The Cell and its Nucleus on a human scale – V

Now that we have the nucleosome in hand, we’ve compacted cellular DNA down by about 1/9  (call it 10 fold).  But we’ve still got another 10,000 fold of compaction to get it all into the nucleus. What’s next?  It’s time to pause and consider that as structures in the cell get larger and larger, they get less regular and less regular. You can do Xray crystallography and solution NMR of  nucleosomes and get something believable.  Obviously the cell and its contents don’t form a crystal, so at some point we’ll have to stop assembling nice structures out of regular building blocks, as DNA was assembled from nucleotides, the double helix from two strands of DNA, the nucleosome from 146 elements of the double helix and 8 proteins. Sadly, that point has probably been reached at the next level of DNA compaction.

This is the infamous and much studied 30 nanoMeter fiber.  Recall that DNA is 2 nanoMeters in width, the nucleosome is 11 nanoMeters wide and 6 nanoMeters high and shaped like a cylinder.  So you can probably only pack 6 nucleosomes in a fiber  30 nanoMeters in diameter.  Call this another factor of 10 of compaction.  This still leaves over a thousandfold more DNA compaction to discover.  Going back to the linguini metaphor, the 30 nanoMeter fiber  (at 3/8 of an inch/2 nanoMeters) would be between 5 and 6 inches thick.  If all 15,000,000 feet of DNA were put into this structure, we’d have 15,000 feet of  6 inch rope in our 150 foot sphere. (To see where the pasta and the numbers come from, go back to the first post in the series).

Unfortunately “The structure of the compacted 30 nanoMeter fiber remains unresolved despite intensive effort” [ Cell vol. 128 pp. 651 -654 ’07 ]. Molecular Biology of the Cell 5th edition p. 217 has some nice pictures of two different models of the fiber.  But models they are.  Consider how you get to see the 30 nanoMeter fiber.  First, kill the cell to get the DNA.  Then swell the chromosomes in hypotonic buffer, then fix them, then dehydrate them with alcohol and then embed them into plastic, and then take an electron micrograph.  30 nanoMeters is 300 Angstroms, less than 1/10 of the smallest wavelength of visible light (4000 Angstroms) so you have to do all this to get a glimpse of them.

After years of effort by many workers, the following paper throws in the towel [ Proc. Natl. Acad. Sci. vol. 105 pp. 19732 – 19737 ’08 ].  They think the compacted chromosome in metaphase (which is thick enough to be seen by visible light and presumably made by mushing the 30 nanoMeter fibers together) is highly disordered, with NO regular arrangement of the nucleosomes.  The whole thing is said to resemble a polymer melt, in which the nucleosomes no longer interact with adjacent nucleosomes on the DNA helix (as they do in the 30 nanoMeter fiber), but with any nucleosome which happens to be nearby.   This fits with the ability of large proteins (topoisomerase IIalpha, condensins — don’t worry if you don’t know what they are, just accept that they are as large or larger than the nucleosome) to diffuse into the chromosome.  A truly tight regular structure would not permit this.

Has this sort of thing happened before?  You bet, and I got into it on day #1 of medical school back in 1962.  Those of you not interested in ancient history or philosophical musings on scientific error, can stop reading now and wait for the next post in the series, which will concern what must happen when DNA is read by chemical complexes far larger than the nucleosome.

I’d just finished two years of graduate work in organic chemistry in a department which eventually produced 6 Nobelists. When I was there they were doing the work that netted them the prize and definitely not resting on their laurels.  Back then electron microscopy of the cell was pretty new, and something called the unit membrane which bounds all eukaryotic cells had been described.  It consisted of two dark lines with a light line sandwiched between,  the whole business being about 70 Angstroms thick. The question at the time was did it represent lipid sandwiched by protein or protein sandwiched by lipid.  Well, I knew as much chemistry as anyone and I tried to figure it out.  One of the chemicals used to get the pictures was good old osmium tetroxide — which said vic-diol to me.  The more I read about fixing, dehydrating and embedding, the more I realized, that no chemist could possibly figure out what was going on.

Those of you knowing some cellular biology will realize that the whole thing was an artifact.  Yes, cells are bounded by the plasma membrane, but it is a lipid bilayer in which float proteins passing through (transmembrane proteins).  There are a few proteins attached to the extracellular surface only (the glypiated proteins), and a few bound to the intracellular surface, but not in the regular fashion seen on the electron micrographs back in the day.

So did anyone write a paper saying that the ‘unit membrane’ was an artifact?  No. People just began to ignore it, and stopped mentioning it.  Well not everyone.  I laid out $100 of the long green 4 years ago for “Basic Neurochemistry” based on fabulous reviews and there on p. 7 is the unit membrane.  I stopped reading at this point.

Pretty harmless.  Not in medicine though.  Here’s one example, but there are many, many more.  The late Michael DeBakey was a great man, a pioneer cardiac surgeon, teacher, medical statesman etc. etc.   His word was law in the profession.  He spent his life opening up narrowed or even occluded arteries (in the heart, or leading to the head).  There are four main vessels leading to the brain, two carotid arteries which carry most of the blood and two vertebral arteries. Narrowing of the carotid artery in the neck is fairly common and can lead to stroke. The carotid is quite accessible and is the pulsing artery you can feel just interior to the angle of the jaw (not too hard now! ).  DeBakey was the first to open a narrow carotid artery up in 1953.

He also said you could open up a completely occluded carotid artery to treat stroke. Surgeons all over the country tried it, but almost everyone who had the procedure died.  So they stopped doing it.  As far as I know, no paper ever appeared contradicting DeBakey.

Why this happened takes us pretty far afield, into brain metabolism etc. etc., and if anyone wants to know, I’ll put it in as a comment. 

It’s been fun socializing with old friends in the past month or so.  Look for the frequency of posts to increase.

Here’s a link to the next paper in the series

Unfortunately as of 3/14 it’s the last paper in the series — there will be more if I ever get to i.

Post a comment or leave a trackback: Trackback URL.


  • Yggdrasil  On April 24, 2010 at 9:19 pm

    I’m curious as to the author’s choice of cell line to study in the Eltsov et al paper you cite. They chose to study HeLa cells, a cell line derived from human cervical cancer. Notably, HeLa cells’ chromosomes are highly abnormal: they have a hypertriploid chromosome number, specific numerical deviations, 20 clonally abnormal chromosomes, and contain multiple copies of HPV sequences integrated at specific sites (Macville et al 1999. Cancer Research 59: 141-150. PMID:9892199). Perhaps the authors’ observations are simply an artifact of studying a cell line with very abnormal chromatin. Furthermore, some have suggested that 30nm fibers are part of euchromatin, which would not be present in mitotic chromosomes. Now, I agree that we don’t have enough evidence to say that 30nm fibers exist, but the PNAS paper you cite is not enough evidence to say that 30nm fibers don’t exist.

  • luysii  On April 26, 2010 at 1:58 pm

    Yggdrasil — Agree. HeLa cells certainly are NOT the obvious choice to study normal chromosome structure. Yet they seem to be the E. Coli of cell biology.

    A recent paper used them to study mitosis to a farethewell [ Nature vol. 464 pp. 684 – 685m 721 – 727 ’10 ]. Each of 21,000 known protein coding genes was silenced by small interfering RNA (siRNA). The cells were then watched for changes in mitosis using time lapse microscopy. It was a huge effort. There are 190,000 time lapse movies of 19,000,000 mitoses. Perhaps HeLa cells are used because they are so easy to grow.

    I think the PNAS paper just tried to provide another explanation of why, despite decades of effort, we don’t have an exact structure for the 30 nanoMeter fiber.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: