Oliver Smithies:

[00:00:00] Here we are in Book ε, beginning — no, not Book ε — Book η, September 6, 1983, going through to October 22nd of 1983, so a fairly short interval of time and a fairly thick book.  We’ll see what it has.

It starts off with, and the TNFUS — maps of various sorts or, in this case, a computer printout.  Beginning to have computer printouts of digestion maps.  So this is a construction of the pSV40 Neo plasmid, TNFUS, which is [00:01:00] SV40 driving a Neo gene — SV40 promoter driving a Neo gene — the origin of replication.  (pause; break in audio) So this book begins [00:02:00] by talking about constructing a pSV40 Neo fusion plasmid, which would use the strong promoter, SV40, driving the Neo gene — so using that, presumably, to clone, in various expression, fragments that we were interested in.  So this is the beginning of that book, Tuesday, September 6, page 1.  (pause) Continuing over several pages [00:03:00] of the same general type.  (pause)

Starting with page‑1 material, which was pSV‑010.  (pause) [00:04:00] Continuing with various digests, through the next few pages.  More experiments with this pSV‑010, Thursday, September 8th, page 21.  (pause) [00:05:00] (pause) On page 23, and Thursday, September 8, which is a BamHI digest plus or minus Vagl2 plus or minus EcoRI, on pSV‑010, etc.  Rather imaginative interpretation of the gel picture sketched below, on page 22.  But not happy with it.  Friday, September 9th, page 25, pSV‑010, [00:06:00] again, again.  Transformations of various plasmids into C600 cells.  (pause)

Minis, on Monday, September 12th, page 33.  ε‑‑ I mean, η‑31 minis process, using standard protocol.  Set up the digest with BamHI.  And the conclusion, on the opposite page, is that — “Highly successful.  2,075 base pairs and 2,915 base pair‑‑” for the two types of candidates, “Grow up #6,” which is called TNFUS.Int1, intervening sequence 1.

Beginning to remember a little bit about [00:08:00] why that was the thought, that an intervening sequence might be needed to get expression.  So here, on Tuesday, September 13th, page 35, construction of TN fusion, intervening sequence 2 into…  Attempt to get it, in one way or another.  Second try, on page 37.  Third try, on page 38.  And large-scale on the Mini #6, on page 41.  But the yield [00:09:00] was too low to continue.  So continuing in the same way.  (laughs) With a comment on page 42, “Messing around with gel techniques.  Conclusion‑‑” etc.  Partial Bams being tried, on Friday, September 16th.  Feeling very happy about it.  “Bam is about ideal.  Supercoil is now zero and one cut is very heavy.”  [00:10:00] (pause)

So here is intervening sequence 1 being talked about, on page 51.  “TN fusion int1, single-Bam cut.  The fragment is good.”  So getting these different intervening sequences under control.  [00:11:00] Ligating pieces together, on Tuesday, September 20th, page 53, with the conclusion that — “Progressing well.”  And as η‑53, Ligation 2.  And many gels, on the following two pages.

Mini digests, on Saturday, September 24th, page 59, with the conclusion that “#20 is what I set out to get,” etc.  [00:12:00] So #20 is happy.  “#14 and #20 re‑digests,” Sunday, September 25th, page 61.  “Proceed to the next step and grow up η‑55, #20, as” TNFUS.int1, intervening seq‑‑ “‑‑int2,” that is, intervening sequence 2.

A whole bunch, again, of restriction maps, printed out on a computer.  Beginning of our use of computer programs to analyze DNA [00:13:00] sequences — or perhaps not the beginning but at least an example of them.

On the September 26 page, 69, one of the few times I had to talk about patenting anything.  “Phoned an idea to EJFCMI, genetic engineering of polyvalent antigen.”  No idea [00:14:00] about any more…  It never came to any conclusion.  To make a plasmid and then have a polymeric antigen and use that then to have a better antigen for making antibodies and the like.  (pause) [00:15:00] (pause)

clNA filling in, on fragment, and Xmar3, etc., etc., Wednesday, September 28, page 75, trying to have a 20-base pair fill-in fragment.  [00:16:00] Hoping to see a map of what this is all about.


Oliver, Jada Lewis is on 15.  Oliver, Jada Lewis on 15.

(break in audio)


So here we are on page 79, Friday, September 30th, continuing to work on making this pSV2 Neo plasmid, driven by the SV40 promoter, and driving the Neo gene.

Then, on page 83, two, three pages later, Monday, October 3rd, looking at a partial digest to make what I — written down as int3 TNFUS.  But it really isn’t and was corrected to be TNFUS.  That’s tk [00:17:00] Neo fusion, a tk Neo fusion gene.  With a diagram showing the tk promoter and the Neo gene and restriction sites that could be used to insert material, if that was desired.  But anyway, there is the tk gene driving the Neo gene.  Was so aimed to be constructed.

Just continuing to do the things needed to make that sort of product.  Fill-in fragments and so on and so forth.  Complete digests of the tk Neo fusion, with the intervening sequence 2, now, on page 95, Wednesday, October 4.  But used the wrong temperature.   (pause) So an Xmar single cut being re‑cut with Smar1, from page η‑97, being looked at, on Thursday, October 6th, page 90.  And just various things needed to try to get the desired product.  Prep gel again, on October 7th, Friday, page 103, to get η‑93 Smar1 single cut — just a necessary thing, to proceed.  And pages and pages of the same general type.  So in a sense, a digression from the targeting.  Because one is waiting to proceed with that separately.  (pause) [00:20:00] Preparative gels.  And Xs single-cut test gel, October 11th, page 113, Tuesday.  “Conclusion:  Well hit.”  Unusual comment for me.  (pause)

Here was a little return to Cla 17, on Thursday, October 13th, page 123, “Single-cut BstXI on Cla [00:21:00] 17.  Make a single cut, S1, ligate, and sort into the control cos.  And beta cut,” etc.  So the desired product is Cla 117.  And the control product is Cla C‑117, meaning control for 117.  So Cla 117 is going to have an Xba side-cut.  And so Cla 117 DNA was cut with Xba1, and for varying lengths of time.  And [00:22:00] eight minutes’ digestion was “close to ideal and contains about 6mg of DNA.”  So, “S1‑nuclease it and proceed” is the conclusion.  But jumping back, the same day, to ligation of TNFUS, on page 125.  So these two things are, one might say, overlapping day by day or even in the same day.  Thursday, October 13th, page 127, η‑123 eight-minute sample, the best digestion being treated with S1 nuclease.

[00:23:00] So a check of the ligations, on page 129, October 14th, Friday.  And labeling with the end.  (pause) This was material that was labeled and about 10% of the label is now slow, large amount of small — of [00:24:00] presumably the radioactive precursors.  But there is a labeled band, which is the product one is wanting.  But not much help on ligation of this material.  So that was looked at again, on page 131, Saturday, October 15, ligation tests, with the “Conclusion:  More ligation on TNFUS, judging by the counts, but still can’t see anything.  Cla 117 is getting better.”  Although I can’t see anything by radioactivity, apparently, there [00:25:00] are plenty of clones.

Minis on Cla 117 and Cla Control‑117 — minis.  “Transformations done with a new method by Natalie Borenstein gave about 50 plaques with 10ml and many with 200ml.  Picked 11” and proceeded.  The conclusion is, unfortunately, however, that “They are not very good minis.  And no signs of the 7.2‑ or 10.1‑KB fragments,” which were diagnostic either of C‑117 or of 117.  [00:26:00] The map being there, to show what this is all about.  TNFUS being digested, on Tuesday, October 18, page 135.

Next page, again, Tuesday, October 18, Cla 117 and Control‑117 minis.  Much better growth.  And these are now heavy minis.  And, “Xba is incompletely digested.”  But [00:27:00] various candidates.  “Five, 9, and 13 are going to be the starting plasmid.  Foruteen is no good.  But conclusion that “One should increase the Xba time and the amount of Xba.”  And did so, on the following page.  Some problems, that are not needed to understand.

October 19th, class work, on hold — “Class work hold,” meaning that I have to stop working for a while in the lab, in order to do class work, teaching.  [00:28:00] And Mike will then go on with various things.

Not completely dead yet, though.  October 19th, Wednesday, page 143, the Cla 117 minis are continued.  But with the “Conclusion:  Six are the same as previous.  Number fourteen gave a 3.1‑KB and a 6.0‑KB fragment.  Neither is predicted.  But 10 is different and gives a 7.5 and a 4.1.  Perhaps the map is wrong and these are correct.  Anyway, try 16 more before retransforming.”  So more minis, from TNFUS [00:29:00] minis.  “Two very promising candidates, #13 and #14.”

Using a different enzyme.  “Check on the tk Neo fusion genes,” on Thursday, October 20th, page 149.  And as again jumping back to Cla 117 and Control‑117 minis, October 20th, page 151, the conclusion that “#26 and #35 or Cla 117.”  “Redid Digest #24 and #36, to see if they are Cla Control‑117.  [00:30:00] So…  And that’s…  “26 and 35 are Cla 117.  And 24 and 36 might be Cla Control‑117.”

The following pages or the last two pages of the book are concerned with the effect of blocked ends on recombination.  “Test by Raju Kucherlapati on a suggestion from me show about 30‑fold increase in recombination of delta left Neo and delta right Neo plasmids,” by introducing a deletion cut, by introducing a cut into delta left.  So by introducing a cut into the plasmid, one could [00:31:00] increase recombination greatly.  And we did, in the end, publish a paper — that cutting within the plasmid increased the recombination many-fold.  And I’ll try to find that reference, in ending this book, Saturday, October 22nd.  [00:31:25]


[00:00:00] — work ended up with a publication, in 1984, that “Homologous Recombination Between Plasmids in Mammalian Cells Can be Enhanced by Treatment of the Input DNA.”  That was in PNAS, 1984, with Raju as the first author and me as the last.  [00:00:26]