Oliver Smithies:[00:00:00] So this is the beginning of Book β, in the Greek alphabet, December 2nd, 1981, going through April 5th, 1981. And we’re getting very close to the beginning of the Nobel Prize work, that is still thinking about origins of replications though at this point, and various probes.
So it starts off December 3rd, Thursday, page 1, looking at probes that are now available. Hot probes, various probes. And, Paula’s probes as well. We have a look at them on page 3, the conclusion, “Greatly improved with clean [00:01:00] fragments,” et cetera. Time series with tests of Paula’s material.
And a cheerful note on page 7, the conclusion, “The hybridization is getting to be quite good.” And Paula shows something at about 300base pairs, and at the start.
More label tests on the following page, 48 hours of hybridization, and 12 hours hybridization. New tests, probes available for G gamma [00:02:00] 1482, Fnu 4R1 fragment, and 462 A gamma Fnu 4R1. “Test each against itself and against the opposite,” is the idea. But, no very definite conclusions.
Longer hybridization and less single-stranded DNA on page 15, with a conclusion that, query, “A bad experiment? The labels very low for some unknown region, but all are cross-experiments, and they didn’t show yesterday, so it was an unsatisfactory day. [00:03:00]”
So repeated on page 17 again with the conclusion, “Very disappointing. Perhaps alpha-109 is a little too stringent,” but there was no label at all in this experiment, except for the probe itself, I think.
Paula’s material tests, purified fragments being looked at on page 19, Friday, December 11th. Conclusion, “IT WORKED FINE,” in capital letters, clear that the present estimate is very close to full-length, about 700bp long. So looking for the fragment that Paula was [00:04:00] interested in.
Evidently grant-writing time, because the note, Tuesday, February 2nd, “Grants written,” page 21. And then, a reprint from David Ward’s lab on enzymatic synthesis, a biotin-labeled polynucleotides, a novel nucleic acid affinity probe, which struck me as being rather exciting, and I thought I would try that type of experiment. So he gave a seminar on December 9th, talking about his iminobiotin methods, and I began to think I might use that, and indeed, [00:05:00] I know that quite a lot of the work in the next book or so is related to that, so my notes are looking at — going to a seminar of his on December 9th, 1981, David Ward, and as usual, when I go — when I went t seminars in those days, I would always take notes of them. But I enjoyed the seminars better if I took notes, and it made me absorb the material, but it was a pleasure, not something to be done under duress, but rather for enjoyment. And so here are two pages of notes from David Ward’s seminar. And I’m trying to [00:06:00] follow some of his work on February 3rd.
And trying to absorb the probes, label material, to Sigma avidin-agarose material that was available, so one could try for absorption.
And, looking at this as a repeat on page 27, “Only nonspecific losses,” is the conclusion. The IB — dUTP, that’s presumably iminobiotin, or biotin derivative, I forgot what that was called, [00:07:00] maybe blocking of the beads, but anyway, “iminobiotin,” not “immuno.”
Iminobiotin incorporation for further checks on the following page. February 5th, Friday, page 29, getting into this experiment. Repeat of the absorption experiments on page 31, and with 13 counts on the beta-binding, and 1300 counts, and only 270 on the T-binding, with the comment, “Clearly, there is some binding, but either need to increase time or amount of agarose. There are not very many counts. We’re only dealing [00:08:00] with 270 counts per minute.”
Count is trouble on the next page. “Better labeling material, thought to be,” on page 37, but there was an error, which is solved on page 59.
And an experiment with heparin, and this was repeated on page 39, an attempt to increase the yield. Apparently, heparin was thought to make things more specific. And, with the conclusion, [00:09:00] because I didn’t see the preceding experiment, but this time, “Once again, heparin decreased the nonspecific incorporation very markedly, but it also appears to abolish the specific binding, so try phosphate buffer in place of heparin, or de-titrate the heparin.”
So here we go working on biotin labeling, iminobiotin labeling. A repeat of the experiments again, as usual. Other conditions using, for binding iminobiotin to streptavidin, Stan Kline’s conditions from Enzo Biochem being looked at. [00:10:00]
Page 49, then the conclusion that “The non-inhibited binding is that, larger molecular weight material, the specific binding is small, but the starting materials looks as if Mbo did not work,” which was commented on earlier, that there would be some check on Mbo. So, as usual, technical problems titrating the beads, February 11th, page 53, and so continuing.
Repeat an Mbo digest with more enzyme. Nothing much, specifically, on page 55.
So, a letter from [00:11:00] Stan Kline. “Dear Oliver, I’m enclosing so many units of iminobiotin-dUTP,” et cetera, so sending material, people were always kind at sharing their precious products. Remarkable good scientists.
Remake of labeled test material, February 12th, page 59. And stuck in here, page 51, is a map, [00:12:00] photographic record of a map of the region between pseudo-beta and the delta gene, where the two Alu sequences are, so I don’t know whose map that is, but it’s certainly a useful map.
A new sample of agarose-avidin on page 62, so I was struggling to make the method work. Eluates and [00:13:00] supernatants looked at on page 67. Nonspecific T-binding is about 3% for all, and whatever “T-binding” is.
Picking up on page 60, or 61, Saturday, February 13th, labeling continued, and then looking at the counts that were obtained when the material was labeled with thymidine T, or with iminobiotin uracil, uridine, iBu, so T is the material that should not bind to the column, and B is the material [00:14:00] that should bind to the column. And the differences in counts on this page are miniscule. T was eluted as 14,000 counts per microliter, and iminobiotin was 16,000 counts, so clearly, it didn’t yet work appreciably.
So, we continue on the following page 63 with a new preparation of a new product of agarose-avidin, to which the biotin will bind, buying this from Vector Laboratories Incorporated.
And the eluates and the supernatants were compared [00:15:00] on the following day, but the conclusion is that the nonspecific T-binding is about 3% for everything, and increased slightly by biotin. And then, all three agarose do fairly well with iminobiotin. This was a test of three different agarose preparations, agarose, streptavidin. But Vector Labs is best, about 12% difference between the specific and non-specific binding, and streptavidin was next best at 8%, and sigma is the worst at six percent. But so then, qualitatively, [00:16:00] by the radiographs, streptavidin column, however, is most like the input. So, continue with the three agars, but omit the T control for the present time.
The end labeling with iminobiotin dU, either by end-labeling or nick translation for size tests being carried out on Tuesday, February 16th, looking at the differences between the end-labeling and the [00:17:00] nick translation on page 71. Suspect an error, because the results were very odd.
And going back to titrating the beads on February 20th. Eluates supernatant looked at on page 75. With a conclusion that, “The eluates are truly representative, except 1V+B looks suspiciously like little b. [00:18:00] In any case, vector gel is either twice or four times better than AH, so a different agarose material, so comparing the different column material, a titration of the vector beads on 77, as asked for on page 75.
Cleaning up of the DNA on page 79, “Non-specific binding is very low, but specific binding is there but small. So make the translated again, and try again from scratch.”
And new [00:19:00] translated DNA, page 81, and 83, continuing to try to get a decent label with specific ability to isolate the DNA that had incorporated iminobiotin.
For example, on page 87, a rather severe side-effect, but still can titrate the beads by examining the supernatants. Counts alone can be misleading. Continuing to try to repeat this problem. Smaller pieces on page 93, Monday, March 8th. [00:20:00] Time series, the following days, on and on, with the same type of experiment. Conclusion: “Promising, the supernatants and eluates are beginning to approach. Beads are not saturated,” et cetera, et cetera.
pH-controlled time series, and on page 107, evidently I talked to David Ward. I have his telephone number there, but he was abroad for awhile, and [00:21:00] David Ward arranged for me to talk to Sam Kline, et cetera, so these yellow page notes, of these telephone conversations. I remember these days getting more and more deeply into the biotin labeling, and less and less happy with the work.
Comments on recapitulation on page 109, “The data collected on the left show progressive improvement, but critical effects appear to be the time of effect, time effect for large pieces, and pH effect for elution and blocking experiments,” et cetera, et cetera. [00:22:00] A reminder that streptavidin and avidin have different pKs, and notes on, my whole notes on what was happening, recapitulating.
“Co-carrier, I need a carrier with iminobiotin!” Thursday, March 18th. Typical comment here, conclusion, “One and a half hour exposure yields even on zero added DNA, are very, very low, as shown by the counts.” [00:23:00] On and on.
Testing different beads again. For example, on Monday, March 22nd, page 129, an experiment with different supernatants and eluates, and different preparations with conclusion, “Nothing wrong with the labeled DNAs, but the results confirm approximately useless eluates,” so there the supernatants were nice, heavily-labeled, but nothing came from the beads.
And a nice little letter [00:24:00] from Stanley Kline on March 24th, saying, “We’ve reviewed your recent proposal, and request concerning streptavidin. We’re very interested in your studies,” et cetera, et cetera. “Since we’re providing this material free of charge, we would like to have the material only used by yourself,” et cetera. But nonetheless, they were puzzled by the lack of specific binding. And might also do this, that, and the other, so a polite letter from Stanley Kline. Evidently, David Ward had a little business of some sort going on, though I don’t know the details of that.
Higher pHs, [00:25:00] and pH test comparison, heavily-labeled material on page 141. The eluates and supernatants being tested. Still, evidently nothing very striking, or else there would have been a comment on it. Titration of the beads with low pHs, in this case. Struggling, struggling.
On with it, and continues, page 149, [00:26:00] 147. “Higher pH absorption and lower pH elution,” Monday April 5th on 147. And the results on page 149. “We’re seeing that the lower pH elution is poorer than 4. pH optimum is 12,” et cetera, et cetera. Problems, problems.
And then, the last couple of pages are testing EcoRI on iminobiotin nick translated material versus the same with thymidine, with T control material. [00:27:00] Because, with the conclusion that, “The substitution must be quite low, hence only about 30% recovery. The EcoRI cutting is very good.” Provisional conclusion, to end the page on Tuesday, April 6th, “Time is OK, maximum reached in 1-2 hours,” et cetera, et cetera. But, nothing very remarkable. And I know what is coming in Book γ, so here we end Book β at this time.