Letter from Rosalind Franklin to H. Fraenkel-Conrat
During her 1954 visit to the United States, Franklin met or renewed acquaintance with many virus researchers, who were able
to send samples of their virus material and brainstorm with her on her x-ray diffraction findings. Heinz Fraenkel-Conrat at
the UC Berkeley Virus Laboratory was especially helpful, sending a variety of heavy-atom substituted TMV preparations. In
this letter Franklin reported on her latest work with mercury-substituted TMV, which indicated a very different number of
protein sub-units than previously believed.
Number of Image Pages:
2 (117,302 Bytes)
1956-01-07 (January 7, 1956)
[University of California, Berkeley]
Original Repository: Churchill Archives Centre. The Papers of Rosalind Franklin
Reproduced from the Franklin Collection at the Churchill Archives Centre with the permission of the copyright holder.
You may remember that you told me, in Brussels, that you found it difficult to get mercury on to all the sulphur atoms in
TMV. The first specimen you sent me had one mercury per 20,000 molecular weight. For the second you did not state the quantity.
Is it possible that there may be, in both specimens as little as one in 23,000? The reason why I ask this is that I now have
quite a lot of evidence that there are neither 31 nor 37, but 46 structural sub-units in three turns of the helix. The simplest
way of reconciling this with the chemical data is to suppose that every third structural sub-unit has an extra S and an extra
end-group associated with it. This extra S (1/4 of the total) might perhaps be that which is inaccessible to mercury.
I think I gave you, in an earlier letter, a preliminary estimate of 57 A for the radius of the mercury. From a quantitative
comparison of the scattering curves of TMV and Hg-TMV we now find the radius to be 56.0+/-0.5 A. It is the comparison of Hg-TMV
with TMV which has provided nearly all the evidence for the existence of 46 units in a repeat. I no longer believe that there
is any evidence that the chemical sub-units are related in pairs by diad axes in the structural units. On the contrary the
Hg-TMV data provide evidence that such diads cannot be present. (We were originally led to consider the existence of diads
on account of the apparent frequency of zeros of intensity, but improved X-ray photographs now show that true zeros of intensity
are, in fact, very rare.) It therefore seems that structural and chemical sub-units are the same thing, but that they are
probably not all strictly identical. The molecular weight of each comes out to be around 23,000.
I should be very interested to hear from you whether you know of any other evidence that the S or the end-groups are not all
equally accessible, or oare otherwise non-equivalent.
The iodinated TMV-Hg and TMV-I which you sent me are orientating very slowly, but will, I think, be good quite soon now. The
Hg-A protein was much more troublesome, and I have left only one capillary in which the birefringence has not entirely disappeared.
In this, orientation is proceeding extremely slowly, but I am still hoping that it will be good enough to be useful one day.