Charge transfer through DNA/DNA duplexes and DNA/RNA hybrids: Complex theoretical and experimental studies

A quantum physics layer of epigenetics: a hypothesis deduced from charge transfer and chirality-induced spin selectivity of DNA

BACKGROUND

Epigenetic mechanisms are informational cellular processes instructing normal and diseased phenotypes. They are associated with DNA but without altering the DNA sequence. Whereas chemical processes like DNA methylation or histone modifications are well-accepted epigenetic mechanisms, we herein propose the existence of an additional quantum physics layer of epigenetics.

RESULTS

We base our hypothesis on theoretical and experimental studies showing quantum phenomena to be active in double-stranded DNA, even under ambient conditions. These phenomena include coherent charge transfer along overlapping pi-orbitals of DNA bases and chirality-induced spin selectivity. Charge transfer via quantum tunneling mediated by overlapping orbitals results in charge delocalization along several neighboring bases, which can even be extended by classical (non-quantum) electron hopping. Such charge transfer is interrupted by flipping base(s) out of the double-strand e.g., by DNA modifying enzymes. Charge delocalization can directly alter DNA recognition by proteins or indirectly by DNA structural changes e.g., kinking. Regarding sequence dependency, charge localization, shown to favor guanines, could influence or even direct epigenetic changes, e.g., modification of cytosines in CpG dinucleotides. Chirality-induced spin selectivity filters electrons for their spin along DNA and, thus, is not only an indicator for quantum coherence but can potentially affect DNA binding properties.

CONCLUSIONS

Quantum effects in DNA are prone to triggering and manipulation by external means. By the hypothesis put forward here, we would like to foster research on "Quantum Epigenetics" at the interface of medicine, biology, biochemistry, and physics to investigate the potential epigenetic impact of quantum physical principles on (human) life.

Physics-inspired machine learning of localized intensive properties

Machine learning (ML) has been widely applied to chemical property prediction, most prominently for the energies and forces in molecules and materials. The strong interest in predicting energies in particular has led to a 'local energy'-based paradigm for modern atomistic ML models, which ensures size-extensivity and a linear scaling of computational cost with system size. However, many electronic properties (such as excitation energies or ionization energies) do not necessarily scale linearly with system size and may even be spatially localized. Using size-extensive models in these cases can lead to large errors. In this work, we explore different strategies for learning intensive and localized properties, using HOMO energies in organic molecules as a representative test case. In particular, we analyze the pooling functions that atomistic neural networks use to predict molecular properties, and suggest an orbital weighted average (OWA) approach that enables the accurate prediction of orbital energies and locations.

Hole Transport in A-form DNA/RNA Hybrid Duplexes

DNA/RNA hybrid duplexes are prevalent in many cellular functions and are an attractive target form for electrochemical biosensing and electric nanodevice. However the electronic conductivities of DNA/RNA hybrid duplex remain relatively unexplored and limited further technological applications. Here cyclopropyl-modified deoxyribose- and ribose-adenosines were developed to explore hole transport (HT) in both DNA duplex and DNA/RNA hybrids by probing the transient hole occupancies on adenine tracts. HT yields through both B-form and A-form double helixes displayed similar shallow distance dependence, although the HT yields of DNA/RNA hybrid duplexes were lower than those of DNA duplexes. The lack of oscillatory periods and direction dependence in HT through both helixes implied efficient hole propagation can be achieved via the hole delocalization and coherent HT over adenine tracts, regardless of the structural variations.

Comparing Charge Transport in Oligonucleotides: RNA:DNA Hybrids and DNA Duplexes

Understanding the electronic properties of oligonucleotide systems is important for applications in nanotechnology, biology, and sensing systems. Here the charge-transport properties of guanine-rich RNA:DNA hybrids are compared to double-stranded DNA (dsDNA) duplexes with identical sequences. The conductance of the RNA:DNA hybrids is ∼10 times higher than the equivalent dsDNA, and conformational differences are determined to be the primary reason for this difference. The conductance of the RNA:DNA hybrids is also found to decrease more rapidly than dsDNA when the length is increased. Ab initio electronic structure and Green's function-based density of states calculations demonstrate that these differences arise because the energy levels are more spatially distributed in the RNA:DNA hybrid but that the number of accessible hopping sites is smaller. These combination results indicate that a simple hopping model that treats each individual guanine as a hopping site is insufficient to explain both a higher conductance and β value for RNA:DNA hybrids, and larger delocalization lengths must be considered.

Long-Range Charge Transport in Adenine-Stacked RNA:DNA Hybrids

An extremely important biological component, RNA:DNA can also be used to design nanoscale structures such as molecular wires. The conductance of single adenine-stacked RNA:DNA hybrids is rapidly and reproducibly measured using the break junction approach. The conductance decreases slightly over a large range of molecular lengths, suggesting that RNA:DNA can be used as an oligonucleotide wire.