Coiling bacterial DNA

By Grant Jacobs 24/08/2010

A chain of proteins hold bacterial DNA in a compacted spiral.

You and I are eukaryotes. Our cells have nuclei, repositories that contain our DNA and the proteins that read them to produce an RNA copy of them.

HeLa cells*** stained for DNA (Source: Wikimedia Commons.)
HeLa cells* stained for DNA (Source: Wikimedia Commons.)

In earlier articles, I’ve mentioned in passing how the enormous length of DNA in our cells is fitted into a nucleus. Our DNA, all 2 metres of it, were you to stretch it out end-to-end – is fitted within a nucleus with a diameter of roughly 6-10 micrometres, about one-millionth of a metre.

The trick is that a DNA molecule is very skinny – it’s only about 2 nanometers wide (2 billionths of a metre wide). Wrap that up around a something handy and it’ll be quite compact.

The ‘something handy’ in eukaryotes are histone proteins. Eight histone proteins associate to form a disk-shaped octamer, wrap DNA almost twice around it and you have a structure called a nucleosome.

(Source: Transcription and Translation.)
DNA wrapped around histone octamers to form nucleosomes. (Source: Transcription and Translation.)

Bacteria don’t have histone proteins. They don’t have nuclei either. Their DNA lies within the ’main’ compartment of their cells (the cytoplasm).

Their DNA isn’t as compacted as eukaryotic DNA, as we’ll see soon, but it is compacted. It’s not just floating loose in there.

In place of histones in some bacteria** is a protein unimaginatively called histone-like nucleoid structuring protein,*** or H-NS for short. (Some gene names can be quite fun, but many are more ordinarily named after what they are thought to be or do.)

This ribbon cartoon shows the Ava protein (3fse) has two domains. (Source: TOPSAN.)
This ribbon cartoon shows the Ava protein (3fse) has two domains. (Source: TOPSAN.)

Researchers have previously determined the atomic structure of parts of the H-NS protein from several species of bacteria. With this and a lot of experimental data, they’ve tried to figure out how it compacts bacterial DNA without coming up with a definitive answer.

Individual H-NS proteins bind together to form larger structures made of many H-NS proteins with the bacterial DNA attached to them. What’s needed is knowing how these proteins associate to form the larger complexes than bind the bacterial DNA.

Proteins are often made of several modules, or domains, each contributing a different part of the function of the protein.

Protein crystals grown in space (Source: Wikimedia Commons.)
Protein crystals grown in space (Source: Wikimedia Commons.)

One way of determining the three-dimensional atomic structure of a protein is to get a lot of highly purified protein, grow crystals of the protein and work out from how x-rays are scattered by the regular array of proteins what the structure of the crystallised protein must be. (It’s called x-ray crystallography if you want to read more. Another method to determine the 3-D atomic structure of a protein is NMR spectroscopy.)

It can be hard to make large amounts of pure proteins, or to crystallise them. Often instead of trying the whole protein, researchers will try just a single domain of a protein, or to leave out some bits in the hope of being able to grow a crystal.

Previous bits of the H-NS protein crystallised were the DNA-binding domain. H-NS also has a region that is important for it to assemble against another H-NS protein to form a two protein complex. This domain is the dimerisation domain. (Dimer = two-mer.)

H-NS dimer, from Arold et al. (2010)
H-NS dimer, from Arold et al. (2010)

Ladbury’s group (he is the senior author on the paper) have determined the structure of a portion of the H-NS protein (residues 1-83) that are able to associate with each-other to form dimers from a bacteria that can cause gastroenteritis, Samonella typhimurium. (You might know this particular gastroenteritis better as samonella infection, which caused by members of the Samonella genus of bacteria. It’s grotty.)

Using molecular modelling, they built models of how these dimers (shown above) might associate with each-other to form a continuous scaffold that DNA could attach onto (next three figures below).

Chained H-NS monomers, from Arold et al. (2010)
Chained H-NS monomers, from Arold et al. (2010)

Each dimer, two-mer of H-NS proteins, form not side by side facing the same way (parallel symmetrical dimers), but end-to-end from opposite directions ’anti-parallel’ to each other.

Superhelical coil of chained H-NS monomers, from Arold et al. (2010)
Superhelical coil of chained H-NS monomers, from Arold et al. (2010)

Chaining these together like a daisy chain forms a long connected chain of proteins.

If you think of the proteins as having a ‘head’ and a ‘tail’, two tails associate from opposite directions (‘Site 1’), then two heads associate (‘Site 2’), then two tails, and so on, to form a continuous series of proteins packed against eachother.

The angles that each individual protein (monomer, one-mer) associate means the chain forms a large spiral or superhelical coil, of about 190Å diameter, each turn of the spiral rising about 280Å (One Ångström is one ten-billionth of a metre.)

Model of H-NS compacting bacterial genome. From Arold et al. (2010)
Model of H-NS compacting bacterial genome. H-NS is showed in brown, with the DNA-binding domains in yellow. Bacterial DNA is grey. The blue amino acids are near the head (N-terminus) of the H-NS protein and the DNA-binding domains near the tail. You can see how the H-NS dimers alternative heads and tails, and how the DNA-bidning domains are on the outside in regular spacing bridging two DNA double-helices. The green (histones) and red (DNA) inset is to show how much more compact eukaryote DNA packing is by comparison. From Arold et al. (2010)
(From Thanbichler et al.)
(From Thanbichler et al.)

The proteins in Arold and colleagues’ crystals don’t include the DNA-binding domain, but other groups have worked out what the DNA-binding domain looks like. When the DNA-binding domains are placed on to their model of the spiral, they appear in regular steps on both sides of the spiral (figure above).

Bacterial DNA is a circle. For two DNA molecules to be bridged by this scaffold, they ought to be a circle folded into a spiral with the two adjacent DNA molecules running in opposite directions.

Thus is model nicely explains what is seen in experimental studies an electron micrographs of prokaryotic (bacterial) DNA.


* If you want to learn more about HeLa cells, try reading Rebecca Skloots book, The Immortal Life of Heriretta Lacks. (The stain used in the image is Hoechst 33258; DNA is stained blue. The cell to the left has compacted chromosomes and replicating.)

** In Enterobacteria

*** H-NS is not the only protein that can compact bacterial genomes. Others include FIS, HU, or polyamines (which are peptides rather than proteins)


Arold, S., Leonard, P., Parkinson, G., & Ladbury, J. (2010). H-NS forms a superhelical protein scaffold for DNA condensation Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.1006966107

Dame, R., Luijsterburg, M., Krin, E., Bertin, P., Wagner, R., & Wuite, G. (2005). DNA Bridging: a Property Shared among H-NS-Like Proteins Journal of Bacteriology, 187 (5), 1845-1848 DOI: 10.1128/JB.187.5.1845-1848.2005

Thanbichler, M., Wang, S., & Shapiro, L. (2005). The bacterial nucleoid: A highly organized and dynamic structure Journal of Cellular Biochemistry, 96 (3), 506-521 DOI: 10.1002/jcb.20519

Other articles on Code for life:

Epigenetics and 3-D gene structure

The scale of cellular life and compacting chromosomes

Temperature-induced hearing loss

Preserving endangered species – of gut microbes

0 Responses to “Coiling bacterial DNA”

  • Great article, but found a mistake: “6-10 micrometres, about one-hundredth of a metre”. Second paragraph.


  • Long story short: sometimes I get thrown out of WordPress and this ditches some writing/edits along with it. I had actually picked that up myself and corrected it while writing, but didn’t realise after it tossing me out that it’d ditched that edit. What a pain, it’ll mean that I now I have read it over *again* to check if there are other edits WP has dropped. Sigh

    (I can see others. This night is already long enough… double sigh)

  • Great post – with some lovely illustrations. I like histone-like proteins, they help to show that despite being a lot smaller and thought of as ‘simpler’ bacteria have similar problems to solve as eukaryotes do.

    By the way, I don’t know if you’ve done this already, but this would be a great post to submit to the MolBio carnival: We’re looking for submissions for the next edition on September 6th.

  • Thanks for the kind thoughts – I missed carnival that last time around. My only concern is that this article was really intended for those that don’t know molecular biology; those that do might find it is a little ‘lite’, even patronising – ? But if you think it’s appropriate, sure!

    This post is in fact part of a lead-in of sorts to another article I want to write about, so you might see that offered to as they relate in an interesting way. I need to find time first…

    I have a couple of older posts that might be possibilities, too.