Oliver Smithies:

So this is book seven, Roman seven, which starts in 1956, beginning again or continuing with the two dimension electrophoresis and bovine serum.  So pages one and two are just looking at some checks on the bovine sires, the sires of the animals that had been tested previously.  Testing starch still as usual, new BDH starch, one hour hydrolysis, Stein Hall starch against November 25th BDH starch with a comment that they were electrophoretically indistinguishable except for [00:01:00] gamma which is rather granular and streaky.  Therefore reserve old for very important tests in 2D work where strength is imperative.  So I still keep some on one side for the very important tests.  Comments on that side.  And Saturday, January 28th, a family again, mother and one year old twins.  Small volumes with difficulty.  Also hemoglobin present, not clear why.  A bit of difficulty with them.  Comments on starch again [00:02:00] and then on pages nine and 10 the breeds of the cattle and the numbers of the cattle, their designation numbers, and which type they belong to.  So it’s a total, I think, of 32 samples of which some were kept as standards, 32 sampled and three sires with the types.

Back to two dimensional electrophoresis.  Trying XXM buffer with narrow strips, just working out how best to make an experiment with the narrow strips.  [00:03:00] So gels were made and the two dimensional electrophoresis and still not perfectly happy with the strip electrophoresis, trying to make that work on pages 13 and 14, looking at strips with different concentrations of buffer.  No, different volumes of sample, from five microliters, 10 microliters, but I called it in millimeters 005 mil, 01 mil, 02, 005, 01, 02 mils to see how much I could put on, and concluding that I could use .01 mils [00:04:00] as 10 microliters for a strip.  Testing some more XXM buffer which is a standard buffer for electrophoresis, the composition of which I don’t remember.  A whole bunch of papers on that.  for some reason or other I tried two samples which were rather unsatisfactory and five samples which were quite good, but I make the comment that the results are once again disappointing, increase the temperature two degrees and run for a longer time.  Trying to get what I think are good samples.  (inaudible) to [00:05:00] me at this point.

Page 22, temperature 26 degrees and 27 degrees in the morning.  Dave Poulik’s serum, running the test strip between some big strips to keep the humidity up for electrophoresis that I get a result there that I quite like because there’s a check mark.  More tests of different filter papers.  Plus or minus a toxin.  Dave Poulik tests plus or minus toxin and toxide.  [00:06:00] Quite pleased with the result.  Friday, February 10th, page 25, 26 and February 11th.  Very promising result though looking at it here it doesn’t look very good.  Tried 1% arga [00:07:00] on Whatman paper, but not very satisfactory.  Family study, February 13th.  Two dimensional electrophoresis again with XXM buffer, but a good check mark on it, on the gel at 12:30.  Nothing very spectacular.  Family studies continuing.  Fraser family [00:08:00] and [name redacted] family.  Tuesday, March 13th, [Sascochak?] lipoproteins.  That’s Andrew Sascochak who was a colleague who worked in Hospital for Sick Children and with whom I had constantly been in touch at various times.  And he had some lipoproteins and I ran gels on his lipoproteins.  We’re doing this again March 14th, page 42 and page 41.  Wet New Brunswick aqueous HCL starch used for testing lipoproteins [00:09:00] as the background staining is so much less with that starch and slices perfectly.  I was happy with it.

We’ve got various results.  Family study the following day, March 15th.  Friday, March 16th, family studies continuing.  Oh, and here I have a note that I got Danish starch and this later on proved to be very important.  Danish acetone HCL starch.  The hemoglobin pattern is good on the Danish, possibly better than usual.  A notation [00:10:00] added in red crayon.  Even better so later on we used Danish starch completely.  Family study Saturday, March 17th.  This sample from Lewis positive or Lewis negative secreted and non-secreted to see if there was any correlation.  No particular good reason given.  No, that’s not correct.  [00:11:00] Let’s go back on that.  Saturday, March 17th.  These were salivas from secretors and non-secretors I was interested in, but don’t quite understand.  I’ll backtrack.  Saturday, March 17th, page 48 and page 47, father and mother [name redacted] with two children, two female 21 and 19 year old children.  And these are Lewis secretors and non-secretors and I wanted to check if the salivas are different — if the secretors are different.

So on Monday, [00:12:00] March 19th, I’m doing a rather normal family gels with the Danish starch.  With the salivas on page 48 salivas were prepared from [O acid?] David Scott and [Otto Hiller?] prepared by filtration through number 54 paper.  Watery fluids were produced.  [00:13:00] And on page 52 I have a sample of DAS saliva.  The starch was packed into a slot loosely, a rather wide slot, and then electrophoresis.  Actually I was looking partly to see if the saliva had any hydrolytic activity with respect to starch.  No enzymatic degradation noted in the starch, but some breakdown of the gel on the negative side of the slit.  Family study on Tuesday, March 20th, no comment.  And a big check gel [00:14:00] on page 56, March 23rd.  One, two, three, four, five, six, seven, 10, 11, 12, 35, 56, 20, 21 samples of 2A versus 2B just to confirm that they’re all right, cross checked and all are correct.

An example on Friday, March 23rd, of detection of anti A and anti B in group zero serum with an [00:15:00] antibody gel, but as usual I don’t think that’s very good and not terribly worthwhile.  So more bovine sample, Saturday, March 24th.  Monday, more bovine checks.  And here I have, this is the first time I mention his name on page 63, telegram received from CGH showing that number six is 2A and number eight is three, mistake was in the previous pedigree.  So I’m beginning to understand the inheritance [00:16:00] and CGH is Hickman, Charlie Hickman whom I never met and who was sending me these blood samples from cattle.  So this is the first time that I’ve put his initials that I know at this point.  So he and I are working on this quite carefully.  More lipoproteins from Andrew Sascochak, March 26th, and some of the results.  Don’t know that they ever led anywhere, but there they are.  Looking at beta lipoproteins again.  And by page 70 I’m looking at samples [00:17:00] again of twins and twin cattle and looking to see if we’ve assigned them.  So we have these four groups, one, two, three, and four, with a possible 2A again.

More lipoproteins.  Another family study coming.  Nothing very notable.  Comment here on the gels, as I do the family studies of April sixth, Friday, page 76, [name redacted] and [name redacted], lipoproteins [00:18:00] and zinc insulin all being done on the same day with the comment here that the gels are a bit weak and so albumin [backs?] too diffuse.  Repeat tomorrow, etc.  At this point 024 [molar bore?] is in the future.  Family studies, typical family, [name redacted] family.  Those who can taste PTC, I think it’s phenylthiocarbamide, I can’t remember exactly what PTC is.  We’ll check and find out.  A test on some PTC tasters, phenylthiocarbamide, [00:19:00] as to whether there was any correlation with one, 2A or 2B.  And have a no/yes thing and so no means not a taster and check means a taster and Miss Lewis was a type one and she was a taster and I was a type two and not a taster.  Similar things there.  DA Scott was a taster so there was no correlation between PTC and the one, 2A, 2B types.

Idaho potato starch tried on April 10th, Tuesday.  Tried scented starch box.  [00:20:00] Attempted to bind grains in an open mesh structure which would still have a wet strength.  This was for loading the sample because I’d been loading the sample with starch grains and got good results and I was trying to make a scented starch block that I could use.  Not very successful, I think.  Tested some rhesus monkeys on Thursday, April the 12th.  Considerable variations in the monkeys, but not like those in humans.  The hemoglobin had [00:21:00] very little effect on the fatness.  Don’t think I followed this up.  Family studies with different starches.  No, it was not a family study.  It was Friday, April 13th, page 92, just a test of different starches.  Six hour starch, five hour starch, four hour starch of Idaho aqueous HCL hydrolyzed starch.  And there’s my synthetic filter paper being tested on Friday, April 13th, with comments on the following page, 94.  [00:22:00] Synthetic filter paper experiments, Roman two, must use coarse grains.  Tested the usual method.  Too much clumping for separation, etc.

I got something that I thought was a beautifully porous, but the wet strength is rather low.  Seemed overdry.  Tried to get a meta method of inserting samples into the gel which later on I found a different method which was to use vertical starch gel electrophoresis, but I haven’t got there yet.  April 16th, Monday, [00:23:00] testing Danish potato starch versus Idaho with a comment that the Idaho is not as good as expected.  My fifth synthetic filter paper, very poor, must try starch [U?] again.  So synthetic filter paper wasn’t working very well.  [00:24:00] I quite liked the Idaho potato starch.  Still messing with starch.  Wednesday, April 18th, page 100, Merck 15% Idaho, 14% starches, strengthen texture very close.  And altering the buffer strength should make them the same.  Various hyperlipimic serum lipoprotein runs from [00:25:00] Sascochak.  Nothing remarkable commented on.

New nomenclature at last.  I’ve now realized that haptoglobin is the correct notation.  New nomenclature on page 105.  HP is the locus.  HPA and HPB for the two genes.  Haptoglobin one, haptoglobin type A is one, haptoglobin type AB is 2A, haptoglobin type B is 2B.  Later on we changed that to haptoglobin one and haptoglobin two.  As haptoglobin is a hemoglobin [00:26:00] binding protein of serum.  Thinking of HPA, HPAB and HPB for the shorthand.  Beginning to think of notation.  I must have at that point heard from Laurell, Carl Bertil Laurell, with the haptoglobin, that my proteins were probably haptoglobin of the type described originally by Max-Fernand Jayle.  I think I’ve talked about him before and his son.  [00:27:00] Tried arrowroot starch.  Always trying different starches.  Summary on page 109, 110 of — written in [Artur?] Hiller’s writing — of who is what type, 2B and 2A.

Looking at possible relationship to Rh disease, family [00:28:00] Ogilvy, erythroblastosis fetalis, Rh disease.  Mother is A, first child is A, second child is A, and their Rh blood groups are little C, little D, little E, little C, big D, big E, little C, big D, big E.  Trying to understand whether there’s any relationship between Rh incompatibility or the development of antibodies which doesn’t always occur in relation to the haptoglobin types.  [00:29:00] Skin sensitivity tests intradermally for secretor versus non-secretor.  I don’t remember exactly why that was done, but here’s an example of a left arm with one, two, three, four, five, six sample being tested for skin sensitivity.  Redness and so on photographed.  This is intradermal hemorrhage, about equivalent to a mosquito bite.  I think I must have injected into my arm the serum from these [00:30:00] persons not realizing I’m making myself possible antibodies against serum protein or whatever.  But anyway I did six samples in my left arm and redness on Thursday.  They were injected at some time or another and at 3:45 to 4:00 pm and then read at midnight.  Redness from the different samples.  One and four, no redness.  One was type one and four was type two, etc.  Photographed.  Showed some bruises finally due to intradermal hemorrhage and still some more on this.  [00:31:00] On Friday, April 27th, looking at them again.  Number four is definitely redder at 20 minutes, but etc.  Three tingled for about an hour, a little red, etc.

Anyway some gels again on page 120 looking at bovine samples just confirming something and a couple of infants that were HP zero.  Couldn’t see anything.  And it confirmed reports [00:32:00] in the literature that haptoglobin is approximately undetectable in newborns, which I had already seen.  Some antibody tests again on April 28th.  Not very revealing.  Aha.  Saturday, April 28th is rather interesting.  Outfit 1140.  These were samples from a baby girl, [name redacted], and the mother [name redacted] and the mother [name redacted] and the two mothers were both type one and I have a note, [00:33:00] an extra band between slow alpha two and beta in both mothers, must check normal postpartum mothers, because these mothers had some problem.  They were Rh mothers, I think.  But this was the beginning of my detection of what later became called the pregnancy band.  So Saturday, April 28th, an extra band between slow alpha two and beta.

Muscular dystrophy family tested Saturday, April 28th, and there’s bovine checks the following day and other samples being checked.  Rh positive and Rh hydrops fetalis again.  [00:34:00] Met baby at first pregnancy, aborted it four months, second OK, third OK.  Was exchanged because of Rh incompatibility.  These were difficult samples in those days.  Babies who had Rh incompatible mothers and got hemolytic jaundice.  These were from the [Humble?] Memorial Hospital.  Another mother, it says the mother of boy [name redacted], will cooperate through a family study.  [00:35:00] No haptoglobin detectable.  May second, Wednesday, page 134 and 133, mother [name redacted].  Note (inaudible) beta band again.  That is in blue crayon, mother [name redacted].  Babies with Rh problems and I remember witnessing the exchange procedure, these very premature and small babies and having complete [00:36:00] blood exchange to get rid of the jaundice.  And they stuck it with probably a 50, maybe 25 milliliter syringe, withdrawing the blood and replacing with — I don’t know whether it would replace with blood or replace with plasma or whatever, but exchange transfusions it was called.

Some more rather obscure tests described [00:37:00] on May 22nd, page 138, people from [Ottawa?] General Hospital.  Rh incompatibility I think again, in St. Joseph Hospital.  Rather poorly documented.  I don’t know why Mrs. — there’s an example on Monday, May seventh, of a Rh positive Mrs. [name redacted] [00:38:00] had a transfusion with E2 Rh positive blood.  This is her first transfusion but she has had several children, etc.  Not very clear.  Mrs. [name redacted] looking for antibodies.  Out of my depth, I think, in this work.  Still these are the Rh incompatible families.  [00:39:00] For example, on Wednesday, May seventh, this Mrs. [name redacted], mother of a baby girl, May seventh, and a girl [name redacted], Dr. Snelling’s patient, born May eighth, exchange transfusion at 6:50 am on the ninth.  And the boy [name redacted] born at 8:30 am, Rh positive, an exchange on 1:07 pm because the mother was Rh negative and they’ve got antibodies against the Rh positive baby and develop jaundice and were in danger.  So these were exchange transfusion before one knew that he could be treated with the right sort of antibody [00:40:00] to get rid of any Rh positive cells from the baby to the mother.  Baby [name redacted] samples. [name redacted] family and [name redacted] family, May 10th.

(Inaudible) experiment begun on page 158.  [00:41:00] More Rh transfusion tests.  I was trying to find out with the help of — I was trying to help Norma Ford Walker find out if the haptoglobin types had any bearing on whether the mother would get an antibody against Rh positiveness or red cells from the baby because by no means do all Rh incompatible pregnancies lead to a problem in the baby.  Didn’t work out to have any bearing. [name redacted] family studied again here, [00:42:00] a new family.  May 17th, a little baby girl, mother with a second replacement.  [00:43:00] These must have been families where there was an Rh positive baby to an Rh negative mother in multiple pregnancies because here is an example on page 173.  The girl [name redacted], third replacement at 12:30 am.  Or maybe they needed to replace blood more than once.  And that’s of course why I got the samples because the first sample I could get was [00:44:00] withdrawn from the baby and I could get those — the serum from those samples and then they would be replaced by donation.

Trying to make the gels in two parts so they wouldn’t have to be sliced, not worth hoarding them up.  Split gels, that’s what I was talking about, further tests on page 180.  [00:45:00] I’m not very happy with it.  Still trouble as the gels stretch so much and are a little too wide for easy handling, etc.  Testing thin gels and turning them over.  With a comment that when split gels were done one gave trouble in splitting, other OK, but discontinuities were negligible, but it was never pursued very much further.  Still trying it though.  [00:46:00] Various family studies from the Rh lab.  Friday, May 25th, page 186, nothing very remarkable, just continuing.  Mother [name redacted] query extra slow alpha two beta band, which I later called the pregnancy band.  That pregnancy band was published in Advances of Protein Chemistry for which I wrote an article [00:47:00] on starch tailored electrophoresis and it was later followed up by quite a number of people, the so-called pregnancy band, and it was later discovered to be something else.

On page 192 Monday, May eighth, [00:48:00] various samples taken again, beginning to understand the presence of the extra pregnancy band.  Mother [name redacted] may have the slow alpha two beta zone.  Later it was understood that it occurred in pregnant women towards the end of pregnancy or shortly after delivery it could be found.  And it was identified although I don’t remember what it proved to be, but it is now well known.  On page 193 and 194 there’s a big summary of samples from many individuals [00:49:00] just confirming that they had been checked or not checked with hemoglobin and what had been observed.  It’s really a working summary of the samples that had been tested.  About, I should think, 40 samples had been looked at.  And the last page on this book is [00:50:00] just a note that OS is an Rh positive non-secretor and that Artur Hiller’s telephone number was changed.  And final checks to be done [name redacted], father [name redacted] and mother [name redacted] and check marks indicating that they had been checked.  So that’s the end of book seven, Roman numeral seven.


— do it right now.


Tell me.  The pregnancy band was later shown to be a proteinase inhibitor.  I don’t know if its function has been yet to come in, but it has been cloned and sequenced.