Oliver Smithies:

[00:00:00] So let’s delete that previous section.  I think that some comment is needed on what was going on in terms of publication during the 1990s as it were.  Because although I talked at great length about what was in my notebooks, gene construct, I haven’t said much about what came out of all this.  So it’s perhaps worth going back a little bit and seeing what was happening.  So the first experiments related to blood pressure began essentially in [00:01:00] — hold it a moment.  [00:02:00] Yes.  The publications were beginning around about 1995.  Masaki Ito and various other people with Tom Coffman at the end.  We started with a paper on the regulation of blood pressure by AT1A receptor for angiotensin II and then Hyung-Suk Kim talked about the — published the “Genetic control of blood pressure and the angiotensinogen locus,” one of the more important papers I think.  Krege in 1995 described the computerized tail-cuff system for measuring blood pressure which I’ve referred to.  [00:03:00] And then in 1995 looked at the effect of the changes in the angiotensinogen-converting enzyme or knocking out the angiotensinogen-converting enzyme and showing that heterozygotes and homozygotes — that the heterozygotes didn’t have much change in blood pressure.  Simon John in 1995 looked at the atrial natriuretic peptide changes and so on.  Then Ed [00:04:00] Shesely was the senior author on a paper in 1996 on the effect on blood pressure of mice lacking endothelial nitric oxide synthase.  Simon John in 1996 was publishing again on blood pressure and fluid-electrolyte balance in mice with reduced or absence in the atrial natriuretic peptide.  [00:05:00] Natriuretic peptide receptor in 1997 with Paula Oliver as the leading author and Nobuyo Maeda as the senior author, Mice lacking natriuretic peptide receptor A, cardiac hypertrophy, sudden death.  Role of the angiotensin Ia receptor and [00:06:00] Ib receptor in 1997 which Tom Coffman was the senior author.  The work which was quite seminal in the angiotensin-converting enzyme gene mutations, blood pressure, and cardiovascular homeostasis in 1997.  [00:07:00] And then changes then going to estrogen receptor and so on.  So there were a number of papers being published in that time.

This is book capital J’ beginning June 15th, 1999 and going up to June 11th of 2000.  So still trying to get to the CRLR construct for the mouse.  Extending the sequence of CLR Mus [00:08:00] on Wednesday, June 16th and making some new primers.  And then continuing with the construct.  Still modifying the Dupdel plasmid removing one of the AscI sites on Tuesday, June 22nd, page seven.

Continuing the same work June 23rd on OS and Roger William Smithies’s seventy-fourth birthday Wednesday, June 23rd, 1999.  [00:09:00] Continuing with the modification of Dupdel single AscI site page 19, Monday, June 28th.  Using the Dupdel now and inserting into Cla site 3’ homology being inserted on Tuesday, June 29th, page 21.  [00:10:00] On page 25, Tuesday, June 29th, I’m now talking about natriuretic peptide receptor C, the so-called clearance receptor.  With a comment.  Since we failed to generate an NPRC duplication Nobuyo suggested removing the intracellular domain, etc., etc.  So make an idea for the next project.  [00:11:00]

Various modifications of Dupdel still being sequenced on page 27, Friday, July 2nd, and going on with CRLR 3’ homology insert on page 29, Friday, July 2nd.  Just straightforward plasmid constructions again.

Now putting or attempting to put in the 5’ homology assembly Saturday, July 3rd.  Inserting into a Cla site the fragment needed, [00:12:00] 1.1-kilobase fragment insert.  Making plasmids.  Making plasmids.

Wednesday, July 7th, page 37, rather satisfactory day.  Conclusion good.  The deduction was correct.  Can use clone 11 as one orientation and clone 14 as the other for this construct.  Correct sequences.  [00:13:00] Getting ready for the final assembly on page 39.  First try of the final assembly of the targeting construct on page 41 with a diagram of what it’s going to look like.  Preparation of a control construct on page 43.  [00:14:00] Continuing thereafter.  So leading on page 49, Saturday, July 7th.  Final sequencing checks on T19.  Looking at the CRL rat sequence compared with the CRL Mus sequence again.  [00:15:00] And various comments on what might be found and what sequences were obtained.  Looking to the sequence where exons 1, 2, and 3 are used as the left-hand homologous sequence and then a deletion and jumping to exon 5 with a chain terminator neo.  So going to essentially delete everything after — or delete a [00:16:00] section of exons 4 and 5.  That basically with some homology after the neo gene.  It’s a straightforward deletion.  Various experiments to confirm the correctness of the constructs.  Following pages.  With primers for the control on page 55, Tuesday, July 20th and Thursday, July 22nd.  [00:17:00] Checking all the critical regions by sequence comparisons with Kathleen.  Only surprise was that I had forgotten that my exon 1 is equivalent to mRNA exon 1 but my exon 2 is probably really exon 3.  Since exon 3 is not in the sequence.  A little bit obscure.  But something odd.  But not apparently important.  More computer modeling again with Nobuyuki Takahashi trying to get some modeling of the tall ascending limb cells.  [00:18:00]

Thinking about surrogate markers for hypertension and hypertrophy on Sunday, August 1st, page 59.  Thinking about several peptides or peptide hormones as surrogates for blood pressure.  The difficulty is the low concentration of many of them.  Limited availability of plasma and the laborious assay.  [00:19:00] Not very promising.  Going back to thinking about targeting efficiently on Monday, August 9th, page 61.  Reviewing the literature for my hemoglobin grant renewal.  Review of the paper claiming that gene targeting is really closer to 10 to the -1 than the usual estimate of 10 to -5.  But looking at it though it didn’t seem to be anything very different between their protocol and our.  But a procedure [00:20:00] to try to vary the voltage used in electroporation being listed on that insert pages to compare this person whose paper we were looking at.  Let’s just see what it is.  The paper was by Templeton, Roberts, and Safer.  “Efficient gene targeting in mouse embryonic stem cells.”  And so we looked at our method and compared it to his method.  [00:21:00]

Not very much different.  But we continued to think about it Monday, August 9th.  Can adapt his protocol to G418 and ganciclovir method.  Page 63, Monday, August 9th.  To compare the two methods.  [00:22:00] Thinking about insulators on Friday, August 27th, page 65.  To test the proliferative advantage concept in the beta-globin region I proposed to surround the transgene with silencers described by Gary Felsenfeld et al.  And will receive the plasmid today and Seigo Hatada will grow it up.  Continuing work on the CRLR exon 4 deletion in the targeting vector page 67, Tuesday, August 31st.  [00:23:00]

More thoughts about modeling Sunday, September 19th.  Trying different models, computer simulation with the Stella modeling.  [00:24:00] More on insulators.  Somewhere around about September 25th, page 73.  So that’s September 25th and the next entry is Friday, October 29th with grants completed nearly three months in doing them and sent off before the deadline.  So thinking about antibody staining of red blood cells.  The idea being to be able to determine a small number of normal cells amongst a large number of sickle cell hemoglobin-containing cells or vice versa.  [00:25:00] With anti-beta A and anti-beta S antibodies.  And I remember spending quite a lot of time on trying to make these work and trying to make the red cells so they wouldn’t lyse under various conditions.  And making them into cross-linked and as I called them hardballs.  So Saturday, October 30th, page 77 it’s hardballs again.  And the antibody staining of hemoglobin has three phases.  Cross-linking, permeabilization, and antibody staining which I was trying to make work.  [00:26:00] Wednesday, December 15th a little note, page 79.  I had to quit the grants, etc. for experimental therapy.  Meaning I had to do an experiment, I was fed up with writing grants.  And so I did an experiment with respect to pMP8neo, which has the following structure.  And modify it to change the structure of this construct.  Modification continued [00:27:00] Friday, December 17th, page 83 and 85, a fresh start.  Still more December 19th and so on.  Here is a construct related to the having the insulators and making a site between two copies of insulators in which to insert the thing that needed to be protected by insulators.  Now another digression or progression, whichever one wants to call it.  On Monday, December 20th, page 91.  Now I’m thinking about silencers.  A nice way to put two silencers into [00:28:00] pSKB1 is to use Xba knockout with a Bam.  Change to a Bam linker, etc.  It should be fairly easy.  Anyway there is the beginning of experiments on pSKB1 deleting the silencers.  Linker insertions, etc., making the construct.  Still working on Christmas Day, Saturday, December 25th, page 99.  Plasmid constructions.  Plasmid construction.  Here Tuesday, December 28th I’m working [00:29:00] and the next entry is Sunday, January 16th.  And what happened in between is quite entertaining because it says at the top of page 105, “Millennium 2000 celebrated in London on Greenwich meridian.”  I went to London with Nobuyo and our friends Marshall Edgell and Cora-Jean Edgell.  And my brother was fanatic about time and place, had us stand with one foot on one side of the Greenwich meridian and one foot on the other side of the Greenwich meridian near the place where the ball really falls, so that we were in two centuries at the same time when 2000 came along.  [00:30:00] We had a very entertaining time and a happy visit.

Back to sequencing again.  Insertion of insulators page 111, January 27th, continued.  Page 113, restarting insulator insertions again because it didn’t work.  Insulator insertion, page 117.  Silencer dimer insertion, page 119, on and on.  Dimer sequencing, page 121.  Possible hematopoietic stem cell [00:31:00] targeting scheme was talked about on page 123, Tuesday, February 29th.  Suggested in my hemoglobin grant renewal using the truncated EPO R1 receptor as a surface antigen to select immunologically for hematopoietic stem cells that had the receptor.  And some various complications of this thought being talked about on page 123.  On 125 I’m now happy.  My grants in review is off.  And I’m back to TNAP c-kit with a clear statement of what that was all about.  Tissue-nonspecific alkaline phosphatase, TNAP [00:32:00] replacement with c-kit, the receptor for stem cell factor.  So replacing the alkaline phosphatase.  The background, TNAP has been known for many years to be a marker for cells that eventually become germ cells.  My thought was since heterozygotes for TNAP gene disruption are essentially normal I can replace the TNAP coding region with c-kit so that the receptor would be made earlier in development than usual and might enable ES cells with this TNAP c-kit to outcompete wild type cells in populating the gonads.  So this is a clear statement of what all those experiments with TNAP c-kit were about.  It’s quite a good summary and easy to understand even [00:33:00] at this time and with the targeting plasmid being very carefully documented and shown.  So page 124 is a good page in the book, Tuesday, April 18th.  A little internal review as it were, Monday, April 24th, page 127 on the failure to get recombinants in C57 black 6 ES cell line.  They were difficult to get recombination with C57 black with B6 ES cell lines at that time.  Hyung-Suk Kim had succeeded in getting gene disruption duplication [00:34:00] C57 black cell line from Switzerland but it was unstable.  It had an unstable karyotype and never went germline.  And Tomoko Hatada has had no success with thaw number two, etc.  So here’s some discussion about why it isn’t working and what we might do about it to make B6 work.  Subsequently of course people found out how to do it and it’s not uncommon, though strain 129 was always better.  Trying to think about isolating hematopoietic stem cells on Friday, April 28th, page 129.  [00:35:00] This was experiments that Seigo Hatada was doing.  He was trying to isolate, to target hematopoietic stem cells in such a way that either they would grow better or poorer.  Particularly of course to get recombination in these cells so that we could hope to correct hemoglobin mutation.  But working with the HPRT gene as a model.  pMP8neo and the HPRT locus.  [00:36:00]

Looking as if somebody was thinking about a patent here, but I don’t remember who because on the opposite page is a construct for generation of single copy transgene of the truncated EPO receptor at the HPRT locus.  From the paper eventually [00:37:00] that was published with Suzanne Kirby’s work with the truncated EPO R1 receptor in the HPRT locus.  But there’s some thought about a patent here because it says a printed version here saying, “This invention provides a general method for isolating hematopoietic stem cells or other cells in which a faulty mutant gene has been corrected by homologous recombination.  Preamble.  “It’s generally accepted that the direct correction of a faulty gene such as mutant form of beta-globin gene which causes sickle-cell anemia would provide an extremely useful means of treating this and other genetic diseases.”  So it’s just as it were trying to think about writing a patent [00:38:00] related to the old idea long since past of correcting genes.  But anyway it talks about a patent that we’re going to make.  Targeting faulty gene upstream sequences.  It never came to anything.  Never did apply for a patent.  Anyway it was too late.  So still the truncated EpoR receptor work was nonetheless useful.  At least valuable in thinking about.  And so page 131, Friday, May 12th I’m considering making tEpoR receptor without the poly(A) sequence.  Suzanne, that’s Suzanne Kirby, started with mouse cDNA plasmid and [00:39:00] excised the coding region, etc., and recloned.  So this is talking about the truncated EPO R1 receptor clone.  So Saturday, May 27th, page 141.  Going back to the tEpoR construct but with a little note.  Returned from a west coast trip with Field Morey the fourth time and with George Bandow, the third time with him.  Super.  [00:40:00] Two of my flying friends, Field being my instructor and George my flying friend of many years.  Still do things together.  So must have been maybe five or six times we had trips of one sort or another.  So back to the tEpoR construct Saturday, May 27th, page 141.  How to proceed and some sequences.  [00:41:00] A rather complicated name for a plasmid that appeared I was making.  Monday, May 29th, bactin EpoR sequence.  This is assembled the 8.1-kilobase B101 sequence.  This is beta-actin-driven tEpoR, bactin EpoR sequence.  And a diagram of what it would look like on page 142.  So 5’ beta-actin gene with its promoter and then exon 1 and intron 1 from beta-actin.  And then fusion to tEpoR plasmid coding region [00:42:00] so that we get a methionine start, etc.  pSKB1.  I suppose that’s probably plasmid Suzanne Kirby B1 plasmid.  bactin EpoR changing sites, page 145, 147.  [00:43:00] This is where we were.  [00:43:22]

OS:

[00:00:00] So we’ll just start again, Jenny.  I’m not worried about it.  I’m restarting this recording on page 81, Thursday, December 16th.  Talking about pSKB1 modifications.  Now pSKB1 is Sarah Bronson’s construct that allowed us to insert a transgene into the HPRT locus as a single copy transgene.  So it depended upon taking ES cells, embryonic stem cells, which were HPRT-, because of the deletion, and then using a construct that would make them [00:01:00] HPRT+, which was easily selectable, at the same time as inserting a single copy transgene into the HPRT locus.  So that pSKB1 was a plasmid designed to insert single copy transgenes into the HPRT locus with high efficiency.  And here on Thursday, December 16th, page 81, I’m just modifying that plasmid.  The plasmid was actually grown by Kathleen Caron as J’81 pSKB1.  But it is Sarah Bronson’s single gene targeting construct and the paper that was published on that was published in [00:02:00] PNAS volume 93 page 9067, August 1996.  So modifying that plasmid in various ways.  Following pages show some modifications which are needed to make it easier to clone things in by putting new cloning sites into it as illustrated on page 86, new cloning sites for pSKB1 deletion and old cloning sites and linearization sites and so on.  So just improving that vector in order to use it to insert silencers on Monday, December 20th, page 91.  A nice way to put two silencers into pSKB1 [00:03:00] is to use a Bam linker, etc.  A linker insertion then on page 95, the following page, Wednesday, December 22nd.  Still working on Christmas Day, Saturday, December 25th inserting a linker into pSKB1 and also insulator dimer insert checking on the results on the following page.  So sequencing it on page 107, Tuesday, January 18th, J’107 pSKB1 AvrII [00:04:00] to Seigo for growth.

Insertion of insulators now on Saturday, January 22nd, page 111 and continuing in the usual way, though abandoned because of sequence results on page 113 Sunday, January 23rd.  Restarted the insulator insertion on the next page.  Usual business.  Find something wrong, restart.  So insulator insertion into Avr eleven, page 117, Friday, January 28th.  And silencer dimer insertion going on.  [00:05:00] Continuing sequencing and so on.  Thinking about a slightly different scheme now for hematopoietic stem cell targeting using a poly(A)-lacking surface antigen,  page 123.  Possibility of using the truncated erythropoietin R receptor as a surface antigen to select immunologically for hematopoietic stem cells that have the receptor.  And so various comments on this possible technique.  Back on page 125 with the grants in review off [00:06:00] Tuesday, April 18th.  The TNAP c-kit.  That’s the alkaline phosphatase c-kit.  Rather nice description of what those whole experiments were about, page 124 appended.  Tissue-nonspecific alkaline phosphatase TNAP replacement with c-kit, the receptor for stem cell factor.  And talks about the background that TNAP has been known for many years to be a marker for cells that eventually become germ cells.  My thought here was since heterozygotes for TNAP gene disruption are essentially normal my thought was to replace the TNAP coding region by c-kit [00:07:00] so that the receptor would be made earlier in development than usual and might enable the ES cells with this TNAP c-kit to outcompete wild type cells in populating the gonads.  So this is an attempt to use the alkaline phosphatase locus TNAP with the c-kit receptor to give an advantage and move something more efficiently into the germline.  I don’t think it ever worked, but that’s what I was aiming to do.  And a fairly good description of what the plasmid looked like on that page.

On page 127 Monday, April 24th I have a page of comments on the failure to get [00:08:00] recombinants in C57 black 6 ES cell lines to go germline.  At that time I had a problem with B6 ES cells.  It’s been overcome since although strain 129 is still easier.  But B6 can be used now.  But in those days we had a problem with it.  And Hyung-Suk Kim succeeded in getting gene disruption and gene duplication in thaw one of the C57 black ES cells, but it was unstable.  It had an unstable karyotype and doesn’t go into the germline.  And Tomoko Hatada has had no success with it.  So here was some thoughts about what might be done to correct it.  [00:09:00] And you may be able to find out how to change the things that made the problem.  And Kim, Tomoko, and Laura Reid would investigate this possibility.  I don’t think we were successful, but that’s what it was.

Following page has a rather strange insertion here.  There is a typewritten up something thinking about a patent.  And this invention provides a general method for isolating hematopoietic stem cells or other cells in which a faulty mutant gene has been corrected by homologous recombination gene targeting.  So trying to patent what had originally been — thinking about patenting what had been the original motivation for all of my gene targeting work.  [00:10:00] I wrote up quite a bit of material.  But nothing was ever submitted anywhere so it never went anyplace.  Although there is a good diagram of how the faulty gene might be corrected with an upstream sequence in the target gene and surface antigen being inserted into the corrected gene so that this would correct the gene and at the same time insert a surface antigen that would enable the cells in which the corrected gene was present to be detected immunologically.  [00:11:00] Quite a neat dream but never put into practice.

Back to truncated EpoR poly(A)-less form.  Suzanne Kirby had started the work and this is further parts of that type of work.  I think that’s where we were at.  Some more checks on sequence B101.  Let’s see what B101 was.  B101.  No.  Don’t know what B101 is.  See if I can find out.[00:12:00] Going back a little ways then.  Truncated EPO R1 site.  Truncated EPO R1 gene being considered on page 131 as I already said.  Mapping.  J’131.  B101.  J’131.  [00:13:00] It was just the name given to the poly(A)-less plasmid sequence with a human beta A gene, the tEpoR gene, and SV40 sequences.  It was a complicated plasmid.  Anyway mapping is on page 135.  [00:14:00] And regrowing it on the following page.  More sequencing of it, page 139, May 15th, Monday.  tEpoR 1 construct continued, Saturday, May 27th with a little note that I’d returned from a west coast with Field Morey and George Bandow.  That this was the fourth trip with Field Morey and the third with George Bandow, my two flying friends, or instructor and flying friend.  And so I’m back to sequencing after that enjoyable type of trip.  Super, exclamation mark.  Flying and having a good time.  [00:15:00] B101 is beta-actin EPO R1 sequence bactin EpoR B101 assembled and there is a map of it on page 142.  It has the 5’ beta-actin gene and promoter exon 1, intron 1, and then the tEpoR gene is now inserted after the beta-actin promoter with a methionine start.  So that’s beta-actin-driven EpoR, which is an 8.1-kilobase fragment.  [00:16:00] I’m going to put that into pSKB1.  Changing sequences in the bactin EpoR in the following pages.  Thinking about using pSKB1 AvrII insertion site for putting in an mRNA-destabilizing sequence.  The idea being to have a destabilizer into a — single copy destabilizer.  [00:17:00] I’m not quite sure what the idea was.  Changing sequences and more sequence determination June 6th, Tuesday, page 155.  Message destabilizer.  Put that on Tuesday, June 6th, page 157, 69-base-pair from c-fos message-destabilizing.  So although it’s not very clear at this point, this became quite important that we learned later on to [00:18:00] change the 3’ untranslated region of the gene to make the message less stable and therefore control the gene using 3’ untranslated region.  This work was with Masao Kakoki brought to very powerful procedure somewhat later.  But this is where the idea was first beginning to mature.  In other words, to insert a single copy gene, but vary the expression level by destabilizing the sequence with an unstable [00:19:00] message 3’ part of the message from the c-fos gene.  So here is the statement.  A message destabilizer.  There is an A-rich element sequence of 69 base pairs from c-fos that is destabilizing.  And I was going to insert that into pSKB1 so that I could control the level of the single copy gene.  Various plasmids being made as intermediates towards that.  [00:20:00] And so on Sunday, June 11th I have on page 161 an acceptor for ARE being thought about and made.  Putting ARE into pSKB on Wednesday, June 14th with a comment J’161 ligation “not much hope, start again tomorrow.”  And another note.  “Canceled.”  But the following pages 162 and 163 is a repeat.  And a rather more helpful outcome, 28, number 3, 8, and 9 are correct, and 31, number 13 and 6 are correct.  So that was a repeat of putting in an Mlu site into [00:21:00] the EcoRI site of presumably pSKB1.  Let’s see if I can see what it was being inserted into.  Anyway there is the end of that book J’, Monday, June [00:22:00] 12th.  [00:22:02]

OS:

[00:00:00] This last page is just inserting an Mlu site into the beta-actin EpoR construct bactin EpoR.  So it’s just a simple insertion of an Mlu site.  But it was successful as the page says.  And that’s the end of book J’.  [00:00:26]