Oliver Smithies:

[00:00:00] Book zeta beginning on April 5th, 1983 and extending to June of the same year.  Just a continuation of what was going on before.  F33 being put into different vector type [00:01:00] So on Thursday, April 7th, page nine we see PGFY slash 226, PFGY226 slash 33 minis being looked at for orientation and the two orientations, forward and reverse, 33F and 33R, being obtained.  [00:02:00] Some more tests of PGFY series on page 13, Friday, April 8th.  So page 13, Friday, April 8th is a check on various plasmids based on the expression vector from Brosius, PGFY.  And so that’s the parent plasmid.  It’s an expression plasmid.  And we’re putting into sequences related to the proline-rich protein [00:03:00] genes.  So we see PGFY.  Page 19, Monday, April 11th, commenting that these are various tests.  These minis very low level expression.  Growth I should say.  And continuing on this so that a number of samples were obtained, and number eight was streaked out as PGFY234 zeta 21, [00:04:00] being this page number.  Preparative gels being made.  Continuing with P33 minis into that expression vector.  [00:04:51]

OS:

So Thursday, April 14th, page 29, continuing to look at the cloning into Brosius’s 234, some of the proline-rich fragments, in this case 33.  And then we have forward and reverse there at the bottom of the page.  Growing them up more on page 31, Friday, April 15th.  And a complete series of 33 clones being looked at on page 33.  Decided various ones to make.  [00:01:00] Continuing with this work Sunday, April 24th, minis on all of the 33F and 33, that’s forward, and 33 reverse.  And on the next page, 42, there’s a sequencing of these various clones being written up by Qiuliang, my visiting scientist from China, Sixiong.  So there’s 236, 234, [00:02:00] etc., all of them sequenced there.  And all are correct except 241 which really is 218 and so on.  So getting sequences to confirm them.

New tests being considered on page 45, Monday, April 25th, how to induce expression from these cultures.  But abandoned as too low for practical use.  So the next page continues to be tests of induction of expression with a conclusion no obvious differences.  Try adding [00:03:00] casamino acids.  I couldn’t get expression yet.  More expression tests with amino acids on Wednesday, April 27th, page 49.  And continuing on the following pages with gels again on Saturday, April 30th, page 53. Overloaded, dilute all these samples, etc.  Repeat.  So there they are looked at again with an SDS gel Laemmli loading buffer.  Gels, gels, gels, Monday, May 2nd, page 55.

New cultures Monday, May 2nd, more cultures on May 3rd.  And gel [00:04:00] repeats for an antibody test.  Rabbit, goat, primary antibody.  Non-specific chaining.  The rat proline-rich proteins are smeary, positive, etc., etc.  [00:05:00] So some material from University of Illinois basically from Raju’s lab.  Oliver.  Letters for Oliver from Brian Morse talking about Raju Kucherlapati.  Various plasmids.  Talked about on page 63, Friday, May 6th for the Xba supF gene rescue trials.  So I was wanting to test whether we could rescue the fragment, the recombinant fragment.  [00:06:00] These are various samples of plasmids which might help.

One of them being called RK32 which is being ligated into a Charon 3A Xba cut with Xba.  And ligating in RK32.  Ligating good.  Proceed.  Packaging.  So packaging the ligated material to the two bacteria C1A [00:07:00] which doesn’t require supF and C600 which doesn’t have a — C1A doesn’t have a suppressor gene in it.  And C600 slash SF8 has the suppressor.  So the titration in C1A is about 6 times 10 to the 3 per ml.  When the suppressor is there the titration is 3 times 10 to the 8 per ml.  So that’s a good, almost 10 to the 5th selection ability for being able to find something with suppressor in it versus an equivalent material that doesn’t contain the suppressor DNA in it.  And so I said [00:08:00] approximately 1 in 100,000 is correct.  It’s a little unclear but the ability to detect supF is not unclear.

Trying to get the Brosius thing to work better.  Trying with something that was thought to help which is Trasylol.  But it didn’t seem to help.  So Brosius gave some suggestions.  [00:09:00] And again now still testing the ability to rescue supF-containing bacteriophages.  So here is page 79, Thursday, May 12th, checking on the amber status of a Charon 3A and the various derivatives.  Charon 3A has mutations in amber mutations.  And the derivatives thereof are being checked here on page 79 [00:10:00] with a conclusion that some rechecking is needed.

More plaques for minis on page 81.  Amber checks being done on Tuesday, May 17th, but difficult to read with any certainty.  Therefore repeat using the old recipe on the following page.  [00:11:00] That’s rather a different experiment.  It’s not a repeat.  However, it’s phage minis on page 83.  So one is 16 preparations tested and with the conclusion that the lysed samples are badly overloaded and not digested.  And so [00:12:00] repeat it again with less material.  And the repeat on the following page with Xba digests comes to the conclusion that Xba is working all right.  But need to try phenol hydroxyquinoline extract and repeat.  Because I do not believe that all of them are Xba-negative.  There was no digestion with Xba on page 89.

However, here we are now on page 91.  DNA coming back from Raju.  Cosos 17 transformed DNA, the real thing.  So on the left-hand page, 90, is a description of the material.  RK41 from Raju, [00:13:00] which was the human cell line EJ.  Was transfected with about 10 micrograms of Cosos 17 and selected for G418 resistance.  Got about 700 colonies which were pooled, grown up, and the DNA was extracted.  So this is DNA from about 700 independent bacteriophages that had resulted from cloning DNA from 700 colonies of mammalian cells.  So it’s a test of a pool of DNA from 700 transformants.  A test run on 100 micrograms of the DNA [00:14:00] with Xba and Xba plus Cla12 as a control.  And the conclusion.  The digest is fine.  Proceed.  Must be several Xba repeat families because the digested DNA RK41/Xba has about half a dozen clear bands indicative of a simple sequence DNA that can be detected with Xba.  That is what the statement means at the bottom of this page.  That there must be several Xba repeat families.  Here is a real test at last.  [00:15:00] Interim test of amber checks on the new phage preparations.  All a single plaque purified.  This is a test on C600 and also test on W857.  With the titers coming out as on 3AX and 3A delta delta X titers are about 1 times 10 to the 9th.  But on the other two phages, 466 and 857, the titers are about 100 times lower.  [00:16:00] Some confusion in the results requiring retests.

So here is tests on RK41 DNA.  Digested with Xba.  And gels run.  So the sample is there.  And the preparations are made fast, middle, and slow from the digested DNA.  [00:17:00] Being 2.8 to 3.8 kilobases presumably.  The purified fragments that were obtained are rerun on a gel.  Was quite a pretty result.  The fast fraction obtained from digesting RK41 with Xba.  The fast fraction has a range in size roughly 5.5 to 6.5 kilobases.  M is 6.5 to 7.7.  And S is 7.7 to about 9.5.  [00:18:00] And with a comment.  It’s better to pool the M and S and rely on phage size pattern, which we’ll see what that means in a moment or two.

234 expression again.  Vectors being considered on page 99, 234 forward, 234 reverse, 279 forward, and 279 reverse, etc.  So Brosius.  That work again.  Alternating with the work on RK41 as we come back on page 101, Friday, May 20th.  [00:19:00] Ligation of RK41 Xba into the Charon phage, which should be able to accept these pieces.  Preligation and postligation look good.  And so there’s a comment that the ligation is very good, proceed.  So packaging RK41 Xba fragments with sonic extract and freeze-thaw lysate which became a very common job in the lab.

So here we have details of that packaging, page 103, Saturday, [00:20:00] May 21st.  Can see the sonic extract and the freeze-thaw lysate and then packaging of course using a cos site in the DNA of the vector.  Mike Koralewski will plate them on C600 SF8 with a suppressor and C1A which doesn’t have the suppressor.  And one will be just plated simply.  And the other will be used and probed with the IVS2 beta probe test.  That’s looking for IVS2 in the rescued — in the packaged phages.

[00:21:00] Couldn’t see anything on page 103 at 11:00 p.m.  Less than 6.1 on C600 and none on 40 microliters of C1A but later found one on C1A, one plaque on C1A.  Meaning of course that it had — since C1A doesn’t have the suppressor, that it already had the suppressor.  This is a real candidate then that’s being talked about with the balloon on page 102.

[00:22:00] Now on page 105, Sunday, May 22nd, more results of packaging or more packaging being carried out.  And the results that five bacteriophage found on C1A plus two previous means I now have seven candidates.  The packaging method worked but was not any better than previously.  Continuing with ligation of the material and looking at the gels.  [00:23:00] So Monday, May 23rd is ligation roman III of RK41/Xba material.  With a nice comment on page 108, the conclusion, the ligation is now as good as it ever gets with these arms of bacteriophage.  So happy conclusion.

Page 111, Monday, May 23rd looking at W3310IG products.  [00:24:00] Start that page again.  So this is page 111, Monday, May 23rd.  Using one of the bacterial strains that Brosius suggested, W3310 and saying that it looks good.  The level is probably greater than that obtained with DH20 and the product is somewhat more uniform than attempts to get antibody expression.  [00:25:00] So the following page, 113, Tuesday, May 24th.  Going back to Raju Kucherlapati Xba fragment.  So we now have nine positive for phages.  That’s to say they have the suppressor gene in them.  And we’ll proceed with the minis.  Plus three C600 phages which don’t have the suppressor gene in them.  So these were packaged ligation too and all plated at all — plate all of the material.

[00:26:00] C600 titer was 1.2 times 10 to the 7.  C1A had 2 plaques on 4-microliter sample, and about 70 on the rest.  So here are more tests of ligating these.  I’ll check that again.  Let’s read this again.  Page 113, Wednesday, 25th.  A total of at least 80 C1A-derived plaques are now available.  More will be available later.  [00:27:00] Looking at hybridization of these phages to the IVS probe.  No positives out of about 40 candidates.  Ligation.  Then another ligation of the remaining material.

So minis were being repeated.  RK41 Xba from the bacteriophages.  But not any good on that attempt, Friday, May 27th.  [00:28:00] Going back to expression in W3310IQ of 279F and 279 reverse.  Trying to get the Brosius expression vectors to work.  Antibody tests on the following day, June 7th, Tuesday.  But no results yet quoted.  On page 121, Thursday, June 9th, [00:29:00] 5.2-k a probe for screening minis being made.  So that’s 50 microcuries of maybe a total of 100 microcuries of dCTP and dTTP as the labeled material and dATP and dGTP cold material.

And got a probe then with 64,000 counts per minute per microliter.  Here are the antibody results on Thursday, June 9th anti vector PRP, etc.  Conclusion.  Start single colony tests using the antibody at 1/1,000.  [00:30:00] So going backwards and forwards between these projects.  Minis on the C1A candidates.  Page 127, Friday, June 10th, Mike had lifted all the latest candidates.  Results are encouraging, 7 to 10 clear positives.  Test out that’s against the 5.2-kilobase probe.  [00:31:00] Test again with the beta IVS probe.  But something seems contradictory on the following page, page 129.  Xba digest of the minis.  Here there are about eight minis.  The minis are no good.  But try again on DP50 stop that.  So here we go again.  On Friday, June 17th.  My sister’s birthday.  Page 131.  Results of screening and minis [00:32:00] and summary of the sets of experiments with RK41.  Which comes out as being how many positives there are in each one.

So something on the order of many positives.  At least 100 being talked of in this summary.  So at the bottom seven positives sub positive for the 5.2-k.  No.  The seven positive phages that were positive to 5.2 k gave six positive results on the gels.  [00:33:00] And number six was negative.  One of the randoms was also positive.  So roughly 1 in 10 of the C1A positive cells are 5.2-k-positive.

And the minis showing that about half are the Cosos 17 5.8-kilobase piece.  Of the others, two may be the same.  So probably two candidates went in within the 5.8-kb piece.  [00:34:00] So comments.  We must destroy the 5.8-kb piece if possible with an interpretation of the results on page 133.  Where did the 5.8-kilobase fragment come from?  It comes from Xba.  Includes supF and delta beta.  And it can be cut with Cla1.  This should be safe since the Cla1 map of the total locuses is it’s given.  So retrying to get around these technical problems.  Remaking of the RK41 Xba with Cla digest.  And in this case obtaining a fast fraction [00:35:00] about 5.5 kilobases.  And a slow fraction about 8.5 kilobases.  Which later on was a trick that was used to determine whether recombination had occurred in the bacteria or had already occurred in the mammalian cells.  This ability to fraction the DNA.  But in this case it’s not being used for that purpose.

And so this is the remaking just going back to the purpose of this experiment.  It was [00:36:00] to make an XbaI digest and fractionate it to get out the 7 to 8-kilobase material and get rid of the 5.8 material, which was a problem.  Cut with Cla if necessary.  More digestions now with Cla of these fragments.  And so RK41/Xba X was digested with Cla1.  But no Cla1 digestion.  And also some comments on the size of the fraction.  With a remark that the current procedure [00:37:00] is turning out to be too risky.

The top end is self-limiting by size.  And the top end will be limited by the chance of supF getting next to an Xba, etc.  Some interpretations of what it might be with a fresh start on Wednesday, June 22nd.  But no detection.  So happy birthday, Roger William Smithies on June 23rd, Thursday.  This would be — let’s see, ’83, the fifty-eighth birthday or something like that.  So starting again.  And taking the RK41 [00:38:00] Z fur 91 DNA and digesting it with Xba.  The digestion is fine.

Ligation trials.  Near full-scale.  On my birthday, Thursday, June 23rd.  Good ligation perceived.  The proportions are good.  Packaging the Z143 material.  Freeze-thaw lysate and sonic extract.  And the usual sort of recipe.  Preliminary results obtained already on Thursday, June 30th.  That about 20% to 30% of the C1A positive cells are also 5.2-k-positive.  [00:39:00] And none are yet IVS2-positive.  Checked 288 on C1A, 104 are 5.2-k, that’s 36%, not bad.  So the method is working.

More ligations on Wednesday, June 29th, page 147 and 149.  About 560 micrograms of RK41 is still available.  But this is all.  And 1 or more milligrams of RK43 and RK44.  Presumably this is new material from Raju of the same general type.  No particular entry of it there.  [00:40:00] The last page of this book is a usual type of pages for describing current stages.  Second version.  September 5th, 1983.  Construction of tk-neo fusion and delta P tk-neo fusion with SV40 origin, etc., etc.  This is for a different type of experiment that we no doubt will talk about later TNFUS [00:41:00] and delta TNFUS these are tk-neo fusion genes and tk-neo fusion gene, the deletion which I expect will become clear eventually.  End of book zeta.  [00:41:21]