We are used to thinking of DNA as a double helix - but this is just one of its many forms. Since Watson and Crick published their model, human cells have found a triple and quadruple DNA helix, as well as crosses, hairpins and other weaving patterns - some are easier to draw than to describe in words.
Watson and Crick weren't the only ones poring over the 3D model of DNA. They weren't even the first. Scraps of biochemical data could be used to construct a variety of molecular forms, and there were many options.
The conditions of the problem were the same for all. At the beginning of 1953, it was already clear how the nucleotide works:
- the remainder of phosphoric acid,
- one of the nitrogenous bases: adenine (A), guanine (G), thymine (T) or cytosine (C).
It was also known that nitrogenous bases were scattered along the chain for a reason: in any DNA molecule, the total amount of adenines and guanines was strictly equal to the amount of thymines and cytosines. In addition, in all X-rays of Rosalind Franklin and Raymond Gosling, regardless of which piece of DNA was imprinted on them, the filament itself had the same thickness. This meant that the shape remains unchanged for any nucleotide sequence.
From these introductory notes, Linus Pauling and Robert Corey put together their model - a triple helix bristling with nitrogenous bases on all sides (biochemists have assigned phosphate and sugar to the role of an internal core). This design looked unstable: it was not clear why the negatively charged phosphate groups in the center of the spiral did not repel each other.
DNA structure according to Pauling and Corey
Bruce Fraser solved this problem by turning the structure inside out: in his version, three threads looked out with phosphates. The nitrogenous bases were turned inward, but Fraser could not explain how they were connected.
The Watson and Crick model with a double helix twisted to the right was the most stable. Like Fraser, the scientists placed the phosphates on the outside and the nitrogenous bases on the inside. There was also a clear principle of their opposition in this model: A on one circuit was always connected with T on another, and G - with C. This explained why the thickness of the structure is stable - the pairs A-T and G-C are about the same size.
Pencil sketch of DNA structure by Francis Crick
Then there were other attempts to reassemble DNA into a new form. The Dutch biochemist Karst Hoogsteen, for example, noticed that it is possible to connect the same pairs of nucleotides with other faces - so the helix also remained stable, but it turned out to be thinner. Other authors have depicted DNA as a spiral with alternating left and right turns, or even as two double helices that form a single quadruple. And although the existence of the Watson-Crick double helix has since been confirmed many times, in the 21st century people continue to speculate about what forms a DNA strand takes inside a cell, where it is much more difficult to see it than in a test tube. True, none of the alternative ideas so far has turned out to be good enough to abandon the classic right-handed double helix.
Watson and Crick did more than just resolve the controversy over the shape of DNA. Their model immediately explained how this form works: a one-to-one correspondence makes each thread a template for the other. Having only one of the chains, it is always possible to restore the second one along it - all modern models of the transfer of genetic information are based on this principle.
Nevertheless, most of the "rejected" ideas turned out to be correct in some way. For almost 70 years of close scrutiny of DNA, almost all possible types of base connections, other spirals and even a left turn have been detected in it.
Roll up to the wrong place
The double helix itself can be structured in different ways. This was noticed by Rosalind Franklin, although she did not assume that there was a spiral in front of her, and even a double one. Under normal conditions, resembling intracellular, the DNA in the biologist's pictures had a "loose" shape, which Franklin called B-DNA. But if the humidity in the test tube fell below 75 percent, the resulting A-DNA was wider and denser.
A (left) and B (right) forms of DNA as seen by Rosalind Franklin
As it turned out later, A-DNA is really twisted tighter: it takes 10 nucleotides to turn the helix, and not 11, as in B-DNA. And they are located not perpendicular to the axis of symmetry of the spiral, but at an angle: if in B-DNA nucleotides are usually depicted as horizontal lines, in A-DNA they should be drawn obliquely.
Watson and Crick chose B-DNA as the basis for their model, and they were right. Later it turned out that the B-variant actually occurs in the cell much more often, and now it is considered the main form of DNA existence, and all deviations are often denoted by the general term "non-B DNA".
Moreover, the real double helix almost never lives up to its idyllic model. In living systems, B-DNA is, as a rule, twisted slightly more than Watson and Crick predicted, and the average number of nucleotides per turn of a helix in it is not 10 or 11, but about 10.5. In addition, individual pairs of nucleotides are constantly deviating from the set "horizontal" (this is called "propeller rotation"), therefore, the spiral is never absolutely smooth and even - here and there on its sides roughness sticks out: the ends of nucleotides at different angles.
"Propeller" turn of nucleotides in B-DNA
Later it turned out that the coils of the spiral can not only lie tighter or more relaxed, but completely twist counterclockwise (for example, the spiral of the Evolution tower in Moscow City, which clearly symbolizes the DNA strand, is twisted to the left). By a strange coincidence, this is exactly the kind of DNA that was seen in 1979, when it was finally possible to examine nucleic acids with high resolution. It was still a double helix, but in a completely different shape: 12 nucleotides per turn, even thinner than B-DNA and twisted not to the right, but to the left. The phosphate groups sticking out on the surface did not form a smooth spiral, but a zigzag, so the new version was called the Z-shape.
A-DNA (left), B-DNA (center), Z-DNA (right)
This, of course, did not mean that the Watson-Crick model was wrong. The Z-form was obtained under rather exotic conditions - in a solution with a high concentration of salts. And in the cell, it is also obtained from B-DNA only under certain circumstances: for example, when the "voltage" on the chain is too high and it must be released. The tension appears due to excessive twisting: the DNA strands are already wrapped relative to each other, but the double helix formed by them winds around some protein (for example, histone), so-called supercoiling occurs. The transition to the Z-form helps to relieve tension and unwind unnecessary turns - and this, in turn, is important so that new proteins can bind to DNA, for example, polymerase during transcription.
Therefore, DNA often takes the Z-form during gene transcription. Moreover, the more Z-DNA in this case, the more active the transcription is. Histones cannot bind to Z-DNA, so no one interferes with polymerase to do their job. And this, by the way, is actively used by tumor cells, in which a left-handed helix appears in time in front of the genes they need.
The Evolution Tower (foreground) looks like left-handed DNA
Then other forms of the double helix were found. Depending on the moisture content, salt content and nucleotide sequence in a particular region, DNA can be even more elongated (E-DNA) or shrink (C- and D-DNA), include metal ions (M-DNA), or be stretched so that instead of nitrogenous bases, phosphate groups (S-DNA) appear in the center of the helix. And after other types of intracellular DNA were added to the list, such as nuclear N-DNA and recombinant R-DNA (which, however, were included in this list not because of their shape, but position in the cell or origin), in the English alphabet for the DNA variants, the letters are almost out. Anyone who decides to open some more non-canonical form will have to choose from five free ones: F, Q, U, V, and Y.
Alphabetical List of DNA Forms
- A-DNA is double-stranded, slightly thicker than B.
- The B-DNA is the one that Watson and Creek built.
- C-DNA is double stranded, 9, 3 nucleotides per turn.
- D-DNA is double-stranded, narrow: 8 nucleotides per turn, contains many thymines.
- E-DNA is double-stranded, even narrower: 15 nucleotides per two turns.
- G-DNA is a quadruple helix with guanine tetrads.
- H-DNA is a triple helix.
- I-DNA is two double helices held together by the attraction of their cytosines.
- J-DNA is another triple helix formed by AC repeats.
- K-DNA - DNA from trypanosomes, especially rich in adenines.
- L-DNA - DNA based on L-deoxyribose (not D- as usual).
- M-DNA - B-DNA complexed with divalent metals.
- N-DNA is nuclear DNA.
- O-DNA is the start point of DNA doubling in bacteriophage λ.
- P-DNA is the Pauling-Corey triple helix.
- R-DNA - recombinant DNA (obtained by inserting a foreign fragment).
- S-DNA is double-stranded, elongated 1.6 times stronger than the B-form.
- T-DNA - similar to the D-form, found in the T2 bacteriophage.
- W-DNA is synonymous with Z-DNA.
- X-DNA is a double-stranded helix formed by AT repeats.
- Z-DNA is double-stranded, left-handed.
Get in the grip
In addition to all kinds of double helix shapes and ways of weaving it, DNA sometimes breaks down into individual strands, which form into hairpins, crosses, and other double-stranded shapes. It also happens that an already existing double helix is overgrown with new neighbors.
In 1985, it turned out that Pauling and Corey were right thirty years ago: the DNA triple helix (H-DNA) exists. However, it is not arranged at all as they expected. In a true triple helix, two chains are connected in the standard Watson-Crick manner, and the third adjoins them laterally, lying in a large groove between the chains. In this case, the nitrogenous bases of the third, additional thread are connected to the main pairs not in the classical way, but as if from the side - by the very bonds predicted by Karst Hoogsteen. He, too, in a way, was right.
The triple helix, like many alternative forms of DNA, also arises in response to supercoiling of the strand. However, unlike the Z-form, it does not support transcription, but, on the contrary, interferes with it. RNA polymerase, which habitually unweaves two strands in front of itself, does not always cope with separating the triplex. Therefore, if a triple helix is formed in a gene or its regulatory regions, it works worse than others.
Variants of the formation of a triple helix. Watson-Crick pairs are shown in black, the extra third nucleotide is highlighted
It also happens that not two or not three, but four DNA strands are connected at once. For this to happen, four guanine nucleotides must meet in one place - it does not matter if they are on two strands of the same strand or on four different strands that are not connected to each other. Each guanine forms a non-classical, Hoogsteen pair with two neighbors, and together they form a square guanine tetrad. If next to them there are other guanines that can create a square, then a stack is formed from them - a stack that holds four DNA strands side by side.
Guanine tetrad (top) and options for the arrangement of chains in the quadruplex (bottom)
All 30 years that have passed since the discovery of quadruplexes, the number of processes in which they are somehow involved is growing. More than two hundred proteins are already known that can selectively recognize guanine tetrads - the latter probably play the role of a kind of genetic markup, another way to regulate the packaging and transcription of genes. For example, they are often found in promoters (regulatory regions from which transcription starts) of different genes. More recently, scientists have even managed to distinguish different types of breast cancer through sets of quadruplexes - which, in turn, depended on which genes in tumor cells were overactive.
The further we look at the DNA molecule, the more we notice deviations from the long-familiar model. The double helix is not the only and not the final structure of DNA, but only one (albeit the most frequent) of the poses that it takes in a continuous dance. Obeying the dictates of the nucleotide sequence, the DNA strand contracts and unclenches, bends, twists and takes on an infinite number of (beautiful) forms. None of them is final: alternative DNA structures transform into each other, compete with the B-form and with each other, obey the signals of cellular proteins and direct their work themselves.
Find and lead
Non-canonical forms of DNA, for all their diversity, do not appear in random places. Stability is given to them by a certain set of nucleotides in their composition, therefore they appear only in those parts of the chain where there is a "convenient" sequence for them.
So, for example, there are certain regions in the DNA that are especially willing to fold into a zigzag. These are the places where G-C pairs alternate: after a left turn in them, every second nucleotide takes an "irregular" shape, hence the broken profile of the entire Z-shape. This means that sequences that tend to take the Z-form can be found right in the text - if you see HZGZGZGZHZHZ, then you are unlikely to go wrong. So in one work, for example, they counted 391 such regions in the human genome.
The places where the triple helix can form can also be recognized by the characteristic nucleotide sequence. The third chain is attached either according to the principle of complementarity - that is, another G is added to the G-C pair, forming G-C * G - or "to its own" - and it turns out G * G-C. Therefore, such a construction often occurs in those places of DNA where several identical (for example, YYYYG) or chemically similar (AGGAAG) nucleotides go in a row and where they form palindromic (mirror) repeats.
In the same way, the appearance of quadruplexes can be predicted from the DNA text. According to the results of only one sequencing (in fact, direct translation of DNA into letters), more than 700 thousand of them were found in the human genome. Not all of them are likely to be found in vivo - for this, the corresponding DNA strands need to be close at one point in the complex cell nucleus - however, this may mean that the four-helical structures have some specific role in the life of the cell.
The formation of alternative forms of DNA does not always benefit the cell: most of them are much less durable than ordinary B-DNA, and break much more often. Therefore, sequences that tend to form non-B forms become sites of genetic instability and increased mutagenesis. Some researchers see this as the engine of evolution - if such regions appear in genes associated with the development of an organism. Others blame alternative forms of DNA for all kinds of diseases associated with random mutations and rearrangements in the genome - from tumors to schizophrenia and autism.
It turns out that DNA contains not only information about the structure of cellular proteins and RNA, but also about what forms this information can take, in addition to the Watson-Crick standard. And these forms, in turn, determine what happens to this information: whether the cell can realize it or the gene, will be forever silent, or even break down altogether, giving rise to some additional mutations.
We will probably learn one day to interfere with this process - we could, for example, build a chain of nucleotides that would mimic the third strand in the helix and "slip" it at the right time in the right place to block the work of some unwanted gene in the cell. There were even bolder proposals - to use the triple helix for targeted genome editing: introduce a nucleotide into the cell that can form a triple helix with the target DNA region and induce the repair system to replace this region with a “healthy” variant from another chromosome.
In the meantime, we are just learning this, it remains to recognize the structure of DNA as another type of information - in addition to genetic (nucleotide "text") and epigenetic (availability of genes for reading) - which carries our genome. And we still have to learn how to work with it, influencing the content through the form, or vice versa.