Oliver Smithies:

[00:00:00] So here we are on book c, beginning October 27th, 1972, and going on through February, so about four months, February of ’73, beginning with attempts at improving the new Fc gamma-2 method, make the constant region.  Continuation of B-143, 0 degrees papain digest [F?] gamma-2, with a comment that it’s an odd result, query due to the dialysis or due to the gel.  [00:01:00] And not clear what the difference is in looking.  The column has obviously worked.

Check on the crystallization products on October 30th, Monday, page 7.  The supernatant and the wash crystal, and supernatant and the wash crystals, and a big, fat band, and a couple of other bands as well.  Clearly one fat band, clear that the small batch, the first crystallization cleans up the product better than the slow.  One was the fast crystallization, [00:02:00] and one was slow crystallization.  And from the two experiments, they don’t look very different at this point. (laughter)  There’s a comment. “Cleans up the product better than the slow!  Or at least as well,” so that’s what it looks like now, doesn’t look any different.

So, A-1 vs. A-2 test, Fc0 to Fc1-5, that’s what I’ve been struggling with in looking at these results, “I cannot understand the heterogeneity of Fc1-4 at the same time as Fc0, thus B-152 gel with B-139 [00:03:00] starch shows identical pattern, as shown by the experiment, down to very B-133, bulk A-1,” so that’s B-139 versus B-133, down to very fine details, and neither shows what I call Fc0.  I can’t understand it.

So trying to solve that problem, with different samples being compared on a gel in page 10 and 11.  But, some extra A-1, A-2, gamma-2 room temperature was put to ice after setting up the gel, and A-2 crystallized, but A-1 did not.  So, they were easy to crystallize, [00:04:00] some of them, more of A-2 crystallized, whereas A-1 did not.

Continue with crystallization on page 13, November 1st.  Different crystallizations.  A gel, showing the results of the crystallization.  Recover the crystals, see 13, 94mg.  [00:05:00]

So here, this is the crystallization, a comment again, so the next day, Thursday, a small crop of nice, small crystals.  So it will come out of solution with 0.1M salt, dialyzed it, and then Friday, flat plate crystals, the [00:06:00] earlier ones, didn’t say what the shape was, (inaudible) amorphous precipitate.

Continuing with this type of work.  Fca, or A2, A2 for x-ray.  But no comment on what was being thought of.  I don’t think I would have been thinking about doing x-ray diffraction experiment, but why it says [00:07:00] “for x-ray,” it’s not at all clear, page 19, November 2nd, Thursday.

Recovered 7mg of washed, dry crystal, on page 20, November 9th.  Carboxymethylcellulose column, on the following day or so, room temperature digests, 1/50 papain.

Trypsin on the Fc, November 7th.  [00:08:00] Different times of digest with no obvious differences of the digestion, on page 27.

Papain comments, November 8th, thinking about where the cleavage might occur, etc.  Possible, cut papain, cleavage point towards the end of the constant region, where there was a sequence, histidine-histidine-lysine-lysine, a possible cleavage point with trypsin, very [similar?] thoughts of [00:09:00] this type being considered on November 8th.

520g of serum were dialyzed against 4L of ammonium phosphate pH 8.5 on November 8th, [00:10:00] so large amounts of material on papain digests of this material on the following page, no significant differences.  Looking for what could one find of cysteine positions, or whether they’re in disulfide bonds, on Fc papain 0 degrees Centigrade crystals, A1, A2-A2, [00:11:00] 7mg that had never seen iodoacetamide, so three SH groups could still be there, or 7mg of material, both are crystals that had been exposed to iodoacetamide to see if one could detect the difference with interpretations.

And some interpretations, Fc4 is not correctly identified, November 13th.

And here is a quite pretty gel on page 41 and 40, test versus anti-A2, and anti-A1, gamma-2 allotyping serum, [00:12:00] and various F1, F2, Fc1, Fc2, etc., Fc5, all the different bands of the constant region.  With a mysterious comment, that, “No difference except IA and IB, they’re now reduced, or papain was reactivated.”

November 14th [00:13:00] the usual sort of error, no error, no DTE, dithioerythritol.

Continuing to try to solve this protein purification and degradation, etc.  Back to beta-2 microglobulin, [00:14:00] dog, and alpha-2 retinol binding protein, on Thursday, November 16th.  Dog serum, Harvey, and then the beta-2 from — just dog beta-2, from Dave Poulik — very pure material, nice-looking material.  So alpha-2 retinol binding protein, and pure human beta-2, and dog beta-2 all are different, but they’re nice, pure proteins.

And here is the sequence now from the sequenator on page [00:15:00] 55, where the sequence can be seen, retinol binding protein, and the beta-2 sequence there.  There’s no similarity, of course, between the alpha retinol binding protein, and the beta-2.

This is getting close to another little story which may turn up in the next few pages.  If not, I’ll tell the story anyway.  But at this point, this is just writing down sequences.  [00:16:00] So we’ll remember, page 54, book c, we might come back to that later.

Working on Thanksgiving Day on Thursday, November 23rd.  Still purifying Fc fragments.  An attempt to do radioactive iodoacetamide on a pool of C-47, pool 3, C-47 protein, using C-14 iodoacetamide to try to find [00:17:00] which bands have a free SH group.  But no comment on the results, although the results are given, of the column and the counts.  It was trying to work out the sequence in front of [VC?] PCV, RZ, PSV, FIF, etc., on the sequenator.  And, some comments are there [00:18:00] what is not correct.  See what happens.

Another strategy, Tuesday, December 12th.  Reduce under mild conditions, add radioactive iodoacetamide, and then excess cold iodoacetamide, etc., and also excess protein, and no excess mercaptoethanol, and then try to find out what’s happening.  [00:19:00]

Peptide isolations, December 14th.  Modified digest, and so on.  Peptide isolation.  Different fractions from the columns being analyzed.  [00:20:00] High voltage electrophoresis being carried out, 2,000 volts applied, recovered on, this is on filter papers, presumably.

Cleaning up of fraction 4, C-81 on page 93, Book c, cleaning up [00:21:00] for that fraction, sequence 615 looks promising.  But the results are over the page, with sequences being written down there.  Presumably, I’m dictating to someone else, because there’s my writing in part, but all of the sequences are written by hand by someone else.  [00:22:00]

(inaudible), etc., on different fractions, trying to purify material, January 19, heavy lambda experiment.  This is lambda bacteriophage, I remember at that time having a hypothesis [00:23:00] of branch DNA, and trying to find in the lab that I visited, that there’s some experiments that I think refer to January 23rd of heavy lambda, that’s bacteriophage lambda.  I’m almost certain that this is in relation to experiments I was thinking of doing with [Bill Doug?] to try to see replication forks, and see if I could get a branch DNA; I had published a hypothesis that branch DNA might account for antibody variability, and we can look this up in a little while and find out where that was published, about this time.  So this was [00:24:00] lambda C-60, 6*10^10; I’m sure that means 6*10^10 plaques per total volume available, and try to get replication (inaudible) and then abort it, and select the recombinants, and try to propagate them again.  So, these are sort of, replication forks have been seen by [Shblasky?], and so there’s a comment that he uses 2.5 hours, I think that it is, at 18,000 revolutions in 19 rotor, and Maria remembers 1.5 hours was OK, so the experiment was to test temperature-sensitive DNA mutant to see [00:25:00] about 10% replication figures can be obtained, the idea being to try to get DNA to make multiple forks.

And so indeed, on the next page, it says, I’m obviously talking to [Bill Doug?] because he recommends using CR-34, a temperature-sensitive mutant.  [Ross Inman?] uses the same strain, and has all the protocol, and if one uses tritiated thymidine, it says that, the notes say that 1 microcurie per microgram, then the phage is stable for months, and tentatively, eight phages can be counted OK.

So I [00:26:00] tried continuing that type of experiment on page 105.  As usual, if I have a hypothesis, I always try to do experiments related to them.  This is an attempt to see whether my branch DNA idea had any merit.  In relation, the experiment was related to testing a hypothesis that I had published in August of 1970, so here we are in 1972, trying to test a hypothesis published in 1970, pathways through networks of branch DNA, the idea being that you could get all the [00:27:00] variable regions, being multiple, and constant regions having gamma-1, gamma-2, mu, and alpha being the different constant regions, with variable numbers of pathways, with the variable number, and so could get variability on one hand, and constantly, the Fcs on the other.  Quite a nice idea, and I have a little glass model of this, that was made by my friend.  Anyway, a glass model of this I still have which is a reminder of one hypothesis.  [00:28:00] There was another model, also a glass model, made of a later hypothesis, both of which have, that one was made professionally, and the other was made by (inaudible).  I might say a friend of my wife, maybe.

So anyway, trying to make these bacteriophages.  [00:29:00] Different variations of the medium being tested, and of the thymidine for the cultures.  Eventually, the idea got to be making lamba-DV, the defective variant.  More sequences on the post-gamma, this was from Dave Poulik, post-gamma, trying to get sequence, and not, altogether successful from the look of it.  [00:30:00]

Here’s an experiment talking about using heavy water, a repeat of the heavy water experiment.  But it doesn’t say anything about the details, so it’s a little impenetrable.

Different mutants being looked at, and titrated on page 113, using different bacteria, C-600 which was a [00:31:00] suppressor 2+, threonine/leucine -/-, most lambda will grow on it.  And then, C-594, equally useful, but suppressor minus, so mutants will not grow well on this.  But I’m accumulating some experience with bacteriophage lamba mutants, which later on, when we get to a time of gene targeting, this experience with mutants was helpful in letting me work quite happily with bacteriophages that did or did not have suppressor mutants in them, so experience with mutant bacteriophages [00:32:00] is quite valuable.  The indicator is, and here’s another thing that has a repercussion later, the indicator was tryptophan broth bacteria grown to saturation, and resuspended in buffer, TM.  I later on found a rather critical note that we’ll come to quite a bit later in book gamma or delta, but it was important not to use saturation, bacteria grown to saturation, but to use bacteria that were in the exponential phase, so here, little things of this sort that are building up experience, [00:33:00] that later on becomes quite important.

Magnesium Tris buffer, and lambda suspension medium formulas for these, important reagents in working with bacteriophages.  FM suspension medium for preserving lambda is another recipe.  This is needed if one’s going to keep the phages for any length of time.  [00:34:00]

Getting help from a person Carol, mentioned several times, but I am afraid I don’t remember who Carol was.  She was probably Bill [Dove’s?] technician.  I might be able to find that out.

And here’s an experiment with deuterium, the heavy phage that I was mentioning.  Test with CR-34, which is lambda C1, [an?] 845 mutant.  Maria started a 30-degree culture in deuterated sugar, so that it would be getting heavy as a result of deuterium to try to separate by deuterium mass changes.  [00:35:00] It was obviously the idea, not going to be easy, I’m sure.

So here is an attempt to do that titration of lambda C187 lysate, from water broth.  Experiment C129, and also from deuterium oxide broth on the two pages on 130 and 131, because the titer from the deuterium experiments is 3*10^7, whereas the water experiment, I got a titer of 9*10^9, so about 300 times more when grown in water [00:36:00] than when grown in deuterium oxide.

More protein sequencing on C95 from fraction 2, lithium iodosalicylate-treated material.

Aha, at last we’re getting to lambda dv now, so Tuesday, January 30th, lambda dv, which is much smaller and easier to handle than lambda itself, to try the latest Hogness method, and also [unreadable?] Pettyjohn made.  Pettyjohn, and Hogness, had made [00:37:00] these phages, all these defective variants.  They weren’t phages, they were plasmids, we would now think of them as simple plasmids.  And, used in electron microscopy.  So lambda dv assay on January 31st, 1973.  I obtained, it says, received on Tuesday, KM424, [rec minus?] lambda dv, etc., etc., from [Sid Hayes?] as a slant, and inoculated it, and titrated.  [00:38:00] Estimated at 5-10*10^9, grew up overnight.

Learning also to make replicates of bacterial cultures using sterile velvet, where would one would take an imprint of the bacterial plate and capture enough bacteria that we can replicate the plate on several of the clean plates.  [00:39:00] Repeating the titration of lambda 029P3, etc., indicator stock, C600.  Replicated on agar.

And then getting to lambda dv DNA tests on February 1st, attempt without DNA precipitation.  I think that’s Ross Ingram, I’d forgotten which Ross, tells me that DNA must have, for lambda, an OD of greater than or equal to 0.05, between 0.05 and 0.01 optical density for convenient spreading [00:40:00] without dilution.  Kaiser gives an optical density of 10, equivalent of 4*10^12 lambda dg/mL, or 0.1 is equivalent to 4*10^10 lambda/mL, assuming 50 [monomer?] square bacteria, etc., etc., this goes on.  But the experiment ends up, I think, it didn’t lyse, it didn’t lyse.  So use log cultures; use logarithmically growing cultures.  OK, and an idea that later on became [00:41:00] important.

An assay, still, on the following page, 145, still negative.  Try colonies, etc., etc.  Sporadic resistance colonies, because they’re looking for resistance that the lambda dv, I think, was carrying a suppressor gene, and that I missed it.

So here on page 147, I’m trying to make lambda dv again with logarithmic cultures, OD about 0.9, equivalent to 3^10^8/mL on tea broth, quickly chilled, etc., etc., onto sucrose, lysozyme, magnesium, and [00:42:00] taking them up in sucrose, and adding lysozyme and magnesium, and neutral detergent bridge.  The viscosity quickly increased, so some E. coli is present, DNA, etc., etc.  And then added predigested pronase, and the viscosity immediately went to water-like, etc.

Spreading tests of the lambda dv on page 149.  Nothing visible.  Ross says, due to bridge 58 getting into the film.  [00:43:00] And there we end this Book c, with an experiment that has not yet worked.