DNA is a double helix: two long strands wound around each other in a spiral, joined by pairs of chemical bases that function like rungs on a twisted ladder. That single sentence captures the iconic image most people carry, and it is correct as far as it goes. But the full structural picture is richer and stranger than the tidy model suggests. DNA bends, stretches, and shifts between different shapes depending on its surroundings. What holds it together is not quite what most textbooks emphasize. And the physical shape of the molecule, down to the width of its grooves, turns out to be just as important for biology as the genetic code written in its bases.
The Double Helix Up Close
Each strand of DNA is built from a repeating backbone of sugar and phosphate groups, with one of four chemical bases attached to each sugar. The four bases are adenine (A), thymine (T), guanine (G), and cytosine (C). The two strands run in opposite directions, a feature confirmed early on through X-ray diffraction of crystalline DNA fibers.1Oxford Academic (Nucleic Acids Research). Backbone-base inclination as a fundamental determinant of nucleic acid self- and cross-pairing This antiparallel arrangement matters because it allows the bases on one strand to pair neatly with the bases on the other: A always pairs with T, and G always pairs with C. These are called Watson-Crick base pairs, and they are held in place by hydrogen bonds between the paired bases.2Journal of Computational Chemistry. Characteristics of the Watson‐Crick type hydrogen‐bonded DNA base pairs: An ab initio molecular orbital study
The two strands coil around each other with a repeating twist, completing one full turn roughly every ten base pairs. This winding creates two grooves that spiral along the outside of the helix: a wider one called the major groove and a narrower one called the minor groove. Both grooves expose parts of the bases, but in different ways, and this turns out to be critical for how the cell’s machinery reads and interacts with DNA. The classic structure was famously confirmed through X-ray fiber diffraction, including the celebrated “Photo 51” captured by Rosalind Franklin and Raymond Gosling, which revealed the helical pattern at a resolution of about 3.4 angstroms, the spacing between stacked base pairs.3PubMed. Three-dimensional double helical DNA structure directly revealed from its X-ray fiber diffraction pattern by iterative phase retrieval
What Actually Holds the Helix Together
Most people learn that the double helix is held together by hydrogen bonds between the base pairs, and that G-C pairs (with three hydrogen bonds each) are stronger than A-T pairs (with two). That is true as far as it goes, but it misses the bigger contributor to the molecule’s stability. The dominant force keeping the double helix intact is not the hydrogen bonding between paired bases, but the stacking interactions between bases sitting on top of one another along each strand.
These stacking interactions arise because the flat, ring-shaped bases pile up like coins in a roll, and the electron clouds of neighboring rings interact in ways that release a small amount of energy. Each individual stacking contact is weak, but they add up across the entire length of the molecule into a substantial stabilizing force.4PubMed. Extended weak bonding interactions in DNA: pi-stacking (base-base), base-backbone, and backbone-backbone interactions Computational studies have confirmed that stacking is the main factor stabilizing the double helix across a range of temperatures and salt concentrations. In fact, A-T base pairing on its own is slightly destabilizing, and G-C pairing contributes almost no net stabilization. The sequence dependence of DNA stability, meaning why some stretches melt apart more easily than others, is driven primarily by how well their bases stack, not by how many hydrogen bonds they form.5PubMed Central. Base-stacking and base-pairing contributions into thermal stability of the DNA double helix
This might sound like the hydrogen bonds do not matter, but they do, just in a different way. Hydrogen bonds enforce the specificity of pairing: A pairs with T and not with G because only the correct partners form the right hydrogen-bond geometry. Stacking provides the energy that keeps the helix wound together; hydrogen bonding provides the logic that keeps the code readable.
DNA Comes in More Than One Shape
The double helix shown in textbooks is specifically the B-form of DNA, the shape the molecule adopts under normal conditions inside a cell. B-DNA is a right-handed helix with about ten base pairs per turn, the bases roughly perpendicular to the helix axis, and a relatively wide major groove. But B-DNA is not the only conformation the molecule can take.
Under low-water conditions, or when certain proteins or drugs bind to it, DNA can shift into the A-form. A-DNA is also right-handed, but it is wider, more compact along its length, and the bases tilt significantly relative to the axis. The transition between B and A forms is reversible, cooperative (meaning a stretch of DNA tends to flip all at once rather than base by base), and depends on the local sequence.6PubMed. Understanding B-DNA to A-DNA transition in the right-handed DNA helix: Perspective from a local to global transition The A-form is not just a laboratory curiosity. Proteins sometimes locally induce the A-form when they bind DNA, and repeated B-to-A transitions have been proposed as a mechanism that helps drive DNA into viral capsids during genome packaging.7PubMed Central. Transitions of Double-Stranded DNA Between the A- and B-Forms
Then there is Z-DNA, which twists in the opposite direction, forming a left-handed helix. Z-DNA tends to form in sequences that alternate between purines and pyrimidines, especially under conditions of high salt or certain chemical modifications. Its biological role is still debated, but it appears to be involved in gene regulation and immune signaling in some contexts.
Beyond the Double Helix
DNA can also fold into shapes that bear little resemblance to the classic double helix. These non-canonical structures include hairpins, where a single strand folds back on itself; cruciforms, where inverted repeat sequences splay out into a cross-shaped arrangement; and multi-stranded structures that involve three or even four strands of DNA coming together.8PubMed Central. Non-canonical DNA structures: Diversity and disease association
The most studied of these are G-quadruplexes and i-motifs. G-quadruplexes form in guanine-rich sequences, where four guanines arrange themselves into a flat quartet stabilized by a special type of hydrogen bonding, and multiple quartets stack on top of each other into a columnar structure. I-motifs form in cytosine-rich sequences under slightly acidic conditions. Both structures have attracted intense interest because they appear at biologically important locations in the genome, including near the start sites of genes and at chromosome ends. Recent work suggests that the stability and behavior of G-quadruplexes and i-motifs inside living cells are tuned by the physical organization of the cell’s interior, fitting them to the seconds-to-minutes timescales of key regulatory processes.9PubMed Central. Phase separation tunes the stability and dynamics of G-quadruplex and i-motif DNA in the nuclei of living cells These non-canonical structures have also been linked to genetic instability, DNA damage, and repair pathways, making them relevant to understanding diseases like cancer.
How Proteins Read DNA’s Shape
When people think about how proteins find specific genes, they usually picture a protein scanning along the DNA and reading the sequence of bases like letters. That does happen, primarily through hydrogen bonds formed in the major groove, where the edges of the base pairs are most exposed. But a large part of protein-DNA recognition depends on something subtler: the physical shape of the DNA itself.
Different sequences of bases produce slightly different local geometries. Some sequences naturally compress the minor groove, while others widen it. Some create gentle bends. These shape variations change the electrostatic properties of the DNA surface, and proteins can detect those differences. A comprehensive analysis of protein-DNA complex structures showed that proteins commonly use arginine residues to bind narrow minor grooves, exploiting the enhanced negative electrostatic potential that narrow grooves create.10PubMed Central. The role of DNA shape in protein-DNA recognition The minor groove on its own offers few chemical handles for distinguishing one base pair from another, so this shape-based readout gives proteins a way to extract sequence information from a groove that would otherwise be nearly featureless to them.
One well-studied example is the protein Fis, which bends DNA sharply when it binds. In the Fis-DNA complex, the minor groove at the center of the binding site compresses to almost half its normal width, allowing the protein’s recognition elements to insert into adjacent major grooves. The minor groove then widens dramatically on the flanking sides before compressing again, creating an oscillating pattern.11Genes & Development. The shape of the DNA minor groove directs binding by the DNA-bending protein Fis The protein is effectively reading the contour of the DNA rather than just its chemical letters.
Packing Two Meters of DNA into a Cell
The total length of DNA in a single human cell, if you unraveled every chromosome and laid it end to end, stretches roughly two meters. Fitting that into a nucleus only a few millionths of a meter wide requires extraordinary packaging. The first level of compaction involves wrapping DNA around protein spools called nucleosomes. Each nucleosome core consists of eight histone proteins (two copies each of four types: H2A, H2B, H3, and H4), and about 146 base pairs of DNA wrap around this core roughly 1.65 times in a left-handed superhelix.12Chemical Reviews. Nucleosome Structure and Function
Nucleosomes are connected by short stretches of “linker” DNA, producing a structure often compared to beads on a string. This fiber then coils and folds into progressively higher-order structures, though the precise geometry of the intermediate stages is still debated. One feature of this compaction is DNA supercoiling: the helix itself becomes coiled in space, like a phone cord that twists and kinks. Supercoiling is not just a passive consequence of packing. It stores mechanical energy that can be released to drive biological processes, and it helps newly replicated DNA molecules separate from each other after copying.13PubMed Central. DNA supercoiling helps to unlink sister duplexes after replication Enzymes called topoisomerases manage supercoiling by cutting the DNA, passing strands through the break, and resealing it.
How Chemical Tags Reshape the Helix
The structure of DNA is not fixed by sequence alone. Chemical modifications can be added to the bases without changing the genetic code, and these modifications alter the molecule’s physical shape. The most common modification in mammalian DNA is the addition of a methyl group to cytosine, typically at sites where a C is followed by a G (called CpG sites). This methylation is a central mechanism of gene regulation: heavily methylated regions tend to be silenced, while unmethylated regions are usually active.
From a structural standpoint, the methyl group protrudes into the major groove, introducing steric crowding that subtly changes the local geometry. Crystal structures of methylated DNA have shown that the methyl group can resist DNA bending and twisting, subtly widen the major groove, and narrow the minor groove.14Briefings in Functional Genomics. Evolving insights on how cytosine methylation affects protein–DNA binding Computational analyses have gone further, showing that CpG methylation significantly alters local DNA shape across multiple parameters. Roll and propeller twist, two measures of how the base pairs tilt and rotate relative to each other, are the features most strongly affected, and the size of the effect depends on the surrounding sequence.15PubMed Central. Systematic prediction of DNA shape changes due to CpG methylation explains epigenetic effects on protein-DNA binding
Beyond methylation, hydroxymethylation of cytosine (adding a hydroxyl-bearing methyl group) has its own structural fingerprint. Both modifications change the molecule’s dynamic landscape, pushing DNA toward different twist and tilt angles at the modified sites.16PubMed Central. Understanding the structural and dynamic consequences of DNA epigenetic modifications: Computational insights into cytosine methylation and hydroxymethylation These shape changes are not cosmetic. They provide specificity for protein binding, meaning that a protein looking for a methylated binding site is partly detecting the altered contour of the DNA, not just the chemical identity of the methyl group.
Melting, Stretching, and Breaking the Helix
Heat a DNA solution slowly and the two strands will eventually separate, a process called denaturation or “melting.” This does not happen all at once. Regions rich in A-T pairs melt first, because they are held by fewer hydrogen bonds and, more importantly, by weaker stacking interactions. As the temperature rises, the proportion of A and T in the melted regions decreases as the more resistant G-C-rich stretches start to open up.17PubMed Central. Determination of melting sequences in DNA and DNA-protein complexes by difference spectra Researchers track melting by measuring how much ultraviolet light the solution absorbs, because single-stranded DNA absorbs more UV than double-stranded DNA. This increase in absorbance, called the hyperchromic effect, arises because the excited electronic states of the bases become more spread out when the bases are no longer stacked and constrained in the double helix.18PubMed. On the nature of DNA hyperchromic effect
You can also pull on DNA mechanically. Under increasing tension, the B-form double helix undergoes a cooperative elongation into a mixture of a stretched form called S-DNA and separated single strands.19Nucleic Acids Research. Force-induced melting and S-DNA pathways for DNA overstretching exhibit distinct kinetics These stretching experiments are done with single-molecule techniques, where a single DNA molecule is tethered between two tiny beads and pulled with controlled force. They have revealed that short DNA fragments are surprisingly flexible, bending more easily than older models predicted for fragments in the range of 15 to 90 base pairs.20PubMed. DNA bending stiffness on small length scales This flexibility matters biologically because DNA must make tight bends to wrap around nucleosomes and to loop when distant regulatory elements need to contact each other.
Structural damage is another way the helix gets disrupted. Ultraviolet light, for instance, causes adjacent pyrimidine bases (C and T) on the same strand to bond together abnormally. The two most common UV lesions are cyclobutane dimers and pyrimidine-pyrimidone (6-4) photoproducts, with the thymine-cytosine (6-4) product being the most abundant of the latter type.21Oxford Academic (Nucleic Acids Research). Structural determination of the ultraviolet light-induced thymine-cytosine pyrimidine-pyrimidone (6–4) photoproduct These lesions distort the helix locally, and cells deploy specialized repair systems that recognize the structural distortion as a signal that something has gone wrong.
DNA Outside the Nucleus
Not all DNA lives in chromosomes. Mitochondria, the energy-producing compartments inside cells, carry their own small genomes. In humans, the mitochondrial genome is a circular DNA molecule roughly 16,000 base pairs long, packed tightly with 13 protein-coding genes alongside the RNA genes needed to express them. Unlike nuclear DNA, mitochondrial DNA is not wrapped around histones. Instead, copies of the mitochondrial genome are clustered into small protein-DNA packages called nucleoids, each containing one or a few genome copies, scattered throughout the mitochondrial network.22PubMed. Mitochondrial DNA nucleoid structure A single human cell can contain hundreds to thousands of copies of this genome. The compaction strategy is entirely different from the histone-based system in the nucleus, relying instead on a smaller set of DNA-binding proteins and the tight curvature of the circular molecule itself.
DNA as a Building Material
The precise and predictable base-pairing rules of DNA have made it an attractive material for nanotechnology. In a technique called DNA origami, a long single strand of DNA is folded into a desired shape using hundreds of short “staple” strands that bind to specific stretches and hold the structure in place. Because A pairs with T and G pairs with C with reliable geometry, designers can program almost any two-dimensional or three-dimensional shape at nanometer-scale precision.23PubMed Central. Self-assembly of DNA origami for nanofabrication, biosensing, drug delivery, and computational storage DNA nanostructures are being explored for applications in drug delivery, biosensing, and even data storage. In these applications, the double helix is not carrying genetic information at all. It is functioning as a programmable scaffold, valued for the same structural regularity that makes it useful to biology.
Seeing DNA Structure With Modern Tools
The understanding of DNA structure has evolved alongside imaging technology. X-ray crystallography provided the original double-helix model and remained the gold standard for decades. More recently, cryo-electron microscopy (cryo-EM) has become a powerful complement. Cryo-EM flash-freezes molecules in solution and images them with an electron beam, avoiding the need to grow crystals. This has allowed researchers to capture DNA in the context of large, dynamic protein complexes that resist crystallization. Structures of enzyme-DNA assemblies have been resolved at near-atomic resolution using cryo-EM, reaching down to about 3 angstroms in some regions.24PubMed Central. Visualizing functional dynamicity in the DNA-dependent protein kinase holoenzyme DNA-PK complex by integrating SAXS with cryo-EM Combining cryo-EM with other techniques like small-angle X-ray scattering has enabled visualization of how DNA-bound protein machines flex and move while doing their work, providing a view of DNA structure not as a frozen snapshot but as a dynamic participant in cellular processes.
Synthetic Alternatives Highlight What Makes Natural DNA Work
One of the sharpest ways to understand why DNA has its particular structure is to look at what happens when you change it. Researchers have synthesized nucleic acid variants with altered sugar backbones to test how structural details affect function. Xylonucleic acid (XyloNA), for example, uses a different sugar configuration than natural DNA. The result is a molecule that still pairs its bases, but adopts a radically different shape: an extended, ladder-like structure with barely any helical twist, a high rise between base pairs, and grooves so wide they are almost indistinguishable from each other.25Nucleic Acids Research. Xylonucleic acid: synthesis, structure, and orthogonal pairing properties Instead of the intrastrand stacking that stabilizes natural DNA, XyloNA relies predominantly on stacking between bases on opposite strands, a zipper-like arrangement that actually provides higher stability in some contexts. XyloNA pairs orthogonally, meaning it does not cross-pair with natural DNA or RNA, making it potentially useful as a bioorthogonal tool that can operate inside cells without interfering with natural nucleic acids. Studying molecules like XyloNA illuminates just how much the specific sugar chemistry and backbone geometry of natural DNA contribute to its helical shape, its groove architecture, and ultimately its ability to interact with the protein machinery of life.