Werner Arber (2015) - Insight into the Laws of Nature for Biological Evolution

I'm trying to cover 150 years of history of science in 30 minutes. Let's start with Gregor Mendel, Charles Darwin, and Friedrich Miescher on three fields - genetics, evolutionary biology, and nucleic acid biochemistry. In the first almost 100 years, not very much advance was being done. At that moment, people looked at phenotypes. Genetics was not known as genetics, as such. Phenotypes means you look at the organisms, what they do, and you identify that with various tools, your eyes and other senses. Then, you come to some conclusions. It is in the 1940s only that genetics started to be done with microorganisms, bacteria, and with particular bacterial viruses,which are called bacteriophages. You see here that there are three important, very rapidly showing up conclusions. First, transformation, as investigated by Avery and his colleagues in the 1940s, showed that isolated DNA molecules sometimes can penetrate into bacterial cells. If these cells are variants of the one which is the so-called donor bacterium, then sometimes some genes are transferred and replace the mutated genes by the original wild-type. This is called transformation. When Avery and his colleagues discovered that in pneumococcal bacteria, what they actually did, it was known that a broken-open donor can transform other bacterial strains which are genetically different. It was known, but they didn't know what was the substance which made that difference. This group of three people, they purified several components. For example, the donor DNA was highly purified, so there were no other biological molecules in that fraction, and only that DNA fraction gave rise to this transformation and no other fraction. You can imagine, there was only a minority of biologists who believed that. They couldn't understand that, because one did expect that it would be not possible on nucleic acids, which have only four different building blocks, four nucleotides, to have that complex information which is the genetic information. The second possibility was that people working around Joshua Lederberg with E. coli bacteria, they could show that, by chance, these could sometimes give rise to coupling the donor cell with the recipient cell, and genes were transferred. It was pretty soon then later shown that transfer is a linear thing in this coupled transfer. That means that DNA molecules must be long filaments. Student Norton Zinder, working with Joshua Lederberg, had the chance to see whether what they had found with E. coli bacteria is also true for Salmonella bacteria. The first results indicated that it might be so. But when they looked more carefully, how the donor DNA was transferred into the recipient bacteria, they found that these were viral particles, bacteriophage particles, P22. Later on, it was shown that these are really able to do what we call now transduction. Bacteria sometimes incorporate in their coat the bacterial DNA rather than the viral DNA. A small minority, but you can select for specific markers. I forgot to say, in 1953, of course, were the two Nature papers in the spring of Watson and Crick showing the double-helical structure of this filament, this DNA. Everything became clear at that moment, and people started to pay attention to that. Slowly, the idea that DNA is really the carrier of genetic information became more strong. I show you here, symbolically, the large chromosome of E. coli bacteria, which carries almost 5 million base pairs. Base pairs as shown above - T, A, G, C, and so on - in various orders, are actually the genomes of these bacteria. Just one single, large DNA molecule. As I mentioned, from conjugation, one could already predict that this was just one molecule and it's circular. In the wrong scale I just showed there a gene, which is composed essentially on the coding region, which is important for synthesizing particular gene products, mostly proteins. There are expression control signals, of course, because these genes have to be expressed when the product may be needed. Then, from what we learned from Watson and Crick, is that actually a mutation, which is causing phenotypic variations, is most likely an alteration in the parental DNA sequences. At that moment, one was not able to identify these sequences yet. I started in the fall of 1953 my postdoctoral work at the University of Geneva. I came in with majoring in experimental physics, and I had to take care daily of a still relatively primitive electron microscope at that time. In the afternoons, when in the morning I had set up the microscope, cleaning the tube, and so on. In the afternoon, we did research. I started research looking at bacteria, at bacteriophages, and, in particular, at mutants of bacteriophage lambda. One of these mutants, which I also incorporated in this, is a particular mutant of lambda, which is a mutant not producing any visible particle in the electron microscope. But it was clear that the genome was there, that the genes became expressed, and some of the genes are actually host genes. My task then was to go not only with the microscope into these studies, but starting to do genetics. A few mutations of bacteriophage lambda were known at that time. I was able, as shown in this projection, that when this lambda genome - this was already known: Lysogenic bacteria are actually hybrids between the host genome and the viral genome. When you shine ultraviolet light on that, virus production is ... But sometimes the excision of the here black viral genome is not at the right size. So that you get the hybrid between part of the viral genome, which was still able to replicate, and a few adjacent bacterial genes which happen to be genes for galactose fermentation. That could be easily selected. The conclusion was: this defective virus was a hybrid still being able to replicate as a plasmid but not making real viral particles because host coat and tail genes were clearly missing and left behind in the donor bacterium. Instead there were these genes for the capacity to ferment galactose. This idea was accepted by, at that moment, relatively few molecular geneticists. And they reflected on it, knowing that the genome's are tremendously long DNA molecules with many genes; E. coli has almost five thousand genes in its genome. How do you study genes? The best is sorting it out as nature does in this particular case and then using that and replicating and harvesting from this bacteria, the DNA starting to see what the structure and finally the function of these genes are. That was in several pets, but the experiment was not so easy, because how do you sort out a gene which you want to study? We were aware already at that time that several factors limit gene acquisition. Namely, the surface of the donor and the recipient must fit together, for example in conjugation, in the phage infection. The phage must be able to infect the recipient cell and so on. When the DNA penetrates from another bacterial cell into the recipient cell, then restriction modification starts to act. This was a phenomenon which has been described but not explained. Then functional capacities, that's the selection level for propagation and for expression of noble genes. It was already concluded, nature uses these possibilities but it's a strategy in small steps of acquisition of foreign genetic information. Nowadays we call that horizontal gene transfer. I just briefly explain the phenomenon of restriction modification here. K strain red, drawn red here, if you grow that bacteriophage longer in K, you have a progeny after an infection of 100 or 200 particles and each one can go back to another still intact host K. So that can go forever. If you, however, change the host, sometimes we have a host which is marked here by zero, which has no restriction modification in the enzymes, you still go with an efficiency of one. But if you go back to the original host K from the host zero, with these phages, you only get only one in ten thousand which is reproductive. And the phage coming out is again red, so can go back to K and so on. This phenomenon is called restriction and putting the right colour is called modification. The conclusion was there must be these two enzymes for DNA restriction and for its modification of the host. The hunt for enzymes started in the 1960s. At the end of the '60s it was finally successful. You see not all of the restriction enzymes work in the same way. The type-one enzymes are more complex. We had already worked with E. coli which has a type-one restriction system that cuts the DNA into fragments but not reproducibly. You see here a DNA molecule having - down there, type one - having two recognition sites, and the recognition sites are such that wherever I put the star here in the post there is a metro group attached, that's the modification. Modification is DNA regulation. So that the metro group attached to one particular nucleotide does not hinder the expression of the gene. The other enzymes of type-two, this is also relatively widespread in nature, is that again you have the recognition sites for the enzyme E co R1 here - it's G-A-A-T-T-C - And when one or both stars there are not metro groups - they are foreign because the donor would have another recognition site - then the DNA is cut as shown here, reproducibly always in the same way. These enzymes had first been isolated from Haemophilus influenzae by my colleague Hamilton Smith who is also at our meeting. And he, together with me, finally got the Nobel Prize in 1978 together with another colleague, Nathan Smith, who had also been working in the same university as Ham Smith and he worked on human DNA cancer virus, S340. And the three of us got the prize for clearing up the restriction systems, being the first to isolate an enzyme which was very useful for genetic engineering as I will show you now. And a third for its applications of medical interest. So at that moment in the early 1970s, several groups started to use that knowledge, taking a vector that can be a viral DNA or another small, we call that plasmid in bacterias sometimes. There are these, and conjugation is actually due to plasmids working as fertility factors. If you take donor DNA from any origin, cut into fragments which are relatively small, you can incorporate it into that vector, transferring it into some bacteria by various tools. And if you are successful, you can multiply that plasmid and also express the genes which are carried on that, including the inserted red gene. So that's genetic engineering. We discussed early in the 1970s the possibility whether there is some risk to do that. Because if you are looking for a yet unknown gene of unknown function which you sort out,you multiply it highly in the lab and study its structure and function, then you have to be careful not to infect you because you don't know. It could be pathogenic or have some other toxic effect. So the question was then finally deliberated at a few scientific meetings, letter to 'Science' - one should really take that very carefully. And in February of 1974, it came to the Asilomar Conference where all these questions were discussed. And the conclusion was: There are short-term and long-term potential risks. Short-term, as I mentioned, pathogenicity and so on and so forth. And there the recommendation was: Just apply the same method as has already been done for a few decades in medical microbiology. If you take from a sick person some bacterial samples, sending them to an analysis laboratory. These people working in this analysis have to be very careful not to be infected also. That was the rule, but then there came the question: If ever some of this recombinant DNA would penetrate, would get into the environment, either by accident escaping from the lab, or by purpose, liberating something in order to produce these particular cheap products. Then it might be possible at that moment - remember we are in the mid-1970s - that it might be able to transfer into other bacterial strains. I didn't realize then that working with bacterial cultures, which grow very rapidly, generation time of - under very good living conditions - half an hour. So in one day you have a large population, and it's easier to do population genetics and studies with bacteria than with higher organisms which have often generation times of many years. I show you here, what at that moment in biological evolution was known. This is neo-Darwinian evolution that creation, spontaneous creation of mutations, that means genetic variation, is the driving force of evolution. Without having any alteration in the genome, you wouldn't have any evolution. The natural selection which is here characterized by the effect of the environment, and the environment is both physical-chemical and biological. All the other living beings in the same eco-system can also influence the selection of new variants. Third, isolation as Charles Darwin had seen on the Galapagos Islands. There, of course, this is geographic isolation; there are also many types of reproductive isolation that you wouldn't be fertile if you just bring in some DNA which is not related to the organism. That modulates the process of evolution. The idea was then in the 1970s, I decided to go together with many colleagues in microbial genetics into studying the molecular processes of genetic variation in order to understand how that works. We already knew at that time that a mutation, here defined as an alteration of the sequence of the nucleotides, is, in fact, not so often favourable in giving a selective advantage. More often you see selective disadvantage, even extreme lethality. Often also an alteration in the nucleotides is neutral, doesn't give any alteration in the phenotype. So we conclude from that, that the mutations should be relatively rare, spontaneous mutations. That's what one really observes also in bacterial cultures in higher organisms later on also. That's a fact that horizontal transfer does occur but relatively rarely. What I show you here is already ... I will come back later on to the details. This is an indication on how the mutations are created. There are a number of mechanisms. Nature is very inventive. You find a lot of different specific processes contributing to variation, and you can subdivide these into strategies of genetic variation, local one or a few adjacent nucleotides, DNA rearrangement and DNA acquisition by horizontal transfer. What I show you here, I'm pretty happy to show you that. This is a contribution to what I heard yesterday on interdisciplinarity. One does know that nucleotides sometimes have short living isomers called tautomeric forms. For example, the adenine can have its hydrogen atom jumping to another site and then it doesn't pair any longer with thymidine, but by chance it pairs with cytosine. At that moment you have, of course, the adenine goes back to its normal, stable standard form, you have a mispairing. Still nowadays most textbooks say, these mutants are replication errors. It's a completely wrong attitude of understanding nature. Nature uses short-living tautomeric forms of nucleotides in order to occasionally create a nucleotide variation and alteration. And, in fact, most individual organisms studied in that respect have been shown to have repair systems to inhibit rapidly the fixation of the wrong pairing and replacing correctly in the mutated strand that nucleotide. But again, cleverly, not with full, hundred percent efficiency so that very rarely you have substitution. Watson and Crick in the fall of 1953 already showed that, and people later on just ignored it for decades. I show you here there a number of rearrangements possible by enzymes, often general recombination. Transposition of mobile genetic elements jumping from one site to another in the genome. Site-specific recombination, these enzymes can give rise to duplications, to deletions. It can give rise to inversions and new fusions and that sometimes also has some effect. In fact, the three differences - local sequence change, DNA arrangement, and DNA acquisition - have different qualities in their contribution to evolution. Local sequence change often is a stepwise improvement of something, if natural selection selects for it. DNA arrangements give rise to new fusions of functional domains or of alternative expression control signal with a functional gene while the horizontal transfer is a sharing in successful developments made by others. That's quite successful. I come now to just mention a few conclusions. The blue type is what Charles Darwin said, that living beings have common - he doesn't say whether one or many - common origins, but it's good to show that tree, and there is good evidence for that With horizontal transfer you can draw between branches of these trees. Transfers which are only once used usually and that gives rise that, as I mentioned, you can profit from something, some gene functions which were developed elsewhere. And I conclude that Charles Darwin is right, we have common origin, but also common future. That's quite essential and I ask you to accept that as an important thing. And that's for me a very important additional argument to safeguard the big biodiversity. Because if we lose that biodiversity where are all these developed genes that could be at some time or another, at future times, helpful for any organism. I just go very rapidly. There are evolution genes which are either variation generators or modulators of the frequency of variation. Then nature also uses non-genetic elements; I just remind you of the tautomeric form of nucleotides. This is a structure of flexibility of biologically active molecule materials. Random encounters: If a virus carries genes from one organism to another, it's a chance to infect that one or another one and so on. And environmental mutagens. Nature uses that and the conclusion is that natural reality takes actively care of biological evolution. It's an active process and we have difficulty in understanding and believing that because it's inefficient. It must be inefficient, otherwise we wouldn't be able to live because our genome has to be of a good stability. Only rare individuals can try-out whether a newly known mutation can be advance or lethality. You see you have these two antagonistic principals, promotion of genetic variation and limitation. And it is really a very nice fine tuning that in the past the evolution genes have made. So we see if we study these things, that in most of the genomes it has been shown that the two kinds of genes are there. It's a duality of the genome to the benefit of individuals: the majority of the genes - the housekeeping genes, accessory genes and developmental genes - in higher organisms. Whereas the evolution genes are actually the source of, in fact, genetic evolution and biodiversity. I will finish here but just to say: If you apply all that knowledge to genetic engineering you will see that genetic engineering is not fundamentally different from what nature does in horizontal and other gene transfer all the time. And in biotechnology, in classical including agriculture, you go into nature, you see which plants can serve for me as a food, which animals are useful for me and you domesticate those. And nowadays you can include your knowledge and sometimes improve some function, making differently higher expression and so on. And introduce a gene of possible interest in another organism in order to harvest its product and use it. That's called domestication of genes. Thank you for your attention.

Werner Arber (2015)

Insight into the Laws of Nature for Biological Evolution

Werner Arber (2015)

Insight into the Laws of Nature for Biological Evolution

Abstract

Both evolutionary biology and genetics have their roots 150 years ago in work with phenotypic variants of plants and animals. In contrast, microbial genetics originating as recently as the 1940s, rapidly revealed that filamentous DNA molecules are the carriers of genetic information. We will discuss further steps in the exploration of structure, functions and evolution of genetic determinants, in particular of bacteria and bacterial viruses. These studies revealed a multitude of specific molecular mechanisms contributing to the spontaneous production of genetic variants. These specific molecular mechanisms can be assigned to three natural strategies of genetic variation with qualitatively different contributions to biological evolution: (1) Local nucleotide sequence alterations can contribute to step-wise improvement of a particular function; (2) segment-wise DNA rearrangements can lead to improvement of available capacities; (3) DNA acquisition by horizontal gene transfer can be viewed as sharing successful developments made by others.
These processes are the drivers of biological evolution and they employ enzymatic activities of so-called evolution genes acting as variation generators and as modulators of the frequency of genetic variation, together with a number of non-genetic elements. On the basis of acquired knowledge on the slow, but steady progress of biological evolution, we can conclude that natural reality takes active care of biological evolution. Increasing evidence indicates that this process of self-organization applies to all kinds of living organisms. Aspects of worldview related to our conclusions will be discussed, as well as impacts on biotechnological applications.

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT


Related Content