In the previous article, we looked at the structure of nucleotides, the basic building blocks of nucleic acids, and how they link together to form the long nucleic acid chains we call DNA and RNA. But once we get this far, an interesting question naturally arises. DNA and RNA are made from similar building blocks, so why do living organisms use two different types of nucleic acid, and what is the relationship between them?
In living organisms today, DNA is clearly the main molecule responsible for storing genetic information over the long term. So when we talk about nucleic acids, it is natural for DNA to receive much of the attention. RNA plays an essential role in carrying the information stored in DNA and using it in the process of making proteins, which may make RNA seem like a supporting nucleic acid that mainly serves DNA. But if we judge RNA only by the role it plays today, we may miss just how remarkable its presence appears to have been during the earliest stages of life. Looking at the relationship between the two, I sometimes find myself thinking of RNA almost as the parent molecule that gave rise to DNA. Why might that be? In this article, we will travel back in time and take a closer look at the RNA world hypothesis—the idea that RNA may have existed before DNA—as well as the characteristics and roles of RNA.
From the Formation of Organic Molecules to the Beginning of Life
Many hypotheses have been proposed to explain how life may have emerged under the conditions of the early Earth. These hypotheses explore different possibilities, and the reasoning behind them, for how inorganic substances could have reacted in the harsh and extreme environment of the early Earth to form organic molecules such as amino acids, lipids, and nucleic acids, and how these molecules might then have become concentrated and polymerized, eventually leading to the emergence of life.
To briefly list a few of these ideas, there is the primordial soup hypothesis, which proposes that organic molecules formed in a “primordial soup” containing a mixture of many inorganic substances; the hydrothermal vent hypothesis, which suggests that complex organic molecules may have been synthesized through reactions involving various minerals near deep-sea hydrothermal vents, where hot water and gases emerge in an environment relatively sheltered from ultraviolet radiation and other radiation; the rock-surface hypothesis, which proposes that inorganic substances were converted into organic molecules and concentrated in particular rocky environments rather than in the ocean or atmosphere; the clay-catalysis hypothesis, in which the surfaces of mineral particles in clay may have concentrated organic molecules and promoted chemical reactions, providing favorable conditions for forming the components of life; the warm little pond hypothesis, which proposes that small, warm ponds created by volcanic activity, with lower salinity than the ocean and a greater capacity to concentrate organic matter, may have provided favorable conditions for chemical concentration and reactions; and the panspermia hypothesis, which proposes that microorganisms or organic molecules arrived on Earth from space carried by meteorites or comets.
These hypotheses all offer ways to explain how increasingly complex molecules could have been synthesized on the early Earth. But that alone does not seem sufficient to explain how life itself emerged. Even if a life-like chemical system happened to arise spontaneously by chance, it would be difficult for that system to develop into life with continuity, persistence, and an evolutionary direction unless it could repeatedly replicate itself and store and preserve information about itself. With this in mind, RNA becomes particularly intriguing as a possible intermediate step in the transition from complex organic molecules to living systems capable of preserving genetic information and synthesizing proteins as organisms do today.
The RNA World Hypothesis
Laboratory experiments designed to reproduce various conditions of the early Earth have already shown that a range of organic molecules can form from relatively simple substances. The early Earth offered many sources of energy—lightning, the energy released as meteorites entered the atmosphere, ultraviolet radiation, other forms of radiation, lava, and hot gases from the deep sea—that could have driven early chemical reactions among basic atoms and molecules. Under various conditions containing simple substances such as water (H₂O), methane (CH₄), ammonia (NH₃), hydrogen, and hydrogen cyanide (HCN), amino acids, nucleobases, and sugar precursors could have formed abiotically. These organic molecules could then have become concentrated and combined into a wider variety of more complex molecules. Through countless rounds of chemical reactions, nucleobases may have joined with sugars to form nucleosides, which could then have acquired phosphate groups to form nucleotides. As these nucleotides linked together, short primitive RNA molecules of varying lengths may have polymerized at random.
Of course, it seems unlikely that only the four standard RNA bases we know today—A, U, G, and C—would have formed naturally in a primordial soup. A variety of bases, base analogues, and different combinations may have been tried, with bases and complementary pairing systems favorable for replication and structural formation more likely to persist through chemical selection. As we can see from complementary base pairing in nucleic acids, we can imagine a primitive form of replication in which complementary nucleotides lined up along a single RNA strand serving as a template, became linked together, and the two strands then separated again. Among these molecules, RNA that replicated faster or more accurately, or remained intact for longer, would probably have been more likely to be chemically selected and gain an advantage over others. RNA can also fold into many different shapes as complementary bases within the same strand are drawn together and form hydrogen bonds. Through such folding, some RNA molecules may have acquired particular three-dimensional structures that enabled them to promote chemical reactions that cleaved or formed bonds within their own chains or in other RNA molecules.
At some point, fatty acids with both hydrophilic and hydrophobic regions may also have spontaneously arranged into double-layered structures, with their hydrophobic regions turning inward away from water to form primitive membranes. If RNA molecules became enclosed within such primitive lipid membranes, RNA and its building materials could have become concentrated and interacted within a confined space separated from the surrounding environment. This could have opened the way for the development of primitive cell-like systems known as protocells.
Over a long period of evolution, the many functions once carried out by RNA alone may gradually have become divided among different types of molecules. RNA’s high reactivity was advantageous for forming diverse structures and carrying out catalytic functions, but it also placed limits on the long-term, stable preservation of genetic information. A more chemically stable molecule, DNA, may therefore have taken over the role of long-term information storage. Proteins, meanwhile, can fold into an even greater variety of structures than RNA and carry out highly precise and rapid catalytic reactions, and they may have taken over most enzymatic functions. In other words, a division of labor emerged in which each role was taken over by the molecule better suited to perform it. These changes may ultimately have led to the basic flow of genetic information familiar to us today: DNA → RNA → protein.
Traces of RNA That Support the RNA World Hypothesis
According to the RNA world hypothesis, RNA may have served as an important bridge in the transition from nonliving chemistry to biological systems. The hypothesis proposes that before life emerged, RNA or RNA-like molecules may have formed first and performed both information-storage and catalytic functions. Today, it is one of the leading and most widely discussed hypotheses for explaining the early stages of the origin of life. The idea did not initially attract widespread support, but interest grew dramatically after the discovery that RNA itself could undergo cleavage and catalyze chemical reactions—in other words, that catalytic RNAs called ribozymes exist. When the two chemists who carried out this work were awarded the Nobel Prize in Chemistry in 1989, the hypothesis received even greater attention. The discovery strongly supported the possibility that, before the division of labor among DNA, RNA, and proteins arose, primitive RNA molecules on the early Earth could have performed several of these functions at the same time.
As evolution continued, DNA rather than RNA took over the long-term storage of genetic information, while most catalytic functions were transferred to proteins. Yet traces of the information-storage and catalytic roles thought to have been performed by early RNA can still be found throughout modern living organisms. First, consider some of the familiar functions of RNA. Interestingly, RNA plays central roles at many points in the process of using the genetic information stored in DNA to make proteins. As the saying goes, “a barber cannot cut his own hair”: proteins do not simply take charge of making proteins themselves. Instead, RNA stands out at several of the most critical steps. mRNA copies and carries information from DNA; ribosomal RNA (rRNA) is a core component of the ribosome, the molecular factory that uses this information to make proteins; and tRNA reads the codons in mRNA and delivers the corresponding amino acids. Together, these roles show that RNA is broadly involved in information transfer as well as structural and catalytic functions. In addition, small nuclear RNA (snRNA) plays a central role in splicing, in which introns are removed from pre-mRNA after transcription and exons can be joined in different combinations. And of course, the RNA world extends far beyond these examples, including regulatory RNAs such as miRNA, siRNA, and lncRNA. If mRNA is the classic example of RNA as an information carrier, one of the strongest signs that RNA has retained a catalytic role can be found in the rRNA of the ribosome. If the existence of ribozymes is important evidence supporting the RNA world hypothesis, ribosomal rRNA is one of the clearest examples in modern organisms of RNA directly participating in catalysis. Ribosomes are shared by all cellular life—bacteria, archaea, and eukaryotes—and they all contain rRNA. Moreover, the core rRNA sequences and structures of the ribosome are highly conserved through evolution. Could the universal presence of ribosomes and the remarkable conservation of their core structure be telling us that ribosomes have been with life since very early in its evolutionary history?
The Ribosome: RNA Acting as an Enzyme (Ribozyme)
There are 20 standard amino acids with a wide variety of chemical groups, and proteins also have the advantage of being able to fold flexibly into complex tertiary and quaternary structures, giving them highly intricate and precise shapes. Thanks to these properties, protein enzymes can catalyze an enormous range of chemical reactions rapidly and precisely, and most enzymatic reactions in modern organisms are therefore carried out by proteins. Yet in some extremely important biochemical reactions, RNA still acts as the catalyst. These RNA-based catalysts are called ribozymes. Their existence directly demonstrates that RNA can not only store information but also perform catalytic functions that promote chemical reactions.
One of the best-known examples of a ribozyme is the ribosome, the molecular machine that synthesizes proteins. A ribosome consists of a small subunit that decodes the nucleotide sequence carried by mRNA and a large subunit that links amino acids together to synthesize a polypeptide. The two subunits normally exist separately and come together to form a functional ribosome during protein synthesis. Ribosomes are present in enormous numbers inside cells. A rapidly proliferating eukaryotic cell may contain as many as 10 million ribosomes. [1] Considering that most cellular functions are carried out through proteins, it is hardly surprising that cells devote a substantial portion of their resources to the machinery that produces them. Ribosomes may float freely through the cytoplasm or attach tightly to the surface of the endoplasmic reticulum (ER), giving it the bumpy appearance that earned it the name rough ER.
Although ribosomes are composed of roughly 40% protein and 60% rRNA, the catalytic core of the peptidyl transferase center—the part that actually catalyzes peptide-bond formation between amino acids—is made of rRNA, not protein. This rRNA promotes peptide-bond formation by positioning the reactants in the proper location and orientation. So saying that the proteins are there mainly to lend a hand is not entirely wrong. If we focus specifically on peptide-bond formation in the ribosome, its catalytic center is an RNA-based ribozyme. It is striking that in the ribosome, a structure shared by all living organisms, rRNA serves as the catalyst at the heart of one of life’s most fundamental events: protein synthesis. The fact that RNA itself catalyzes the process that makes proteins, rather than this central reaction being catalyzed by a protein, is considered an important piece of evidence supporting the RNA world hypothesis that RNA may have performed catalytic functions before proteins did.
Why Are Ribozymes So Important?
In living organisms today, the synthesis of RNA and DNA depends heavily on the activity of various protein enzymes. Yet producing those enzyme proteins requires genetic information stored in nucleic acids. In other words, proteins are needed to make nucleic acids, while nucleic acids are needed to make proteins. We are faced with the classic chicken-or-egg dilemma. This raises the possibility that during the earliest stages of life, there may have been molecules capable of replicating themselves and preserving information without depending on protein enzymes and a DNA-based genetic system like those used today. For this reason, a substance capable of both storing information and acting as a catalyst to promote chemical reactions becomes extremely important in explaining how early life could have emerged. Ribozymes provide direct evidence that RNA can combine information storage and catalytic activity within the same type of molecule. In this view, RNA is not simply a molecule that exists to carry genetic information from DNA. Instead, it becomes a leading candidate for a central component of the early chemical systems that may have opened the way for life to emerge.
Self-Splicing Introns
The process in which introns are removed from pre-mRNA transcribed from DNA and the exons are joined together is called splicing. When exons are joined in different combinations to produce different mRNAs and proteins, the process is called alternative splicing. At the catalytic center of the spliceosome, the molecular complex that carries out splicing, small nuclear RNAs (snRNAs) play essential roles alongside proteins. It has been proposed that this RNA-based catalytic mechanism may be evolutionarily related to self-splicing introns.
Self-splicing introns are RNAs that can promote their own cleavage and the joining of exons through the three-dimensional structure of the RNA itself, without any assistance from protein enzymes, making them classic examples of ribozymes. In these reactions, the RNA folds precisely to create an active site, and a particular OH group acts as a nucleophile to drive a phosphoryl-transfer reaction. This shows that RNA bases and the ribose sugar can perform some of the catalytic roles that amino acid side chains perform in protein enzymes. Self-splicing introns therefore provide another important example of RNA carrying out catalytic functions without depending on protein enzymes.
Phosphodiester Bonds: Formation of an RNA Chain
When an RNA chain is synthesized in our cells, RNA polymerase uses ribonucleoside triphosphates (NTPs: ATP, GTP, CTP, and UTP) as substrates. During polymerization, the 3′ OH of the ribose at the end of the growing chain attacks the α-phosphate of an incoming NTP, forming a new bond while the two terminal phosphates are released as pyrophosphate (PPi). As a result, a single phosphate becomes bonded to two sugars, forming the 3′–5′ phosphodiester bond that links them together, and the RNA chain is extended one nucleotide at a time. This is a nucleotidyl-transfer reaction catalyzed by a polymerase, and the energy that drives bond formation comes from the triphosphate structure of the incoming NTP together with the subsequent hydrolysis of PPi.
However, it has also been proposed that even on the early Earth, before enzymes existed, individual nucleotides could have spontaneously linked together to form short chains, or oligonucleotides, and various experiments have provided support for this possibility. Oligomers containing about ten nucleotides have been reported to form within several days under dry alkaline conditions [2], and experiments have also shown that nucleotides can polymerize nonenzymatically into oligomers under repeated wet–dry cycles, in which water evaporates and is then replenished. [3]
Is DNA an Upgraded Version of RNA?
At first glance, the difference between RNA and DNA seems obvious. RNA usually exists as a single strand, whereas DNA typically forms a double helix of two intertwined strands. But this is not the most fundamental distinction between the two molecules. One of the more fundamental differences lies in the structure of the sugars they contain. The ribose in RNA has an OH group on its 2′ carbon, whereas the 2′-deoxyribose in DNA has a hydrogen at that position instead of an OH group. The names ribonucleic acid and deoxyribonucleic acid themselves arise from this very difference.
The 2′-OH Group and Self-Cleavage
The ribose sugar in RNA has an OH group on its 2′ carbon. This small functional group has a major influence on the chemical properties of RNA. Under certain environmental conditions, the 2′-OH group can lose a proton, or become deprotonated, producing a negatively charged 2′-O⁻. The resulting oxygen, with its increased electron density, can donate an electron pair and act as a strong nucleophile seeking to form a new bond. Close to this nucleophile lies an electrophilic phosphorus (P) atom in a phosphate group of the RNA backbone. If the 2′-O⁻ attacks this phosphorus atom, the phosphodiester bond in the RNA backbone can be cleaved.
Metal ions such as Mg²⁺ can play important roles in promoting these self-cleavage reactions within RNA molecules. Mg²⁺ can interact with surrounding water molecules and with various oxygen atoms in RNA, helping to create a chemical environment in which the 2′-OH can more readily participate in the reaction. For example, by influencing the local chemical environment or acid–base chemistry around the 2′-OH, Mg²⁺ may facilitate the loss of a proton (H⁺) from the 2′-OH and thereby promote formation of the nucleophilic 2′-O⁻. The precise role of Mg²⁺ in this process, however, can vary depending on the particular RNA or ribozyme involved.
RNA phosphate groups also carry negative charges, so when they approach one another, electrostatic repulsion tends to push them apart. Positively charged Mg²⁺ can partially shield these negative charges around the phosphate groups and reduce this electrostatic repulsion. This can help RNA fold into the structures required for catalytic reactions and maintain those structures in a stable form. At a catalytic site, Mg²⁺ may also help stabilize negative charges that develop during the reaction, creating an environment that makes the reaction easier to proceed. In this respect, the role of Mg²⁺ resembles its role in kinase reactions, where it helps stabilize the negative charges around the phosphate groups of ATP and facilitates phosphate transfer. In this way, metal ions such as Mg²⁺ serve as important cofactors in many biochemical reactions involving phosphate groups.
Now let us look at how the RNA backbone is actually cleaved. The 2′-O⁻ formed when the 2′-OH loses a proton makes a nucleophilic attack on the phosphorus atom of the adjacent phosphate group. During this process, an unstable transition state forms in which five oxygen atoms are associated with the central phosphorus, and in some RNA-catalyzed reactions Mg²⁺ can help the reaction proceed by stabilizing the charges that develop in this state. A new bond then forms between the 2′-O and the phosphorus at the same time that the existing bond between the phosphorus and the 5′-O in the RNA backbone is broken. The RNA chain is consequently cleaved, leaving a 2′,3′-cyclic phosphate terminus on one side and a 5′-OH terminus on the other. Depending on the conditions, the cyclic phosphate can subsequently undergo hydrolysis.
Was RNA’s Instability Really Only a Disadvantage?
However, the fact that a molecule always carries within itself the possibility of cleaving its own backbone is a significant disadvantage when the goal is to preserve and store genetic information safely. If self-cleavage occurs frequently, it becomes difficult to maintain long nucleic acid chains in a stable form. As evolution moved toward overcoming this chemical instability of RNA, DNA may have emerged using 2′-deoxyribose, in which the OH group at the reactive 2′ position is replaced by hydrogen. With the 2′-OH group removed, the internal nucleophilic attack possible in RNA is greatly suppressed, giving DNA a much more stable sugar–phosphate backbone. At neutral pH and physiological Mg²⁺ concentrations, the phosphodiester bonds of DNA have been reported to be about 200 times more stable than those of RNA. [4] By greatly reducing the risk of self-cleavage through internal nucleophilic attack, DNA gained much greater chemical stability than RNA and became far better suited to preserving long stretches of genetic information over extended periods. This increased chemical stability would have provided an important foundation for DNA eventually becoming the molecule responsible for long-term genetic information storage.
Interestingly, however, from another perspective, the instability caused by the 2′-OH group of ribose may not have been purely a disadvantage. In fact, this high reactivity may have played a crucial role in making RNA what it is. Structurally and chemically, this tiny functional group may have given RNA the ability to perform functions resembling those of enzymes. The ability to undergo chemical reactions within its own backbone, or to fold into diverse structures that promote particular reactions without the involvement of an external enzyme, resembles the catalytic capabilities of protein enzymes. The capacity to promote reactions at particular sites in itself or in other RNA molecules while also forming a wide variety of structures may have been especially useful in a primitive environment before protein enzymes appeared. During the earliest stages of life, countless experiments and failures must have occurred, and chemical flexibility—the ability to generate new structures and reactions—may have mattered more than long-term stability. In such an environment, RNA’s structural flexibility and high reactivity may actually have been advantages that opened up a wider range of possibilities. The 2′-OH group of RNA was therefore, on the one hand, a weakness that had to be overcome for long-term information storage, but on the other hand, it was also an important chemical feature that helped make RNA catalysis possible. The very same property can be an advantage or a disadvantage depending on the circumstances.
RNA Formation and Structure: RNA Folding
Unlike DNA, in which two complementary strands pair with each other to form a relatively stable double helix, RNA usually exists as a single strand. Yet it is precisely this single-stranded nature that allows sequences located far apart within the same RNA molecule to pair with one another. As a result, RNA can fold back on itself in many different ways and form diverse and complex structures, much like proteins. An RNA chain made of many linked nucleotides can adopt a variety of shapes depending on its length and sequence. Just as a long thread can bend and fold, an RNA strand can fold into many different forms, and complementary bases within the strand can pair with one another to create local double-stranded regions. These base pairs form through hydrogen bonds, while neighboring bases are further stabilized by base-stacking interactions as they align roughly parallel to one another. Hydrogen bonding is a polar interaction involving electronegative atoms and allows complementary bases to recognize and pair with one another. Base stacking, in contrast, arises as the flat bases with their aromatic ring structures align and stack roughly parallel to one another, and it makes a major contribution to the stability of RNA structure. We will look at hydrogen bonding and base stacking in much greater detail in the article on nitrogenous bases.
The ability of RNA to fold and rearrange itself in many different ways greatly expands the range of functions it can perform. Just as structure determines function in architecture, and shape determines function in proteins, structure is also a key factor closely linked to function in RNA. Proteins are organized hierarchically from their amino acid sequences into primary, secondary, tertiary, and quaternary structures. RNA can likewise be described hierarchically in terms of primary, secondary, and tertiary structure, and it can also form higher-order complexes involving multiple RNA molecules or combinations of RNA and proteins. In fact, RNA can form structures that are every bit as complex and diverse as those of proteins.
Primary to Quaternary Structure of RNA
The primary structure of RNA is the linear nucleotide sequence that makes up the RNA molecule. This sequence determines which bases can form complementary pairs with one another and provides the fundamental basis for RNA folding.
Secondary structure refers to local patterns of base pairing formed when complementary bases establish hydrogen bonds. The most basic example is a stem, or helix, a local double-stranded region formed when complementary bases within a single RNA strand pair with one another. These stem structures resemble double-stranded DNA, although, more precisely, the typical DNA double helix is predominantly B-form, whereas RNA double helices generally adopt an A-form structure.
When an unpaired region occurs between paired stem regions, it can form a loop, and the combination of a stem and a loop creates a hairpin structure. If unpaired bases occur within a stem, they can form a bulge or an internal loop. When several stem structures meet at a single point, they can also form a multibranch loop. These different patterns of folding make up the secondary structure of RNA. Calling them secondary structures does not mean that the actual molecule is flat. Rather, a secondary-structure diagram is a simplified kind of connectivity map showing which bases pair with one another. The actual RNA molecule, of course, exists in three-dimensional space. Tertiary structure refers to the three-dimensional shape that emerges when these secondary-structure elements interact further with one another. In other words, stems, loops, and other elements fold and twist through space to create a complex three-dimensional structure.
One characteristic fold commonly found in RNA tertiary structure is the pseudoknot. A pseudoknot forms when bases within a loop establish additional base pairs with another region of the RNA, creating two or more interlocked stem structures. Several types of pseudoknot exist, but the H-type pseudoknot is the best known. In an H-type pseudoknot, for example, bases in the loop of an RNA hairpin pair with complementary bases in another region outside the hairpin, forming a second stem. In this way, regions of the RNA chain that were originally separated become additionally connected, causing the structure to become intertwined. Although this may appear to be a relatively simple fold, these crossing interactions can increase the stability of the RNA structure and provide functionally important structural platforms for biological activity. Pseudoknots are found in many functional RNAs and ribozymes and play important roles in the structure and function of various RNAs, including telomerase RNA. [5]
Finally, when multiple RNA molecules interact with one another, or when RNA associates with proteins, they can form larger molecular complexes. This is sometimes described as quaternary structure, although the term is not used as strictly or universally for RNA as it is for proteins. A representative example is the ribosome discussed earlier, a massive complex composed of RNA and proteins.
Recent research has shown that a substantial portion of the human genome is in fact transcribed into RNA. As we saw in the previous article, in addition to protein-coding genes, the genome contains many genes for functional noncoding RNAs, including tRNA, rRNA, snRNA, miRNA, and lncRNA. These RNA molecules play important roles in a wide range of biological processes, including the regulation of gene expression, RNA processing, and translational regulation. The functional diversity of RNA also comes from more than simply its ability to form complex three-dimensional structures. RNA can change its structure in response to a variety of cellular conditions and signals, including proteins, small-molecule ligands, metal ions, changes in temperature, and changes in pH. This structural flexibility and dynamic behavior make RNA an important molecule capable of carrying out sophisticated regulatory functions. In the next article, as we explore nitrogenous bases, we will take a closer look at hydrogen bonding, which helps stabilize RNA and DNA structures, as well as the equally important principle of base stacking.
RNA’s high reactivity may have been a major advantage in the environment of early life, allowing a wide variety of structures and reactions to arise. But once life needed to preserve larger amounts of genetic information more reliably and for longer periods, that same reactivity may have become a limitation. A more stable molecule, DNA, may have overcome this limitation and taken over long-term information storage, while proteins—with their ability to carry out a far wider variety of chemical reactions rapidly and precisely—came to perform most catalytic functions. In this way, these molecular roles may gradually have become divided among DNA, RNA, and proteins. Even so, RNA continues to perform essential roles and functions throughout life today.
[References]
[1] https://www.britannica.com/science/ribosome
[2] RNA Oligomerisation without Added Catalyst from 2’,3’- Cyclic Nucleotides by Drying at Air-Water Interfaces
https://research-explorer.ista.ac.at/download/17078/17355/2023_ChemSystemsChem_Dass.pdf
[3] High-Yield Prebiotic Polymerization of 2′,3′-Cyclic Nucleotides under Wet–Dry Cycling
doi: 10.1021/acscentsci.5c00488
[4] RNA Stability: A Review of the Role of Structural Features and Environmental Conditions
https://doi.org/10.3390/molecules29245978
[5] Pseudoknots: RNA Structures with Diverse Functions
https://doi.org/10.1371/journal.pbio.0030213
[Image sources]
1. Le Vay K, Mutschler H. The difficult case of an RNA-only origin of life.
https://doi.org/10.1042/ETLS20190024, CC BY 4.03. Cev455. RNA Hydrolysis.
https://commons.wikimedia.org/wiki/File:RNA_Hydrolysis.png, CC BY-SA 3.0
4-1. Mamuye AL, Merelli E, Tesei L. A Graph Grammar for Modelling RNA Folding.
https://arxiv.org/abs/1612.01639, CC BY 4.0
4-2 Legendre A, Angel E, Tahi F. Bi-objective integer programming for RNA secondary structure prediction with pseudoknots.
https://doi.org/10.1186/s12859-018-2007-7, CC BY 4.0
4-3 Staple DW, Butcher SE. Pseudoknots: RNA Structures with Diverse Functions.
https://doi.org/10.1371/journal.pbio.0030213, CC BY
5. Petrov AS, Bernier CR, Gulen B, Waterbury CC, Hershkovits E, et al. Secondary Structures of rRNAs from All Three Domains of Life.
https://doi.org/10.1371/journal.pone.0088222, CC BY 4.0
