Oliver Smithies:

[00:00:00] Book z, starts on August 24th, 1980 and goes through June of the same year.  It begins with trying to work out the conditions to anneal the S1-labeled probe to DNA, a little obscure page, getting involved in the rate of hybridization, at the length of time needed to anneal a probe to a DNA bound on a filter.  And, several conclusions from this Sunday, August 24th experiment, “40 [00:01:00] degrees Centigrade just begins to cut off the label with the S1.  30 and lower, essentially the same as if no enzyme, so use 37 degrees,” and conclusion too that “The conditions for S1 nuclease are very good, as judged by the zero,” et cetera.

So some calculations of specific activity on Monday, August 25th, on page 3, talking about what the amount of label that was needed to detect a band.  And, here’s some comments of what other people used.  For example, Gordon Dixon used cDNA [00:02:00] at 0.5×10^8 counts per minute per microgram, and Harold Weintraub used about 0.5×10^8 also, with each of — with up to 100μg of DNA, so that’s the cDNA was the counts, and the amount of DNA was up 100μg.  Then high specific activity could be made in the scheme for trying to make high specific activity [of/for/the?] DNA.

Going on with this general type of investigation, Monday August 25th, page 7, “‘Single’ copy gene probe for S1 tests.” [00:03:00] Starting with the plasmid containing 2.[6?]kb fragment of G gamma, and what might be done.  And, the result of the tests say that there was a very poor run, but proceed anyway.  That gel run, so this type of test continuing on Tuesday, August 26th, and thereafter.

Some ideas of new probes, different conditions on page 15, Thursday, August 28th.  3’ labeling, or 5’ labeling. [00:04:00] And one of which depends on nick translation from a four-base overlap, and the other depends on a 3’-5’ exonuclease in the absence of triphosphates, et cetera.

So, these general types of experiments being continued.  So a new labeling procedure continued on page 21, Friday, August 29th.  “Looks very good, increase the time with enzymes, after looking at the DNA on the gels” is the conclusion.  The usual problems trying to get good probes.

September 2nd, test of the length of the 3’ label.  “Results [00:05:00] show that higher molecular weight than the 20% gel which was being used here, acrylamide gel, was higher than the gel could handle, so repeat with lower amount of acrylamide.”  And the following page shows, instead of 20% acrylamide, an 8% acrylamide gel, with the results showing that the size range 200-400bp is almost up to the maximum, et cetera.  Trying to get good probes.

Further tests on new product, Friday September 5th, and on Sunday, September 7th, a nice comment here, “Set off for [00:06:00] Europe in a Cessna-210 with Field Morey.”  That was the beginning of a very exciting trip, where we set off from, in the end, Goose Bay, Labrador, and flew to Reykjavik, Iceland, without any extra tanks in the aircraft, and landed with very little fuel, but we did beat the world record for flight time of an aircraft of that standing.  Of course, I wasn’t really very important in this trip; it was Field Morey, but it was very enjoyable.  And, we went in a Cessna-210, a single-engine aircraft, and there is a rather nice image of Field Morey [00:07:00] and myself somewhere in the library of images, getting ready for early departure.  I think it was in the Nobel lecture, actually, I think it got into the Nobel lecture later on.  But it’s saying, that trip lasting about three weeks, which ended after more than 10,000 miles, and after more than 60 hours of flying.  Two world records, Goose Bay Laboratory to Reykjavik, Iceland, 8 hours and 17 minutes.  The old hours was 8 hours and 27 minutes, so we beat the record by 10 minutes, and we had between three and seven gallons left, and Reykjavik, Iceland, to Prestwick, Scotland, which at that time, was still being used as a landing place on trips from overseas, [00:08:00] and that was four hours and 59 minutes, whereas the old record was an hour longer, so that one was by a good, big margin.  Still our 17 minute record held up for really 20 years, so it was quite a feat.  And certainly was enjoyable and exciting, and took to that some other account of my days of flying.

Back to work on Wednesday, October 1st.  “Repeat of new end labels with cold fill-in.”  And getting counts, somewhat reasonable, 1.3×10^6, 0.9×10^6 on page 35. [00:09:00] Thinking of changing T4 DNA polymerase in place of E. coli to increase the exonuclease activity, technical (inaudible).

And seem to have been satisfactory, as judged by page 41, which stated, “The results are very clean, products by all methods,” et cetera, et cetera.  Next, by increasing temperature.  And the following says, “Test of increased temperature.”

Reviewed on Monday, October 6th, page 45, the temperature effects [00:10:00], with comments on the different possible things that are happening.  And the following page, “Minimal labeling anywhere.  Perhaps it’s [via an EDTA?]??”  Obviously puzzled, a couple of exclamation marks.  This was a repeat of Thursday, October, with very different — “It must be the Sephadex.  Do a simple test,” et cetera.

And then the result on the next page, “OK again.  Most likely the problem is the Sephadex (inaudible).  Repeat the experiment with several different temperatures. [00:11:00] 1.3×10^6 counts, with Klenow at 15 degrees.”

Back on track, hopefully, on Tuesday, October 7th, and with a comment, the conclusion, “It’s beginning to work.”  And I’m getting something of the order of 10^6 counts.

And beginning to think about the Alu family on — that’s a family of repeats common in eukaryotic DNA, on Wednesday, October 8th, page 55.  [I can?] test TK, pBR322 plus or minus [00:12:00] Alu for ability to transform, to see whether Alu made any difference.  With some diagrams about what might be tried.

Exonuclease being tested, the following pages.  More refined ExoIII tests, and so on, on Friday October 10th, page 63, time series ExoIII.  And, temperature effects.  And the conclusion on page 65, “The reaction is virtually complete in 15 minutes of 30 degrees, (inaudible) test,” et cetera, 37 for 30 minutes versus [00:13:00] no Exo (inaudible).”  Usual technical struggles.

Repeating with Exo under milder conditions on page 71, and new Sephadex procedure.  Tried to get better columns.  Similar things continuing October 16th, Thursday, page 77.  And lots of radioactive bands on material made on page 71. [00:14:00] With Exo under milder conditions, and testing the product on Thursday, October 16th.  A shorter time re-run, et cetera, on the following day.

And rethinking on Saturday, October 18th, page 83, with the feeling that the problem is that the Klenow end labeling for length determination fails to score molecules that are completely ExoIII.  It just catches the intermediates, so that poor progression in the time series. [00:15:00]

Some tests with M13, MP7, considering working again with the pseudo alpha gene 30.5, which I called “pseudo alpha 3” for short.  “Need M13 subclones of alpha 1, pseudo alpha 3, and pseudo alpha 4, M13 clones.”  This is the beginning of thinking about using M13 bacteriophage, single-stranded [00:16:00] bacteriophage that has the advantage then that one can make complimentary DNA, and this was — so this is start of that thought, on Monday, October 20th, page 13.  And trying to get at least three different clones from into M13, as mentioned, alpha 1, pseudo alpha 3, and pseudo alpha 4, which one of the first is 4.2kb, the next 2.3, and the next 1.7kb.

And so, it begins with Tuesday, October 20th, trying to clone the 30.5 band bagel piece into [00:17:00] M13.

Looking at gels Charon 3A delta lac, and Charon 3A were containing 30.5 marker mix on page 89 to get some good markers of size.  And, concluding from this that “The cuts are complete, proceed to cut out the [doublet?], which is 3928 and a 4480 fragment” [00:18:00] that I evidently knew was suitable from Charon 3A 30.5.

Second digest of the fragments on the following day, Wednesday, October 27th.  Cutting out different pieces, and checking on the fragments, and DS, presumable double-stranded M13, MP7, Bam digest, [00:19:00] with the ligation being evidently satisfactory on Friday, October 24th, but with a comment that, “The ligation can good, but can expect a fairly high background of blue plaques from the uncut M13,” so M13 gives blue plaques, and because presumably, it had a lac gene in the genome, which disappears in ligation of something in between the promoter, and the lac gene.

Saturday, October 25th, and Monday, October 27th, M13 transfection. [00:20:00] “Nice set of plaques, but no color yet.”  So, the plate were left at room temperature to see if the color developed, and by Sunday, October 26th, “Two standard plates of [37?] are showing blues, about 100 per plate, and whites, about three per plate,” meaning something on the order of 3% of them would be here, as thought to have captured some DNA.  So repeat of the transfection, Monday, October 27th, and more experiments of the same type, (inaudible) M13 plaque, and others, conclusion, “Success.  5 and 8 are of one type, [00:21:00] and 1, 3, 4, 6, and 7 are of other.  Only one is negative.  Do double-stranded DNA gel.”  This is page 101, Monday, October 27th.

More, now thinking about the primers to use for transcripts of 30.5 on page 103, Tuesday, October 28th.  And continuing with the problem, (inaudible) gels and digestion of sample, Thursday, [00:22:00] October 30th, with a peak — prep gel preparation shown on the following page, 110, with cutting out one, two, three, four different fragments.

Thinking about Alu family binding proteins on page 113, Sunday, November 23rd.  Thinking about the possibility that single-stranded regions formed by RNA polymerase reactivity [00:23:00] from Alu sites are recognition sites for DNA-binding proteins used in replication, a rather complicated hypothesis.  The experiments would be to make single-stranded M13 of bone Alu strands and use them as definitive column substrates, et cetera, for purifying the proteins of they existed.  And a mention of [Shi Shang He?] for the first time, a scholar from China who was very energetic, and did some good work in the lab.

And now, a rather interesting beginning of this whole idea of replication which [00:24:00] then consumed me for quite some time was the idea of origins of replication being in the notebook on Tuesday, April 7th, page 115.  (inaudible) question is, “To distinguish the possibility of just a single-stranded — a linear piece of double-stranded DNA, and DNA in which there was already a replication fork,” with the usual hypothesis as to how the two strands are replicated at the fork.  So the idea, the question is, “Distinguishing non-replication DNA from a replication piece of DNA with a replication fork, [00:25:00] cut with an enzyme, first RI, and DNA polymerase label, and expect to see that it will be possible to capture the single-stranded DNA on paper,” et cetera.  “Further refinement, might cut with a four base enzyme, and map the [hot?] pieces to get sites of origin and direction of replication.”  So here, I’m thinking about origins of replication and the direction of replication.

So, working with Dixie Mager.  She synchronized cells with 14 hours in 2mM thymidine, [00:26:00] et cetera, and with K52 clone [four?] cells, and she could synchronize them, unsynchronize, and then, in the right way, get them in the S period, and et cetera, et cetera, a method of synchronizing cells using thymidine blockage.

With Saturday, April 11th, 119 page, “First phenol layer is so clean, and volume of cells so small that one phenol is probably adequate,” et cetera.  And, look at, on the opposite page, comments on, “Look at the product in the electron [00:27:00] microscope.  Should be about one of a hundred forks.”

And, here then in page 121, there’s a gap which is made through [gene?] pause to rewrite and regroup, and then to publish a hypothesis of origins of replication controlling gene transcription that I’ve referred to earlier, and so, there is a diagram of this scheme on page 121.  And, we’ll look and see where the reference is.  This is writing a paper which eventually appeared in The Journal of Cellular [00:28:00] Physiology supplement in 1982, the control of globin and other eukaryotic genes in which I’ve spoken about earlier, the hypothesis being events that eukaryotes control different parts of their genome by the selective use of origins of replication, which I have spoken about earlier, though I don’t know just where about.  But there in my book, page 121, is the Figure 1 of that paper.  And so here I am beginning to test as usual a hypothesis.  The hypothesis is a bit wild, and the tests are quite consuming to try to find out whether globin genes when transcribed — when replicated from upstream of the gene [00:29:00] are active, and when replicated from downstream of the gene are inactive.

And so here on Friday, June 26th, page 123, is a statement of a problem that the aim is to see if Alu sequences, because that thinking of Alu sequences as being the start of replication, and particularly, the A gamma delta Alu pair, which I would call gamma delta Alu, are potentially origins of replication, so there is a map of the A gamma pseudo beta 1, delta beta region, pseudo beta 1 being a pseudo gene, 30.5, and delta, and beta, and in between is the 3.1 piece [00:30:00] with an Alu sequence, and the idea being to clone candidates into pSVO10 vectors, “pSVO” being “Simian Virus Origin of replication, 10 vector,” so that one could then replicate the DNA, and depending on how the plasmid was made, in two directions.  So this is the beginning of the test to test this idea.

And page 125, continuing the origin of replication vector, which is now being called “Ori-vector,” and ORI, test vector [00:31:00] for Ori-vector.  The idea (inaudible) being that one could test whether a piece of DNA can act as an origin of replication, and so having a [redoubt-readout?], which at the moment is not specified.

So I’m attempting to clone into the Ori test DNA, pSVO10, HindIII, [00:32:00] et cetera, on page 129.  Test gels on the following page.  “The ligation is good, but the Bam is bad.”  Page 131.  And rechecking on page 133.

Plasmid checks on page 137, June 29th.  And 131, 139, and continuing to check these plasmids that had been put into Ori-vector. [00:33:00]

So, Friday July 3rd, SV Ori continued.  And looking at various preparations.  “Digesting — no digestion with HindIII of two different candidates.  One more [00:34:00] HindIII than advertised.”

So restarting the work on Ori-vector on Friday, July 3rd, “Ori-vector restart,” with a conclusion that, “The HindIII cut is OK, and the ligation is OK,” and so, Saturday, July 4th, tests, the conclusion again, that “R1 is good,” et cetera.

Similar comment on page 153, July 6th, “Ori tests continue.  The ligation is OK.  The starting proportions were good.” [00:35:00] And that ends Book z.