In the previous article, we looked at the two grooves formed along the surface of the DNA double helix: the major groove and the minor groove. The bases that carry DNA’s genetic information face one another inside the double helix and form hydrogen bonds, but some of the chemical features along their edges remain exposed to the outside through these two grooves. DNA does not exist on its own. Inside the extremely cramped space of the nucleus, it is constantly interacting with countless proteins—from the histones that help package its enormously long strands, to the many transcription factors involved in transferring genetic information, to the repair teams called in immediately when DNA is damaged, and even to drugs that we introduce from outside the body for treatment. And for any of these interactions to begin, proteins first have to read the DNA. From this perspective, the major and minor grooves are particularly important.
So how do DNA-binding proteins make use of them? In the wide-open major groove, the different patterns of hydrogen-bond donors and acceptors, methyl groups, and other chemical features of each base pair are relatively well exposed. This allows DNA-binding proteins to directly read these chemical fingerprints and identify particular base sequences. The minor groove is narrower than the major groove, so the base information available for this kind of direct reading is more limited. Even so, structural features of DNA itself—such as groove width, curvature, and the local electrostatic environment—can also provide important information to proteins. In the previous article, we saw how proteins use both base readout, in which they directly read the chemical features of the bases, and shape readout, in which they read the shape of DNA, to find where they need to bind, and how the two grooves contribute to this process.
But after learning all this, another question came to mind. We said that proteins can read the “shape” of DNA—but does DNA always have the same shape? Surprisingly, the familiar double helix we have known since our school days is actually just one of several structures that DNA can adopt. Depending on its surrounding environment, base sequence, and structural conditions within the molecule itself, DNA can form different types of double helices. And it does not stop there. As we will explore in later articles, DNA can even form structures beyond the double helix, including triple helices and quadruplexes. So now I want to find out how DNA can take on these different shapes, and why it can even form non-canonical DNA structures such as triple helices and quadruplexes. Let us begin by comparing the three double-helical forms—A-, B-, and Z-DNA—and looking at the molecular structural factors that determine their shapes.
DNA Is a Dynamic Molecule, Not a Fixed Structure
DNA is a molecule for storing information. The sequences that provide the instructions for making proteins and functional RNAs are regions that can directly give rise to particular products, yet they make up only a relatively small portion of the entire genome. Large parts of the remaining genome still do not have clearly established functions. But according to our current understanding of molecular biology, much of the genome appears to do more than simply store information: it helps provide the conditions that allow DNA to function properly. It must help regulate when, where, and how much a gene is expressed, allow replication to begin accurately and be completed without errors, create the right conditions for genetic recombination when diversity is needed, and signal for repair proteins to gain access when damage occurs. The functions of DNA extend much further than we might first imagine. And to carry out transcription, replication, and repair with such precision in the constantly changing environment of a cell, DNA cannot simply insist on remaining a “double-twisted rod.” It needs the flexibility to change its shape as circumstances change. The DNA structure we usually picture is simply the most common and standard form among many possible structures. Hydrogen bonding between bases can also take different forms. The A-T and G-C base pairs discussed above use Watson–Crick base pairing, the most common and fundamental mode of base pairing, but bases can also form additional hydrogen bonds in other ways. Through these interactions, nucleic acids can form structures beyond the double helix, including triple helices and quadruplexes. These non-canonical modes of interaction add even more diversity to the functions of DNA.
Above all, the fundamental reason DNA can shift between different shapes as circumstances change and better serve its functions lies in the mechanical flexibility of the DNA molecule itself. A nucleotide is made of a phosphate group, a sugar, and a base, and there are several rotatable bonds between these components. The torsion angles of the sugar-phosphate backbone, changes in the shape of the sugar ring, and the orientation of the base around the N-glycosidic bond between the sugar and base can all contribute to structural change. It is a little like a jointed doll whose arms and legs can be moved into different combinations to create many different movements and poses. DNA likewise uses this structural freedom to form a variety of three-dimensional structures. Depending on which angles and arrangements are adopted, it can take on entirely different structural states, which in turn affect how it interacts with proteins.
So let us move beyond the simple image of a double helix and look at DNA as a real three-dimensional structure with its own geometry. From the overall cylindrical shape of the helix down to small local regions, DNA has considerable structural variability, and this is precisely what makes shape readout possible. Because DNA does not always have exactly the same shape, proteins can use a strategy of reading not only its sequence but also its shape. Now let us look, one by one, at the factors that make this structural flexibility possible.
Classification Based on Double-Helix Shape: A-, B-, and Z-DNA
The three-dimensional structure of DNA can change through three major structural factors. DNA is a large molecule made of a long chain of nucleotides, each consisting of a base, a sugar, and a phosphate group. Its three-dimensional structure is determined by a combination of the torsion angles along the sugar-phosphate backbone, the conformation of the sugar ring, and the orientation of the base around the N-glycosidic bond between the sugar and base. Different combinations of these features give rise to different DNA double-helix structures, which can be broadly classified into three forms: A-, B-, and Z-DNA. Before looking at each structure, let us first understand the factors that create these structural differences.
C2′-endo vs. C3′-endo Puckering of the Ribose Sugar
One of the most important differences between A- and B-form double helices lies in the shape of the sugar ring—in other words, its puckering. (The word puckering is also used in sewing when thread pulls the fabric together, making it wrinkle or gather.) In A-DNA, the deoxyribose sugar mainly adopts a C3′-endo conformation, in which the C3′ carbon protrudes out of the average plane of the sugar ring. In B-DNA, the deoxyribose sugar mainly adopts a C2′-endo conformation, with the C2′ carbon protruding out of the average plane. To understand why this difference arises, let us first look at some basic properties of the sugar ring.
As we saw in the earlier series on carbohydrates, sugars are generally more stable in cyclic rather than linear forms under physiological conditions, and ring formation creates a new chiral center known as the anomeric carbon. For a D-sugar, if the OH group on the anomeric carbon lies on the same side as the CH₂OH group, it is the β anomer; if it lies on the opposite side, it is the α anomer. We looked at this process and the α and β anomers in detail in an earlier article.
α and β anomers related link: Carbohydrates 4: Hemiacetal vs. Hemiketal (Featuring the Cyclization of Fructose)
The same basic α/β stereochemical relationship also applies to the sugars in nucleic acids. The difference is that when a nucleoside is formed, the OH group on the anomeric carbon, C1′, does not remain there. Instead, a base is attached at this position through an N-glycosidic bond. So just as α and β in carbohydrates are distinguished by the direction of the OH group at the anomeric carbon, in nucleic acids they are distinguished by the relative orientation of the base attached to C1′. The nucleosides found in natural nucleic acids have β-N-glycosidic bonds. Remembering the spatial relationship between α and β that we saw in carbohydrates therefore makes it much easier to understand how a base is attached to the sugar in a nucleic acid.
Five-membered sugar rings such as ribose cannot remain perfectly flat. As the single bonds within the ring rotate together (ν0, ν1, ν2, ν3, and ν4), some atoms move out of the average plane of the ring, producing a three-dimensional shape known as puckering. If the atom that moves out of the ring plane lies on the same side as the substituent toward C5′, the conformation is called endo; if it lies on the opposite side, it is called exo. For example, when the C2′ carbon protrudes toward the same side as the C5′ substituent, the sugar adopts a C2′-endo conformation. When C3′ protrudes toward that side, it adopts a C3′-endo conformation. C2′-endo is also called the S (South) type, while C3′-endo is called the N (North) type. Many three-dimensional conformations are possible, but these are two of the most common forms of sugar puckering.
Now imagine nucleotides linked together in a long chain. DNA lacks the OH group at the C2′ position of the sugar, so it can adopt a C2′-endo pucker without the same steric constraint. RNA, however, has an OH group at C2′. If it adopts a C2′-endo conformation, this OH group can come into unfavorable contact with nearby phosphate groups or neighboring atoms, making the conformation energetically less favorable. As a result, C2′-endo is much less favored in RNA, while the C3′-endo pucker is strongly preferred. This preference for C3′-endo changes the geometry of the sugar-phosphate backbone, shortens the distance between neighboring phosphates, and causes the base pairs to become more tilted. This geometry is known as the A-form helix, and double-stranded RNA and RNA:DNA hybrids adopt this type of double-helical structure.
Base Orientation: anti vs. syn
The base and sugar are connected through an N-glycosidic bond (C1′–N). The base hangs from the C1′ carbon of the sugar and can change its orientation by rotating around this bond axis. Two representative orientations are called anti and syn. When the base ring rotates outward from the ribose plane and points away from the sugar, it is in the anti orientation. When the base folds over and lies closer above the sugar, it is in the syn orientation. Put simply, if the base stands out beside the sugar, it is anti; if it folds over the sugar and overlaps it, it is syn. In the image below, these two orientations can be distinguished by looking at the relative positions of H1′, the hydrogen attached to C1′ of the sugar, and H8, the hydrogen at position 8 of guanine. We will see below how this difference in orientation affects the structure of Z-DNA.
The Sugar-Phosphate Backbone: Six Torsion Angles
The sugar-phosphate backbone places strong constraints on the overall structure of the chain, whether the nucleic acid is single- or double-stranded. At the same time, because the backbone is connected through a series of single bonds, the atoms joined by these bonds can in principle rotate around their bond axes, giving the backbone considerable flexibility. The extent of this rotation is described by torsion angles, and the geometry of the DNA sugar-phosphate backbone is defined by a combination of six torsion angles, designated α through ζ. In simple terms, we can think of a torsion angle as describing how much a bond is twisted. Together, these torsion angles determine how the chain as a whole winds and twists. The image above shows that even when the structure of the sugar ring itself is kept fixed, changes in these torsion angles can cause the sugar-phosphate backbone to twist into different shapes. In other words, even with the same chemical composition, the three-dimensional arrangement of the backbone can change considerably depending on the direction and angle of rotation around its bonds. Of course, a real molecule cannot adopt every theoretically possible combination. Steric clashes between atoms and energetically unfavorable arrangements limit the conformations that are actually observed. Nor does the sugar-phosphate backbone become permanently fixed once a particular geometry is formed. Because phosphate groups carry negative charges, factors such as surrounding cations that help reduce repulsion between phosphates, together with the puckering state of the sugar rings, continually influence the backbone and allow it to adjust toward lower-energy arrangements. These subtle adjustments also contribute to the formation of A-, B-, and Z-form DNA helices.
DNA is not locked into a single shape. It can shift among different double-helical structures known as A-, B-, and Z-form DNA. Changes in sugar puckering, rotation along the sugar-phosphate backbone, and the orientation of the bases attached to the sugars can alter the overall structure of the double helix. This is more than simply a change in appearance: it also changes the way DNA interacts with its surroundings. Let us now look more closely at how these DNA structures differ and what functions and roles they have.
B-DNA: The Standard Double-Helix Structure
DNA inside the cell can undergo structural changes depending on biological processes and its local environment, and under certain conditions it can adopt alternative helical structures other than the B-form. Among these forms, however, B-DNA is the most common and stable under physiological cellular conditions—and it is the structure most familiar to us. It is the standard form that usually comes to mind when we picture DNA. In an aqueous environment without unusual conditions, DNA naturally tends to adopt the B-form structure.
In B-DNA, the deoxyribose sugar mainly adopts a C2′-endo pucker. In this structure, the distance between neighboring phosphates along the sugar-phosphate backbone remains relatively large, while the base pairs are arranged almost perpendicular to the helix axis. As a result, the spacing between base pairs remains regular, producing a long, smooth right-handed helix. This arrangement is favorable for maintaining stable base-stacking interactions and the geometry of the phosphate backbone.
One important feature of B-DNA is the clear distinction between the major groove and minor groove that we looked at earlier. The wide and accessible major groove exposes a rich pattern of chemical features, including hydrogen-bond donors and acceptors and methyl groups, allowing proteins to read base-sequence information relatively directly. Proteins can also bind to the narrower and deeper minor groove, where some recognize structural features such as DNA curvature and overall shape. In other words, the structure of B-DNA allows proteins to use both direct reading of base-sequence information and recognition of DNA shape. Because most cellular DNA exists in the B-form, most DNA-binding proteins—including transcription factors, regulatory proteins, and proteins involved in replication and repair—have evolved to recognize B-DNA. In this sense, the B-form is the most standard and common DNA structure, providing the basic architecture best suited for storing genetic information and allowing that information to be read.
B-DNA is commonly found in the water-rich physiological environment of the cell and remains stable when sufficient water is available. If hydration decreases or DNA is exposed to dehydrating conditions, however, the hydration environment around the molecule changes and A-DNA can become relatively more stable than B-DNA. As the sugar pucker shifts from C2′-endo toward C3′-endo, an A-form helix can form.
A-DNA: A Double-Helix Structure Found in RNA-Containing Nucleic Acids
The A-form structure is commonly found in double-stranded RNA (dsRNA), dehydrated DNA, and RNA:DNA hybrids that form transiently during processes such as transcription and replication. One of the major differences between DNA and RNA lies in their sugars. The sugar in DNA, deoxyribose, lacks an oxygen at the C2′ position, whereas the ribose sugar in RNA carries a 2′-OH group. This seemingly small difference has a major effect on the three-dimensional shape of the sugar. In RNA, the 2′-OH group has steric and electrostatic interactions with nearby phosphate groups and the sugar-phosphate backbone, making the C3′-endo sugar pucker energetically more stable and strongly preferred. A shift from the C2′-endo pucker common in B-DNA to C3′-endo changes the arrangement of the sugar-phosphate backbone. The distance between neighboring phosphates (P-P) becomes shorter than in B-DNA, and the base pairs stack at a greater tilt—about 20°—relative to the helix axis. As a result, the entire helix becomes more compact and compressed than B-DNA. As these geometric changes accumulate, the A-form helix develops a very narrow, deep major groove and a wide, shallow minor groove, giving it a groove geometry distinctly different from that of B-DNA.
Because of this groove geometry, an A-form double helix is less suited than B-DNA to protein recognition that depends on precisely reading base-sequence information through the major groove. Instead, because the A-form naturally appears in particular nucleic-acid states such as RNA:DNA hybrids and dsRNA, cells can use these structural features to allow particular enzymes to recognize these structures.
RNA:DNA Hybrids and R-Loops
During transcription, the DNA double helix opens locally and one of its strands serves as a template for RNA synthesis. Sometimes, part of the newly synthesized RNA binds back to the DNA template strand that was just transcribed. The resulting structure is called an RNA:DNA hybrid. While this hybrid remains in place, the opposite DNA strand, which would normally pair again with its complementary DNA strand, remains exposed as single-stranded DNA. The result is a three-stranded structure containing both an RNA:DNA hybrid and an exposed single-stranded DNA (ssDNA) strand. This structure is called an R-loop. R-loops occur naturally in cells and can even serve useful functions. B cells, for example, make functional use of R-loops during class-switch recombination of immunoglobulin genes to generate antibody diversity.[1] Problems arise when R-loops form excessively or persist for too long. If they remain unresolved, the transcription and replication machinery can collide, causing replication stress and potentially leading to the accumulation of mutations or DNA damage. Because RNA contains a 2′-OH group, an RNA:DNA hybrid cannot readily maintain B-DNA geometry and instead naturally adopts an A-form double helix. Thus, within an R-loop, the RNA:DNA hybrid region locally takes on A-form helical geometry.
Resolving R-Loops with RNase H and RPA
Cells have enzymes that process RNA:DNA hybrids, and one of the best-known examples is RNase H1 (Ribonuclease H1). RNase H1 recognizes the structure of an RNA:DNA hybrid, binds specifically to it, and selectively cleaves the RNA strand. Once the RNA is cleaved, the hybrid becomes destabilized and the R-loop can be resolved. During this process, the exposed single-stranded DNA needs to be stabilized because it can pair again or fold into secondary structures. This is where RPA (Replication Protein A) comes in. RPA binds single-stranded DNA with high affinity, stabilizing the exposed DNA strand and helping prevent unwanted reannealing or secondary-structure formation.
But RPA’s role does not end there. Recent research has shown that RPA can bind the single-stranded DNA that becomes transiently exposed within an RNA:DNA hybrid and promote separation of the hybrid. In doing so, it not only helps RNase H1 cleave the RNA more rapidly, but also allows RNase H1 to remain engaged and continue cleaving the RNA processively rather than easily dissociating after a single cut.[2] A similar role is seen during DNA replication. When a helicase unwinds double-stranded DNA, the two DNA strands temporarily become single-stranded, and single-stranded DNA-binding proteins rapidly bind and protect the exposed DNA. RPA performs this kind of single-strand-binding function in eukaryotic cells.
The Biological Significance of the A-Form Structure
The A-form can be understood as a structural state that arises naturally from the chemical properties of RNA. Cells can then make use of these structural differences to activate enzymes that recognize and process particular situations such as RNA:DNA hybrids. Rather than being a structure suited for transcription factors to precisely read base sequences, as in B-form DNA, the A-form is a characteristic geometry found in RNA-containing nucleic-acid structures. Cells can use this geometry as part of the mechanisms that recognize and process hybrid nucleic acids.
Z-DNA: An Alternative Structure for Relieving Supercoiling Stress
Unlike B- and A-form helices, both of which are right-handed, Z-DNA is a highly distinctive left-handed double helix. Rather than simply being another DNA form that happens to twist in the opposite direction, Z-DNA can be understood as an alternative structural state that DNA can adopt when it is subjected to strong torsional stress.
The distinctive structure of Z-DNA is closely related to the way the bases are oriented relative to their sugars—in other words, the syn and anti orientations. A base is attached to the C1′ carbon of the sugar through an N-glycosidic bond, and its ring can point in different directions around this bond axis. When the base extends away from the sugar, it is in the anti orientation; when it folds over and lies closer to the sugar, it is in the syn orientation. In B-DNA, most bases adopt the anti orientation, helping create a relatively uniform and smooth right-handed helix. Z-DNA follows a different pattern. Particularly in sequences with alternating G-C base pairs, guanine (G) adopts the syn orientation while cytosine (C) remains anti. This alternating syn–anti arrangement means that instead of progressing smoothly around the helix axis, the bases stack in an alternating, zigzagging pattern. As a result, the sugar-phosphate backbone takes on a characteristic zigzag shape rather than a smooth helical path. It is from this distinctive backbone geometry that Z-DNA gets its name.
So when does DNA switch into this unusual and seemingly less stable structure? The answer is closely connected to a physical phenomenon called supercoiling. DNA does not exist simply as a double helix. The double helix itself can become further wound or unwound, storing torsional energy in the molecule. This is what we call supercoiling. If DNA becomes more tightly wound than its original state, it develops positive supercoiling; if it becomes underwound, it develops negative supercoiling. The word “underwound” might make it sound as though the DNA has simply loosened up and become more relaxed, but that is not actually the case. Imagine holding both ends of a coiled telephone cord so that neither end can rotate freely, and then twisting the cord in the direction that would unwind its original coils. The cord tries to unwind, but because its ends cannot rotate freely, it cannot fully release that force. Instead, the cord itself begins to twist or bend to absorb the strain. Negative supercoiling in DNA can be understood in much the same way: a force is acting in the direction that tends to unwind the double helix. If the DNA cannot freely rotate at its ends to release that force—as is often the case for constrained DNA inside the cell—the force does not simply disappear. It remains stored in the DNA as torsional stress.
During DNA replication and transcription, the two DNA strands are locally opened while the replication or transcription machinery moves along the molecule. This can build up torsional strain in the surrounding DNA. During transcription in particular, as RNA polymerase moves along DNA, positive supercoiling tends to accumulate ahead of it and negative supercoiling behind it. In regions where transcription is rapid and intense, this negative supercoiling can build up considerably over a short period of time, making it increasingly energetically costly for the DNA to remain entirely in its normal right-handed B-form. But if a nearby sequence is particularly favorable for Z-DNA formation, another option becomes available. Part of the right-handed B-DNA can switch into left-handed Z-DNA, allowing some of the accumulated torsional strain to be relieved through the structural transition itself, without actually cutting and reconnecting the DNA. In other words, rather than forcing the DNA to remain entirely right-handed while bearing the torsional strain, a local region can reverse its helical direction and shift into a more energetically favorable structural state.[3]
If this torsional strain is not relieved, the DNA double helix can become structurally unstable and interfere with the progression of the replication and transcription machinery. One way to relieve this strain is through topoisomerases, enzymes that cut and rejoin DNA to remove or redistribute supercoiling. Another is for DNA itself to undergo a structural transition that absorbs some of the torsional energy. Conversion to Z-DNA can serve as one such structural response. In alternating sequences such as (CG)n, strong negative supercoiling can make conversion from a right-handed B-form helix to a left-handed Z-form helix energetically more favorable than remaining in the B-form. Z-DNA can therefore be understood as an alternative structural state capable of absorbing some of the torsional energy when DNA becomes excessively twisted and stressed.
Under normal physiological conditions inside the cell, DNA generally remains in the B-form. But in regions where transcription suddenly becomes very active or chromatin is rapidly reorganized, sequences favorable for Z-DNA formation can temporarily switch into the Z-form. This structural transition is dynamic: after helping relieve some of the accumulated torsional energy, the region can return to B-DNA. From this perspective, Z-DNA is less a structure designed for reading genetic information than a specialized structural state that reflects strong torsional stress within DNA.
The Innate Immune System and Z-Form Nucleic Acids
We have seen that under certain conditions, a region of ordinary B-DNA can switch into the left-handed Z-DNA form. Interestingly, however, this Z-form structure is not limited to DNA. Double-stranded RNA (dsRNA) normally forms a right-handed A-form helix, but some RNA can also adopt a left-handed Z-RNA structure under particular conditions. The left-handed structures formed by DNA and RNA are collectively referred to as Z-form nucleic acids.
Z-RNA also has an intriguing connection with our innate immune system, so it is worth taking a closer look. Many viruses use RNA as genetic material and produce various forms of RNA as they replicate and multiply. Fortunately, our cells are equipped with several types of innate immune RNA sensors that can detect characteristic features of these RNAs and recognize that an invasion may be underway. These sensors can respond not only to whether RNA is single- or double-stranded, but also to features such as its length, structure, and chemical characteristics. For example, RIG-I mainly detects relatively short double-stranded RNA, whereas MDA5 mainly detects long double-stranded RNA. When these sensors detect signs of viral infection, the cell activates antiviral defenses, including the production of cytokines such as type I interferons, which help prevent viral replication and spread.[4]
But this defense system faces a difficult problem: these kinds of RNA are not produced only by viruses. Our own cells can also generate double-stranded RNA (dsRNA) when complementary RNA sequences pair with one another during normal RNA metabolism. If a cell cannot properly distinguish its own RNA from viral RNA that has entered from outside, it could trigger an immune response against its own RNA even when no infection is present. It needs to detect viruses effectively while at the same time avoiding reactions against its own RNA. So how does our body deal with this problem?
ADAR1 and ZBP1
One important protein that helps prevent such mistaken immune responses is ADAR1 (Adenosine Deaminase Acting on RNA 1). ADAR1 binds to double-stranded RNA produced within our own cells and converts adenosine (A) in the RNA to inosine (I), a process known as A-to-I RNA editing. This editing changes RNA base pairing and the properties of the double-stranded RNA, providing an important safeguard that helps prevent endogenous RNA produced by our own cells from being mistaken for viral RNA by innate immune sensors such as MDA5.
But ADAR1 has an even more interesting feature that connects directly to the Z-form nucleic acids we are discussing. One of the two forms of ADAR1, p150, contains a Zα domain at its N-terminus. Another protein called ZBP1 (Z-DNA-binding protein 1) also contains a Zα domain. What makes this domain interesting is that rather than reading a particular base sequence, it recognizes and binds the distinctive left-handed structure of Z-form nucleic acids. This is the shape readout we discussed in the previous article. Yet the consequences of Z-RNA binding by these two proteins are very different. In situations such as viral infection, ZBP1 can detect Z-form nucleic acids such as Z-RNA and trigger necroptosis, an inflammatory form of programmed cell death, to eliminate infected cells. ADAR1, by contrast, binds Z-RNA and helps suppress unnecessary activation of ZBP1.[4] Incidentally, ZBP1 got its name because it was first identified as a Z-DNA-binding protein, but it is now understood to recognize Z-form nucleic acids more broadly.
The story does not end there. A protective mechanism such as ADAR1, which normally helps prevent healthy cells from mistaking their own RNA for viral RNA and launching an unnecessary immune response, can instead be exploited by some cancer cells. In cancer cells, gene expression and RNA metabolism can become disrupted, producing double-stranded RNAs capable of stimulating the immune system. If these dsRNAs remain exposed, they can be detected by innate immune RNA sensors, triggering immune responses that interfere with the growth and survival of the cancer cells. But if ADAR1 suppresses the immunostimulatory properties of these RNAs, cancer cells can escape this alarm. In other words, the ADAR1 “false-alarm prevention system” that normally protects healthy cells from reacting against their own RNA can, in some cancers, be turned to the cancer cell’s advantage by helping it evade immune surveillance.
As we saw above, when ZBP1 is activated by recognizing Z-RNA, it can trigger cell death in cancer cells, and ADAR1 can suppress this response. Studies have shown that removing ADAR1 or disrupting its function in cancer cells can allow Z-RNA to accumulate, where it can be detected by ZBP1 and trigger cell death. For a cancer cell, then, ADAR1 can act as more than simply an RNA-editing enzyme. It can serve as a kind of shield, preventing its abnormal RNA from sending danger signals to the immune system or activating ZBP1-mediated cell-death pathways. For this reason, ADAR1 has become an important subject of research into cancer immune evasion and resistance to immunotherapy.[5]
What I find especially interesting about these three helical forms is that different nucleic-acid structures present different binding surfaces to proteins. DNA is not simply a medium for storing information; the proteins it can interact with can also depend on the shape it adopts. Proteins can use not only the base sequence itself but also the structure created by that sequence and its surrounding environment as a clue for recognition. In that sense, we might think of DNA shape itself as another layer of information.
The three helical forms appear in different structural and biological contexts. B-DNA is the most common basic form under physiological aqueous conditions and serves as the standard structure in which genetic information is stored and read inside the cell. The A-form, by contrast, arises naturally in RNA-containing nucleic-acid structures and is seen in situations such as RNA:DNA hybrids and double-stranded RNA. Z-DNA is an alternative structure that can form under strong torsional stress and in particular sequence contexts.
DNA structure, then, is not simply a static form. It can be thought of as a dynamic structural language that reflects the many physical and biochemical conditions occurring inside the cell. Understanding these structural differences is important not only in basic biology but also in drug design. When designing molecules that target nucleic acids, the structural form of the target must be taken into account. A molecule designed to recognize the major groove of B-DNA, for example, may not bind properly to an RNA:DNA hybrid or an A-form helix. Without an understanding of nucleic-acid structure, both binding affinity and selectivity for the intended target can suffer. Understanding the different structural states of DNA is therefore an important starting point not only for explaining molecular biological phenomena, but also for developing precise molecular-targeted therapeutic strategies. For reference, the structural features of A-, B-, and Z-DNA that we have explored so far are summarized in the table below.
[References]
[1] R-Loops and Its Chro-Mates: The Strange Case of Dr. Jekyll and Mr. Hyde
[2] RPA transforms RNase H1 to a bidirectional exoribonuclease for processive RNA–DNA hybrid cleavage
[3] Formation and biological implications of Z-DNA
https://doi.org/10.1016/j.tig.2025.07.006
[4] Innate immune responses to RNA: sensing and signaling
https://doi.org/10.3389/fimmu.2024.1287940
https://doi.org/10.1038/s41586-022-04753-7
[Image sources]
Image 1Challenges and opportunities in technologies and methods for lncRNA structure determination — CC BY 4.0
DNA Orbit Animated Clean — CC BY-SA 4.0
Image 2
Molecular Basis for Chiral Selection in RNA Aminoacylation — CC BY
Sugar Puckering Drives G-Quadruplex Refolding: Implications for V-Shaped Loops — CC BY 4.0
Image 3
Structural Probes in Quadruplex Nucleic Acid Structure Determination by NMR — CC BY
Image 4
Colorado State University BC401 – Aptamers — adapted
Image 5
Insights into the Molecular Structure, Stability, and Biological Significance of Non-Canonical DNA Forms, with a Focus on G-Quadruplexes and i-Motifs — CC BY 4.0
Image 6
R-Loops and Its Chro-Mates: The Strange Case of Dr. Jekyll and Mr. Hyde — CC BY 4.0
RPA transforms RNase H1 to a bidirectional exoribonuclease for processive RNA–DNA hybrid cleavage — CC BY
Image 7
DNA Structure: A-, B- and Z-DNA Helix Families — Table 2
Selected helical parameters for various polymorphs of DNA polynucleotides — Table 1.4
