Klug here reported on recent conferences devoted to DNA sequencing and structure, and described the critical response to evidence
that chromosomal DNA formed a supersolenoid--a coiled cylindrical structure made of smaller such coiled structures--presented
by Crick's two co-authors of a recent paper on chromosome structure, Arne Leth Bak and Jesper Zeuthen. Klug concluded
that "if your name hadn't been attached to it [the idea of a supersolenoid], it would probably have been disbelieved."
Item is a photocopy.
Number of Image Pages:
4 (541,331 Bytes)
1977-03-11 (March 11, 1977)
Original Repository: Wellcome Library for the History and Understanding of Medicine. Francis Harry Compton Crick Papers
Thank you for your letters of 17 January and 23 February. I am sorry I have not written earlier, but I am only now beginning
to surface after my visit to the States and the meetings in London.
Firstly, about your note on Len's work. I have discussed the various points you raised in detail with Len and he is going
to do the controls you mention. He is away for two weeks and will be doing them immediately on his return. In the meantime,
we do not understand your statement that the removal of nicks by the enzyme might fuzz his resu1t. In any case he plans to
repeat the experiment in which the nicks have been earlier sealed with ligase. As for the point about a single band moving
down in toto, he cannot of course detect a small level of upward movement but he is pretty confident that almost all does
I have already written your paper with Bak and Zeuthen. Bak described the matter quite clearly at the Royal Society, but did
put forward their picture of a double helix for the final level of folding. Callan answered him firmly but politely that he
could not believe that the telomere turns back and tucks into the centromere. Zeuthen spoke at the EMBO meeting where there
was much more time and he was more cautious in his approach on this point. Indeed I found Bak rather fixed in his ideas and
Zeuthen altogether more open to argument, and I think he is the brighter man. I had raised the question in public as to whether
the opening up in 1 M hexyleneglycol could be an artifact, but only later, in private conversation, did Bak say that they
had observed the phenomenon in other buffers. On the whole the response to the supersolenoid was one of caution and I think
that if your name hadn't been attached to it, it would probably have been disbelieved. Someone at the back at the Royal
Society meeting raised the relationship of Bak and Zeuthen's work to that of Stubblefield and Wray, presumably the same
point about which I wrote to you on 28 January. Bak's answer wasn't at all clear.
The meetings themselves went quite well. At the Royal Society Discussion there must have been 400-500 people present on both
days, with younger people sitting in the aisles, on the platform and perched all around. Zachau had nothing new, but Chambon
presented some results on reconstitution with H3 and H4 only. This leads to particles of about 75-80 A diameter (compared
with 95 A for semisomes) and they seem able to organize 120-130 base pairs on their own, indeed rather like the results claimed
by Felsenfeld in another type of experiment. I saw Felsenfeld briefly in Boston who told me they had yet another paper in
press on this subject. Varshavsky turned out to be very impressive, although in his talk he showed so many gel patterns, mostly
unlabelled, that it became very difficult to follow. However, he came to the Lab. the following week with Engelhardt and gave
a much clearer seminar. He spoke mostly about three different types of monomers and the sub-nucleosomal fractions. There is
a wealth of detail and, rather than summarize it all, I am sending you a copy of his manuscript. I must say I don't know
what conclusions one can draw from it but there are some rather interesting associations of non-histone proteins with some
of the fractions. One generalization is that the smaller fragments of DNA seen to be associated with about one histone each.
Varehavsky speaks very good English and was altogether very impressive. I am glad we invited him but I rather fear that he
may run into trouble in the future since he is rather outspoken about conditions there and his own situation (Having blurted
quite a lot out over lunch here as though he were impelled to do so, he then asked us not to repeat anything in case the word
got back. He thinks he may not be able to come to the Cold Spring Harbor meeting because both Georgiev and Mirzabekov will
be going. I am writing a letter of thanks to Engelhardt in which I will end by saying I hope to see Varshavsky at CSH.
Roger gave an excellent talk. What he showed was there was no phasing, but the most striking result came out as a by-product.
He uses a combination of exonuclease 3 (which only removes one strand of double stranded DNA) plus S1 (which only acts on
single stranded DNA) to trim monomers and dimers down to a limit in separate experiments. The monomer band sharpens up beautifully
and this is the control. The dimer does not, so the conclusion is that there is a variable length of DNA between two different
nucleosomes. I regard this result as conclusive, and Roger pointed out that this explained why phasing would in any case be
impossible, although in the discussion people raised the question as to whether the phasing could be restored in the next
higher level of structure, say at the dodecamer level.
I suppose one could invent other explanations such as the two free ends of a UNA on the dimer corning close together and therefore
inhibiting trimming, but I prefer Roger's simple explanation. The rest of the meeting went very well I thought. Many of
the people had already been to two earlier meetings on the subject of DNA sequences (the Harden Conference last September
and one in Switzerland in January) but, to most of the audience, a lot of the results were new and very well presented. Hogness'
work in particular was spectacular but I assume you know all about that. There was a lot of discussion as to whether beads
are present on the DNA during transcription and Franke was adamant on the basis of e.m. evidence that they are not. Laird
was guarded. The situation wasn't made any clearer by Joel Gottesfeld's paper. He spoke very clearly and I must say
I really can't see what is wrong with the experiments now that he experiment seems to be that he can convert the 14S particle
to the 11S particle in the same tube by treatment with RNase. He is writing this up for Nucleic Acids Research and I will
make sure that you get a copy so that, even if it all turns out to be wrong, it won't be obvious at this stage.
I think the hit of the meeting was Ashburner who substituted for Brian Clarke who was ill. Ashburner went into a very detailed
analysis of the zest white region of drosophila and pointed out that there are strong reasons to dispute the one gene-one
band hypothesis. He points out that tests for saturation would be difficult or invalid (since the phenomenon didn't obey
poisson statistics), that George Lefebre had shown that many mutations were without macroscopic phenotype, that there was
at least one band which produced three enzymes and also (adding that he could practically hear you calling from California)
that the largest bands were probably too large for a single gene and that they were unlikely to contain special bands. I must
say that I hope there isn't a straw man involved here. He gave the impression that the one gene-one band hypothesis was
so sloppy as not to be worthwhile, but surely one can modify it to one band-one of genes operating together et sim. However
I must say that looking at some of Hogness' results there doesn't seem to be a lot of spare DNA in the genes he has
looked at, which are now 18S + 28S RNA, the histone genes and heat shock genes. FHCC: space between histone gene about twice
as long as coding sequence.
Worcel gave a good talk and essentially described his model which you had already alerted me to. It is rather clever but I
am beginning to think that clever conjectures by bright people, without any supporting evidence, are rapidly becoming the
bane of the chromatin field. What Worcel didn't talk about was the parallel between bacteria and drosophila but I tried
to make this clear in some remarks I made.
At the EMBO Workshop, most of the stuff was structural, and the only new thing was a model by Pardon and Richards who presented
a model to fit the neutron and scattering data in solution. This was a flat disk of about 50 A thick by 110 A in diameter
with the DNA right on the outside, two turns, about 30 A apart. It may not be such nonsense as some of the audience thought
because people are fixed on the idea of a spherical bead. The Workshop itself was not a great success because there were so
many wanting to speak that I simply decided to allow them 10 to 15 minutes each. There was a good thing by Keller (I enclose
a copy of his abstract). Bruce Ponder also described his work. This was the first time I had followed it in detail. I found
it amazing that he got the same pattern of eco R1 bands whichever cut of the original monomer material he took, i.e. the 140-170
base pair region or 170-200. Since one gets discrete bands in the eco R1 digestion, one seems forced to conclude that there
is a correlation between the alternative positions of the eco R1 site and the alternative lengths of the micrococcal nuclease
That's all for now. Sydney and Max have told me about your plans and although we haven't talked about it the word
seems to have got around.
I will write soon about progress with the crystals. We now have some evidence that the packing I described to you in my letters
of September and October is correct but we don't know what the units are that are being packed, either nucleosomes or
semi-nucleosomes. I hope to prepare an internal memo on the subject and will send you a copy. In the meantime, we are trying
to grow bigger crystals to measure the density, of which we only have limits.