Your cells keep their genes in DNA and make short working copies of them in RNA. Both molecules are polymers: long chains built from repeating units called nucleotides. This section takes one nucleotide apart, joins nucleotides into a chain, and then pairs two chains into a double helix.
None of the chemistry is new. The base hangs from the sugar's anomeric carbon, just as the second sugar does in a glycoside. Phosphate esters join one unit to the next. Hydrogen bonds and stacking between flat rings hold two chains together. The information they carry belongs to biology, but every bond in them is one this course has already met.
Three pieces, two names
Each nucleotide has three pieces. A five-carbon sugar sits in the middle. On one side it carries a base, and on the other a phosphate. The base, often called a nitrogenous base, is a flat ring system containing several nitrogens. Take them one at a time, starting with the sugar.
RNA uses the sugar ribose. DNA uses 2-deoxyribose, which is ribose with the OH on C2 replaced by a plain H. "Deoxy" means "without an oxygen", and the 2 says which carbon lost it. That one oxygen is the only difference between the two sugars, and it gives RNA and DNA their names: ribonucleic and deoxyribonucleic acid.
In both polymers the sugar is closed into a five-membered ring, a furanose, with an oxygen as one ring atom. C1 is the anomeric carbon, as in the carbohydrates section earlier in this chapter. It was the aldehyde carbon before the ring closed, and it is the only ring carbon bonded to two oxygens.
Once a base is attached, the sugar's carbons are numbered with a prime: C1′ (read "C1 prime") through C5′. The base has numbered atoms of its own, and the primes keep the two sets apart. So "the 2′ OH" always means the OH on the sugar's C2, never an atom of the base.
Now attach a base, here adenine. One of its nitrogens takes the place of the OH on C1′. The base sits on the same side of the ring as C5′ (the CH2 group), which is the β position in the language of the carbohydrates section. Every nucleotide in DNA and RNA has its base in the β position.
The new bond joins the anomeric carbon to a nitrogen, so it is called an N-glycosidic bond. It is a glycosidic bond with nitrogen in place of the second oxygen. C1′ keeps its bond to the ring oxygen and gains one to nitrogen, so it is still an acetal-type carbon. Like the acetals in Carbonyl Chemistry, the bond survives base. Warm aqueous acid cuts it, mainly when the base is adenine or guanine; the bond to the other bases resists acid far better.
A base joined to a sugar is a nucleoside. Adding a phosphate finishes the unit. The phosphate forms an ester with the OH on C5′, called the 5′ OH, and the nucleoside becomes a nucleotide.
The names follow one pattern. Adenine on ribose is adenosine, and on 2-deoxyribose it is deoxyadenosine. One phosphate on adenosine's 5′ OH gives adenosine monophosphate, AMP. A chain of three phosphates in the same place gives adenosine triphosphate, ATP. The deoxy versions are dAMP and dATP, and the other bases follow the same pattern:
| Base | Nucleoside with ribose (RNA) | Nucleoside with 2-deoxyribose (DNA) |
|---|---|---|
| Adenine (A) | adenosine | deoxyadenosine |
| Guanine (G) | guanosine | deoxyguanosine |
| Cytosine (C) | cytidine | deoxycytidine |
| Uracil (U), RNA only | uridine | — |
| Thymine (T), DNA only | — | thymidine (deoxythymidine) |
The bases, in two families
DNA uses four bases: adenine (A), guanine (G), cytosine (C) and thymine (T). RNA uses A, G and C as well, but uracil (U) takes the place of thymine. The five bases fall into two families by their rings.
- Purines have two fused rings: a six-membered ring that shares one edge with a five-membered ring. Adenine and guanine are purines.
- Pyrimidines have a single six-membered ring. Cytosine, thymine and uracil are pyrimidines.
A purine bonds to C1′ through its N9, and a pyrimidine through its N1. Thymine and uracil differ by a single methyl group on C5: thymine has it and uracil does not. Each base is flat. Every ring atom is sp2, and the rings are aromatic, so the ring atoms and the atoms bonded to them all lie in one plane. That flatness matters when the bases stack up inside the double helix.
The backbone is a chain of phosphate diesters
Now join two nucleotides. The phosphate on the 5′ OH of one nucleotide forms a second ester, this time with the 3′ OH of the nucleotide before it. One phosphorus then carries ester links to two different sugars. That group is a phosphodiester, and the link runs C3′–O–P–O–C5′.
Repeat the link and you get the backbone of the chain: sugar, phosphate, sugar, phosphate, with a base hanging off every sugar.
Each phosphodiester still has one O–H on its phosphorus, and that proton is strongly acidic, with a pKa of about 1. At the pH of a cell it is gone, so every phosphate in the backbone carries one negative charge. Losing those protons is what makes them acids. It also makes DNA and RNA polyanions, ions with many negative charges along their length. The charges left behind are why they dissolve well in water and why they move toward the positive electrode in an electric field. Gel electrophoresis, which separates pieces of DNA by size, relies on that movement.
The two ends of a chain are different. At one end, the 5′ carbon is not joined to another nucleotide and carries an OH or a phosphate. At the other end, the 3′ OH is free. A chain therefore has a direction, and a sequence is written from the 5′ end to the 3′ end by convention. A peptide works the same way, written from its N-terminus to its C-terminus. Reversing the letters describes a different molecule: 5′-AG-3′ and 5′-GA-3′ are two different dinucleotides.
How a chain grows
ATP is a nucleotide that does chemistry instead of storing information. Its three phosphates sit in a row on the 5′ oxygen. The first is joined to the sugar by an ester. Each of the other two is joined to the phosphate before it through a shared oxygen, P–O–P. That link is an anhydride: two acids joined through one oxygen, the phosphorus version of the anhydrides in Carboxylic Acids & Derivatives.
The anhydride link is the reactive one. When it breaks next to the first phosphate, the other two leave together as diphosphate. Diphosphate carries its negative charge spread over many oxygens, and spread-out charge makes a weak base, just as it did for a carboxylate in Acids & Bases. A weak base is a good leaving group.
An enzyme called a DNA polymerase builds DNA with this reaction. Its building blocks are the four deoxynucleoside triphosphates, or dNTPs: dATP, dGTP, dCTP and dTTP. The 3′ OH at the end of the growing chain attacks the first phosphorus of an incoming dNTP, the one bonded to its 5′ oxygen. Diphosphate leaves, and the chain is one nucleotide longer. Each new nucleotide is added at the 3′ end, which is why chains grow 5′ to 3′.
RNA's extra OH, and why base cuts RNA
Warm dilute base breaks RNA down into single nucleotides, while DNA comes through the same treatment intact. The two backbones are the same phosphodiesters, so the difference has to be the one group RNA has and DNA lacks: the 2′ OH.
Base removes the proton from the 2′ OH, leaving an alkoxide, O−. Three atoms separate that oxygen from the phosphorus on the same sugar's 3′ oxygen: C2′, C3′ and O3′. When the 2′ O− attacks that phosphorus, it closes a ring of five atoms: O2′, C2′, C3′, O3′ and P. The 5′ oxygen of the next nucleotide leaves, and the chain is cut. The ring left behind is a 2′,3′-cyclic phosphate.
The attack is fast because it is intramolecular: the nucleophile belongs to the same molecule and is held right beside the phosphorus. Hydroxide from the solution would have to meet the phosphorus by chance, and it reacts enormously more slowly. In base the cyclic phosphate then opens as well, so the final products are a mixture of nucleotides with the phosphate on either C2′ or C3′.
DNA has no oxygen at C2′, so it has no internal nucleophile, and its phosphodiesters hydrolyze extremely slowly. The difference suits what each molecule does. DNA holds genes for the life of a cell and passes them on when it divides, and its backbone lacks the group that would cut it. Many RNA copies are used for a short time and then broken down, and their backbone keeps that group.
Two strands, running in opposite directions
In a cell, DNA is usually two strands wound around each other in a double helix. To see how the strands are arranged, picture the helix untwisted into a ladder. The two sugar–phosphate backbones are the rails. They sit on the outside, where their charged phosphates meet water. The bases point inward and pair across the middle, and each pair is one rung.
The two strands run in opposite directions, an arrangement called antiparallel. Read one rail from top to bottom and you go 5′ to 3′. Read the other rail the same way and you go 3′ to 5′. In this arrangement the bases of the two strands face each other with their pairing edges lined up.
Base pairing: hydrogen bonds choose the partner
Each base pairs with one partner only. A pairs with T, and G pairs with C. In RNA, U takes the place of T and pairs with A. What decides the partner is hydrogen bonding. A hydrogen-bond donor is an N–H or O–H whose H can point at a lone pair on a nearby atom. A hydrogen-bond acceptor is the N or O that supplies that lone pair. Two bases pair when the donors on one line up with the acceptors on the other.
A and T line up two such donor–acceptor pairs, and G and C line up three.
| Pair | Hydrogen bonds | Note |
|---|---|---|
| A–T | 2 | In RNA the pair is A–U |
| G–C | 3 | The stronger pair |
Every pair is a purine with a pyrimidine. So every rung has the same width, and the helix keeps the same diameter along its whole length. Two purines would be too wide to fit between the backbones, and two pyrimidines too narrow to reach across.
Pairing also explains Chargaff's rules, which were measured years before anyone knew the structure. In any double-stranded DNA, the amount of A equals the amount of T, and the amount of G equals the amount of C. Every A has a T across from it, and every G a C, so the counts have to match.
One DNA strand reads 5′-ATGC-3′. Give the complementary strand.
Pair each base. A pairs with T, T with A, G with C and C with G, which gives TACG.
Get the direction right. The strands are antiparallel, so TACG, read alongside the original, runs 3′ to 5′. Written the conventional way, 5′ to 3′, the answer is reversed: 5′-GCAT-3′.
The reversal is where these questions are usually lost. Pairing the bases is mechanical; the real test is remembering to write the answer in the other direction.
Stacking: what else holds the helix together
Hydrogen bonds choose the partner, but they are not the only force holding the two strands together. The flat base pairs lie one on top of the next up the middle of the helix, like a stack of coins. Neighboring pairs attract each other through London dispersion forces between their closely packed ring faces, and the stack keeps those faces out of water. This base stacking contributes more to the stability of the helix than the hydrogen bonds between partners do. Part of the reason is that each base was already hydrogen-bonded to water before it paired, so pairing gains less than the bonds themselves suggest.
Heating double-stranded DNA until its strands separate is called melting. A stretch rich in G and C melts at a higher temperature than one rich in A and T. Most courses explain this with the hydrogen-bond count: three for each G–C pair against two for each A–T pair. Stacking points the same way, because stacks that contain G–C pairs are also more stable. Both effects make GC-rich DNA harder to pull apart.
When a base moves one hydrogen
Look at thymine's C4. It is double-bonded to an oxygen, and the N3 beside it carries an H. An N–H next to a C=O inside a ring is a cyclic amide, called a lactam. Guanine and uracil have one too.
A lactam has a tautomer, an isomer that differs only in where one H sits and where one double bond is, like the keto and enol forms in Enolate Chemistry. Move the H from N3 onto the oxygen of C4, and the C=O becomes C–OH while the C–N becomes C=N. This form is the lactim. Adenine and cytosine have a similar option: an H from the NH2 can move onto a ring nitrogen, which turns the amino group, C–NH2, into an imino group, C=NH.
The equilibrium lies very far toward the lactam and amino forms. Those are the forms drawn everywhere on this page, and their donors and acceptors give the standard pairs.
The rare form has a donor and an acceptor in swapped places, so it pairs with a different partner. Along its pairing edge, the lactam of thymine reads acceptor, donor, acceptor, which matches adenine. The lactim reads donor, acceptor, acceptor, which matches guanine. If a T is in its lactim form at the moment DNA is copied, a G can be placed across from it. When that G is copied in turn, it pairs with C, and the DNA now has a G–C pair where an A–T pair used to be. A change of one base pair like this is a point mutation.
Watson and Crick suggested rare tautomers as a cause of copying errors in 1953, and this is still the model most courses teach. Other kinds of mispairing are known as well, so rare tautomers are one source of point mutations rather than the only one.
Why weak interactions suit the job
Hydrogen bonds and stacking are both weak. A single hydrogen bond has only a few percent of the strength of a covalent bond, and ordinary thermal motion at body temperature is enough to break one.
That weakness is useful. To be read or copied, the two strands must come apart and later come back together. If covalent bonds held them, the information could not be opened without breaking the molecule. Thousands of weak interactions give a helix that is stable as a whole, yet an enzyme can open a short stretch without breaking a single covalent bond. Protein folding rests on the same idea (see the peptides and proteins section earlier in this chapter): interactions that are weak one at a time can be decisive in bulk.
What carries forward
This is the last section of the biomolecules chapter, and it sums up what the chapter was for. A nucleotide is a sugar holding a nitrogen-containing base through an acetal-type bond, with a phosphate ester on its 5′ carbon. A double helix is hydrogen bonding, stacking and shapes that fit. None of it needs new chemistry. It needs the chemistry you already have, applied to molecules that happen to carry life's information.