Oliver Smithies:

[00:00:00] So this is Book j, October 8th, 1974, going to December of the same year.  Continuing to try to prepare 125I‑labeled…  Papain-shedding experiment is the first one described.  So still trying to get a different sample.  And the autoradiograph on page two is substantially better than many others — and fairly good counts.  And the reason being that that a lo‑‑ a reasonable amount of counts put onto the gel, so that the — [00:01:00] something on the order of 600,000 counts was available.  Beta­‑2 microglobulin is labeled, the ca‑‑ be a carrier there, the radioactivity.  So it looks as if the experiments are beginning to work, anyway, on this one.  But no particular comment on it, just a little line showing beta-2‑M — and where it was labeled — with a question mark about wha‑‑ there’s a lot of label that is not beta‑2‑M, with a question mark on it.  [00:02:00] So this is continuing, on page five.  “Dave precipitated –” that’s Dave McKean — “precipitated half a ml with antibody from the [j‑1?] harvest and obtained ten di– 74,000 counts were dissolved easily in chlorohydrate,” etc.  And exposed from Friday to Saturday and Saturday to Monday.  The results are, “No beta‑2‑M yet in the antibody.  But new 125I beta‑2‑M is good but contains some impurities.  Plus or minus carrier has no effect on the electrophersi– But the only successful antibody precipitation was the three-day plus [00:03:00] carrier [i‑87?] experiment.  Therefore retest the latest 125I.”  There’s lots of labeled but they’re not of the product that we are hoping to get.  This papain strip method is — papain is in the medium and is presumably thought to release beta‑2 microglobulin.  (pause)

Continuing this type of experiment with [00:04:00] antibody precipitates, autoradiography and stain.  A rough repeat with iodinated beta‑2 macroglobulin, papain with cold beta‑2‑M, on page 13.  (pause)  [00:05:00] And worrying about the susceptibility of beta‑2 microglobulin to papain, on Tuesday, October 15th.  But the conclusion is that “This enzyme-substrate ration, there is not breakdown.  Therefore, decrease the enzyme or add beta‑2 microglobulin.”

The unknown band was taken out for sequencing.  [00:06:00] Or attempted to take it out.  But the conclusion was, having cut the band out, that (laughs) “This band is dialyzable in heavy,” underlined.  The dialysate had about 400,000 counts per minute in it.  A liquid in the gel had something like 15,000 count.  (pause)  Still 125I papain strip, with 0.1 of papain [00:07:00] and carrier, beta‑2 microglobulin in the medium — and with a result that about 120 microcuries were released and about 9 microcuries were left on the cell.  And so it looks as if it was possible to decrease the papain more.  Although, unfortunately, antibody precipitation with the harvested material — “No appreciable counts in the antibody precipitate.”

(laughs)  And Wednesday, Thursday, October 17th, page 21, a gel — autoradiograph with lots of counts but [00:08:00] with the resul‑‑ “Results suggest a miniscule trace extra of beta‑2‑M.  Continue by decreasing the enzyme ten times more” — despite the comment on the following page.

So I check on — of a…  “A band gave about 120,000 counts per minute.  [Centrifuged?].  But plus lauryl hydrate and some aluminum lactate [to do?] — to Braunitzer sequencing.”  [00:09:00] With the comment, “OK.  Purification must be greater than 100‑fold.”

And lower-still-papain strip being considered.  “And test zero papain next.”  (long pause)  [00:10:00] Repeat, Tuesday, October 22nd, with zero papain.  With, on the opposite page, a yield approximately 9 microcuries.  Gel on the following page.  But saying that the product was — looking at the sample of 1.6×106 counts, with no papain, and then second extract with 1/1000th and then third with 1/100th of papain, [00:11:00] and with a control.  And the counts are definitely released with the papain, as it says.  “The product was only liberated by papain.”  But it didn’t give anything that looked like beta‑2 microglobulin.

Trying mercaptoethanol, on the stored material, page 37.  Eight-molar urea, aluminum lactate tests, on page 39.  Chloral hydrate extraction, on 41, continuing this general testing.  Typical gel on page 43, with lots of counts in the j‑21 [00:12:00] control material.  But no beta‑2 microglobulin in the harvested product.  Repeating the experiment without carrier.  Total counts in the precipitate were about 0.5% to 0.7%.  But nothing was remarkable enough to have any very promising comments.

[00:13:00] Monday, October 28th, chloramine‑T trial.  (pause)  With a comment on the page 51, with a couple of circles, “Looks positive by electrophoresis but a tad fa‑‑”  Then the interpretation on page 50 that “Previous experiments labeled lactoperoxidase,” etc., etc., “[Either?] experiment,” etc., comments on the result.  [00:14:00] But a material that was obtained on the autoradiogra‑‑ is a little bit faster than beta‑2 microglobulin, as it says in the comment there.  “Yield less than 1 mc.”

And Dave McKean was sequencing these proteins, presumably by hand, with Edman degradation.  Because we’re not yet using the sequenator on these radioactive materials, as far as I recollect.  But [00:15:00] I don’t know whether we were using the sequenator, at this point, or not.  It’s probable that we were — since it was…  Yeah, we were probably using the sequenator, although it’s not immediately clear, that.  Although if I look at what is described later, the sequencing was with quantitative sequenator runs.  So all these sequences were obtained with the Edman-Begg sequenator that I’ve talked about earlier.  [00:16:00] And so, going back to page 53, October 29, “Antibody precipitation not successful by counts.”  But we have there the comment, “One kept overnight for Dave to sequence.”  Continuing to try to get label into beta‑2 microglobulin.  Reruns, on page 57.  “Expose at least 24 hours.  Better 48 hours.”  But quite a lot of count.  [00:17:00]

Comment on page 65, “The latest set of results show that for present purposes lactoperoxidase and chloramine‑T give the same result.  So the components are real.  Now necessary to check 356 versus [GM‑17?]” — those are the two cell lines — “And then return to HeLa to get a known positive.”  On page 69, “Dave repeated the antibody with [00:18:00] 1.4ml and got 52,000 counts.”  Mm-hmm.

Formula on how to make aluminum lactate buffer.  It was always (laughs) rather entertaining and somewhat difficult.  Because you had to take aluminum foil and then activate it by coating it with a thin film of mercury.  So, “Immerse the foil in mercury chloride to activate the foil.  Pour off the mercury chloride, being careful not to loose any foil.  And be careful, because mercury chloride is poison,” etc., etc.  This is after Sung and Smithies in Biopolymers.  We’d made aluminum lactate buffer.

[00:19:00] Page 75, continuation of the earlier experiment, with the comment, “Most of the antibody counts are not in beta‑2 microglobulin.”  [Check?] of new lactoperoxidase.  “[Either?] cells retested.”  And so on.  Huge amounts of counts, on page 87, judging from the autoradiographs.  Seventy-two-hour exposure.  Enormous dark film.  But only a minute trace [00:20:00] of beta‑2 microglobulin — with 72‑hour exposure.  “Further exposure failed to make this more definite.”  Never give up.  HeLa, papain, page 93.  Large-scale HeLa experiments, on November 26.  And continuing in the same vein.  “Pulse-chase intrinsic label.  And shed without antibody,” using 35S methionine this time, [00:21:00] on page 105.  Thanksgiving Day.  Another Thanksgiving Day experiment — not the first, on page 109.  We’ve seen experiments on Thanksgiving Day before.  “The results suggest a beta‑2‑M in original 24 medium and in the Triton extract.  — Very much better dried,” the gel.  Still, all these experiments are [00:22:00] basically incremental, if even tha– Methionine labeling, 35S, on page 117, repeated.  “See j‑120.”  (pause)  And comments on the results.  Two hundred microcuries of the 35S methionine, on page 121.  Used quite a lot of lead, in those days.

[00:23:00] But Thursday, December 5th, using goat anti-rabbit Fc and polyethylene glycol, etc.  Six-day exposures, 17‑hour exposure.  Sloping gel.  But it’s…  “The count is successful again.”  But successful in what sense is not immediately clear, perhaps that there is some beta‑2 microglobulin in the [00:24:00] supernatant antibody precipitate.  “Was medium supernatant harve‑‑ and Triton harvest.”  And the supernatant antibody precipitate, which had 100,000 counts.  And there is definitely a band that looks as if it is in the same place as the 125I‑labeled material.

Following experiment was, “Repeat sterile 125I pulse-chase,” with a big comment on the top, “Abandon.”  [00:25:00] [Octoloni?] tests again, on Sunday, December 8th, trying to see what things meant, and with the conclusion, “The results approximately equivalent to the electrophoretic result suggest that 24 hours is several times the average turnover time of some part of beta‑2 microglobulin,” etc., “This method is probably the fastest to study 35S‑labeled material.”

And as we got to the end here, of this book — we’re nearing the end — many gels being tested in a turnover [00:26:00] study — with no clear statement as to what they mean.  (long pause)  Monday, December 23rd, purified anti-beta‑2 microglobulin.  [00:27:00] “Dave made gamma‑2,” etc., etc.  Alkaline gel tests for 356 versus GM‑17, very long run.  Lots of counts but no comments on what the results meant — as we reach the end of this book.  Last page — or next to the last page, December 26th, page 147, is an Octoloni test, [00:28:00] with no written interpretation.  But the result is shown.  And the last page is Pharmacia cyanogen bromide-activated Sepharose, used for coupling, etc. — to try to label antibody, autocleave the methionine.  The comment is that “Pharmacia cyanogen bromide-active Sepharose recommends coupling in 0.1‑molar sodium bicarbonate, sodium chloride,” etc.  [00:29:00] But that is the end of this book — and the end of what we are going to do today.  [00:29:07]


[00:00:00] It’s perhaps useful, at this point, to reemphasize the motivation behind all of these experiments with attempting to label proteins in small amounts and sequence them.  So just reading the introduction to the final paper, on membrane antigen sequencing, “It’s worthwhile –” said, “Many laboratories are directing their efforts towards determining the primary structures of membrane proteins and cell-surface antigens.  But the proteins usually constitute only a small fraction, making isolation of sufficient material difficult.”  And then…  And that was why we tried to label them. It’s — “The third approach to the problem employs radioisotopes synthetically incorporated [00:01:00] throughout the molecules of interest” — and that only small amount is needed.  And the results are insensitive to unlabeled contaminants, etc., etc.  And that was what we were trying to do.  Intrinsic labeling of proteins for radioisotope sequencing.  (pause) [00:02:00] (pause) Still running, Jenn.  [00:02:27]