I am here tonight to talk about Linus Pauling. He was a very great scientist. He was in his time the world's leading chemist. I knew him personally, but I cannot say that I knew him well. We met on a number of occasions. Of course, he was somewhat older than I, and moreover when we first interacted I was still a graduate student at a somewhat advanced age because of the war, and he was the dominant figure in chemistry.
I am not going to talk on this occasion, or really on any occasion, about some of the important parts of his work that is to say, how he applied quantum mechanics to chemistry. This was an absolutely fundamental advance, and he was the leader of it. Nor am I going to say anything about his humanitarian interests.
What I am going to talk about is his impact on molecular biology, but I am going to talk about it fairly informally, because you are not all molecular biologists. I thought I would start off with a few reminiscences before going on to the science. The first question I ask myself is not when did I first meet Linus, but when did I first hear about him? In 1946 I was sitting in an office in London, in the British Admiralty, trying to decide what to do and wondering whether I should go into what we now call Molecular Biology. I had not had much organic chemistry in school; I remembered there were hydrocarbons and various things like that, but I did not know what an amino acid was. My knowledge was at a very elementary level. Across my desk came an article, or rather I think the account of an article, in Chemical and Engineering News, which said that hydrogen bonds were very important in biology. Well, I did not know two things: I did not know what a hydrogen bond was, and I had never heard of the author who had a rather peculiar name; it was Linus Pauling. That was the first occasion I came across his name. Of course, I asked around, and I heard that a lot of other people, not just ignorant physicists like myself, knew a lot about him.
Not long after that, I bought a textbook, Linus Pauling's General Chemistry, and this was an eye-opener to me, because while I had done some chemistry at high school, I had not done any at university. High school chemistry in those days was taught really as a subject where you had to memorize a lot of odd things that happened. I exaggerate a little, but do remember it was high school; no doubt, if I had gone on further it would have fallen into place more. But when I read Linus' book, I was very impressed by the systematic way everything was organized. I do not think I ever did learn much of the inorganic chemistry, by the way, but I learned all about the strong bonds, and the weak bonds, including the hydrogen bond, of course. And that really is almost all, I would say, the organic chemistry I know today. By now I have forgotten most of it as well, but what I knew then, I think I got from that book.
I next noticed Linus' influence when I went to join Max Perutz's lab. Before that I had done a year or so in a tissue culture lab just to get some acquaintance with biology. There, it was clear they were all very conscious of what Linus Pauling was doing, because not many laboratories were using X-ray diffraction in those days, certainly not to try and do protein structures, as Perutz and John Centra were doing. But even on organic molecules they were not doing so much, and therefore Pauling was a major presence in the background of anything that happened in the lab. There was a friendly rivalry between the group in Cambridge, in the Cavendish lab, and Pauling's lab at Caltech. I say it was friendly because, although, as you know, science is supposed to be cooperative, there is nevertheless a certain amount of competition, because everybody wants to publish a little sooner than somebody else. It was friendly because, for example, in the office that Jim Watson and I shared, there were several other people, five or six, and one of them was Jerry Donohue who had worked with Linus Pauling and had come to work with us; and a little later, Peter Pauling, who had come to get his doctorate with us.
Now when did I first actually meet Linus? That was some time in Cambridge. I am not exactly sure when, but I think he visited Cambridge before the year of the double helix. I explained to him how I was thinking ŕ-helices packed together in coiled coils, and it turned out he had been thinking along those lines. And at that stage, I think I did have, also, an interview with him. Like all people of his distinction, his time was rather taken up, so the interview actually took place in an automobile while he was going from one place to another. As a result of that interview it was agreed that after I had gone to Brooklyn Polytechnic for a year, 1953-54, I would go on to Caltech. In the upshot, that did not happen.
I think perhaps the most memorable occasion, though, was in 1953 when we had just gotten the double helical structure, and Linus came to England. He was on the way to a Solvay Conference. In those days there were not so many conferences. Of the ones that there were, the Solvay, which was in Brussels, was an extremely distinguished one, and it had been established for some time; it was mainly for physicists. Linus was on his way to it when he came to Cambridge. I remember that the laboratory was closed because it was Good Friday, and we showed him the model and discussed it with him. He went on then to the Solvay Conference and I think said there that our model might be on the right lines.
On at least one of these visits, and it may have been that one, he came to dinner in our house. I clearly remember because our dining room was rather small, and he was telling me about bits of things he was doing, which I remember I did not fully understand. I did know him to that degree, but I really should say that I knew Peter Pauling much better. I had not realized until I recently read an article by Peter that he apparently used to come to lunch with us quite often, because the college he was at did not serve lunch on Sundays. So on Sundays Peter was apt to turn up looking hungry. Somewhere about that period, I think it was perhaps a little later, Linda Pauling lived in our house for a few months and helped with the children. I felt that I knew Peter and Linda much better than I knew their father, whom I did see on subsequent occasions. We exchanged words, but I never got to know Linus. I think the difference of years, on the one hand, and the difference of distinction during that period, on the other, did make a certain barrier. I very much regret not having had the opportunity to go and work at Caltech, because once you are thrown together, then some of these barriers come down.
Having said all that, let me now say something about science. Looking back, what was really striking about that period was not, as we take it now rather obviously, that chemistry and physical chemistry, and the physics associated with those subjects, are essential to understanding biology at the molecular level. Not everybody agreed with that, in particular Max Delbrck, who was a physicist who went into biology. He also went to Caltech, and he also won a Nobel prize, and was the leader of the so-called bacteriophage or phage group. He believed rather the opposite. He was hoping to discover new laws of physics which emerged when he looked at these extremely mysterious biological processes of how you get replication, which in those days seemed utterly baffling. In fact, when the DNA structure came out, Delbrck had a double reaction: he was very enthusiastic about it I have seen one or two letters that he wrote, apart from what he told us, saying that he thought it was very revealing but he also said that he was very disappointed in a way, because it was just like a tinker-toy. Instead of having a profound new bit of physics, all he had found was just a few things fitting together. But that was really, I think, Linus' main contribution. The chemists, after all, traditionally were much more concerned with the strong bonds, the homopolar bonds and the coulombic bonds, and things of that sort. Only later on did they become interested in the van der Waals forces and hydrogen bonds. And those are the ones that fit molecules together in a physical sense, in one way or another, with different degrees, of course, of specificity. Pauling believed that that was going to be the key to everything, the way that things fitted together, using these weak forces. And that was the whole point of the argument of the article that I mentioned at the beginning: to say it was the hydrogen bonds and the other weak forces that were going to be of major importance for biology, and not just the strong bonds, although, of course, the shape of the molecule is determined largely by the stronger bonds.
Well, it was not just that Linus had these ideas. He also started important research programs at Caltech in order to further them. If you want to know the shape of something for instance, of an organic molecule you must know accurately the distance between two atoms, and the angles that you form when there are three atoms that is to say, the strong bonds, although as you know you can get rotation around some of them so it does not totally fix the structure. So he started a program of looking at small organic molecules at Caltech, looking at their structure by X-ray crystallography. They had done several molecules, and they were able, therefore, to know that the carbon-carbon distance was normally 1.54 . Linus laid the essential foundations for this point of view. It was not that other people were not doing some of it, but I think he really pioneered it and pushed it.
The other great thing that I think Linus did was to realize what you can do by building molecular models. The models, of course, have to be to scale, the distances right, the angles reasonably right, and there must be the necessary degrees of freedom when you have single bonds. But what you would do today on the computer those days you did by having these little metal models that showed you the sorts of configurations a molecule might have. In some cases, as with benzene, it fixed the whole structure, but in other cases, as with sugar or something like that, there was more than one structure. And in certain other cases the molecule was very floppy, but Linus did realize the importance of that. He was a much more professional model builder than one or two of the people who had gone before him, like the Englishman William Astbury, who had tried to make models of polypeptide chains, but had done it in rather a sloppy way.
I was too new to understand all of that at the time, but my more senior colleagues, Sir Lawrence Bragg, who won a Nobel prize at 25, and Perutz and Kendrew, also thought it would be a good idea to try and see if there were regular configurations of the polypeptide chain by doing model building, using the parameters of distances and bond angles and certain assumptions. They started to do that while I watched on from the wings doing something else. They decided, of course, that such a model would be a helix, because the essence of a helix is that it repeats, so that you necessarily get a screw axis which leads to a helix, or as mathematicians would say a "degenerate" helix, meaning a straight line or a circle. But they knew it was going to be some sort of helix, and they knew more or less how to do the model building. Unfortunately, they did not know that the so-called peptide or amide bond was planar. They were misled when they asked a theoretical chemist, who ought to have known better, who told them these bonds could wobble about. Linus knew better. So they tried to build various helices and here they had a bit of really bad luck. You will notice in this story as we go on that there are various people who do have bad luck because they make an excusable mistake or they are misled by the data.
It so happens that it was thought that there was a particular X-ray diffraction diagram, which is what you get off hair and nails and so forth, which was called the ŕ pattern by Astbury, who discovered it. And the X-ray data showed that there was a repeat at a certain distance, at 5.1 or 5.15 . You can get that quite unambiguously. If there is a screw axis, it has to be an integer screw axis, such as a two-fold axis or a three-fold axis. It is well known that, except in very special circumstances that have been discovered more recently, you only get two-, three-, four-, or six-fold axes in crystal structures. Bragg realized that if you have a repeating pattern, you can have a four-fold rotation axis. A four-fold rotation axis means that if you turn it through 90 degrees, the pattern looks the same. You cannot have a five-fold axis in a crystal. This is a very old principle. I do not know who first established it. It could well go back to Euler, and I think before that to the Greeks. But if you have an isolated helix, that restriction does not apply; you can have a five-fold one. They thought it must have an integer axis because of this X-ray spot. So they built all the possible models, and none of them looked right. It was very disappointing. They had a good idea. They were trying to build models, but they all looked clumsy, or awkward, or had holes in the middle, or did not do this or that. Nevertheless they wrote up a paper on it, and it so happened that I read the paper, not for the usual reason, but because they all went away when the proofs came, and they said, "Would you please read the proofs?" So I read the proofs, but I did not understand what I was reading. At least I understood it at the superficial level, but not at what I would call the critical level.
Linus was two jumps ahead on this problem. He had realized that whereas they had made the peptide bond floppy it should be more rigid although it is not perhaps as rigid as all that and it should be planar. And he had realized that you did not need to have an integer axis. That was essentially the way the ŕ-helix was discovered. There were some X-ray data that did hint this, which the group at Cambridge ignored, and we now know the ŕ-helix repeats after 3.6 residue. Linus published a paper on the ŕ-helix, and when it came out, of course, my colleagues at Cambridge realized they had missed the bus. They had actually built an ŕ-helix and twisted the poor thing to give it a four-fold repeat instead of 3.6, and it looked just awful. Bragg was very cast down. He went around saying, "the biggest mistake of my scientific career." That time Linus really did hit the nail on the head. For the structure, which Astbury put forward, Linus put forward a better one. Instead of being a completely planar structure, he made it a pleated sheet. There were one or two other structures which did not turn out so well in the series that he published, but the ŕ-helix was really a major achievement.
Now let us come to the case of DNA. Peter has said in a letter that he thought his father really did not care much about nucleic acid, that it was just another structure to be solved. I am not sure that is really true. I think you could have said that of Bragg, but I am not sure that it was true of Linus. But anyway, he wanted to solve the structure of DNA, and this is where he had the bad luck. Because he only had some old X-ray pictures of Astbury's to work on. Now it turned out that Rosalind Franklin in London, by very careful work she was a very good experimentalist had shown there were two forms of DNA. There was a wet form of DNA, and if you dried it a little so there was less water there, then something happened and you got another form called the A form. And when Astbury took the pictures, he had not been very careful about what the humidity was, so the actual X-ray picture he had was a mixture of the A and B forms. In other words, the diffraction pattern corresponded to something which did not exist at all. It was a jumble of two things which were quite distinct. It was not too surprising that, when Linus tried to solve that structure using that pattern, he did not come up with the right answer.
He also made another error, I think, in that he took the density as the dry density. We all know now that there are two chains in the structure of DNA, but if you take a dry density you can persuade yourself it might be three. Indeed, Jim Watson and I were not sure whether it was two or three (although Jim produced a very convincing argument, which he never refers to, though it is a very good argument). It was more likely to be two. Linus thought it was three. And then he did something which I asked him about in later years; and it was just as baffling to him as it was to me and to Jim. He put a hydrogen on the phosphate group. Now, in the paper, this hydrogen was used to form a hydrogen bond which held the structure together. I was flabbergasted. I thought that because the phosphate had a negative charge, there should be no hydrogen on it. I thought I must consult a book of chemistry, and I took down the only chemistry book I had, which was General Chemistry by Linus Pauling, and reassured myself that you only got a hydrogen there at a very acid pH, about pH 1, whereas, of course, these structures existed near neutral pH, or pH 7. I asked Linus in later years, at least twice, and he said he did not know why he did that. It was just a mental aberration. So that was one of the reasons that he did not get the structure of DNA. If he had had Rosalind Franklin's X-ray data, I am quite sure he would have solved the structure in no time at all, because he had all the right background. And I do not think he would have put the hydrogen on the phosphate group.
Perhaps I should explain to you what his problem was in solving structures of this type: the ŕ-helix on the one hand, and the structure of DNA on the other. In both cases, they are only general structures. In the case of the ŕ-helix, we are not talking about the side chains; we are talking about the polypeptide backbone, which is a simple repeating polymer, and there are not too many atoms in it. In fact, you can define the structure by just two parameters the rotation about two of those angles in the backbone so there are only really two parameters you have to decide on. Of course, you must get them right for the hydrogen bonds, so it is not too difficult. It is much easier to solve a polymer structure where everything is joined end to end than the structure of little molecules in a crystal, which can be any old way. Moreover, there was the idea that it should be a helix. Some people thought what we got from Linus in doing DNA was the idea that it was a helix. That is quite wrong. Everybody was thinking about helices at the time; it was the obvious, sensible thing to do. So what did we get from him? Well, Linus realized, first of all, as we did, that the structure is not defined by the chemical formula, because we have these two parameters, the bonds that can rotate; so you only have part of the information from the molecular side to solve the structure. "But surely," you would say, "if you have an X-ray diffraction diagram, that should tell you the structure." Not so. Essentially what scatters the X-rays is the density of the electrons, so what you are measuring is the electron density. And it is perfectly true that if you knew the electron density you could calculate the diffraction pattern. The diffraction pattern is a sort of spatial Fourier analysis of the electron density. And in order to calculate it, you have to know the amplitude of each of these imaginary waves which build up the electron density, and also what is called the phase the relation of one to the other. What the X-ray data give you is the amplitudes, but it does not give you directly the phases, at least in the ordinary way. In other words, you only have half the information from the X-ray pattern. This is called "the phase problem."
You thus have two sources of information. One is what you know about the structure from its chemical formula and the bond distances and angles. That only has part of the information, because of the rotation. You also have these diffraction data, which have only part of the information. What Linus realized was that you should be able to combine these two with a bit of imaginative guessing this is the only way of putting it so that you come up with a plausible structure, or set of structures; and once you had a structure, then you could calculate what the x-ray pattern should be; and you could see that it satisfied things in a molecular way. And that was the lesson that Jim Watson and I learned from Linus Pauling. It was a very important and profound lesson. Our colleagues in London, Maurice Wilkins and Rosalind Franklin, were trying to do it by other methods, by what is called the Patterson synthesis. They had not gotten around to using the method which had been so strongly pioneered by Linus Pauling.
I think that gives you a little glimpse of where my work overlapped with Linus' work. But you must realize that he had made contributions to molecular biology before that, and he went on to make more after that. He was interested at an earlier period in what happens when you denature proteins what happens when you boil an egg, as I tell my wife. After all, it is a little funny. You have the white of an egg, and it is all gooey and runny, and then you heat it, and it all goes solid. So what is happening? Of course, we all know now that the protein molecules, which are in a concentrated solution in the white of the egg, have groups inside which do not like water (they are hydrophobic), and groups on the outside which do like water. If you heat them up and jostle them around, you open up the groups that do not like water and they combine with other ones that have been opened up which also do not like water; and you get a sort of gel, a very solid jelly structure. And essentially Linus had the right idea then when he proposed that this was the general basis of protein denaturation. It was not a change in the strong chemical bonds of proteins. It was a change in the way the weak forces interacted how, in other words, the molecules were folded, and how everything got jumbled when they got unfolded. So that was certainly a very important idea which has turned out to be broadly on the right lines.
Linus also had the right idea about antibodies. He understood that the antigen a little molecule which stimulates the production of antibodies fits into a particular cavity. He did have the wrong idea about how that cavity was constructed, but on the other hand, putting it forward, I think, stimulated the field so that eventually the right idea was found.
There was another occasion during the period when I was still a graduate student, when I was also very influenced by him. I was over in another laboratory, because the Cavendish lab was a physics lab, and their idea of chemistry was heating something up to 800 degrees centigrade. I had to go somewhere else when I wanted to do biochemistry. On this occasion, I was in the process of preparing some hemoglobin. In those days you did it with a big centrifuge, and it took some time to run. As I was waiting for this to run, I came across an article in Nature about sickle-cell hemoglobin, by Linus Pauling, in which he said it was a molecular disease. It had been shown quite recently by geneticists that indeed the gene for it was inherited in a Mendelian manner. I will not go into the details, but it looked as if it were caused by a single gene. What Linus and his co-workers had shown was that there was a change in the hemoglobin molecule, so that under electrophoresis it moved at a different rate. In other words, one of the charge residues had been changed. And I remember how excited I was at this idea, because this was the sort of thing I was very interested in. I was very interested in genes. Here was presumably a gene product. It looked much the same, as far as one could tell, in many of its properties, but there clearly had been some change.
As a result of that article, when Vernon Ingram came to join us at the lab, he and I wanted to find out what happened when there was a mutation in a gene. Did it affect that particular protein which we believed was controlled by that gene? We did not make any progress on that. We tried working on lysozymes and various other things, but Vernon received a specimen, I think through Perutz, of sickle-cell anemia hemoglobin, and he was able to show that Linus was right. There was a change in a single amino acid. Here we had a very dramatic case of a disease where, if you have two bad copies of the gene, you probably at least in those days would not live much beyond your teens. And yet it is a change, as we know now, of just one base pair in the DNA, one amino acid in the protein.
You may ask how it is possible that a small molecular change can do something which will kill you. But you realize that in the red cell, first of all, there are many copies made of the gene onto the messenger RNAs, so there is amplification there. But there is amplification before that, because from the fertilized egg there come many, many red cells. There are an enormous number of red cells being made by a stem cell, because the cell is dividing, so you have got a lot of red cells in your blood which carry the hemoglobin. Then the mRNAs produce many copies of the molecule. Thus, in fact, you get a large mass of not very good molecule, even though it is a very tiny change in the egg and the sperm. Just one or two atoms here and there is enough to make a change which is potentially lethal. Nowadays, of course, these ideas are commonplace, but in those days that was not the case. You may be surprised to know that most people who worked on protein chemistry wanted to know merely what the composition was. Fred Sanger had just been doing the sequence of the first protein to be sequenced, which was insulin. Protein biochemists really had no idea their subject had anything to do with genetics. Those of you who are in the field may find this absolutely astonishing, because now it is commonplace. One of the major functions of genes is for each one to code for a particular protein. That was not known in those days, so this was a very dramatic idea. In fact, as a result of it, Sydney Brenner, Seymour Benzer, and I gave lectures to Fred Sanger's group on the elements of genetics. It was the first time they realized that they had to learn something about the subject. Nowadays, of course, all of you who go into these things learn it right from the beginning and take it as a matter of course. And it was very much sparked off by this discovery of Linus that sickle-cell hemoglobin was, as he said, a molecular disease.
I hope I have given you some idea of the things he did. He did other things as well. With Emile Zuckerkandl, he had the very important idea of the molecular clock. That is only perhaps an approximation, and one should not exaggerate, but it has turned out to be much truer than people thought at the time. Mutations in a gene tend to accumulate at a fairly regular rate over time. Of course, it is a stochastic process, so it is not really regular; but the average is regular, so to some extent you can date things in evolution by seeing how many changes there are between the ends of a Linnean tree between two creatures, and that will give you some idea what their common ancestor was. And although, as I say, this has to be used with care, that was another basic idea.
I think the message one gets from this was that Linus Pauling was enormously fertile in the way he developed his ideas. He was asked a few years ago, "How do you get ideas?" And he gave I think what is the correct reply. He said, "If you want to have good ideas you must have many ideas. Most of them will be wrong, and what you have to learn is which ones to throw away." Having the ideas is the easy part; he did not say this I am simply interpreting here what he probably thought of as the easy part. The difficult and rather intuitive part is to know which ones to hang on to and which ones to throw away. It is clear from what I have said that Linus Pauling was not always right in his ideas. But my belief is that, in most cases, if somebody is always right in his ideas you find that he does not have much to say. It is an expression of somebody's fertility that he does produce quite a number of ideas, and I think Linus Pauling's score is pretty high. I do not know what Linus would have said about it, but I certainly find in myself that, as you get older, this intuitive knowledge of which ones to discard perhaps weakens a little. Maybe with some of his later ideas, although they were on the right lines, he perhaps might have clung to them a little too strongly. I find myself doing just the same.
How should we summarize Linus' contribution? I do not think, as I said earlier, that it is right to discuss the impact of Linus Pauling on molecular biology. Rather, he was one of the founders of molecular biology. It was not that it existed in some way, and he simply made a contribution. He was one of the founders who got the whole discipline going. And he got it going because, first, he understood chemistry and physical chemistry and he believed that that was the right way to think about these processes not in terms of mysterious forces. And he got it going, second, because he was genuinely interested in biological things. He looked out into the biological world to see where he would apply the right set of ideas. In addition, therefore, to his enormous contributions to chemistry and his humanitarian work, I think we should celebrate him as one of the founders of molecular biology, which as you know is flourishing today.
© Copyright 1996, Special Collections, Oregon State University Libraries, Corvallis, OR 97331-4501
All rights reserved, including the right of reproduction in whole or in part in any form.