averaged, native state ensembles of macroscopic nucleic acid states remains to be determined.
As part of an effort to address this deficiency, we report here experimental evidence for the presence of discrete microstates in metastable triplet repeat bulge looped Ω-DNAs of potential biological significance. The specific triplet repeat bulge looped Ω-DNA species investigated mimic slipped DNA structures corresponding to intermediates in the processes that lead to DNA expansion in triplet repeat diseases (36). Incorrect processing by the cellular replication and repair machinery of such transiently formed slipped structures, in their macro and/or microstate ensembles, is believed to cause DNA expansion, the genotype that ultimately leads to the phenotype associated with DNA expansion diseases (37–50). The presence of discrete and kinetically stable microstates within the ensemble of the triplet repeat macrostate is therefore of potential significance for understanding the biology of triplet repeat diseases.
Results and Discussion
We have shown that the 2 complementary (CAG)6 and (CTG)6 triplet repeat bulge looped Ω-DNAs diagrammed in Fig. 1A do not spontaneously rearrange to their thermodynamically more stable duplex states when combined in solution, but rather remain trapped in their metastable looped structures (36).
Fig. 1.
Interconversions of DNA states. (A) Schematic representation of the different DNA conformations and the transformations between them. (B) Calorimetric excess heat capacity curve for the first DSC heating scan of a 1:1 mixture of 2 complementary, triplet-repeat bulge looped Ω-DNAs. The dotted vertical lines delimit the thermal domains in which particular DNA conformations are selectively populated and the thermal regions in which conformational transformations are observed. The relevant DNA states are shown in diagram form. Note the unusual exothermic peak ≈50 °C, which is reflective of strand exchange between the 2 Ω-DNAs to form the 2 fully paired duplex states.
Strand exchange to the duplex state requires input of energy, occurring only at elevated temperatures. At the so-called exchange temperature, Tex, the strand exchange reaction proceeds irreversibly, with a release of heat energy. We detect this strand exchange between the 2 metastable bulge looped Ω-DNAs to form the 2 fully paired duplex states by monitoring the unusual exothermic peak in the DSC curves that, as expected for an irreversible process, is absent in the second heating scan (Fig. 1B).
These observations are consistent with the energy landscape of these looped, repeat sequences containing local minima (here the Ω-DNA structures), in addition to the global Watson and Crick duplex minima. Such metastable states may well present processing problems for the machinery of replication and repair. Further heating of the system results in conventional duplex melting, fully consistent with expectation.
The landscape for these transformations is depicted in the energy funnel shown in Fig. 2. The arrow in this figure tracks the transformations noted above; namely, from the ensemble of high energy single strands (corresponding to the upper outer rim of the funnel), to the initial valley, which corresponds to the ensemble of metastable Ω-DNA structures (kinetically protected by a significant energy barrier), to the deep global minima associated with the ensemble of thermodynamically most stable duplex states.
Fig. 2.
Idealized representation of an energy landscape, showing features related to the triplet repeat DNA conformations investigated here (4). For clarity of presentation and ease of interpretation, the energy wells and the idealized trajectories between them are shown as smooth surfaces. The experimental data reported here, however, reveal that these surfaces are not smooth. The specific trajectory by which the DNA strands combine/fold and the shape of the folding landscape will depend on the temperature. At temperatures above exchange (landscape not shown), the barrier between the local Ω-DNA minimum, and the global duplex minimum would be diminished such that the Ω-DNA state no longer would be populated. [Reproduced with permission from ref. 4 (Copyright 1998, Wiley, New York).]
To further map the energy landscape of the metastable DNA molecules, we conducted a series of thermal incubations below the exchange temperature. Through such measurements we have been able to resolve populations of microstates of differential energetics that comprise the metastable Ω-DNA macrostate, consistent with a rough energy landscape (micro hills and valleys) within the upper “macro valley” representing the metastable Ω-DNA macrostate in Fig. 2. For each measurement needed to detect and energetically resolve this microstate fine structure, we incubated for an hour an equimolar mixture of the complementary CAG and CTG metastable Ω-DNAs at a series of temperatures below the exchange temperature, followed by cooling the sample down and rescanning. Because strand exchange between the 2 Ω-DNAs is irreversible, any exotherm detected in the postincubation second scan reflects the unreacted/unexchanged fractions of the original 2 Ω-DNAs. The results obtained here for different incubation temperatures are shown in Figs. 3A and 3B and are described below.
Fig. 3.
Incubation induced redistribution of DNA states. (A) Superposition of experimental excess heat capacity curves for a 1:1 mixture of the complementary, triplet repeat bulge looped Ω-DNAs after 1 h of preincubation at different temperatures. Incubation temperatures (Tin) are indicated in the figure. Note the differences in excess heat capacity curves in the temperature range where we observe the exothermic peak near 50 °C. Similar differences in the excess heat capacity curves are not observed at other temperatures. (B) Expanded view of the exothermic peak shown in Fig. 3A that highlights the incubation induced shift to higher temperatures of the exchange peak. The incubation temperatures (Tin) and the measured exchange temperature (Texin) are indicated in the figure.
We find that when the sample is incubated at temperatures well below the onset of the exchange reaction, the Ω-DNAs do not undergo strand exchange during the incubation period, as evidenced by the presence of the full exothermic peak in the DSC rescan curve (the black curve in Fig. 3A). Conversely, when the sample is incubated at temperatures above the exchange temperature, all of the Ω-DNA molecules undergo strand exchange, as evidenced by no signal being observed for residual, unreacted Ω-DNAs in the DSC rescan (the magenta curve in Fig. 3A). However, when the samples are incubated at a temperature close to the onset of the exchange exotherm, some residual fraction of the original exchange exotherm is detected upon rescan after cooling; evidence for some, but not all, of the Ω-DNA metastable state undergoing conversion to the more stable duplex state during incubation (the red, green, blue, and cyan curves in Fig. 3A) The fraction of the remaining Ω-DNA detected by the DSC exotherm scales with the incubation temperature, consistent with slow strand exchange kinetics, even close to the exchange temperature.
Most intriguing is that we observe shifts to higher exchange temperatures in the rescan of the incubated samples relative to the exchange temperatures of the exotherm in the initial scan of samples that were not preincubated (Fig. 3B). We propose that this increase in the exchange temperature for the residual/unexchanged Ω-DNA reflects the selective removal via strand exchange during preincubation of the low stability/high energy subfraction of the Ω-DNAs, thereby leaving the high stability/low energy substates unchanged. Our data are consistent with these low stability/high energy, unexchanged Ω- substates not being repopulated during the time course of our experiments. The lack of redistribution between the different microstates and the slow exchange kinetics at temperatures close to the exchange temperature may reflect energetically costly unfolding of a part of the loop domain before refolding to a new configuration or before strand exchange is possible (51, 52).
Collectively, our results suggest that preincubation leads to a selective enrichment of the high stability/low energy subpopulations of the microstate ensembles that constitute the macroscopic, metastable Ω-DNA states. We have made similar observations, using UV melting experiments (data not shown). In this case, our data show that a triplet repeat Ω-DNA metastable macrostate is populated by energetically discrete microstates separated by significant energy barriers. This conclusion means that the apparently smooth upper valley depicted in Fig. 2 for the metastable Ω-DNA states actually is quite rough, being itself composed of micro hills and valleys that constitute the ensemble of states corresponding to the 2 metastable Ω-DNA macrostates as indicated in Fig. 4.
For the CAG and CTG triple repeat bulge loop DNAs studied here, the presence of discrete microstates with significant energy barriers between them may contribute to the propensity of triplet DNA sequences to undergo DNA expansion. It has been suggested that DNA expansion is a consequence of challenges to the DNA replication and repair machinery, which has been optimized by evolution for dealing with duplex DNA, and therefore has difficulty processing bulge loop substrates (37–50, 53, 54). Our results suggest that in addition to an unusual metastable macrostate, the machinery of replication and repair also may need to recognize and process an ensemble of complex energetic and perhaps structural substates. Incorrect recognition, binding, and/or processing of some or all of these substates may challenge the DNA replication and repair machinery in a manner that facilitates expansion of the triplet repeat sequences. More generally, our results reveal the existence of a rough energy landscape for a metastable macrostate that is composed of an ensemble of energetically discrete microstates. This conclusion is also supported by strategically placed lesions within the Ω-DNA structure that modulate the distribution of microstates, while retaining the metastable macrostate (unpublished data).
In summary, we have shown via calorimetric measurements that the energy landscape of nucleic acids can be rough, and that some nucleic acid states, like those of proteins, can be composed of discrete microstates. It can be envisioned that discrete substates within the native state ensemble play important roles in the biology of other non-B-DNA secondary structures, such as G-quadruplexes (55–57), iDNAs (58–61), cruciforms (62–64), and triple helical H-DNA like structures (65–70), inter alia (41, 67, 71–76). Given the diverse range of secondary structural elements and highly folded 3 dimensional structures common to many biologically important macromolecules, the realization of rough energy landscapes and discrete microstates for nucleic acids should prove important for understanding the broad range of mechanisms that modulate nucleic acid biology.
Materials and Methods
Materials.
Oligonucleotides were synthesized on a 10 μmole scale by standard phosphoramidite chemistry, using an Äkta DNA synthesizer, and were purified by repeated DMT on/DMT off reverse phase HPLC, as described in refs. 52 and 61. The purities of the oligonucleotides were assessed by analytical HPLC and ion spray mass spectroscopy, and were found to be better than 98% by mass spectroscopy. Purified oligonucleotides were dialyzed against at least 2 changes of buffer containing NaCl, 10 mM Cacodylic acid/Na-Cacodylate, and 0.1 mM Na2 EDTA to yield a final concentration of 100 mM in Na+ cations, using dispo-dialyzers with MWCO 500 da (Spectrum). DNA extinction coefficients were determined by phosphate assay (77, 78) under denaturing conditions (90 °C) and were found to be: εX(CAG)6Y (260 nm, 90 °C) = 368,400 M−1·cm−1; εY′ (CTG)6X′ (260 nm, 90 °C) = 342,900 M−1·cm−1; εXY (260 nm, 90 °C) = 190,400 M−1·cm−1; εY′X′ (260 nm, 90 °C) = 186,200 M−1·cm−1.
DSC Studies.
DSC studies were conducted using a NanoDSCII differential scanning calorimeter (Calorimetry Science) with a nominal cell volume of 0.3 mL (79). Just before filling the DSC cell, equimolar concentrations of preformed complementary Ω-DNA oligonucleotides were combined to yield a final concentration 50 μM in strand. The sample was scanned once from 0 °C to the relevant incubation temperature at a rate of 1 °C/min, and then incubated at the incubation temperature for 1 h, followed by cooling back to 0 °C. The power needed to maintain sample and reference cells at constant temperature during incubation exhibited a complex multiexponential time dependence with a non-Arrhenius dependence on incubation temperature. After incubation, the sample was scanned at least twice from 0 °C to 100 °C at a constant heating rate of 1 °C/min, while recording the excess power required to keep sample cell and reference cells at the same temperature. After conversion to heat capacity units and subtractions of buffer/buffer scans, the raw DSC traces were normalized for DNA concentration and analyzed using Origin software.
UV Melting Studies.
UV absorption versus temperature studies were conducted using a Aviv DS14 UV/VIS spectrophotometer (Aviv Biomedical). Samples in 1-cm cells were heated from 0 °C to the incubation temperature in steps of 0.5 °C, and allowed to incubate for 1 h. Samples where than cooled back to 0 °C and heated in a second heating scan from 0 °C to 95 °C in steps of 0.5 °C. Upon reaching each set temperature, samples were equilibrated for 1 min before recording the absorption at 260 nm, with a 5-sec averaging time, resulting in a nominal heating rate of 0.1 °C/min. Oligonucleotide strand concentration was 1.5 μM.
UV Mixing Studies.
UV absorption versus time studies were conducted using a Varian 4000 UV/Vis spectrophotometer (Varian), equipped with a peltier controlled multicell cell holder. Complementary Ω-DNA samples were placed in separate compartments of a partitioned UV cuvette (path length 1 cm) and allowed to equilibrate at the relevant incubation temperature for a minimum of 10 min. After temperature equilibration, samples were rapidly mixed and the change in absorbance with time was recorded every second for 5 h. The UV absorption versus time profiles showed a complex multiexponential decay with non-Arrhenius temperature dependence, and an overall hypochromic change that depended on incubation temperature.
Acknowledgments
We thank Drs. Roger Jones, Barbara Gaffney (Rutgers University), and G. Eric Plum (Rutgers University and IBET, Inc.) for advice and many helpful discussions. This work was supported by National Institutes of Health Grants GM23509, GM34469, and CA47995 (to K.J.B.)
Footnotes
- 1To whom correspondence should be addressed. E-mail: kjbdna{at}rci.rutgers.edu
-
Author contributions: J.V., H.H.K., and K.J.B. designed research; J.V. performed research; J.V., H.H.K., and K.J.B. analyzed data; and J.V. and K.J.B. wrote the paper.
-
The authors declare no conflict of interest.
-
Freely available online through the PNAS open access option.
- © 2008 by The National Academy of Sciences of the USA
References
DNA energy landscapes via calorimetric detection of microstate ensembles of metastable macrostates and triplet repeat diseases
Sign up for Article Alerts
Jump to section
You May Also be Interested in
Better explicating the strengths and shortcomings of these models will help refine projections and improve transparency in the years ahead.
Image credit: Witsawat.S.
Using CRISPR, researchers showed that a region some used to label “junk DNA” has a major role in a rare genetic disorder.
Image credit: Nathan Devery.
Mara Reed and Michael Manga explore why Yellowstone's Steamboat Geyser resumed erupting in 2018.
A study demonstrates how two enzymes—MHETase and PETase—work synergistically to depolymerize the plastic pollutant PET.
Image credit: Aaron McGeehan (artist).