In molecular biology, the double helix is the structure created by two strands of nucleic acid molecules like DNA. This double-helix shape forms part of the molecule's secondary structure and helps determine its tertiary structure.

The DNA double-helix is made of nucleotides that pair with each other. In B-DNA, the most common type of double-helix found in nature, the structure twists to the right and has about 10–10.5 base pairs in each full twist. The DNA double-helix has two grooves: a major groove and a minor groove. In B-DNA, the major groove is larger than the minor groove. Because of this size difference, many proteins that attach to B-DNA do so by binding to the larger major groove.

The double-helix structure of DNA was first discovered by James Watson and Francis Crick, using research by Rosalind Franklin, Raymond Gosling, Maurice Wilkins, and others. The term "double helix" became widely known after Watson's book, The Double Helix: A Personal Account of the Discovery of the Structure of DNA, was published in 1968.

History

The double-helix model of DNA structure was first published in the journal Nature by James Watson and Francis Crick in 1953. This work was based on the research of Rosalind Franklin and her student Raymond Gosling, who took an important X-ray diffraction image of DNA called "Photo 51." It also relied on the work of Maurice Wilkins, Alexander Stokes, and Herbert Wilson, as well as chemical information from Erwin Chargaff. Before this discovery, Linus Pauling and his collaborator Robert Corey incorrectly suggested that DNA had a triple-stranded structure.

Understanding that DNA has a double-helix shape explained how genetic information is stored and copied in living things. This discovery is widely regarded as one of the most important scientific achievements of the 20th century. James Watson, Francis Crick, and Maurice Wilkins each received one-third of the 1962 Nobel Prize in Physiology or Medicine for their roles in this discovery.

Nucleic acid hybridization

Hybridization is the process where base pairs join together to create a double helix structure. Melting is the process where the connections between the two strands of the double helix are broken, causing the strands to separate. These connections are weak and can be easily broken using gentle heating, enzymes, or physical force. Melting happens more easily in certain areas of the nucleic acid. Sections with more T and A bases are easier to melt than sections with more C and G bases. Some specific base pairs, such as T-A and T-G, are also more likely to melt. These features are used in DNA, such as the TATA sequence near the start of many genes, to help RNA polymerase separate the DNA strands during transcription.

Separating DNA strands using gentle heating, as in the polymerase chain reaction (PCR), works well for DNA molecules with fewer than about 10,000 base pairs (10 kbp). However, the twisting of DNA strands makes it harder to separate long segments. Cells solve this problem by using DNA-melting enzymes called helicases together with topoisomerases. Topoisomerases can cut one strand of DNA to allow it to rotate around the other strand. Helicases then unwind the strands to help enzymes like DNA polymerase read the DNA sequence.

Base-pair geometry

The shape of a base pair step in a nucleic acid molecule can be described using six measurements: shift, slide, rise, tilt, roll, and twist. These measurements show the exact position and direction of each base pair in the molecule compared to the one before it along the helix. Together, these measurements explain the helix shape of the molecule. When the normal structure of DNA or RNA is changed, the differences in these measurements can describe the change.

For each base pair, compared to the one before it, the following base pair geometries are important:

• Shear
• Stretch
• Stagger
• Buckle
• Propeller: one base spins around the other in the same pair.
• Opening
• Shift: movement along a line in the base-pair plane, from the minor groove to the major groove.
• Slide: movement along a line in the base-pair plane, from one strand to the other.
• Rise: movement along the helix axis.
• Tilt: rotation around the shift axis.
• Roll: rotation around the slide axis.
• Twist: rotation around the rise axis.
• X-displacement
• Y-displacement
• Inclination
• Tip
• Pitch: the height of one full turn of the helix.

The measurements rise and twist determine whether the helix is left-handed or right-handed and its pitch. Other measurements can be zero. Slide and shift are usually small in B-DNA but are larger in A- and Z-DNA. Roll and tilt cause base pairs to be less parallel, and these are usually small.

The term "tilt" has sometimes been used differently in scientific writing. It refers to how the axis of the first base pair between strands bends away from being perpendicular to the helix axis. This is the same as slide between base pairs and is correctly called "inclination" in helix-based measurements.

Helix geometries

At least three DNA shapes are found in nature: A-DNA, B-DNA, and Z-DNA. The B form, discovered by James Watson and Francis Crick, is most common in cells. It is 23.7 Å wide and extends 34 Å for every 10 base pairs. The double helix twists to the right, completing one full turn every 10.4–10.5 base pairs in solution. This twisting pattern, called the helical pitch, depends on forces between bases in the DNA chain.

A-DNA and Z-DNA have different shapes and sizes compared to B-DNA, but they still form helical structures. A-DNA was once thought to appear only in lab samples without water or in DNA-RNA hybrids. However, DNA dehydration also happens inside living cells, and A-DNA now has known biological roles. Some DNA segments that are chemically modified for regulation may form Z-DNA, which twists in the opposite direction compared to A-DNA and B-DNA. Evidence also shows that protein-DNA complexes can create Z-DNA structures.

Other DNA shapes, such as C-DNA, E-DNA, L-DNA, P-DNA, S-DNA, and others, have been described. Only the letters F, Q, U, V, and Y remain to name any future DNA forms. Most of these shapes are made in labs and not found in natural biological systems. Some DNA forms have three or four strands, like G-quadruplex and i-motif structures.

Two strands twist together to form the DNA backbone. Another double helix can be seen by looking at the spaces, or grooves, between the strands. These grooves are next to the base pairs and may serve as binding sites. Since the strands are not directly opposite each other, the grooves are different sizes. The larger groove, called the major groove, is 22 Å wide, while the smaller groove, the minor groove, is 12 Å wide. The minor groove is narrower, making the edges of the bases more accessible in the major groove. Proteins that bind to specific DNA sequences often interact with the sides of the bases exposed in the major groove. This pattern changes in unusual DNA shapes, but the terms "major" and "minor" grooves always refer to size differences compared to the standard B form.

In the late 1970s, scientists briefly considered non-helical models to explain challenges in DNA replication in plasmids and chromatin. However, these models were not accepted after experiments like X-ray crystallography of DNA and discoveries about topoisomerases. Today, the double-helix model is widely accepted, and non-helical models are not supported by the scientific community.

Bending

DNA is a stiff molecule that is often compared to a worm-like chain. It can bend, twist, and be compressed, and these actions affect how DNA behaves inside a cell. The stiffness of DNA when twisted is important for forming circular DNA and for how proteins attach to DNA. The stiffness of DNA when bent is important for wrapping DNA around proteins and forming circles. Compression is not important unless very high forces are applied.

In water, DNA is not rigid. It constantly changes shape because of heat and collisions with water molecules. This makes it hard to measure DNA's rigidity using traditional methods. Instead, scientists use a value called the persistence length to describe how stiff DNA is. This value can be measured directly using a tool called an atomic force microscope. In water, the average persistence length of DNA is about 50 nanometers (or 150 base pairs). This value can range from 45 to 60 nanometers (or 132 to 176 base pairs). The persistence length can change depending on temperature, water conditions, and DNA length. This makes DNA a moderately stiff molecule.

The persistence length of a DNA segment depends on its sequence. This variation is mainly due to how the bases in DNA stack together and how they extend into the minor and major grooves of DNA.

At lengths larger than the persistence length, DNA behaves in a way that matches standard models used to study polymers, such as the Kratky-Porod model. These models show that DNA follows Hooke's law when bent under very small forces. For DNA segments shorter than the persistence length, the bending force is nearly constant, and DNA behaves differently than predicted by the Kratky-Porod model.

This behavior makes it easier to form circular DNA from small DNA molecules and increases the chance of finding highly bent sections of DNA.

DNA often prefers to bend in a specific direction, a property called anisotropic bending. This preference is due to the arrangement of bases in DNA. A random sequence of bases results in no preferred bending direction, called isotropic bending.

The preferred bending direction of DNA is determined by how stable the bases are when stacked on top of each other. If unstable stacking occurs on one side of the DNA helix, DNA will tend to bend away from that side. As the bend becomes sharper, steric hindrance and the ability of residues to move relative to each other, especially in the minor groove, also influence bending. Adenine (A) and thymine (T) bases are often found in the minor groove on the inside of bends. This effect is especially noticeable in DNA-protein interactions, such as in nucleosome particles.

Some DNA molecules have a strong preference for bending in a specific direction, making them intrinsically bent. This was first observed in trypanosomatid kinetoplast DNA. Typical sequences that cause intrinsic bending include stretches of 4–6 A and T bases separated by sections rich in guanine (G) and cytosine (C) bases. These sequences keep A and T bases aligned with the minor groove on one side of the molecule. For example, the intrinsically bent structure is caused by the "propeller twist" of base pairs, which allows unusual hydrogen bonds between base pairs. At higher temperatures, this structure breaks down, and the intrinsic bend disappears.

All DNA that bends in a preferred direction has, on average, a longer persistence length and greater axial stiffness. This increased rigidity prevents random bending, which would make the molecule behave isotropically.

DNA circularization depends on both its bending stiffness and its torsional stiffness. For DNA to form a circle, it must be long enough to bend into a full circle and have the correct number of base pairs so that the ends align properly for bonding. The ideal length for DNA circularization is about 400 base pairs (136 nanometers), with an exact number of helical turns, such as multiples of 10.4 base pairs. If the number of base pairs is not a whole number, it creates a significant energy barrier for circularization. For example, a DNA molecule with 312 base pairs (10.4 x 30) will circularize much faster than one with 317 base pairs (10.4 x 30.5).

Short circularized DNA segments do not bend uniformly. Instead, bending is concentrated in 1–2 kinks that form in AT-rich sections. If a nick is present in the DNA, bending is localized to the nick site.

Stretching

DNA can stretch when pulled because of the energy from heat in the surrounding water. In solution, DNA constantly changes shape due to the movement of water molecules and the heat from the solvent. Because of this, DNA tends to stay in a tangled, compact form rather than being stretched out. When force is applied, DNA straightens. Scientists have studied this behavior using tools like optical tweezers and found that DNA acts similarly to a model called the Kratky-Porod worm-like chain under normal energy levels.

When DNA is pulled strongly and twisted, it may change into a different structure where the bases spread apart and the phosphate groups move toward the center. This structure is called P-form DNA, named after Linus Pauling, who first suggested it as a possible DNA shape.

Studies of DNA stretching without twisting show that DNA can change into other structures called S-form DNA. These structures are not yet fully understood because it is difficult to observe them in detail while they are being stretched. However, many computer models have been used to study them.

Possible S-DNA structures include those where the DNA bases stay connected through stacking and hydrogen bonds (common in GC-rich DNA) but allow stretching by tilting. Other structures involve partial separation of the bases (common in AT-rich DNA) while keeping them overall connected.

A proposed structure for DNA involves regular breaks in the base pairs, with one break every three base pairs. This structure, called Σ-DNA, keeps the bases flat and allows stretching. The name Σ-DNA comes from the Greek letter Sigma, which has three points resembling the three base pairs grouped together. This form is more likely to occur in DNA sequences with GNC patterns, which are thought to be important in evolution.

Supercoiling and topology

The B form of the DNA helix completes one full twist (360°) every 10.4 to 10.5 base pairs when there is no twisting force. However, many biological processes can create twisting force. A DNA segment that has more than the normal amount of twisting is called positively supercoiled, while one with less twisting is called negatively supercoiled. In living cells, DNA is usually negatively supercoiled, which helps the DNA strands separate (unwind) when making RNA during transcription.

Most DNA inside cells is topologically restricted. DNA is often found in closed loops, like plasmids in bacteria, which are completely enclosed. Long DNA molecules also behave as if they are in closed loops because their movement is limited. Even linear DNA segments are often attached to proteins or structures, such as cell membranes, forming closed loops.

Francis Crick first explained the importance of linking numbers in DNA supercoiling in a 1976 paper. He described three key values used to analyze DNA topology:

• L = linking number: The number of times one DNA strand wraps around the other. This is always a whole number in a closed loop and stays the same in a closed DNA domain.
• T = twist: The total number of turns in the DNA double helix. Normally, this number is close to the number of base pairs divided by 10.5, unless special substances (like ethidium bromide) change the DNA’s stiffness.
• W = writhe: The number of times the DNA helix twists around its superhelical axis.

The relationship between these values is: L = T + W and ΔL = ΔT + ΔW. Any change in twist (T) in a closed DNA domain must be balanced by a change in writhe (W), and vice versa. This balance creates complex DNA structures. For example, a circular DNA molecule with zero writhe is perfectly circular. If its twist is increased or decreased through supercoiling, the writhe changes, causing the DNA to form plectonemic or toroidal coils.

When the ends of a DNA segment are joined to form a circle, the DNA strands become topologically linked. This means the strands cannot be separated without breaking one of them, such as by heating. Enzymes called topoisomerases help untangle these linked DNA strands by cutting one or both strands so another segment can pass through. This process is essential for copying circular DNA and for certain types of DNA recombination in linear DNA.

For many years, scientists were unsure why eukaryotic DNA had leftover supercoiling. This mystery was called the "linking number paradox." However, when experiments showed that DNA wraps around a protein structure (the histone octamer) in a left-handed, over-twisted way within the nucleosome, the paradox was resolved.