Oliver Smithies:

[00:00:00] This is the beginning of book D, capital D, 1960 still.  University of Wisconsin.  So it goes from the end of 1960 to the beginning of 1961 starting on Sunday November 20th.  Purifying haptoglobin still with a column that has — let’s see.  That it was 200 milliliters of 1-1, 2-1, and 2-2 were onto DEAE cellulose at pH 4 as usual.  Then 15 milliliters of [204 and 207?] split in mercaptoethanol acetone solution.  [00:01:00] The 2-2 was quite yellow on the DEAE, etc.  And sample was heated.  One sample was heated at 70 degrees.  All samples were diluted with acetic acid and sodium hydroxide.  And here is a beginning comment of what’s happening.  The gel is an 8 molar urea formic acid and starts with book C151 [pool?] and 1-1 and 2-2 samples with a comment that what was believed to be Hp2 is strong in 1-1.  [00:02:00] Therefore repeat with pure haptoglobins.  And the fingerprints confirmed that the material that we got out was not haptoglobin.  More column splitting tests on the following day.  Monday 21st.  Very little split product is visible.  Messy.  Beginning to have some decent images on page seven November 23rd.  Test of mercaptoethanol versus sodium [sulfite?] versus no splitting.  [00:03:00] This is with an acetone column.  And with a rather messy-looking gel.  Showing that splitting is incomplete in sodium [sulfite?] and even — question even complete in mercaptoethanol.  So leave overnight.  And other tests, other pH tests of the gels.  Formic acid gels.  No improvement over 0.05, 0.01 formic acid, 8 molar urea.  Some haptoglobin samples tested November 15th.  And overnight splitting now.  We’re back toward the overnight splitting result.

[00:04:00] On Friday November 26th quite complicated result which from current knowledge looks very unpromising.  But at the time it’s not sure.  The result showed that the Hp in quotation mark is not reduced.  [Only the?] impurities [and etc.?] so it was not a very clear result.  Azide [cup in?] tested on November 25th.  More extensive tests of the azide.  November 26th.  [00:05:00] Difficult to interpret.  Need to use better haptoglobin to test things.  Page 19.  First-class 2-2 is required for this experiment or better use purified Hp.  Hemoglobin interchange still possibility.  Testing on Sunday 27th.  No interchange is the result on Monday November 28th page 23.  Trying to get hemoglobin 1-1 — haptoglobin 1-1 and haptoglobin 2-2 to interchange and give haptoglobin 2-1 but it doesn’t happen.

[00:06:00] Some retesting of different materials.  Formamide.  Nonaqueous hemoglobin reaction, etc.  Let’s go back.  I’ll go back to [00:07:00] after page 23 which was showing no interchange then going to page 25 November — Monday November 28th began to think about nonaqueous hemoglobin reactions with haptoglobin to see if — whether this would affect the reaction.  The conclusion being this formamide hemoglobin does not bind haptoglobin.  So if the formamide had been treated with — if the hemoglobin had been treated with formamide it would no longer bind haptoglobin.  Also trying again acid chloride reaction with hemoglobin SOCl2 (inaudible) value.  [00:08:00] Continuing on November 30th.  Trying to CMC carboxymethylcellulose pH 7.  [00:09:00] But no particular joy.  Formamide retest again on page 41 December 4th.  Beginning to think about testing tannic acid-treated diethylaminoethylcellulose or [TEAC?] starch, etc.  So para-nitrobenzylcellulose being considered on page 45 December 5th.

[00:10:00] Cellulose trinitrate.  Nitrocellulose made with sulfuric acid and nitric acid and water and cellulose making guncotton.  Some comments on page 51 of doubtful value.  [FCOCL?] column, light cellulose nitrate column, heavy cellulose nitrate column.  Toronto benzyl nitrocellulose column, etc.  [00:11:00] There’s no indication that it was of any value (inaudible) [succinylcellulose?] page 59.  [00:12:00] Really trying very many variants.  Saturday December 10th decided to try to obtain quote heavy cellulose trinitrate again and then immobilize the hemoglobin chemically.  Attempted the nitration again.  The reaction got out of hand as the [sulcoflox?] would not wet.  Tried again with water.  This had been in water [sulcoflox?].  And said there the temperatures are too high.  Quite dangerous material.  [00:13:00] Testing them December 12th.  Page 71 for example.  May have to work in the cold, the doubly coated hemoglobin appears to go brown quite rapidly.  Made up small new lot and kept it room temperature and (inaudible) and in ice.  Hold up these experiments until I understand this better.  [00:14:00] Still trying cellulose [tetranitrile?] trinitrate with an ammonium salt.  Column was completely devoid of brown at the end of the experiment.  Clear pink but no haptoglobin [remove?] abandon this approach.  Page 77.  Testing para-nitrobenzylcellulose December 15th.

[00:15:00] Very heavy haptoglobin types on December 20th, page 95.  Very clear special sample from Elo Giblett of the Johnson modified and the 2-2 and 2-1.  That Elo Giblett sample was of course later on [00:16:00] proved to be very interesting.  December 21st trying to resynthesize 2-2.  Gordon Dixon and George Connell report that bisulfite HSO3 — that sulfite HSO3 has no effect on haptoglobin.  But sulfite plus chloromercuric ClHgCOO- gives some interesting reactions.  The reactions being the one that Rupert Cecil studied many years ago.  R — when — that I knew about when I was a graduate student.  That’s RSSR’ [00:17:00] plus HSO3 goes to RSH plus [SO3 minus?] SR’.  And parachloromercuribenzoate then can combine with the SH material to give RSHg benzene ring carboxylate.

Tested some of these ideas.  Still working on that idea on December 22nd page 103 testing with sodium sulfite and sodium sulfide.  But [00:18:00] left them together and with the result that there’s still no reaction left for several days of haptoglobin 1-1, 2-2, and a mixture of 1-1 and 2-2 with sodium sulfide and sodium sulfite.  But no images.  Says Friday December 23rd [sulfite?] parachloromercuribenzoate in alkaline borate buffer and so on.  [00:19:00] Considering the reaction of aminoethylthiol mercaptoethylamine.  That’s decided to try RSSR’ with S [parenthesis H?] EtNH3+ at pH 8.5 for splitting and resynthesis.  Beginning to realize that we might get a reversible depolymerization and resynthesize by the disulfide interchange.  And this later proved to be very possible and quite revealing and I had a pretty paper that is maybe being presaged in these pages.

[00:20:00] Page 113 sodium sulfide-treated 1-1 and 2-2 mixture and sodium sulfite-treated mixture.  And minimum treatment of 2-1 with mercaptoethylamine.  It appears to be not very much happening.  And the following pages show much the same sort of thing.  [00:21:00] And with the comment on page 114 this is sodium sulfide and sodium sulfite with copper.  Still no cleavage.  Repeat with mercaptoethanol as neither mercaptoethylamine or SH- are effective.  More column splitting tests on page 119 with a comment no differences, abandon column splitting for the present.  A wise decision.  December 30th, Friday, reductant tests.  Still testing different reductants.  This is testing mercaptoethanol at alkaline pH in borate buffer [00:22:00] at different concentrations, 0.5, 1, and 1.5 molar.  Those are the solutions that were made up, not tested.  They were tested at 0.02 molar as usual.  So two gels were run.  One was a borate buffer gel.  [00:23:00] The other was an 8 molar formic acid gel.  The borate buffer gel shows very little effect of the control SH- mercaptoethylamine.  Mercaptoethanol mercaptoethylamine mercaptobenzoic mercaptoacetate.  But to the — I’ll try this again.  Just stop for a minute.  On Friday December 30th pages 120 and 121 [00:24:00] there were two experiments going on, the reductant tests.  One is the concentration of urea 0.5 molar, 1 molar, 1.5 molar urea with hundredth molar mercaptoethanol in alkaline pH with the comment that some reaction in the zero increasing to about 20% in 1.5 molar urea.  The product looks clean and like the PCMB [sulfite?] product.  Migrates a little slower.  So this was lower urea than usual.  And then the other test was a test of without any urea.  Was a test of mercaptoethanol and mercaptoacetic acid, sodium sulfide, and mercaptoethylamine, etc.  [00:25:00] And shows very little effect although — that I can see — although the comment is that mercaptoethanol reaction is complete and is clean.  Mercaptoacetic to completion but not quite clean, etc.  I can’t make that conclusion anymore looking at the gels.  But that’s the conclusion that I wrote down.

[00:26:00] Page 123 has several gels looking at these various products.  Varying the concentration of urea on January 2nd, Monday.  Not obvious what these experiments were trying to do.  [00:27:00] Mercaptoethanol concentration tests on Tuesday January 3rd.  Varying the concentration of mercaptoethanol from 0.05 molar up to 0.4 molar.  In this case is [00:28:00] very clear reduction of the — that is to say depolymerization is really quite clear in 0.4 molar urea with — no, 0.4 molar mercaptoethanol in 0.1 molar borate without any urea present.  That the polymers are completely — it’s a complete depolymerization occurs.  So [00:29:00] there is a clear indication that one can get a stable product that appears monomeric just by reduction with mercaptoethanol and no urea.  So (inaudible) a two-stage process.  For example the comment on page 129 is the results show that 0.2 molar mercaptoethanol would be about ideal for maximum cleavage and minimum wreckage.  And the wreckage being the creation of the irreversible destruction of the haptoglobin polymers.

[00:30:00] This is really quite well shown on page 131 with a very promising result of haptoglobin being partly depolymerized.  It’s not completely depolymerized.  But the red crayon shows that there is an increase in the faster-migrating material and the polymeric material has disappeared.  So I am understanding that the haptoglobin can be dissociated into a monomeric form [00:31:00] that is still native protein.  And with a comment on page 133.  Note that low mercaptoethanol concentration convert haptoglobin 2-1 to haptoglobin 2-1 modified.  This is really a misinterpretation.  But it certainly looks that way.  Beginning to think that the reaction might be reversible.  This dissociation, depolymerization [00:32:00] page 135.  The results from the D133 experiment showed that 2-2 is especially prone to query repolymerization.  Actually detected as smearing.  So check with hemoglobin and repeat.  This is an attempt on Thursday January 5th to get that result.

And then on page 137 we come to a very beautiful pair of gels that may be even the ones that were used for publication.  Because Thursday January 7th all types splitting.  And they were dissolved in alkaline urea.  [00:33:00] Mercaptoethanol was added.  And then iodine — free iodoacetic acid was added and then the gel was run with 8 molar urea, formic acid, January 3rd gel.  And the result is really very beautiful.  I believe it’s the gel that was finally used in the publication but I will check that.  [00:33:31]

OS:

[00:00:00] Picking up the account on page 136 and 137.  Thursday January 5th.  These are tests of splitting all the types of haptoglobin that were then available.  And the gels are really remarkably high quality and are the forerunners of what we later published.  So page 137 there’s all types of splitting, 1-1, 2-1 modified, 2-1, and the Johnson type, and the 2-2 samples that had been previously purified were dissolved in 8 molar urea and mercaptoethanol added and iodoacetamide and then [00:01:00] the gel was run in 8 molar urea, 0.0501 formic acid, the standard acid gel.  And the results are really quite beautiful.  There were two images.  They’re both basically the same.  One is just trimmed copy of the other.  But the results show beautifully the 1F and 1S and 2 and the different type of gene 2 product from the Johnson phenotype.  Of course at this point we don’t yet know what all that means.  But the gels are really quite spectacular.  Although I wasn’t completely satisfied because I say on these later [sets?] the Hp2 is not symmetrical, presumably not, etc., etc.  [00:02:00] Not quite content ever with perfect gel.  But they really are — looking at them now from looking backwards they are very — it’s a very beautiful gel.  One which was formic acid was the January 3rd gel.  And one was — sorry.  January 6th gel.  And one was a January 7th gel.  I don’t know which the image is from.  But it’s perhaps understandable on the next page because this is particularly talking about 1F and 1S.  So by now because of George Connell’s work we really understood that haptoglobin — the 1-1 type really had three different forms.  [00:03:00] It could be 1F-1F or 1S-1F or [1S-1S?].  And so this is — this gel that was run on Friday January 6th was Hp1F versus 1S.  The gel there was actually the one that was published in George Connell’s paper in 1962, Figure 1 gel is exactly that one.  So it’s rather nice to see it in the notebook and in the paper at the same time.

The images on 136 and 137 were not published.  Those are — [00:04:00] that particular comparison was run in a later gel.  But anyway Hp1F versus 1S we clearly understand now that there are three forms of the gene, Hp1F, Hp1S, and Hp2 — oh.  Four forms of the gene, Hp1F, Hp1S, Hp2, Hp2 modified, and the Johnson type.  But don’t yet know what those differences are in terms of structure of the protein.  1F and 1S checked on the unsplit protein and not possible to tell the differences between them on a regular [00:05:00] 8 molar urea formic acid gel unsplit, the differences are not detectable in this gel.  Though later we did find a way of doing it.  But only — no.  Correction.  Not in the unsplit material.  More tests on hemoglobin binding of material.

[00:06:00] Some rather ineffectual tests on page 143 Saturday January 28th, 142 and 143, on some different hemoglobins.  [00:07:00] January 19th 145 Thursday some attempts to test SH transfer in 8 molar urea formate.  So haptoglobin 2-2 was treated in the usual way with 8 molar urea and mercaptoethanol and iodoacetamide and then various — with or without iodoacetamide.  And various checks were made of these mixtures with the comment that iodoacetamide has no effect.  [00:08:00] Some gamma globulin made for Lois Kitze.  Lois gamma globulin on page 147.

Back to thinking of polymerizing hemoglobin and [chelation?] of hemoglobin via polymers, carbodiimide.  Still trying to couple with bromoethylamine.  But apparently (inaudible) trying to use (inaudible) to make a cellulose with an [00:09:00] [imide C?] ethylbromide side chain.  Not obvious why.  No coupling.  Still no coupling.  Abandoned anyway.  And the end of this book is with solvent for fingerprint.  So I must have been thinking of beginning to do fingerprints myself.  Because I made [there this?] solvent.  Fingerprint.  I recollect that time trying to begin to do fingerprints and being very unhappy with even getting good fingerprints with the hemoglobins.  Not even getting good electrophoresis with hemoglobin.  So I did some work on getting a better buffer to run hemoglobin samples.  [00:10:00] And that probably will appear in the following books.  But here we’re ending this book with getting solvents ready for fingerprints and having trypsin and (inaudible) digest and digesting various materials or just it’s basically the recipe to trypsinize and digest or hydrolyze with HCl, etc. and [Paolo?] reactions and the different ways of detecting fingerprints.  So beginning to think about making fingerprints myself.  Book D is ended.  [00:10:54]