Oliver Smithies:[00:00:00] So this is book lambda which begins on July 17th, 1985 and goes on almost a year to April — well, three quarters of a year. April 1986. Starts with a fresh start on Kpn delta beta 17. Old preparation of Cla17 and no supercoil left and about 50% yield. And then proceeding then to make a blunt end with T4 polymerase. [00:01:00] Continuing Kpn digest Thursday, July 18th. Disappointing page. That was page five. Disappointing. No sign of Kpn action. Nonetheless continue to use it. The Kpn delta beta transformation on page seven. But good transformation with Gina and Nobuyo’s preparations.
So here we come on page seven, Thursday, July 18th, [00:02:00] Kpn delta beta transformation using cells, recA bacterial cells, competent cells, HB101 is the name of the cells. One preparation from Gina and one from Nobuyo. But checked Nobuyo’s cells and they were not recA. So there was a difference between Gina’s material and Nobuyo’s. [00:03:00] On page eight there’s an insert describing a phage clone lambda BD12, which is partial Sau3A-digested genomic DNA from a patient with hereditary persistence of fetal hemoglobin cloned into the BamHI site of lambda phage EMBL3A. This is material from Kathleen Delgrosso and commented on the material. [00:04:00] It’s 10.8-kilobase Xba fragment which includes a 2.3-kilobase RI fragment of the delta gene.
So it was plated on DP50 supF Dennis Perreiro bacteria. Minis were obtained. And analyzed on the following page, Saturday, July 20th, page 11. Checks on two of them. [00:05:00] Because of the ones that were checked, number 14 and number 19 make sense. Kpn cuts make sense. And this was continuing then on page 13 and 12, July 21st, Sunday.
Delta beta K117 it was called. That is kappa 117. No, it’s not. It’s K for Kpn 117. Delta beta K117. Number 14 was grown up from a single colony. Probably in the recA+ strain. And various checks on it on colonies 14 — on candidates 14 and 19 [00:06:00] confirming the correct site on both of them. No. That’s incorrect. On number 14. Correct on number 14.
Making a supF fragment again. Going back a long way. HpaII supF fragment 380 base pairs, pi VX, etc. Comment on. Cut is fine. [00:07:00] Trying to get the fragment on page 19, Friday, July 26th. Nearly missed. It is in S. Use C plaque in the future. An insert from Harvard. Letter from Harvard. Talking about two plasmid DNA samples. And this is from Rick Meyers with sending me two plasmid DNAs. [00:08:00] One he used for making a labeled RNA probe to detect the sickle cell change in DNA. And the second one is one that he made from a lambda clone given to him by Stuart Orkin. So these are sickle cell DNAs basically. And the next page is from Joe Peters. Letter MRC. Howard Ditcut, Oxfordshire in having a nice little letter. The DNA that he’s sending is from the liver of a mutant mouse Hbbd4 homozygote. [00:09:00] So hemoglobin, a mouse hemoglobin gene.
And evidently feeling happy on Saturday, November 9th. So that’s a jump from July 26th. No entry until November 9th. Saturday, November 9th. With a comment the grants written. Course on its way. So I’d been writing grants and teaching. [00:10:00] And this is looking at the Xba fragment for unmodified DNA. Human DNA 11-kilobase fragment that had been used in the targeting as the target locus with a TaqI fragment of delta being about 3 kilobases in it. And so digesting with TaqI to isolate that 3-kilobase fragment.
With a map on the opposite page, 24, of delta beta supF insertion into the cos site. Next to the cos site of [00:11:00] this – so wanting to prepare two fragments. One S fragment from delta beta. TaqI TaqI fragment which I call capital delta little delta. And fragment for the supF being HpaII HpaII fragment. Small fragment already talked about. [00:12:00] And reasonably clear map of what has been done.
So cutting fragments out from gels on Sunday, November 10th, page 27. And proceeding to insert them into this construction. The aim being to find a way of detecting gene conversion versus crossing over, with a scheme that will perhaps become clear although already confident; [00:13:00] so Monday the 29th is inserting PvuII TaqI piece into a Kpn Hpa or assembling these four fragments into a bacteriophage. Scheme on how to, which is pursued experimentally on page 31. Repeating ligations page 33. And now beginning to think about using the gene targeting for other purposes. This is a beginning of a whole new thought on Monday, December 30th, page 35. [00:14:00] Talks about the EK cell system, that’s Evans-Kaufman cell system. Which eventually became the embryonic stem cell for which Martin Evans received the portion of the Nobel Prize along with Mario Capecchi and myself. And so here on page 34 is a reprint of his very important paper on the formation of germ-line chimeras from embryo-derived teratocarcinoma cells. The beginning of this thought. Not yet the cells that he became famous for. So with Allan Bradley, Martin Evans, [00:15:00] Matthew Kaufman, and Liz Robertson. Showed that they could get germ-line chimeras from embryo-derived teratocarcinoma cells. That’s interesting because that was basically the same type of thing that Beatrice Mintz had published on but it’s referred to in Allan Bradley’s paper. It’s reference 18 to Mintz and Illmensee.
So it’s really talking about the embryo-derived pluripotential cell which exhibit both a normal karyotype and a high differentiation ability. That’s referring to Evans and Kaufman. [00:16:00] Critical paper in Nature volume 292, page 19 — year 1981, page 154. That’s the Evans and Kaufman paper of ES cells. This paper is as it were in a sense a backtrack to showing that you could get germ-line chimeras even if you started with teratocarcinoma cell lines. And I don’t know why I have picked this out for inclusion. But I’ll just look and see what context the comment on where Beatrice Mintz’s work fits into it. I’ve always been interested in that side of things. [00:17:00] Because it’s just referred to as the published rates of construction of germ-line chimeras using embryonic carcinoma cell line have been disappointingly poor. And that’s where the reference to Beatrice Mintz’s work is. But anyway going back to the important part. Page 35 is the EK system, the Evans-Kaufman cell system. And that you need a feeder layer of STO fibroblasts. These are HPRT-. And they are thioguanine- and ouabain-resistant. Need to be subcultured and frozen. And EK cells [00:18:00] from Martin Evans CC1.2 originally received the end of November. So that’s a month earlier than this entry. But it’s rather enjoyable reference even though it isn’t documented here that Martin Evans brought cells to me personally. He was over in the United States and he visited my lab and he brought the cells in his shirt pocket in a little tube and came into the lab and pulled out this tube from his left-hand shirt pocket and said, “Here they are, Oliver.” Gave me his embryonic carcinoma — his embryonic stem cells as they later became called. And so this is talking about them and saying they were originally received the end of November. And one flask was contaminated and one was [00:19:00] trypsinized and frozen, eight ampules. So we got these cells from Martin Evans and began to think about using them because the gene targeting frequency was so low as to be useless for gene therapy but maybe could be used to make mouse mutants and so the plan commented on the bottom of the page, 35, is to use these to get HPRT- by recombination. In other words to start with HPRT+ and use recombination to get HPRT- and get chimeras or germ line by the blastocyst route. So here was the thought that we would go from HPRT+ cells to HPRT-. And as [00:20:00] it will turn out later on this is in fact not the best way to do this type of experiment. It’s much better in trying to work out the system to go from HPRT- back to HPRT+. The background is much lower.
So here we are on page 37, Saturday, January 4th, 1986. Now making a delta beta supF fragment. The 3-kilobase fragment. And cloning it into a plasmid called pAT153, cloning that fragment. And Nobuyo had a preparation of this material, which we [00:21:00] attempted to give with a comment. Should have given longer time for Kpn. But can live with this digestion of Kpn ready to do the subcloning. And then Kpn linkers being put onto delta beta supF. The Kpn linker coming from — going back to theta 79 where that was available. That one, Jenny, no, next one, yeah, so we must have made it on theta 79. [00:22:00] Yeah, this is to make Kpn linker addition as described on that page, book theta. Now making a different one. Kpn linker. Because we still couldn’t synthesize these linkers at will or buy them. So digestion with Kpn after insertion. Hopefully the Kpn linker. But no visible changes. So get rid of the linkers. That is carried out on Wednesday, January 8th, page 45. [00:23:00] Prep gel to remove the linkers and supF and get the linker supF product. Feeling rather happy about it evidently. Because on page 44 it says conclusion proceed with the cloning.
Ligation is on the next page but could have been a minority fraction that was not kept. However, proceed with the transformation. So transformation is on page 49, Friday, January 10th. These HB101 recA cells made competent by Susan McEvoy. And the transformation was carried out with lambda 47 pBR and with a control on [00:24:00] those, meaning the DNA was from lambda 47 ligation or from control of pBR32 control DNA. And picking colonies Sunday, January 12th, hybridizing them to lambda 53 supF-labeled probe. With a comment on the opposite page. Not an uncommon one. [00:25:00] Start again. Kpn linker stage. And Phil will do this. So supF fragment. Nick translation page 53 and cloning and rescuing, etc. On page 55. So continuing this. [00:26:00] On page 59 rechecking phage GW149.1 which comes to the conclusion that it cannot be the beta piece but a comment with a little arrow saying OK. Confirm by IVS2 blot by Vicky as negative. It cannot be the beta piece. So usual progressive experiments.
Trying to rescue the Xba Xba fragment on [00:27:00] Thursday, January 30th, page 63. With an experiment on Thursday, January 30th, page 65. Taking delta delta X Xba and lambda 63 Xba pool, etc., etc. And packaging to get an appropriate phage. Usual sonic extract and freeze-thaw lysate, etc., etc. [00:28:00] On Friday, January 31st, page 67. Looking at the ligations of various things done on page lambda 65. A whole bunch of different ligations with a conclusion. I must have labeled the samples in the reverse order. But the conclusion of the experiment is that the ligation was poor. Packaging was only about 4 times 10 to the 6 per microgram.[00:29:00] When plated on the right bacteria. If plated on C-1a only nine phages were positive versus where it says about six phages per microgram. Versus 4 times 10 to the 6 when tested on suppressor strain. So repeated ligations, page 69. So continuing. And more tests on Monday, February 3rd, page 73. And [00:30:00] trying to inactivate Xba with increasing heating but used too low a temperature. Making more delta delta X Xba material from Charon 3A delta delta X.
More ligation tests Tuesday, February 4th and remake of classic buffer which Nobuyo and I continued to use for [00:31:00] very many years, a p95 buffer, that’s little book p, way back. And it gives the formula for that buffer, which is a ligation buffer. And both of us over the years kept tubes of it and used it for very many years. Tris, magnesium chloride, DTT, and ATP. So the lambda 77 ligation E tested on Wednesday, February 19th, page 81. About 5 micrograms of ligation. Approximately 25 micro tube, etc., etc. Previous titer was 1 times 10 to the 8 per microgram. [00:32:00] So redesign of the TNFUS series. Fusion. Thymidine kinase neomycin fusion series. To make TH hygromycin fusion, page 83, Wednesday, February 19th. How to get a different promoter driving — the same promoter driving a different fusion gene.[00:33:00] Rescreening of RGB31.5B 98 colonies continued from book kappa 81. Going back to work with the targeted colonies of mammalian DNA. New genomic DNA available from RGB31.5B lambda 87. Concentration is high, 475 micrograms per ml and suitably labeled. [00:34:00] Testing the DNA that was obtained. Bulk packaging recipe on Monday, February 24th, page 93. [00:35:00] So here we are packaging again on Monday, February 24th, page 95. The experiment is 31.5B. Packaging that material and control DNA which is Charon 3A delta lac. And the ligation of 31.5B is the lambda 88 ligation, and plated on C-1a. [00:36:00] C-1a plating. Had the idea of using maltose but it didn’t seem to be very helpful. Says not good in the comments and abandon in the plates. This is C-1a plus maltose. Plaques are smaller with maltose. And it’s talking about the yield of phages. So DP50 supF bacteria with 31.5B package material, 1 times 10 to the 9 phages. And the control gave — was equivalent to 2.1 times 10 to the 7 per microgram. [00:37:00] Plaated on DP50 supF but plated on C-1a gave 1.3 times 10 to the 4 total meaning that 1.3 times 10 to the -5 of the total have recovered the suppressor gene, 300 times better than the kappa 35 ligation. And 10 times higher than usual. So very satisfactory result.
So talking about it on page 97, Tuesday, February 25th, looks as if single packaging mix works well with large-scale. This is a [00:38:00] single packaging which I failed to see that it was — packaging was chi 92 SMR, whatever that might be. Have to find out what that is. This continuing on page 95, Monday, February 24th, this was using a single packaging mix called SMR chi or lambda 92. [00:39:00] Because lambda 92 SMR is single packaging mix. Let’s see what it is. [00:40:00] This SMR is a packaging mix which Phil Howells had. A single packaging mixture. Didn’t need the freeze-thaw lysate and the sonic extract. So continuing on page 95 February 24th, the DNA, the whole of lambda 88 ligation, packaged with this mixture. Lambda 92 SMR. With and without maltose. And although as I said before maltose turned out to be no good the single packaging mix worked quite well. [00:41:00] And so going on to page 97. Remarks. Results of large-scale plating of the lambda 88 31.5 beta and I think I’ve said this already but it looks as if the single packaging mix works well with large-scale. And a summary of what was obtained on page 96. Hybridizing to 5.2-kb fragment as usual and to beta IVS2 onto film. Developed on Friday, 5.2-k was no good, doubtful positive on IVS2. [00:42:00] So it didn’t actually work anyway.
Tuesday, February 25th, 99 page. Going on with the beta S hygromycin resistance design plasmid. What I’m trying to make. Two-strain packaging. Going back to the two-strain packaging with usual sonic extract and freeze-thaw lysate on page 101 of the lambda 89 ligation of 31.5B and control. [00:43:00] Usual. In this case some maltose was added. Not a very high yield but continuing. Hybridized at 5.2 and the beta IVS2. Saying from the phage titer that it was — leave on film until Sunday if negative. In parentheses expected. Only hybridize all the filters [00:44:00] to the lambda 105 probe. Fresh probes being made page 105, 5.2-kb probe, the usual one, 1.1 times 10 to the 8 counts. And beta IVS2 probe 1.2 times 10 to the 8. So one was 60,000 counts per minute per microliter. And the other 8,000, pretty good.
So rehybridizing the lambda 88 filters to these two probes. [00:45:00] Wash by Phil. Films on page with no immediate comment on the result except review on the following page, Thursday, March 6th of the 31.5B experiments, with the conclusion that the best estimate of the first lip was that 74 were positive for 5.2-k probe but no positives for beta IVS2. And the [00:46:00] second packaging lambda 89 at least 76 hybridize OK to 5.2-k. And beta IVS2 hybridization looked clean. Four or five spots corresponding to plaques. One plaque was stronger. But uncertain alignment. And the overall conclusion is that 31.5B can now be abandoned except as needed for development of method. More than 200 5.2-k-positive plaques have now been inspected. And the unwritten statement is there and found negative for hybridizing to the IVS2 probe indicative of targeting.[00:47:00] Comparing two strains and one strain hybridization on page 113. Sonic extract and freeze-thaw lysate for the two-strain. And the other was preparation nine from Charlie Blystad with a comment that the two-strain gave 10 times fewer phage on titration [00:48:00] than the one-strain ligation.
What to expect then if that was correct as commented, as documented, on Monday, March 10th, page 115. Expect about 200 doing the two-strain bacteriophages and 30,000 on the one-strain if those numbers were right.[00:49:00] So titration data looked to reconfirm the one-strain was better. Tuesday, March 11th titration data. And on the opposite page one-strain has about three times better yield per amount of material than the two-strain. One-strain linear above 6 microliters with a multiplicity of infection greater than 0.015 and the two-strain linear above 24 microliters with a multiplicity very similar 0.02. [00:50:00] So different technical things in getting packaging to work efficiently with the two methods. So returning on Tuesday, April 8th, page 118 to freeze-thaw lysate. No. Sorry. On F lambda 45 Kpn delta beta supF various ligations. Cloning with Kpn linkers. Trying to get this delta beta piece made.
Checking on it on page 121 with a comment that it’s probably there. Probably in there. So on. Ligating again page 123. Repeated linkers on page 125, no signs. Test it again tomorrow in bulk single [00:51:00] page 129 lambda 123, lambda 125 are no good. So usual negatives are mixed with the positive.
Blunt ending with T4 polymerase to make a Sal linker on page 131. And continuing in the following pages with similar experiments. Now Klenow polymerase on TaqI ends page 137 but no sign of ligation. Try PEG. Getting a bit desperate. PEG attempt at rescue on page 139.[00:52:00] Restarting on page 141. Klenow and linker page 143. Just slugging away rather unhappily I think. So having obtained apparently on page 144 a Sal PvuII delta beta TaqI Sal fragment delta beta on SalI fragment always the problem of not at that time being able to buy linkers or make them synthetically. Made life considerably more tricky.
Same sort of thing continuing. And here we are calf intestinal alkaline phosphatase page 151, Saturday, April 26th on delta beta kappa 117. [00:53:00] Whether that’s k117 at this point don’t know. Not important. Prep gels for a new linker with the delta beta S. New fragment, delta beta S. On BstXI SalI SalI ends. Check on the products prior to ligation page 155. Ligations on 157. And says transformations again on Monday, April 25th. That’s April 28th rather, page 159.[00:54:00] Repeated Tuesday, April 28th with Ron Gregg’s fresh cells, not happy with the results. So finally a sort of conclusion of the plasmid minis summarized on page 161. Wednesday, April 30th, minis delta — capital delta little delta and delta beta S. How to proceed. Trying to make these two. [00:55:00] And so the book is ending on page 163 with minis on delta beta S with a big cross out and scrapped. And the opposite page 162, April 30th, the day before. Retransformation with the rest of the delta beta S ligations. But evidently not too satisfactory end to this book. [00:55:28]