2.1 B-form DNA #
2.1.1 The dimensions that matter #
Under physiological conditions the usual structure is B-DNA: a right-handed helix, strands antiparallel, bases inside, phosphates outside. Adenine pairs with thymine by two hydrogen bonds, guanine with cytosine by three, so GΒ·C-rich DNA melts less readily.
Two numbers are used for the rest of the course. The rise is about 0.34 nm per base pair. There are about 10.5 base pairs per turn in solution; fibre diffraction gave the older figure of 10.0, which still appears in problems. One turn therefore advances about 3.6 nm, and the familiar “3.4 nm pitch” is the same measurement rounded from 10 bp per turn. The helix is about 2 nm wide. A haploid human genome of about 3.1 Γ 10^9 base pairs is about a metre of DNA, and the diploid nucleus holds about two metres. coli* chromosome of about 4.6 Γ 10^6 base pairs is about 1.5 mm of DNA inside a cell one or two micrometres long. Packaging is why chromatin and supercoiling exist.
A-form helices are shorter and wider. RNA duplexes and DNAβRNA hybrids adopt them, so a transcription bubble is not B-DNA. It is not the default archive.
2.2 Supercoiling and topoisomerases #
2.2.1 Linking number, twist, and writhe #
In a closed circular duplex, or a linear duplex whose ends are held, the strands cannot be separated without a break. The linking number Lk counts how many times one strand winds about the other. The relaxed value Lk0 is the number of base pairs divided by about 10.5. Supercoiling is ΞLk = Lk β Lk0, and the superhelical density Ο = ΞLk / Lk0 is about β0.06 on the *E.
Negative supercoiling favours strand separation, because melting absorbs some of the underwinding. Initiation of replication and of transcription uses that help.
2.2.2 Topoisomerase I and topoisomerase II #
Topoisomerases change Lk by a transient covalent bond between an active-site tyrosine and a DNA phosphate. The energy of the backbone bond is stored, so the break can be resealed without a separate ligase.
| Type I | Type II | |
|---|---|---|
| Break | one strand | both strands |
| Change in Lk | steps of 1 | steps of 2 |
| ATP | not required for breakage and reunion | required |
| Bacterial roles | topo I relaxes negative supercoils | DNA gyrase introduces negative supercoils; topo IV decatenates daughters |
| Eukaryotic roles | topo IB relaxes both signs | topo II relaxes and decatenates; there is no gyrase |
Bacteria keep DNA underwound with gyrase. Eukaryotic topo II does not introduce negative supercoils; nucleosomes supply an equivalent wrap. Fluoroquinolones inhibit bacterial gyrase and so stop replication and transcription. They are not a general chemical destruction of DNA. After replication, circular daughter molecules are catenated. Without that reaction the products cannot be segregated.
2.3 Nucleosomes #
The study of Nucleosomes provides fundamental principles governing cellular architecture, molecular interactions, and physiological regulation.

Nucleosome Structure And Subunits
2.3.1 The histone octamer and about 147 bp #
The nucleosome core is an octamer of two copies each of histones H2A, H2B, H3, and H4. They are small and basic, which is why they bind the acidic backbone. About 147 bp of DNA wraps the octamer in about 1.7 left-handed turns. Histone H1 binds the linker where DNA enters and leaves; H1 is not part of the octamer. Linker length varies, often a few tens of base pairs, so the nucleosome repeat, core plus linker, is often about 180β200 bp. “200 bp per nucleosome” means that repeat, not the 147 bp on the octamer. At low salt the chain looks like beads on a string, about 10 nm across.

Histone Modifications
Each nucleosome is a local negative supercoil. Strip histones from a closed circular DNA that was assembled into nucleosomes and the linking number shows the underwinding. Bacteria do not use this particle. Their nucleoid is compacted by supercoiling and by small proteins such as HU, IHF, H-NS, and Fis, which bend or bridge DNA but do not form an octamer. The centromere is marked by the H3 variant CENP-A; the kinetochore built on it is cell biology.
2.3.2 The 30 nm fibre, and what a living nucleus does instead #
A solenoid or a zigzag of nucleosomes, about 30 nm wide, is what chromatin can form in vitro and in some fixed preparations. Both models describe that material. They are not a good description of most chromatin in a living nucleus.
Cryo-electron microscopy and methods that measure which sequences lie near one another show irregular clusters of nucleosomes and loops, with a density that changes as genes turn on and off. Higher-order folding in vivo is dynamic. It is not a single static 30 nm fibre. The negative claim is not: do not memorise the 30 nm fibre as a physiological polymer on a par with the nucleosome.
2.3.3 Euchromatin and heterochromatin #
Euchromatin is the more open fraction, enriched for genes that are active or able to be activated, and it tends to replicate earlier in S phase. Heterochromatin is compact, poorer in transcription, and often later-replicating. The contrast is a state of the same DNA, not a different chemistry.

Chromatin Structure
Constitutive heterochromatin stays compact in most cells: centromeres, telomeres, and long repeats. Facultative heterochromatin is conditional. A gene moved next to heterochromatin in Drosophila can be silenced in some cells and not in others (position-effect variegation). Neighbourhood, not only local sequence, sets the state. How the state is copied is Chapter 11.
201704 Chromatin Structure
2.4 Histone modifications #
The study of Histone modifications provides fundamental principles governing cellular architecture, molecular interactions, and physiological regulation.

Histone Stem Loop Structure Caenorhabditis Elegans
2.4.1 A code that recruits proteins #
Histone N-terminal tails carry modifications, mostly on lysines. The useful description is not “acetyl means on, methyl means off”.

Early Genetic Code
Acetylation, written by histone acetyltransferases and removed by deacetylases, neutralises the positive charge of lysine, so the tail grips DNA less tightly. It also binds bromodomains. Methylation does not remove that charge. Phosphorylation, notably H3S10 in mitosis, adds a further layer.
| Mark | Usually associated with | A reader or writer to remember |
|---|---|---|
| H3K4me3 | active promoters | PHD fingers; Set1/MLL-family methyltransferases |
| H3K36me3 | transcribed gene bodies | left in the wake of elongating RNA polymerase II |
| H3K27ac | active enhancers and promoters | bromodomains; p300/CBP |
| H3K9ac | open chromatin | bromodomains |
| H3K27me3 | Polycomb repression | Polycomb chromodomains; written by PRC2 |
| H3K9me3 | constitutive heterochromatin | HP1; Suv39-family enzymes |
| H3S10ph | mitotic chromatin | coincides with condensed chromosomes |
These are tendencies. A promoter can carry H3K4me3 and still be silent if H3K27me3 and a repressor dominate nearby. Calling the pattern a code is fair only in the sense that readers decode combinations. It is not a legend with one meaning printed beside each mark.
2.5 Key terms, summary, and self-check #
Key terms. B-form; rise; linking number; gyrase; nucleosome; H1; euchromatin; heterochromatin; bromodomain; writer and reader.
Summary. B-DNA rises about 0.34 nm per base pair and turns about once every 10.5 bp, so a diploid human nucleus holds about two metres of DNA. Negative supercoiling favours melting. Bacterial gyrase introduces it; eukaryotic topo II does not, and nucleosomes supply the wrap. The octamer binds about 147 bp. A static 30 nm fibre is an in-vitro description, not the dynamic, looped folding of a living nucleus. Euchromatin and heterochromatin are states. Acetylation loosens charge and recruits bromodomains; methylation mainly recruits readers, and the same class of mark means different things on different residues.
Self-check.
1. About how long is a haploid human genome of 3.1 Γ 10^9 bp, and which two B-form numbers fix that length?
2. Why does negative supercoiling help initiation, and why is eukaryotic topo II the wrong enzyme to credit with introducing it?
3. What is in the histone octamer, about how much DNA does the core particle wrap, and where does H1 sit?
4. Why should the 30 nm fibre not be drawn as the structure inside a living nucleus?
5. Why is “H3K9me3 off, H3K27ac on” too crude a legend?
—
—
—