Oliver Smithies:
[00:00:00] This is Book XXII, Roman XXII, no date on it, except 1959. Starts October 27. Trying again, 8‑[molar?] urea, formic acid, and haptoglobin, a new solution — with various other things. In this gel, there is a new solution of haptoglobin, acid, urea, from so-and-so. And glucagon and urea… And so on. Then no particular comments on it.And then a two-dimensional gel again, 8‑molar urea in the main gel and formate in the first dimension, with or without [00:01:00] iodo– It says, “No iodoacetami‑‑” And the comment that “All of the fast components are soluble in [50:50:10?], after dying. So that that’s one of the problems, is that the fast components wash out of the gel rather easily. They don’t form very stable complexes with Amido Black, when in 50:50:10 buffer.
The urea gels were always a little bit of a pain. They took a long time to set. And Wednesday, October 27th, with a comment on preparing them. “8‑molar urea gel made by solid urea and starch [00:02:00] mix,” and “rapidly stirred into the buffer. And sets in about 2 minutes. Heating at 50 liquefies it readily. And heating at 100 gives a pH 6 after 10 minutes and gassing.” In other words, the urea had decomposed. So… This is just saying that the 8‑molar ur‑‑ gel is made by solid urea plus starch, which are mixed and rapidly stirred into the buffer. And it sets in about 2 minutes. “Can be easily controlled and pouring is possible. At 1 to 1.5 minutes, pH 4.” Heating, again, as I said, liquefies it. [00:03:00] And at 50°, it’s safe. But 100° is too much. Heating at 70° for 6 minutes, the pH changes from 3.90 to 3.96, not very bad. Some gels run but they were not kept, though it says, “Very promising” and “Unsure why [the back goes on?].” This is 6 minutes at 70°, more than 30 minutes at 50, and 9 minutes at 50 and no heating. And various comments on them. So it’s evidently messing around with making the gel.
Thursday, October [00:04:00] 29, shows that a gel made at [0.5‑, 0.1‑?], 8‑molar urea, by cold-mixing, and heated at 50°, the pH stayed really pretty constant up to 20 minutes but at 40 minutes began to get a little bit higher pH. And 180 minutes, the pH had changed from 3.95 to 4.16. And the comment, “Thus, a reaction probably occurs appreciably with any times, at 50°. But try 35°.” But at 20° degrees, the extrapolation said that the rate would be [00:05:00] negligible. So trying different ways of making the gel. A heated urea gel. (long pause)
The gel made by cold-mixing, [00:06:00] room temperature setting, 8‑molar urea, was not kept. No image of it. But the 0.05-molar formic acid gel, the result is shown. And all the same whether the stock solution had been heated or the urea re‑crystallized. Or unheated stock urea, unheated re‑crystallized urea didn’t really make any difference. Trying to work out whether the urea made any difference. (pause) “…cold — 8‑molar urea, cold-mix, cold-set tested.” [00:07:00] Still the same pattern. But the current was very low.
More tests, formate gel, urea gel, on Monday, November 22nd (sic), trying to get things to work. A two-dimensional gel, on November 3rd. Rather poor two-dimensional. (pause) [00:08:00] Trying dimethylsulfate, not obviously why. Reference to E.A. Werner, Journal of — Chemical Society, with a comment, “O-methylisourea is the — and picric acid…” Six grams of urea, 12.6 grams of Me2SO4. Have my doubts but there’s something wrong with it. The urea was heated to 100, at 120, and the dimethylsulfate was added very slowly, with stirring. “A viscous mat. Needle crystals of [00:09:00] picrate,” [the?] sodium picrate. And a gel was run, with 8‑molar urea, cold-mix, with haptoglobin with water, haptoglobin with urea, haptoglobin with isourea, with a poor run. But the front doubled back multiple… In other words, nothing unusual. Attempted re‑crystallizing isourea. Potassium cyanate was considered. And… So the November 5th gels, with haptoglobin, with isourea and urea, [00:10:00] potassium cyanate, and isopicrate. And there really wasn’t any obvious difference. All of the — all of the patterns are basically the same, as far as the fast components are concerned.
Various amino acids were tried, for no obvious reason, in the formic acid, sodium hydroxide gel, arginine with urea, asparagine with or without urea, aspartic with or without urea, glutamine with and without, glutamic acid, etc. Glucose [00:11:00] amine, histidine, lysine, and lysine, acetone. And the results, Monday, the 9th. “Only lysine, O‑methylisourea sulfate look promising.” The gel was run Monday, the 9th. Very little difference.
Eight-molar urea, Tuesday, November 10th, Wednesday, November 11. Again, nothing showing anything very different, usual multiple bands, in the [00:12:00] slow component, and two in the fast, when using haptoglobin 1‑1. (long pause) Sunday, November 15th, an oxalic acid gel, pH 3.23, with 8‑molar urea, and then acetic acid gel, pH [00:13:00] 5.3, with 8‑molar urea, testing various substances. And clearly, the 8‑molar urea, oxalic acid is much better than the acetic acid gel.
Still worrying about multiple bands in the constant region. So Wednesday, November 18th, “To determine if the multiple bands are due to loss of NH2 from iodoacetamide.” Using carbodiimide. (pause) [00:14:00] (pause) No comments, really, on the results, which is shown on Wednesday, November 18th, a page or so afterwards. Wednesday, November 18th. Says, “Test the [00:15:00] isolation procedure.” So this is the beginning of tests that eventually made sense. Fresh serum was prepared, Tuesday to Wednesday, and desalted and put on diethylamine — DEA column, etc., eluted in the usual way, and then treated with urea and mercaptoethanol and iodoacetamide. Well, no mention of iodoacetamide, actually. And [00:16:00] it isn’t clear where the serum came from, at that point. (pause) And the gel is shown, without much comment, in fact, without any comment. Rather messy page in the book, [00:17:00] where some filter paper chromatography sheets present have really damaged the surrounding book, rather. Trying chromatography of glutamic acid with urea, glutamine with urea. And no differences plus or minus urea. Formamide tests again. (pause) [00:18:00] Checks of with and without iodoacetamide.
On Thursday, November 26th. Still two bands in the haptoglobin 1‑1. Maybe the serum was from me. Because I think I’m a heterozygote for 1F‑1S, which was part of the problem — still not recognized. Two-dimensional gels, on Wednesday, November 25th, [00:19:00] showing nothing unexpected. And again, on Wednesday, November 25th. And the fast components easily wash out of the gel, if one is not careful. And then, Friday, November 27th, from this date, all the haptoglobin 1‑1 is a new lot. [00:20:00] Trying sodium sulfide again. And here, on Saturday, Sunday, 29th, a check of various preparations — with no interpretation of the results listed. [00:21:00] (pause) Getting ready for some more two-dimensional gel. And they occur on the following few pages, with nothing very much learned from them. (pause) [00:22:00] Monday, Tuesday, November 30th, December 1st, “To try to check if acidification removes the multiple zones.” I was worrying about the multiple zones — and trying different things and still multiple. This was after making them acidic with glacial acetic acid versus not making them acidic. (pause) [00:23:00] (pause)
Wednesday, December 2nd, 3rd, haptoglobin, second grade, and a new lot of haptoglobin, both 1‑1, with a comment that — testing sodium sulfide again — and “A very good run.” So that’s increasing amounts of sodium sulfide, with second-grade haptoglobin 1‑1 and with first-grade haptoglobin 1-1. And [00:24:00] the result is independent of the concentration of sodium sulfide and is equivalent to previous results, two major fast components and two minor one. So, “Try two-dimensional gel.” I’m still getting two bands at the front. But there are some complications in this one. There are too many fast bands — that doesn’t seem to be worrying me but it would worry me now. Very complicated patterns with the sodium sulfide, on the gels run Wednesday, 2nd and 3rd. They’re being made ready [00:25:00] two-dimensional. And the two-dimensional gels show nothing very striking. Everything just migrates into the gel, you might say diagonally, as expected, if nothing different happened.
“[Na+‑SH–?] versus mercaptoethanol suggests that a second type of bond is split by mercaptoethanol. Maybe a thioester interchange occurs.” Think thioester… [00:26:00] Tests of the hydroxylamine — with a result with no obvious differences plus or minus hydroxylamine, in this way. “Test with it for longer periods.” But the sodium sulfide patterns are much more complicated than [00:27:00] the mercaptoethanol ones. Thinking about checking a shifting of SS or SCHO — and sodium boron hydride reduction, on December 9th. Two-dimensional gels on the following page, with nothing remarkable found from them. Quite a lot of work. (pause) [00:28:00]
Interviewer:
Go ahead. How are you?
OS:
Further tests, December 10th, of SCO. Iodoacetamide, hydroxylamine, sodium sulfide into the gel. And trying to understand where the double band comes from, in the fast component. [00:29:00] (long pause) “Possible…” Friday, December 11th, trying to get the cleavage of S‑S bonds, followed by iodoacetamide very rapidly. So this is 8‑molar urea, mercaptoethanol, and then samples go into iodoacetamide at interval, 70 seconds, 120 seconds, 185, 4 [00:30:00] minutes, 8 minutes, 32-minute, 60-minute. And not a hoot of different, no differences in haptoglobin. “Very fast reaction.” So it doesn’t matter. The reaction is very rapid, as I’ve already said before. But it’s rather interesting that this haptoglobin 1‑1 has not two bands in the same proportion as previously — but not noted. Test of SCO and mercaptoethanol — no, and — yeah, mercaptoethanol. No effect [00:31:00] with iodoacetamide, in this experiment. Thioglycolate retests. Not much comment. (pause) [00:32:00] Trying Sephadex to get rid of the urea, December 15th. (pause) Beginning to think again about reoxidation, split haptoglobin. Dialyze and try to oxidize. Try with ascorbic acid. [00:33:00] Many different things tested. Looking for fractionation — on a column. (long pause)
[00:34:00] Running a gel with 8‑molar urea, formic acid, containing 0.02-molar mercaptoethanol. The sodium sulfide product is converted completely into the same product as is obtained with mercaptoethanol, with and without iodoacetamide. So that the comment is that… And the‑‑ and then the other gel is 8‑molar urea, formic acid, with 2‑millimolar sodium sulfide. So one is with 20-millimolar mercaptoethanol and 8‑molar urea, formic acid and the other is with 2‑millimolar sodium sulfide, formic acid, and 8‑molar urea. And the sodium sulfide gel still shows a difference between whether [00:35:00] the product was reduced with mercaptoethanol or sodium sulfide, with or without mercaptoethanol. But if the gel contains 0.02-molar mercaptoethanol, everything goes the same. So the mercaptoethanol effect is reversible, in the presence of free SH groups — so that the sodium sulfide product, which is complicated, can be simplified by mercaptoethanol.A little comment from George here, “All of the haptoglobin preparations done during November are Mrs. [name redacted]. I think they are quite good.” He’s sending a [00:36:00] haptoglobin preparation of a family — [name redacted] family — no, [name redacted] family. This is separate. There is a family being tested, on Wednesday, December 16th, just a straightforward family test. And the other is just a note saying where the haptoglobin came from.
Thursday, December 17th, is a good page. Because here it’s an attempt to use SDS precipitation to fractionate the [00:37:00] alpha chains from the beta chains. So this is 1:55 p.m., mercaptoethanol on the haptoglobin 1‑1, and then reduced in 8‑molar urea, borate and iodoacetamide added and then onto a Sephadex column — and then dispensed to different concentrations of sodium lauryl sulfate. And then precipitations — and centrifuge to get rid of the precipitate. And so this is an attempt to purify the haptoglobin alpha chains by precipitation. And the gel is really quite nice but [00:38:00] fairly critical, on Wednesday, December 16th, is the beginning of how we later purified the alpha chains. Untreated Sephadex eluate is shown — and then with — is 0.0005-molar, then 0.002, etc. — 0.00215, 0.0003-molar SDS, and so on. And although it’s very critical concentration, there are very good alpha chains, as we later call them, with 3‑millimolar sodium lauryl sulfi‑‑ SDS. Thus, the promising tube is 0.002 or 0.00215. [00:39:00] But it’s very critical. “Extremely good fractionation. 0.003 is almost pure,” is the comment. The start of being able to get pure alpha chains.
Tests with SDS, the sodium lauryl sulfate, with type 2‑2. So to recapitulate, the material is freed from urea by passing down a column and then is treated with SDS — and this particular [00:40:00] method. And in this case, 0.002 is very clean, as shown by the gel on Friday, December 18th. And with a note that “The slow fraction is already decreased in the dialyzate,” suggesting that it will aggregate very considerably, even without the lauryl sulfate.
(laughs) Back to testing with hemoglobin, pre-splitting. [00:41:00] So Saturday, December 19, the 2‑2 purified sodium lauryl sulfate fraction, plus or minus hemoglobin, was made. But I forgot that this is acid — and so added borate. And the result is shown, on the gel. But benzidine is in the bor– This is a borate gel on the material. And the hemoglobin is clearly [00:42:00] still present and reacting. There is no indication that adding the SDS-purified material makes any difference. But there’s no comment on it.
Trying to let the Sephadex column stand. Thinking about the precipitation that occurs in the dialyzate. So run it in [00:43:00] the column and wait, is the idea. So 0.5mls of a mixture is put on the top of a column, 3ml wash, rapidly. And nothing will come out, because it doesn’t come out until 0.6mls more is added. And so there is a region of at least a clear zone of [one side — to be?] with hemoglobin at the bottom. This is an attempt to see whether any binding occurs. [00:44:00] This Sephadex eluate, 1‑1. One-one-C, it’s called — is [probably — is group?]. And 2‑2 Sephadex eluate. No fraction in the 1‑1 — no fractionation in 1‑1 and very little in 2‑2, on the column.
United the fractions with the eluate of the Sephadex column. So the Sephadex column now has been allowed to remove everything. [00:45:00] And the haptoglobin, consisting still of the large-molecular-weight and the small-molecular-weight material, has been prepared. And the comment that “They were united and left in the cold” and they’d almost gelled in the cold — was the comment. So, Sunday, December 20, the recovery of the sodium lauryl sulfate precipitate from the 18th, 3.2mls. (pause) And [00:46:00] a comment of George’s, Monday, December 21st. “GEC found 2‑2 split and freeze-dried could be extracted with ethanol and then [filtrate?] buffer,” to get the fast only. So that’s another way of getting the fast component. (pause) Comment of…
[00:47:00-00:49:00] So here was an experiment in trying different methods of fractionation, again. This is on Monday, December 21st, following some ideas of George Connell, where 2‑2 haptoglobin had been run through Sephadex, and to get it in 1‑millimolar formic acid, 20% neutralized. So it’s acidic but very low ionic strength. And tried the same buffers, for isoelectric precipitation, without success. Added a little mercaptoethanol to each, to try to get the slow component to [00:50:00] precipitate. Etc., etc. But there was some promise. Because formic acid extract, acetic acid Tris extract — because the Tris, pH 7.5, gave a precipitation — small — plus a clear supernatant. So I tested it on the gel. And the results were quite promising — although I say, “Very promising indeed.” But looking at them now, I would not feel quite so happy about them — that formic acid gave about 90% fast and 10% slow — the formic acid extract. [00:51:00] And the… Yeah. (pause) Formic acid being on the right-hand side of the — of the photograph, the top of the gel. The gel is, in fact, oriented rather awkwardly, to see the result. I think I’m going to stop there and change the illustration. No, I won’t do that. The result is that the formic acid extract is [00:52:00] quite promising, and looking at it now. But my interpretation at the time was that it’s “Very promising indeed.” Then testing a two-dimensional gel, with 1‑1 extract. [Impossibly?] complicated, the gel. With not much comment on them. Could be different. Could be equivalent. A column, again. [00:53:00] (pause)And Tuesday, December 29th, a fairly extensive attempt to purify again on column. Fifty milligrams of 1‑1 FB, pure from November ’59, was reduced in 8‑molar urea, borate, with mercaptoethanol and then iodoacetamide, onto a Sephadex column, 25 centimeters-by-1.2 centimeters, in formic acid, and fractionated. [00:54:00] And then various amounts of sodium lauryl sulfate were added to the acidic eluate. And the result shows that precipitation occurs when the SDS or the lauryl sulfate reached about 0.015-molar. And the later fractions were used for electrophoresis tests. And the electrophoresis tests showed a very satisfactory result — that [00:55:00] there really is quite good fractionation again with the material, at higher concentrations than 0.02-molar — are pretty well free from the heavy chain. Two gels are shown. One may be the staining after being in 50:50:10, to wash out the fast components. It’s not complete absence of the heavy chain. But it’s fairly good.
And trying [00:56:00] it on 2‑1, haptoglobin 2‑1, again and haptoglobin 2‑2 also, with the gels being run on probably Wednesday, December 30th, or thereabout. The 2‑1 purification is really not very good. The 2‑2 is considerably better. So why one worked and the other didn’t is not clear. (pause) [00:57:00] And, in fact, the next page addresses that problem. So, “As in view of the poor amount of separation of 2‑2 in the 0.0501 formic and known aggregation of (inaudible) under these conditions, try 2‑molar urea and 4‑molar urea for the Johnson type.” So beginning to think it wasn’t quite so simple. And so, on Friday, January 1st, I did an approximate precipitation curve, in 2‑molar urea, with the 2‑2. And the comment is, “Thus the urea has lengthened the gap between the slow and fast. It’s quite a nice separation.” “There’s a rather long tail,” is the comment. [00:58:00] The gel shows that the fractionation is not complete. But the long tail of the curve is probably due to the fast component. Purity is fairly good, on the top of the tail. Looking at the gel now, it looks quite good.
And testing 4‑molar. And again, an idealized curve drawn. But no precipitation but cloudiness, with the sodium lauryl sulfate — [00:59:00] at first. But “Very clean, [at?] 30%,” is the comment. So there is… Let’s go over that again, the 2‑2 haptoglobin with 4‑molar urea. And then the… There is a fairly clean material obtained, at 0.034 or 0.032 sodium lauryl sulfate, and therefore promising [01:00:00] for the Johnson type.
And the last entry of this book, which is on Friday, January the 1st, working New Year’s Day, is “Tested of fractionation of sodium boron hydride on haptoglobin.” So dialyzates with haptoglobin, reduced with sodium boron hydride, and with a comment that “Sodium boron hydride has no effect.” And that’s the end of this book. [01:00:45]