Letter from Kai O. Pedersen to Michael Heidelberger
Twice, in 1934 and 1936, Heidelberger received a John Simon Guggenheim Memorial Fellowship for a stay of six weeks at the
University of Uppsala in Sweden, where he used the first ultracentrifuge, constructed by Theodor Svedberg and Kai O. Pedersen,
to determine the molecular weight of various proteins, including, during his second visit, antibodies in horse antipneumococcal
sera. In experiments completed by Pedersen after Heidelberger's departure and discussed by Pedersen in this letter, the
two found that these antibodies were of high molecular weight, another indication that they were, in fact, proteins.
Number of Image Pages:
4 (300,276 Bytes)
1936-10-20 (October 20, 1936)
Pedersen, Kai O.
Reproduced with permission of Lennart Pedersen.
Medical Subject Headings (MeSH):
Antigens and Antibodies: Heidelberger and The Rise of Quantitative Immunochemistry, 1928-1954
[Handwritten note:] Do you think I may get a copy of the MS when you have sent it to the editor. I am very anxious to know
the improvements you are going to make and to hear what you think of the final experimental results. K.O.P.
Dear Professor Heidelberger.
Yesterday I mailed the MS to you (as registered letter) and this morning we have sent your handwritten MS together with a
summary of the Rabbit and Horse Sera runs beside a copy of the two most characteristic sedimentation diagrams from each run
(this was mailed as registered "affarshandlingar" this morning). I hope you will get these at the same time as this
letter, otherwise they must come shortly after. For the case you have not got the first letter I add a copy of this.
As regards the MS I reckon that you are going to bring it into its final shape, so I have not at all bothered about getting
it into the correct language, and I hope that you will change it in all the ways which you think advisable. - The MS has come
out rather long, but, I think it difficult to shorten it much without, excluding things that are valuable for the understanding
of the experimental results. I was a little in doubt about the best way of stating the experimental results, since I was afraid
that the tables would not give a correct picture of' the centrifuge experiments. Finally I arrived at these diagrams,
and I think that they really give a fairly good picture of the results. Svedberg and Tiselius said that it was a great advantage
for them to have them, when they were reading the MS. - As to the typographical performance of these diagrams they will have
to be rather much dimished, probably at least to 1/3 scale. The lines for the rectangles must be heavily drawn in order that
they may be easily distinguished from the text. It is probably best to print the text in the diagrams. The diagrams I send
you are just rough-draughts in order to save time, if I should have got the fine dawn done it would probably have meant a
delay of another week, since our drawer from now until Christmas is piled up with work, but I hope you can get it done easily
in New York.
I have been a little afraid whether the MS should be more than the 25 printed pages they state as a maximum in J. Exp. Med.
But do you not think that all the description of the experiments could be printed with "petit", I think so. If it
is still too long, do you think it would be possible to divide it into two papers? The one dealing with the rabbit experiments
and the other with those done on horse sera.
You will probably observe that the new sedimentation constants given in several cases are different from the ones calculated
when you left. Slight changes may be due to the fact that we have recalculated most of the runs and in some cases we have
got additional exposures comparated. But the main cause of change is due to the additional correction which I have applied
on these experiments. It is a very difficult problem in what manner to correct the sedimentation constants in the mere concentrated
solutions and in the mixtures (f .inst . serurn). I am working on the problem and hope to come to some conclusion sooner or
later, but so far I have not got any definite results. In lack of results I have used the same method as v. Mutzenbecker used
for correcting his sedimentation constants for serumalbumin and serumglobulin to infinite dilution, viz. by multiplying the
(for salt and temp. corrected) s20 with the viscosity increase due to the protein itself. That means I have multiplied the
salt-corrected s20 by protein/buffer solution where protein is the viscosity of a protein solution having the same concentration
and composition as the centrifuge solution has where the molecule corresponding to this s, sediments; or in another way: s
has been corrected for the protein concentration c1, s2 for protein concentration c1 + c2 and sN has been corrected for the
protein concentration c1 + c2 + . . . + cN. The values for ___ for cvalbumin, serumalbumin and serumglobulin have been taken
from determination of the viscosity in pure solutions of these proteins, whereas the ___ values for the faster sedimenting
substances have been put equal to the values for serumglobulin.
The ____ values for s20 are the values which are corrected in the way just described, and these values are given in the diagrams.
In my earlier letters I have not had time to answer some of your questions in your letters of September 4. and September 13.
I shall try t do this now.
First so very many thanks for the nice blue thermos bottle which was used for sending the serum samples from London to Upsala.
We have used it already several times and enjoyed it very much; it followed us on a weekend trip in the canoe in the middle
of September. As to the 10 sh stamp, I had given it to my mother-in-law before I got your letter, where you asked me what
Mr. Lonn thought about it. My mother-in-law is very glad to get stamps, so we always collect our stamps for her.
As to the runs with the polysaccharide (expt.23 and 28), the result could be summarized as follows: In the experiment with
the 1% solution (23) we got a very sharp peak, whereas in the dilute solution (expt.28) the sedimentation started with a peak,
but this peak became broader and broader and almost disappeared after a while. Even with the very scale distances it was
impossible to get any definite peak on the curves. This behavior is very typical for the "threads" molecules. Even
in very polydisperse solutions one gets only a single and sharp boundary, when the concentration of the thread molecules is
not very dilute. This state of affairs has been treated by Signer and Gross (Helv. Chim. Acta 17, 59, 335 and 726 (1934))
for the polystyrol systems. The polysaccharide is therefore after all quite polydisperse. I wonder whether an osmotic pressure
measurement will tell so very much in this case, since the polysaccharide undoubtedly must be very polydisperse, and I am
afraid the osmotic pressure mean value for M even may come out with the wrong order of magnitude, the osmotic pressure measurement
often aces with the thread molecules. A diffusion experiment will be of little value too in this case.
It is very interesting that our experimental results show that we chose the right thing, when we decided to carry out the
experiments in moderate dilute solutions. If we had carried out the experiments in concentrated solutions as McFarlane would
suggest, then I am afraid that the result of the whole investigation would have been very meager. I have often told him that
it is wrong to stick to a single concentration.
I think it must be a mistake when you speak about the s20 = 2.8 * 10 ^-13 for EA in Experiment 21. The value is quite normal,
but on the whole the values for the EA are not so very sure, because we often did not run the centrifuge long enough.
I add some proposals for text for the Figures on a separate sheet.
I have had piled up with work the whole autumn, and it has been even worse after the return of Professor Svedberg from U.S.A.
He was very enthusiastic about his journey. He was very interested in the work done with the airdriven centrifuge at the Rockefeller
Institution, and I think that perhaps there will be some sort of cooperation between the two institutes. Dr. Baur is probably
going to write a chapter about the airdriven centrifuge in the centrifuge monography.
With the kindest regards to you, Mrs. Heidelberger and Charlie from Ida and
Kai O. Pedersen
[Handwritten note:] P.S. I should ask you to order 200 reprints for our institute (150 for the lab 50 for me) and perhaps
you could send us a list with the names of the people you are sending reprints to yourself in order that we are not sending
to the same persons.